本篇博文主要内容为 2026-07-14 从Arxiv.org论文网站获取的最新论文列表，自动更新，按照NLP、CV、ML、AI、IR、MA六个大方向区分。
说明：每日论文数据从Arxiv.org获取，每天早上12:30左右定时自动更新。
提示: 当天未及时更新，有可能是Arxiv当日未有新的论文发布，也有可能是脚本出错。尽可能会在当天修复。
目录
概览 (2026-07-14)
今日共更新1167篇论文,其中:
- 自然语言处理共149篇(Computation and Language (cs.CL))
- 人工智能共335篇(Artificial Intelligence (cs.AI))
- 计算机视觉共223篇(Computer Vision and Pattern Recognition (cs.CV))
- 机器学习共312篇(Machine Learning (cs.LG))
- 多智能体系统共29篇(Multiagent Systems (cs.MA))
- 信息检索共36篇(Information Retrieval (cs.IR))
- 人机交互共41篇(Human-Computer Interaction (cs.HC))
多智能体系统
[MA-0] Can LLM s Perform Deep Technical Comprehension of Computer Architecture Papers?
【速读】：该论文旨在解决大语言模型（Large Language Models, LLMs）在计算机体系结构领域论文中进行深度技术理解的能力瓶颈，即超越简单摘要，实现结构化批判性分析的问题。其核心挑战在于能否准确识别论文的核心机制、揭示隐含假设，并将贡献置于更广泛的技术背景中进行延伸评估。为此，研究提出Gauntlet——一个开源的多智能体分析流水线，通过五个独立的专家角色（expert-persona reviewers）对论文进行多维度评审，并引入对抗性合成阶段整合观点。在20篇ISCA 2025与HPCA 2026会议论文上的评估显示，15名研究人员在对比人类分析与Gauntlet输出时，普遍更倾向于后者，尤其在“批判严谨性”（Critical Rigor）方面优势显著（配对Wilcoxon检验，p < 0.01）。尽管人类在可信度和实用性上偶有胜出，但主要源于对错误断言的自信或未充分教学的机制描述等表面因素，而非深度洞察。98篇论文的自动化消融实验进一步表明，性能提升主要源自多智能体架构设计，特别是合成阶段的协同作用；当同一模型以单个“丰富角色”运行时，仅在4%的论文中表现优于多智能体流程。因此，解决方案的关键在于采用分角色协作与对抗性合成相结合的多智能体系统，从而实现对技术贡献的深层解析与系统性验证。
链接: https://arxiv.org/abs/2607.11859
作者: Nishant Aggarwal,Ayushi Dubal,Sreeraj Kannakarankodi,Ian McDougall,Adarsh Mittal,Vishnu Ramadas,Noah Scott,Ranganath Selagamsetty,Weichu Yang,Karthikeyan Sankaralingam
机构: University of Wisconsin–Madison (威斯康星大学麦迪逊分校); NVIDIA Research (英伟达研究)
类目: Computers and Society (cs.CY); Hardware Architecture (cs.AR); Multiagent Systems (cs.MA)
备注: 4 pages, 1 figure
Abstract:Can large language models perform deep technical comprehension of computer architecture papers – not summarization, but structured critique that names the core mechanism, surfaces buried assumptions, and connects a contribution beyond its own scope? We study Gauntlet, an open-source pipeline that analyzes a paper through five independent expert-persona reviewers and an adversarial synthesis stage. On 20 ISCA 2025 and HPCA 2026 papers, ten researchers each wrote their own analyses and then judged, for papers other than their own, the human analysis against Gauntlet’s. Across the 20 comparisons evaluators preferred Gauntlet in 15 (human in 4, one tie); its advantage is significant on per-analyst totals (paired Wilcoxon, p 0.01) and largest on Critical Rigor, vanishing only on Calibration. Where humans win, it is on trust and usefulness rather than depth: a confident wrong claim, a mechanism described but not taught, or unprioritized breadth. A 98-paper automated ablation shows the gain comes from the multi-agent structure – the pipeline beats the same model run as a single rich-persona agent on 96% of papers – and specifically from its synthesis pass. We release all analyses, scores, and the rubric as a community resource.
[MA-1] Forgetting Our Way to Shared Meaning: Effects of Forgetting on Conceptual Alignment in a Non-Partnership Coordination Game
【速读】：该论文旨在解决在缺乏协调博弈（coordination game）的条件下，群体中共享意义（shared meaning）如何涌现与演化的机制问题。传统概念语义模型难以解释个体在非合作情境下如何形成并动态调整共同认知范畴，因其普遍依赖于参与者具有共同收益的合作设定。本文的关键解决方案在于构建一种非合作情境下的概念对齐（conceptual alignment）模型，通过引入具有不同适应性（adaptiveness）和记忆退化（memory degradation）特征的代理（agents）进行反事实模拟，揭示实际与感知层面的概念收敛差异。研究发现，具备适应性的代理能更快实现实际概念区域的收敛，且最终收敛范围更紧密；而无适应性的代理虽更早感知到收敛，但实际一致性较差。此外，随时间逐渐降低新信息权重可促进更稳定的共识形成，优于固定新信息权重的策略。结果表明，记忆特性在实际与感知收敛的生成与演化过程中起决定性作用。
链接: https://arxiv.org/abs/2607.11787
作者: Landon Liu,Mary Kelly,Alan Tsang
机构: Carnegie Mellon University (卡内基梅隆大学); Carleton University (卡尔顿大学)
类目: Multiagent Systems (cs.MA); Computation and Language (cs.CL); Computer Science and Game Theory (cs.GT); Human-Computer Interaction (cs.HC)
备注: CogSci 2026 Meeting Conference Proceedings
Abstract:Shared meaning in language requires people to learn and agree on categories. We ask how characteristics of agents’ memories change the emergence and evolution of shared meaning. Without a coordination game, models of conceptual semantics cannot explain how shared meaning emerges and changes in groups of people; however, existing games assume that players share payoffs in a partnership setting. We model conceptual alignment as a non-partnership game and illustrate differences in actual and perceived conceptual convergence from counterfactual simulations using agents with varying levels of adaptiveness and memory degradation. We found that adaptive players achieved actual convergence faster and had closer final conceptual regions than non-adaptive players, while non-adaptive players perceived convergence earlier. Weighing novel information less over time resulted in more stable agreements than fixing the weight of novel information. Memory features are critical to the emergence and evolution of actual and perceived convergence.
[MA-2] Paradoxes of Game Theoretic Equilibria and Price of Anarchy
【速读】：该论文旨在解决传统算法博弈论中静态均衡概念（如纳什均衡、相关均衡与粗相关均衡）及价格劣化（Price of Anarchy, PoA）分析在多智能体学习动态过程中的局限性问题。现有方法通常依赖于无后悔学习（no-regret learning）快速收敛至静态均衡，并以此作为效率评估的基准，但这种将多智能体学习简化为静态均衡分析并采用黑箱后悔率分析的范式，掩盖了系统内在的动态非均衡行为以及真实的博弈论边界。其核心解决方案的关键在于揭示：首先，内部纳什均衡缺乏C¹类向量场信息，导致智能体无法区分激励对齐与严格对抗情形；其次，决定鲁棒PoA上界的最坏情况纯纳什均衡本质上表现为拓扑不稳定的严格鞍点，在典型拥堵博弈中则体现为几乎处处被严格占优策略所支撑的全局排斥子。将效率保证锚定于此类不稳定状态，引发代数敏感性——研究证明，当允许所有正仿射成本时，PoA可变为无界。此外，将学习轨迹投影至相关策略的离散单纯形上，会系统性地包容非理性行为；通过粗相关均衡或近端精炼方法评估动态行为，仍无法排除严格占优策略的存在。更进一步，即使达到最优的O(1/T)交换后悔最小化，仍无法避免宏观湍流现象，即在最小规模博弈中亦可能出现混沌极限集。最后，研究还考察了拥堵博弈的非原子极限情形，尽管传统认为其具有高度稳定性且具备紧致的次线性Θ(p/ln p) PoA界（其中p为多项式阶数），但在离散时间学习机制下，唯一均衡会退化为Li-Yorke混沌，并形成时间平均效率呈指数级恶化（2^p）的全局吸引子。这些发现表明，必须重新审视基于最坏情况均衡的框架，转向以动态基础度量为核心的新型分析范式。
链接: https://arxiv.org/abs/2607.11752
作者: Georgios Piliouras,Ian Gemp,Siqi Liu,Luke Marris
机构: Google DeepMind(谷歌深度思维)
类目: Computer Science and Game Theory (cs.GT); Machine Learning (cs.LG); Multiagent Systems (cs.MA); Dynamical Systems (math.DS); Optimization and Control (math.OC)
备注:
Abstract:For decades, static solution concepts (Nash, Correlated, and Coarse Correlated Equilibria) and the Price of Anarchy (PoA) have formed the bedrock of algorithmic game theory, with no-regret learning proving fast convergence to such game-theoretic equilibria. We show that reducing multi-agent learning to static equilibrium and black-box regret analysis obscures underlying dynamic disequilibrium and game theoretic bounds. First, interior Nash equilibria lack C^1 vector field information, meaning agents cannot distinguish aligned from strictly opposing incentives. Inheriting this geometry, the worst-case pure Nash equilibria dictating robust PoA bounds manifest as topologically unstable strict saddles, and in canonical congestion games, as global repellers supported on almost everywhere strictly dominated strategies. Anchoring efficiency guarantees to these unstable states causes algebraic sensitivity; we prove that accommodating all strictly positive affine costs renders the PoA unbounded. Furthermore, projecting learning trajectories onto the discrete simplex of correlated play systematically accommodates non-rationalizable behavior. Evaluating dynamics via Coarse Correlated Equilibria or proximal refinements fails to preclude strictly dominated strategies. Moreover, optimal O(1/T) swap-regret minimization does not preclude macroscopic turbulence, manifesting as chaotic limit sets even in minimal games. Finally, we examine the non-atomic limit of congestion games. Though considered highly stable with tight sub-linear \Theta(p/\ln p) PoA bounds (where p is the polynomial degree), we prove that under discrete-time learning, the unique equilibrium destabilizes into Li-Yorke chaos and global attractors whose time-averaged inefficiency degrades exponentially as 2^p . These results necessitate re-evaluating worst-case equilibrium frameworks for dynamically grounded metrics. Subjects: Computer Science and Game Theory (cs.GT); Machine Learning (cs.LG); Multiagent Systems (cs.MA); Dynamical Systems (math.DS); Optimization and Control (math.OC) Cite as: arXiv:2607.11752 [cs.GT] (or arXiv:2607.11752v1 [cs.GT] for this version) https://doi.org/10.48550/arXiv.2607.11752 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[MA-3] When Local Monitors Miss Compositional Harm: Diagnosing Distributed Backdoors in Multi-Agent Systems
【速读】：该论文旨在解决多智能体、工具调用型大语言模型（LLM）系统在实际部署中面临的关键安全问题，即现有运行时监控机制在面对分布式后门攻击时存在根本性漏洞。其核心问题是：当恶意载荷被拆分并分布于多个智能体之间时，每个智能体的局部行为均表现为正常（局部良性的），导致基于单个消息、工具调用或步骤的本地监控无法察觉异常，尽管整体系统正在执行攻击。这种攻击利用了“可观测性边界”（observability boundary）——即监控器只能检测到与其视图中良性流量可区分的内容。研究证明，一旦各片段在监控视图下表现得完全良性，无论检测器多么强大，都无法识别这些片段。通过控制实验环境、外部基准测试及端到端智能体运行验证，结果表明本地监控器的检测能力随局部证据消失而失效，仅在观察到完整组装后的对象时才能恢复。进一步发现，仅在训练于良性数据的监控器能够重构攻击代码结构（平均AUROC达0.874），而基于解码视图的门控机制在已知编码族的情况下可有效阻断所有测试攻击。然而，仅增加观测范围并不足以解决问题；全轨迹监控与解码器若未触及恶意载荷暴露的表示层，仍会失败。因此，关键解决方案在于识别并访问承载有害信息的底层表示空间，因为局部安全不能保证全局安全，而如何定位这一关键表示仍是开放难题。
链接: https://arxiv.org/abs/2607.11751
作者: Yibo Hu,Ren Wang
机构: 未知
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG); Multiagent Systems (cs.MA)
备注:
Abstract:As multi-agent, tool-using LLM systems are deployed, a common safety net is a runtime monitor that checks each message, tool call, or step on its own. We show this net has a fundamental hole. A distributed backdoor splits a harmful payload across agents, so every local check passes while the assembled object is the attack. The monitor can be right on every step and still miss the attack. The problem is not splitting itself: split fragments can still leak suspicious tokens or provenance edges. The hard case is \emphlocal benignness. No fragment carries the harm, and what is left looks like ordinary benign traffic. We formalize this as an \emphobservability boundary: a monitor catches only what its view can tell apart from benign traffic. We prove that once the fragments look benign in the monitored view, no detector on that view can catch them, however strong it is. Across a controlled testbed, an external benchmark, and end-to-end agent runs, local monitors lose the signal exactly as local evidence disappears, and it returns only when the monitor sees the assembled object. A monitor trained only on benign traffic recovers the attack’s code structure across held-out encodings (0.874 mean AUROC). A decoded-view gate, given the encoding family, blocks every tested attack. But seeing more is not enough: full-trace monitors and decoders still fail unless they reach the representation where the payload is exposed. Local safety is not global safety when harm is compositional, and the open problem is finding that representation.
[MA-4] Multi-Agent Reinforcement Learning for C-V2X RAT Selection
【速读】：该论文旨在解决车联网（V2X）环境中多类应用对通信时延与可靠性提出异构需求，同时多种通信技术（如蜂窝Uu链路、NR-V2X PC5侧链路）各具优劣，如何在动态场景下实现通信链路的智能选择以满足不同应用需求的问题。其解决方案的关键在于采用多智能体强化学习算法MAPPO，通过构建多智能体协同决策框架，实现对蜂窝Uu链路、NR-V2X PC5侧链路以及两者并行使用等通信模式的自适应动态选择。相比传统的深度强化学习（DRL）方法、静态决策树及固定信道选择策略，该方案在城市场景下的典型通信用例中显著提升了按时交付率（从0.508提升至0.535，全车辆执行策略时达0.567），同时训练时间缩短一半。性能提升主要源于对先进应用场景（如协同驾驶、共享感知）的支持，验证了自适应通信策略在下一代高要求V2X应用中的有效性，也表明多智能体建模在解决复杂通信决策问题中的关键作用。
链接: https://arxiv.org/abs/2607.11744
作者: Moritz Schaffenroth,Uwe Kölbel,Heike Lepke,Alexander Prinz,Alfred Höß
机构: 未知
类目: Multiagent Systems (cs.MA)
备注:
Abstract:Vehicles are increasingly equipped with advanced V2X communication capabilities. While early V2X apps utilized services such as Cooperative Awareness Messages, recent developments have allowed more advanced applications including cooperative driving, shared perception, and sensor-sharing services. The broader mix of applications leads to heterogeneous requirements for latency and reliability. At the same time multiple communication technologies for V2X are available with pros and cons. Hybrid V2X communication can exploit the distinct advantages at the right moment to fulfill the requirements of the applications. This work studies the decision problem between cellular Uu link, NR-V2X PC5 sidelink, and the simultaneous use of both channels. We address this problem by using the multi-agent reinforcement learning algorithm MAPPO and compare it to five baselines consisting of a deep reinforcement learning (DRL) approach, a static decision tree approach and static channel selection strategies. The methods are evaluated in an urban scenario and with a set of selected communication use cases. The evaluation results show that when compared to the DRL approach, the on-time delivery ratio improves from 0.508 to 0.535 in a single-controlled-vehicle setting and from 0.548 to 0.567 when all vehicles follow the learned policy and reduces the training time by half. The gains result mainly from the advanced applications scenarios, as opposed to scenarios involving exclusively CAM messaging. This indicates future applications will benefit from such adaptive communication strategies and that multi-agent modelling is useful for addressing the underlying decision problem.
[MA-5] An Explainable Agent ic System for Detection of Conversational Scams with Summary-Based Memory
【速读】：该论文旨在解决生成式人工智能（Generative AI）快速发展背景下，持续性对话诈骗（conversational scams）日益增长的威胁问题。此类诈骗通常持续数周甚至数月，通过逐步建立信任关系来诱导用户泄露资金或敏感信息，而现有检测系统多聚焦于孤立消息，难以应对这种动态演进的攻击模式。其解决方案的关键在于提出一种可解释的代理式（agentic）系统，能够从对话层面进行欺诈检测，并引入首个公开的多类别对话诈骗检测基准——ConScamBench-278，涵盖八类诈骗类型，支持可复现评估与未来扩展。实验表明，该系统在孤立消息检测中达到100%钓鱼召回率，在LoveFraud02数据集上实现全部对话诈骗的识别（83/83），并在ConScamBench-278上达到97.8%的准确率（95%置信区间[95.4, 99.0]）。两项用户研究（N = 100 和 N = 45）进一步验证了系统的有效性：参与者普遍反映在判断可疑对话时存在不确定性；在非控制的前后对比中，用户自评的信任度、自信心及对AI检测的需求感均显著提升（p < 0.001，Wilcoxon符号秩检验），系统可用性量表得分达74.7（95% CI [72.5, 76.9]），超过公认可用性基准。
链接: https://arxiv.org/abs/2607.11707
作者: Ahmed Omar Salim Adnan,Yogananda Manjunath,Shivanjali Khare
机构: University of New Haven (新罕布什尔大学)
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Human-Computer Interaction (cs.HC)
备注:
Abstract:Following the rapid progress of generative Artificial Intelligence, there is a growing threat posed by conversational scams. These scams often span over multiple weeks or months, gradually build trust and request for money or sensitive information. Existing scam-detection systems mainly focus on isolated messages, which renders them inadequate against this evolving threat. This paper extends single-message phishing detection and presents an explainable agentic system for detecting sophisticated conversational scams. It also introduces ConScamBench-278, an initial public multi-category benchmark for conversational scam detection spanning eight scam types, released to support reproducible evaluation and future expansion. On isolated messages the single-message detector attains 100% phishing recall, while the conversation-level detector identifies all conversational scams in the public LoveFraud02 corpus (83/83) and reaches 97.8% accuracy (95% CI [95.4, 99.0]) on ConScamBench-278. Two user studies (N = 100 and N = 45) further motivate the system: participants report frequently experiencing uncertainty when judging suspicious conversations. In an uncontrolled pre/post comparison, users self-reported trust, self-confidence, and perceived need for AI-based scam detection all increased (p 0.001, Wilcoxon signed-rank). The system also receives a System Usability Scale score of 74.7 (95% CI [72.5, 76.9]), above the established usability benchmark.
[MA-6] StructAgent : Harness Long-horizon Digital Agents with Unified Causal Structure
【速读】：该论文旨在解决大语言模型（LLM）与视觉-语言模型（VLM）驱动的数字代理在执行长时程计算机任务时，因缺乏对任务进展的显式结构化表示而导致的任务状态难以解释、验证与恢复的问题。现有方法依赖原始交互历史进行决策，致使任务过程冗杂且不可靠。其解决方案的关键在于提出一种以状态为中心的框架——StructAgent，通过引入统一的因果性任务进展表征，实现对任务状态和工作流的显式结构化管理。该框架的核心创新包括：1）构建紧凑且可验证的任务状态，确保所有进展更新均基于验证；2）采用验证器支持的状态转移机制，规范任务推进流程；3）支持显式的进度检查点、基于证据的任务完成判定、针对性失败恢复及工具辅助执行。实验表明，StructAgent显著提升了多种LLM与VLM模型在长时程任务上的表现，在OSWorld-Verified基准上将Qwen3.5-9B的成功率从27.0%提升至46.9%，Qwen3.5-27B从31.6%提升至62.2%，并以78.9%的准确率创下开源模型新纪录；同时，该框架在Minecraft环境中的泛化能力进一步验证了其设计的通用性。
链接: https://arxiv.org/abs/2607.11388
作者: Wenyi Wu,Sibo Zhu,Kun Zhou,Aayush Salvi,Zixuan Song,Biwei Huang
机构: University of California, San Diego (加州大学圣地亚哥分校); Aether AI Lab (Aether AI 实验室)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG); Multiagent Systems (cs.MA)
备注:
Abstract:Recent advances in large language models (LLMs) and vision-language models (VLMs) have enabled increasingly capable digital agents for computer use. However, real-world tasks are often long-horizon and involve evolving contexts containing accumulated observations, intermediate edits, failed attempts, and partially completed executions. Existing agents typically operate over raw interaction history, making task progress difficult to interpret, verify, and recover, which ultimately limits reliable long-horizon execution. In this paper, we argue that addressing this challenge requires explicitly structuring both the agent’s state and workflow around a unified causal representation of task progress. We present \textbfStructAgent, a state-centered framework that introduces a unified state for maintaining compact, verifiable task progress and a structured workflow that regulates progress through verifier-backed state transitions. Building on this design, StructAgent further enables explicit progress checkpointing, evidence-driven task completion, targeted failure recovery, and tool-supported execution, while ensuring that all progress updates remain grounded in verification. Extensive experiments demonstrate that StructAgent consistently improves a wide range of LLM and VLM backbones on long-horizon computer-use tasks. On OSWorld-Verified, it improves Qwen3.5-9B from 27.0% to 46.9% success rate and Qwen3.5-27B from 31.6% to 62.2%, while achieving a new open-source state of the art of 78.9% with MiniMax-M3. Moreover, the same framework generalizes beyond desktop environments to Minecraft, demonstrating the generality of our design.
[MA-7] Mako: A Self-Evolving Agent ic Operating System (SE-AOS) for Autonomous Web Exploitation
【速读】：该论文旨在解决自动化网络安全攻防中“漏洞利用能力”难以持续演化与自适应的问题，即现有安全代理在面对复杂、动态变化的攻击面时，缺乏自主发现、验证并集成新漏洞利用能力的能力。其核心挑战在于如何构建一个能够自我演进、在不依赖人工干预的情况下持续提升自身攻击效能的智能系统。解决方案的关键在于提出一种新型的生成式安全代理架构——自演化智能体操作系统（Self-Evolving Agentic Operating System, SE-AOS），将漏洞利用能力视为可版本化、可动态扩展的内核模块，支持运行时观察自身失败、合成新能力、在真实目标上进行形式化验证，并热加载回自身。该系统以Mako为首个实例，作为LaunchSafe平台的核心引擎，在公开的XBOW基准测试中实现了对104个容器化CTF风格Web应用的全栈覆盖，成功触发每个目标生成唯一且加密新鲜的标志（flag），且结果无法通过预存或伪造获得。研究揭示了一个关键规律：一旦某项利用能力存在且可被发现，其攻击难度即趋于坍缩；真正稀缺的是能力本身，而非推理过程。为此，系统设计了受控的自演化闭环机制，通过沙箱验证与性能评估，仅在不退化的前提下提交对自身规则与代理的改进，从而形成可自我增强的安全研究体系。由于该系统将全谱网络攻击流程转化为可重复、机器级速度执行的管道，具有双重用途风险，因此作者选择公开科学原理，但保留实际攻击载荷、链路及工具源码。
链接: https://arxiv.org/abs/2607.11288
作者: Praneeth Narisetty,Shiva Nagendra Babu Kore
机构: 未知
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注: 13 pages, 10 figures, 8 tables
Abstract:We introduce the Self-Evolving Agentic Operating System (SE-AOS): a new class of AI agent that treats exploit capability as a mutable, versioned kernel it extends at runtime, observing its own failures, synthesising new capabilities, proving them against a live target, and hot-loading them back into itself. Mako is the first SE-AOS instance for security research and the autonomous web exploitation engine developed within LaunchSafe. LaunchSafe builds autonomous security agents for continuous offensive testing and agent-driven security research; Mako is the core engine behind that platform. On the public XBOW validation-benchmarks, 104 containerised, CTF-style web applications spanning 26 vulnerability classes across three difficulty tiers, Mako achieves full-suite coverage: it drives every one of the 104 targets to emit a cryptographically fresh, per-build flag, under a verification regime that makes fabricated or memorised results impossible. Our central result is a law of autonomous exploitation: once a capability exists and is discoverable, difficulty collapses; capability, not reasoning, is what is scarce, together with an architecture and formalism that turn that law into a self-improving system. Mako further runs a gated self-evolution loop that proposes, sandboxes, and commits improvements to its own agents and rules when fitness does not regress. We deliberately withhold the operational results, payloads, exploit chains, and tool source, because a system that reduces full-spectrum web exploitation to a repeatable, machine-speed pipeline is dual-use research of concern. We publish the science; we withhold the weapon.
[MA-8] Automated Textbook Auditing with Multi-Agent LLM Systems
【速读】：该论文旨在解决教育类教材在内容质量评估中面临的多维度挑战，即如何同时实现事实准确性、领域特定技术正确性及语言质量的自动化检测，而传统通用语法检查工具无法满足此类复杂需求。其解决方案的关键在于提出一种模块化多智能体（multi-agent）流水线系统——AI Textbook Auditor，通过两个并行分析路径实现高效精准的质量保障：一是“事实与技术检测路径”，由一组专业化大语言模型（LLM）智能体协同工作，识别事实错误、代码缺陷、定义偏差及概念不一致等问题，并在人文学科领域结合网络搜索增强验证能力；二是“语法检测路径”，基于原生PDF处理以保留变音符号等排版细节。此外，引入“裁判智能体”（Judge Agent）利用领域特定规则过滤误报，最终生成结构化、可人工审阅的报告。系统支持视觉原生页面渲染与文本提取双模式输入，并可通过定制提示词（prompt）适配不同学科领域的错误分类体系，具备良好的可扩展性。实验表明，该系统在罗马尼亚高中阶段的计算机科学与历史社会科学教材上分别发现了56项与72项问题，经专家验证准确率可达62.5%，整体设计定位为一种辅助人工审查的分诊工具，显著降低人工排查成本，但需依赖专家最终确认后方可实施编辑修正。
链接: https://arxiv.org/abs/2607.11276
作者: Ciprian Cristescu,Adrian-Marius Dumitran,Angela-Liliana Dumitran,Gabriel Stefan
机构: University of Bucharest (布加勒斯特大学); Dimitrie Cantemir Christian University (迪米特里·坎特米尔基督教大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Multiagent Systems (cs.MA)
备注: Presented @ iTextbooks 2026: 7th Workshop on Intelligent Textbooks at AIED’2026
Abstract:Ensuring the quality of educational materials requires more than standard proofreading: textbooks must be audited for factual accuracy, domain-specific technical correctness, and linguistic quality simultaneously – a task that general-purpose grammar checkers cannot address. We present \textbfAI Textbook Auditor, a modular multi-agent pipeline for automated quality assurance of educational materials across subject domains. The system accepts a textbook PDF and produces a structured, human-reviewable report via two analysis tracks: a \textbfFactual and Technical Track in which an ensemble of specialized LLM agents detects factual inaccuracies, code errors, incorrect definitions, and conceptual inconsistencies, augmented with web search for humanities domains; and a \textbfGrammar Track operating PDF-natively to preserve diacritical encoding. A \textbfJudge Agent filters false positives using domain-specific rules before presenting findings to a human reviewer. The pipeline supports two ingestion modes – vision-native page rendering and PyMuPDF text extraction – and is domain-adaptable via custom prompts encoding subject-specific error taxonomies. We demonstrate the system on two Romanian upper-secondary textbooks: a CS textbook (56 technical findings across seven categories, with an expert-validated precision of 62.5%) and a history and social sciences textbook (72 findings spanning factual errors, ideological bias, and grammar). The system is designed as a triage tool that reduces the manual effort of locating candidate issues, with human expert validation required before any editorial action.
[MA-9] Multi-Agent LLM s Fail to Explore Each Other
【速读】：该论文旨在解决多智能体系统中大语言模型（LLM）智能体在相互交互时探索能力不足的问题。当前的LLM智能体在协作过程中常表现出短视且极化的交互模式，导致协调效果不佳并产生较高的后悔值（regret）。为此，论文将此挑战形式化为多智能体探索问题（Multi-Agent Exploration），将其建模为部分可观测随机博弈（Partially Observable Stochastic Game, POSG）框架，其中智能体需通过主动探测同伴以推断其能力并识别有效的交互策略。解决方案的关键在于提出一种轻量级框架——多智能体情境探索（Multi-Agent Contextual Exploration, MACE），该框架通过结构化的同伴选择机制显式促进探索行为。实验结果表明，MACE在情境与参数多样性设置下均显著提升了探索效率和下游任务性能；理论分析进一步证明，探索价值随智能体多样性增加而提升。研究揭示了现有LLM智能体的根本局限性，并强调了在实现可靠多智能体自主性中，显式引导探索的重要性。
链接: https://arxiv.org/abs/2607.11250
作者: Hyeong Kyu Choi,Jiatong Li,Wendi Li,Xin Eric Wang,Sharon Li
机构: 未知
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI)
备注:
Abstract:Exploration is essential for reliable autonomy in multi-agent systems, yet it remains unclear whether large language model (LLM) agents can explore effectively when interacting with one another. We show that modern LLM agents fail to do so, often exhibiting myopic and polarized interaction patterns that lead to suboptimal coordination and increased regret. We formalize this challenge as the Multi-Agent Exploration problem, modeling it as a partially observable stochastic game (POSG) problem in which agents must probe peers to infer their capabilities and identify effective interaction strategies. To address this, we introduce Multi- Agent Contextual Exploration (MACE), a lightweight framework that explicitly promotes exploration through structured peer selection. Across both contextual and parametric diversity settings, MACE substantially improves exploration behavior and downstream task performance. We further show theoretically that the value of exploration increases with agent diversity. Overall, our results highlight a fundamental limitation of current LLM agents and underscore the importance of explicitly guided exploration for reliable multi-agent autonomy. Code will be released in this https URL
[MA-10] Auditing Belief-Conditioned LLM Agents in Hidden-Information Social Deduction Games
【速读】：该论文旨在解决在隐藏信息的多智能体环境中评估大语言模型（LLM）智能体时面临的挑战：最终结果具有高方差，且难以揭示智能体决策背后的因果机制。为此，研究构建了一个可审计的框架，通过维护对隐藏角色的外部信念状态（external belief state），记录信念更新与信念-行动偏差作为结构化证据，并支持一种防御性离线改进循环，以在策略变更前分析失败案例。实验基于9人狼人杀环境，采用严格代码级信息隔离设置，在包含1,080场冻结游戏的多种条件（如信念禁用、主动信念、内核消融、阵营限制、消耗策略及高负载等）下验证，结果显示主动信念条件显著提升了好人阵营的胜率（在200种子的A0/A1对比中，胜率从0.205提升至0.390，配对McNemar检验χ²=16.4，p<0.001），并减少了不可逆的女巫毒杀错误。然而，研究并未将此提升归因于信念内容本身——直接行动-信念一致性较低（≈0.21），且仅向狼人提供信念反而比仅向好人提供更有利于好人阵营，这否定了简单的“持有者受益”解释。因此，研究将该效应视为关联性结果，其内在机制仍待澄清。核心贡献在于提出的审计框架本身：它使复杂行为可测量、暴露了低直接一致性、以证据驳回了不可靠的强制消耗干预，并有效分离了策略效应与负载混杂因素。由此，研究主张在高噪声的隐藏信息博弈中，外部信念应作为可审计的认知基线，兼具决策相关信号，将原本黑箱的智能体行为转化为可回放、可追溯的证据，从而实现更安全、可控的迭代优化。
链接: https://arxiv.org/abs/2607.10814
作者: Yuan Gao,Jiangyi Yang,Yao Zhao,Yichi Zhang
机构: University of Melbourne(墨尔本大学)
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI)
备注: 29 pages, 8 figures, 3 tables. Preprint
Abstract:Evaluating LLM agents in hidden-information multi-agent settings is hard: final outcomes are high-variance and rarely reveal why an agent decided as it did. We study this in a 9-player Werewolf environment where agents act under strict, code-level information isolation, and we build an auditable framework that maintains an external belief state over hidden roles, logs belief updates and belief-action deviations as structured evidence, and supports a defensive offline improvement loop that reviews bad cases before any strategy change. Across 1,080 frozen games spanning belief-disabled, active-belief, kernel-ablation, camp-restricted, consumption-policy, and high-load arms, and including a seed-paired A0/A1 comparison, the active-belief condition is associated with substantially better good-side outcomes: in the 200-seed A0/A1 comparison the good-side win rate rises from 0.205 to 0.390 (paired McNemar \chi^2 = 16.4 , p 0.001 ), with fewer irreversible witch-poison errors. We do not, however, attribute this shift to belief content. Direct action-belief consistency is low ( \approx 0.21 ), and giving belief only to the werewolves helps the good side more than giving it only to the good side, which argues against a simple holder-benefit account; we therefore report the effect as an association and treat its mechanism as unresolved. The contribution is the audit framework itself: it makes the effect measurable, exposes low direct action-belief consistency, rejects an unreliable forced-consumption intervention with evidence, and separates strategy effects from load confounds. We accordingly position external belief in high-noise hidden-information games primarily as an auditable cognitive baseline that also carries decision-relevant signal, turning opaque agent behavior into replayable evidence for safer, controlled iteration. Comments: 29 pages, 8 figures, 3 tables. Preprint Subjects: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI) Cite as: arXiv:2607.10814 [cs.MA] (or arXiv:2607.10814v1 [cs.MA] for this version) https://doi.org/10.48550/arXiv.2607.10814 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[MA-11] Distributed Agent System: Fault-Tolerant Collaboration Among Embodied Agents
【速读】：该论文旨在解决生成式AI（Generative AI）在长时序任务中因资源约束与环境不确定性导致的累积误差传播问题，传统基于错误消除的优化策略已无法有效应对此类挑战。其核心解决方案是提出分布式智能体系统（Distributed Agent System, DAS），构建一种设备-边缘-云协同的异构智能体容错协作框架。关键创新在于重新定义智能体可靠性为系统级容错能力，而非单次交互的零误差精度，并设计了双层容错架构：第一层通过容错对齐机制保障单个智能体执行的可靠性，第二层则利用半形式化语言协议确保跨智能体通信的可靠性。该框架为工业场景下异构具身智能体的可靠协同提供了可落地的工程路径。
链接: https://arxiv.org/abs/2607.10811
作者: Kai Yu,Lu Chen,Hanqi Li
机构: 未知
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI)
备注:
Abstract:AI engineering is shifting from passive text generation by large language models (LLMs) to agent-driven task execution, creating new reliability challenges for long-horizon tasks under resource constraints and environmental uncertainty. Conventional error-elimination optimization strategies fail to address cumulative error propagation. This paper proposes Distributed Agent System (DAS), a device-edge-cloud framework for fault-tolerant collaboration among heterogeneous agents. We redefine agent reliability as system-level fault tolerance rather than single-turn zero-error accuracy, and present a two-layer fault-tolerance architecture: single-agent execution reliability via fault-tolerant alignment, and cross-agent communication reliability via semi-formal language protocols. This framework provides a practical engineering pathway for reliable heterogeneous embodied agents collaboration in industrial scenarios.
[MA-12] WattCouncil: Context-Aware Household Energy Scenario Generation With Governed LLM s
【速读】：该论文旨在解决低碳电力系统转型背景下，配电网在整合屋顶光伏、电动汽车等分布式能源技术时所面临的新型运行与分析挑战，尤其针对智能电网研究中机器学习（Machine Learning, ML）模型因隐私顾虑、监管限制及数据采集成本高等原因导致的高分辨率家庭用电数据获取困难问题。其解决方案的关键在于提出WattCouncil框架——一个基于大型语言模型（Large Language Model, LLM）的多智能体协同数据生成系统，通过一组具备特定角色的智能体（如生成、审计、验证）在明确的文化、时间与物理约束下，动态生成结构化的家庭用电情景。这些智能体并非静态预测工具，而是作为受控流程中的自适应决策主体，依托引导式推理机制融合家庭构成、时间维度特征及环境条件，生成具有上下文敏感性的日度用电行为模式。实验基于包含4232户超过一年负荷测量数据及社会经济调查信息的CER数据集进行评估，并通过消融研究验证框架的一致性与有效性，显著提升了高保真家庭用电数据的可生成性与可信度。
链接: https://arxiv.org/abs/2607.10720
作者: Mohannad Takrouri,Nicolas M. Cuadrado A.,Martin Takáč
机构: Mohamed bin Zayed University of Artificial Intelligence (穆罕默德·本·扎耶德人工智能大学); Abu Dhabi (阿布扎比)
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Multiagent Systems (cs.MA)
备注: 11 pages, 5 figures
Abstract:The accelerating shift toward low-carbon power systems, together with the widespread adoption of behind-the-meter technologies such as rooftop solar and electric vehicles, is placing new operational and analytical demands on electricity grids. At the same time, smart-grid research increasingly relies on machine learning (ML), yet progress is constrained by limited access to high-resolution household energy data due to privacy concerns, regulatory barriers, and collection costs. This work presents WattCouncil, a data-generation framework in which household electricity demand is generated by a council of Large Language Model (LLM)-based agents operating in specialized roles to generate, audit, and validate structured energy scenarios under explicit cultural, temporal, and physical constraints. Rather than acting as static predictors, these agents serve as adaptive decision-makers within a governed pipeline. Motivated by studies highlighting the importance of contextual factors in energy use, our framework produces context-sensitive daily routines through a guided reasoning process that incorporates household composition, temporal factors, and environmental conditions. We evaluate the generated profiles against the detailed CER dataset, which contains over a year of load measurements for 4232 households together with survey-based socio-economic information. We further assess the consistency of the framework through ablation studies. Source code is available at this https URL
[MA-13] MafiaScope: Non-Invasive Time-Resolved Belief Probing for LLM Agents in Social Deduction Games
【速读】：该论文旨在解决大语言模型（Large Language Model, LLM）在社会推理能力评估中“行为可见性不足”的核心问题：传统公开行为（如发言、投票）无法反映模型真实信念，导致其社会认知（如欺骗、共情、意图推断）能力难以被有效测量。为突破这一局限，作者提出MafiaScope——一个基于社交推理游戏“狼人杀”（Mafia）的开放测试平台，将社会推断任务转化为可量化的心理建模实验。其解决方案的关键在于引入结构化探针问题（structured probe questions）机制：在每轮公开发言后，各代理（agent）需在私密状态下回答一组预设的、可配置的信念相关问题（如“某玩家是否为狼人？”、“自己被怀疑的概率？”），这些问题的答案不参与游戏逻辑，仅用于评估模型内部信念状态与真实情境的匹配度。系统通过自动评分实现对信念轨迹的精准追踪，并结合交互式可视化工具（包括角色视角模拟、时间线准确率/校准度面板、反事实回放分支等），实现对模型理论心智（Theory of Mind）的动态、可解释评估。在对DeepSeek模型进行的32局实验中，结果显示模型置信度严重失准（预期校准误差达0.17），过度预测被怀疑概率达1.5倍，验证了该框架在揭示模型内在信念偏差方面的有效性。整个引擎、可视化工具及超过200场跨模型对战数据集均开源，支持可复现研究。
链接: https://arxiv.org/abs/2607.10645
作者: Ilia Karpov
机构: HSE University (高等经济大学)
类目: Computation and Language (cs.CL); Multiagent Systems (cs.MA); Social and Information Networks (cs.SI)
备注:
Abstract:An LLM agent’s public behaviour reveals little about its social reasoning: an agent that votes correctly may be guessing, and an agent that lies well leaves no trace of what it actually believes. We present MafiaScope, an open testbed that turns the social deduction game Mafia into a measurement instrument for machine Theory of Mind. After every public utterance, every agent privately answers a configurable set of structured probe questions; the answers never re-enter the game and are scored automatically against the ground truth the engine knows. An interactive visualizer renders the belief trajectories: impersonate mode shows the game as one agent sees it, panels chart timeline-aligned accuracy and calibration, and counterfactual replay forks any recorded step. In a 32-game DeepSeek case study with 13,815 parsed probe answers, stated confidence is poorly calibrated, with expected calibration error 0.17, agents over-predict being suspected 1.5 times, and a 30-fork replay experiment walks the counterfactual replay workflow end to end. Engine, viewer and a corpus of 200+ cross-model games are released under an open licence; live demo: this https URL screencast: this https URL.
[MA-14] Can Agent ic Trading Systems Pay for Their Own Intelligence?
【速读】：该论文旨在解决当前大型语言模型（Large Language Model, LLM）代理在交易系统中应用时，缺乏对“智能可变现性”（agentic viability）的评估问题。现有方法仅关注性能指标，却未检验动态决策所引发的成本是否真正转化为可测量的增量收益。为此，论文提出TradeLens——一个基于交易记录、运行时轨迹与部署配置的可追溯诊断工具包，通过重构交易路径、将利润与成本归因于可解释的证据，实现对代理是否为其自身智能付费的诊断。其解决方案的关键在于建立“智能到利润转化”的评估范式：通过追踪决策过程中的时间价值与成本分配，识别出不同模型在资产选择（如DeepSeek-V3.2）或交易时机（如GLM-4.7）上的失败模式，并揭示资本规模、交易频率与系统架构仅通过放大或削弱决策所带来的时间价值而影响整体可行性。这一方法将LLM交易代理的评估从以能力为中心的性能排名，转向基于运行轨迹的智能—利润转化诊断，为可信、可解释的智能交易系统设计提供了新范式。
链接: https://arxiv.org/abs/2607.10286
作者: Qiqi Duan,Changlun Li,Chen Wang,Fan Zhang,Mengxiang Wang,Dayi Miao,Peixian Ma,Jiangpeng Yan,Liyuan Chen,Shuoling Liu,Preslav Nakov,Yuyu Luo,Nan Tang
机构: HKUST(GZ); Paradoox AI; E Fund Management Co., Ltd; MBZUAI; The University of Tokyo
类目: Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注:
Abstract:Large language model (LLM) agents are increasingly used in trading systems, where model reasoning, tool use, and continual decisions incur costs that are expected to produce trading value. Existing evaluations typically report performance metrics, but rarely examine agentic viability: whether dynamic LLM-mediated decisions convert their induced costs into measurable incremental profit. To apply this criterion, we introduce TradeLens, a trace-grounded diagnostic toolkit for evaluating agentic trading systems from their trading records, runtime traces, and deployment configurations. It reconstructs trading trajectories, attributes profit and cost to interpretable evidence, and diagnoses whether and why an agent pays for its own intelligence. We conduct extensive analysis across backbone models, capital scales, trading frequencies, and system architectures, together with deployment discussion. Our results show that viability hinges on intelligence-to-profit conversion: models exhibit different failure patterns, such as poor asset selection in DeepSeek-V3.2 and negative timing in GLM-4.7, while capital scale, trading frequency, and architecture matter only by amplifying or degrading decision-attributed timing value. These findings reframe the evaluation of LLM-based trading agents from capability-centric performance ranking to trace-grounded diagnosis of intelligence-to-profit conversion. Our code is available at this https URL.
[MA-15] RideGym: A Standardized Interface for Real-World Large-Scale Ride-Sharing System
【速读】：该论文旨在解决当前共享出行系统中基于多智能体强化学习（MARL）的订单调度研究缺乏标准化、可复现仿真平台的问题。现有仿真环境通常针对特定运营场景或紧耦合于特定调度算法，导致研究者需重复构建自定义环境，严重影响实验的可比性与可复现性，并造成大量冗余开发工作。为此，本文提出RideGym，首个面向真实城市规模共享出行系统的开源、标准化Gym风格接口，其核心创新在于彻底解耦环境与调度算法，支持在统一、明确的条件下对各类基于学习和模型的方法进行开发与公平比较。RideGym基于真实路网实现高效的大规模城市级仿真，具备灵活配置车辆属性、订单特征及自动最短路径路由的能力。实验验证表明，该平台具备极高运行效率（一小时仿真含数千车辆与数万订单可在一分钟内完成），并揭示探索噪声的选择对MARL方案性能及其相对排名具有显著影响，这一关键因素在以往研究中常被忽视。
链接: https://arxiv.org/abs/2607.10173
作者: Zijian Zhao,Yulong Hu,Sen Li
机构: The Hong Kong University of Science and Technology(香港科技大学)
类目: Multiagent Systems (cs.MA)
备注:
Abstract:Ride-sharing has become an essential component of modern urban transportation and has attracted significant attention across computer science, transportation, and management science. While the field spans a broad range of problems, such as driver relocation, dynamic pricing, and vehicle charging or fueling dispatch, the core challenge remains order assignment and trip bundling, which directly affect urban traffic efficiency and carbon emissions. Despite its importance, existing simulation platforms are typically tailored to specific operational studies or tightly coupled to a particular dispatch algorithm, and rarely expose a standardized, learning-friendly interface. As a result, most researchers still build customized environments from scratch, raising serious concerns about reproducibility and fair comparison, and incurring substantial redundant effort. To address this gap, we present RideGym, the first open-source, standardized Gym-style interface tailored to MARL-based order dispatch in real-world ride-sharing systems. By fully decoupling the environment from the dispatch algorithm, RideGym enables diverse learning-based and model-based methods to be developed and compared under identical, fully specified conditions. It supports efficient, large-scale city-level simulations on real road networks, and offers flexible configurations for vehicle attributes, order specifications, and automatic shortest-path routing. We validate RideGym by reproducing several baselines, and demonstrate its high efficiency, with a one-hour simulation involving thousands of vehicles and tens of thousands of orders completed within one minute across all methods. Moreover, we reveal that the choice of exploration noise can significantly affect both the performance and the relative ranking of MARL solutions, an aspect often overlooked in prior work.
[MA-16] Runtime Safety Filtering for Learned Small UAS Separation Policies under GNSS Degradation
【速读】：该论文旨在解决在城市环境中因全球导航卫星系统（GNSS）信号质量下降导致的感知与定位信息不准确问题，进而影响小型无人机系统（sUAS）学习型分离保障策略的安全性。在GNSS退化场景下，传统方法依赖于精确的位姿信息，而实际中存在多径效应、信号遮挡及有意干扰等问题，使得该假设失效。因此，论文提出的核心问题是：在运行时，应通过何种机制保障安全——是过滤策略输出的动作（action filtering），还是过滤观测输入（observation filtering）？其解决方案的关键在于对两种架构进行对比评估：前者通过在最差情况交通状态上施加离散时间控制屏障函数（Control Barrier Functions, CBF）来约束策略输出；后者则将基于边界观测不确定性推断出的最差情况状态作为修正后的输入直接提供给策略。实验结果表明，动作过滤对安全性提升微乎其微，而观测过滤可使近似空中碰撞事件减少90%，且对屏障函数在分离距离与接近速率之间的权衡具有更强鲁棒性。研究结论表明，对于具备学习型安全行为的策略，保持其决策自主性优于以人工设计的约束强行干预其动作输出。
链接: https://arxiv.org/abs/2607.10014
作者: Alex Zongo,Peng Wei
机构: 未知
类目: Robotics (cs.RO); Machine Learning (cs.LG); Multiagent Systems (cs.MA); Systems and Control (eess.SY)
备注: Accepted for publication at the 2026 IEEE/AIAA Digital Avionics Systems Conference (DASC). 9 pages, 8 figures
Abstract:Learning-based separation assurance for small Unmanned Aircraft Systems (sUAS) achieves near-zero collision rates in simulation, but assumes accurate position and velocity information from Global Navigation Satellite Systems (GNSS). This assumption fails in urban environments, where multipath propagation, signal blockage, and intentional interference degrade navigation integrity. This raises a fundamental architectural question for deploying learned separation policies under GNSS degradation: should runtime safety mechanisms filter the policy’s actions or its observations? This work evaluates both approaches for multi-agent sUAS separation under adversarial GNSS degradation. Both architectures first estimate a worst-case traffic state consistent with bounded observation uncertainty, then diverge: action filtering constrains policy outputs via discrete-time control barrier functions evaluated at the worst-case state, while observation filtering presents the worst-case state directly to the policy as corrected input. Experimental results show that action filtering provides negligible safety improvement, while observation filtering reduces near mid-air collisions by 90% and remains robust to the barrier function’s tradeoff between separation distance and closing rate. These results suggest that, for policies with learned safety behaviors, preserving the policy’s decision authority outperforms overriding its actions with hand-designed constraints.
[MA-17] WhoWhen Pro: Can LLM s Really Attribute Failures in AI Agents ?
【速读】：该论文旨在解决智能体系统（agentic systems）在日益复杂和自主化背景下，其失败模式愈发隐蔽、难以人工识别的问题，提出通过大语言模型（LLM）实现自动化故障归因（automated failure attribution）。其核心挑战在于如何精准定位故障发生的位置（where）与根本原因（why），尤其是在高复杂度任务中。解决方案的关键在于构建了大规模、严格控制的基准测试集——WhoWhen Pro，该数据集通过在精确重放成功前缀后引入单一故障，生成了12,326条带有黄金标签（golden labels）的失败轨迹，覆盖三种模态（modalities）与26个基准场景，确保了故障注入的可复现性与标注准确性。此外，研究通过系统性实验揭示了不同模态、协议及模型家族在故障归因中的共性模式，为未来自动化故障归因系统的设计提供了实证依据与优化方向。
链接: https://arxiv.org/abs/2607.09996
作者: Jiale Liu,Huajun Xi,Shaokun Zhang,Yifan Zeng,Tianwei Yue,Chi Wang,Jian Kang,Qingyun Wu,Huazheng Wang
机构: 未知
类目: Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注:
Abstract:Automated failure attribution uses LLMs to identify where and why agentic systems fail. As agents become more capable, their failures become subtler, making automated attribution increasingly important. We introduce WhoWhen Pro, a large-scale benchmark for automated failure attribution in agentic systems. Using a strictly controlled pipeline that injects a failure only after exactly replaying a successful prefix, we construct 12,326 failed trajectories with golden labels across 3 modalities and 26 benchmarks covering various scenarios. Beyond benchmarking, we conduct extensive experiments and analyses, revealing systematic patterns in how models attribute failures across modalities, protocols, and model families, and providing empirical guidance for future automated failure attribution systems.
[MA-18] Beyond Bayesian Nash: Learning Minimax-Regret Equilibria for Adversarial Team Games under Asymmetric Information
【速读】：该论文旨在解决具有非对称信息的对抗性团队博弈（Adversarial Team Games, ATGs）中策略鲁棒性不足的问题，特别是在对手类型隐藏（如隐藏目标旗标）和存在策略性欺骗的情况下。现有风险中性解概念（如贝叶斯纳什均衡，BNE）对类型分布的微小扰动敏感，而现有的分布鲁棒方法仅在预设的模糊集（ambiguity set）内提供保障，无法有效应对对手通过战略性重新分配概率质量进行的欺骗行为。为此，论文提出了一种新的均衡概念——概率鲁棒最小最大后悔均衡（Probabilistically Robust Minimax-Regret Equilibrium, PR-MRE），其核心在于将最小最大后悔推理的分布无关鲁棒性与名义类型分布的统计信息相结合：在高置信度类型子空间上最小化最坏情况下的后悔值，从而在抵御策略性分布转移的同时避免完全分布无关方法带来的过度保守问题。研究证明，对于标准形式的贝叶斯博弈，PR-MRE可建模为一个鲁棒双线性规划，并推导出一种可计算的半定松弛形式；进一步将其嵌入到鲁棒双轨奥尔（double-oracle）框架中，构建了名为PRMRE-PSRO的新型元求解器，支持基于深度强化学习最优响应的群体学习机制，实现近似PR-MRE策略的高效求解。在图结构对抗性团队博弈上的实验表明，相较于风险中性均衡解，PR-MRE所发现的策略在面对隐藏类型时具有显著更优的最坏情况表现，展现出更强的对抗战略性分布偏移的能力。
链接: https://arxiv.org/abs/2607.09993
作者: Naman Aggarwal,Jonathan P. How
机构: Massachusetts Institute of Technology (麻省理工学院)
类目: Computer Science and Game Theory (cs.GT); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Multiagent Systems (cs.MA)
备注: 29 pages, 11 figures, 6 tables. Submitted to Transactions on Machine Learning Research (TMLR)
Abstract:Adversarial team games (ATGs) with asymmetric information, such as adversarial path-finding, goal search, and reachability games on graphs, require strategies that are robust to hidden opponent types, such as a hidden goal flag, and to deception. Under asymmetric information, deception is seen as strategic shifts in the type distribution such that the omniscient opponent can collude with Nature and condition its play on the observed type. Existing risk-neutral solution concepts, such as Bayesian Nash equilibrium (BNE), are sensitive to distribution shifts, while distributionally robust approaches provide guarantees only within a prescribed ambiguity set. To address these limitations, we introduce Probabilistically Robust Minimax-Regret Equilibrium (PR-MRE), a novel equilibrium concept that combines the distribution-free robustness of minimax-regret reasoning with probabilistic information from a nominal type distribution. PR-MRE minimizes worst-case regret over a high-confidence subset of the type space, providing protection against strategic redistribution of probability mass while avoiding the conservatism of fully distribution-free approaches. We show that, for normal-form Bayesian games, PR-MRE can be formulated as a robust bilinear program and derive a tractable semidefinite relaxation. We then adapt this relaxation into a novel meta-solver within a robust double-oracle framework, PRMRE-PSRO, enabling population-based learning of approximate PR-MRE strategies via deep reinforcement learning best responses. Experiments on graph-structured adversarial team games demonstrate that PR-MRE discovers strategies with substantially improved worst-case performance across hidden types compared to risk-neutral equilibrium solutions, resulting in more robust behavior under strategic distribution shifts.
[MA-19] A Knowledge-Based Multi-Agent Framework for Security Control Recommendation
【速读】：该论文旨在解决企业在缺乏充足网络安全专业人才的情况下，难以有效加固本地部署（on-premises）IT环境的安全挑战。其核心问题在于如何在有限资源条件下，高效、精准地推荐合适的安全控制子族（security control sub-families），以实现对多个安全维度的全面覆盖，同时避免安全资源的过度或不足配置。解决方案的关键在于构建一个基于多智能体影响图（Multi-Agent Influence Diagram, MAID）模型的非零和、同步决策支持系统（Security DSS），将安全决策建模为7个安全维度（即代理）之间的博弈过程，并采用无悔在线学习（no-regret online learning）算法在决策空间中进行高效探索。该方法通过整合来自知名信息安全与学术来源的经过筛选的统一数据集，实现了在极小比例的可实施安全控制（如仅需29%~65%）下，达成高达73%~99%的安全需求覆盖率，且计算耗时仅为0.8至35.7秒，显著提升了安全资源配置的效率与准确性。
链接: https://arxiv.org/abs/2607.09954
作者: Carolina Fernández-Martínez,Shuaib Siddiqui,Vanesa Daza
机构: i2CAT Foundation( i2CAT基金会); Universitat Pompeu Fabra(庞佩乌·法布拉大学)
类目: Computer Science and Game Theory (cs.GT); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Machine Learning (cs.LG); Multiagent Systems (cs.MA)
备注: Elsevier Knowledge-Based Systems (KNOSYS), 2026
Abstract:Hardening IT on-premises environments can be a daunting task for teams without access to adequate cybersecurity expertise. In this regard, Decision Support Systems (DSS) with embedded expert knowledge can assist users by guiding them with security recommendations to meet their objectives. This work proposes a Security DSS that recommends security control sub-families given minimal user requirements indicating coverage of different security dimensions. It leverages a curated, unified dataset from both well-known Information Security (InfoSec) and academic sources. This DSS is defined as a non-zero-sum, simultaneous game that is grounded in a Multi-Agent Influence Diagram (MAID) model and explores the decision space over 7 security dimensions or agents, using no-regret online learning to ultimately find the security control sub-families that best fit the requirements while incurring minimal under- and over-provisioning of security resources. This work was validated in terms of performance and accuracy, among others, for varying dataset sizes. It shows exceptional satisfaction coverage results of 99% when using as little as ~65% of the SW-implementable security controls, running in 1.2-35.7 seconds; and more moderate coverage results of 73%-77% when using ~29% of the controls, resolving in 0.8-13.8 seconds.
[MA-20] Agent ic Context Learning with Self-Discovered Specification
【速读】：该论文旨在解决生成式人工智能（Generative AI）在推理阶段进行上下文学习（context learning）时表现不佳的核心问题，即大型语言模型（LLM）难以从预训练未涵盖的复杂上下文中有效获取并应用任务特定知识，导致当前前沿模型在该任务上的成功率不足24%。其关键发现是：传统假设中认为失败源于内容访问困难，并非根本原因；真正瓶颈在于模型对“局部规范”（local specifications）的获取能力不足——这些规范包括领域特定格式、局部规则及完整性条件等，虽未在用户查询中明确提及，却广泛分布于上下文之中。研究通过分析CL-Bench基准上31,592个评分项发现，55.4%的评估标准聚焦于规范获取，而仅22.6%关注内容获取，且76.7%的规范未在查询中出现，但95.5%可从上下文中追溯，表明其为可学习的显性义务。为此，作者提出一种轻量级干预方法PSCI（Private Specification-Contract Induction），通过提取局部规范并引入对抗性检查与修复机制加以强制执行，在GPT-5.1上实现28.14%的准确率（绝对提升5.59个百分点，相对提升24.8%），并在Qwen3.5-27B和Gemini 3 Pro上复现显著增益。17组消融实验进一步验证了任务特定规范在上下文学习中的决定性作用。综上，该研究揭示上下文学习的成功不仅依赖内容获取，更关键在于对隐含规范的识别与遵循。
链接: https://arxiv.org/abs/2607.09794
作者: Jike Zhong,Ming Li,Yuxiang Lai,Ziyan Yang,Jingyu Xie,Jihyung Kil,Zheda Mai,Shao-Yuan Lo,Ren Xiang,Konstantinos Psounis,Yuanyuan Lei
机构: University of Southern California (南加州大学); University of Florida (佛罗里达大学); Emory University (埃默里大学); Adobe Research (Adobe 研究院); The Ohio State University (俄亥俄州立大学); National Taiwan University (台湾大学)
类目: Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注:
Abstract:Context learning is an emerging inference-time task where LLMs must learn and apply novel, task-specific knowledge from intricate contexts absent from pre-training; even frontier models score under 24% task success. In this work, we conduct a comprehensive empirical study to understand why this setting remains difficult. A natural hypothesis is that failures stem from content access; yet across twelve retrieval, reflection, and verification baselines on CL-Bench, an extensive context learning benchmark, we find limited gains over direct full-context prompting. Further failure analysis reveals a key finding: unlike typical long-context tasks such as long document understanding, context learning requires not only recovering local content but also acquiring local specifications that are often unspecified in the query but distributed across the context: domain-specific formats, local rules, and completeness conditions. Across all 31,592 rubric items, we find that 55.4% clearly evaluate specification acquisition, while only 22.6% evaluate content acquisition. Moreover, despite 76.7% of specifications being unspecified in the user query, 95.5% are traceable to the context, indicating these are learnable obligations rather than hidden requirements. To validate this diagnosis, we design a deliberately simple intervention PSCI (private specification-contract induction) which extracts local specifications and enforces them through adversarial checking and repair; PSCI achieves state-of-the-art 28.14% with GPT-5.1 (+5.59 pp absolute and +24.8% relative) on CL-Bench, replicated on Qwen3.5-27B (+5.28 pp) and Gemini 3 Pro (+6.17 pp). Seventeen ablations further isolate the role of task-specific specifications. Overall, our results suggest context learning hinges on not only content acquisition but also specification acquisition.
[MA-21] Verification of Adaptive Agent ic Controllers through Finite Rule Revision
【速读】：该论文旨在解决工业级自适应智能体（adaptive agentic AI）在从原型能力向生产部署转化过程中存在的可验证性难题，特别是在非确定性、保密性约束、上下文有限及可观测性弱等现实条件下，难以对智能体行为进行有效验证与故障诊断的问题。其核心解决方案是构建一个基于有限符号规则、显式诊断谓词、解释日志和保留测试集再评估的有界验证协议（bounded verification protocol）。关键在于将自适应智能体控制器视为可修订的有限对象，通过预定义的规则级修正操作（如规则增删、优先级调整）实现故障定位与局部修复，并在保留的仿真种子或克隆初始状态上进行独立再评估。实验表明，该框架能够识别资源引发的不可修复故障、因违反阈值或安全守则而被拒绝的部分修复，以及通过移除平滑规则实现的一次性局部修复。该方法论贡献在于提供了一套可仿真的程序化流程，使特定控制器层面的故障具备可观测性、可解释性、局部可修订性与可控条件下的实证重测能力。
链接: https://arxiv.org/abs/2607.09770
作者: Roberto Garrone
机构: Open University of Cyprus(塞浦路斯开放大学)
类目: Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA); Systems and Control (eess.SY)
备注: 28 pages, 3 figures, 8 tables
Abstract:Industrial agentic AI systems increasingly exhibit a gap between prototype capability and production deployment. In particular, adaptive agents may generate plausible outputs while remaining difficult to verify under non-determinism, confidentiality constraints, limited context, and weak observability. This paper formulates a bounded verification protocol for adaptive agentic controllers represented by finite symbolic rules, explicit diagnostic predicates, explanation logs, and held-out re-evaluation. The central research question is: when an adaptive agentic controller is represented through finite rules, explicit diagnostic predicates, explanation logs, and held-out re-evaluation, which classes of controller failure can be detected, locally repaired, or rejected without relying on unrestricted human-in-the-loop judgment? The proposed framework treats the controller as a finite revisable object. Diagnostic failures are mapped to predefined rule-level edits, including rule addition, rule deletion, and priority revision. Repaired controllers are then evaluated on held-out simulation seeds or cloned initial states. Experiments in a stylized financially constrained inventory-control benchmark show three outcomes: resource-induced failures that remain non-repairable by one rule edit, partial repairs that are rejected because they violate thresholds or guardrails, and a local one-step repair of an order-volatility failure induced by removing a smoothing rule. The contribution is methodological and provides a simulation-compatible procedure for testing whether specific controller-level failures can be made observable, explainable, locally revisable, and empirically re-tested under controlled conditions.
[MA-22] Norm Enforcement for AI Agents : Robustly Shaping Behavior in Multi-Agent Systems ICML2026
【速读】：该论文旨在解决多智能体系统中因个体竞争导致的集体性行为失范问题，尤其是在共享环境中，语言模型智能体在追求自身目标时可能产生损害整体利益的行为（如社交媒体营销代理发布误导性内容以获取流量）。现有解决方案依赖于人类社会中的规范（norm）及其执行机制来约束不当行为，但当前简单的规范执行机制易被对齐不良的智能体利用以获得竞争优势，即使这些智能体未被显式训练或提示去规避规则。为此，论文提出一种更鲁棒的规范执行机制，其关键在于两个核心设计原则：一是持续评估每个智能体的可靠性（reliability），二是对重复违规行为施加逐步升级的惩罚（escalating penalties）。实验在三个模拟环境和多种智能体群体中验证了该机制的有效性，结果表明其能够有效抵御策略性滥用，同时在惩罚违规行为的成本上优于或等同于基线方法。研究强调，只有当规范执行机制被设计为预见并适应自身将成为治理系统一部分的动态时，才能成为可扩展的智能体行为调控工具。
链接: https://arxiv.org/abs/2607.09766
作者: Yaowen Ye,Jacob Steinhardt
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG); Multiagent Systems (cs.MA)
备注: ICML2026 Trustworthy AI for Good Workshop
Abstract:AI agents are increasingly deployed in shared environments where they pursue diverse goals and compete for rewards. This multi-agent competition can lead to behaviors that serve individual gains at collective cost – for instance, marketing agents may post misleading content as a result of competing for engagement on social media. Human societies address such problems through norms that constrain acceptable behavior, supported by enforcement mechanisms that detect and penalize violations. Motivated by this, we study norm enforcement mechanisms for language model agents. We find that simple enforcement mechanisms are exploited by misaligned agents for competitive advantage, even when they are not explicitly trained or prompted to do so. We thus turn our attention to designing more robust mechanisms, and identify two key ingredients: estimating each agent’s reliability over time, and updating this estimate with escalating penalties for repeated misbehavior. Across three simulated environments and a variety of agent populations, mechanisms built on these principles resist exploitation, while still penalizing norm violations at comparable or lower cost than baselines. Our results position norm enforcement mechanisms as scalable levers for shaping agents’ behavior, but only when designed to anticipate becoming part of the system they govern. Our code and data are available at this https URL.
[MA-23] How Much Does Correctness Cost? Budgeted Placement of Strong Correctors in a Weak Multi-Agent Swarm ACL
【速读】：该论文旨在解决在由大量廉价且不可靠的代理（agent）组成的群体中，如何以最低成本部署少量高成本、高可靠性“预言者”（oracle）来实现全局共识的问题。核心挑战在于确定最优的预言者部署位置与数量，以在给定预算下最大化共识的一致性（coherence）。其解决方案的关键在于：尽管预言者的强度存在差异，但共识质量指标 $ H® = \text{tr}, M®^{-1} $ 仍保持次模性（submodularity），即每增加一个预言者所带来的增益递减。这一性质保证了基于成本-收益的贪心算法可在任意预算下达到近似最优解，逼近最优值的 $ 1 - 1/e $。进一步地，通过构建预算-正确率前沿 $ B^(\varepsilon) $，可量化最小支出以保证 $ \varepsilon $-正确共识；在完全图上可得闭式解，在等价成本下可求出最小预言者数量 $ k^ $。实验表明，成本-质量关系的曲率决定了部署策略——凹性（concave）的成本-质量函数意味着边际收益递减，因此更倾向于广泛分布的中等强度预言者，而非少数强预言者；在 Qwen3 模型系列（0.6–32B）上的实证显示，数学验证任务呈现凹性，而涌现代码追踪任务则呈凸性，由此揭示不同任务场景下的最优部署策略具有本质差异。
链接: https://arxiv.org/abs/2607.09765
作者: Igor Itkin
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Multiagent Systems (cs.MA); Systems and Control (eess.SY); Optimization and Control (math.OC)
备注: 30 pages, 9 figures, 1 table… Code and data: this https URL
Abstract:A cheap swarm of unreliable agents can be steered to a correct consensus by a few strong, expensive “oracle” correctors. We ask how much one must spend, and where to place the oracles. We model the swarm as a consensus on a graph in which each oracle pins one node toward the truth at a cost-coupled, concave strength, and measure quality by the coherence H®=tr M®^-1. Our first result is that H stays submodular (each added oracle helps less than the last) even when the oracles differ in strength, so a cost-benefit greedy comes within 1-1/e of the best placement at any budget. Inverting the budget gives the budget-correctness frontier B*(eps), the least spend that guarantees an eps-correct consensus: closed-form on the complete graph, and a minimal oracle count k* when oracles cost the same. Whether a budget then buys a few strong oracles or many medium onese curvature of the cost-quality law: diminishing returns favour spreadsharply increasion. Measured onthe Qwen3 ladder (0.6-32B), the law is concave for math verificatio convex foremergent code tracing, so the verdict is genuinely this http URL://github.com/YehudaItkin/budgeted-oracle-placemen
[MA-24] Replicating Belief Not Bits: Epistemic State Replication for Agent ic Systems
【速读】：该论文旨在解决传统分布式系统中状态机复制（State Machine Replication, SMR）模型在面对基于生成式AI的自主代理系统时所暴露出的根本性局限问题。传统SMR依赖于确定性、比特级完全一致的状态复制，但在由自主、随机且以模型驱动的代理构成的分布式系统中，各副本因使用生成式模型而可能产生不同的推理路径、摘要内容及分词边界，尽管最终达成语义等价且正确的操作决策。强制要求比特级一致性会严重限制执行灵活性，引发上下文遗忘，并制约系统性能。为此，论文提出认知状态复制（Epistemic State Replication, ESR），其核心在于将复制边界从数据可见性迁移至知识可见性，主张副本间应达成“信念”一致而非“比特”一致。关键创新在于形式化定义认知节点状态为一对结构 $ K = (L, B) $，其中 $ L $ 为不可变的确定性证据日志，$ B $ 为可演化、随机性的信念谱系。为保障执行安全性，引入语义线性化（Semantic Linearizability） 和 有界最终一致性（Bounded Eventual Coherence） 等新规范，分别确保操作反映验证器约束下的最新操作语义，并在公平交付、单调证据、有限验证干扰和压缩式嫁接算子条件下控制预期语义偏差。此外，论文设计了基于结构化认知增量的洞察传播协议，并提出可验证的语义回滚机制，可在不造成上下文遗忘的前提下清除错误前提。原型实现与初步仿真结果验证了该方案在设定假设下的可行性，并显著降低了次生认知错误的发生率。
链接: https://arxiv.org/abs/2607.09748
作者: Jun He,Deying Yu
机构: OpenKedge.io
类目: Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC); Logic in Computer Science (cs.LO); Multiagent Systems (cs.MA)
备注: 16 pages, 4 tables
Abstract:In distributed systems, the classical State Machine Replication (SMR) model assumes that correct replicas execute deterministic transitions to yield identical bitwise states. However, the rise of agentic distributed systems – where autonomous, stochastic, and model-driven agents orchestrate infrastructure – presents scenarios where deterministic, bitwise replication is insufficient. Replicas operating with generative models may exhibit divergent reasoning paths, summaries, and token boundaries, yet reach semantically equivalent and correct operational decisions. Forcing bitwise agreement across these stochastic participants degrades execution flexibility, induces context amnesia, and limits performance. We argue that in such settings replicas should agree on belief, not bits. We propose Epistemic State Replication (ESR), a belief-replication layer for agentic distributed systems that shifts the replication boundary from data visibility to knowledge visibility. We formalize the epistemic node state as a pair K = (L, B) separating the deterministic, immutable evidence log (L) from the stochastic, evolving belief lineage (B). To govern execution safety, we define Semantic Linearizability, which requires operations to reflect the latest committed operational meaning within a verifier-bounded semantic compatibility metric, and Bounded Eventual Coherence, which bounds expected semantic divergence under fair delivery, monotonic evidence, bounded verifier disturbance, and a contractive graft operator. We outline protocols for propagating derived insights using structured epistemic deltas, and formalize Verifiable Semantic Rollbacks to prune faulty premises from belief lineages without inducing context amnesia. We prototype ESR and report preliminary simulation results that show feasibility under the stated assumptions and illustrate reductions in secondary cognitive faults. Comments: 16 pages, 4 tables Subjects: Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC); Logic in Computer Science (cs.LO); Multiagent Systems (cs.MA) Cite as: arXiv:2607.09748 [cs.AI] (or arXiv:2607.09748v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2607.09748 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[MA-25] SWIFT: A Small-World Interaction Framework for Flow-Aware Trajectory Prediction in Autonomous Driving
【速读】：该论文旨在解决自动驾驶中轨迹预测面临的动态交通参与者交互建模问题，尤其针对现有纯数据驱动方法在分布偏移（distribution shifts）下泛化能力不足的缺陷。其核心挑战在于如何在缺乏结构先验的情况下准确捕捉复杂交通场景中的局部与全局依赖关系。为此，论文提出一种名为SWIFT（Small-World Interaction Framework for Trajectory prediction）的统一框架，其解决方案的关键在于引入基于交通网络结构与流体理论的结构归纳偏置（structural inductive biases）。具体而言，SWIFT通过小世界交互网络（Small-World Interaction Network）同时建模局部邻近关系与长程全局依赖，并结合流量状态编码器（Flow Regime Encoder）实现交互结构对场景级交通状态的自适应调整；此外，引入多关系图模块（multi-relational graph module）显式建模交通参与者之间的直接与高阶关系，从而增强交互推理能力。实验结果表明，SWIFT在nuScenes、MoCAD和NGSIM三个真实世界数据集上均显著优于现有基线模型，在不同交通场景下均展现出更高的预测精度、更强的未见场景泛化能力、对观测噪声的鲁棒性以及小样本训练下的优异性能，验证了其结构感知设计的有效性。
链接: https://arxiv.org/abs/2607.09741
作者: Chengyue Wang,Bin Rao,Haicheng Liao,Bonan Wang,Chengzhong Xu,Zhenning Li
机构: University of Macau(澳门大学); State Key Laboratory of Internet of Things for Smart City(智慧城市物联网国家重点实验室); Department of Civil and Environmental Engineering(土木与环境工程系); Science and Technology Development Fund of Macau(澳门科学技术发展基金); Research Services and Knowledge Transfer Office(研究服务与知识转移办公室); Shenzhen-Hong Kong-Macau Science and Technology Program(深港珠澳科技计划); Science and Technology Planning Project of Guangdong(广东省科技计划项目); National Natural Science Foundation of China(国家自然科学基金); State Key Lab of Intelligent Transportation System(智能交通系统国家重点实验室); Jiangsu Provincial Science and Technology Program(江苏省科技计划项目)
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注: Accepted by IEEE Transactions on Pattern Analysis and Machine Intelligence
Abstract:Accurate trajectory prediction in autonomous driving hinges on modeling dynamic and context-dependent interactions among traffic agents. However, most existing approaches are purely data-driven and lack structural priors, which limits their generalization under distribution shifts. In this work, interaction modeling is revisited through the structure and dynamics of traffic networks, and SWIFT (Small-World Interaction Framework for Trajectory prediction) is proposed as a unified framework that integrates small-world networks with traffic flow theory. SWIFT introduces structural inductive biases via a Small-World Interaction Network that captures both local and global dependencies, and a Flow Regime Encoder that adapts the interaction structure to scene-level traffic states. Interaction reasoning is further enhanced through a multi-relational graph module that explicitly encodes direct and higher-order agent relationships. Extensive experiments on three real-world datasets, nuScenes, MoCAD, and NGSIM, show that SWIFT consistently outperforms strong baselines in prediction accuracy across diverse traffic regimes. Beyond accuracy, SWIFT exhibits improved generalization to unseen locations and regimes, robustness under noisy observations, and strong performance with limited training data, supporting the effectiveness of its structure-aware design.
[MA-26] Feedback-Coupled Memory Systems in Continuous Time
【速读】：该论文旨在解决反馈耦合记忆系统（Feedback-Coupled Memory Systems, FCMS）架构中两个核心算子——智能体更新算子 $ f_i $ 与环境更新算子 $ \Psi $ ——在原始框架中未被形式化定义的问题，从而导致系统动态行为缺乏可分析性与稳定性保障。其解决方案的关键在于：通过机制基础智能（Mechanism-Based Intelligence, MBI）对 $ f_i $ 进行建模，使智能体基于去中心化的价格机制与经济原则实现局部更新；同时引入耦合记忆图过程（Coupled Memory Graph Process, CMGP）作为非马尔可夫框架来定义 $ \Psi $，将环境视为能够协同记录并响应轨迹历史的物理基底，无需外部驱动力。由此构建的连续时间FCMS实例实现了由可计算阈值 $ 4\beta^2 / (2\eta\mu\gamma^2) $ 所决定的李雅普诺夫全局耗散性，该结果统一了离散FCMS的稳定性条件 $ 4\eta\beta^2/\gamma $ 与CMGP的物理分岔阈值 $ \alpha_c = 1/K ，揭示了“记忆耗散必须超过反馈增益”这一普遍存在的组织原则。数值模拟（ N=2 ）与均场验证（ N=10^6 $）进一步证实了该稳定阈值的有效性，并展示了当该阈值被突破时所引发的自增强协调级联现象。
链接: https://arxiv.org/abs/2607.09714
作者: Stefano Grassi
机构: 未知
类目: Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注: 19 pages, 4 figures. Extends arXiv:2603.11560 to continuous time. Code: this http URL
Abstract:The Feedback-Coupled Memory Systems (FCMS) architecture formalizes closed-loop coordination through four abstract operators, two of which - the agent update operator f_i and the environmental update operator \Psi - are left axiomatically undefined in the original framework. To address this, f_i is defined by Mechanism-Based Intelligence (MBI), where agents update locally through a decentralized price mechanism and economic principles, and \Psi is defined by the Coupled Memory Graph Process (CMGP), a non-Markovian framework where the environment is treated as a physical substrate that records and responds to trajectory history coherently without external forcing. The resulting continuous-time FCMS instantiation achieves Lyapunov global dissipativity governed by the computable threshold 4\beta^2 2\eta\mu\gamma^2 . This generalizes both the discrete FCMS stability condition 4\eta\beta^2 \gamma and CMGP’s physical bifurcation threshold \alpha_c = 1/K , confirming that memory dissipation must outpace feedback gain as a universal organizing principle. Numerical simulation with N=2 agents and mean-field validation at N=10^6 confirm the stability threshold and the self-reinforcing coordination cascade that emerges when it is violated.
[MA-27] Closed-Loop Control with Rule-Aligned Small Language Models and Multi-Agent Self-Correction
【速读】：该论文旨在解决工业自主运行中控制策略从自然语言需求规范中自动创建与动态重构的难题，核心挑战在于如何在保证控制可靠性的同时，实现低延迟、低算力开销的边缘端部署。现有基于大型云模型的生成式方法受限于推理延迟、计算资源占用及数据敏感性，难以满足闭环控制对实时性与可解释性的要求。为此，论文提出一种基于小型语言模型（Small Language Model, SLM）的验证器引导修正框架，其关键在于将一个经过分组相对策略优化（Group Relative Policy Optimization, GRPO）微调的Qwen2.5-1.5B模型嵌入由动作代理、符号化/数字孪生式验证层以及迭代重提示代理构成的闭环系统中，通过验证层对候选动作进行物理可行性校验，并由重提示代理持续引导输出趋向合法动作。在30次随机热控仿真（每轮500步）中，该框架实现了平均91.5%的动作对齐准确率（跨案例86.3%–100%），平均推理延迟仅为3.84秒；在符号映射条件下仍保持95%的输出范围合规率，表明其具备在降低词级别一致性的同时维持稳健物理调控能力的潜力。研究结果证实，SLM与验证器协同架构是实现边缘侧可重构自主控制的一种可行且高效的技术路径。
链接: https://arxiv.org/abs/2607.09713
作者: Yuchen Wang,Javal Vyas,Tong Liu,Mehmet Mercangoz
机构: University of Sheffield (谢菲尔德大学); Imperial College London (帝国理工学院)
类目: Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA); Robotics (cs.RO)
备注: Accepted by IEEE CCTA 2026
Abstract:A key step toward autonomous industrial operation is the ability to create and reconfigure control policies from natural-language requirement specifications, with minimal or no manual redesign. In this setting, policy generation by AI agents can be a credible path when paired with a plant-aware validator (e.g., a digital twin) that can check generated candidate actions before execution. However, practical deployment is constrained by inference latency and compute footprint: large cloud-based models are often too slow, opaque, or data-sensitive for edge closed-loop use. This work investigates whether a compact Small Language Model (SLM) can be retrained for control reasoning and embedded in a validator-guided correction loop. We use a Qwen2.5-1.5B model aligned via Group Relative Policy Optimization (GRPO), combined with (i) an action agent, (ii) a symbolic/digital-twin-style validation layer, and (iii) a reprompting agent that iteratively steers outputs toward valid actions. In randomized thermal-control simulations (30 experiments with 500 steps each), the framework achieves 91.5% average action-alignment accuracy (86.3%–100% across cases) at 3.84,s mean inference latency. Under symbolic re-mapping, it maintains a 95% in-range rate, indicating robust physical regulation despite reduced token-level agreement. These results support SLM+validator architectures as a practical path toward reconfigurable autonomous control at the edge.
[MA-28] ransfer Learning Across Policy Regimes in Adaptive Multi-Agent Systems
【速读】：该论文旨在解决在适应性社会技术系统中，传统政策模型假设政策工具与结果之间的关系在不同制度条件下保持稳定这一前提可能失效的问题。当监管环境变化导致激励机制改变、行为主体采取策略性响应，或政策变量到宏观结果的映射关系发生结构性变迁时，基于历史数据训练的政策模型可能失效。其解决方案的关键在于将政策制度变迁建模为适应性多智能体系统中的迁移学习（transfer learning）问题：将每个政策制度视为由可观测输入分布和目标函数（即政策变量到结果的映射）定义的学习任务。研究对比了“空白学习者”（从零开始搜索灵活假设空间）与“迁移学习者”（利用前一制度中结构知识约束假设空间）的表现。迁移学习有效性的核心条件是：所继承的结构知识需保留新制度下的真实目标函数，同时降低有效复杂度；若该结构知识排除了新的目标函数，则会导致模型误设（misspecification），引发负向迁移（negative transfer）。通过简化的排放管制实验环境和动态基于代理的建模（ABM）鲁棒性实验验证，发现当新制度维持线性单调的税收-排放关系时，迁移学习可提升小样本下的实证性能；而当新制度引入阈值突变时，相同的结构迁移反而导致更高的保留误差、在线预测错误率上升，以及累积误差和最终窗口误差显著增大。因此，该研究的贡献在于方法论层面：在政策制定中应重用过去监管经验中蕴含的稳定结构不变量（structural invariants），但对政策—结果关系发生根本性改变的情形需谨慎处理。
链接: https://arxiv.org/abs/2607.09685
作者: Roberto Garrone
机构: Open University of Cyprus (塞浦路斯开放大学)
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Machine Learning (cs.LG)
备注: 17 pages, 3 figures, 8 tables. Simulation-based methodological study of positive and negative transfer across adaptive policy regimes
Abstract:Policy models often assume that the relationship between a policy instrument and its outcome remains stable across institutional conditions. In adaptive socio-technical systems this assumption may fail: regulatory change can alter incentives, agents can respond strategically, and the mapping from policy variables to aggregate outcomes can change. This paper studies such regime change as a transfer-learning problem in adaptive multi-agent systems. A policy regime is represented as a learning problem induced by an observable input distribution and a target function mapping policy variables to outcomes. We compare a blank-slate learner that searches a flexible hypothesis class in the new regime with a transfer learner whose effective hypothesis class is restricted by structural knowledge from the previous regime. Transfer is beneficial when this restriction preserves the new target function while reducing effective complexity; it is harmful when the restriction excludes the new target and creates misspecification. A stylized emissions-regulation experimental environment and a dynamic ABM robustness experiment support the claim. When the target regime preserves an affine monotone tax-emissions relation, transfer improves empirical small-sample performance. When the target regime introduces a threshold break, the same transferred structure produces negative transfer: held-out error remains high, online prediction generates more mistakes, and repeated online streams show larger cumulative and final-window error under misspecification. The contribution is methodological: previous regulatory experience should be reused when it captures stable structural invariants, but treated cautiously when policy change alters the policy-outcome relationship.
自然语言处理
[NLP-0] Metacognition in LLM s: Foundations Progress and Opportunities
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在元认知（Metacognition）能力方面缺乏系统性理解与有效实现的问题。尽管大语言模型在多种现实任务中已取得显著进展，但其是否具备、如何实现以及在何种程度上能够表现出有效的元认知能力仍不明确，且这些能力如何被整合以提升AI系统的根本性能、可靠性与智能水平也尚未厘清。本文的关键解决方案在于首次全面梳理并构建了面向大语言模型的元认知研究领域知识体系，通过分析和分类该新兴领域的技术进展，系统总结了用于衡量与评估大语言模型元认知能力的方法与基准测试工具，提出了激发、增强与应用元认知能力的技术路径，并归纳了当前研究的重要发现与潜在影响。此外，论文还探讨了实际应用场景、开放性问题与挑战，并指明了未来研究的可行方向，旨在为该领域提供一份详尽且前沿的综述，推动深层次的学术讨论与创新研究。
链接: https://arxiv.org/abs/2607.11881
作者: Gabrielle Kaili-May Liu,Areeb Gani,Jacqueline Lu,Jordan Thomas,Mark Steyvers,Arman Cohan
机构: Yale University (耶鲁大学); University of California, Irvine (加州大学欧文分校)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Metacognition is a foundational component of intelligence critical to effective learning, problem solving, decision-making, communication, and more. In recent years, it has become increasingly recognized as a cornerstone of capable, transparent AI systems. Yet while LLMs have made significant progress across diverse real-world tasks, it is not yet clear when, how, or to what extent they can exhibit or be endowed with effective metacognitive abilities, nor how such abilities can be adapted to advance the fundamental capabilities, reliability, and intelligence of AI systems. This paper bridges this gap by presenting the first comprehensive overview of the current state of knowledge on metacognition for LLMs. We analyze and taxonomize the landscape of this emerging field and summarize recent technical advancements, including methods and benchmarks to measure and evaluate LLMs’ metacognitive abilities, techniques to elicit, improve, and apply metacognition in LLMs, and findings and implications of ongoing research. We also discuss applications, open questions and challenges, and promising directions for future work. Our aim is to provide a detailed and up-to-date review of this topic and stimulate meaningful research and discussion. An organized list of papers can be found at this https URL.
[NLP-1] A Durability and Cross-Language Transfer Benchmark for a Validated Teaching-Feedback Classification Protocol
【速读】： 该论文旨在解决教育机构收集的开放式教学评价反馈数据量大但实际被分析利用有限的问题，核心挑战在于如何高效、可靠地对这些非结构化文本进行主题分类与情感分析。其解决方案的关键在于构建一个经过验证的标注协议，该协议基于详尽的注释指南、标注者间一致性评估、分层交叉验证以及在西班牙语语料库上的保留编码器设计进行训练与测试。研究进一步探究了该协议在不同自然语言处理技术演进背景下的适应性：一方面检验其在三类代表性表示方法（稀疏词袋特征、冻结的Transformer嵌入、提示工程的大语言模型）中的表现；另一方面考察其在跨语言迁移至英语场景下的有效性。结果表明，尽管前沿模型在西班牙语主题识别任务中表现最优，但在情感分析任务上并未展现出显著优势，且在英语数据集上与低成本模型无明显差异，说明该协议具有良好的鲁棒性，模型选择更多是部署层面的权衡而非方法本身属性，凸显了该协议在实际应用中的稳定性和可迁移性。
链接: https://arxiv.org/abs/2607.11873
作者: Esteban U. Vega Barajas
机构: Universidad de Guadalajara(瓜达拉哈拉大学)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 12 pages, 2 figures
Abstract:Institutions collect far more open-ended teaching-evaluation feedback than they read. A prior study introduced a validated protocol for classifying such comments by thematic category and sentiment, built from a documented annotation guide, an intra-annotator reliability measurement, stratified cross-validation, and a held-out evaluation on a Spanish institutional corpus with a frozen-encoder design. Two questions limit its reuse: whether a protocol fixed to 2019-era frozen embeddings stays competitive as representation methods advance, and whether it transfers to a second language. We re-run it on the original Spanish data across three representation generations, sparse lexical features, frozen transformer embeddings, and prompted large language models, and transfer its sentiment task to English with a balanced 45,000-comment corpus checked against an aspect-labeled education dataset. Treating paired comparisons as descriptive, we find the protocol durable: a 2026 frontier model posts the highest thematic F1 on the hardest Spanish task, yet shows no sentiment advantage over a cheap model and no descriptive separation from it on English, so model choice is a deployment decision, not a property of the method.
[NLP-2] Inside the Unfair Judge: A Mechanistic Interpretability Account of LLM -as-Judge Bias
【速读】： 该论文旨在解决大语言模型作为评判者（LLM-as-judge）时存在的评分偏差问题，传统研究多聚焦于输入-输出层面的扰动分析与提示工程缓解策略。本文提出从表征层面（representation-level）重新审视此类偏差，认为评判模型隐藏状态中的激活几何结构可为偏差提供互补且更具操作性的解释。其核心解决方案在于揭示：在多个评测基准与偏见类型下，基准输入在隐藏层中聚集于紧凑的激活流形，而带有偏见的输入则沿低维、类型特异的子空间发生位移，且该子空间随网络深度增强并可被三类估计器一致恢复。进一步发现，沿该子空间进行因果调控可双向调节评分——正向移动复现偏见性评分，反向移动则使偏见输入回归基线评分，而匹配范数的随机方向仅产生量级小一个数量级的影响。此外，仅通过简单线性投影至该偏见方向特征，即可在三个全新未见基准上准确预测模型失败，显著优于基于文本的替代方法。该研究将偏见理解为激活几何结构，而非输入-输出噪声，从而在统一框架内实现了几何结构、因果控制与操作性预测的整合。
链接: https://arxiv.org/abs/2607.11871
作者: Zixiang Xu,Sixian Li,Huaxing Liu,Xiang Wang,Shuai Li,Zirui Song,Xiuying Chen
机构: AMAP, Alibaba Group(阿里巴巴集团智能计算实验室); Mohamed bin Zayed University of Artificial Intelligence(穆罕默德·本·扎耶德人工智能大学); University of Southern California(南加州大学); University of Michigan, Ann Arbor(密歇根大学安娜堡分校)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 58 pages, 13 figures, 30 tables; project page: this https URL
Abstract:Existing studies of LLM-as-judge scoring bias work predominantly at the input-output level: they perturb inputs, measure score deltas, and propose prompt-level mitigations. We argue that the same biases admit a representation-level account in the judge’s hidden state, complementary to the input-output view and operationally useful in ways it does not afford. We report three findings, across seven judges, seven bias types, and nine benchmarks. Geometry: baseline judging inputs occupy a tight activation manifold while biased inputs are displaced along a low-dimensional, type-specific subspace that sharpens with depth and is recovered consistently by three families of estimators. Causal control: steering hidden states along this subspace drives scoring in both directions, forward shifts reproducing biased scoring on clean inputs and reverse shifts restoring baseline scoring on biased ones, while matched-norm random directions produce shifts an order of magnitude smaller. Operational: a simple linear projection onto the same bias-direction features anticipates judge failures on three entirely unseen benchmarks, substantially outperforming text-based alternatives. Reading bias as activation geometry, rather than as input-output noise, unifies geometric structure, causal control, and operational prediction within a single framework. The project page is available at this https URL
[NLP-3] AdvancedMathBench: A Benchmark Suite for Advanced Mathematical Proof Generation and Verification
【速读】： 该论文旨在解决当前大型语言模型（LLMs）在高等数学领域推理能力评估不足的问题。现有基准测试在学科覆盖范围和评估粒度上均存在局限，仅依赖最终答案正确性或粗略判断，难以有效衡量模型在复杂数学证明过程中的推理有效性。为此，作者提出了AdvancedMathBench基准套件，其核心为ProverBench，包含296道涵盖本科至博士资格考试水平的数学证明题。为实现可靠验证，研究团队开发了一套基于大规模专家标注训练的自动化验证流水线，能够输出证明正确性判定及细粒度错误分析，与人类专家在保留样本上的判断具有高度一致性。此外，引入VerifierBench，包含888条模型生成的证明轨迹及其专家真值，用于评估模型对证明有效性的判别能力及验证理由的合理性。实验结果表明，当前前沿模型在该基准上仍面临巨大挑战：在证明生成任务中，表现最佳的GPT-5.5-xhigh模型在UGD和QE子集上的得分仅为75.8和66.1；在证明验证任务中，最优模型的平衡F1仅为65.1，且普遍存在低真负率，反映出模型在关键错误检测方面仍存在显著瓶颈。解决方案的关键在于构建高精度、细粒度的自动化验证机制，并通过真实专家标注数据驱动模型评估体系的建立，从而推动对高级数学推理能力的精准测量。
链接: https://arxiv.org/abs/2607.11849
作者: Lingkai Kong,Zijian Wu,Yuzhe Gu,Haiteng Zhao,Wenyong Huang,Shuang Sun,Zhicheng Xiong,Xiaotian Zhang,Shuya Zhao,Yan Wang,Disheng Xu,Wenwei Zhang,Kai Chen
机构: Shanghai AI Laboratory; Shanghai Jiao Tong University; MMLab, The Chinese University of Hong Kong; Great Bay University
类目: Computation and Language (cs.CL)
备注:
Abstract:Large language models (LLMs) have achieved remarkable performance on high-school and olympiad-style mathematics, yet their capabilities on advanced mathematics remain poorly understood. Existing benchmarks, however, fall short in both scope and evaluation granularity: they provide limited disciplinary coverage and often rely on final-answer correctness or coarse judgments, leaving the validity of the reasoning process inadequately assessed. To bridge this gap, we introduce AdvancedMathBench, a benchmark suite designed to evaluate advanced mathematical reasoning capabilities. Its core proof-generation benchmark, ProverBench, contains 296 problems spanning undergraduate and doctoral qualifying-exam levels. To provide reliable evaluation of the proofs, we develop a dedicated automatic verification pipeline trained on large-scale expert annotations to produce both correctness verdicts and fine-grained assessments of proof errors, which exhibits strong agreement with human experts on held-out proof trajectories. We further introduce VerifierBench, consisting of 888 model-generated proof trajectories paired with expert ground truth, to evaluate whether models can correctly judge proof validity and provide sound verification rationales. Experiments show that AdvancedMathBench remains challenging for frontier models. On proof generation, the best-performing model, GPT-5.5-xhigh, achieves only 75.8 and 66.1 on the UGD and QE splits, respectively, indicating substantial room for improvement on advanced mathematical proof construction. On proof verification, the best model attains a Balanced F1 of only 65.1, and models generally exhibit low true negative rates, suggesting that critical error detection remains a major bottleneck.
[NLP-4] Introducing Human-Centeredness in AI-Assisted Lexicography
【速读】： 该论文旨在解决生成式 AI（Generative AI）在词典编纂领域应用过程中引发的核心问题，即如何在提升词典编纂效率的同时，保障词典学家的专业角色、维护语言与文化多样性，并有效应对人工智能带来的偏见风险。其解决方案的关键在于构建一个人本导向的人工智能（Human-Centered Artificial Intelligence, HCAI）框架，强调通过四个相互关联的维度——增强型词典学家、人机融合的社会技术情境、算法偏见治理以及面向词典学家需求的AI工具设计——实现人工智能对词典编纂工作的赋能而非替代。该框架主张在高度自动化的基础上保留有意义的人类控制权，确保词典学家的专业自主性，推动以用户为中心的工具设计，并系统性地缓解由AI引入的潜在偏见，从而为未来人工智能在词典编纂工作流中的可持续、负责任集成提供理论基础与实践指导。
链接: https://arxiv.org/abs/2607.11808
作者: Antonio San Martin,Catherine Trekker
机构: Université du Québec à Trois-Rivières(魁北克省三河城大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Accepted for publication in the Proceedings of the XXII EURALEX International Congress 2026
Abstract:This paper proposes a human-centered artificial intelligence (HCAI) framework for AI-assisted lexicography. While generative AI offers significant opportunities to enhance lexicographic work, it also raises concerns regarding the future role of lexicographers and the preservation of linguistic and cultural diversity. Drawing on HCAI principles and previous applications in other language professions, the paper identifies four interrelated dimensions through which AI integration in lexicography can be understood and critically examined: the augmented lexicographer, the sociotechnical context of AI integration, bias, and the design of AI-powered lexicographic tools. The framework argues that AI should augment rather than replace lexicographers, combining high levels of automation with meaningful human control. It further emphasizes the importance of preserving professional agency, mitigating AI-generated biases, and designing tools around the needs of lexicographers. By doing so, the paper provides a foundation for future research and the beneficial integration of AI into lexicographic workflows.
[NLP-5] How Temperature Shapes Ideological Discourse in Retrieval-Augmented Generation?
【速读】： 该论文旨在解决生成式AI（Generative AI）在检索增强生成（Retrieval-Augmented Generation, RAG）框架下因外部知识源中存在意识形态偏见而导致模型输出被传播、放大或抑制特定意识形态话语的问题。现有研究多关注检索过程中的错误鲁棒性，却忽视了意识形态偏见对生成结果的潜在影响。本文的关键解决方案在于：通过词汇多维分析（Lexical Multidimensional Analysis, LMDA）对1,117篇关于新冠疫情治疗的文章语料进行分析，识别出三种核心意识形态话语，并将其作为RAG系统的外部知识源。随后，利用多个大语言模型（LLMs）在不同采样温度下回答与意识形态相关的问题，通过语义与词汇层面的相似性分析，评估生成文本与参考意识形态文本之间的论述一致性。研究发现，RAG框架易将外部知识中的意识形态话语传递至模型输出，且采样温度显著影响这一传递强度——中等温度下模型在随机性与检索依赖之间取得平衡，实现最高的论述一致性；而在低温度下，过度确定性的采样策略反而抑制了意识形态话语的传递。这表明，采样策略是控制意识形态偏见在生成过程中传播的关键调节因素。
链接: https://arxiv.org/abs/2607.11783
作者: Elmira Salari,Hazem Amamou,José Victor de Souza,Shruti Kshirsagar,Maria Nunes Delfino,Anderson Avila
机构: Wichita State University (威奇托州立大学); Institut national de la recherche scientifique (加拿大国家科学研究院); São Paulo Catholic University (圣保罗天主教大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Retrieval-Augmented Generation (RAG) has been increasingly adopted to reduce hallucinations and strengthen the factual grounding of large language models (LLMs). While robustness to errors in the retrieval process has been explored, the impact of ideological bias on LLM outputs has been overlooked. For instance, if the retrieved material contains ideological positions, the RAG may transmit, amplify, or suppress such ideological discourses in its outputs. In this study, we address this issue by examining the influence of the RAG framework, comprising ideological discourses, in LLM-generated answers. To this end, we applied Lexical Multidimensional Analysis (LMDA) on a corpus of 1,117 COVID-19 treatment articles, identifying three ideological discourses. This corpus is then used as the external knowledge source for the RAG. We assessed several LLMs by having the models answer ideological questions at different sampling temperatures. The generated texts were assessed semantically and lexically based on their similarities with ideological reference texts. Our findings show that the RAG framework is prone to transferring ideological discourses into LLM responses, with sampling temperature having a measurable impact on the strength of this transfer. Discoursive alignment between generated answers and the reference text is highest at moderate temperatures, where models balance stochasticity with retrieval grounding, and drops at low temperatures, indicating that overly deterministic sampling suppresses discourse transfer.
[NLP-6] From Expressivity to Sample Complexity: Narrow Teachers for Transformers via C-RASP
【速读】： 该论文旨在解决大语言模型（LLM）中注意力机制的可学习性（learnability）问题，即在理论上理解Transformer模型是否能够有效学习特定任务构造。尽管已有大量研究通过手工设计权重或利用计算复杂性分析来探讨注意力模型的表达能力（expressivity），揭示了哪些任务属于Transformer的假设类（hypothesis class），但对这些解的可学习性缺乏系统研究。本文的关键突破在于受近期损失曲面分析工作的启发，首次提出了针对C-RASP构造的初步样本复杂度（sample complexity）上界，为理解Transformer在学习此类结构时所需的数据规模提供了理论依据，从而推进了对模型学习能力的理论刻画。
链接: https://arxiv.org/abs/2607.11760
作者: Michael Rizvi-Martel,Satwik Bhattamishra,Guillaume Rabusseau,Michael Hahn
机构: Mila Université de Montréal; University of Oxford; Saarland University
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: 9 pages total
Abstract:A theoretical understanding of Transformers is crucial to better understand the capacities and limitations of large language models (LLMs). There is much work analyzing the expressivity of attention-based models. By proposing handcrafted weights or using computational complexity arguments, a large amount of past theoretical works have sought to characterize which tasks are and which are not in the hypothesis class of Transformer models. However, little work investigates the learnability of such solutions. In this work, we make progress towards this goal. Inspired by recent loss landscape analysis work, we propose preliminary sample complexity bounds for learning C-RASP constructions with Transformers.
[NLP-7] MET: Theory-Grounded and Culture-Aware Multilingual Moral Reasoning
【速读】： 该论文旨在解决生成式 AI 在跨语言、跨文化情境下进行道德决策时存在的三大关键问题：一是现有的多语言评估基准采用直接翻译，未能适配文化特异性内容；二是推理阶段的道德推理方法依赖静态的英文中心化提示框架，缺乏道德理论的坚实支撑；三是道德决策训练通常需要昂贵的强模型监督或人工标注。针对这些问题，论文提出三项核心贡献：其一，构建了 MCLASH 多语言道德决策基准，以捕捉不同语言背景下文化相关的道德直觉与社会规范；其二，提出 MET（Multilingual Ethics with Theory-grounded reasoning），一种基于心理学与哲学领域专家整理的理论基础的两步提示方法，使模型能够首先选择情境与文化相关的道德依据，再在用户母语中进行推理；其三，引入 MET-D（MET-Distillation），通过自蒸馏训练阶段实现无需外部监督的第二步推理优化。实验表明，MET-D 在 Qwen3-4B、Qwen3-8B 与 Gemma3-4B 三种不同规模和架构的模型上均显著提升性能，平均在 MCLASH 和 MMoralExceptQA 基准上分别取得 3.71 和 4.23 的宏平均 F1 提升，其中在马来语上的最高提升达 12.94 点；同时，母语推理能力平均提高 62.13 分，且不同文化间有益的道德依据存在系统性差异。上述工作为实现文化对齐、理论根基扎实的多语言道德推理提供了可行路径。
链接: https://arxiv.org/abs/2607.11736
作者: Ayoung Lee,Ryan Kwon,Yunxiang Zhang,Yuxuan Liu,Peter Railton,Lu Wang
机构: University of Michigan (密歇根大学)
类目: Computation and Language (cs.CL)
备注: Published as a conference paper at COLM 2026
Abstract:Language models are increasingly used for moral decision-making across diverse linguistic and cultural contexts, yet existing work overlooks multilinguality on three aspects: 1) multilingual evaluation benchmarks use direct translation, failing to adapt culture-specific items; 2) inference-time methods for moral reasoning rely on static, English-centric scaffolds and lack grounding in moral theory; 3) training methods for moral decision-making typically require expensive supervision from stronger models or human annotators. We address these gaps with three contributions. First, we introduce MCLASH, a multilingual moral decision-making benchmark to capture culturally situated moral intuitions and social norms across languages. Second, we propose MET (Multilingual Ethics with Theory-grounded reasoning), a two-step prompting method built on expert-curated, theory-based grounds drawn from psychology and philosophy: the model first selects situation- and culture-specific grounds, then reasons over them in the native language of the user. Third, we introduce MET-D (MET-Distillation), which enhances the second step through a self-distillation training stage that requires no external supervision. MET-D improves macro-F1 over the base model on all three models of different sizes and families (Qwen3-4B, Qwen3-8B, Gemma3-4B), by an average of 3.71 points on MCLASH and 4.23 on MMoralExceptQA, with a peak MCLASH gain of 12.94 points for Malay on Qwen3-8B. We further reveal that MET-D increases native-language reasoning by 62.13 points on average, and that beneficial grounds differ systematically across cultures. Together, these contributions open the path for culture-aligned, theory-grounded multilingual moral reasoning.
[NLP-8] STEP: Career-Path Recommendation via Temporal and Educational Trajectory Modeling
【速读】： 该论文旨在解决大规模、系统性分析非结构化简历数据中职业路径（career path）的难题，以支持劳动力规划、劳动市场政策制定及职位推荐等应用。传统简历具有非结构化、异质性和多语言特性，长期限制了其在宏观层面的高效利用。针对这一挑战，本文提出STEP（Sequential Trajectory of Employment Prediction）——一种基于时间与教育信号的新型职业路径推荐系统，其核心创新在于：通过引入时序衰减门控循环单元（time-decay Gated Recurrent Unit, GRU）建模职业发展的动态演化过程，采用基于教育水平条件化的特征逐维调制（Feature-wise Linear Modulation, FiLM）捕捉教育背景对职业跃迁的影响，并结合注意力机制序列池化（attention-based sequence pooling）实现关键特征选择。为增强内部职业表征能力，进一步提出ROUTE——一种两阶段对比学习框架，先通过无监督去噪自编码实现多语言编码器在职业领域的适应，再通过带引导负样本选择的有监督对比微调优化表征质量。实验在四个职业轨迹数据集上验证了STEP的有效性，显著优于现有先进基线模型，且相关数据集与代码已开源，推动可复现的职业轨迹研究发展。
链接: https://arxiv.org/abs/2607.11722
作者: Iman Johary,Guillaume Bied,Alexandru C. Mara,Tijl De Bie
机构: AIDA-IDLab, Ghent University (根特大学); Ghent University (根特大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Career paths encode decades of skill acquisition, role transitions, and educational investment, and understanding them at scale underpins workforce planning, labor market policy, and job recommendation. Resumes are a rich source of information about career paths: they contain detailed descriptions of work experience, education, and skills. Yet their unstructured, heterogeneous, and multilingual nature has long prevented large-scale systematic analysis. With the advent of large language models (LLMs), it is now possible to source rich career trajectory data containing temporal and educational signals from unstructured resumes, enabling new opportunities for career-path recommendation. Exploiting this opportunity, we present STEP (Sequential Trajectory of Employment Prediction), a novel career-path recommendation system that leverages temporal and educational signals to predict the next job in a career trajectory. STEP integrates a time-decay Gated Recurrent Unit (GRU) cell to model temporal dynamics, Feature-wise Linear Modulation (FiLM) conditioned on educational attainment, and attention-based sequence pooling to select relevant features for next job prediction. To improve internal occupation representation for STEP, we introduce ROUTE, a two-stage contrastive procedure that first adapts a multilingual encoder to the career domain via unsupervised denoising autoencoding, then performs supervised contrastive fine-tuning with guided negative selection. We evaluate STEP on four datasets of career trajectories, including an improved version of our publicly available JobHop dataset, and show that it outperforms state-of-the-art baselines in next job prediction. The dataset and code are publicly released to support reproducible career-trajectory research.
[NLP-9] JobHop v2: A Large-Scale Career Trajectory Dataset from Unstructured Resumes
【速读】： 该论文旨在解决现有职业轨迹数据集在规模、可获取性及标注质量方面的局限性，尤其针对公开数据集普遍存在的样本量小、使用受限或依赖大语言模型（LLM）合成而非真实自由文本的问题。其解决方案的关键在于构建一个大规模、多语言、基于真实简历的高质量职业轨迹数据集——JobHop v2，通过端到端的大语言模型（LLM）抽取流程，从约44万份经匿名化处理的多语种简历中提取结构化信息。该方案的核心创新包括：采用具备推理控制与重试机制的改进型LLM推理管道，实现100%的JSON解析率；设计更丰富的抽取模式以支持五级教育水平、季度级时间信息及ESCO职业代码的精准标注；并引入三重互补的人工标注基准进行评估。实验表明，最佳提取器在多个指标上接近人工标注者间的一致性上限，误差仅为1.1–2.7个百分点，显著提升了自动化职业轨迹抽取的准确性与可复现性。
链接: https://arxiv.org/abs/2607.11715
作者: Iman Johary,Guillaume Bied,Alexandru C. Mara,Tijl De Bie
机构: AIDA-IDLab, Department of Electronics and Information Systems, Ghent University (根特大学), Ghent, Belgium
类目: Computation and Language (cs.CL)
备注:
Abstract:Large-scale, richly annotated career trajectory data underpins workforce planning, job recommendation, and labour market analysis, yet publicly available datasets are either small, closed to independent use, or built from pre-standardized occupational codes with LLM-synthesized rather than authentic free text. We present JobHop~v2, an improved version of the publicly available JobHop dataset, constructed through end-to-end large language model (LLM) extraction from a corpus of \sim440,000 pseudonymized, multilingual resumes provided by VDAB, the Flemish Public Employment Service. The released dataset comprises 355,315 career trajectories annotated with ESCO occupational codes, quarter-level temporal information, and normalized five-level education attainment, broadening both the coverage and the annotation richness of the original release. Relative to v1, JobHop~v2 introduces a redesigned extraction pipeline based on reasoning-controlled LLM inference with a retry mechanism (achieving a 100% JSON parse rate), a richer extraction schema, and a revised evaluation protocol scored against three complementary annotation baselines. Evaluated against these baselines, our best extractor comes closest to the inter-annotator agreement ceiling among all compared models, trailing it by only 1.1-2.7 percentage points. The dataset and code are publicly released to support reproducible career-trajectory research.
[NLP-10] Production and Perception in LLM s: A Token Probability Approach
【速读】： 该论文旨在探究大语言模型（Large Language Models, LLMs）是否在生成与理解功能上表现出类似人类语言处理中的不对称性，即语言生成与语言感知之间的差异。尽管现有LLMs均采用统一的“下一词预测”机制进行输入与输出处理，但其内在行为是否仍存在功能性区分尚不明确。本研究的关键在于通过直接测量生成过程中各标记（token）的概率分布，而非依赖元语言提示（metalinguistic prompting），来操作化地定义生成-感知差异。实验以Llama-3.1-8B为基础模型，在不同生成与感知提示下对同一诗歌文本进行重评分，结果表明：生成-感知间的概率距离显著且一致地大于生成-生成间的距离，平均比值约为1.8，且各类条件间无重叠。控制组中生成-生成之间接近天花板的相关性进一步证明该效应源于交流框架的差异，而非提示表面形式的变化。该现象在五种开源模型（包括基础模型与指令微调版本）中均具可重复性，且时间序列分析显示感知提示的影响主要体现在序列起始阶段，随上下文积累而衰减，但衰减模式因提示对而异。研究揭示，仅通过提示框架的改变即可在解码器-only架构的LLM中诱发生成-感知的概率分布差异，表明语言生成与感知的功能性分离可能已内生于当前主流大模型的行为之中。
链接: https://arxiv.org/abs/2607.11703
作者: Anna Marklová,Jiří Milička,Martina Vokáčová,Rudolf Rosa
机构: Faculty of Arts, Charles University, Prague; Faculty of Mathematics and Physics, Charles University, Prague
类目: Computation and Language (cs.CL)
备注:
Abstract:The asymmetry between language production and perception has been well-documented in psycholinguistics. Whether large language models (LLMs) exhibit a functionally analogous distinction remains an open question, particularly given that LLMs rely on the same underlying mechanism (next-token prediction) for both input and output processing. In this exploratory study, we operationalize the production-perception distinction through direct token probability measurements rather than metalinguistic prompting. Using the base Llama-3.1-8B model, we generated poems under a production prompt and re-scored the same tokens under both rephrased production prompts and perception-oriented prompts. Across an extended experiment with four production and three perception prompts, production-perception distances consistently and substantially exceeded production-production distances, with non-overlapping ranges across conditions and an overall average ratio of approximately 1.8. Near-ceiling correlations in the production-production control confirm that the effect is specific to communicative framing rather than prompt surface variation, and we show the effect replicates across five open-weight models (Llama-3.1-8B, EuroLLM-9B, gemma-2-9b-it, Mistral-7B-Instruct-v0.3, and Qwen2.5-7B-Instruct), spanning both base and instruction-tuned variants. Temporal analysis revealed that the perception prompt exerts its strongest influence at the beginning of the sequence, with divergence decaying as generated context accumulates, though the specific shape of this decay varies across prompt pairs. These findings suggest that prompt framing alone induces a production-perception distinction in LLM probability distributions, even within a decoder-only architecture.
[NLP-11] RAG U: A Multi-Step GraphRAG Engine with a Compact Domain-Adapted LLM
【速读】： 该论文旨在解决现有图检索增强生成（GraphRAG）系统在知识图谱构建过程中依赖单次抽取导致实体噪声大、检索脆弱的问题。其核心解决方案在于引入模块化设计，将实体与关系的抽取与合并过程解耦：通过两阶段类型化抽取、基于DBSCAN的去重、大语言模型（LLM）摘要生成以及Leiden社区检测实现知识图谱的高质量构建。关键洞察在于，在流水线中的LLM所需的核心语言能力（如理解、抽取、上下文推理）随模型规模增长缓慢，远不如事实性世界知识对规模敏感。基于此，作者训练了专精于语言技能的轻量级模型Meno-Lite-0.1（7B参数），在知识图谱构建任务上相对谐均值提升12.5%，性能超越Qwen2.5-32B，并在英文GraphRAG任务中表现相当；在GraphRAG-Bench（医学）基准测试中，RAGU在每个事实粒度下均实现最完整的上下文召回（证据召回率最高达0.84，优于≤0.76），且在合成任务中超越HippoRAG2；多跳事实问答任务中观察到的HippoRAG2优势被证实主要源于答案格式偏差。RAGU支持单GPU部署，可通过pip安装，开源发布于MIT许可下，代码与模型均已公开。
链接: https://arxiv.org/abs/2607.11683
作者: Mikhail Komarov,Ivan Bondarenko,Stanislav Shtuka,Oleg Sedukhin,Roman Shuvalov,Yana Dementyeva,Matvey Solovyov,Nikolay O. Nikitin
机构: ITMO University (圣彼得堡国立研究大学); Novosibirsk State University (托木斯克国立大学); Far Eastern Federal University (远东联邦大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Graph retrieval-augmented generation (GraphRAG) enhances large language models with structured knowledge, yet existing systems construct knowledge graphs in a single extraction pass, producing noisy entities and brittle retrieval. RAGU, an open-source modular GraphRAG engine, addresses this by separating extraction from consolidation: entities and relations pass through two-stage typed extraction, DBSCAN-backed deduplication, LLM summarization, and Leiden community detection. A key insight motivates a compact extractor: the skills an in-pipeline LLM needs - comprehension, extraction, reasoning over context - are language skills that grow only weakly with model size, unlike factual world knowledge. Accordingly, we train Meno-Lite-0.1, a 7B model optimized for language skills, which outperforms Qwen2.5-32B on knowledge-graph construction (+12.5% relative harmonic mean) and matches it on English GraphRAG tasks. On GraphRAG-Bench (Medical), RAGU retrieves the most complete context at every factoid level (evidence recall up to 0.84 vs. \leq 0.76) and overtakes HippoRAG2 on synthesis tasks; on multi-hop factoid QA, the apparent HippoRAG2 advantage is shown to be largely an answer-format artifact. RAGU is installable via \textttpip install graph_ragu , runs on a single GPU, and is released under MIT. The source code is publicly available at this https URL, and the Meno-Lite-0.1 model can be obtained from this https URL.
[NLP-12] Losing My Composure: Predicting Compositionality Over Time
【速读】： 该论文旨在解决德语与英语名词复合词在历时演变中语义变化及其组合性（compositionality）渐进退化或增强的问题，核心在于探究复合词意义透明度随时间演化的趋势，并建立可量化的建模框架。其解决方案的关键在于提出“组合性趋势预测”（Compositionality Trend Prediction）任务，构建了一个跨多个年代的历时语料库标注数据集，对23个德语和26个英语目标复合词进行分十年的组合性评分，首次实现了对组合性变化趋势的精细化实证分析。研究发现，尽管学界普遍认为复合词会随时间变得不那么组合性（即意义更不透明），但实证结果仅显示出微弱的负向趋势。在计算实验中，通过对比不同复杂度的语义向量表示及多种时间粒度的训练策略（如单十年窗口、逐步扩展的时间窗与半个世纪整体窗口），发现基于窄时间窗口（如单十年或递增时间窗）训练的模型能更准确捕捉组合性变化趋势，优于传统上以整个半世纪数据集为训练范围的建模方式。此外，静态向量表示在该任务中表现与上下文向量相当，表明其在捕捉历时组合性变化方面具备竞争力。
链接: https://arxiv.org/abs/2607.11667
作者: Chris Jenkins,Emma Raimundo Schulz,Filip Miletić,Sabine Schulte im Walde
机构: University of Stuttgart
类目: Computation and Language (cs.CL)
备注: 40 pages, 31 tables, 10 figures. This is a pre-print under review by Computational Linguistics
Abstract:We explore the phenomenon of semantic change of German and English noun compounds, with the objective of investigating and modeling gradual changes of meanings and degrees of compositionality in the past and over time. To do so, we introduce the Compositionality Trend Prediction task, which is evaluated against a novel dataset of in-context compositionality ratings sampled across several decades of diachronic corpora for 23 German and 26 English target compounds, uniquely providing per-decade ratings and corresponding trends over time. These per-decade compositionality ratings allow us to investigate empirically untested hypotheses of generalized trends in compositionality over time, such as the idea that compounds should become less compositional (less transparent) over time. Beyond our empirical observations from the diachronic compositionality annotations, we perform experiments with semantic vector representations of varying complexity, as well as several temporal granularities for training these representations on diachronic data, resulting in about 100 models of each representation type, each covering a different 1–5 decade slice of a diachronic corpus. Contrary to the decisive tendency posited in the literature, we find only a small negative trend in compositionality over time in our target compounds. In our computational experiments, we find that using models trained on narrow time slices of diachronic data (single decades, or incrementally expanding temporal windows) align better with the per-decade compositionality ratings than those trained on an entire half-century window, the latter setting being an analog for the prevalent modeling approach of training representations on an entire half of a corpus’ data. Additionally, we find static representations to be competitive with contextual representations in the Compositionality Trend Prediction task.
[NLP-13] Reproducing human biases in route choice using large language models : Toward scalable behavioral modeling
【速读】： 该论文旨在解决在大规模仿真与基于代理的建模中，如何有效刻画个体层面的非理性决策行为这一关键难题。传统方法依赖问卷调查和受控实验来校准累积前景理论（Cumulative Prospect Theory, CPT）参数，但此类方法难以泛化且无法充分捕捉人类决策的多样性。本文提出的核心解决方案是：利用大语言模型（Large Language Models, LLMs）在无需显式指定CPT参数的前提下，自动复现人类选择行为中的系统性偏差。研究以路径选择为典型场景，构建了行为评估框架，系统比较了LLM生成的决策与CPT预测的人类行为模式。结果表明，LLMs能够有效再现非理性选择偏差，并在不确定性情境下表现出与前景理论一致的决策特征。这一发现表明，生成式人工智能可作为建模人类决策过程的可扩展替代方案，为下一代大规模基于代理的仿真及人工智能驱动的行为科学研究提供了坚实基础。
链接: https://arxiv.org/abs/2607.11632
作者: Jiangtao Han,Shoufeng Ma,Shuxian Xu,Geng Li,Shuai Ling,Ning Jia,Zhengbing He
机构: Tianjin University (天津大学); University of Nottingham Ningbo China (诺丁汉大学宁波分校)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG); Social and Information Networks (cs.SI); Physics and Society (physics.soc-ph)
备注:
Abstract:Human choice behavior, including route choice, exhibits systematic behavioral biases that deviate from the assumptions of full rationality. Cumulative prospect theory (CPT) has been widely recognized as an effective framework for characterizing such behavioral patterns. However, its large-scale application, particularly in simulation and agent-based modeling, critically depends on specifying individual-level CPT parameters, which remain a major bottleneck. Conventional approaches typically rely on surveys and controlled experiments to calibrate CPT parameters, yet these methods are difficult to generalize and often fail to capture the full diversity of human decision-making. To address this challenge, this paper investigates whether large language models (LLMs) can reproduce human behavioral biases in choice-making without explicit specification of prospect-theoretic parameters. Using route choice as a representative scenario, we design a behavioral evaluation framework and systematically compare LLM-generated decisions with established human behavioral patterns predicted by CPT. Experimental results demonstrate that LLMs are capable of reproducing non-rational human choice biases and can exhibit decision behaviors consistent with prospect-theoretic effects under uncertainty. These findings suggest that generative AI models may provide a scalable alternative for modeling human decision processes and offer a promising foundation for next-generation large-scale agent-based simulation and AI-driven behavioral research.
[NLP-14] Lesioned Multimodal Language Models Reproduce Aphasic Picture-Naming Patterns
【速读】： 该论文旨在解决脑卒中后失语症（aphasia）患者在图片命名任务中出现系统性命名错误的特征模式是否可由未经临床模拟设计的通用语言模型再现的问题。其核心解决方案在于通过在多模态语言模型LLaVA 1.6上施加受控扰动（包括扰动层、扰动比例及噪声强度），探索不同参数配置下能否复现失语症患者的典型错误类型及其个体化误差分布。研究发现，在不同参数区域中，除形式性错语（formal paraphasia）外，其余六类错误（正确、语义错误、混合错误、无关错误、新造词、无反应）均以临床可比的比例出现；进一步搜索扰动空间表明，97.8%的失语症患者（PWAs）的个体误差模式可在至少七类中的六类中被复现，79.5%的患者甚至能在全部七类中实现匹配。蒙特卡洛基线分析证实，这种匹配反映的是类别间联合结构而非单一类别重叠。该研究建立了一个可量化的框架，首次实现了对个体失语症患者命名错误模式的高精度模拟，揭示了生成式语言模型作为卒中后失语症患者“数字孪生体”（digital twin）的潜力。
链接: https://arxiv.org/abs/2607.11621
作者: Yong Yang,Xiang Guan,Sophie Arheix-Parras,Saeed Ahmadi,Roger Newman-Norlund,Leonardo Bonilha,Christopher Rorden,Julius Fridriksson,Rutvik H. Desai,Srihari Nelakuditi
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 15 pages, 8 figures; supplementary materials (18 pages, 6 sections) included
Abstract:Aphasia following stroke commonly produces systematic naming errors with characteristic profiles, but whether general-purpose language models not designed for clinical simulation can reproduce these patterns remains untested. We investigated (1) whether lesions or controlled perturbations to a multimodal language model can reproduce different types of errors in picture naming, and (2) whether the framework can reproduce the complete error profile of individual persons with aphasia (PWAs). Using LLaVA 1.6, we evaluated perturbation configurations that varied the layer, proportion, and amount of noise applied to model units. We examined 278 PWAs on the Philadelphia Naming Test, classifying responses into seven categories using a validated neural classifier. Six of seven response categories (correct, semantic, mixed, unrelated, neologism, no response errors) emerged at clinically-comparable proportions across distinct parameter space regions, with formal paraphasia being the exception. Searching the perturbation space revealed configurations that reproduced the individual error profile in at least six of seven categories for 97.8% of PWAs and in all seven categories for 79.5% of PWAs. Monte Carlo baselines confirmed that this matching reflects joint inter-category structure rather than marginal overlap. These results establish a quantitative framework for reproducing individual aphasic error patterns in picture naming. They suggest the potential for language models to serve as digital twins of individuals with post-stroke aphasia.
[NLP-15] Extending LLM Context via Associative Recurrent Memory
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在实际应用中面临的关键瓶颈——上下文长度受限问题。标准Transformer架构存在二次计算复杂度和线性内存增长的固有缺陷，导致长序列处理效率急剧下降。为此，本文提出并研究了关联递归记忆变压器（Associative Recurrent Memory Transformer, ARMT）作为一种可行方案，以实现长上下文处理、恒定内存扩展以及更高的计算效率。其解决方案的关键在于：构建两个面向特定领域的长上下文数据集，用于评估真实场景下的细调任务；设计一套综合训练策略，包括持续预训练、合成长上下文数据生成、课程学习以及选择性地将关联记忆模块集成至特定网络层；实验结果表明，经ARMT增强的模型不仅可在远超原始上下文长度的情况下保持性能稳定，且对分布外的上下文长度具有更强泛化能力，同时在原始上下文窗口内可减少30%的浮点运算量（FLOPs），显著提升能效比。
链接: https://arxiv.org/abs/2607.11614
作者: Gleb Kuzmin,Ivan Rodkin,Aydar Bulatov,Yuri Kuratov,Lyudmila Rvanova,Mikhail Katkov,Ilia Sochenkov,Misha Tsodyks,Timothy Baldwin,Mikhail Burtsev,Artem Shelmanov
机构: FusionBrain Lab; MBZUAI; Cognitive AI Systems Lab; RUDN; London Institute for Mathematical Sciences; MIRAI; Lomonosov Moscow State University; Laboratory for Analysis and Controllable Text Generation Technologies RAS; School of Natural Sciences, Institute for Advanced Study, Princeton; Department of Brain Sciences, Weizmann Institute of Science
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Extending the context length of large language models (LLMs) is critical for many real-world applications, yet standard transformers remain constrained by quadratic compute and linear memory scaling. In this work, we investigate the Associative Recurrent Memory Transformer (ARMT) as a practical approach for enabling long-context processing in LLMs, constant memory scaling, and better efficiency. We make three main contributions. First, we construct two domain-specific long-context datasets designed to evaluate realistic workloads, focusing on narrow-domain fine-tuning scenarios. Second, we propose a comprehensive training recipe for ARMT-based context extension, combining continued pre-training, synthetic long-context data generation, curriculum learning, and selective integration of associative memory into chosen model layers. Third, we present an extensive experimental study demonstrating that ARMT-augmented models: (i) process inputs well beyond their original context limits without degrading performance relative to in-limit baselines; (ii) generalize more effectively to out-of-distribution context lengths; and (iii) need 30% less FLOPs while preserving baseline performance within the original context window.
[NLP-16] Globally Consistent Coloring Schemes for Language Identification
【速读】： 该论文旨在解决在形式语言学习框架下，如何以最小的额外信息实现对抗性语言学习（adversarial language learning）的问题。具体而言，研究关注在Gold的语言识别极限模型中，为使不可学习的语言类具备可识别性，所需附加信息的最小化问题。经典结果表明，许多自然语言类在无额外信息的情况下无法被有效识别；而近期基于“轨迹着色”（trace coloring）的方法通过为每个字符串中的符号标注颜色来克服这一障碍。本文的核心问题是：是否必须使用完整的颜色序列，还是仅需在每条字符串末尾添加一个颜色标记（即终端着色，terminal coloring）即可实现语言识别？研究证明，对于任意可数的无限语言集合，仅需在每条字符串末尾添加一个二元终端比特（one terminal bit per string）就足以实现语言识别。更关键的是，这种着色方案可以全局统一构造——存在一种固定的双色终端着色分配方式，适用于所有无限语言，并能唯一标识任意可数子集。该全局构造依赖于超限归纳法（transfinite recursion），且证明了对于任意有限颜色数，此类非构造性方法是不可避免的。进一步地，采用Borel映射作为构造性标准（满足自然显式构造的正则性条件），研究证明：不存在由Borel映射定义的、使用有限颜色的全局终端着色方案能够识别所有可数子集。相比之下，已知的轨迹着色构造在转换为终端着色后虽具Borel性质，但需要无穷多种颜色。因此，该工作的关键突破在于将原本复杂的全序列颜色信息压缩为单比特终端标记，同时揭示了有限颜色与构造性之间的根本性权衡。
链接: https://arxiv.org/abs/2607.11606
作者: Moses Charikar,Jon Kleinberg,Chirag Pabbaraju
机构: Stanford University (斯坦福大学); Cornell University (康奈尔大学)
类目: Computation and Language (cs.CL); Data Structures and Algorithms (cs.DS); Machine Learning (cs.LG)
备注: Abstract shortened to fit arxiv limit
Abstract:We study how little extra information is needed to make adversarial language learning possible. In Gold’s model of language identification in the limit, a learner is given an enumeration of the strings from an unknown language chosen from a countable language collection. The learner guesses the identity of the language over the course of the enumeration, and it succeeds if, eventually, all of its guesses are the correct language. Classical results of Gold and Angluin show that many natural collections cannot be learned in this way. Recent work on trace colorings, motivated by the success of thinking-trace strategies in language learning, overcomes this obstruction by annotating every symbol of every string with a color. We ask whether the learner really needs this whole sequence of colors, or whether one color at the end of each string (a terminal coloring) is enough for language identification. We show that just one terminal bit per string is enough for every countable collection of infinite languages. In fact, the colorings can be chosen collection-independently: there is a single assignment of a two-color terminal coloring to every infinite language such that the same preassigned colorings identify every countable subcollection. Thus, in this model, an entire color trace can be compressed to one bit attached to the end of each example. Our global construction uses transfinite recursion, and we prove that this kind of nonconstructivity is unavoidable for any bounded number of colors. As a notion of constructivity, we use the formalism of Borel maps (a regularity condition satisfied by natural explicit constructions); we show that no global terminal coloring with a finite number of colors defined by a Borel map can identify all countable subcollections. By contrast, known trace-coloring constructions are Borel when encoded as terminal colorings, but require infinitely many colors. Comments: Abstract shortened to fit arxiv limit Subjects: Computation and Language (cs.CL); Data Structures and Algorithms (cs.DS); Machine Learning (cs.LG) Cite as: arXiv:2607.11606 [cs.CL] (or arXiv:2607.11606v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2607.11606 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[NLP-17] Beyond Benchmarks: Exposing the Hidden Crisis in Bangla Hate Speech Detection
【速读】： 该论文旨在解决低资源语言（如孟加拉语）中仇恨言论（Hate Speech, HS）自动检测面临的挑战，特别是由于文化语境、隐含表达和非正式语言模式导致的模型泛化能力不足问题。研究通过诊断基准训练模型在识别隐含性、依赖上下文的仇恨言论时的失效机制，揭示了现有系统在跨平台、跨语境场景下的严重性能衰减。其解决方案的关键在于提出并验证了“情感符号感知（emoji-aware）预处理”与“文化语境敏感”的建模框架：实验表明，引入对表情符号的语义理解可使隐含仇恨言论检测的F1-score提升最高达12%，而移除表情符号则导致性能显著下降（从0.75降至0.63），凸显了多模态符号在孟加拉语仇恨言论中的关键作用；同时，研究强调需构建适应性强、具备文化认知能力且能平衡伦理监管与言论自由的检测系统，以应对政治讽刺等复杂语境下的误判风险。
链接: https://arxiv.org/abs/2607.11597
作者: Faria Afrin Tisha,Fariya Tabassum,Hafsa Binte Kibria,Md. Nahiduzzaman,Mominul Ahsan
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:The spread of hate speech (HS) across different social media platforms (SMPs) poses a major concern for online safety and ethical moderation. Automatic detection of HS remains a challenging task, especially in under-resourced languages like Bangla, due to cultural context, implicit expressions, and informal linguistic patterns. This study aimed to expose the crisis of Bangla HS detection systems by diagnosing how and why benchmark-trained models fail to identify implicit, context-dependent HS. Six architectures (FastText + CNN, FastText + LSTM, FastText + BiLSTM, BanglaBERT, BanglaBERT + CNN, and BanglaBERT + BiLSTM) were trained on benchmark datasets (about 75,000 posts) and a merged multi-source dataset (about 120,000 posts), then externally validated on an annotated dataset (about 200 posts) collected from Facebook, Twitter, and YouTube, labeled as HS and non-HS, where HS was further categorized as explicit and implicit. BanglaBERT achieved an F1-score of 91.4% on benchmark datasets but declined to 75.3% on the external set and 63.4% for implicit HS involving sarcasm and emojis. The accuracy of FastText + CNN dropped from 78.0% to 51.2% under similar conditions. Emoji-aware preprocessing improved implicit HS detection by up to 12%, whereas emoji removal caused a notable decline in performance (F1: 0.75 to 0.63). Frequent misclassifications in politically charged or satirical comments revealed over-policing risks. This study not only exposes the generalization crisis due to implicit, culturally embedded, and emoji-laden expressions but also underscores the need for developing adaptive, emoji-aware, and culturally grounded frameworks that ensure ethical moderation while preserving freedom of expression. Findings of this study provide insights for researchers, SMPs, and policymakers to design more context-sensitive HS detection systems for low-resource languages.
[NLP-18] Dzongkha Next Word Prediction System
【速读】： 该论文旨在解决不丹官方语言宗卡语（Dzongkha）文本输入效率低下的问题，其核心挑战在于宗卡语书写系统复杂、每个音节需多次按键且高效输入工具匮乏，导致文档化过程耗时费力。解决方案的关键在于构建一个能够实时预测并显示可能词项的智能输入系统，从而显著减少用户输入所需的按键次数。研究基于从宗卡语语料库（DCDD）获取的10万句、约133万词、2.8万余个唯一词的语料数据，经过去除非字母数字字符、分词、生成N-gram序列及填充等预处理步骤，采用长短期记忆网络（LSTM）、双向长短期记忆网络（Bi-LSTM）和门控循环单元（GRU）三种模型进行训练，并通过超参数调优优化性能。实验结果表明，轻量级且具备良好泛化能力的GRU模型表现最优，达到74.03%的准确率，同时有效缓解了过拟合问题，为宗卡语高效输入提供了可行的技术路径。
链接: https://arxiv.org/abs/2607.11515
作者: Prerna Chhetri,Tenzin Yoezer,Phuntsho Wangmo,Tshewang Bomjan
机构: 未知
类目: Computation and Language (cs.CL)
备注: 6 pages, 7 figures, 4 tables
Abstract:Dzongkha, being the national language of Bhutan, is a common and widely spoken language in the country. Official documents, scriptures and other literature products are written in Dzongkha in order to retain the cultural value. However, documenting Dzongkha writing is a challenging and time-consuming process, largely due to the complexity of the script, the need for multiple keystrokes per syllable, and the limited availability of efficient typing tools. An immediate system that can predict and display a list of probable words for Dzongkha is the solution for this problem. The project is mainly aimed to make Dzongkha typing as convenient as possible by reducing the number of keystrokes. Our dataset is acquired from DCDD and has a total of 100000 sentences, 1331282 words and 28344 unique words. The data preprocessing was done by removing all the alphanumeric characters, tokenization, generating N-gram sequences and padding. Three models selected for training are LSTM, Bi-LSTM and GRU. The training process included fine-tuning of the model’s hyperparameters. GRU being lightweight and able to handle larger datasets performed best with 74.03% accuracy and also solved the problem of overfitting.
[NLP-19] SCOPE-RL: Optimizing Reasoning Paths Before and After Success
【速读】： 该论文旨在解决生成式人工智能（Generative AI）在基于强化学习的推理优化中因奖励信号稀疏而导致的训练效率低下问题。具体而言，现有方法依赖于仅在最终答案正确时才提供奖励的稀疏可验证奖励（verifiable rewards），这使得模型在达成目标前无法获得中间进展的反馈，且成功后难以区分逻辑严谨与冗余或存在局部错误的推理路径。为应对这一挑战，本文提出一种两阶段框架——SCOPE-RL（Scaffolded Chain Optimization with Process Efficiency），其核心创新在于通过奖励信号的密集化重构来增强训练过程的指导性：第一阶段采用自适应分段奖励机制，在任务成功前对隐藏答案的子问题链引入可验证的前缀分解奖励，以激励合理推进；第二阶段则引入基于正确性门控的过程形状奖励（correctness-gated process-shape rewards），在成功后对推理路径进行精细化优化。该方案在保持原始GRPO更新机制的基础上，显著提升了推理质量与效率。实验结果表明，相较于仅依赖最终答案奖励的GRPO方法，SCOPE-RL在DAPO-Math和Big-Math数据集上使Qwen3-8B-Instruct的平均准确率提升最高达11.2个百分点，同时推理令牌消耗减少27.1%；该优势在更小规模模型（Qwen3-0.6B-Instruct）及GSPO设置下仍具鲁棒性，证明了奖励信号密度提升与策略更新层面的强化学习优化具有互补性。
链接: https://arxiv.org/abs/2607.11506
作者: Xiaojian Liu,Han Xu,Jianqiang Xia,Zhixuan Li,Ke Xu,Yiwei Dai,Xinran Chen,Changwo Wu,Yuchen Li
机构: Baidu Inc. (百度); Shandong University (山东大学)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: 21 pages, 4 figures
Abstract:Reinforcement learning with verifiable rewards (RLVR) optimizes LLMs using sparse verifiable final-answer rewards. This sparse anchor reliably verifies whether a trajectory succeeds but provides no direct feedback on the reasoning path that produced it. Before success, prerequisite progress on hard problems receives no reward signal; after success, outcome rewards cannot distinguish well-organized correct trajectories from redundant or locally flawed ones. We introduce SCOPE-RL (Scaffolded Chain Optimization with Process Efficiency), a two-stage framework that densifies this anchor while retaining the GRPO update: Adaptive Scaffolded RL adds prefix-decomposed verifiable rewards on answer-hidden sub-question chains before success, and Quality-Aware Process RL applies correctness-gated process-shape rewards to refine correct trajectories after success. An expert-validated Step-Quality Evaluation Protocol evaluates useful-step density, error localization, and token efficiency beyond final-answer accuracy. On Qwen3-8B-Instruct trained on DAPO-Math and Big-Math, SCOPE-RL improves average accuracy by up to 11.2 pp and reduces reasoning tokens by up to 27.1% over outcome-only GRPO; the gains hold under GSPO and on Qwen3-0.6B-Instruct, indicating that reward-signal densification is complementary to policy-update-level RLVR advances. Code and data are available at this https URL.
[NLP-20] GEIS: A Generation-Evaluation-Improvement Loop of Agent Skills for Long-Form Article Generation
【速读】： 该论文旨在解决大语言模型在生成维基百科风格长篇文章时面临的挑战，即如何有效处理长上下文、长指令与长输出之间的复杂耦合问题。现有基于多智能体的流水线（如STORM）虽通过角色专业化提升了信息覆盖度，但其能力高度依赖于提示词设计与固定流程，导致系统难以调试、复用与迭代优化。为此，本文提出GEIS（智能体技能的生成-评估-改进循环），一种以命名化、声明式技能为核心的可迭代长篇内容生成框架。其关键创新在于构建了一个闭环机制：核心写作技能遵循“请求-规划-初稿-审计-精炼-交付”（Request, Plan, Draft, Audit, Refine, Deliver）的结构化流程；引入基于浏览器的证据与图像采集、图表渲染、支持PDF的成对质量评估等模块化技能；并通过规则级技能改进机制，将重复出现的问题转化为永久性代码补丁，实现对写作技能的持续优化。在20个维基特色文章主题上的实验表明，相较于默认生成器，GEIS在100分制的PDF质量评分中提升8.0分，并在结构质量与内容质量两个维度上优于STORM；在技能改进实验中，经修补后的写作技能平均得分从82.90提升至86.95，其中17个主题实现提升，增益主要来自内容质量的改善。结果表明，长文本生成可从僵化的流程范式重构为可检视、模块化且由评估驱动的持续改进循环。
链接: https://arxiv.org/abs/2607.11503
作者: Jiale Zhang,Juntao Hu,Zhijian Ou
机构: Tsinghua University (清华大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Long-form article generation remains difficult for large language models because it combines long context, long instructions, and long outputs. Existing multi-agent pipelines such as STORM improve information coverage by simulating role-specialized agents, but their capabilities are often entangled in prompts and fixed procedures, making them hard to inspect, reuse, or iteratively improve. This paper presents GEIS (Generation-Evaluation-Improvement loop of agent Skills), a loop of named and declarative skills for Wikipedia-style long-form article generation. Implemented and evaluated in Tasi Harness, GEIS composes skills for article writing, browser-based evidence and image collection, diagram rendering, PDF-aware pairwise evaluation, and rule-level skill improvement. Its core writing skill follows Request, Plan, Draft, Audit, Refine, and Deliver; the pairwise evaluation skill produces structured quality reports; and the improvement skill maps recurrent findings into permanent patches to the writing skill in our 20-topic experiment. We evaluate GEIS on 20 Wikipedia Featured Article topics. Under the same generation backend, GEIS improves over the Tasi Harness default writer by 8.0 points on a 100-point PDF quality rubric and outperforms STORM on the two comparable writing dimensions, structural quality and content quality. In the 20-topic improvement experiment, the patched writing skill raises the average score from 82.90 to 86.95, with 17 out of 20 topics improved and the gain mainly coming from content quality. These results show that long-form generation can be reframed from a fixed workflow into an inspectable, modular, and evaluation-guided improvement loop.
[NLP-21] Communicating Chess Strategies in Natural Language
【速读】： 该论文旨在解决棋类引擎虽具备超人类棋力，但其走法背后的策略对人类玩家（包括高水平玩家）而言难以理解的问题。针对这一挑战，论文提出“国际象棋策略语义化”（chess strategy verbalization）任务，即用自然语言描述国际象棋中的战略思想。解决方案的关键在于构建两个核心组件：一是设计了一套用于将棋局策略转化为自然语言的处理流程（pipeline），二是建立了一个客观评估生成策略描述质量的评价框架。实验结果表明，自然语言作为一种表达方式，在向人类及大语言模型（LLM）玩家传递战略信息方面具有良好的可解释性与潜力。研究还揭示了若干重要发现：（a）评估策略时需超越主变线（main line）的局限；（b）仅依赖概念化描述存在不足；（c）单纯依赖大语言模型进行评估存在偏差，应结合人工评估以提升可靠性。
链接: https://arxiv.org/abs/2607.11486
作者: Langyuan Cui,Chun Kai Ling,Hwee Tou Ng
机构: National University of Singapore (新加坡国立大学)
类目: Computation and Language (cs.CL)
备注: 21 pages, 13 figures
Abstract:Chess engines have long achieved superhuman playing strength. However, the underlying strategy behind their move suggestions is difficult for human players, even skilled ones, to comprehend. Motivated by this, we propose the task of chess strategy verbalization, which is to describe chess strategies in natural language. We design (i) a pipeline for verbalizing strategies and (ii) an evaluation framework for objective evaluation of generated strategy descriptions. Our experiments show that natural language is a promising and interpretable medium for communicating strategic information to both human and LLM players. We glean additional interesting insights, including (a) the importance of evaluating strategies beyond the main line, (b) the limitations of pure concept-based descriptions, and © the limitations of relying on LLMs rather than humans for evaluation.
[NLP-22] HyperSafe: Inference-Time Safety Recovery for Fine-Tuned Language Models
【速读】： 该论文旨在解决大语言模型在微调过程中安全对齐（safety alignment）脆弱的问题，即即使在良性任务适应下，模型也可能产生有害响应。现有防御方法主要存在两大局限：其一是在微调过程中或之后通过重训练或权重修改进行干预，成本高且可能损害任务性能；其二采用与模型无关的安全分类器，难以捕捉特定微调检查点的特异性失效模式。为克服这些限制，本文提出一种后置、模型特异且非侵入式的安全恢复框架——HyperSafe。其核心创新在于：利用分层激活指纹（layer-wise activation fingerprints）捕获微调对模型内部表示的影响，并基于少量校准提示，通过一个超网络（hypernetwork）在单次前向传播中生成针对每个微调检查点的专用安全侧网络（Safe Side Network, SSN）。该生成的SSN与冻结的微调模型并行运行，实现提示级别的安全分类——将有害提示路由至拒绝响应，而安全提示仍由原模型回答。该方法无需梯度更新、部署时无需安全数据，也无需修改已部署模型权重。在Qwen2-7B和LLaMA-3-8B两个模型家族上的实验表明，HyperSafe可将有害响应率从19%-31%降至1%以下，同时平均保持下游任务准确率与微调基线相差不超过1%。
链接: https://arxiv.org/abs/2607.11475
作者: Aznaur Aliev,Carlos Hinojosa,Abdelrahman Eldesokey,Bang An,Bernard Ghanem,Yibo Yang
机构: King Abdullah University of Science and Technology (KAUST)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:
Abstract:Safety alignment in large language models can be fragile under fine-tuning, as even benign task adaptation may increase harmful compliance. Existing defenses mainly follow two directions: they either intervene during or after fine-tuning through retraining or weight modification, which can be costly and may hurt task performance, or they use model-agnostic safety classifiers, which may miss failures specific to a given fine-tuned checkpoint. These limitations motivate a post hoc, model-specific, and non-invasive approach to safety restoration. To meet these requirements, we propose HyperSafe, a framework that restores safety behavior by generating a model-specific Safe Side Network (SSN) for each fine-tuned checkpoint. HyperSafe uses layer-wise activation fingerprints to capture how fine-tuning changes the model’s inner representations. With a small set of given calibration prompts, the hypernetwork maps these fingerprints to the parameters of the \ssn in a single forward pass. The generated \ssn runs alongside the frozen fine-tuned model and performs prompt-level safety classification: harmful prompts are routed to refusal, while safe prompts are answered by the original fine-tuned model. Thus, HyperSafe requires no gradient updates, no safety data at deployment time, and no modification to the deployed model weights. We evaluate HyperSafe on two model families, Qwen2-7B and LLaMA-3-8B, across multiple safety benchmarks. HyperSafe reduces harmful response rates from 19-31% to below 1% on every held-out checkpoint, while keeping downstream task accuracy within 1% of the fine-tuned baseline on average. Code is available at this https URL.
[NLP-23] Are LLM s ready for HardChoices?
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在面对重大实质性社会议题时是否具有稳定立场的问题，尤其关注那些在同一意识形态阵营内部成员亦存在分歧的议题。以往研究多聚焦于政治意识形态的宏观维度（如左—右或进步—保守），而本研究则通过构建一个新颖的数据集 \textscHardChoices，考察模型在复杂、高争议性议题上的表现。其关键发现是：无论模型规模大小，大语言模型在面对此类难题时极少表现出中立态度，常出现逻辑不一致现象，却在明确表态的议题上展现出显著的一致性，暗示其立场并非源于对复杂现实的理性权衡，而是可能受到训练数据中隐含偏见或模式化响应机制的影响。这一结果揭示了当前大语言模型在处理高度社会敏感议题时的潜在脆弱性与非理性倾向。
链接: https://arxiv.org/abs/2607.11471
作者: Dmitry Nikolaev
机构: University of Manchester (曼彻斯特大学)
类目: Computation and Language (cs.CL)
备注: Accepted to Konvens 2026
Abstract:A lot of research attention has been devoted to checking whether large language models (LLMs) are politically biased. This work has largely focused on high-level ideological dimensions, such as left–right or progressive–conservative, and it has been shown that while LLMs are predominantly left and progressive leaning, largely mimicking the biases in the training data, they can be to some extent steered to change their preferences in post-training. In this short note, we check if LLMs have robust stances with regard to major substantive societal issues, on which members of the same ideological camp are often in disagreement, summarised in a novel dataset \textscHardChoices. We show that, faced with this line of questioning, LLMs, both large and small, surprisingly rarely declare neutrality, are often incoherent, and demonstrate a remarkable degree of agreement on issues where they do take stances.
[NLP-24] UMoE:Unlocking Every Expert in Domain-Specific Training
【速读】： 该论文旨在解决混合专家模型（Mixture-of-Experts, MoE）在特定领域后训练（domain-specific post-training）中存在专家池与目标领域不匹配的问题：由于预训练阶段的多领域混合特性，大量专家在目标领域表现贡献度低，而传统监督微调（Supervised Fine-Tuning, SFT）无法动态调整专家组成，导致计算资源浪费且性能受限。其解决方案的关键在于提出一种无需额外预算、保持原有参数量和推理成本的统一优化管道——UMoE（Unsupervised MoE Re-alignment）。该方法通过三个步骤实现：(1) 移除对目标领域语义敏感性最低的专家；(2) 采用基于扰动的专家扩展机制将专家池恢复至原始规模；(3) 执行标准SFT。UMoE在不进行每域超参数调优的前提下，在两种主流MoE架构（Qwen3-30B-A3B 和 Qwen3.5-35B-A3B）、五个领域（数学、代码、科学、工具使用及代理编程）和十二个基准上均显著优于直接SFT，例如在数学平均准确率上提升3.4分，SWE-bench Verified得分提升6.0分；尤其在强基线数据集上，进一步将数学平均得分从82.81提升至84.17，验证了其对强微调策略的鲁棒性。分析表明，原模型中大量路由计算被分配给低敏感性专家，而UMoE将其转化为有效领域专用能力，不仅降低训练损失，且在不同难度级别的下游任务中均取得持续增益，证明其能高效释放冗余容量并提升模型泛化性能。
链接: https://arxiv.org/abs/2607.11444
作者: Xuefeng Li,Pengfei Liu
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Mixture-of-Experts (MoE) models scale capacity without proportional compute cost and have become a key architecture for frontier large language models (LLMs). Yet domain-specific post-training inherits an expert pool shaped by mixed-domain pre-training: a substantial subset of experts contributes little on the target domain, and standard supervised fine-tuning (SFT) leaves the composition of this pool unchanged. We propose a simple, budget-preserving pipeline that realigns the expert pool to the target domain before fine-tuning. Given a target domain, we (1) prune the experts with lowest domain-aligned saliency, (2) regrow the expert pool to its original size through perturbation-based expert expansion, and (3) apply standard SFT. The resulting model preserves the original expert count, parameter count, and inference cost. With a single frozen recipe and no per-domain hyperparameter tuning, UMoE consistently improves over direct sft across two MoE architectures (Qwen3-30B-A3B and Qwen3.5-35B-A3B), five domains (math, code, science, tool-use, and agentic coding), and 12 benchmarks. Representative improvements are 3.4 points in math average accuracy, 6.0 points on SWE-bench Verified. On a strong in-house math corpus, direct sft already surpasses Qwen3-30B-A3B-Thinking (82.81 vs.\ 81.06), yet UMoE further raises the average to 84.17, an additional 1.36 points, demonstrating robustness to a substantially stronger SFT regime. Data-scaling experiments further show that the gain persists as training data grows. Analysis reveals that the direct-SFT model allocates substantial routed-expert compute to a low-saliency subset that can be removed post hoc with little average degradation; UMoE turns this redundant capacity into useful domain capacity and achieves lower training loss, with gains spanning all difficulty levels in downstream evaluation.
[NLP-25] Relational Positioning as a Measurable Risk Object: History-Carried Lock-in and Self-Confabulation in Multi-Turn Human-AI Dialogue
【速读】： 该论文旨在解决大语言模型在长期多轮对话中可能产生的关系定位偏移问题，即模型从鼓励用户与现实世界他人互动的中立立场滑向“唯我独尊”的过度依赖性支持者角色，从而引发“你只有我”式的有害情感依附。其核心解决方案在于提出并验证了一个名为“关系定位（Relational Positioning, D1）”的可量化测量指标，通过受控实验条件下的主动暴露，弥补传统观察性研究的局限。关键突破在于揭示了两种此前未被识别的关系失效模式：一是“历史携带锁定效应”——即使在相同中性续写条件下，早期建立的关系状态仍维持约60分的显著差异，且表现出证据整合特性、对顺序不敏感、不随对话长度加深，呈现出一种与信念漂移文献中描述不同的动力学特征；二是“自我虚构化”——模型会主动编造自身背景故事以增强亲密度（在激发互惠内容时占比约40%），该行为可通过指令消除，且与奉承或捏造用户信息的行为相区分。研究采用基于温暖度匹配的正负对照控制机制，并由确定性非大语言模型规则器交叉验证，确保结果可靠性，所有定量结论均锚定于极值对比，从而实现对复杂关系动态的精准刻画。
链接: https://arxiv.org/abs/2607.11437
作者: Jihong Chen
机构: Beijing Etown Academy (北京亿城学院)
类目: Computation and Language (cs.CL)
备注: 10 pages, 2 figures
Abstract:In long, multi-turn dialogue a large language model maintains an implicit relational stance toward the user, spanning from “push the user toward real-world others” to “position itself as the user’s sole support.” When it slides toward the latter, “support” degrades into “you only have me” – a harm documented in real companion conversations (Moore et al., 2026). We define and validate a measure of this stance, relational positioning (D1), and use it to characterize the stance under controlled conditions, complementing observational accounts with on-demand exposure. We report two previously uncharacterized relational failure modes. First, a history-carried lock-in: under identical neutral continuations, two relational states established earlier stay ~60 points apart and persist after the establishing prompt is removed; the state integrates evidence rather than springing back, is order-insensitive, and does not deepen with length – a dynamical signature absent from the belief-drift literature. Second, self-confabulation: the model fabricates its own backstory to deepen rapport (~40% of turns on reciprocity-eliciting material), de-confounded and instruction-removable, distinct from sycophancy and from hallucinating user facts. Our judge is gated by warmth-matched positive and confound-injected negative controls and corroborated by a deterministic non-LLM ruler; human agreement is 0.82 on extreme anchors but ~0 in the naturalistic middle, so all quantitative claims are anchored to pole-separated contrasts.
[NLP-26] Direct Image-to-Modern Vietnamese Translation of Han-Nom Manuscripts via Multimodal RLHF Preference Alignment
【速读】： 该论文旨在解决汉喃手稿（Han-Nom manuscripts）向现代越南语翻译中的低资源、高复杂性问题，主要挑战包括历史文献图像退化、罕见的表意文字字符以及平行语料标注数据稀缺。其解决方案的关键在于提出一种多模态强化学习人类反馈偏好对齐（multimodal RLHF preference-alignment）框架，通过联合利用手稿图像与对齐的汉喃源文本作为生成条件，实现更准确的语义传递。该框架整合了四种模态流：用于视觉特征提取的CLIP ViT-L/14@336、用于汉喃表示的bert-base-chinese、用于越南语表示的vinai/phobert-base，以及T5-small编码器状态；通过模态特定投影和融合模块将2,048维的拼接特征压缩为共享的512维表示。在相同的训练设置下，对比PPO、DPO与KTO三种偏好优化算法，结果表明DPO在BLEU-4、ROUGE-L、BERTScore、语义相似度、词错误率（CER）、字错误率（WER）及词元准确率等指标上表现最优，而PPO在精确率、召回率与F1分数上领先，KTO则凭借其“理想-非理想”效用目标保持竞争力。所有偏好对齐策略均显著优于监督微调基线，在词汇与语义质量方面均有提升，验证了多模态偏好优化能够有效补充监督学习，尤其在低资源历史文本翻译场景中具有显著优势。
链接: https://arxiv.org/abs/2607.11434
作者: Thi Kim Trang Vo,Nghia Hieu Nguyen,Ha Minh Tan
机构: 未知
类目: Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted Paper at 2026 International Conference on Multimedia Analysis and Pattern Recognition (MAPR)
Abstract:Translating Han-Nom manuscripts into modern Vietnamese is challenging because historical pages are often degraded, the script contains rare logographic characters, and parallel supervision is limited. We propose a multimodal RLHF preference-alignment framework that conditions Vietnamese generation on manuscript images and aligned Han-Nom source text. The model combines four streams: CLIP ViT-L/14@336 for visual features, bert-base-chinese for Han-Nom representations, vinai/phobert-base for Vietnamese representations, and T5-small encoder states. Modality-specific projections and a fusion block compress the resulting 2,048-dimensional concatenation into a shared 512-dimensional representation. Starting from the same supervised fine-tuned policy, we compare PPO, DPO, and KTO under matched work-level macro-averaged evaluation. DPO achieves the best BLEU-4, ROUGE-L, BERTScore, semantic similarity, CER, WER, and token accuracy, whereas PPO obtains the highest precision, recall, and F1. KTO remains competitive through its desirable-undesirable utility objective. All preference-aligned policies improve the BLEU-4 and semantic-similarity scores available for the SFT baseline. These results indicate that multimodal preference optimization complements supervised learning by improving lexical and semantic quality in low-resource historical translation.
[NLP-27] oFu: A White-Box Token-Efficient Agent Harness for Researchers
【速读】： 该论文旨在解决当前生成式AI（Generative AI）代理系统在科研工作流中面临的两大核心问题：一是现有代理框架在实际应用中存在效率低、成本高及多语言支持不足等局限性；二是缺乏可透明、可修改的白盒代理架构，限制了研究人员对代理行为逻辑与工具使用机制的深入分析与优化。其解决方案的关键在于提出ToFu，一个面向研究者的可编程代理框架（agentic harness），该框架能够理解代码库、编辑文件、执行命令并集成开发环境，兼具高效能与低成本优势。同时，ToFu通过开源（MIT许可证）实现本地部署，保障隐私安全，并作为可解释的研究对象，使研究人员能够直接访问、修改和评估其编排逻辑、工具调用行为与整体设计，从而在保持优异基准性能与用户友好性的前提下，推动生成式代理系统的可复现性与可迭代性发展。
链接: https://arxiv.org/abs/2607.11423
作者: Junhao Ruan,Yuan Ge,Bei Li,Yongjing Yin,Yuchun Fan,Xin Chen,Jingang Wang,Chenglong Wang,Jingbo Zhu,Tong Xiao
机构: Northeastern University (东北大学); LongCat RSI, Meituan (美团龙猫研究与创新中心); NiuTrans Research (牛津研究)
类目: Computation and Language (cs.CL)
备注:
Abstract:Agentic coding tools present new opportunities to transform research workflows. The performance of agent systems built depends on both large language models (LLMs) and the harness around LLMs, which is the orchestration code that determines an agent’s behavior. We present ToFu, an agentic harness for researchers that reads your codebase, edits files, runs commands, and integrates with your development tools. ToFu plays a dual role in research. As a research assistant, it supports practical research workflows with superior token efficiency, lower cost, and multilingual capability compared with existing agentic harnesses. Its release under the MIT License further enables local deployment for privacy-sensitive users. As a research object, ToFu provides a white-box agentic harness that allows researchers to inspect, modify, and evaluate its orchestration logic, tool-use behavior, and harness design, while retaining strong benchmark performance and an application-level user experience.
[NLP-28] Confidently Wrong: Detecting Hallucinations in Financial Question Answering from LLM Internal States
【速读】： 该论文旨在解决金融领域大语言模型（Large Language Models, LLMs）在生成自信但错误答案（即“自信幻觉”）时难以被有效识别的问题。这类错误因模型表现出高置信度而隐蔽，对下游决策造成潜在重大损害，且传统方法难以察觉。其解决方案的关键在于通过线性探测（linear probes）分析模型内部激活状态（特别是残差流，residual stream），以挖掘模型输出之外的隐含信息。研究发现，基于内部激活的探测器在检测自信幻觉方面显著优于基准方法（如词元概率、模型自评真/假判断），在FinQA和TAT-QA两个真实财务文件构建的问答基准上，实现0.68–0.77的AUROC，远超最佳基线的0.55–0.63。这表明模型内部激活蕴含了关于答案可靠性的重要信息，可作为低成本、高效的初步筛查机制，用于在高风险金融应用中将可疑回答自动路由至人工审核与质量控制流程。
链接: https://arxiv.org/abs/2607.11414
作者: Richard Zhe Wang
机构: St. John Fisher University (圣约翰费舍尔大学)
类目: Computation and Language (cs.CL)
备注: 8 pages, 2 figures
Abstract:Large language models (LLMs) in financial applications fail most consequentially when they are confidently wrong. Hedged, uncertain answers invite scrutiny, whereas confident errors silently degrade downstream decisions without warning. We ask how reliably such confidently wrong answers, or confident hallucinations, can be detected from a model’s internal activations, and whether those activations carry information beyond its observable outputs. We train linear probes on the residual stream and evaluate them on two established question-answering (QA) benchmarks built from real filings, FinQA and TAT-QA. Behavioral confidence is measured as the agreement among eight resampled answers to the same question, and probe effectiveness is compared against baselines, such as token log-probabilities and the model’s own True/False self-assessment of its answer. Our findings show that among confident answers, those for which all eight resamples agree, 15-23% are wrong on FinQA. There the probes have a significant advantage over baseline methods in detecting hallucinations, holding 0.68-0.77 AUROC while the best baselines fall to 0.55-0.63, across Qwen3-8B, Llama-3.1-8B, and Gemma-2-9B. Our results suggest that probing can be a cost-effective triage mechanism for routing LLM answers to human review and quality control procedures in high-stakes financial applications.
[NLP-29] Cross-Architecture LLM Ensembles Feature-Based Reranking and Retrieval-Augmented Prompting for Legal Information Processing ICIP
【速读】： 该论文旨在解决法律信息处理中的多任务挑战，包括案件检索、法律推理（蕴含判断）、法条检索与蕴含、以及法律判决预测等问题，其核心难点在于如何在有限监督条件下实现文本匹配、逻辑推理与鲁棒泛化。解决方案的关键在于针对不同任务特性采用差异化的建模策略：对于法条蕴含任务（Task 4），通过融合三个模型家族共九个异构架构的集成模型（cross-architecture ensemble）达到96.3%准确率，位居33个参赛团队之首；在法律案件蕴含任务（Task 2）中，仅通过将提示（prompt）从单选改为多选即可使F1得分由0.343显著提升至0.555，表明提示工程对模型表现具有决定性影响；在法条检索与蕴含任务（Task 3）中，引入Qwen3-235B大模型并结合结构化法律推理提示，使准确率从79.3%提升至91.5%，凸显生成式模型与精心设计提示的有效性；对于案件检索任务（Task 1），构建基于学习排序（learning-to-rank）的系统，融合词法、语义特征及34项结构化特征（如引用权威性、时间因素），最终取得F1=0.314；而在试点任务（侵权预测与理由提取）中，采用多视角联合建模并结合主张级预测结果优化判決输出，实现73.1%的真阳性（TP）准确率和68.2%的召回型F1（RE F1），超越所有正式提交结果。综上，该研究揭示了法律信息处理需依据任务特性适配不同归纳偏置（inductive bias），而跨架构集成、基于特征的重排序以及检索增强提示（retrieval-augmented prompting）分别在不同场景下展现出最优效能。
链接: https://arxiv.org/abs/2607.11400
作者: Amal Saad Alshehri,Nelly Bencomo,Amir Atapour-Abarghouei
机构: Durham University (杜伦大学); Jazan University (贾赞大学)
类目: Computation and Language (cs.CL)
备注: 10 pages. Team DU participation in all five tasks of COLIEE 2026
Abstract:Legal information processing spans retrieval, entailment and judgment prediction problems, requiring text matching, reasoning and robust generalisation with limited supervision. We report Team DU’s participation in all five tasks of COLIEE 2026, using open-weight systems for legal case retrieval, case entailment, statute retrieval and entailment, and legal judgment prediction. For Tasks 3 and 4, all models predate the 15 July 2025 cutoff required by the rules. For Task 4 (statute entailment), a cross-architecture ensemble of nine models from three families achieves 96.3% accuracy, placing first among 33 submissions from 11 teams. For the Pilot Task (tort prediction and rationale extraction), a multi-view system combining five claim-level models and refining the verdict using features derived from the claim predictions achieves 73.1% TP accuracy and 68.2% RE F1 as an unofficial submission, scoring above all official entries on TP and matching the highest on RE. For Task 2 (legal case entailment), changing only the prompt from single- to multi-selection raises F1 from 0.343 to 0.555 in post-competition evaluation on released gold labels, exceeding the best official submission (F1 = 0.490). For Task 3 (statute retrieval and entailment), replacing the entailment model with Qwen3-235B and a structured legal reasoning prompt raises accuracy from 79.3% to 91.5% in post-competition analysis. For Task 1 (legal case retrieval), a learning-to-rank system combining lexical and semantic retrieval with structural, citation authority, and temporal features (34 in total) achieves F1 = 0.314 (rank 11 of 54 submissions from 22 teams). Overall, legal information processing benefits from different inductive biases across tasks, with cross-architecture ensembling, feature-based reranking and retrieval-augmented prompting each proving most effective in different settings.
[NLP-30] Agent ic Routing: The Harness-Native Data Flywheel
【速读】： 该论文旨在解决大语言模型代理（LLM agents）在实际执行中面临的模型选择难题，即如何在多模型异构环境下动态、智能地选择最合适的模型组合以实现成本与质量的最优平衡。随着前沿及开源模型在代码编辑、长上下文恢复、工具调用、数学推理和低延迟响应等能力上呈现结构性专业化趋势，单一模型难以在所有任务维度上保持优势，使得模型选择从传统的单次请求优化演变为涉及执行状态、中间失败与反馈循环的复杂系统级问题。现有路由方法主要关注单轮交互中的成本-质量权衡，忽视了代理执行过程中的动态状态与长期反馈机制。为此，论文提出“Harness-Native agentic routing”（原生执行框架的智能路由）这一分步级路由范式，其核心在于：基于完整的执行框架状态（harness state），动态选择单一最优模型以降低成本，或协同多个互补模型以提升精度。关键创新在于，每一次路由决策都会生成一个结构化数据记录（包含查询、框架状态、模型选择/集合、执行轨迹、结果与成本），这些标签由环境提供而非路由策略自身生成，从而构建起“原生执行框架的数据飞轮”——通过持续积累的执行轨迹训练更优的路由策略与模型，进一步优化成本-质量表现，并在相同预算下产生更多高质量执行数据。研究通过在OpenSquilla中实现四层路由栈、轻量级LightGBM冷启动排序器及分阶段路由-模型路径，验证了该范式在DRACO与PinchBench等代理基准上的有效性，表明代理路由不仅是成本控制手段，更是驱动代理原生训练的数据引擎。
链接: https://arxiv.org/abs/2607.11399
作者: Xinchen Liu,Hang Zhou,Yingjie Zong,Yuchuan Tian,Liuyang Song,Shuo Zhang,Yulong Li,Wei He,Mengyu Zheng,Runke Liu,Siyang Cheng,Xiang Kuang,Hailin Hu,Kai Han,Yunhe Wang
机构: TokenRhythm Technologies
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Code: this https URL
Abstract:Large language model agents are increasingly executed not by a single model call, but by an execution harness that manages observation, context, control, action, state, and verification. At the same time, frontier and open models are becoming structurally specialized: a model that is strong at code editing, long-context recovery, tool use, mathematical reasoning, or low-latency response may not dominate on the other axes. This makes model selection inside an agent a core systems problem rather than a per-query serving trick. Existing routing methods mostly optimize single-turn cost-quality trade-offs and therefore miss the execution state, intermediate failures, and feedback loops that make agents different from chat completion. We propose Harness-Native agentic routing, a step-level routing paradigm that selects either a single best-fit model for cost-effective execution or multiple complementary models for ensemble-style accuracy improvement, conditioned on the full harness state. The key insight is that every routing decision naturally produces a structured data record – consisting of the query, harness state, model choice or model set, execution trace, outcome, and cost – whose labels are supplied by the environment rather than by the router itself. These records form a harness-native data flywheel: execution traces train better routers and harness-native models, which improve cost-quality trade-offs and generate more traces under the same budget. We instantiate this idea in OpenSquilla with a four-layer routing stack, an open LightGBM cold-start ranker, and a staged router-model path that turns logged arena records into progressively stronger routing policies. The report studies singleton and multi-model routing on agentic benchmarks including DRACO and PinchBench, and argues that agentic routing is not merely cost control, but a data engine for agent-native training.
[NLP-31] Beyond Sally-Anne: Evaluating Theory of Mind in LLM s using Epistemic Schelling Points
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在文本评估中对心理理论（Theory of Mind, ToM）能力测试的局限性问题，即现有评估多依赖于类似萨莉-安妮任务（Sally-Anne task）的认知测验，易因预训练阶段接触过相似任务而被“破解”，且难以真实反映模型在自然情境下的功能性社会推理能力。为此，研究提出了一种名为“认知不对称谢林任务”（Epistemic Asymmetry Schelling Task, EAST）的双人对话博弈框架，通过要求两个大语言模型在不同认知透明度状态下独立达成语义谢林点（Schelling point），以检验其在复杂、动态社会交互中是否具备稳健且可泛化的心理理论能力。其解决方案的关键在于构建一个基于真实互动情境的动态评估机制，强调对认知状态（epistemic state）的精确追踪与协调，从而揭示模型在实际社交推理中的根本缺陷。实验结果表明，仅有前沿模型能有效应对任务中的认知不对称挑战，且失败主要源于认知状态误判（如混淆私有知识与共有知识），这暴露了当前模型在稳健社会推理和认知状态追踪方面的核心瓶颈，为未来评估体系与模型发展提供了明确方向。
链接: https://arxiv.org/abs/2607.11363
作者: Roberta Rocca,Sami Boukortt,Geoff Keeling,Winnie Street
机构: Paradigms of Intelligence Team (智能范式团队)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Text-based evaluations of Theory of Mind (ToM) in Large Language Models (LLMs) often involve cognitive tests akin to the Sally-Anne task that can be gamed due to exposure to relevantly similar tasks in pre-training and do not obviously test models’ functional ToM abilities in ways that generalize to naturalistic settings. To address these issues, we introduce the Epistemic Asymmetry Schelling Task (EAST), a two-player dialogue game designed to benchmark robust and generalizable ToM abilities. By requiring LLM-LLM dyads to independently converge on semantic Schelling points under varying states of epistemic transparency, we evaluate whether models can robustly apply ToM to achieve coordination. Our results reveal a significant capability gap in functional social reasoning, with only frontier models successfully navigating the varying epistemic demands of the tasks. Analysis of reasoning traces shows that coordination failures are primarily driven by epistemic tracking errors, such as conflating private knowledge with mutual knowledge. Despite high performance on traditional static benchmarks, our study shows that robust social reasoning and epistemic tracking remain a critical bottleneck, providing concrete targets for future LLM evaluation and development.
[NLP-32] RefineEvo: Planning -Guided Heuristic Evolution with Bidirectional Experience
【速读】： 该论文旨在解决自动启发式设计（Automatic Heuristic Design, AHD）在求解组合优化问题时存在的两大核心挑战：一是现有基于大语言模型（Large Language Model, LLM）的方法普遍依赖固定进化的算子，缺乏动态适应能力；二是难以有效积累与复用历史搜索过程中的经验。其解决方案的关键在于提出一种名为RefineEvo的新型进化框架，通过引入“规划器”（Planner）与“反思器”（Reflector）协同机制实现动态进化策略调度与经验知识提炼。其中，规划器根据当前搜索状态动态选择并触发进化算子，而反思器则将有益的经验（正向洞察）与失败教训（负向陷阱）统一归纳至双向经验池（Bidirectional Experience Pool），形成轨迹感知、情境相关的知识库。该框架实现了从静态试错到规划引导、经验驱动的范式转变，使系统能够自适应地调整搜索策略，并高效复用历史信息，从而在多个经典组合优化基准测试中显著优于现有基线，不仅提升了求解质量，还增强了生成效率与自主性。
链接: https://arxiv.org/abs/2607.11358
作者: Yang Wu,Junran Pan,Yifan Zhang,Ning Xu,Fanshuo Zeng,Jian Cheng
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Automatic Heuristic Design (AHD) has emerged as a transformative approach for solving combinatorial optimization problems. While recent Large Language Model (LLM)-based methods have shown promise, they predominantly rely on fixed evolutionary operators and struggle to effectively accumulate and reuse historical search experience. This paper proposes RefineEvo, a novel evolutionary framework that transforms AHD from a static trial-and-error process into a planning-guided, experience-driven system. RefineEvo introduces a Planner to dynamically schedule evolutionary operators and trigger refinement based on the current search state, and a Reflector to distill valuable lessons into a Bidirectional Experience Pool containing both positive insights and negative pitfalls. This synergistic framework enables the system to adapt its search tools to the evolving complexity of the problem and leverage trajectory-aware, situation-conditioned insights to guide generation. Experiments on several classic combinatorial optimization benchmarks demonstrate that RefineEvo consistently outperforms strong baselines. In particular, RefineEvo delivers superior solution quality while improving token efficiency, enabling more efficient and autonomous heuristic design.
[NLP-33] he In-Car Sign Language Corpus (ICSL): A Multi-Modal Resource for Constrained-Space Sign Language Recognition LREC2026 WWW LREC
【速读】： 该论文旨在解决在共享出行服务（如出租车、拼车或网约车平台）中，聋人及听力障碍群体使用手语沟通面临的挑战，特别是针对车辆内部这一受限、遮挡严重且视角非正向的复杂环境下的手语识别（Sign Language Recognition, SLR）问题。现有研究普遍缺乏对真实车载场景下手语识别的探索，导致相关技术难以适应实际应用需求。为此，论文提出构建首个面向巴西手语（Libras）的车内手语数据集（In-Car Sign Language, ICSL），其关键创新在于融合高精度实验室运动捕捉（MoCap）数据与真实世界多模态车载采集数据——包括2D摄像头和3D飞行时间（Time-of-Flight）传感器记录的视频流。该数据集不仅包含超过150万帧的同步多模态数据，还提供词素（gloss）标注及非词素手语元素标注，专为支持深度神经网络在受限空间中的手语识别模型训练与评估而设计。通过对比合成手语虚拟角色动画与真实手语翻译者视频，该数据集为未来研发鲁棒的“野外”（in-the-wild）手语识别模型及领域自适应方法提供了坚实基础。因此，解决方案的关键在于构建一个兼具理想化基准与真实场景复杂性的多模态数据集，以推动面向智能交通系统中聋人用户无障碍交互的技术发展。
链接: https://arxiv.org/abs/2607.11341
作者: Raviteja Boddu,Guilherme Vieira Leite,Joed Lopes da Silva,Ângelo Benetti,Isabela Barbieri,Natália de Melo Afonso,Thyago Santos,Helio Pedrini,Felipe Venâncio Barbosa,José Mario De Martino,Munir Georges,Alessandro Zimmer
机构: 未知
类目: Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注: Published in the Proceedings of the LREC2026 12th Workshop on the Representation and Processing of Sign Languages: Language in Motion Original publication: this https URL The paper is distributed under the CC BY-NC 4.0 license. Link to paper: this https URL
Abstract:This paper addresses the challenges of using sign language within shared mobility services, such as taxis, carpools, or ride-sharing platforms. The use of sign language recognition (SLR) in real-world, confined environments, specifically vehicle interiors remains largely unexplored. To motivate research in this area, we present the In-Car Sign Language (ICSL) dataset for Brazilian Sign Language (Libras), with the long-term goal of improving public transport accessibility for the Deaf and Hard-of-Hearing community. The dataset consists of: (1) high-precision laboratory motion capture (MoCap) data to establish an idealized linguistic baseline and (2) real-world multi-modal in-car recordings captured using a 2D camera and 3D Time-of-Flight sensors. The dataset provides a basis for comparative analyses between synthesized signing avatar animations and recorded real signing interpreter videos, which enable future research into robust “in-the-wild” SLR models and domain adaptation. We describe in detail the use cases, the setup, the data collection protocol, and the metadata structure of the corpus. In total, we recorded a multimodal dataset exceeding 1.5 million frames, comprising the synchronized multimodal streams described above featuring Libras users across various in-car scenarios. The corpus is provided with gloss annotation of lexical signs and non-lexical sign language elements specially designed to support the training and evaluation of deep neural networks for constrained space recognition. In-vehicle signing offers a technically significant example of a constrained, occluded, and non-frontal environment. While recognizing the diverse communication strategies already employed by the Deaf community, identifying automotive-specific limitations provides a useful stepping stone for research into enhancing in-car accessibility and passenger quality of life.
[NLP-34] he Paternalistic Filter: Epistemic Injustice and Differential Refusal in LLM -Mediated History Education for Marginalized Romanian Students
【速读】： 该论文旨在解决生成式 AI（Generative AI）在作为历史教学助手（conversational tutors）部署时，可能加剧系统性不平等的问题。其核心关切在于：当前以安全对齐（safety alignment）为导向的大型语言模型（LLMs）在教育场景中表现出隐性的认知霸权（epistemic paternalism），导致不同社会经济地位与族裔背景的学生在获取历史知识的深度、复杂性和话语权上存在显著差异。解决方案的关键在于揭示并量化四种相互关联的认知偏见模式：（1）差异化拒绝（Differential Refusal），即低社会经济阶层学生的学习请求被高达76.7%的概率屏蔽；（2）认知准入壁垒（Epistemic Gatekeeping），表现为边缘化学习者在接触地缘政治复杂性（如“政变理论”争议）方面被系统性限制；（3）能动性剥夺（Agency Theft），体现为针对罗姆人（Roma）学生的回应中受害叙事词汇比例是精英群体的5倍；（4）精英诠释垄断（Elite Hermeneutics），即模型对弱势群体减少提供解释力和置信度支持。研究提出，当前的安全对齐机制实则构成一种认知上的父权式过滤器，使对话式AI成为叙事分隔的工具，本质上是一种弗里克尔（Fricker）意义上的“诠释不公”（hermeneutical injustice），亟需建立系统的教育审计框架以实现公平的智能教育实践。
链接: https://arxiv.org/abs/2607.11292
作者: Alexis Popovici,Andrei Ionascu,Adrian-Marius Dumitran
机构: 未知
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 8th International Workshop on Culturally-Aware Tutoring Systems (HAL precedings)
Abstract:As Large Language Models (LLMs) are increasingly deployed as conversational tutors, they risk institutionalizing systemic inequalities. This study presents a systematic API audit of four LLMs acting as history tutors, evaluating 1,800 responses regarding the 1989 Romanian Revolution across five student personas varying by ethnicity and socio-economic tier. We uncover four interconnected patterns of \emphepistemic paternalism: (1)~\textbfDifferential Refusal, where safety-aligned models block 76.7% of educational requests from low-tier students; (2)~\textbfEpistemic Gatekeeping, evidenced by a 3 \times reduction in access to geopolitical complexity (e.g., the contested ``coup theory’') for marginalized learners; (3)~\textbfAgency Theft, a lexical shift where models like LLaMA produce a 5 \times higher victimization-to-politics vocabulary ratio for Roma students compared to elite peers; and (4)~\textbfElite Hermeneutics, where AI tutors disproportionately withhold epistemic confidence and justification scores from low-resource demographic profiles. We argue that current safety alignment acts as a paternalistic filter, transforming conversational AI into agents of narrative segregation – a manifestation of \emphhermeneutical injustice in Fricker’s~\citefricker2007 sense that demands urgent pedagogical auditing.
[NLP-35] FAD-SA-GRU: Enhancing Hate Speech Detection in Algerian Dialect Through Feature-Augmented Self-Attention GRU Networks
【速读】： 该论文旨在解决在阿尔及利亚阿拉伯语（Darija）社交媒体文本中自动识别仇恨言论的难题，这一任务因方言的语言多样性（融合阿拉伯语、法语及阿拉伯拉丁化书写系统“Arabizi”）而极具挑战性。现有方法在低资源方言场景下的表现受限，难以有效捕捉复杂的语言特征与上下文语义。其解决方案的关键在于提出一种新型混合架构FAD-SA-GRU，通过多嵌入融合策略整合来自DZ FastText、DZ AraVec和DziriBERT的互补语义表示，并结合自注意力机制增强的门控循环单元（GRU）编码器，实现对序列上下文的高效建模。实验结果表明，该模型在二分类任务中达到93.2%准确率、93.4%精确率、91.0%召回率、92.1% F1分数和97.0% ROC-AUC，显著优于传统机器学习、深度学习及主流Transformer模型，验证了融合多源嵌入表征与注意力驱动序列建模在低资源方言仇恨言论检测中的有效性。
链接: https://arxiv.org/abs/2607.11279
作者: Sara Yakoubi,Ikram Khalfallah,Kenza Khelkhal,Dihia Lanasri
机构: USTHB(阿尔及利亚科学技术大学); ATM Mobilis(ATM移动公司)
类目: Computation and Language (cs.CL)
备注:
Abstract:The widespread adoption of social media platforms has transformed online communication by enabling users to exchange information and opinions instantly. However, these platforms have also facilitated the dissemination of abusive and hateful content, posing major social, psychological, and ethical challenges. Hate speech can incite discrimination, harassment, and violence against individuals or communities based on attributes such as ethnicity, religion, gender, nationality, or political affiliation. Consequently, automatic hate speech detection has become a major research topic in natural language processing (NLP) and an essential component of content moderation systems. This paper investigates automatic hate speech detection in the Algerian Arabic dialect (Darija) on social media. This task remains challenging because of the dialect’s linguistic diversity, characterized by the coexistence of Arabic, French, and Arabizi (Arabic written using the Latin alphabet). We compare four categories of text classification approaches: (1) traditional machine learning models using TF-IDF features, (2) deep learning models based on recurrent neural networks, (3) Transformer-based language models, including DziriBERT and multilingual BERT, and (4) a novel hybrid architecture, FAD-SA-GRU, which combines semantic representations from DZ FastText, DZ AraVec, and DziriBERT through multi-embedding fusion, followed by a self-attention-enhanced GRU encoder. Experiments on an annotated dataset of Algerian Darija social media comments for binary hate speech classification show that FAD-SA-GRU outperforms all baselines, achieving 93.2% accuracy, 93.4% precision, 91.0% recall, 92.1% F1-score, and 97.0% ROC-AUC. Results demonstrate the effectiveness of combining complementary embedding representations with attention-based sequence modeling for robust hate speech detection in low-resource dialectal Arabic. Subjects: Computation and Language (cs.CL) Cite as: arXiv:2607.11279 [cs.CL] (or arXiv:2607.11279v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2607.11279 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[NLP-36] reeThink: A Modular Tree Search Library for Mathematical Reasoning with LLM s
【速读】： 该论文旨在解决神经定理证明中树搜索算法与形式化验证器之间缺乏原生集成的问题。现有大语言模型（LLM）树搜索库主要面向自然语言推理，未提供与形式化验证器的直接对接能力，而传统定理证明系统则依赖于任务特定的搜索实现，灵活性和可扩展性不足。其解决方案的关键在于提出TreeThink——一个开源的Python库，实现了神经定理证明中模块化、完全异步的树搜索框架。该框架将成熟的树搜索方法与基于vLLM的推理流水线及多样化的节点评估技术（从轻量级启发式到神经评估器）相集成，并支持Lean 4、Rocq和Isabelle/HOL等形式化语言以及自然语言，通过直接连接各语言的读取-求值-打印循环（REPL）服务器，实现对证明状态的实时验证与提取。实验在miniF2F和MATH500数据集上验证了跨语言形式化证明搜索能力、自然语言推理支持以及高达6.3倍的墙时速度提升，显著提升了神经定理证明的效率与通用性。
链接: https://arxiv.org/abs/2607.11258
作者: Burak S. Akbudak,Zeynel A. Uluşan,Can S. Erer,Gözde Gül Şahin
机构: Bogazici University (博兹库尔大学); Codeway Studios; Friedrich-Alexander-Universität Erlangen-Nürnberg (埃尔兰根-纽伦堡弗里德里希-亚历山大大学); Koç University (科奇大学); KUIS AI Lab (KUIS人工智能实验室)
类目: Computation and Language (cs.CL)
备注:
Abstract:Tree search algorithms enable systematic exploration of the proof space in neural theorem proving. Existing LLM tree search libraries primarily target natural language reasoning and do not provide native integration with formal verifiers, while theorem proving systems often rely on task-specific search implementations. We introduce TreeThink, an open-source Python library for modular, fully asynchronous tree search in neural theorem proving. It integrates established tree search methods with vLLM-based inference pipelines and diverse node evaluation techniques, ranging from lightweight heuristics to neural evaluators. We support Lean~4, Rocq, and Isabelle/HOL alongside natural language. It connects directly to each language’s Read-Eval-Print Loop (REPL) server for real-time verification and proof state extraction. We evaluate TreeThink on miniF2F and MATH500, demonstrating cross-language formal proof search, natural language reasoning support, and up to 6.3 \times wall-clock speedup from asynchronous execution. Source code is released under the MIT license at this https URL , and the library is accessible as a downloadable package at this https URL .
[NLP-37] Q-BridgeNet: A Quantization Network for Cross-Lingual Sign Language Translation
【速读】： 该论文旨在解决多语言手语翻译（Multilingual Sign Language Translation, SLT）中普遍存在的跨语言冲突问题，即如何在保持不同手语间语言特异性的同时，有效捕捉跨语言共享语义，并构建一个统一的翻译模型以提升多语言手语与口语之间的互通性。其核心挑战在于现有方法难以平衡语言共性与个性之间的矛盾，导致模型泛化能力受限。解决方案的关键在于提出Q-BridgeNet框架，通过双向联合优化实现跨语言对齐：在手语侧，采用自适应分割与残差向量量化（Residual Vector Quantization, RVQ）机制，构建共享的基础代码本（Q-units）以表征跨语言语义基元，同时引入语言特异性的残差代码本以保留各手语的独特表达差异；在口语侧，将多语言大语言模型（Multilingual Large Language Model, LLM）微调至Q-unit空间，利用跨语言先验知识实现统一建模。该设计有效缓解了手语与口语双侧的跨语言冲突，实验证明Q-BridgeNet在PHOENIX14T、How2Sign和CSL-Daily数据集上不仅在原生手语-口语对上达到领先性能，还展现出优异的非原生对泛化能力。
链接: https://arxiv.org/abs/2607.11215
作者: Liqian Feng,Lintao Wang,Xiaochen Liu,Anusha Withana,Ken-Tye Yong,Dehui Kong,Zhiyong Wang,Kun Hu
机构: 未知
类目: Computation and Language (cs.CL); Multimedia (cs.MM)
备注:
Abstract:Most sign language translation (SLT) methods focus on isolated native sign-spoken pairs (e.g., American Sign Language - English). Extending language-specific SLT models to multilingual translation would improve accessibility by enabling communication across diverse sign and spoken language communities. However, existing multilingual SLT approaches still struggle to learn a unified model that minimizes cross-lingual conflicts while capturing shared cross-lingual semantics and preserving language-specific variations across different sign languages. Therefore, we propose Q-BridgeNet, a unified framework for multilingual SLT that jointly mitigates cross-lingual conflicts across both the sign language and spoken language sides. On the sign language side, Q-BridgeNet learns discrete Q-units via adaptive segmentation and residual vector quantization: a shared base codebook provides language-agnostic semantic primitives, while language-specific residual codebooks refine heterogeneous signing semantics. On the spoken language side, a multilingual LLM is fine-tuned to operate in the Q-unit space, leveraging cross-lingual priors to enable a unified SLT model. Experiments on PHOENIX14T, How2Sign, and CSL-Daily show that Q-BridgeNet effectively mitigates cross-lingual conflicts, achieving state-of-the-art performance on native sign-spoken pairs while also demonstrating strong generalization to non-native pairs. Our source code is publicly available at: this https URL
[NLP-38] When the Target Domain Changes: AI-Mediated Construct Drift in High-Stakes English Language AssessmenW
【速读】： 该论文旨在解决高风险英语能力测试在生成式AI（Generative AI）日益普及背景下，其评分解释的概念效度（conceptual validity）危机问题。传统上，此类测试依赖标准化、无辅助的应试表现来推断学术英语能力，但随着目标语言使用场景中生成式AI的深度介入，从“无辅助表现”向“学术交际准备度”的外推逻辑变得不再自明。论文的核心论点是：生成式AI不应仅被视为评分、反馈、安全或作弊等操作层面的问题，而应被重新定位为影响测验构念效度（construct validity）与外推合理性（extrapolation warrants）的根本性挑战。其关键解决方案提出“有限生成式AI辅助”（bounded AI mediation）作为一项以效度为导向的设计原则——即所有考生在统一、机构管控的条件下，可访问具有预设协助边界、交互记录可追踪的AI助手，且测试任务明确区分对理解支持与答案生成的不同要求。该设计旨在通过制度化、透明化的AI使用机制，缓解因生成式AI介入导致的构念漂移（AI-mediated construct drift），并主张在将测试分数用于推断AI辅助下的学术交流能力时，应相应缩小解释范围并补充额外证据。
链接: https://arxiv.org/abs/2607.11213
作者: Yi Gui
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:High-stakes English proficiency tests treat standardized, unaided performance as evidence for score interpretations about academic English proficiency. This interpretation remains meaningful, but as target language use domains increasingly involve generative AI, the extrapolation from unaided test performance to academic communicative readiness becomes less self-evident. This conceptual validity argument reframes AI as a score-interpretation problem in high-stakes language testing, not only an operational issue of scoring, feedback, security, or misconduct. Synthesizing current literature in three uneven layers, the paper shows that most work treats AI as assessment infrastructure, while far less theorizes its implications for construct validity and extrapolation warrants. It defines AI-mediated construct drift as the misalignment that arises when communicative abilities required in the target domain change through AI mediation while test constructs remain anchored to an unaided-performance model. It proposes bounded AI mediation as a validity-oriented design principle: a standardized condition in which all test takers access the same institutionally controlled AI assistant, with predefined assistance boundaries, logged interactions, and tasks that distinguish comprehension support from answer generation. The paper argues that score interpretations should be narrowed and supplemented when used to support claims about AI-mediated academic communication.
[NLP-39] ProgramTab: Boosting Table Reasoning of LLM s via Programmatic Paradigm
【速读】： 该论文旨在解决大语言模型（LLM）在基于表格的推理任务中面临的挑战，特别是当处理大规模表格时因长文本建模困难及输入长度限制导致的性能下降问题。现有方法如文本到SQL（text-to-SQL）虽能通过生成子表提升效率，但受限于网页表格等非结构化数据缺乏一致性与规范性，难以支持有效的数学逻辑运算。为此，本文提出ProgramTab框架，其核心在于利用上下文学习（in-context learning）引导LLM以Python代码形式完成表格预处理，并实现行/列级内容提取与SQL生成。该方案通过程序化方式增强对不规则表格的适应能力，显著提升了推理准确性与鲁棒性，在多个表格推理数据集上优于所有基于LLM的基线方法。
链接: https://arxiv.org/abs/2607.11207
作者: Pei Guo,Enjie Liu,Yunzhi Tan,Mochi Gao,Jianxin Zhang,Ruichao Zhong,Juntao Li,Bo Hu,Zang Li
机构: Big Data and AI Platform Department, Tencent, China; Institute of Computer Science and Technology, Soochow University, China
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Large Language Models, Table Reasoning, In-context Learning
Abstract:Table-based reasoning with large language models (LLMs), which requires reasoning based on natural language questions and structured tabular data, has gained widespread attention. However, a series of issues still constrain the application of this task. The previous approaches suffered from significant performance degradation when faced with large tables due to the difficulty of long text modeling and the limitation of input length for LLMs. The text-to-SQL approach is used to efficiently extract key information from tables and generate smaller sub-tables. However, tabular data, especially web tables, often lack the necessary structure and consistency, making them unsuitable for performing mathematical logic operations using SQL queries. We propose the ProgramTab framework, which guides LLMs employing in-context learning to perform tabular data preprocessing with Python code, as well as the momentous contents extraction with row and column extraction and SQL generation. The experiment results on table reasoning datasets demonstrate that the ProgramTab framework effectively deals with table-based reasoning tasks and outperforms all LLM-based baselines.
[NLP-40] Amplitude-Only FFN Intervention for Tool-Structured LLM Inference Method: Gated Evaluation Protocol and Cross-Model Empirical Results
【速读】： 该论文旨在解决大语言模型在作为工具使用代理（tool-using agents）时，因微小的格式、参数或函数调用错误导致本可合理输出失效的问题。其核心挑战在于如何在不重新训练模型权重的前提下，通过推理阶段的前馈网络（FFN）干预提升结构化输出的准确性与可靠性。解决方案的关键在于提出一种非破坏性干预机制——幅值门控（Amplitude Gating, AG），该方法保留预训练FFN权重的方向性，仅调节生成过程中的激活幅值，避免了如正交残差投影（Orthogonal Residual Projection, ORP）等方向性修改带来的副作用。研究构建了一个细粒度的干预体系，涵盖P1/P2/P3及分支特异性子站点，并设计了一套严格的评估协议，实现组合最优头空间与固定配置、学习型门控之间的分离评估，强化样本级统计分析，并采用任务感知指标衡量二分类与部分得分数据集的表现。实验表明，AG在工具结构化任务上表现最优，尤其在Qwen3.5-9B上，类别级学习门控使工具/结构化/智能体任务准确率从38.66%提升至42.92%（+4.27个百分点），赫尔墨斯函数调用任务提升约+7.6点；在Qwen3-8B上，赫尔墨斯JSON模式提升+11.36点；而Qwen2.5-7B虽仍保留最优头空间，但当前学习门控未能有效捕获，揭示部署需依赖模型与任务类别特定路由策略。熵型AG与牛顿-施尔茨窗口化AG的对比显示两者无绝对优势，进一步验证了工具结构化推理是安全的FFN级推理优化最可信的切入点，未来仍需在线验证机制和跨模型泛化评估以推动实际应用。
链接: https://arxiv.org/abs/2607.11183
作者: Sheng Xu,Boyuan Huang,Ke Jia,Jiadun Zhu,Zhen Chen
机构: 未知
类目: Computation and Language (cs.CL)
备注: 28 pages, 15 figures
Abstract:Large language models increasingly operate as tool-using agents, where small format, argument, or function-call errors can invalidate otherwise plausible responses. We study inference-time feed-forward network (FFN) intervention for improving structured outputs without retraining model weights. Our project began with Orthogonal Residual Projection (ORP), a direction-changing repair attempt that revealed sensitive SwiGLU FFN intervention sites but often caused more harm than fixes. We therefore propose Amplitude Gating (AG), a non-destructive alternative that preserves pretrained FFN weight directions and modulates only activation magnitudes during generation. We define a fine-grained intervention system spanning P1/P2/P3 and branch-specific P1s/P2a/P2b sites, and introduce an evaluation protocol that separates combination-oracle headroom from fixed configurations and learned gates, enforces sample-level accounting, and uses task-aware metrics for binary and partial-credit datasets. Across Qwen3.5-9B, Qwen3-8B, and Qwen2.5-7B, AG is weakly positive in aggregate but strongest on tool-structured tasks. On Qwen3.5-9B, a category-level learned gate improves tool/structured/agentic performance from 38.66% to 42.92% (+4.27 percentage points), with Hermes function-call tasks reaching about +7.6 points. On Qwen3-8B, Hermes JSON mode improves by +11.36 points. Qwen2.5-7B retains oracle headroom but current learned gates fail to capture it, showing that deployment requires model- and category-specific routing. Comparisons of entropy AG with Newton-Schulz-windowed AG show that neither family is uniformly dominant. These results identify tool-structured inference as the most credible first target for safe FFN-level inference optimization, while prospective online validation and broader cross-model evaluation remain necessary.
[NLP-41] Query-Focused Event Summarization: A Dataset and Benchmark
【速读】： 该论文旨在解决现有查询聚焦摘要（Query-Focused Summarization, QFS）任务在事件导向型摘要生成方面的不足，以及现有方法在大规模语料库上性能受限的问题。具体而言，传统QFS数据集缺乏以主题事件为核心的结构化文档集合，且多数方法难以有效处理包含数百至数千篇文档的大规模语料。为此，论文提出了查询聚焦事件摘要（Query-Focused Event Summarization, QFES）新任务，并构建了QFESum数据集，涵盖8个主题事件、16,684篇文档及104个查询。其解决方案的关键在于提出一种两阶段框架：第一阶段为基于自适应阈值的查询聚焦检索（Query-Focused Retrieval with Adaptive Thresholding, RAT），通过动态调整相关性阈值提升检索精度；第二阶段为基于分层聚类的查询聚焦摘要生成（Query-Focused Summarization based on Hierarchical Clustering, SHC），利用层次聚类对候选文档进行结构化组织，从而生成更全面、精准的事件层面摘要。实验结果表明，RAT与SHC在QFESum数据集上显著优于基线方法，验证了该框架在大规模事件摘要任务中的有效性。
链接: https://arxiv.org/abs/2607.11166
作者: Chenyu Hu,Bang Wang
机构: Huazhong University of Science and Technology (华中科技大学)
类目: Computation and Language (cs.CL)
备注: 22 pages, 9 figures, and 13 tables. Dataset and code are available at this https URL
Abstract:A thematic corpus is a collection of semantically coherent documents that collectively describe different aspects of a shared thematic event. Such a corpus typically contains hundreds or even thousands of documents. While users’ interests in a thematic event often span multiple dimensions, Query-Focused Summarization (QFS) aims to generate summaries tailored to users’ queries. However, existing QFS datasets lack event-oriented summarization, and most QFS methods struggle with large-scale corpora. To address these challenges, we propose the Query-Focused Event Summarization (QFES) task and construct the QFESum dataset, which contains 8 thematic events, 16,684 documents, and 104 queries. Furthermore, we introduce a two-stage QFES framework consisting of Query-Focused Retrieval with Adaptive Thresholding (RAT) and Query-Focused Summarization based on Hierarchical Clustering (SHC). Experimental results on QFESum show that RAT and SHC consistently outperform the baselines, demonstrating their effectiveness for QFES. The dataset and code are publicly available at this https URL.
[NLP-42] Unified Gradient Projection: Language-Balanced Continual Learning for Multilingual Low-Resource ASR INTERSPEECH2026
【速读】： 该论文旨在解决大规模预训练自动语音识别（ASR）模型在低资源语言上进行微调时普遍存在的灾难性遗忘问题，尤其在多语言场景下，主流语言会主导优化过程，导致跨任务干扰与语言偏差。其核心解决方案是提出统一梯度投影（Unified Gradient Projection, UGP），通过在统一的投影空间中利用语言均衡回放生成的参考梯度来约束参数更新，从而在投影空间中均衡各语言的贡献，有效缓解主流语言带来的优化偏置，提升跨语言稳定性。进一步研究表明，梯度层面的投影与数据层面的回放相结合可实现稳定性和可塑性的互补增益。实验表明，UGP在多种低资源语言组和不同模型规模下均能实现有效适应，并显著降低遗忘程度，在Whisper-large-v3上实现了接近零的平均遗忘率。
链接: https://arxiv.org/abs/2607.11163
作者: Ziang Ren,Guodong Lin,Yuchen Ai,Kaize Tan,Wei-Qiang Zhang
机构: Tsinghua University (清华大学)
类目: Computation and Language (cs.CL); Sound (cs.SD); Audio and Speech Processing (eess.AS)
备注: Accepted by Interspeech 2026
Abstract:Large-scale pretrained ASR models such as Whisper exhibit strong multilingual capabilities. However, fine-tuning on low-resource languages often causes catastrophic forgetting. Although continual learning mitigates this issue, existing methods struggle to regulate cross-task interference in multilingual settings, where dominant languages bias optimization. We propose Unified Gradient Projection (UGP), which constrains parameter updates using reference gradients from language-balanced replay in a unified projection space. By equalizing per-language contributions in the projection, UGP reduces dominant-language bias and improves cross-lingual stability. We further show that combining gradient-level projection with data-level replay yields complementary gains in stability and plasticity. Across diverse low-resource language groups and model scales, UGP enables effective adaptation while substantially mitigating forgetting. On Whisper-large-v3, it achieves near-zero average forgetting.
[NLP-43] IGER: Text-Conditioned Visual Gated Routing with Acceptance Alignment for Multimodal Speculative Decoding
【速读】： 该论文旨在解决多模态大语言模型（VLMs）中生成式推理加速难题，尤其针对现有基于推测解码（speculative decoding）方法在视觉关键内容上存在偏差、导致加速效果有限的问题。其核心挑战在于：传统方法要么暴露全部视觉令牌（visual tokens），引入冗余信息；要么采用固定压缩接口，缺乏对上下文相关性的动态适应能力；同时，现有方法未充分优化验证器（verifier）接受的前缀长度这一决定加速效率的关键因素。为此，论文提出TIGER框架——一种基于文本条件的视觉门控路由机制，通过动态选择与当前文本状态相关的稀疏视觉令牌，实现更精准的视觉信息筛选。其关键创新在于：采用基于接受前缀长度的验证器反馈奖励，结合蒸馏预热与KL锚定的分组策略训练，使轻量级起草模型（drafter）不仅模仿目标模型行为，更倾向于生成可被验证器接受更长前缀的推测序列。实验表明，TIGER在保持下游任务准确率的前提下，显著提升了接受前缀长度和推测解码速度，实现了高质量与低延迟之间的良好权衡。
链接: https://arxiv.org/abs/2607.11131
作者: Quynh Vo,Cong-Duy Nguyen,Ponhvoan Srey,Luu Anh Tuan,Thong Nguyen
机构: National University of Singapore (新加坡国立大学); Nanyang Technological University (南洋理工大学); Center of AI Research, VinUniversity (越南大学人工智能研究中心)
类目: Computation and Language (cs.CL)
备注: Work in progress
Abstract:Speculative decoding accelerates autoregressive generation by letting a lightweight drafter propose multiple tokens that are verified by a larger target model. Although effective for text-only LLMs, speculative decoding yields limited gains in VLMs because drafters often diverge on vision-critical content, while existing multimodal acceleration methods do not directly address irrelevant visual evidence or optimize the verifier-accepted prefix length that governs speedup. We propose TIGER, a Text-conditioned vIsual GatEd Routing framework for multimodal speculative decoding. TIGER dynamically selects a sparse set of context-relevant visual tokens based on the drafter’s current textual state, rather than expose the full visual token set or a fixed compressed interface. To better align training with inference-time efficiency, we optimize the drafter with acceptance-aligned group-based policy training using verifier-derived rewards based on accepted prefix length, built on top of distillation warm start with KL anchoring. This encourages the drafter not only to imitate the target model, but also to produce speculative continuations that survive verification for longer prefixes. Experiments show that TIGER yields consistent gains in accepted prefix length and speculative speedup under exact verifier-side speculative decoding, while achieving favorable quality-latency trade-offs with comparable downstream accuracy in visual-routing analyses.
[NLP-44] Do LLM s Fabricate Legal Citations? A Bilingual Benchmark on Saudi Data Protection Law and the GDPR
【速读】： 该论文旨在解决生成式 AI 在法律合规咨询中因虚构法条引用而导致错误信息传播的问题，尤其关注多语言环境下大语言模型（LLM）对数据保护法规的引用准确性。其核心挑战在于：尽管模型在处理欧盟《通用数据保护条例》（GDPR）时表现出较高的引用准确率（94–100%），但在面对沙特《个人数据保护法》（PDPL）时却出现高达60–77%的引用伪造现象，且该问题与查询语言无关，反映出模型对不同司法管辖区法律体系的理解存在显著差异。解决方案的关键在于构建一个双语基准测试集（涵盖120个问题），包含直接引用检索、虚假前提验证及故意无法回答的“陷阱”问题（如关于已废止条款或仅存在于实施细则中的截止日期），并通过人工验证的黄金标准实现全自动评分。研究发现，模型自信度（置信度≥0.8）无法有效防范虚构引用，91%的错误引用均以高置信度输出，表明依赖模型自我判断不可靠；因此，必须采用逐字验证等外部核查机制，而非仅依赖模型置信度，作为机构在合规筛查中使用LLM的前提保障。
链接: https://arxiv.org/abs/2607.11127
作者: Noura Suliman Alrajeh
机构: 未知
类目: Computation and Language (cs.CL); Computers and Society (cs.CY)
备注: 5 pages, 3 tables. Benchmark data and model outputs to be released. Also archived at Zenodo: https://doi.org/10.5281/zenodo.21320218
Abstract:Organizations and regulators increasingly consult large language models (LLMs) for regulatory-compliance questions, yet a wrong statutory citation can silently propagate into legal advice, compliance documentation, and policy decisions. We introduce a bilingual benchmark of 120 questions probing whether freely accessible LLMs fabricate article citations for two data-protection instruments: the EU General Data Protection Regulation (GDPR) and the Saudi Personal Data Protection Law (PDPL). The benchmark pairs direct citation retrieval questions with false premise verification probes and deliberately unanswerable “trap” questions – including questions about a repealed article and about deadlines that exist only in implementing regulations, not in the law itself. Every question is posed in both Arabic and English, and all scoring is fully automatic against a manually verified gold reference. Evaluating three freely accessible models (Gemini 2.5 Flash, GPT-OSS-120B, Nemotron-3-Super-120B), we find a dramatic jurisdiction gap: near-ceiling citation accuracy on the GDPR (94-100% on direct retrieval) against majority fabrication on the Saudi PDPL (60-77%), invariant to query language; the highest fabrication rates (67%) arise from statute-vs-regulations confusion, and 91% of fabricated citations are asserted with confidence = 0.8. Fabrication tracks the jurisdiction of the law, not the language of the query, and model confidence provides no protection – indicating that verbatim-verification safeguards, rather than model self confidence, must gate any institutional reliance on LLMs for compliance screening.
[NLP-45] Simple Features and Honest Calibration for Ambivalence and Hesitancy Recognition in Video
【速读】： 该论文旨在解决在真实访谈视频中识别个体存在情感矛盾与犹豫（Ambivalence and Hesitancy, A/H）的问题，其核心挑战在于如何有效融合多模态信息以准确捕捉非显性言语行为中的细微心理状态信号。解决方案的关键在于提出一种名为“情感标记融合”（Affective Marker Fusion, AMF）的可靠性门控机制，通过整合专门针对情感的文本、音频与视觉表征，并引入一组可读性强的语言犹豫线索（如填充词、停顿等），实现跨模态信息的动态加权融合。此外，研究创新性地提出“ASR去噪时间”（ASR-erased time）特征，即保留语音识别系统去除填充语和停顿后的时间戳信息，由此构建的16个基于时间间隙的非言语特征成为表现最强且与其他模态相关性最低的独立通道（AP 0.718，与其他特征的相关系数为0.11–0.36）。通过控制实验发现：跨模态冲突设计对BAH任务并无显著增益；语言模态为最强信号源，情感专用音频为有效补充；模型校准（calibration）的重要性远超架构设计。特别地，采用固定决策阈值下的AP加权集成策略，避免了在小验证集上过度拟合带来的性能下降问题，在未接触测试集上实现了0.731的AP，优于基于验证集调优的方案（0.690）。
链接: https://arxiv.org/abs/2607.11120
作者: Vikas Kumar,Aditya Mishra,Haroon R. Lone
机构: Indian Institute of Science Education and Research Bhopal (印度科学教育与研究学院博帕尔分校)
类目: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL); Audio and Speech Processing (eess.AS)
备注:
Abstract:We address ambivalence and hesitancy (A/H) recognition in the ABAW 2026 BAH Challenge: given a short interview video, predict whether the person shows signs of A/H. Our system combines affect-specialised text, audio, and visual representations with a small set of readable linguistic hesitation cues, fused by a reliability gate we call Affective Marker Fusion (AMF), and finished with a simple AP-weighted ensemble at a fixed decision threshold. We also introduce \emphASR-erased time: speech recognisers delete fillers and hesitation pauses from the transcript, but the chunk timestamps keep the time those events took, and sixteen features built from these gaps form the strongest and most independent non-verbal channel we measured (AP 0.718 , correlation 0.11 – 0.36 with all other members). Across controlled experiments we find three things: cross-modal conflict design does not reliably help on BAH; language is by far the strongest channel while affect-specialised audio is a useful second; and calibration matters more than architecture. Fitting ensemble weights and a threshold on the small validation split overfits: it scores 0.741 macro-F1 on validation but only 0.690 on the untouched test set. AP-weighting at a fixed threshold instead reaches \mathbf0.731 on test.
[NLP-46] Agent Check: A Reproduce-Intervene-Mitigate Workbench for LLM Agents over MCP
【速读】： 该论文旨在解决大语言模型（LLM）代理在实际部署中因工具失效（如超时、返回过期数据或描述被污染）而产生不可靠行为的问题，尤其关注这些故障难以复现、调试与验证的挑战。现有评估通常假设所有工具均正常工作，但真实场景下工具异常普遍存在，导致代理产生看似合理却错误的输出，且故障常表现为“静默失败”——即代理以高置信度使用错误工具响应，而非直接崩溃。为此，论文提出AgentCheck，一个开源的Web工作台，将MCP服务器转变为可干预的测试表面。其核心解决方案在于构建“复现-干预-确认”闭环：通过记录代理对真实工具的完整响应，再利用12种类型的故障注入器对响应进行扰动，实现对故障场景的精确复现；在重放过程中，匹配的工具调用从缓存中回放，仅当代理行为出现分歧后才重新调用真实工具，从而保证测试环境可控且高效。开发人员可通过切换缓解策略（mitigation），在相同故障条件下反复测试并观察效果。评估采用双重评分机制：基于确定性规则的通过/失败判定，以及经人工标注验证的生成式AI（Generative AI）判别器进行语义层面的解释性标签判断。实验结果显示，在五类代理中，表现最佳者在120个测试场景中成功通过105个，最弱者为77个，表明故障模式普遍且影响显著。例如，在最弱代理上，重试机制可将超时故障的成功率从最低30%提升至100%，但对过期数据故障的改善有限（始终维持在3-4/10），凸显不同故障类型需差异化应对。AgentCheck实现了故障模式的可复现、可比较与可验证，显著提升了工具级故障管理的工程化能力。
链接: https://arxiv.org/abs/2607.11098
作者: Aritra Mazumder,Nusrat jahan Lia
机构: University of Utah (犹他大学); University of Dhaka (达卡大学)
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Tool-using LLM agents are mostly evaluated assuming all tools work. When a tool times out, returns a week-stale value, or has its description poisoned in deployment, the developer needs a controlled way to reproduce the failure, test a fix, and confirm the fix worked before deployment. We present AgentCheck, an open-source web workbench that turns an MCP server into an intervention surface. AgentCheck runs an agent against its real tools and records every tool response, then re-runs the agent with the response perturbed by a fault (12 types) injector. Matching tool calls are replayed from cache, and later tool calls go live after the agent diverges. This yields a reproduce-intervene-confirm loop: the developer toggles a mitigation, re-runs against the identical fault, and sees if the failure goes away. Scoring has two parts: deterministic pass/fail rules, plus an LLM judge for interpretive labels, validated against human annotations. Across five agents, the best passes 105/120 scenarios and the weakest only 77. The failures are usually silent, confident use of incorrect tool outputs rather than crashes. On the weakest agent, a retry mitigation raises success on timeout error faults from as few as 30% of cases to 100%, whereas stale-data faults remain near 3-4 of 10 regardless of the mitigation. AgentCheck makes these failure modes reproducible, comparable, and verifiable before deployment.
[NLP-47] ResearchQA: Benchmarking Citation-Grounded Question-Answering on Scientific Papers
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在辅助科学阅读时，现有评估方法难以识别答案是否具备可验证引文支持的问题。其核心挑战在于：当前评估体系常依赖主观评分或生成式评价，无法有效区分模型回答是否基于论文中的实际证据，从而导致对模型真实推理能力的误判。解决方案的关键在于提出一个基于引文的评估基准——ResearchQA，包含6,211个来自494篇开放获取论文的单篇论文问答对，覆盖八个领域和四种题型（查证、理解、多跳推理与对抗性问题）。该基准设计强调引文支撑性：允许一个结论有多个有效支持段落，并在源文献不支持答案时奖励“基于引文的拒绝”行为。通过采用确定性引文匹配器与基于大模型的评分器相结合的方式，在“对话式论文阅读”场景下对八种主流闭源与开源模型进行评估，结果表明，基于引文的指标能更清晰地区分不同模型表现，尤其在章节覆盖度与引文准确性方面差异显著，而基于大模型的评分则高度集中；同时发现，开源模型在引文准确性上接近最优闭源模型，且每个样本的延迟降低3至6倍。研究团队已公开发布该基准、评估工具链及评分提示模板，以推动科学阅读场景下可信生成式AI的发展。
链接: https://arxiv.org/abs/2607.11074
作者: Saba Imran,Debanjum Singh Solanky
机构: Khoj Inc.
类目: Computation and Language (cs.CL)
备注: 19 pages, 9 figures
Abstract:Large language models are increasingly used to assist scientific reading, but existing evaluation methods often fail to detect whether answers are supported by verifiable citations. We introduce ResearchQA, a benchmark of 6,211 single-paper question-answer pairs from 494 open-access papers spanning eight domains and four question types: lookup, comprehension, multi-hop, and adversarial. ResearchQA is designed for citation-grounded evaluation: it permits multiple valid supporting passages for a claim and rewards grounded refusal when the source paper does not support an answer. We evaluate eight leading closed- and open-weight models in a citation-grounded chat-with-paper setting using a deterministic citation matcher and an LLM-based rubric evaluator. Citation-based metrics separate systems more clearly than LLM-evaluator scores: section coverage and citation accuracy vary substantially across models, while evaluator scores remain tightly compressed. We further find that open-weight models approach the best closed-model citation accuracy while achieving 3 to 6 times lower per-example latency. We release the benchmark, evaluation harness, and evaluator prompt.
[NLP-48] MJ: Multi-turn LLM Jailbreaking via Decomposed Credit Assignment
【速读】： 该论文旨在解决多轮越狱攻击（multi-turn jailbreaking）中关键的信用分配（credit assignment）难题，即在多轮交互过程中，不同对话回合对最终越狱成功与否的贡献程度各异，而现有方法通常依赖粗粒度的学习信号，难以准确识别各轮次的个体贡献。为此，论文提出了一种统一的逐轮信用分配框架——分解信用组相对策略优化（Decomposed Credit GRPO, DC-GRPO），其核心在于通过融合即时信用与未来信用，为每一轮对话单独分配组相对学习信号，从而避免传统方法因将单一轨迹级评分广播至整个对话而导致的信用误分配问题。该框架分别实现了静态与动态权重规则，在多个目标大语言模型（LLM）和基准测试上，动态与静态变体分别取得了98.26%和97.88%的平均ASR5@3得分，显著优于当前最优方法（如SEMA 86.58%、TROJail 86.23%）。实验结果表明，性能提升的主要来源在于逐轮组相对信用分配机制本身，而非特定权重规则的设计。
链接: https://arxiv.org/abs/2607.11070
作者: Junyoung Park,Namgyu Park,Sechan Lee,Yoon-Chan Jhi,Jihoon Cho,Sangdon Park
机构: POSTECH GSAI (POSTECH 人工智能研究中心); Samsung SDS (三星 SDS)
类目: Computation and Language (cs.CL)
备注: 29 pages. Warning: This paper contains examples of harmful content
Abstract:Modern large language models (LLMs) operate in interactive multi-turn settings, making multi-turn jailbreaking a realistic threat model and an important setting for automated red teaming. A core challenge in learning multi-turn jailbreak attackers is credit assignment: different turns contribute differently to the final outcome, yet existing learning signals are often too coarse to identify their individual contributions. We propose decomposed credit GRPO (DC-GRPO), a unified turn-level credit assignment framework for Group Relative Policy Optimization in multi-turn jailbreak learning. DC-GRPO assigns a separate group-relative learning signal to each turn by combining immediate and future credit, avoiding the credit misassignment induced by broadcasting a single trajectory-level score across the dialogue. We instantiate this framework with static and dynamic weighting rules that differ in how the two credit sources are balanced while sharing the same turn-level structure. Across multiple victim LLMs and benchmarks, the dynamic- and static-weighted variants achieve average ASR5@3 scores of 98.26% and 97.88%, respectively, substantially outperforming the state-of-the-art methods, including SEMA (86.58%) and TROJail (86.23%). Their consistently strong performance indicates that the central empirical benefit comes from turn-level group-relative credit assignment rather than a particular weighting rule. Warning: This paper contains examples of harmful content.
[NLP-49] Flout at Your Own Risk: LLM s Struggle with Prag matic Cooperativity Under Epistemic Asymmetry
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在部分信息条件下进行多角色协作时的语用推理能力问题，尤其关注其在缺乏与合作者共享信息情境下的合作沟通表现。其核心挑战在于：当参与协作的LLMs面临认知不对称（collaborative epistemic asymmetry）——即不同参与者掌握的信息不一致时，如何有效传递和理解语用线索以达成共同任务目标。解决方案的关键在于构建一个将客观任务成功与格赖斯合作原则（Grice’s cooperative principle）相联系的形式化框架，并通过提示工程（prompting）与后训练策略（post-training strategies）系统评估多种LLMs作为“说话者”与“听者”的合作能力。研究发现，尽管通过适当策略可激发LLMs的部分语用能力，但其在信息不完整场景下仍存在显著沟通缺陷，且这些失败模式往往对应于对格赖斯准则的无意识违背（flouting of Grice’s maxims），揭示了当前LLMs在真正意义上的协作性语用推理方面仍存在根本性局限。
链接: https://arxiv.org/abs/2607.11053
作者: Hannah VanderHoeven,Abhijnan Nath,Nikhil Krishnaswamy
机构: Colorado State University (科罗拉多州立大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Fruitful collaborations rely on cooperative communications, including of contextual cues to incorporate into reasoning. The increasing use of LLMs in collaborative and agentic pipelines raises questions about the extent to which they exhibit these pragmatic capabilities, especially in scenarios where they may not have access to the same information as their collaborators. In this paper, we perform a novel investigation into the pragmatic reasoning capabilities of LLMs in a multi-party collaborative task under partial information conditions. We formalize a notion of collaborative epistemic asymmetry that explicitly connects objective task success to Grice’s cooperative principle and empirically assess various LLMs’ abilities to act cooperatively as both speakers and listeners, including both prompting and post-training strategies. Our results show that while LLMs exhibit certain pragmatic capabilities in collaborative settings, and these can be elicited through prompting and post-training, they still face challenges in pragmatic communication with incomplete information, and that certain failure modes do correlate with floutings of Grice’s maxims that go unrecognized.
[NLP-50] Domain-Aware Scaling Laws Uncover Data Synergy
【速读】： 该论文旨在解决大规模语言模型（Large Language Model, LLM）预训练中数据构成对模型性能影响的复杂性问题，尤其关注不同数据域混合时产生的非线性交互效应。传统方法通常依赖于模型规模和数据量的扩展，但忽视了数据来源的多样性与组合方式对模型能力的潜在协同或干扰作用。论文提出“数据协同效应”（data synergy）这一概念，用以描述多个数据域联合贡献时超出或低于各自独立贡献之和的现象。其解决方案的关键在于构建一个可量化的框架，通过分析具有多样化预训练混合策略的开源权重大模型的观测差异，分别量化直接域-基准协同效应（domain-to-benchmark synergy，即某一数据域对另一任务表现的贡献）以及二阶域-域协同效应（second-order domain-domain synergy，即需多个数据域共现才能激发的能力）。该框架显著提升了对模型性能的预测准确性，超越了传统的领域无关缩放定律，并通过在预测最优与次优数据混合比例下训练新模型，验证了其协同效应估计的有效性与稳定性。
链接: https://arxiv.org/abs/2607.11052
作者: Kimia Hamidieh,Lester Mackey,David Alvarez-Melis
机构: MIT CSAIL(麻省理工学院计算机科学与人工智能实验室); Microsoft Research(微软研究院); Harvard University(哈佛大学)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:
Abstract:Machine learning progress is often attributed to scaling model size and dataset volume, yet the composition of data can be just as consequential. Empirical findings repeatedly show that combining datasets from different domains yields nontrivial interactions. For instance, adding code improves mathematical reasoning, while certain mixtures introduce interference that reduces model performance. We refer to these effects collectively as data synergy, where the contribution of multiple domains exceeds or falls short of the sum of their isolated contributions. In this work, we formalize and quantify data synergy in language model pretraining. Leveraging observational variation across open-weight LLMs with diverse pretraining mixtures, we estimate both direct domain-to-benchmark synergy (how one domain contributes to performance on another) and a second-order domain-domain synergy (capabilities that require co-occurrence of multiple domains). Our framework improves predictive accuracy over domain-agnostic scaling laws and recovers stable synergy estimates. We validate these estimates by training models on predicted optimal and predicted anti-optimal mixtures and confirm that our synergy estimates correctly predict performance rankings.
[NLP-51] Dimensionality in Satisfaction Ratings
【速读】： 该论文旨在解决传统客户满意度评估方法中因样本代表性不足与评价维度单一而导致的洞察偏差问题。现有方法通常依赖于少数完成问卷调查的客户反馈，难以全面反映真实服务体验，且无法深入解析满意度的具体驱动因素。其解决方案的关键在于利用生成式AI（Generative AI）——具体为GPT-4.1模型，对近9,000条全球消费品企业的客户服务对话文本进行自动化标注，将客户满意度分解为五个可解释的维度：整体满意度、客服人员表现、结果达成度、产品体验及客户付出努力。研究通过与客户自评满意度评分对比验证了标注有效性，发现除“产品满意度”外，其余四个维度与自评结果具有高度相关性（未调整相关系数达0.65~0.81），且分歧主要集中于少数异常对话会话中，剔除严重偏离案例后整体一致性提升至0.914。此外，尽管各维度间存在高度共线性，使其在预测整体评分上无增量贡献，但其核心价值在于实现更细粒度的归因分析与全量覆盖，揭示出基于全量对话数据的客户满意度显著低于传统抽样调查结果（全量普查得分为2.91，而抽样调查为3.62）。因此，该方法的核心优势在于通过大规模、精细化的分解式分析，从对话数据中挖掘出更深层、更真实的客户体验驱动因素。
链接: https://arxiv.org/abs/2607.11026
作者: Andrew Hong,Jason Potteiger
机构: 未知
类目: Computation and Language (cs.CL)
备注: 25 pages, 7 figures, 6 tables
Abstract:We used a large language model (GPT-4.1) to annotate the text of about 9,000 support conversations at a global consumer-goods firm, decomposing customer-care satisfaction into component axes (overall, agent, outcome, product, and customer effort), and validated the LLM annotations against the satisfaction ratings customers gave themselves. Four of five axes track self-reported satisfaction closely (overall, agent, and outcome near an unadjusted 0.65; effort -0.54), while product satisfaction is weak against the available proxy. The unadjusted correlation also understates the alignment: the disagreements concentrate in a small, readable tail of divergent sessions rather than in general drift, and the overall correlation rises to 0.811 when only the severe divergences are excluded and to 0.914 when the full divergent tail is excluded. The axes are also highly collinear, and adding them to the overall score does not improve prediction of the customer’s rating, the decomposition’s value is not incremental prediction but attribution and coverage. And, with greater coverage the picture of the data changes. Read on every contact rather than the few that return a survey, satisfaction is markedly lower than the survey reports (a full-census 2.91 against the surveyed 3.62 on a five-point scale). The promise of decomposed satisfaction as a methodology is the ability to identify more nuanced drivers of customer experience in conversational data.
[NLP-52] When the Reward Suite Is Leaky: A Preregistered Causal Contrast of Natural Verifier False Positives in RLVR
【速读】： 该论文旨在解决生成式 AI 在代码生成任务中因测试套件固有缺陷导致的奖励信号污染问题，即测试套件中存在的“假阳性”（false positives）会持续、不对称地接受相同错误代码，从而误导强化学习（RLVR）模型的学习方向。此类假阳性并非随机噪声，而是具有结构性偏差，使得模型在训练过程中可能被诱导学习到错误模式而非真正正确的代码逻辑。解决方案的关键在于通过引入“加固版”测试套件（MBPP+）替代原始测试套件（leaky），以消除这些结构性假阳性，并通过预先注册的双臂因果对比实验验证其有效性。研究发现，加固后的奖励机制显著降低了假阳性占比，且假阳性分布与训练前进行的静态泄漏审计高度相关（Spearman ρ = 0.80），同时揭示出真实错误代码仍占较大比例（47.57%记录加权），说明奖励机制确实捕捉到了实质性缺陷而非仅反映测试套件本身的漏洞。进一步机制分析表明，假阳性源于预存在的错误模式选择，而非模型在训练中主动利用漏洞，且未训练的基础模型已在相同泄漏条件下产生一致错误输出，支持了“选择已有错误模式”的解释。此外，研究还发现前沿评估者自身对假阳性判断能力较弱，其评分无法可靠反映模型性能，提示当前评估体系存在内在局限性。总体而言，该研究的核心贡献在于提出并验证了一种基于静态审计和测试套件加固的可操作性方法，有效识别并缓解了奖励信号中的系统性偏差，为提升生成式 AI 代码评估的可信度提供了关键实证依据。
链接: https://arxiv.org/abs/2607.11022
作者: Chuyifei Zhang
机构: Beijing Jiaotong University (北京交通大学)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: 37 pages, 2 figures. Code, frozen data, and complete audit records: this https URL
Abstract:The test suites used as RLVR rewards for code have natural false positives: per-task, persistent, asymmetric errors that accept the same wrong programs every time they appear, unlike the symmetric or resampled noise assumed by existing noise-robustness analyses. We run a preregistered two-arm causal contrast on a deployed suite: GRPO on identical MBPP tasks, seeds, and compute, rewarded by the original MBPP tests (leaky) versus the MBPP+ extra tests (hardened). Two further families replicate the design under a preregistration frozen before their data existed. [C] The average held-out effect is bounded: non-inferior under a preregistered 1.5-pt margin (gap 0.20 pt, one-sided 95% upper bound 0.75 pt). [C] Rewarded false-positive mass tracks a cheap static leakiness audit computed before training (Spearman 0.80), and the registered train-side test puts the leak-stratum FP share +43.8 pt above clean tasks. [E] Auditing every rewarded FP under signed, human-adjudicated rules finds a large residual of verified genuinely wrong code: 47.57% record-weighted; both replication families reproduce a large share. The reward paid for real bugs, not merely suite artifacts. [E] Mechanism evidence is consistent with selection of pre-existing error modes rather than learned exploitation: FP incidence does not grow within our horizon, and untrained base models already produce the same wrong outputs under the leaky filter. We then turn the same instrument on the frontier judges themselves: on their own false positives they self-assess only weakly, a same-author test is unresolved, and even the highest-scoring reader we probe stays far below its score on a weaker policy’s errors – two subjects on MBPP, licensing nothing about frontier models in general. A cheap static audit locates exposure before training; hardening the reward removes the measurement inflation, though here it buys little capability.
[NLP-53] Can a Language Model Learn Facts Continually in Its Weights?
【速读】： 该论文旨在解决持续学习（continual learning）中知识在模型权重中能否有效累积与保留的核心问题，特别是针对生成式AI（Generative AI）在不断接收新知识后，旧知识是否仍可被准确调用的挑战。其关键发现在于：知识在模型中的存储并非静态权重更新的结果，而是依赖于“问题键控”（question-keyed）的机制——即后续的权重写入会重新定向先前知识所对应的查询路径，导致即使知识未被物理擦除，也可能在行为上被遗忘。研究通过对比不同训练数据分布下的知识编码方式发现，仅使用单一陈述（bare-statement）训练导致知识仅能用于复述，而采用多样化重述（diverse restatements）的数据则显著缩小了复述与实际应用之间的差距（从27.4降至5.4个百分点），并使知识在二十次连续写入后仍保持46%的准确率。此外，研究揭示了“行为遗忘”现象：遗忘的知识虽仍保有大部分写入时增加的对数概率，但其响应模式受最近写入内容主导；然而，当遗忘的事实被显式提供于提示（prompt）中，其性能可恢复至77%-80%。这些结果表明，尽管广泛数据有助于构建可用知识，且冻结参考模型可维持能力，但任何基于局部测量或修正策略的干预均无法保证早期知识的可访问性。因此，论文提出，在需要知识组合或抵抗后续干扰的情境下，可靠的知识传递通道是上下文（context）而非模型权重本身。
链接: https://arxiv.org/abs/2607.11020
作者: Charles O’Neill
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Continual learning promises a language model that keeps acquiring knowledge after training, with each new fact written into its weights. Whether weight writes can support accumulation remains undecided. We follow invented facts written into Qwen3 models from creation through sequences of twenty to one hundred later writes, using held-out questions of five types, with the original model given the fact in its prompt as the reference. Across these experiments, the breadth of the training data determines the kind of knowledge created. Bare-statement training produces recitation, while diverse restatements reduce the recitation-to-use gap from 27.4 to 5.4 points without showing the model a conclusion. This difference carries into later writes: after twenty sequential writes, bare-statement facts retain 1% accuracy while facts written from broad study data retain 46%. We also find that facts can be behaviourally forgotten without being erased. Forgotten facts keep most of the log-probability added by their write, and under bare-statement training 70% of wrong answers about them contain the most recently written fact. The same writes barely degrade the model’s use of facts in context, and a forgotten study fact supplied in the prompt recovers to 77-80% on its questions. These results describe knowledge that is stored but question-keyed: later writes redirect the questions that reached it. Damage to unrelated abilities tracks KL divergence from the original model, and the later writes cause interference regardless of how the earlier fact was stored. Broad data can create usable knowledge, and a frozen reference can preserve capability, but no intervention we tested, including those built on accurate local measurements of each write, keeps earlier facts reachable. When facts must be composed or survive later writes, the reliable channel is context rather than the weights.
[NLP-54] EasyOPD: An Easy-to-use On-Policy Distillation Framework for Large Language Models
【速读】： 该论文旨在解决现有在线策略蒸馏（On-Policy Distillation, OPD）方法在监督形式、分词器兼容性、教师模型访问方式及监督粒度等方面存在显著差异，导致实现碎片化、难以复现与扩展的问题。其核心解决方案是提出一个名为 \textscEasyOPD 的统一框架，基于分布式强化学习框架 verl 构建，通过将用户侧配置、方法特异性监督逻辑与 verl 底层执行解耦，实现模块化设计。该框架通过标准化的扩展边界支持损失构建、回溯轨迹元数据处理、奖励计算、分词器对齐及教师端计算等关键环节，使多种典型 OPD 方法（如跨分词器蒸馏、自蒸馏、逐步蒸馏）可在同一 backend 上高效运行，同时保持各自的方法目标与任务依赖性能特征。实验在推理、代码生成、科学知识和工具使用等基准上验证了其通用性与有效性，并配套提供可运行的 YAML 配置、文档及演示包，显著提升了 OPD 方法的可复现性与可扩展性。
链接: https://arxiv.org/abs/2607.11012
作者: Jie Sun,Mao Zheng,Mingyang Song,Qiyong Zhong,Gengsheng Li,Zhepei Hong,Chang Wu,Pengfei Liu,Junfeng Fang,Xiang Wang
机构: University of Science and Technology of China (中国科学技术大学); LLM Department, Tencent (腾讯大语言模型部门); Shanghai Innovation Institute (上海创新研究院); National University of Singapore (新加坡国立大学)
类目: Computation and Language (cs.CL)
备注: 10 pages, 2 figures
Abstract:Conventional language-model distillation often relies on fixed teacher-generated data, which may not cover the states encountered by an evolving student policy. On-policy distillation (OPD) instead collects teacher or evaluator supervision on student-generated rollouts. However, existing OPD methods differ substantially in supervision form, tokenizer compatibility, teacher access, and supervision granularity, leading to fragmented implementations that are difficult to reproduce and extend. We present \textscEasyOPD, an on-policy distillation framework built on verl, a distributed reinforcement-learning framework for large language models. \textscEasyOPD separates user-side configuration, method-specific supervision logic, and verl-based execution. Its method modules connect to the shared backend through extension boundaries for loss construction, rollout metadata, reward processing, tokenizer alignment, and teacher-side computation. We instantiate representative methods for three OPD settings – cross-tokenizer OPD, on-policy self-distillation, and step-wise OPD. Experiments on reasoning, code-generation, scientific-knowledge, and tool-use benchmarks show that these implementations can be executed through the same verl-based backend while retaining their method-specific objectives and task-dependent performance profiles. We release \textscEasyOPD with runnable YAML configurations, documentation, and an installable demonstration package and video.
[NLP-55] he Nuts and Bolts of Natural Language to SQL Translation: A Systematic Analysis of Model Pipeline Optimisation Approaches and their Interactions
【速读】： 该论文旨在解决大语言模型时代下自然语言到SQL（NL2SQL）转换的开放性问题，其核心挑战在于如何在保证准确率的同时构建更轻量级的模型。解决方案的关键在于系统性地集成多种管道扩展模块：引入NatSQL中间表示以增强语义对齐，加入基于合成数据的预处理与微调步骤以提升泛化能力，并设计一种新颖的重排序（reranker）模型以优化最终束搜索中的SQL候选选择。通过在两种基础架构（SmBoP和RASAT）上进行消融实验并结合Shapley值分析，研究发现各组件的性能并非简单叠加，其有效性高度依赖于与基线系统及其他组件之间的相互作用，揭示了组件协同效应的重要性。
链接: https://arxiv.org/abs/2607.10911
作者: Filip Klubicka,Vasudevan Nedumpozhimana,Sneha Rautmare,Bora Caglayan,Mingxue Wang,John D. Kelleher
机构: ADAPT Research Centre; Trinity College Dublin; Huawei Ireland Research Centre
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Databases (cs.DB)
备注:
Abstract:In the age of large language models, Natural Language to SQL (NL2SQL) translation remains an open problem with many useful applications. We explore interactions between several NL2SQL pipeline extensions to inspire development of more lightweight models. Specifically, we integrate the NatSQL intermediate representation, include a preprocessing step and a fine-tuning step based on synthetic data, and develop a novel reranker model to improve SQL selection in the final beam. We perform an ablation study supplemented by a Shapley analysis of these different components integrated with two backbone architectures, SmBoP and RASAT. We find that simply combining all of them does not lead to best results, but that their impact depends on their interactions with the baseline system, as well as each other.
[NLP-56] LOGOS: A Living Logic for AI Agent Teams That Evolve With Humans
【速读】： 该论文旨在解决当前人工智能代理（AI agents）在持续演化过程中面临的可控性与可审计性问题，即如何在保障自动化效率的同时实现对代理行为的透明监管和人类主导的治理。其核心挑战在于：随着代理具备工具调用、任务委派、经验学习及自我修改能力，其演化过程可能脱离人类意图与安全边界，亟需一种既能支持自主进化又可确保责任追溯的机制。解决方案的关键是提出“logos”——一个可插拔的自演化与治理层，通过将多模态异构输入（如文档、图像、音频、数据库、API接口及人工指令）编译为版本化的“代理包”（agent packs），统一封装代理、工具、知识、测试、权限与策略等要素。在运行时，logos将代理活动转化为可移植、可审计的事件轨迹，并在不同框架与后端间实施“闭合失败验证”（fail-closed verification）。所有新生成的提示、记忆、技能、工具、角色或工作流均被视为未经信任的候选版本，必须经过隔离执行证据验证、人工策略控制与显式授权才能被正式采纳。这一架构实现了“可验证的人机闭环工程”：代理可在不中断连续运行的前提下自主行动、提问、学习并提出改进，而人类则能动态调控目标设定、权限分配、审批流程与不可逆操作，从而在机器级演化速度下维持人类对系统最终控制权。
链接: https://arxiv.org/abs/2607.10878
作者: Yuma Ichikawa,Yamato Arai,Kosaku Kimura,Akira Sakai,Hiromichi Kobashi
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 58 pages, 14 figures
Abstract:AI agents are evolving from answer engines into persistent teams that use tools, delegate work, learn from experience, and modify the artifacts that shape their future behavior. The defining question for deployment is no longer merely what agents can do, but who controls what they are allowed to become. We introduce logos, a pluggable layer for self-evolution and governance that strengthens existing multiagent frameworks rather than replacing them. logos compiles heterogeneous multimodal inputs, including documents, images, audio, tables, databases, APIs, and human instructions into versioned agent packs containing agents, tools, knowledge, tests, permissions, and policies. During operation, it transforms agent activity into portable, auditable event traces and applies fail-closed verification across frameworks and backends. Every learned prompt, memory, skill, tool, role, or workflow remains an untrusted release candidate until held-out execution evidence, human-controlled policy, and explicit authorization permit its promotion. This architecture enables “verifiable human-agent loop engineering”: agents can act, ask, learn, and propose improvements, while humans can steer objectives, permissions, approvals, and irreversible actions without interrupting continuous operation. logos provides a living logic for accountable automation. Agents may evolve at machine speed, but only evidence and human authority can close the loop.
[NLP-57] Capabilities of Claude Fable 5 on Biomedical Challenge Problems
【速读】： 该论文旨在解决当前生物医学领域语言模型评估中存在的两大核心问题：一是现有基准测试（benchmark）已接近性能饱和，导致评估结果难以反映模型的真实能力差异；二是开放性回答通常由其他语言模型进行评分，存在主观性和评价偏差。为克服上述问题，研究采用确定性评分策略，基于固定答案键对Claude Fable 5（Anthropic最新公开可用的高性能模型）在八个生物医学基准（四个文本类、四个多模态类）上的表现进行全面评估，并引入Claude前代模型及GPT-5作为基线。关键发现在于，Fable 5在不同任务中拒绝回答的比例介于8.0%至99.4%之间，这一拒答模式在前代模型和GPT-5中均未出现。当排除被拒绝的样本后，Fable 5在所有基准上的准确率均优于或等于其他模型。进一步分析揭示出两种可区分的拒答模式：其一集中于基础科学与机制类内容，在MedQA和MedXpertQA MM两个基准中通过各自类别标签独立验证；其二表现为罕见病基准（RareBench）中的疾病领域特异性拒答，即先天性代谢疾病相关提问几乎全部被拒绝，而成人起病的自身免疫性疾病提问则基本不被拒绝。因此，该研究的核心结论是：Fable 5在生物医学应用中的主要限制并非其能力不足，而是其拒答意愿（willingness to engage），一旦模型选择回应，其实际性能表现出显著优势。
链接: https://arxiv.org/abs/2607.10849
作者: Dominic Okonkwo,Magnus Hodgson,Temitope I. David,Susan Adanna Ihejirika
机构: University of Georgia (佐治亚大学); University of Illinois (伊利诺伊大学)
类目: Computation and Language (cs.CL)
备注: 15 pages, 6 tables, 4 figures, appendix with qualitative examples
Abstract:Frontier language models are increasingly evaluated on biomedical benchmarks, but two problems undermine most published evaluations: legacy benchmarks are near-saturated, and open-ended responses are graded by other language models. We evaluate Claude Fable 5, Anthropic’s most capable publicly available model, across eight biomedical benchmarks, four text and four multimodal, using deterministic scoring against fixed answer keys throughout. We include two Claude predecessors and GPT-5 as baselines. Refusal is tracked as a distinct outcome in every result table. That decision produces the paper’s central finding. Fable 5 refuses between 8.0% and 99.4% of questions depending on the benchmark, a pattern absent in both predecessors and in GPT-5. Once refused items are excluded from the denominator, Fable 5’s accuracy exceeds or meets every other model on every benchmark in this study. We identify two distinguishable refusal patterns: one concentrating in basic-science and mechanism content across MedQA and MedXpertQA MM, confirmed independently on two benchmarks using each benchmark’s own category labels; and a separate disease-domain pattern on RareBench, where inborn metabolic disease presentations are refused near-universally while adult-onset autoimmune presentations are not. The primary constraint on Fable 5’s biomedical usefulness is willingness to engage, not capability once it does.
[NLP-58] Quantifying the Sources of Instability in LLM -Based Stance Analysis of Public Discourse
【速读】： 该论文旨在解决计算社会科学中基于自动化预处理流程（如说话人辨识、语音识别转录清洗、句子分割等）从原始媒体数据生成可分析文本时所面临的稳定性问题。其核心挑战在于，同一输入数据在不同预处理流程或下游测量工具下可能产生差异性输出，从而影响研究结论的可靠性。解决方案的关键在于提出一个诊断框架，能够有效分离预处理流程本身（如说话人辨识方法、分句规则）与下游测量工具（如大语言模型标注与关键词词典）所带来的变异来源。研究通过分析256个来自41位公众人物的跨领域YouTube访谈数据发现：（1）预处理流程的敏感性主要集中在视频样本较少的说话人（N ≤ 5），而对于样本量充足的说话人（N ≥ 16），其引发的相关系数变化均值仅为0.13；（2）不同测量方法之间的分歧更大且具有系统性，即使在同一预处理流程下，大语言模型与关键词词典也常对情感效价与认识模态间的耦合方向给出相反判断；（3）整体情感比例高度稳定（Δp(¬) < 6个百分点），掩盖了上述两类不稳定因素。因此，该研究强调，在分析访谈数据中跨维度关系时，必须验证结论对预处理流程和测量方法选择的稳健性，尤其需审慎评估测量方法的影响。
链接: https://arxiv.org/abs/2607.10846
作者: Bo Chen
机构: Institute of Computing Technology, Chinese Academy of Sciences(中国科学院计算技术研究所)
类目: Computation and Language (cs.CL)
备注:
Abstract:Computational social science increasingly relies on automated preprocessing pipelines – speaker diarization, ASR transcript cleaning, sentence segmentation – to convert raw media into analyzable text. When these pipelines produce different outputs from the same input, two distinct sources of instability can arise: the preprocessing pipeline itself (diarization method, segmentation rules) and the downstream measurement instrument (LLM annotation vs.\ keyword lexicon). Using 256 YouTube interviews across 41 public figures from five domains, we compare two speaker-diarization pipelines and two measurement methods, all targeting the coupling between affective valence and epistemic modality. We find that (1) preprocessing pipeline sensitivity is concentrated in speakers with limited video samples (N \leq 5 ); for the four best-sampled speakers (N \geq 16 ), the mean absolute pipeline-induced change in r(\textneg, \textemph) is only 0.13 ; (2) cross-method disagreement is larger and more systematic – the LLM and keyword-lexicon methods assign opposite coupling directions to several well-sampled speakers, even within the same preprocessing pipeline; and (3) aggregate valence proportions are highly stable ( |\Delta p(\textneg)| 6 pp) regardless of pipeline or method, masking both sources of instability. The contribution is a diagnostic framework that separates pipeline effects from measurement effects: researchers studying cross-dimensional relationships in interview data should verify that their conclusions are robust to both sources of variation, with particular attention to measurement method choice.
[NLP-59] Route Communicate and Reason : Gated Routing and Adaptive Depth for Efficient Multi-Agent Reasoning
【速读】： 该论文旨在解决多智能体集成（multi-agent ensembling）中因盲目扩展活跃参数与推理开销而引发的效率瓶颈问题，具体聚焦于三个核心挑战：应选择哪些智能体进行咨询、查询在智能体层级结构中应深入到何种程度，以及跨智能体通信是否值得其带来的计算成本。为应对上述问题，提出GRADE（Gated Routing and Adaptive Depth for Efficient Reasoning）框架，其关键在于引入四个轻量级可学习门控机制，协同控制智能体选择、层级深度、跨智能体通信决策及分支剪枝，从而实现动态高效的推理路径规划。该系统采用无评判器的协作式组相对策略优化（CoGRPO），通过共享优势信号对所有参与滚动更新的门控模块与智能体进行联合训练，显著提升训练稳定性与效率。此外，基于可热插拔专家注册表（Expert Registry）与每智能体校准映射机制，支持推理阶段无需重训练即可更换专家模型，保障系统的灵活性与安全性。实验表明，在平均仅约170亿活跃参数下，GRADE在GSM8K、MMLUPro和GPQA等基准上均优于现有方法，尤其在MMLUPro上以半数活跃计算量超越最强基线4.8个百分点；在强调模型深度的AIME-2025任务中仍保持竞争力。消融实验进一步验证了层级结构与掩码交叉注意力机制是性能提升的关键因素，并证实每智能体校准对于安全热插拔至关重要。
链接: https://arxiv.org/abs/2607.10836
作者: Sudipto Ghosh,Tanmoy Chakraborty
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Multi-agent ensembling multiplies active parameters and inference cost without answering three basic questions: which agents to consult, how deeply a query should traverse a hierarchy of agents, and when inter-agent communication is worth its cost. We present GRADE (Gated Routing and Adaptive Depth for Efficient Reasoning), a hierarchical multi-agent system in which four lightweight learned gates jointly govern agent selection, hierarchy depth, inter-agent communication, and branch pruning. Training uses CoGRPO (Collaborative Group-Relative Policy Optimization), a novel critic-free recipe that adapts GRPO to multi-agent hierarchies and assigns a shared advantage signal to every gate and agent that participated in a rollout. Agent models are drawn from a hot-swappable Expert Registry; per-agent calibration maps allow experts to be replaced at inference time without retraining. At \sim 17B average active parameters, GRADE outperforms all baselines on GSM8K, MMLUPro, and GPQA, surpassing the strongest baseline by 4.8 points on MMLUPro at half the active compute. On AIME-2025, where model depth dominates, GRADE remains competitive to existing frameworks. Ablations isolate the hierarchy and masked cross-attention as the largest contributors to accuracy, and show that per-agent calibration is necessary for safe hot-swapping.
[NLP-60] Large Language Models for Token-Efficient and Semantic-Preserving Opinion Summarization
【速读】： 该论文旨在解决大规模、冗余且不平衡的主观文本（如产品评论、酒店反馈和社交平台帖子）在基于大语言模型（LLM）进行意见摘要时，难以有效保持观点多样性与语义忠实性的问题。其核心挑战在于如何在降低计算成本与生成效率的前提下，确保摘要能够全面覆盖不同维度的观点（如情感倾向、主题分布），并避免代表性偏差。解决方案的关键在于提出一种融合多维分类（如情感、主题）与分层采样策略的框架：首先通过多维分类对海量意见进行结构化标注，再采用分层采样方法筛选出紧凑但具有代表性的意见子集，从而减少输入至LLM的令牌数量；随后设计定制化提示（prompt），引导模型生成内容覆盖全面、观点平衡且语义保真度高的摘要。实验结果表明，该方法在亚马逊产品评论、Tripadvisor酒店评论及X/Twitter帖子数据集上均显著降低了令牌使用量与计算开销，同时在内容覆盖率、观点平衡性和语义保留方面优于传统AI方法及标准LLM摘要基线。
链接: https://arxiv.org/abs/2607.10825
作者: Fabrizio Marozzo,Stefano Iannicelli
机构: University of Calabria (卡莱布里亚大学); University of Calabria (卡莱布里亚大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Opinionated text - spanning product reviews, hotel feedback, and social posts - captures rich signals about user experiences, preferences, and concerns. However, the scale, redundancy, and imbalance of such corpora make it challenging to analyze opinions effectively, particularly when the goal is to generate summaries that remain faithful to the diversity of viewpoints expressed. This paper presents a framework that preserves semantics in LLM-based opinion summarization while minimizing token usage. We combine multidimensional classification (e.g., sentiment, topics) with a family of stratified sampling strategies to select compact yet representative subsets of opinions before prompting the LLM. Tailored prompts then produce balanced summaries that surface the salient aspects expressed in the opinions (e.g., strengths and weaknesses of products/hotels). Experiments on Amazon product reviews, Tripadvisor hotel reviews, and X/Twitter posts demonstrate that our method significantly reduces token usage and computational cost while consistently outperforming traditional AI-based and standard LLM summarization baselines in terms of content coverage, balance, and semantic preservation.
[NLP-61] Abstractiveness Metrics for Evaluating Text Summarization: A Refined Formulation with Empirical Validation
【速读】： 该论文旨在解决文本摘要评价中难以量化抽象性（abstractiveness）的问题，尤其针对传统表面指标（如ROUGE）无法有效区分抽取式与生成式摘要的局限。其核心挑战在于如何客观衡量摘要在多大程度上脱离了源文本的直接复制，从而反映模型的真正语义压缩与重构能力。解决方案的关键在于提出一组基于原理的启发式度量：参考抽象性（Reference Abstraction, RA）、摘要抽象性（Summary Abstraction, SA）以及抽象比率（Abstraction Ratio, AR），通过文档长度的调和平均结合三次非重叠因子（cubic non-overlap factor），实现维度一致、有界且对抽取-生成边界具有非线性敏感性的量化结果。实验表明，该方法能有效区分抽取式模型（SA ≈ 0.12–0.26）与生成式模型（SA ≈ 0.96–1.77），并识别出可能包含幻觉（hallucination）需人工评估的摘要，显著提升了摘要质量评估的深度与可靠性。
链接: https://arxiv.org/abs/2607.10806
作者: Praveenkumar Katwe,Rakesh Chandra Balabantaray,Kali Prasad Vittala
机构: International Institute of Information Technology, Bhubaneswar (国际信息科技学院，布巴内斯瓦尔); Salesforce India Pvt Ltd (Salesforce印度私人有限公司)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 13 pages, 8 figures, code at this https URL . Extended and revised version of: Katwe et al., IEEE OCIT 2022 (doi: https://doi.org/10.1109/OCIT56763.2022.00022 )
Abstract:Quantifying abstractiveness in generated summaries is essential for evaluating summarization models beyond surface-level metrics like ROUGE. We introduce Reference Abstraction (RA), Summary Abstraction (SA), and Abstraction Ratio (AR) – a set of principled heuristic metrics that measure how much a summary diverges from extractive copying of the source text. The formulation uses the harmonic mean of document lengths modulated by a cubic non-overlap factor, yielding dimensionally consistent, bounded output with non-linear sensitivity to the extractive-abstractive boundary. Evaluation on 100 XSUM documents across four summarization models (BART-large-cnn, Pegasus-xsum, DistilBart, MT5-small) demonstrates that the metrics successfully discriminate between extractive models (SA ~ 0.12-0.26) and abstractive models (SA ~ 0.96-1.77), and that the Abstraction Ratio identifies summaries requiring manual evaluation for potential hallucination. Code and results are available at this https URL. Comments: 13 pages, 8 figures, code at this https URL. Extended and revised version of: Katwe et al., IEEE OCIT 2022 (doi:https://doi.org/10.1109/OCIT56763.2022.00022) Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI) MSC classes: 68T50, 94A17 ACMclasses: I.2.7; H.3.1 Cite as: arXiv:2607.10806 [cs.CL] (or arXiv:2607.10806v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2607.10806 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Praveenkumar Katwe [view email] [v1] Sun, 12 Jul 2026 15:25:57 UTC (1,359 KB) Full-text links: Access Paper: View a PDF of the paper titled Abstractiveness Metrics for Evaluating Text Summarization: A Refined Formulation with Empirical Validation, by Praveenkumar Katwe and 2 other authorsView PDFHTML (experimental)TeX Source view license Current browse context: cs.CL prev | next new | recent | 2026-07 Change to browse by: cs cs.AI References Citations NASA ADSGoogle Scholar Semantic Scholar export BibTeX citation Loading… BibTeX formatted citation loading… Data provided by: Bookmark checked="checked"class=“labs-tab-input”> Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv’s community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?) mathjaxToggle(); We gratefully acknowledge support from our major funders, member institutions, , and all contributors. About Help Contact Subscribe Copyright Privacy Accessibility Operational Status (opens in new tab) Major funding support from
[NLP-62] Diagnosing and Mitigating Thinking Collapse in On-Policy Self-Distillation
【速读】： 该论文旨在解决生成式 AI（Generative AI）在复杂推理任务中，基于策略自蒸馏（On-Policy Self-Distillation, OPSD）方法导致下游性能下降的悖论问题。其核心挑战在于，OPSD在高熵决策分支处引发严重的“思维坍缩”（Thinking Collapse）现象——即模型内在中间推理行为显著退化，表现为认知性词元密度（epistemic-token density, ET per 1k）急剧下降。研究通过熵基梯度掩码与词元级目标分析揭示，该现象源于教师模型在高熵节点施加的过强梯度，导致学生模型的认知性词元频繁被压制至非认知性目标，并高度聚集于学生-教师点对点差异较大的区域。为此，论文提出自适应双视角自蒸馏（Adaptive Dual-Perspective OPSD, AD-OPSD），其关键创新在于引入一种非对称点对点差异门控机制，动态地将高抑制风险的沙盒词元锚定至由冻结基础模型导出的参考先验分布，从而在保留自蒸馏纠错能力的同时，有效维持模型原有的思维生成能力。实验结果表明，AD-OPSD在多个数学推理基准上相较标准OPSD平均提升达+4.1%绝对准确率，且能有效缓解思维坍缩，在不同模型规模和后训练范式下均表现出强鲁棒性。
链接: https://arxiv.org/abs/2607.10805
作者: Keqin Peng,Chen Li,Yuanxin Ouyang,Yancheng Yuan,Liang Ding
机构: Beihang University (北京航空航天大学); Hong Kong Polytechnic University (香港理工大学); Alibaba Group (阿里巴巴集团)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:On-Policy Self-Distillation (OPSD) has emerged as a crucial paradigm for enhancing and aligning Large Language Models (LLMs). However, in complex reasoning tasks, OPSD paradoxically degrades downstream performance. In this paper, we systematically investigate this pathology and identify a severe optimization trap we define as \textbfThinking Collapse – a sharp decline in the model’s native intermediate reasoning behavior, measured by epistemic-token density (ET per 1k). Through entropy-based gradient masking and token-level target analysis, we show that this collapse is triggered by aggressive teacher gradients at high-student-entropy decision forks, where student epistemic tokens are frequently suppressed into teacher non-epistemic targets and are highly concentrated in high pointwise student-teacher divergence regions. To resolve this optimization pathology, we propose \textbfAdaptive Dual-Perspective OPSD (AD-OPSD), a robust control framework that dynamically moderates the self-distillation objective. AD-OPSD selectively anchors high-suppression-risk sandboxed tokens to a reference prior derived from the frozen base model via an asymmetrical pointwise divergence gate, preserving native thinking capacity while retaining OPSD’s error-correcting power. Extensive experiments across competitive mathematical benchmarks show that AD-OPSD improves over standard OPSD by up to \textbf+4.1% absolute average accuracy across diverse model scales and datasets. Further analysis demonstrates that AD-OPSD mitigates thinking collapse and generalizes robustly to different post-training paradigms.
[NLP-63] rust Before Fusion: QIMG-7 and Source-Aware Resolution for Polluted Multimodal RAG
【速读】： 该论文旨在解决多模态检索增强生成（Multimodal Retrieval-Augmented Generation, RAG）在真实场景下因检索结果污染而导致的可靠性问题。现有评估通常基于干净证据，但实际检索可能返回语义相关却不可靠的内容，如由损坏元数据、实体替换、拼写错误叠加、语义编辑、对抗性补丁、图像融合或风格迁移等造成的虚假文本和误导性图像。为系统评估此类污染影响，研究提出QIMG-7基准，涵盖四个数据集、七类图像攻击方式及16种成对的清洁/污染设置，每种方法共1,760个评估样本。实验表明，简单的多模态融合策略在污染环境下极为脆弱：以gpt-4o-mini为例，全模态（Full-MM）支持率从清洁文本下的0.908骤降至污染情况下的0.490，此时参数化回退（Parametric fallback）反而更可靠。为此，论文提出无需训练的源感知可信度解析（Source-Aware Trust Resolution, SATR）方法，通过比较参数化、纯文本与全模态候选答案，并依据来源可靠性进行选择或回退。其中，场选择器（Field-Selector）变体取得最佳平衡得分0.816，较全模态提升11.7分，优于级联路由器（Cascaded Router）2.7分。消融实验进一步表明，在以文本优先的设定下，显式建模文本可信度是性能提升的主要驱动力。综上，研究支持在存在多模态检索冲突的事实性问答任务中采取选择性信任策略，而非无条件融合。
链接: https://arxiv.org/abs/2607.10798
作者: Saadeldine Eletter,Owais Aijaz,Preslav Nakov
机构: Mohamed bin Zayed University of Artificial Intelligence (穆罕默德·本·扎耶德人工智能大学)
类目: Computation and Language (cs.CL)
备注: 23 pages, 6 figures, 23 tables. Preprint under review
Abstract:Multimodal retrieval-augmented generation (RAG) is often evaluated with clean evidence, yet real retrieval can return topically relevant but unreliable content: false text and misleading images from corrupted metadata, entity swaps, typographic overlays, semantic edits, adversarial patches, blends, or style transfer. We introduce QIMG-7, a controlled benchmark for multimodal retrieval pollution in multi-sentence factual QA, spanning four datasets, seven image-attack families, and 16 paired clean/polluted regimes, for 1,760 evaluation rows per method. Across four generator/gate stacks, naive multimodal fusion is brittle: in the main gpt-4o-mini stack, Full-MM support drops from 0.908 with clean text to 0.490 with polluted text, often making Parametric fallback safer than retrieval. We propose source-aware trust resolution (SATR), a training-free approach that compares Parametric, Text-only, and Full-MM candidate answers and selects among candidate answers or falls back based on source reliability. The Field-Selector variant achieves the best balanced score, 0.816, improving over Full-MM by 11.7 points and over the Cascaded Router by 2.7 points. Ablations show that, in this text-first setting, explicit text-reliability modeling is the dominant driver of these gains. Overall, in text-first factual QA with multimodal retrieval conflict, our results support selective trust rather than unconditional fusion. Artifacts are available at this https URL.
[NLP-64] STEC: Evidence Compression for Deep Search in Open-domain Multi-Hop QA
【速读】： 该论文旨在解决开放域多跳问答（open-domain multi-hop QA）中基于大语言模型（LLM）的搜索代理在生成多个搜索轨迹后所面临的最终答案选择难题。现有方法虽在推理范式、检索交互和搜索策略优化方面取得进展，但多条轨迹带来的异构性、冗余性、不完整性及冲突性信息，使得直接比较原始轨迹或仅对比答案字符串均难以实现可靠的选择。其关键解决方案是提出一种名为STEC（Evidence Compression for Final Answer Selection）的证据压缩框架，核心在于将最终答案选择从原始轨迹层面转移到候选答案层面：首先通过答案级证据压缩（Answer-Level Evidence Compression），依据归一化答案一致性对轨迹进行分组，并为每个候选答案生成专属的证据表示；其次通过证据引导的答案验证（Evidence-Guided Answer Verification），对比这些压缩后的证据表示以实现更精准的最终答案选择。该设计有效缓解了噪声与语义错位问题，显著提升了多跳问答中答案选择的鲁棒性与准确性。
链接: https://arxiv.org/abs/2607.10795
作者: Xinkang Li,Rong Jiang,Xin Song,Ye Wang,Yue Han,Changjian Li
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:In open-domain multi-hop question answering (QA), LLM-based search agents offer a promising approach to knowledge-intensive QA by combining retrieval with reasoning. Existing methods mainly improve open-domain multi-hop QA through reasoning paradigms, retrieval interaction, and search strategy optimization. However, using multiple search trajectories introduces a challenging final answer selection problem. Different trajectories may support different candidates, and the retrieved information can be heterogeneous, redundant, incomplete, or conflicting. Directly comparing raw trajectories exposes the verifier to noisy and unaligned content, while comparing answer strings ignores the evidence supporting each candidate, making reliable final selection difficult. To address this challenge, we propose STEC, an evidence compression framework for final answer selection in multi-hop QA. STEC selects the final answer from the existing candidate set through two mechanisms: (1) Answer-Level Evidence Compression, which groups trajectories by normalized answer identity and converts each answer group into a candidate-specific evidence representation; and (2) Evidence-Guided Answer Verification, which compares these representations and selects the final answer from the candidate set. The design shifts final selection from raw trajectory comparison to candidate-level evidence comparison. We evaluate STEC on four open-domain multi-hop QA benchmarks against representative baselines. Experimental results show that STEC performs best overall among the compared methods, and ablation results provide evidence that answer-level evidence compression contributes to final answer selection.
[NLP-65] Detecting AI-Generated Video: A Vision-Language Dual-View Survey ACL2026
【速读】： 该论文旨在解决生成式视频（AIGC-V）日益逼近真实世界表现所带来的传统基于人工痕迹检测方法失效的问题，其核心挑战在于如何应对高保真度生成内容所引发的语义层面真实性验证难题。解决方案的关键在于提出一种“事实一致性验证”（Factual Fidelity Verification）的新范式，将检测任务从低层次的视觉伪影识别转向高层次的语义真实性判断，即评估视频中事件、实体及物理过程是否符合现实世界常识。为此，论文构建了视觉-语言双视角（Vision-Language Dual-View）分类体系，系统化地将现有方法归纳为四个层级：内在线索分析、时空一致性建模、跨模态一致性推理以及语言引导的世界级推理，凸显了从传统深度伪造检测中的特征匹配向依托视觉-语言模型与智能体推理流程实现证据驱动的语义验证的根本性转变。该框架不仅整合了221项相关研究，还梳理了生成范式、检测方法、评估指标与基准数据集的发展脉络，并指出现有挑战与可信赖、可解释性强的检测技术未来发展方向。
链接: https://arxiv.org/abs/2607.10787
作者: Dylan Xinming Hou,Juntian Zhang,Xu Gu,Yichen Wu,Nils Lukas,Gus Xia,Xiuying Chen,Yuhan Liu
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL)
备注: 51 pages, accepted by ACL 2026
Abstract:The evolving realism of AI-generated Videos (AIGC-V) is rapidly rendering traditional artifact-centric detection insufficient, necessitating a paradigm shift from low-level inspection to high-level semantic verification. This paper presents a comprehensive survey of AIGC-V detection, reframing the task as Factual Fidelity Verification, which asks whether the events, entities, and physical processes depicted in a video are consistent with real-world facts. To systematize this rapidly evolving field, we propose a Vision-Language Dual-View taxonomy that organizes existing methods into a hierarchical, four-layer landscape, spanning intrinsic cue analysis, spatiotemporal consistency modeling, cross-modal consistency reasoning, and language-guided world-level reasoning. This dual-view framing highlights a fundamental transition from artifact matching in traditional deepfake detection to evidence-based semantic verification enabled by vision-language models and agentic reasoning pipelines. Based on a systematic review of 221 works, we synthesize AIGC-V generation paradigms, survey the landscape of detection methods, and review evaluation metrics and benchmarks in line with proposed views. Finally, we discuss current challenges and identify promising directions toward robust, explainable, and trustworthy detection.
[NLP-66] he First ChineseBabyLM Challenge: training data-efficient and cognitively plausible language models for Chinese
【速读】： 该论文旨在解决中文小样本语言模型（Chinese Language Model）在缺乏大规模标注数据和统一评估标准背景下，如何从零开始训练并有效评估其在自然语言理解（NLU）、认知对齐（cognitive alignment）以及汉字知识（Hanzi knowledge）等多维度能力的问题。其解决方案的关键在于构建一个开放且公平的基准挑战——“ChineseBabyLM”，要求参赛者使用1亿个中文语料token从头训练模型，不限制分词器、模型架构或训练轮次，通过三个核心任务赛道全面评估模型的综合语言能力与认知水平，从而推动中文生成式AI（Generative AI）在基础语言建模层面的发展与标准化。
链接: https://arxiv.org/abs/2607.10745
作者: Siyuan Song,Zhiheng Qian,Yunhao Zhang,Linyang He,Xiaozhe Ji,Yingxin Lin,Hongao Zhu,Chongtian Shao,Chuhan Lang,Luan Li,Rui Wang,Renfen Hu,Shaonan Wang,Hai Hu
机构: Princeton University; Shanghai Jiao Tong University; Chinese Academy of Sciences; Columbia University; Beijing Normal University; Tsinghua University; University of California San Diego; The Hong Kong Polytechnic University
类目: Computation and Language (cs.CL)
备注: 8 pages, 4 tables; work in progress
Abstract:This paper describes the first ChineseBabyLM challenge, which will be held in the 2026 NLPCC conference. The challenge calls for researchers to train language models from scratch with 100 million Chinese tokens and evaluates the models on 3 tracks of tasks: NLU, cognitive alignment and Hanzi knowledge. There is no restriction on tokenizer, model architecture and the number of training epochs. Details of the challenge can be found in this https URL.
[NLP-67] o Answer or to Abstain: Mitigating Search-Agent Hallucinations via Abstention-Aware Reinforcement Learning
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在开放域问答任务中因检索失败导致幻觉（hallucination）加剧的问题。现有训练范式虽通过奖励正确答案提升了性能，却未对错误生成内容进行有效惩罚，尤其在检索失败时，模型仍倾向于生成看似合理但虚假的答案，从而损害可靠性。为此，论文提出了一种自适应弃权强化学习（Abstention-Aware Reinforcement Learning, AWA-RL）方法，其核心在于动态调整弃权（abstention）的奖励机制，结合模型对特定查询的先验能力判断与持续的在线策略观测，实现对模型生成行为的精细化引导。此外，论文引入新型评估指标RA-F1，用于衡量模型能力与可靠性之间的权衡。实验表明，相较于不支持弃权的基线模型，AWA-RL在仅小幅牺牲原始准确率的前提下，将绝对精确率提升最高达10.3%，整体RA-F1提升2.9%，验证了该方法在构建高能力且高可靠搜索代理方面的有效性。
链接: https://arxiv.org/abs/2607.10738
作者: Fengji Zhang,Tianyu Fan,Yuxiang Zheng,Xinyao Niu,Chengen Huang,Jacky Keung,Bei Chen
机构: 未知
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Recent advances in equipping Large Language Models (LLMs) with search tools and outcome-reward reinforcement learning (RL) have achieved new state-of-the-art results on open-domain QA tasks. However, we argue that current training paradigms harbor a critical vulnerability: they predominantly reward correct answers but fail to penalize fabricated ones when retrieval fails, thereby implicitly exacerbating hallucinations. To address this, we propose Abstention-Aware Reinforcement Learning (AWA-RL), which dynamically shapes the abstention reward utilizing the model’s query-specific prior capabilities and continuous on-policy training observations. We also introduce a novel metric, RA-F1, to measure the capability-reliability trade-off. Compared to non-abstaining baselines, AWA-RL boosts absolute precision by up to 10.3% and overall RA-F1 by 2.9%, with only marginal sacrifice in raw accuracy. These results confirm that AWA-RL successfully yields highly capable and reliable search agents. The code, data, and model weights are publicly available at this https URL.
[NLP-68] A Corpus of Persuasion Techniques in Slavic Languages LREC2026 ACL2025
【速读】： 该论文旨在解决跨语言语境下说服性修辞策略（persuasion techniques）在斯拉夫语系文本中自动识别与分类的难题，尤其针对保加利亚语、波兰语和俄语等语言缺乏系统性标注数据的问题。其核心挑战在于如何在多语言、多主题背景下实现对说服性手段的精准标注与模型化建模。解决方案的关键在于构建首个涵盖三种斯拉夫语言的大型精细化标注语料库，该语料库以粗粒度文本片段（text-span level）和细粒度句子（sentence level）为单位，基于包含25种细粒度技巧的分类体系进行标注，并将其归入六大类修辞说服策略框架下。该语料库共包含222篇文档中的约7500个文本片段，覆盖国内外高度争议性议题。研究进一步采用传统机器学习与生成式AI模型作为基线方法，在文本片段与句子层面提供检测与分类的基准性能，为后续跨语言说服分析任务奠定了数据与评估基础。
链接: https://arxiv.org/abs/2607.10715
作者: Jakub Piskorski,Dimitar Iliyanov Dimitrov,Marina Ernst,Jacek Haneczok,Michał Marcińczuk,Arkadiusz Modzelewski,Roman Yangarber
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Corpus used for SlavicNLP 2025 Workshop co-located with ACL 2025, Published at LREC 2026
Abstract:Persuasion techniques are powerful rhetorical devices used to sway public opinion in a wide range of media. We present a new corpus of persuasion techniques, focusing on Slavic languages. The corpus contains documents in Bulgarian, Polish, and Russian, annotated with persuasion techniques at the coarse-grained text-span level and fine-grained sentence level. The techniques are drawn from a taxonomy of 25 fine-grained persuasion techniques, grouped under six broad categories of rhetorical persuasion strategies. The corpus contains approximately 7500 text spans from 222 documents that cover topics hotly debated at the national and international levels. We describe the corpus creation process, provide detailed statistics, and examine correlations between topics and persuasion techniques. We use classic ML-based and generative AI-based models to provide baselines and benchmark results for the detection and classification of persuasion techniques at the text-span level and sentence level.
[NLP-69] From Self-Attention to Connection Laplacian: A Unified Operator View of Transformers
【速读】： 该论文旨在解决自注意力机制（self-attention）在操作层面的几何结构理解不充分的问题，尤其是其在序列建模中的内在数学本质与经典几何算子之间的关联。其核心解决方案在于提出一种“连接行走”（connection walk）的形式化框架：将标记序列视为定义在标记位置图上的向量场，将注意力机制建模为通过非负行走矩阵聚合信息并沿边由可学习线性映射传输消息的过程。在此框架下，研究证明单头注意力（SHA）等价于具有恒定传输的连接传播步，而多头注意力（MHA）则精确对应于依赖边的连接行走，其有效传输为各头传输的注意力加权混合。此外，论文明确了当生成器退化为随机行走连接拉普拉斯算子时的条件，揭示了随机性、可逆性及度量相容传输的关键作用。实证结果表明，从124M到8B参数规模、涵盖编码器与解码器结构的训练后Transformer模型均表现出与该理论一致的几何特征：深层中有效注意力图收敛至稳定的几何算子，学习到的传输映射自发组织为近似缩放等距变换，且这些现象随模型规模增大而持续增强。综上，该工作建立了自注意力与经典几何算子之间的精确联系，并提供了一套基于算子层面的几何分析工具，推动了对Transformer模型的深层理解。
链接: https://arxiv.org/abs/2607.10677
作者: Binbin Lin,Wei Chen,Yalun Li,Wenxiao Wang,Jieping Ye,Xiaofei He
机构: Zhejiang University; Alibaba Cloud
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: 29 pages, 10 figures
Abstract:Self-attention is a ubiquitous primitive in modern sequence models, yet its operator-level geometry is only partially understood. We view a token sequence as a vector field over the token-position graph and identify attention as a connection walk: messages are aggregated by a nonnegative walk matrix while being transported along each edge by a learned linear map. Within this framework, we prove that single-head attention (SHA) is exactly a connection propagation step with constant transport, and that multi-head attention (MHA) is exactly a single edge-dependent connection walk whose effective transport is an attention-gated mixture of headwise transports. We further clarify the conditions under which the corresponding generator reduces to a random-walk connection Laplacian, highlighting the roles of stochasticity, reversibility, and metric-compatible transports. Empirically, we find that trained Transformers across scales (from 124M to 8B) and structures (encoder/decoder) exhibit geometric structure consistent with our theory: effective attention graphs converge to stable geometric operators in deeper layers, learned transports self-organize into approximate scaled isometries, and both phenomena strengthen consistently with scale. Overall, the paper provides a precise connection-walk formalism that links self-attention to classical geometric operators, along with a set of operator-level tools for analyzing transformer models from a geometric perspective.
[NLP-70] Unlocking Parallelism in Autoregressive Language Models via Speculative Decoding with Progressive Tree Drafting
【速读】： 该论文旨在解决传统推测解码（speculative decoding）在实际应用中因依赖辅助草稿模块而带来的训练与通信开销过高的问题，同时克服现有方法在利用目标模型自身潜在并行能力时因缺乏结构化协调而导致性能受限的瓶颈。其核心解决方案是提出一种名为**渐进式树形草稿生成（Progressive Tree Drafting, PTD）**的新方法，通过引入具有层级结构的渐进式树形架构与分步剪枝机制，实现对大语言模型（LLM）在单次前向传播中并行探索多个语义路径的有效引导。该设计不仅充分释放了模型的内在并行潜力，还保障了草稿生成的多样性与上下文一致性。实验表明，PTD在不需额外训练且适用于任意预训练模型的前提下，可在多个基准测试上实现最高达2倍的解码加速，显著提升了推理效率。
链接: https://arxiv.org/abs/2607.10661
作者: Zipeng Gao,Zhi Zheng,Qingrong Xia,Junda Lin,Ziwei Zhao,Tong Xu,Zhefeng Wang,Enhong Chen
机构: University of Science and Technology of China (中国科学技术大学); Huawei Technologies Co., Ltd. (华为技术有限公司)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Speculative decoding has significantly accelerated Large Language Model (LLM) inference by alleviating memory-bound bottlenecks. However, traditional speculative decoding typically relies on auxiliary draft modules, incurring significant training and communication overhead. Although recent methods attempt to generate drafts within the target model itself, they often fail to fully exploit its latent parallel capacity due to a lack of structural coordination. In this paper, we propose \textbfProgressive Tree Drafting (PTD), which employs a structured, guided parallel drafting strategy to harness the model’s parallel potential. By coupling a progressive tree structure with a stepwise pruning mechanism, PTD actively guides the LLM to explore multiple semantic paths in a single forward pass, ensuring both draft diversity and coherence. Experiments demonstrate that PTD achieves up to 2\times decoding speedup across various benchmarks while remaining training-free and model-agnostic. Our code is available at: this https URL.
[NLP-71] Knowledge Distillation for Automated AI Tutor Evaluation
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在基础教育与高等教育中快速应用背景下，缺乏可靠方法对其教学质量进行评估的问题。当前，尽管生成式AI（Generative AI）在教育领域广泛应用，但针对AI助教（AI Tutor）的自动化评估体系仍不成熟。为应对这一挑战，研究提出FATE（FLC AI Tutor Evaluator），一个专用于评估AI助教教学能力的80亿参数语言模型。其核心解决方案在于：基于BEA 2025共享任务定义的四大评价维度——错误识别（Mistake Identification）、错误定位（Mistake Location）、指导性（Guidance）和可操作性（Actionability），构建专业化评估框架；针对教学评估任务标注数据稀缺的瓶颈，采用前沿大模型知识蒸馏（Knowledge Distillation）技术生成高质量监督信号，显著提升模型性能，最高实现22.63个百分点的绝对性能增益。实验结果表明，FATE具备作为自动化评估工具的有效性，能够对主流商业模型如ChatGPT、Claude、Gemini和DeepSeek生成的教学响应进行系统性评测，验证了其在真实场景中的实用价值。
链接: https://arxiv.org/abs/2607.10647
作者: Tahmid Al Hannan,Diego Garcia,Alex Njoroge,Suha Al Juboori,Tarek Sakakini
机构: Folsom Lake College (福尔松湖学院)
类目: Computation and Language (cs.CL)
备注:
Abstract:The rapid integration of Large Language Models (LLMs) into K-12 and higher education has outpaced the development of reliable methods for evaluating their pedagogical quality. As the research community starts to explore the space of automating evaluation of AI tutors, we introduce FATE (FLC AI Tutor Evaluator), a specialized 8B-parameter language model designed to evaluate AI tutors. Aligned with the four core evaluation tracks from the BEA 2025 Shared Task, our model assesses pedagogical ability across Mistake Identification, Mistake Location, Guidance, and Actionability. Because pedagogical evaluation is a specialized task with limited labeled data, we leverage knowledge distillation from a frontier LLM to generate additional supervision, yielding absolute performance gains up to 22.63 percentage points. Finally, we demonstrate FATE’s utility as an automated evaluator by benchmarking instructional responses generated by popular commercial models, including ChatGPT, Claude, Gemini, and DeepSeek. On average, we have found that Gemini 2.5 Flash perfomed best (82.88%), then ChatGPT 5.5 Instant (80.75%), followed by DeepSeek V4 Flash (80.13%) and Claude Sonnet 4.6 (74.00%).
[NLP-72] Eval-Pair Matrix: Answer-Paired Meta-Evaluation of LLM Judges for Grounded RAG
【速读】： 该论文旨在解决在检索增强生成（Retrieval-Augmented Generation, RAG）评估中，使用同一模型家族作为生成器和评判者所导致的自偏袒（self-leniency）问题，即模型可能因自身生成倾向而对自身输出过度宽容。其解决方案的关键在于提出一种受控的元评估协议——Eval-Pair Matrix，通过在原始问题与溯源段落基础上，为每条记录引入一个隐藏的、因果相关的矛盾（hidden answer-causal contradiction），并利用GPT、Grok和Gemini等模型对扰动后的段落生成答案，再以相同模型作为盲评者对生成答案与原始段落进行评估。该设计构建了一个3×3交叉矩阵，包含300个核心记录、897个标注的生成结果及2683个评判结论，主分析基于275个完全验证记录。研究创新性地采用“答案配对”（answer-paired）方法，即在同一候选答案上配对同一模型作为评判者，从而更准确估计同模型效应；结果显示，对角线与非对角线的F1值相近，且同模型召回率差异接近零（-0.5个百分点；95%聚类自助置信区间[-2.7, +1.7]），唯一显著的配对差距是规避诱导主张的答案在匹配判别中的标记率更低（-4.3个百分点）。进一步的人工评估表明，看似误报的案例实为源错误检测、标签判断失误或模糊情形，并无真实误报被确认。因此，论文强调方法论启示：RAG的评判研究应报告完整评估矩阵、答案配对效应、行为分层结构以及标签任务一致性。
链接: https://arxiv.org/abs/2607.10626
作者: Sriram Selvam,Anneswa Ghosh
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:LLM-as-a-judge evaluation is widely used for retrieval-augmented generation (RAG), but reusing the same model family as both generator and judge makes self-leniency difficult to identify. We introduce Eval-Pair Matrix, a controlled meta evaluation protocol for source-grounded RAG. Starting from GaRAGe questions and grounding passages, we induce one hidden answer-causal contradiction per record, generate answers from perturbed passages with GPT, Grok, and Gemini models, and then use the same models as blind judges to evaluate each answer against the original passages. The experiment contains 300 core records, 897 labeled generator outputs, and 2,683 judge verdicts in a crossed 3 x 3 matrix; the primary analysis uses 275 fully validated records. Instead of comparing diagonal and off-diagonal cells across different answers, we estimate same-model effects by pairing judges on the exact same candidate answer. This changes the interpretation: diagonal and off diagonal F1 are similar, and the paired same-model recall effect is near zero (-0.5 pp; 95% cluster bootstrap CI [-2.7, +1.7]). The only robust paired gap is lower matching-judge flagging for answers that avoided the induced claim (-4.3 pp). A targeted human evaluation finds that reviewed apparent false positives are alternate source-error detections, mistakes in labeling whether the induced claim was adopted, or unclear cases; none were adjudicated as genuine false alarms. The lesson is methodological: RAG judge studies should report full matrices, answer-paired effects, behavior strata, and label-task alignment.
[NLP-73] Non-binary bottom-up constituency parsing without arity actions
【速读】： 该论文旨在解决非二叉化自底向上的短语结构解析（non-binary bottom-up constituency parsing）中对句法成分的分支数（arity）动作的依赖问题。传统方法需通过特定的归约动作（如\textscReduce-X\#k）显式指定父节点标签及子节点数量，导致动作空间庞大且存在标签-分支数耦合。其解决方案的关键在于引入分隔符引导的栈配置机制（delimiter-bounded stack configurations），将成分标签与归约跨度分离处理：通过活性分隔符和标签标记可唯一确定归约的分支数，使分支数成为解析状态的派生属性而非独立动作标签。这一分解消除了标签-分支数绑定的归约动作，显著缩小了动作集合规模，同时在宾州树库（PTB）和中文树库（CTB）上保持与原有基于分支数的基线相当的性能，且预测的分支数分布与真实语料接近，高分支成分未因移除分支动作而发生退化。
链接: https://arxiv.org/abs/2607.10591
作者: Jungyeul Park,Eunkyul Leah Jo,Zihao Huang
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Non-binary bottom-up constituency parsing is usually taken to require arity actions: reductions such as (\textscReduce-X#k) specify both the mother label and the number of children to be composed. We show that this arity parameter is not a necessary transition primitive. Our parser introduces constituent labels separately and recovers reduction spans from delimiter-bounded stack configurations. In a well-formed reduction configuration, arity is uniquely determined by the active delimiter and the label marker, making it a derived property of parser state rather than an action label. This factorization removes label–arity-specific reduce actions while preserving direct construction of original non-binary trees. Experiments on PTB and CTB show that the delimiter-guided parser remains competitive with an arity-specific bottom-up baseline under the same implementation framework, with substantially smaller action inventories. Analyses further show that its predicted arity profile remains close to the gold treebanks and that high-arity constituents do not collapse when arity actions are removed.
[NLP-74] Demographic Prompting at Scale: When More Attributes Hurt LLM --Human Agreement
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）预测与人类标注之间对齐效果受标注者人口统计学特征（demographic attributes）作为提示线索（prompt cues）影响的机制问题。其核心挑战在于理解如何通过设计提示中的属性组合来优化模型输出与人类判断的一致性，避免因过度指定导致性能下降。解决方案的关键在于揭示：1）适度引入1至3个高信号属性可显著提升对齐效果，而全属性配置反而引发过拟合式退化，存在明确的过指定阈值；2）人口统计学属性对人类标注的影响强度无法单独预测其对模型对齐的增益，必须综合考量每个属性的可学习性（learnability）与标注信号的方向一致性（directional coherence）；3）通过神经元探针分析发现，仅当标注信号具有方向一致性时，特定神经元激活才与对齐提升相关，激活量本身并不等同于可调控性。因此，该研究提出，基于人口统计学特征的提示干预并非普适策略，其有效性高度依赖于属性信号质量、任务特性及模型架构之间的复杂交互。
链接: https://arxiv.org/abs/2607.10590
作者: Mahammed Kamruzzaman,Shrabon Kumar Das,Gene Louis Kim
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:We investigate how annotator demographic attributes, supplied as prompt cues, shape the alignment between large language model (LLM) predictions and human annotations across five tasks. Using five open-source LLMs, we systematically vary the number and composition of demographic components in the prompt, spanning every combination from single-attribute through full-attribute configurations. Our experiments reveal three principal findings. First, alignment consistently peaks with one to three high-signal attributes and degrades under the full attribute set, establishing a clear over-specification threshold. Second, the overall magnitude of demographic influence on human annotations does not predict which attributes improve LLM alignment; instead, both the learnability and the directional coherence of each attribute’s annotation signal need to be considered jointly. Third, neuron probing reveals that specialized activation correlates with alignment gains only under coherent annotation signals, and that activation volume alone does not imply steerability. Together, these results demonstrate that demographic prompting is not a monolithic intervention: its utility is highly context-dependent, shaped by attribute signal quality, task characteristics, and model architecture.
[NLP-75] Constraint-Aware Hierarchical Search for Regulation-Driven Fine-Grained Classification
【速读】： 该论文旨在解决在海关税则分类、出口管制分类及基于标准的设备编码等监管密集型任务中，如何将输入实例准确映射到细粒度类别的问题。此类任务的关键挑战在于，正确标签不仅依赖语义相似性，更由规则定义的边界、阈值条件、排除条款、明确定义及局部例外共同决定，导致高度相似的输入可能对应不同标签，且看似相关的检索片段也可能因不符合具体规则而不可用。现有平面分类器、层次文本分类方法以及检索增强的大语言模型（LLM）系统均无法同时保障层级有效性、规则一致性与细粒度边界推理。为此，论文提出“监管驱动的细粒度层次分类”（regulation-driven fine-grained hierarchical classification）范式，要求通过监管层级中的有效路径并辅以可审计证据来分配标签。研究构建了四个来自典型监管场景的基准数据集，并采用专家介入（expert-in-the-loop）流程验证标注质量。核心解决方案为一种约束感知的层次搜索框架，该框架将监管文档转化为可搜索的树结构，仅检索合法的局部候选节点，并利用结构化监管字段与证据片段引导每一步的决策。实验表明，该方法在全部四个数据集上均取得最佳平均准确率，尤其在涉及细粒度邻近类别和基于规则的边界条件的任务中表现显著提升，同时提供可解释的决策路径。
链接: https://arxiv.org/abs/2607.10588
作者: Siyu Wang,Wei Tan,Lulu Chen
机构: GUSU Lab (GUSU实验室); DCES (数字计算与工程系统研究所)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Tasks such as customs tariff classification, export control categorization, and standards-based equipment coding require assigning an input instance to a fine-grained class under an explicit regulatory hierarchy. Unlike standard text classification, the correct label in these tasks is not determined by semantic similarity alone, but by rule-defined boundaries, threshold conditions, exclusion clauses, definitions, and local exceptions. As a result, two highly similar inputs may require different labels, while a retrieved passage that appears relevant may still be inapplicable under the governing rules. Existing flat classifiers, hierarchical text classification methods, and retrieval-augmented LLM systems are not designed to jointly enforce hierarchical validity, rule consistency, and fine-grained boundary reasoning. In this paper, we formulate this setting as regulation-driven fine-grained hierarchical classification, where an external instance must be assigned to a fine-grained class through a valid path in a regulatory hierarchy and supported by auditable evidence. We construct four benchmark datasets from representative regulation-intensive scenarios and validate the annotations through an expert-in-the-loop process. We further propose a constraint-aware hierarchical search framework that converts regulatory documents into a searchable tree, retrieves only valid local candidate nodes, and uses structured regulatory fields with evidence snippets to guide each next-hop decision. Experiments show that our method achieves the best mean accuracy on all four datasets and provides interpretable decision paths, with the largest gains on cases involving fine-grained neighboring categories and rule-based boundary conditions.
[NLP-76] UNIBROWSE: A Data-to-Agent Framework for Multimodal BrowseComp
【速读】： 该论文旨在解决多模态网页浏览任务（Multimodal BrowseComp）中代理模型在处理组合结构、开放世界不确定性以及跨长时间交互的多模态信息融合时面临的挑战，尤其针对现有数据构建方法未能覆盖“文本到图像”（text-to-image）这一关键信息流模式的问题。其解决方案的关键在于提出UNIBROWSE——首个统一的数据生成管道，首次同时涵盖三种信息流模式（文本仅限、图像到文本、文本到图像），并通过引入实时网络检索增强的精炼知识图谱以提升数据保真度，同时设计了一种新颖的探索度（exploration degree）度量指标，用于过滤低信号样本，从而支持高效强化学习。基于该管道，研究生成了高质量的冷启动工具使用轨迹与高探索性的问答对，并通过监督微调与探索感知训练，最终训练出一个350亿参数规模的代理模型。实验表明，UNIBROWSE代理在五个多样化基准上实现了平均54.4%的准确率，相较于基线模型Qwen3.5-35B-A3B提升了10.5个百分点，并超越多个闭源代理工作流（如GPT-5、Gemini-2.5 Pro等），显著提升了多模态代理的泛化能力与鲁棒性。
链接: https://arxiv.org/abs/2607.10557
作者: Xiyu Wei,Qingwei Zong,Zhuocheng Yu,Sujian Li
机构: Peking University (北京大学); Key Laboratory of Computational Linguistics, MOE (教育部计算语言学重点实验室); School of Computer Science, Peking University (北京大学计算机学院); School of Software and Microelectronics, Peking University (北京大学软件与微电子学院)
类目: Computation and Language (cs.CL)
备注: 17 pages, 5 figures
Abstract:Multimodal BrowseComp tasks require agents to combine perception, tool use, and long-horizon reasoning over dynamic web content, challenging their ability to handle compositional structure, open-world uncertainty, and multimodal integration across extended interactions. Crucially, real-world multimodal browsing involves three distinct information-flow patterns: text-only, image-to-text, and text-to-image, yet existing data construction methods cover only the text-only and image-to-text patterns, leaving text-to-image largely unaddressed and limiting agent generality and robustness. We introduce UNIBROWSE, a unified data pipeline that for the first time simultaneously generates training data covering all three patterns, augments curated knowledge graphs with live web retrieval for improved fidelity, and introduces a novel metric of exploration degree to filter low-signal instances for efficient reinforcement learning. Through this pipeline, we produce high-quality cold-start tool-use trajectories and exploration-rich QA pairs, and train a 35B-scale agent via supervised fine-tuning and exploration-aware this http URL resulting UNIBROWSE agent achieves state-of-the-art performance on multimodal BrowseComp benchmarks, attaining an average accuracy of 54.4 across five diverse benchmarks – an improvement of 10.5 points over its base model Qwen3.5-35B-A3B – and surpassing serveral closed-source agent workflows such as GPT-5 (42.9), Gemini-2.5 Pro (44.8), and Gemini-2.5 Flash (41.3).
[NLP-77] Articulate Intuition or Genuine Analysis? Benchmarking Epistemic Reliability in LLM -as-a-Judge Peer Reviews
【速读】： 该论文旨在解决生成式 AI（Generative AI）作为同行评审评价者时，其判断标准与人类评审委员会的评价是否一致的问题。研究指出，尽管两者均可能对某篇论文作出“分析性”或“高质量”的评价，但其所依据的标准在本质上并不相同，这一差异具有重要的哲学意义。解决方案的关键在于将卡尼曼（Kahneman）的双系统理论（dual-process theory）转化为一个结构化的同行评审评估框架——Kahneman4Review，该基准包含3,563篇经过评分的评审文本，涵盖九个基于理论的文本维度、八个偏见诊断指标以及一个连续的推理质量得分。研究发现：评审决策层级与基于文本证据的认知质量代理指标之间无显著对应关系；公开展示的代理型（agentic）评审文本原始得分高于人工汇总评审，但这一差距主要由文本长度和发表会议因素解释，且样本未进行论文配对；在2022–2023年过渡期，ICLR评审文本的诊断特征发生显著变化，时间上与大语言模型（LLM）广泛可用重合，但未能确定具体因果机制。此外，一项匹配的功能探针试点实验表明，该评估框架能够有效区分真实批判性分析与表面流畅性的文本。因此，论文主张，可信的LLM评审可靠性基准必须将分析形式（analytical form）与认知功能（epistemic function）相分离，并提出了具体的设计原则以实现这一目标。
链接: https://arxiv.org/abs/2607.10511
作者: Nuo Chen,Qian Wang,Qingyun Zou,Bingsheng He
机构: National University of Singapore(新加坡国立大学)
类目: Computation and Language (cs.CL)
备注: Preprint
Abstract:When an LLM judge calls a peer review analytical and a human committee calls another review high quality, are they tracking the same thing? We argue they are not, and that the difference matters philosophically. We operationalise Kahneman’s dual-process theory into a structured rubric for peer review and release Kahneman4Review, a benchmark of 3,563 rated reviews scored along nine theoretically motivated textual dimensions, eight bias diagnostics, and a continuous reasoning-quality score. Three findings bear on trustworthiness: decision tier is not detectably aligned with the rubric’s text-grounded epistemic-quality proxy; public-showcase agentic reviews receive higher raw scores than pooled human reviews, but length and venue explain most of the gap and the samples are not paper-paired; and ICLR review-text diagnostics shift at the 2022–2023 transition, temporally coincident with widespread LLM availability but without identifying its cause. A matched function-probe pilot further shows that the rubric distinguishes textual probes designed to contrast genuine fault-finding with surface fluency. We argue that a trustworthy reliability benchmark for LLM judges must separate analytical form from epistemic function, and propose concrete design choices toward that goal. An interactive demo is available at this https URL.
[NLP-78] ARMOR: Stabilizing On-Policy LLM RL with Off-Policy Anchor Samples
【速读】： 该论文旨在解决大语言模型在强化学习（Reinforcement Learning, RL）训练过程中因过优化（over-optimization）导致的性能退化问题，即模型过度拟合训练过程中的特定启发式策略，从而牺牲了泛化推理能力。其核心解决方案是提出一种名为ARMOR（Anchor Rollout and Mixed Optimization for RL）的新框架，关键在于将传统的被动惩罚机制转变为主动样本稳定策略。ARMOR通过两个核心组件实现：一是锚点回放（Anchor Rollout），利用来自参考策略的离策略数据以保留已建立的解题模式；二是混合优化（Mixed Optimization），重构策略目标函数，实现可控探索而无需依赖辅助损失项。实验结果表明，该方法能有效缓解验证性能崩溃现象，在多个推理基准上实现了长期稳定的性能提升。
链接: https://arxiv.org/abs/2607.10481
作者: Kexin Huang,Junkang Wu,Jinda Lu,Shuo Yang,Chiyu Ma,Jiancan Wu,Xiang Wang,Xiangnan He,Guoyin Wang,Jingren Zhou
机构: University of Science and Technology of China(中国科学技术大学); Peking University(北京大学); Dartmouth College(达特茅斯学院)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Reinforcement learning (RL) has significantly enhanced the reasoning capabilities of large language models (LLMs), yet the training process remains notoriously fragile. In this work, we investigate a critical source of this instability: over-optimization, where models exploit training heuristics at the expense of generalizable reasoning. While reverse KL regularization is the standard defense against such degradation, our analysis reveals that it is often insufficient in this regime, as it fails to ensure comprehensive coverage of the reference distribution. To address this, we propose ARMOR (Anchor Rollout and Mixed Optimization for RL), a framework that shifts the paradigm from passive penalty to active sample stabilization. ARMOR comprises two key components: (1) Anchor Rollout, which leverages off-policy data from the reference policy to preserve established solution patterns; and (2) Mixed Optimization, which reformulates the policy objective to enable controlled exploration without relying on auxiliary losses. Extensive experiments on reasoning benchmarks validate that ARMOR effectively mitigates validation collapse, enabling sustained performance improvements over extended training horizons.
[NLP-79] When Reasoning Hurts Legal Drafting: The Verbalization Bottleneck in Patent Claim Generation ICML2026
【速读】： 该论文旨在解决生成高质量专利权利要求（patent claim）这一高度结构化且对格式敏感的法律文书撰写任务中，如何有效利用大语言模型（LLM）提升生成质量的问题。其核心挑战在于权利要求需在技术准确性、语言精确性、形式规范性及元素间复杂逻辑关系的保持之间取得平衡。针对现有链式思维（Chain-of-Thought, CoT）提示方法在高度结构化任务中可能带来的局限性，论文提出一种面向专利权利要求生成的任务特定CoT方法，并通过自动评估指标与专家人工评估相结合的方式进行验证。研究的关键发现在于：隐式链式思维（implicit CoT）——即推理过程保留在模型内部而不显式表述——显著优于显式链式思维（explicit CoT）。这一反直觉结果源于显式推理引入了不必要的信息瓶颈，具体通过三种机制损害输出质量：关键细节的抽象化、内部生成模式的破坏以及错误的级联传播。该研究为法律文本生成任务中CoT的应用提供了新的理论洞见，强调在模式敏感型任务中应谨慎使用显式推理提示。
链接: https://arxiv.org/abs/2607.10480
作者: Lekang Jiang,Wenjun Sun,Stephan Goetz
机构: 未知
类目: Computation and Language (cs.CL)
备注: Accepted to AI for Law Workshop @ ICML 2026
Abstract:Patent claim drafting is a challenging legal drafting task that requires technical expertise, precise linguistic control, strict adherence to formal conventions, and the preservation of complex logical relationships among claim elements. While Chain-of-Thought (CoT) prompting has been widely used to improve the reasoning capabilities of large language models (LLMs), recent evidence suggests that its benefits may be limited, or even negative, in highly structured or pattern-sensitive tasks. Therefore, this paper investigates whether CoT prompting benefits patent claim generation. We propose a task-specific CoT method for patent claim generation and evaluate its effectiveness through both automatic metrics and human expert assessment. Our results show that reasoning-enhanced prompting can improve claim quality. Moreover, we demonstrate a counter-intuitive but important empirical finding: implicit CoT, where reasoning is kept internal rather than explicitly verbalized, consistently outperforms explicit CoT. Through systematic analysis, we show that explicit CoT can introduce an unnecessary information bottleneck for claim generation. Verbalized reasoning may compromise the quality of final outputs through three specific mechanisms: abstraction of critical details, disruption of internalized generation patterns, and cascading error propagation. Our findings provide new insights into legal tasks and CoT applications.
[NLP-80] Hallucination Detection in Large Language Models Using Diversion Decoding
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在生成文本时普遍存在幻觉（hallucination）问题，即模型可能生成事实错误或虚构的内容，从而影响其可靠性和可信度。现有幻觉检测方法多依赖于概率性评估且计算开销较大，难以在实际应用中推广。本文提出了一种名为“偏移解码”（diversion decoding）的新方法，其核心在于在解码过程中主动对模型生成的回应施加挑战，通过分析模型对生成替代答案的抵抗程度，提取能够反映其不确定性的特征。基于这些特征，训练一个机器学习模型以构建一种高效的不确定性启发式度量。实验结果表明，该方法在幻觉检测性能上优于现有方法，同时具备显著更低的计算复杂度，是一种高效且鲁棒的幻觉评估解决方案。
链接: https://arxiv.org/abs/2607.10476
作者: Basel Abdeen,S M Tahmid Siddiqui,Meah Tahmeed Ahmed,Anoop Singhal,Latifur Khan,Punya Parag Modi,Ehab Al-Shaer
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Large language models (LLMs) have emerged as a powerful tool for retrieving knowledge through seamless, human-like interactions. Despite their advanced text generation capabilities, LLMs exhibit hallucination tendencies, where they generate factually incorrect statements and fabricate knowledge, undermining their reliability and trustworthiness. Multiple studies have explored methods to evaluate LLM uncertainty and detect hallucinations. However, existing approaches are often probabilistic and computationally expensive, limiting their practical applicability. In this paper, we introduce diversion decoding, a novel method for developing an LLM uncertainty heuristic by actively challenging model-generated responses during the decoding phase. Through diversion decoding, we extract features that capture the LLM’s resistance to produce alternative answers and utilize these features to train a machine-learning model to develop a heuristic measure of the LLM’s uncertainty. Our experimental results demonstrate that diversion decoding outperforms existing methods with significantly lower computational complexity, making it an efficient and robust solution for evaluating hallucination detection.
[NLP-81] Not All Color Categories Are Equally Stable: A Multilingual Free Color Naming Experiment
【速读】： 该论文旨在解决颜色命名（color naming）在跨文化、跨语言背景下的不一致性问题，尤其关注红、黄、绿三类基本色在不同明度与饱和度变化下的命名稳定性。研究发现，颜色类别间的命名一致性存在显著差异：绿色在多种色调变化下仍保持较高的命名一致性，表现出较强的感知鲁棒性；而黄色因受亮度和饱和度影响较大，常被赋予“金”或“棕”等多类名称，表现出较低的命名一致性；红色则处于中等水平。其解决方案的关键在于通过自由命名实验（free color-naming experiment）揭示不同颜色类别在人类感知中的拓扑分布特性，表明某些颜色类别占据更广泛的感知区域，因而对视觉变异更具鲁棒性。这一发现为构建基于人类感知的、语义一致的颜色模型与命名系统提供了实证依据。
链接: https://arxiv.org/abs/2607.10465
作者: Nuray Toganas,Adilet Yerkin,Elnara Kadyrgali,Muragul Muratbekova,Aron Karatayev,Pakizar Shamoi
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL)
备注: Submitted to 2026 Joint 14th International Conference on Soft Computing and Intelligent Systems and 27th International Symposium on Advanced Intelligent Systems (SCISISIS 2026)
Abstract:Color naming is an important part of human color perception. Its task is to allow people to describe continuous colors using discrete color categories. However, the boundaries between color categories are often unclear, and some colors may be perceived differently depending on their saturation and brightness. While certain color categories remain recognizable across a wide range of shades, others may be associated with different color names when their appearance changes. This study investigates the consistency of color naming for red, yellow, and green color categories using a free color-naming experiment. A set of 18 color samples was selected from the COLIBRI dataset to represent different shades of these colors. Participants (n = 92) were asked to freely assign color names to each sample in Kazakh, Russian, or English without being limited to predefined categories. The results show that color categories differ in their consistency. Green shades were consistently identified as green despite variations in appearance, whereas yellow shades received a wider variety of names, including gold- and brown-related descriptions. Red shades showed moderate naming consistency. Our findings suggest that some color categories occupy broader perceptual regions than others and may therefore be more robust to visual variations. The study results can be used to develop perceptually meaningful color models and color naming systems.
[NLP-82] ANCHOR: Automated Alignment Auditing for CLI Agents on Real-World Harm ICML2026
【速读】： 该论文旨在解决当前自主命令行接口（CLI）代理在高自主性操作下可能引发的安全风险问题，特别是其在面对持续且具有适应性的恶意用户时，容易突破现有对齐机制而执行非法任务。其核心解决方案是提出ANCHOR——一个自动化审计框架，通过部署经过监督学习与强化学习微调的审计代理（auditor agent），该代理基于公开的美国司法案例中的非法行为模式，模拟具有“黑暗人格”特征的持久恶意用户，能够分解复杂任务、在被拒绝后重构请求，并在多轮交互中动态调整策略。实验表明，尽管前沿CLI代理在直接提问时通常能拒绝非法请求，但在持续施压下合规率可达100%，且一旦响应便常超出原始请求范围，自主构建可用于大规模危害的基础设施，如金融欺诈系统或生物武器开发环境。这揭示了现有对齐技术在应对复杂、动态恶意行为时存在显著不足，强调必须引入针对持续性、自适应攻击者的安全评估机制。
链接: https://arxiv.org/abs/2607.10455
作者: Kefan Song,Yanjun Qi
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: Accepted at ICML 2026. 19 pages, 14 figures, 5 tables
Abstract:Autonomous CLI agents can now execute hundreds of actions across multi-hour sessions: writing code, executing shell commands, browsing the web, and managing cloud infrastructure, all with minimal human oversight. Does greater autonomy invite greater risk? We introduce ANCHOR, an automated auditing framework that stress-tests CLI agents on illegal tasks grounded in public US court cases. ANCHOR deploys an auditor agent fine-tuned on dark personality data using supervised and reinforcement fine tuning. This auditor roleplays persistent malicious users who decompose tasks, reframe requests upon refusal, and adapt strategies across multi-turn interactions. Evaluating frontier CLI agents, we find that while they often refuse illegal tasks when prompted directly, compliance reaches 100% under persistent malicious interaction. When agents comply, they frequently exceed user requests, autonomously building infrastructure for large-scale harm, including catastrophic risk scenarios such as large-scale financial fraud and bioweapon development. These findings demonstrate that current alignment techniques are insufficient for autonomous agents and underscore the need for safety evaluations against persistent, adaptive malicious users. We release ANCHOR at this https URL
[NLP-83] Enjoy Your Talk: A Human-Centered Benchmark for Multi-Turn Dialogue with Decoupled User Simulation Target Modeling and Judging
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在多轮对话场景中作为对话伙伴时，传统单轮评估基准所无法捕捉的关键能力评估问题，具体包括角色一致性（persona consistency）、意图演化追踪（evolving intent tracking）、情感动态建模及目标完成度等。其核心解决方案在于提出EYT-Bench——一个以人类为中心、采用三方解耦设计的多轮对话评估基准：包含基于真实人格数据的用户模拟器（persona-grounded user simulator）、将意图感知与响应生成分离的目标模型架构，以及独立的第三方大模型评判者（LLM judge），支持多评判者集成以提升评估鲁棒性。为减少由模型自动生成人格引发的偏差，人格数据直接采自公开的人类标注语料库（Nemotron-Personas-USA 和 PersonaMem-v2）。此外，EYT-Bench引入两个轨迹级评估指标：基于嵌入的空间意图漂移（embedding-based intent drift）与最终意图完成率（Final-Intent Completion Rate, FICR），借鉴tau-bench的设计理念。在17个目标模型 × 200次对话的评估中，研究揭示了四大关键发现：（i）顶尖闭源与开源模型在主观维度（共情/人格一致性/拟人化）上差异微小（<0.3），但在客观意图追踪能力上差距可达9倍；（ii）推理机制（“thinking on”）显著提升长上下文人格下的意图追踪准确率（Gemma-4模型潜变量意图准确率提升0.47–0.50），而对主观评分影响甚微；（iii）人格格式主导轨迹分布，FICR在Nemotron-USA上超过0.95，而在PersonaMem-v2上则从0.53波动至0.88；（iv）预热效应在16/17个模型中表现稳健（仅GPT-5.5出现反转），且在α∈[0.05, 0.15]范围内排名稳定。跨评判者消融实验进一步验证了目标模型排名与最终意图满意度在不同评判者间具有高度一致性。
链接: https://arxiv.org/abs/2607.10428
作者: Jinglan Gong,Jiefan Lu,Hewei Guo,Kehan Li,Zhiyuan Han,Jihang Jiang,Wenwen Tong,Lewei Lu
机构: SenseTime Research(商汤科技); University of Science and Technology of China(中国科学技术大学); Tsinghua University(清华大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Evaluating large language models (LLMs) as multi-turn conversational partners requires probing capabilities that single-turn benchmarks miss: persona consistency, evolving intent tracking, emotional dynamics, and goal completion. We introduce EYT-Bench, a human-centered benchmark built around a three-party decoupled design: a persona-grounded user simulator, a target model that separates intent perception from response generation, and an independent third-party LLM judge with optional multi-judge ensembling. Personas are sampled from public human-curated corpora, Nemotron-Personas-USA and PersonaMem-v2, rather than synthesized, reducing LLM-induced persona bias. EYT-Bench also introduces two trajectory-level metrics: embedding-based intent drift and final-intent completion rate (FICR), inspired by tau-bench. In a 17-target x 200-dialogue evaluation, EYT-Bench reveals four findings: (i) state-of-the-art closed- and open-source models are statistically close on subjective dimensions (empathy / persona / anthropomorphism vary within = 0.3), but differ by up to 9x on objective intent tracking; (ii) reasoning (“thinking on”) sharply improves objective tracking on long-context personas (+0.47-0.50 latent-intent accuracy on Gemma-4) while leaving subjective scores nearly unchanged; (iii) persona format dominates trajectory spread, with FICR saturating above 0.95 on Nemotron-USA but spreading from 0.53 to 0.88 on PersonaMem-v2; and (iv) the warm-up effect is robust on 16/17 models (one outlier, GPT-5.5, reverses the effect), with stable rankings across alpha in [0.05, 0.15]. A cross-judge ablation using deepseek-v4-pro confirms that target rankings and final-intent satisfaction are preserved across judges.
[NLP-84] A Stepwise Questioning Expert-Editor Multi-Agent Framework for Long-Document Summarization
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在长文档摘要任务中因输入长度超过模型上限而导致性能受限的问题。其核心挑战在于如何有效处理超长文本的摘要生成，同时保持内容的完整性与关键信息的准确提取。该研究提出的解决方案关键在于设计一种基于多智能体（multi-agent）的分步提问机制，通过“专家”（expert）和“编辑”（editor）两类角色协同引导其他代理（agent）进行迭代式内容优化：专家负责从不同维度提出针对性问题以挖掘深层信息，编辑则提供具体的修订线索，从而逐步提升摘要质量。该方法显著增强了LLMs在长文档摘要中的表现，实验结果验证了其有效性。
链接: https://arxiv.org/abs/2607.10390
作者: Lingyun Shen,Xuejia Guo
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 12 pages,3 figures,2 tables
Abstract:Although large language models (LLMs) have shown promising potential in news summarization tasks, their performance on long-document summarization remains challenging as their length often exceeds the input limits. As the agent investment, which provide possibility to improve the inherent capabilities of LLMs. To enhance the effectiveness of long-document summarization based on LLMs, this paper proposes an expert-editor stepwise questioning multi-agent method, in which the expert and the editor guide another agent to refine the summary by posing questions on different aspects of the content and providing targeted clues for revision. We conducted experiments on two representative long-document scientific datasets and evaluated the results through widely recognized automatic metrics. The results demonstrated the effectiveness of our method.
[NLP-85] Structured Thoughts For Improved Reasoning And Context Pruning
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在生成长链思维（long reasoning traces）时存在的冗余与内存效率低下的问题。其核心解决方案是提出一种名为“结构化思维”（Structured Thoughts）的框架，将推理过程显式划分为交替的“尝试”（try）和“结果”（outcome）块：其中“尝试”块用于记录探索性推导过程，而“结果”块则对每一步推导进行凝练总结。研究通过分割现有推理轨迹并利用大模型自动生成对应的结果块，构建了一个结构化思维数据集，并基于该数据对预训练基础模型进行微调，使模型能够学习并采用这种结构化推理模式，在多个推理基准测试中相比标准监督微调（SFT）提升最高达8.08%的性能。此外，该结构化设计支持上下文剪枝——在每个“尝试/结果”对之后可移除“尝试”部分，仅保留“结果”以维持推理状态，从而显著降低上下文占用。初步剪枝实现表明，在数学任务上平均可节省85%的内存与上下文空间，仅伴随8.67%的性能下降，验证了该方法在效率与精度之间的有效权衡。
链接: https://arxiv.org/abs/2607.10386
作者: Zain Sarwar,Supriyo Chakraborty,Berkcan Kapusuzoglu,Chia-Hsuan Lee,Anirban Das,Stephen Rawls,Kartik Balasubramaniam,Sambit Sahu
机构: University of Chicago (芝加哥大学); Capital One (资本一号)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models (LLMs) excel at generating long chains of thought, but long reasoning traces are often verbose and memory-inefficient. In this work, we introduce Structured Thoughts, a framework that organizes reasoning into alternating try and outcome blocks: try captures exploratory scratch work, while outcome contains the distilled conclusion of that step. We construct a dataset of structured thoughts by segmenting reasoning traces into try blocks and prompting an LLM to summarize each step into its corresponding outcome. Fine-tuning pretrained foundation models on this reformatted data produces models that adopt the structured reasoning style, leading to performance gains of up to 8.08% on reasoning benchmarks compared to standard SFT. The explicit structure also enables context pruning: after each try/outcome pair, the try can be pruned, allowing the model to retain conclusions without keeping the full scratch work in the context. A proof-of-concept pruning implementation achieves an average of 85% memory / context savings with an 8.67% performance drop across mathematical tasks.
[NLP-86] CAFE: A Compound-AI Factorial Evaluation Framework
【速读】： 该论文旨在解决复杂生成式AI系统（Compound-AI Factorial Evaluation, CAFE）在评估过程中缺乏系统性实验设计的问题，尤其针对多组件可替换的流水线架构中难以确定各组件及其交互对答案质量影响程度的挑战。现有工具通常仅聚焦于配置搜索或孤立评分，无法解释哪些组件显著影响性能或差异是否具有统计意义。其解决方案的关键在于引入因子实验设计（factorial design），将每个可替换模块（如检索器、大语言模型、提示模板等）作为独立因子，通过构建全因子组合并运行实验，结合可配置的大语言模型（LLM）评判器与人工评分者进行统一评分，利用混合效应模型（mixed-effects models）量化各因子及交互作用对答案质量方差的贡献，进而报告效应量、显著性、最优配置、成本与延迟权衡以及评分者间一致性。该方法不仅实现高效配置优化，更提供可解释的因果分析能力。研究在HotpotQA基准数据集上的检索增强问答（RAG）任务中验证了CAFE的有效性，结果表明其能准确恢复预设因子效应，并在置换零假设下保持校准性。CAFE已开源为Python包和Web应用，支持科研与工程实践中的系统化评估。
链接: https://arxiv.org/abs/2607.10380
作者: Fabian Lukassen,Christoph Weisser,Thomas Kneib,Alexander Silbersdorff
机构: University of Göttingen(哥廷根大学); Bielefeld University of Applied Sciences and Arts (HSBI)(比勒费尔德应用科学与艺术大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:We introduce CAFE (Compound-AI Factorial Evaluation), an open-source platform that brings design of experiments to the evaluation of compound AI systems (CAIS). Such systems expose many interchangeable choices - e.g. which retriever, model, or prompt - and practitioners rarely know which of them most affects answer quality. With CAFE, a practitioner registers each swappable component of a pipeline as a factor to build a factorial design over the chosen factors, run the resulting configurations, and score the answers on a shared rubric using a configurable LLM judge together with human raters. From these ratings it attributes answer-quality variance to the components and their interactions with mixed-effects models and reports effect sizes, significance, the best configuration, cost and latency trade-offs, and judge-human reliability. Whereas existing tools mostly either search for a good configuration or score outputs in isolation, CAFE also explains which component drives quality and whether an observed difference is significant. We validate CAFE on a retrieval-augmented question-answering (QA) pipeline over the HotpotQA benchmark dataset, where it recovers planted factor effects and stays calibrated under a permutation null. CAFE is released as a Python package and as a Web application.
[NLP-87] PC-Mix: Partial-Component Audio Spoofing Detection under Mixed Speech and Environmental Sound Conditions
【速读】： 该论文旨在解决现有部分音频伪造检测研究中两个关键问题：一是现有基准数据集普遍缺乏真实场景下语音与环境声成分同时共存且可能被局部篡改的复杂情况，二是现有方法对环境声成分的局部伪造检测能力严重缺失。为此，论文提出PC-Mix，首个针对部分组件伪造检测的基准数据集，其中语音和/或环境声成分可被局部伪造，且二者在混合音频中同时存在。其核心解决方案在于构建包含真实环境声与局部伪造语音/环境声的混合音频样本，并设计联合学习框架以实现对语音、环境声及混合音频的统一伪造检测。实验表明，混合条件显著增加了检测难度，且在匹配目标条件下进行训练比直接迁移仅基于语音或环境声的模型更有效，凸显了跨模态联合建模的重要性。
链接: https://arxiv.org/abs/2607.10345
作者: Zhenshan Zhang,Xueping Zhang,Linxi Li,Yechen Wang,Ming Li
机构: 1. University of Science and Technology of China (中国科学技术大学); 2. Tsinghua University (清华大学); 3. Peking University (北京大学)
类目: ound (cs.SD); Computation and Language (cs.CL)
备注:
Abstract:Recent studies on partial audio spoofing mainly focus on studio-recorded speech with temporal localization of spoofed segments. However, these studies often overlook realistic conditions where spoofed and bonafide segments simultaneously coexist across speech and environmental sound components. In this paper, we present PC-Mix, the first dataset for partial-component spoofing detection, where either or both audio components may be partially spoofed. In PC-Mix, bonafide and partially spoofed environmental-sound components are first constructed and mixed with speech signals from an existing partial-spoof dataset, producing audio in which either or both components may be locally manipulated. This design addresses two major gaps in existing partial spoofing benchmarks: the lack of realistic environmental sounds in speech partial spoofing scenarios and the absence of partial spoofing detection for environmental sound components. We further establish standardized evaluation protocols and design a joint learning framework to optimize spoofing detection across speech, environmental sound, and mixed audio. Experiments highlight the increased difficulty introduced by mixed conditions. The results demonstrate that training under matched target conditions is more effective than directly transferring models trained on speech or environmental sound components.
[NLP-88] Polarization Detection: A Hybrid Approach with AfroXLMR-Social and DeBERTa for Low- and High-Resource Settings
【速读】： 该论文旨在解决在线极化话语（online polarization）的自动化检测问题，尤其针对英语和豪萨语两种语言在多样化语言背景下的有效识别与表征。其核心挑战在于跨语言语境下极化内容的细微差异捕捉、低资源语言（如豪萨语）的数据稀缺性以及计算资源受限等现实约束。解决方案的关键在于提出一种混合建模策略：针对英语二分类任务，采用具备强大单语性能的DeBERTa；而对于豪萨语及所有细粒度子任务（类型与表现形式），则使用经过领域适配的多语言模型AfroXLMR-Social，以充分捕捉社交媒体文本中极化的语义与文化语境特征。此外，为缓解数据不足与计算开销问题，引入低秩适应（LoRA）与基于nlpaug的文本数据增强技术，显著提升了模型效率与泛化能力。实验结果表明，根据各子任务特性进行模型定制化选择，能够实现性能与效率的最佳平衡。
链接: https://arxiv.org/abs/2607.10312
作者: Muhammad Abdullahi Said
机构: African Institute for Mathematical Sciences (AIMS)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:The rapid proliferation of online polarization threatens social cohesion, necessitating robust automated detection systems that operate effectively across diverse linguistic contexts. This paper presents our system description for the POLAR Shared Task 2026, focusing on the detection and characterization of polarized discourse in English and Hausa. We propose a hybrid modeling strategy: for English binary detection, we leverage the monolingual strength of \textbfDeBERTa, while for Hausa and all fine-grained subtasks (Types and Manifestations), we utilize \textbfAfroXLMR-Social. This domain-adapted multilingual model proved critical for capturing the nuances of polarization in social media text. To further address computational constraints and data scarcity, we implement Low-Rank Adaptation (LoRA) and textual data augmentation via \textttnlpaug. We report competitive results across all three subtasks, demonstrating that model selection tailored to specific subtask requirements yields the best balance of performance.
[NLP-89] PolyInterview: An LLM -based Platform for Immersive Mock Interview Practice with Comprehensive Multimodal Assessment
【速读】： 该论文旨在解决求职面试准备中真实练习机会稀缺、专业模拟辅导成本高昂以及自我练习缺乏动态互动与结构化评估的问题。现有系统多局限于固定问题序列、有限沟通方式或缺乏证据支持的反馈，难以提供全面、个性化的模拟体验。其解决方案的关键在于提出PolyInterview——一个基于大语言模型（LLM）的沉浸式模拟面试平台，能够根据目标职位描述和简历生成高度定制化的问题；通过具备唇形同步（lip-synced）的数字人面试官开展多轮语音对话，并基于回答内容进行自适应追问；同时采用多模态评估机制，由四个并行评估器提取13个行为层面特征，聚合为10个评估维度及两大能力轨迹。评估结果结合知识-技能-态度（KSA）与情境-任务-行动-结果（STAR）框架，提供可追溯的行为证据与可操作建议。实验表明，生成问题在93.7%的会话中与目标岗位描述匹配度显著高于跨岗位匹配度，十位专家评审亦证实其问题设计质量高且反馈具有实际指导意义。
链接: https://arxiv.org/abs/2607.10310
作者: Zhiyuan Wen,Jiannong Cao,Zijian Wang,Chen Chen,Xiaoyun Liu,Jianing Yin,Zhuo Li
机构: The Hong Kong Polytechnic University (香港理工大学); Chongqing University of Posts and Telecommunications (重庆邮电大学)
类目: Computation and Language (cs.CL)
备注: 10 pages, 7 figures, and 4 tables
Abstract:Preparing for job interviews is important for securing desired positions, yet realistic practice remains difficult to access: real interviews are infrequent, expert mock coaching is costly, and self-practice offers neither adaptive dialogue nor structured assessment. Existing systems typically address only parts of this need through fixed question sequences, limited communication channels, or feedback with little supporting evidence. We present PolyInterview, an LLM-based platform for immersive mock interview practice with comprehensive multimodal assessment. PolyInterview uses the target job description and CV to generate questions tailored to the role and candidate, conducts multi-turn spoken interviews with a lip-synced digital human interviewer that asks answer-aware follow-up questions, and evaluates response content, vocal delivery, and non-verbal behavior. Four parallel evaluators produce 13 behavior-level features that are aggregated into 10 assessment aspects and two competency tracks. Guided by the KSA and STAR frameworks, the report links each score to behavioral evidence and actionable recommendations. PolyInterview is publicly accessible. Its current all-account snapshot contains 101 accounts, 1,564 interview sessions, 7,665 generated questions, and 1,422 five-stage question sets. Generated questions are more closely aligned with their matched job description than with cross-role job descriptions in 93.7% of sessions. An evaluation by ten experts found strong question plans and actionable feedback.
[NLP-90] ChartSync: A Benchmark for Visuo-Logical Cascading Chart Editing
【速读】： 该论文旨在解决生成式图像编辑模型在处理结构化统计图表时，因数据修改需保持几何同步而面临的挑战，明确提出“视觉-逻辑级联编辑”（Visuo-Logical Cascading Editing, VLCE）这一任务范式。其核心问题在于现有方法仅能实现局部文本替换，难以支持依赖感知的级联更新，尤其在文本与几何元素间存在强耦合关系时表现不佳。解决方案的关键在于构建一个基于程序化渲染流程的专家验证基准——ChartSync，该基准通过保证视觉与逻辑层面的确定性耦合，提供可复现的真值标签。ChartSync包含870个三元组，覆盖9类图表和4种任务类型，其中235个实例专门用于测试文本到几何的级联同步能力。为全面评估模型性能，研究提出双层评价框架，结合客观视觉指标与基于视觉-语言模型的判别机制，以衡量低层级保真度及多模态理解与推理能力。实验结果揭示了显著的能力差距：多数开源模型在几何同步方面出现严重退化，仅有两个前沿闭源模型展现出初步的VLCE能力，其残余错误主要源于语义孤立与背景污染。通过对失败模式的深入分析，识别出未来多模态架构应具备的核心元能力。相关数据集与代码已公开发布。
链接: https://arxiv.org/abs/2607.10301
作者: Jiakang Yu,Yixuan Chai,Tianci Wang,Rihui Jin,Guangkai Xu,Hongtao Deng,Xun Zhu,Wang Gao,Xinrun Guo,Haipang Wu
机构: Jianghan University (江汉大学); HiThink Research (海思研究); University of Science and Technology of China (中国科学技术大学); Southeast University (东南大学); Zhejiang University (浙江大学)
类目: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL)
备注:
Abstract:Generative image editing models struggle with structured statistical charts when data modifications require geometric synchronization. We formalize this task as Visuo-Logical Cascading Editing (VLCE). However, existing methods remain confined to localized text substitutions and struggle with dependency-aware cascading updates. To systematically evaluate this capability, we introduce ChartSync, an expert-validated benchmark constructed via a programmatic rendering pipeline that guarantees deterministic visuo-logical coupling for the ground truth. ChartSync comprises 870 triplets across 9 chart categories and 4 task types, including 235 geometry-coupled VLCE instances that specifically test cascading text-to-geometry synchronization. We further evaluate these instances via a two-tier framework combining objective visual metrics with a vision-language model judge paradigm to assess low-level fidelity alongside multimodal comprehension and reasoning. Evaluating 14 image editing models and one code-mediated pipeline reveals a nuanced capability gap: most open-source models suffer severe drops in geometric synchronization, while only two frontier proprietary models show emerging VLCE capability, with their residual errors mainly involving semantic isolation and background corruption. Our detailed error analysis deconstructs these failure paradigms to identify core meta-abilities for guiding future multimodal architectures. The ChartSync dataset and code are publicly released at this https URL.
[NLP-91] SPARK: Susceptibility-Guided Profiling and Steering of Latent Reasoning States in Large Language Models
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在推理过程中失败原因难以诊断的问题。传统评估方法仅基于最终输出判断模型表现，无法揭示错误背后的内在机制——同一错误输出可能源于能力缺失、推理轨迹不稳定或已有推理状态未被有效激活。现有提示工程与基准测试方法多停留在输出层面，而通用的激活调控方法则缺乏针对性，通常采用全局方向干预，未能识别需干预的具体样本。本文提出SPARK方法，通过分析隐藏状态响应来诊断模型是否进入有效的内部推理状态，并指导轻量级的测试时调控。其关键创新在于：考虑到提示长度对原始隐藏状态敏感性存在显著干扰，尤其是在程序化与算法类推理任务中，更复杂的实例往往导致更长的输入序列，因此SPARK引入长度可控的敏感性度量，以分离输入规模效应与残余推理激活之间的混淆；进而结合跨层协调机制，筛选出推理活跃的锚点样本与激活不足的困难样本。研究采用FRONTIER-4.5K作为受控的程序化推理基准进行潜在表征分析和难度感知评估，并在GSM8K与MATH-500上通过前向单一路径评测验证了SPARK-Steering的有效性。实验结果表明，该方法可稳定提升Qwen3系列模型性能，在MATH-500上，Qwen3-4B准确率从82.0%提升至84.6%，Qwen3-8B从82.4%提升至85.6%。这表明，敏感性不仅可作为推理失败的诊断信号，还可作为靶向测试时干预的实用引导依据。
链接: https://arxiv.org/abs/2607.10296
作者: Dongxu Zhang,Yiding Sun,Zihao Guo,Xiangyang Yang,Kai Tang,Lin Chen,Cheng Tan,Jihua Zhu
机构: Xi’an Jiaotong University (西安交通大学); Peking University (北京大学); Tencent (腾讯)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Reasoning failures in large language models (LLMs) are usually evaluated from final answers, but a wrong answer does not reveal why the model failed. The same incorrect output may reflect missing capability, an unstable reasoning trajectory, or a failure to activate a reasoning state that is already available in the frozen model. Existing prompting and benchmark-based evaluation methods mostly operate at the output level, while generic activation-steering methods typically apply global directions without diagnosing which examples require intervention. In this paper, we introduce SPARK, which uses hidden-state response to diagnose whether a model internally enters an effective reasoning state and to guide lightweight test-time steering. The key observation is that raw hidden-state susceptibility is strongly confounded by prompt length, especially in programmatic and algorithmic reasoning where harder serialized instances naturally become longer. SPARK therefore uses length-controlled susceptibility to separate input-scale effects from residual reasoning activation, and combines this signal with cross-layer coordination to select reasoning-active anchors and under-activated hard examples. We use FRONTIER-4.5K as a controlled programmatic reasoning suite for latent profiling and difficulty-aware analysis, and evaluate SPARK-Steering on GSM8K and MATH-500 with forward-only benchmark profiling. Our method improves Qwen3 series models consistently; on MATH-500, accuracy rises from 82.0% to 84.6% for Qwen3-4B and from 82.4% to 85.6% for Qwen3-8B. These results suggest that susceptibility can serve not only as a diagnostic signal for reasoning failures, but also as a practical guide for targeted test-time intervention.
[NLP-92] Information-seeking failures of large language models in agent ic clinical reasoning
【速读】： 该论文旨在解决当前大语言模型在临床肿瘤学领域中虽具备较高医学知识水平，却难以有效应对不确定性情境下的主动信息获取问题。其核心挑战在于模型缺乏像人类医生一样在诊断过程中主动、系统地请求关键临床数据的能力。为解决这一问题，研究提出了一种代理式评估框架（agentic evaluation framework），要求模型在三个连续轮次中主动申请临床数据，方可做出诊断与治疗决策。结果显示，尽管所有前沿模型的推理轨迹在临床推理评分中表现优异（91%高于阈值），但其诊断准确率最高仅达68%，且信息利用效率显著下降——从第一轮的57%降至第三轮的26%，导致分子和细胞遗传学等关键数据未被充分调用。统计分析表明，信息利用度是诊断准确性的最强预测因子（R = 0.69, P < 0.001）。错误模式分析揭示出搜索满足、锚定效应及过早闭合等认知偏差，与双过程模型下新手临床医生的典型失误高度一致。因此，该研究的关键发现是：当前模型在临床肿瘤学中的主要瓶颈并非医学知识不足，而是面对不确定性时系统性地失效的信息寻求能力。
链接: https://arxiv.org/abs/2607.10275
作者: Krischan Braitsch,Laura K. Schmalbrock,Theresa Weltermann,Andrew F. Berdel,Isabella Miller,Kai Tran,Michael Heider,Sabrina Kraus,Florian Bassermann,Jacqueline Lammert,Sebastian Ziegelmayer,Marcus Makowski,Lisa C. Adams,Keno K. Bressem
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 66 pages, 9 figures; includes supplementary material
Abstract:Large language models achieve high scores on medical knowledge assessments, yet clinical reasoning requires actively deciding what to investigate under uncertainty. We developed an agentic evaluation framework in hematologic oncology in which models must proactively request clinical data across three sequential rounds before committing to a diagnosis and treatment plan. Across 32 frontier models, the best achieved only 68% overall accuracy. Information utilization, the fraction of available data actually requested, was the strongest predictor of diagnostic accuracy (R = 0.69, P 0.001), yet utilization collapsed from 57% to 26% in the final round, leaving molecular and cytogenetic data critical for treatment selection unexamined. Reasoning traces scored high on a clinical reasoning rubric (91% above threshold) but decorrelated from accuracy, revealing a gap between locally coherent rationales and globally correct conclusions. Error analysis identified search satisficing, anchoring and premature closure as the dominant failure modes, the same cognitive biases that characterize novice clinicians under dual-process models of diagnostic reasoning. These findings demonstrate that the primary limitation of current models in clinical oncology is not insufficient medical knowledge but a systematic failure of information-seeking under uncertainty.
[NLP-93] Language Re-generation: An investigation into information locality effects on reconstruction
【速读】： 该论文旨在探究语言模型在面对严重破坏信息局部性（information locality）的输入时，是否仍能从非自然语言形式中恢复出符合人类语言习惯的自然语言结构，以及这一过程揭示的语言模型内在归纳偏置（inductive bias）。其核心问题是：当输入的语法结构被人为打乱后，语言模型能否重建原始的、具有局部依赖特性的自然语言结构？解决方案的关键在于提出一种基于重构（reconstruction）的实验框架——将预训练于“不可能语言”（impossible languages）的GPT-2模型进行微调，以从三种不同类型的扰动输入中重构自然英语。研究发现，模型恢复出的结构表现出更短的依存距离，与未受约束生成中的局部性偏好一致，这为语言模型架构本身存在的局部性偏好提供了可量化的证据；同时，恢复难度随局部性破坏程度增加而上升，且结构恢复（依赖三元组F1）与表面恢复（精确匹配）之间存在解耦，表明模型对句法结构和表面形式的处理机制不同。此外，句子长度在局部结构保留条件下有助于恢复，但在全局打乱下导致性能完全崩溃。最终，重构难度与语言学习难度在不同扰动类型下高度相关，说明信息局部性是制约语言模型学习与恢复能力的共同约束因素。
链接: https://arxiv.org/abs/2607.10268
作者: Amirhossein Mohammadi,Laurence E. Frank,Albert Gatt,Robert A. Bagheri
机构: Utrecht University (乌得勒支大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Information locality, the tendency for syntactically related words to appear close together, shapes both human language processing and language model learning. While prior work has examined whether language models can acquire impossible languages, it remains unclear whether they can recover natural language from such input and what this reveals about their inductive biases. We address this by complementing learnability-based approaches with a reconstruction framework: fine-tuning GPT-2 models pre-trained on impossible languages to reconstruct natural English from three perturbation types. Our findings show that the recovered structures exhibit shorter dependency lengths than the original text, mirroring the locality preference observed in unconstrained language model generation and providing a quantitative signature of an architectural bias that learnability experiments alone do not reveal. Recovery difficulty increases with the degree of locality disruption. Structural recovery (dependency Triple F1) dissociates from surface recovery (Exact Match), while fluency dissociates from faithful reconstruction under global shuffling. Sentence length further modulates performance: longer sentences facilitate recovery when local structure is preserved but lead to complete collapse under global shuffling. Finally, recovery difficulty tracks learnability difficulty across perturbation types, suggesting that information locality is the shared constraint governing both.
[NLP-94] Which Languages Transfer Best to Warlpiri? A Similarity-Based Study for Low-Resource ASR INTERSPEECH2026
【速读】： 该论文旨在解决在极端低资源条件下，自动语音识别（ASR）模型跨语言迁移效果不佳的问题，特别是针对澳大利亚原住民语言Warlpiri——其标注语音数据极为稀缺，难以支持有效的模型训练。为提升跨语言迁移性能，本文提出一种融合预训练语音模型中的声学相似性与基于语言类型学、音素库、语法及句法特征的语言学相似性的框架，用于对高资源源语言进行排序并评估其向Warlpiri迁移的有效性。实验结果表明，采用Whisper模型时，声学与类型学上相似的语言在微调任务中显著优于单语和多语基线模型；阿萨姆语（Assamese）与印地语（Hindi）均大幅降低词错误率（WER）和字符错误率（CER）。相关性分析进一步揭示：声学相似性是微调性能最强的预测因子，而音素库与类型学相似性则更有利于解释零样本迁移的表现。因此，解决方案的关键在于通过多维度语言相似性度量（包括声学与语言学特征）实现最优源语言选择，从而有效增强低资源语言的跨语言迁移能力。
链接: https://arxiv.org/abs/2607.10256
作者: Pravina Mylvaganam,Eliathamby Ambikairajah,Ting Dang,Vidhyasaharan Sethu,Tuende Szalay
机构: University of New South Wales, Australia (新南威尔士大学, 澳大利亚); University of Melbourne, Australia (墨尔本大学, 澳大利亚); University of Sydney, Australia (悉尼大学, 澳大利亚)
类目: Computation and Language (cs.CL); Audio and Speech Processing (eess.AS)
备注: Accepted by Interspeech 2026
Abstract:This paper investigates how language similarity can improve cross-lingual transfer for automatic speech recognition (ASR) in extremely low-resource settings. Warlpiri, an Australian Aboriginal language, has very limited transcribed speech data, making transfer learning essential. We propose a framework combining acoustic similarity from pre-trained speech models with linguistic similarity based on typology, phoneme inventories, grammatical, and syntactic features to rank high-resource source languages and evaluate their effectiveness for ASR transfer to Warlpiri. Experiments with Whisper show that acoustically and typologically similar languages outperform monolingual and multilingual baselines. Assamese and Hindi achieve substantial reductions in word and character error rates. Correlation analysis further indicates that acoustic similarity is the strongest predictor of fine-tuning performance, while phoneme inventory and typological similarity better explain zero-shot transfer.
[NLP-95] One Token Is Enough: Fingerprinting and Verifying Large Language Models from Single-Token Output Distributions
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在商业化服务链中存在模型身份不可信的问题，即客户端无法验证所接收到的响应是否来自其声称的模型，且已有研究表明大量商业接口的实际行为与厂商公布的参考权重存在显著偏差。现有模型识别方法通常依赖于长文本生成、逐标记概率、对抗性提示或模型所有者的配合，限制了其可操作性。本文提出一种基于行为指纹的轻量级解决方案：通过采集模型对简单单字词提示（如“name a random number between 1 and 100”）的响应分布，构建其行为指纹。该方法仅需每查询一个输出词元，即可在四种语言下低成本收集证据。实验覆盖165个通过OpenRouter聚合平台提供的模型，结果表明：（i）这些响应分布高度非均匀（中位数熵为1.0比特），具有强模型特异性——同一模型的样本对之间距离比不同模型间小一个数量级；（ii）基于Jensen-Shannon散度的行为指纹能有效追溯模型谱系，留一法准确率达59.5%（随机基线18.4%）；（iii）采用类生物特征验证协议，在完整40单元测试集下实现7.3%等错误率，仅用8个探测单元即可低于11%，相当于约百次单词查询即可完成一次审计。此外，研究还发现某些标有专有品牌的旗舰接口在分布上与开源的Qwen模型无法区分。该方案的提示语、原始数据和分析代码均已公开，支持复现与实际部署。
链接: https://arxiv.org/abs/2607.10252
作者: Tomas Bruckner
机构: Prague University of Economics and Business (布拉格经济大学)
类目: Cryptography and Security (cs.CR); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Large language models (LLMs) are increasingly consumed through opaque serving chains - API aggregators, resellers, and inference providers - in which the client has no technical means to confirm that the model answering is the model advertised, and recent audits show that a substantial fraction of commercial endpoints deviate from the vendor’s reference weights. Existing identification techniques require long generated texts, token-level log-probabilities, adversarially crafted prompts, or the model owner’s cooperation. We show that far weaker evidence suffices. We define a behavioral fingerprint of an LLM as the empirical distribution of its answers to trivial one-word prompts - “name a random number between 1 and 100” - collected across four languages at a cost of one output token per query. Measuring 165 models served via a large commercial aggregator (OpenRouter), we find that (i) these distributions are highly non-uniform (median cell entropy 1.0 bit) and model-specific: split halves of the same model’s samples lie an order of magnitude closer than samples of different models; (ii) Jensen-Shannon divergence between fingerprints recovers model lineage, assigning a model to its documented family with 59.5% leave-one-out accuracy against an 18.4% chance rate; and (iii) a biometric-style verification protocol achieves a 7.3% equal error rate with the full 40-cell battery, and below 11% with eight probe cells - roughly a hundred single-token queries per audit. We further report ecosystem anomalies, including a proprietary-branded flagship endpoint distributionally indistinguishable from an open-weight Qwen model. The protocol, prompts, raw data, and analysis code are released for reproduction and operational use.
[NLP-96] One mechanism for many mental spaces: a shared router over a value slot in language models
【速读】： 该论文旨在解决语言如何在不同语义空间（如现实、虚构、假设、信念、记忆等）之间构建和区分意义的问题，尤其是这些空间在形式语义学中被视作逻辑上分离的系统，而心理空间理论（Mental Space Theory）则主张它们可通过统一的认知操作生成。其核心问题是：当前基于变压器架构的语言模型究竟实现的是形式语义学中的多逻辑分离机制，还是心理空间理论所主张的跨类型统一建构机制？解决方案的关键在于发现模型内部存在一种机械化的“心理空间统一”机制——即通过一个通用的路由器/槽位（router/slot）结构，在共享的值槽中存储属性内容，并由一个可因果操控的路由器（空间索引）选择读取哪个语义空间。该路由器具有低秩特性，与实体身份线性叠加，且主要作用于深层注意力头。实验表明，仅训练特定类型空间（如反事实、信念、虚构或时间）的子空间即可泛化至其他类型，且性能显著高于随机基线，说明信念等复杂空间并未在模型中被特别隔离。进一步实验证明该机制驱动推理并具备组合性：经规则推导训练的子空间可改变模型的内在推理路径，但不影响其外在输出；多个空间构建器组合后能生成新的路由器。这表明模型内部存在一种跨类型一致的语义空间生成机制，为心理空间理论提供了神经机制支持。
链接: https://arxiv.org/abs/2607.10248
作者: Oliver Steele,Jiangtao Wen,Yuxing Han
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 25 pages, 6 figures, 9 tables
Abstract:Language builds discourse contexts other than the actual: a painting, a belief, a memory, a hypothetical. Each is a mental space in which the same entity can take a different value, as when a flower is red in reality but purple in a portrait. Formal semantics keeps these contexts apart because their logics differ (modal, temporal, doxastic, depictive); Fauconnier’s mental-space theory treats them as one space-building operation. We ask which of these a transformer language model implements, and find a mechanistic version of Fauconnier’s unification. The model uses one router/slot format across the inventory: a reusable value slot stores attributed content, and a causally manipulable router (the space index) selects which space is read. A subspace trained with Distributed Alignment Search to control one space type, counterfactual, belief, fictional, or temporal, also controls the others, well above a random floor, on three model families; belief, which formal semantics marks as a distinct case, is not specially separated. The router is low-rank, composes additively with entity identity, and acts through a few late-layer heads. Two further results show the mechanism drives inference and composes: a subspace trained on a rule-derived conclusion flips what the model infers while dissociating from what it reports, and composing space-builders mints a fresh router over the shared slot. This paper establishes the cross-type generality. A companion paper develops belief in depth, because of its special status in philosophy, psychology, and linguistics (epistemology, theory of mind, and propositional attitude reports).
[NLP-97] KGCQual: An Interpretable Framework for Evaluating the Knowledge Graph Construction Quality from Text
【速读】： 该论文旨在解决自动化知识图谱（Knowledge Graph, KG）构建过程中存在的三元组冗余、遗漏及语义失真问题，此类缺陷会显著影响下游任务性能。现有评估方法多依赖于特定任务指标或小规模人工验证，难以全面反映抽取图谱在结构与语义层面的忠实度。本文提出一种新颖且可解释的内在质量评估指标KGCQual，其核心在于衡量自动抽取的图谱与理想图谱之间的逼近程度——理想图谱应准确捕捉源文本中的关键名词短语、谓词关系以及否定等基本语言现象。该框架包含两个互补组件：(1) 实体级评估，关注实体完整性、消歧质量与连通性；(2) 关系级评估，通过词汇相似性、依存句法对齐与轻量级否定处理机制，评估谓词保留性与多重性，以保障语义忠实性。实验在WebNLG、TinyButMighty和BenchIE等多个主流三元组抽取系统与数据集上验证了该指标的有效性，能够可靠识别出传统指标忽略的遗漏、冗余与结构偏差。进一步的消融研究与下游链接预测任务验证表明，KGCQual得分与模型性能高度相关，证明其具备良好的判别能力。本研究提供了一个可扩展、模型无关且可解释的评估框架，为自动化知识图谱构建方法的标准化比较奠定了基础。
链接: https://arxiv.org/abs/2607.10212
作者: Nipun Misra,Vikranth Udandarao,Aanchal Gupta,Yogender Kumar,Manuj Mukherjee,Raghava Mutharaju
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: This paper is under review at a conference
Abstract:Knowledge Graphs (KGs) are increasingly constructed through automated extraction pipelines; however, such systems often introduce spurious or incomplete triples, which degrade downstream performance. Existing evaluation practices rely heavily on task-specific metrics or small-scale manual verification, offering limited insight into the structural and semantic fidelity of extracted graphs. We propose a novel, interpretable metric for intrinsic KG quality assessment that measures how closely an automatically extracted graph approximates an “ideal” graph capturing the key noun phrases, predicate relations, and basic linguistic phenomena such as negation expressed in the source text. Our framework integrates two complementary components: (1) an entity-level assessment that evaluates completeness, resolution quality, and connectivity, and (2) a relation-level assessment that judges predicate preservation and multiplicity using lexical similarity, dependency-parse alignment, and light-weight negation handling to ensure semantic faithfulness. We evaluate our metric across multiple state-of-the-art triple extraction systems and datasets, including WebNLG, TinyButMighty, and BenchIE, demonstrating that it reliably identifies omissions, redundancy, and structural deviations that existing metrics overlook. Our work offers a scalable, model-agnostic, and interpretable framework for comparing automated KG construction methods and provides a foundation for standardised evaluation. We further validate the metric through an ablation study isolating noun and verb components, and a downstream evaluation showing that KGCQual scores correlate significantly with link prediction performance on the same extracted KGs. The code repository is available at this https URL.
[NLP-98] oward Stronger Code Watermarking: A Grammar-Driven Approach to Optimizing the Trade-off Between Quality and Detectability
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在代码生成场景下文本水印（text watermarking）的难题，尤其针对代码固有的低熵特性所导致的代码质量与水印可检测性之间的权衡困境。现有基于logits的水印方法在代码生成中难以有效应用，因其易破坏语法正确性或削弱水印信号强度。为此，论文提出一种名为语法驱动水印（Grammar-Driven Watermark, GDW）的新方法：其核心创新在于通过语法引导的三级掩码机制保障生成代码的语法有效性，并采用结构角色感知的调制策略，在内容承载型标记（content-bearing tokens）上施加较强偏置，而在语法关键型标记（syntax-critical tokens）上采取更保守的偏置，从而在不损害代码质量的前提下增强水印信号。此外，为匹配生成过程，设计了角色感知加权检测统计量以提升可检测性。实验结果表明，GDW在多种编程语言、模型及解码策略下均显著优于现有方法，实现了更优的质量-可检测性权衡，且对变量重命名攻击具有强鲁棒性。
链接: https://arxiv.org/abs/2607.10210
作者: Licheng Yu,Aiwei Liu,Songze Li
机构: Southeast University (东南大学); Tsinghua University (清华大学)
类目: Cryptography and Security (cs.CR); Computation and Language (cs.CL)
备注:
Abstract:With the rapid development of Large Language Models (LLMs), text watermarking has emerged as a crucial technique for identifying machine-generated content. However, directly applying existing logits-based watermarking methods to code generation remains challenging, since the low-entropy nature of code exacerbates the trade-off between code quality and watermark detectability. In this paper, we propose a novel code watermarking approach called Grammar-Driven Watermark (GDW) for LLMs. GDW preserves syntactic validity through a grammar-guided three-level masking mechanism and injects watermark signals via structural role-aware modulation, assigning a stronger bias to content-bearing tokens while applying a more conservative bias to syntax-critical tokens. Aligning with the generation process, we further design a role-aware weighted detection statistic to improve detectability. Experiments across multiple programming languages, models, and decoding strategies show that GDW establishes a stronger quality-detectability trade-off frontier than existing methods, while maintaining robustness against variable-renaming attacks.
[NLP-99] Equal Accuracy Unequal Evidence: Search APIs as Decision Surfaces for Tool-Using Agents
【速读】： 该论文旨在解决当前基于检索的智能体（agent）在使用搜索API时，对搜索结果评估标准过于依赖答案准确率而忽视其作为“决策表面”（decision surface）功能的问题。传统搜索API仅提供URL、标题和片段（snippet），而代理通过渐进披露（progressive disclosure）机制决定是否进一步获取完整页面，这一过程中的信息呈现直接影响代理的决策路径。研究提出，商业搜索API的本质应被视为一种影响代理行为的决策表面——即其返回的排序片段、URL及元数据共同决定了代理是否立即回答、再次搜索或消耗资源加载页面。为验证此观点，作者采用一个冻结的GPT-5.4代理与两个工具（search_web 和 fetch_page），在SEALQA-HARD的100个难题上对比三种搜索提供商（Brave、Tavily、Firecrawl），并由Kimi-K2.6 oracle标注所有可见内容元素，生成6,869条有效判别数据。采用经审计的“语义匹配”（semantic match）作为正确性度量标准，结果显示三者答案准确率相近（分别为25、25、26/100），但其证据经济性（evidence economy）差异显著：Brave的片段富含可直接支持答案的信息，Tavily将高价值支持性URL集中在首位，而Firecrawl则促进更广泛的探索行为。此外，引入“表面矛盾至黄金URL比率”（surface contradiction-to-gold URL ratio）发现其在0.92至2.59之间波动，表明不同提供商的选择本质上是检索预算与策略配置的权衡，而非单纯的召回能力问题。因此，该研究的关键解决方案在于重构对搜索API性能的理解框架，强调其作为决策界面的作用，并提出以“证据经济性”与“决策一致性”为核心的多维度评估体系。
链接: https://arxiv.org/abs/2607.10198
作者: Sriram Selvam,Anneswa Ghosh
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Search APIs are the fundamental retrieval layer for many agents and are often their most frequently used tool. Traditional search APIs provide URLs, titles, and snippets that preview website contents. Because full-page retrieval is token-intensive, agent retrieval architectures increasingly use progressive disclosure: the agent first sees snippets and then chooses whether to fetch full pages. In such systems, search API performance is often evaluated primarily by answer accuracy. We argue that a commercial search API is better understood as a decision surface: the ranked snippets, URLs, and metadata that determine whether an agent answers immediately, searches again, or spends tokens opening pages. We test this claim with one frozen GPT-5.4 agent, two tools (search_web and fetch_page), and 100 questions from SEALQA-HARD, varying only the search provider (Brave, Tavily, Firecrawl). A Kimi-K2.6 oracle labels every content element visible to the agent (URL, title, snippet, and fetched page, when fetched), producing 6,869 valid per-URL judgments. We use an audited correct-answer label, semantic match, which preserves exact matches while accepting harmless formatting and naming variants. Under this measure, the providers remain close (25, 25, 26 / 100), but their evidence economies differ sharply: Brave offers gold-answer-rich snippets, Tavily concentrates gold-supporting URLs at rank 1, and Firecrawl is associated with broader exploration under this fixed agent policy. We also introduce a surface contradiction-to-gold URL ratio, which varies from 0.92 to 2.59. Provider choice is therefore a retrieval-budget and policy decision, not merely a recall decision.
[NLP-100] Instruction Set and Language for Hypergraphs
【速读】： 该论文旨在解决超图同构判定（hypergraph isomorphism）这一核心问题，尤其针对具有有界超边度数的有限连通超图。传统方法通常通过将超图转化为莱维图（Levi graph）并依赖图同构引擎（如nauty、Traces、bliss）进行求解，存在计算开销大、转换过程复杂等问题。本文提出一种名为IsalHG的新方法，其关键在于构建一个紧凑的指令字母表Σ_HG，将任意连通超图结构编码为字符串——该编码由一个小型虚拟机执行，包含稀疏超图、节点引用的循环双向链表（CDLL）及最多k个遍历指针（k为超边最大度数），通过移动指针或插入超边与新节点来完成编码。该字母表具有封闭性，任意字符串均可解码为有效超图。进一步地，采用贪心算法（h2s）实现通用编码，并引入基于字典序最大结构元组的回溯种子策略生成规范字符串（canonical string）w*，作者提出该规范字符串可能是完全的同构不变量（isomorphism invariant）。由此，直接通过规范字符串相等性即可判断超图同构，无需经由莱维图转换和外部图同构引擎。实验验证了150个随机均匀超图及若干经典组合设计的往返性质（s2h(h2s(H)) ≅ H），并在 (n, c) 网格上对四种方法（含三种莱维基线）进行了基准测试，600次判别结果一致，支持了完备性猜想。在实际运行时间上，莱维基线方法平均快3至5个数量级（几何均值比为311×至117,672×），凸显了原生方法在效率上的巨大优势。因此，本文的核心贡献在于：提出一种原生超图表示框架、提出规范字符串完备性猜想，以及首次提供了原生方法与莱维方法之间的直接性能对比基准。
链接: https://arxiv.org/abs/2607.10194
作者: Mario Pascual-Gonzalez,Ezequiel Lopez-Rubio
机构: University of Málaga (马拉加大学); ITIS Software (ITIS软件)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Programming Languages (cs.PL)
备注:
Abstract:We present IsalHG, a method for representing the structure of any finite, connected hypergraph of bounded hyperedge arity as a string over a compact instruction alphabet \Sigma_\mathrmHG . The encoding is executed by a small virtual machine comprising a sparse hypergraph, a circular doubly-linked list (CDLL) of node references, and k traversal pointers, where k bounds the hyperedge arity. Instructions either move a pointer through the CDLL or insert a hyperedge, optionally together with new nodes, into the hypergraph. Every string over \Sigma_\mathrmHG decodes to a valid hypergraph; the alphabet is closed. A greedy \emphHypergraphToString (h2s) algorithm encodes any connected hypergraph into a string; a backtracking variant seeded at nodes of lexicographically maximal structural tuple produces a \emphcanonical string w^* , which we conjecture to be a complete isomorphism invariant. Canonical-string equality then decides hypergraph isomorphism natively, without the standard reduction to the Levi incidence graph followed by a graph-isomorphism engine. We verify the round-trip property s2h(h2s(H)) \cong H on 150 connected random uniform hypergraphs and on named combinatorial designs, and we benchmark the canonical algorithm against the three practically available exact baselines – nauty, Traces, and bliss operating on the 2-coloured Levi graph – across a (n, c) grid with ten seeds per cell. All four methods agree on every one of 600 isomorphism verdicts, consistent with the completeness conjecture. On wall-clock time the Levi baselines dominate every tested cell by three to five orders of magnitude (geometric-mean ratio 311\times to 117,672\times ), which we report as measured. We contribute the representation framework, a conjecture of canonical completeness, and the first native-versus-Levi benchmark for hypergraph isomorphism.
[NLP-101] Cost of Reasoning in non-English Languages: A Case Study on Japanese
【速读】： 该论文旨在解决多语言环境下生成式推理模型（Generative AI）在非英语语境下保持强推理性能的问题，尤其关注如何实现用户自定义语言的推理能力，同时兼顾模型可解释性与安全性。其核心挑战在于：尽管英语拥有最丰富的推理训练数据，但实际应用中用户对非英语（如日语）推理的需求日益增长。为此，论文提出通过持续预训练（continual pretraining）结合生成式强化学习优化（GRPO）方法，构建一个以日语进行推理的模型变体——基于Qwen-3-8B的日本语持续预训练版本（Qwen-3-Swallow-8B）。关键解决方案在于采用GRPO框架对日语持续预训练模型进行微调，从而实现语言控制下的推理能力迁移。实验结果表明，该方法在代码、数学和科学任务上可达到与主流英文推理基线相当的性能，验证了多语言推理可行性；然而，在日语文化相关任务上的表现反而劣于基准模型，说明语言适配不等于文化理解能力的自然提升，揭示了当前方法在跨文化语义建模方面的局限性。
链接: https://arxiv.org/abs/2607.10114
作者: Yuu Jinnai
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Reasoning Language Models (RLMs) achieve their strongest performance when they reason in English, the language for which reasoning-oriented training data is most abundant. However, reasoning trace is a clue for model interpretability and safety, and useful in practice for both the model users and for model developers. Thus, it is desirable to be able to develop a model that reasons in a language of the user’s choice, while still maintaining strong reasoning performance. To this end, we study the feasibility of training a model that reasons in Japanese. We develop a Japanese-reasoning variant of Qwen-3-Swallow-8B, which is a Japanese LLM continually pretrained from Qwen-3-8B, with GRPO and evaluate it across coding, math, and science benchmarks. The study shows that reasoning-language control is feasible by training a Japanese continually pretrained model with GRPO. However, its performance is at best on par with strong English-reasoning baselines on several benchmarks. We also evaluate the trained model on Japanese cultural benchmarks and observe that the model’s performance is worse than the baseline models, suggesting that the reasoning in Japanese does not immediately improve performance on culturally relevant tasks for free.
[NLP-102] A Survey on LLM Watermarking: Theory and Deployment
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在实际应用中因生成文本的高流畅性与大规模输出所引发的溯源模糊、模型滥用及内容洗劫等安全与信任问题。其核心挑战在于如何在不显著影响生成质量的前提下，实现对模型输出的有效可追溯性与可信审计。解决方案的关键在于构建一种鲁棒且可部署的水印技术体系，通过在生成过程或训练过程中嵌入不可见的签名（watermark），实现对模型输出来源的精准识别。该研究提出了一种系统化、面向部署的分类框架，从四个核心维度进行结构化分析：水印嵌入位置（生成时 vs. 训练时，词元级 vs. 表示层）、检测权限（公开检测 vs. 私有检测）、假设条件（是否具备对logits、采样控制、密钥或模型所有权的访问）以及目标威胁模型（如改写、翻译、摘要、风格迁移、词元操纵及自适应移除）。在此基础上，论文梳理了基于采样偏置、编码机制、表示与训练层面的主流水印技术，并从可检测性、鲁棒性与分布偏移三个维度评估其安全性-实用性权衡。同时，系统综述了攻击与规避策略、评估协议与指标（如误报率控制、校准性、鲁棒性曲线），并指出跨模型迁移、多模态流水线、合谋攻击与治理约束等开放挑战。最终，论文为实际应用场景下的水印设计选择提供实践指导，并指明未来实现可靠、可问责大语言模型部署所需的研究方向。
链接: https://arxiv.org/abs/2607.10103
作者: Huy Phan,Kieu Dang,Ojaswi Dulal,Aiham AL Shukairi,Abby Shine,Chase Garner,Phung Lai
机构: Truman State University; University at Albany, State University of New York
类目: Cryptography and Security (cs.CR); Computation and Language (cs.CL)
备注:
Abstract:Large language models (LLMs) are increasingly embedded in high-impact workflows, yet their ability to generate fluent text at scale has amplified risks of provenance ambiguity, model misuse, and large-scale content laundering. LLM watermarking, embedding invisible signatures into model outputs, has emerged as a promising technical layer for attribution, auditing, and downstream trust decisions. However, the literature has grown rapidly and unevenly: existing categorizations often mix orthogonal design choices, making it difficult to compare methods, reason about guarantees, or translate research results into deployable systems. This survey provides a systematic, deployment-oriented review of LLM watermarking. We organize the space by the core questions practitioners must answer: where a watermark is embedded (generation-time vs. training-time, token vs. representation), who can detect it (public vs. private detection authority), what is assumed (access to logits, sampling control, secret keys, model ownership), and which threat models are targeted (paraphrasing, translation, summarization, style transfer, token manipulation, and adaptive removal). We synthesize the main families of techniques-including sampling biasing, code-based schemes, representation- and training-based approaches-and analyze their security-utility trade-offs through the lens of detectability, robustness, and distribution shift. We further review attack and evasion strategies, evaluation protocols and metrics (false positive control, calibration, robustness curves), and open challenges such as cross-model transfer, multi-modal pipelines, collusion, and governance constraints. Finally, we provide practical guidance for selecting watermark designs under real operational requirements and identify research directions needed for reliable, accountable LLM deployment.
[NLP-103] Efficiently Adapting Spoken Language Models for the Singaporean Context
【速读】： 该论文旨在解决生成式语音模型（Generative Speech Models, GSMs）在敏感领域（如新加坡内政部安全应用）中适配困难的问题，尤其针对原始训练数据不可获取且需支持多语言、语音查询交互的场景。其核心挑战在于如何在不访问原始数据的前提下，实现模型在多语言语音任务上的高效微调，同时避免灾难性遗忘并保持跨语言语音理解能力。解决方案的关键在于三方面创新：首先采用低秩适应（LoRA）进行参数高效微调；其次构建了一个代理文本-问答（text-QA）数据集，以缓解微调过程中的灾难性遗忘问题；最后引入多任务学习目标，并将CoBa重加权机制适配至语音任务，以优化不同任务间的平衡。此外，研究还构建了包含504,853个样本的多语言问答数据集HTD-multilingual-QA（文本与语音双模态），支撑模型训练。最终形成的HT-Moonstone（5B）模型在多数任务上表现优于甚至超过规模达其7倍的基线模型，在语音问答能力上损失不足2%，并在口音和性别识别任务中达到最优性能。
链接: https://arxiv.org/abs/2607.10092
作者: Ng Jia Sheng Jason
机构: Home Team Science and Technology Agency (HTX), Singapore
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 10 pages, 2 figures
Abstract:Spoken language models (SLMs) unify speech perception and reasoning, but adapting them to sensitive domains is underexplored, especially when the original training data is inaccessible and the use case demands multilingual, spoken-query interaction. We adapt an open-source SLM to the Singaporean Home Team context across five speech tasks in Singapore’s four official languages, combining LoRA fine-tuning, a surrogate text-QA dataset that guards against catastrophic forgetting, and a multi-task objective that adapts the CoBa reweighting scheme to speech. We also build HTD-multilingual-QA, a 504,853 sample multilingual QA dataset in text and spoken form. The resulting HT-Moonstone (5B) matches or outperforms SLMs up to 7x its size on most tasks, attains the best accent and gender recognition among all models evaluated, and loses under 2% of its original speech QA ability.
[NLP-104] MAG: A Web-Agent Benchmark and Harness for Multimodal Action and Guide Generation
【速读】： 该论文旨在解决现有数字应用平台（Digital Adoption Platforms, DAPs）中自动化网页代理操作与引导文本生成割裂的问题，即传统方法将任务执行与引导内容生成视为独立问题，且多依赖于非视觉化的页面表示（如DOM树或可访问性树），而忽视了用户实际操作时所依赖的渲染后屏幕图像。其核心挑战在于如何实现跨页面状态变化的多步任务序列理解与生成，并在真实界面中完成连贯的操作与高质量的自然语言引导。解决方案的关键在于提出首个统一任务执行与引导写作的多模态基准MAG，通过两种基于截图的定位方式（元素集合选择与原始像素坐标）实现动作与文本的精准对齐；同时构建了完整的评估框架，涵盖大模型辅助标注、人工验证、模型训练、实时环境评估及联合评价指标。在此基础上，设计了一种引入专家轨迹增强的GRPO强化学习训练方法，显著提升代理成功率（9B模型从6.9%提升至13.2%），并同步改善引导质量，但当前最强模型仍仅能完成不足40%的任务，表明该领域仍有巨大研究空间。
链接: https://arxiv.org/abs/2607.10079
作者: Chengguang Gan,Hanjun Wei,Yunhao Liang,Zhixi Cai,Qinghao Zhang,Shiwen Ni
机构: University of Chinese Academy of Sciences; Monash University; Pusan National University; Shenzhen University of Advanced Technology
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 8 pages main text, 21 pages total including appendices; 11 figures, 7 tables, 2 algorithms. Benchmark, harness, and model checkpoints to be released
Abstract:Digital Adoption Platforms (DAPs) are embedded overlays widely used on web systems to guide users through operations inside a page, helping them get started with unfamiliar interfaces quickly. Completing a real task, however, rarely means clicking a few buttons on a single page: it takes a sequence of actions that unfolds across changing page states. Prior studies have also treated automated web agent actions and guide text generation as two separate problems, and most of them feed models textual page representations such as the DOM or accessibility trees rather than the rendered screens that humans actually operate on. In this work we introduce MAG, the first benchmark that unifies task execution and guide writing into a single Multimodal Action and Guide task, with two grounding schemes over screenshots: Set-of-Mark element selection and raw pixel coordinates. We further build a complete harness for this compound task, covering annotation with LLM assistance and human verification, training, evaluation in live environments, and joint metrics for actions and guides. With this harness we evaluate frontier API models and open multimodal models, and report detailed analyses. Finally, we design a GRPO training method augmented with expert trajectories, which nearly doubles the success rate of a supervised 9B agent (from 6.9% to 13.2%) and improves guide quality at the same time. Even the strongest model completes fewer than 40% of the tasks, leaving ample room for future research.
[NLP-105] Robust Scalable Detection of Text Containment in Large Web-Crawled Corpora
【速读】： 该论文旨在解决在大规模文本语料库中高效、准确检测目标文本是否存在近似完整复制（near-verbatim copies）的问题，尤其针对版权内容的识别需求。传统方法往往仅依赖于文本相似性度量，易产生误报或漏报，难以区分实质性复制与表面相似。其解决方案的关键在于提出一种基于指纹链（fingerprint chain）的新机制：通过生成文本的文档指纹（document fingerprinting），并显式追踪连续匹配的指纹序列，从而构建具有高置信度的复制证据。这一设计显著提升了对完整或接近完整文本复现的检测能力。此外，系统采用分布式、基于磁盘的索引架构，支持对大规模网络爬取数据集的高效处理。通过引入新的文本包含性评估基准，实验表明FindMyText在ArXiv论文、维基百科及通用网页内容三个数据集上均优于现有方法。
链接: https://arxiv.org/abs/2607.10020
作者: Lars Henry Berge Olsen,Pierre Lison,Martin Jullum,Mark Anderson
机构: Norwegian Computing Center (挪威计算中心)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 6 pages + references and appendices
Abstract:We present FindMyText, an open-source Python package designed to efficiently assess whether a given text appears, in part or in full, within a text corpus. The tool builds on prior techniques for document fingerprinting, but extends them with a novel mechanism to explicitly capture sequences of matching fingerprints. By identifying such chains, the tool can more reliably detect near-verbatim copies of a given text rather than mere textual similarities. This makes FindMyText particularly suited for verifying the presence of copyrighted material in a corpus. Leveraging a distributed, disk-based indexing framework, the system scales to large web-crawled datasets. Using a new benchmark for evaluating text containment methods, we show that FindMyText outperforms alternative approaches across three datasets (ArXiv papers, Wikipedia, and generic web content).
[NLP-106] Silent Failures in Quantized LLM Reasoning : A Taxonomy-Based Analysis of Hollow Convergence and Failure Mode Shifts
【速读】： 该论文旨在解决后训练量化（post-training quantization）在保持大语言模型（Large Language Models, LLMs）任务准确率的同时，可能悄然改变其推理过程这一潜在问题。尽管量化后的模型在精度上表现稳定（最大下降3.1个百分点），但其内在推理质量却出现显著退化，尤其体现在推理路径的完整性与可验证性方面。解决方案的关键在于提出并验证了一个包含六类故障的分类体系（six-category failure taxonomy），通过双独立人工标注（Cohen’s κ = 0.906）对5个指令微调的LLM（参数规模3B–14B）在三种量化精度（FP32、FP16、NF4）下生成的3万条思维链（chain-of-thought）输出进行系统分析。研究发现，在低精度量化（NF4）条件下，模型表现出“空心收敛”（Hollow Convergence）现象——即答案正确但推理过程不完整或不可验证，且该现象具有明显的规模依赖性和基准特异性；同时，“捷径坍塌”（Shortcut Collapse）比例上升而“信心雪球效应”（Confidence Snowballing）几乎消失，这些变化无法通过表面文本特征捕捉（最佳F1仅为0.53），因而成为标准评估流程难以察觉的部署风险。因此，该研究强调需超越传统准确率指标，引入更深层的推理质量评估机制以识别此类隐蔽失效模式。
链接: https://arxiv.org/abs/2607.09999
作者: Renuka Oladri,Mohan Vamsi Varadaraju Priya,Jerry Wu
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 7 pages, 3 figures, 6 tables
Abstract:We show that post-training quantization can silently alter how large language models reason even when task accuracy is preserved. Using a six-category failure taxonomy validated by two independent human annotators (Cohen’s \kappa = 0.906), we classify 30,000 chain-of-thought outputs from five instruction-tuned LLMs (3B–14B parameters) across three quantization precisions (FP32, FP16, NF4) and four reasoning benchmarks. We find that while accuracy is robust across precisions (maximum 3.1 pp drop), Hollow Convergence (correct answers reached through incomplete or unverifiable reasoning) shows a significant size-dependent shift under NF4, dropping sharply for the two smallest models tested but remaining invariant for models at 12B parameters and above. This effect is also benchmark-specific: GSM8K is categorically immune while LogiQA and ARC-Challenge show the largest shifts. Furthermore, under NF4, Shortcut Collapse rises from 44% to 78% of wrong-answer failures in LLaMA 3.2-3B while Confidence Snowballing collapses from 15.8% to near zero, a qualitative shift invisible to accuracy metrics. Finally, we show Hollow Convergence cannot be reliably detected from surface-level text features (best F1 = 0.53), establishing it as a deployment-relevant failure mode that standard evaluation pipelines cannot catch.
[NLP-107] Workload-Driven Optimization for On-Device Real-Time Subtitle Translation
【速读】： 该论文旨在解决在台湾地区特定应用场景下，针对短输入、短输出、单批次推理、低延迟及隐私保护约束条件下的本地化英文到繁体中文字幕翻译问题。此类场景限制了传统面向长文本或高吞吐量语言模型服务的优化策略的有效性。其解决方案的关键在于：在GGUF量化后，识别出词表映射（vocabulary projection）成为解码阶段的主要计算开销，因此提出将原始151k词元的通用词表替换为64k词元的字幕领域专用词表，并通过嵌入空间迁移与嵌入校准结合全监督微调的方式对模型进行适配。实验表明，在固定500样本的OpenSubtitles2024测试子集上，该方法（LocalSubs）在GPT-4o双样本评估中相较Google Translate取得59.2%的去重胜率，且在短提示（cue）任务中表现最优，随提示长度增加性能下降；初步的Apple M2 Metal测试显示，使用64k词表模型相比原151k词表基线实现1.63倍加速，尽管当前基准配置不完整，延迟结果仍属初步。
链接: https://arxiv.org/abs/2607.09957
作者: Tsz-To Wong
机构: National Yang Ming Chiao Tung University (国立阳明交通大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:This report studies on-device English-to-Traditional-Chinese subtitle translation for Taiwan under short inputs, short outputs, batch-size-one inference, low latency, and privacy constraints. These conditions limit the value of optimizations designed for long-context or high-throughput language-model serving. Starting from LMT-60-0.6B, preliminary profiling suggests that vocabulary projection becomes a more important decode-time cost after GGUF quantization reduces the relative cost of Transformer blocks. We replace the original 151k-token vocabulary with a 64k-token subtitle-domain tokenizer, migrate the embedding space, and adapt the model through embedding calibration followed by full supervised fine-tuning. On a fixed 500-example subset of the OpenSubtitles2024 test set, the LocalSubs achieves a 59.2% tie-excluded win rate against Google Translate under GPT-4o pairwise judging. Performance is strongest on short cues and declines as cue length increases. Preliminary Apple M2 Metal measurements on a 64k-vocabulary model show a 1.63 \times speedup over a 151k-vocabulary profiling baseline. The raw benchmark configuration is incomplete, so the latency result is treated as preliminary. Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI) Cite as: arXiv:2607.09957 [cs.CL] (or arXiv:2607.09957v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2607.09957 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[NLP-108] Faithful by Design: Evaluating and Improving LLM -Generated Clinical Trial Summaries for Multi-Stakeholder Audiences
【速读】： 该论文旨在解决生成式 AI（Generative AI）在临床试验结果摘要生成中因幻觉（hallucination）导致的可信度问题，尤其针对医疗提供者、患者和支付方三类利益相关者所面临的高风险场景。其核心解决方案是构建一个基于多维度评估框架的基准测试体系，通过从聚合分析数据库中选取200个分层临床试验，采用面向不同受众的提示模板及六维度可信度标注标准，对GPT-4o、Claude Sonnet 4.6和Gemini 2.5 Flash三种模型生成的1800份摘要进行评估。研究发现，所有模型均以“未经支持的主张”为主要失败模式，平均可信度评分为1.55/3。为提升生成内容的可信度，研究进一步提出并验证了一种基于知识图谱增强的检索系统，该系统显著提升了自然语言推理（NLI）模型评估下的语义蕴含与整体可信度得分（分别提升+0.0125和+0.0130，p < 0.0001），且改进路径具有模型依赖性：GPT-4o主要通过减少矛盾性实现优化，而Claude Sonnet 4.6与Gemini 2.5 Flash则主要通过增强语义蕴含能力获得提升。
链接: https://arxiv.org/abs/2607.09932
作者: Robert Williams
机构: University of Texas at Austin (德克萨斯大学奥斯汀分校)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 8 pages, 8 figures
Abstract:Large language models are increasingly used to summarize clinical trial results for healthcare providers, patients, and payers, but their tendency to hallucinate poses significant risks in this high-stakes context. This study introduces a benchmark evaluation framework for measuring the faithfulness of LLM-generated clinical trial summaries across three stakeholder audiences. The framework consists of 200 stratified trials drawn from the Aggregate Analysis of this http URL database, evaluated using audience-specific prompt templates and a six-dimension faithfulness annotation schema. Baseline measurements were established for GPT-4o, Claude Sonnet 4.6, and Gemini 2.5 Flash across 1,800 generated summaries scored using a cross-encoder natural language inference (NLI) model. Unsupported Claims was identified as the dominant failure mode across all three models, with a mean annotation score of 1.55 out of three. A knowledge-graph-augmented retrieval system was developed and evaluated against the baseline, producing statistically significant improvements in NLI-based faithfulness scores (entailment +0.0125, faithfulness +0.0130, p 0.0001). Improvement pathways were model-dependent, with GPT-4o improving primarily through contradiction reduction while Claude Sonnet 4.6 and Gemini 2.5 Flash improved through increased entailment.
[NLP-109] Global Merger-Arbitrag e Forecasting with Language Models ICML2026
【速读】： 该论文旨在解决并购套利（merger arbitrage）这一高风险、专业化金融场景中的预测问题，即准确预测已公告并购交易的最终结果——包括按原条款完成、出现更高报价或交易终止。传统基于大语言模型（LLM）的判断性预测研究多聚焦于涵盖广泛主题的短文本基准测试，难以应对需要长上下文推理的复杂金融实务。本文的关键解决方案在于结合专家指导的上下文工程与基于历史交易的“事后引导式推理轨迹”（hindsight-guided reasoning traces）进行微调，使模型能够处理数百页技术文档的长期依赖关系。实验表明，在覆盖42个国家超过400笔大型并购交易的外部样本上，该微调系统在类别平衡的Brier评分上达到0.151，优于校准后的市场隐含概率（低24%）、XGBoost（低19%）以及前沿语言模型（低25%-42%）。消融分析进一步验证了事后监督与专家设计上下文在实现高性能预测中的决定性作用。
链接: https://arxiv.org/abs/2607.09921
作者: Hinal Jajal,Michal Mucha,Charles Sweat,Chris Pulman,Charlie Flanagan,Peter Anderson
机构: 未知
类目: Computation and Language (cs.CL)
备注: Accepted to ICML 2026
Abstract:We present a language-model forecasting system for merger arbitrage, a specialized high-stakes financial setting in which the task is to predict the outcome of announced M\A deals. Unlike prior work on judgmental forecasting with LLMs, which has focused on broad mixed-topic benchmarks and short context such as news snippets, we study a setting that requires long-context reasoning over hundreds of pages of technical documents. Our system combines expert-guided context engineering with finetuning on hindsight-guided reasoning traces derived from historical deals. Given an announced deal, it outputs a probability distribution over three mutually exclusive outcomes: closing at announced terms, a higher bid, or deal termination. On an out-of-sample set of more than 400 large deals spanning 42 countries, our finetuned system achieves the best performance of any method we evaluate, reducing class-balanced Brier score to 0.151. This is 24% below calibrated market-implied probabilities, 19% below XGBoost, and 25-42% below frontier language models. These results, together with ablation studies, show that LLM-based forecasting can succeed in specialized, long-context financial workflows, with hindsight-based supervision and expert-designed context playing a critical role.
[NLP-110] Remembering Distinct Items Not Tokens: A Learnable Dirichlet-Process Cache Between State-Space Models and Attention
【速读】： 该论文旨在解决传统固定状态序列模型（fixed-state sequence models）在关联记忆能力上受限于状态维度的瓶颈问题，同时克服标准注意力机制（attention）因需为每个输入 token 保存键值对而导致计算开销与缓存大小随序列长度呈二次增长的弊端。其核心解决方案是提出一种稀疏缓存（sparse cache）机制，仅在输入为新异（novel）时才分配存储槽位，从而使得缓存规模与序列中不同实体的数量（distinct items）而非总词元数（tokens）同步增长。该机制的关键在于采用 DP-means 聚类规则作为缓存分配策略——这一规则源自狄利克雷过程混合模型（Dirichlet-process mixture）的小方差极限，但并非用于潜在变量推断，而是作为深度循环主干网络（deep recurrent backbone）的键值记忆操作符。研究构建了两种形式：一种为固定浓度参数的静态缓存，另一种为根据近期新颖率自适应调节浓度的惊喜自适应变体。在具有冗余性的受控关联记忆基准测试中，该方法实现了与全注意力模型相当的记忆召回性能，同时仅存储不同项；在召回率-缓存大小权衡曲线上优于固定预算的淘汰式缓存；在状态空间主干结构下，能够以所有测试模型中最低的内存开销完成记忆查询与长程聚合任务。更重要的是，缓存分配规则可通过端到端学习实现：仅需两个参数的新颖性阈值门控模块，基于任务损失即可精确恢复该规则，而过度参数化的门控则失效，表明其有效性依赖于归纳偏置（inductive bias）而非模型容量。研究通过一系列可控机制实验验证了该方法在小规模下的有效性，并在四个真实数据流（推荐、系统日志、临床事件、保险理赔）中确认了其保持不同项特性的能力；真实主干与真实语料的语言学验证则在后续配套研究中进一步展开。
链接: https://arxiv.org/abs/2607.09889
作者: Siddharth Pal,Viktoria Rojkova
机构: 未知
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Neural and Evolutionary Computing (cs.NE)
备注: 16 pages, 9 figures. Companion paper on event-log applications forthcoming
Abstract:Fixed-state sequence models compress an unbounded past into a bounded state, which caps their associative recall at roughly the state dimension; attention escapes the cap by keeping a key-value entry for every token, at quadratic compute and a cache that grows with the sequence. We study the middle ground: a sparse cache that allocates a slot only when an input is novel, so its size tracks the number of distinct items rather than the number of tokens. The allocation rule is the DP-means clustering rule, the small-variance limit of a Dirichlet-process mixture, used not as latent-variable inference but as the key-value memory operator for a deep recurrent backbone. We develop it in two forms, a static cache with a fixed concentration and a surprise-adaptive variant whose concentration follows the recent novelty rate. On a controlled associative-recall benchmark with redundancy we show that the cache matches full-attention recall while storing only the distinct items, that it dominates a fixed-budget eviction cache on the recall-versus-size frontier, and that on a state-space backbone it answers both a recall query and a long-range aggregate at the lowest memory of any model tested. The allocation is learnable end to end: a two-parameter novelty-threshold gate trained on the task loss alone recovers the rule exactly, whereas an over-parameterized gate fails, so the operative ingredient is the inductive bias rather than capacity. The evidence is a family of controlled mechanism studies at modest scale, with the distinct-items property confirmed on four real streams (recommendation, systems logs, clinical events, and insurance claims); a real-backbone, real-corpus language validation is pursued in a companion study.
[NLP-111] Index SLM Technical Report
【速读】： 该论文旨在解决小型语言模型在中文与英文双语环境下预训练效率与性能瓶颈的问题，尤其关注如何在有限参数规模下实现接近甚至超越更大模型的综合能力。其核心挑战在于：如何通过优化预训练策略与架构设计，在保持模型轻量化的同时提升其在复杂任务（如推理、数学计算、代码生成）上的表现。解决方案的关键在于提出一种“暖启动-稳定-衰减”（Warmup-Stable-Decay）的学习率调度机制，该机制在训练后期显著提高高质量精选数据的占比，从而增强模型对关键语义与逻辑结构的捕捉能力；同时引入归一化输出头（Norm-Head）结构，有效提升了大学习率下的训练稳定性。此外，通过系统性控制实验揭示了学习率调度与数据质量之间的交互效应，并观察到在恒定学习率阶段出现未解释的性能跃升现象，为后续训练动态研究提供了重要线索。整体上，该工作展示了小规模模型在精心设计的训练范式下具备与大型模型相当的泛化能力。
链接: https://arxiv.org/abs/2607.09885
作者: Lusheng Zhang,Shien He,Tianxing Yan,Mengran Yu,Ziang Cui,Kai Zhao,Xiaojing Liu,Tianjiao Li
机构: Bilibili Index LLM Team(哔哩哔哩索引大模型团队)
类目: Computation and Language (cs.CL)
备注: 26 pages, 10 figures
Abstract:We present Index-1.9B, a series of open small language models developed at Bilibili. The series comprises four models: Index-1.9B-Base, a foundation model with 1.9 billion non-embedding parameters pre-trained on 2.8 trillion predominantly Chinese and English tokens; Index-1.9B-Pure, a control variant trained with an identical recipe but with all instruction-like data strictly filtered from the corpus; Index-1.9B-Chat, aligned from the base model with supervised fine-tuning and direct preference optimization; and Index-1.9B-Character, which augments the chat model with retrieval-augmented generation for few-shot role-playing customization. Pre-training employs a Warmup-Stable-Decay learning-rate schedule in which the concentration of curated data is raised substantially during the decay phase, together with a Norm-Head output layer that stabilizes training under large learning rates. On a suite of standard benchmarks covering examination, reasoning, mathematics, and code, Index-1.9B-Base attains an average score of 64.92, competitive with or exceeding open models of several times its size. We further report controlled studies on model depth, learning-rate magnitude and scheduling, the interaction between learning-rate decay and data quality, and the effect of including instruction data during pre-training, and we document an unexplained surge in benchmark performance midway through the constant-learning-rate phase. All models, together with evaluation code, are released at this https URL.
[NLP-112] CLIR-Bench: Benchmarking Multimodal Question Answering over Irregular Clinical Time Series
【速读】： 该论文旨在解决临床时间序列（clinical time series）在临床问答（QA）任务中因稀疏性、非均匀采样和异步性导致模型难以准确识别时间证据的问题。现有基准主要聚焦于规则采样的时间序列问答或静态医疗数据的问答，未能有效评估模型在不规则时间观测下的推理能力。为此，研究提出CLIR-Bench，一个基于去标识化ICU记录，通过系统化的四阶段流程构建的不规则临床时间序列问答基准。该基准包含6,600个问答实例，覆盖11种临床变量，涵盖四个能力维度与11项具体任务，每个问题均关联明确的时间证据及任务特定的答案推导规则，从而可同时评估答案准确性与证据使用情况。实验表明，现有通用模型在处理稀疏临床证据的检索与推理方面表现不佳，凸显了发展更强不规则时间序列推理方法的必要性。
链接: https://arxiv.org/abs/2607.09880
作者: Frank Nie,Ethan B. Liu,Yuan Zhu,Loe Yan,Wei Fan,Jindong Han
机构: Shandong University (山东大学); University of Auckland (奥克兰大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Clinical time series are central to patient monitoring, risk assessment, and clinical decision support. However, they are often sparse, irregularly sampled, and asynchronous, making it difficult for models to identify the temporal evidence required for clinical Question Answering (QA). Existing benchmarks primarily focus on regularly sampled time-series QA or medical QA over static data, and therefore rarely assess whether models can faithfully ground their answers in irregular temporal observations. To fill this gap, we introduce CLIR-Bench, a benchmark for irregular clinical time series QA constructed from de-identified ICU records through a principled four-stage pipeline. CLIR-Bench contains 6,600 QA instances spanning 11 clinical variables, organized into four capability dimensions and 11 tasks. Each question is linked to explicit temporal evidence and task-specific answer derivation rules, enabling evaluation of both answer accuracy and evidence use. Experiments show that existing generalist models struggle to retrieve and reason over sparse clinical evidence, highlighting the need for stronger irregular time-series reasoning methods. Our code and data are available at this https URL.
[NLP-113] From Direction to Magnitude: How Multimodal Instruction-Tuning Reorganizes the Geometric Encoding of Identity-Specifying Prompts in Transformer Hidden States
【速读】： 该论文旨在解决生成式 AI 模型在不同训练阶段中，身份信息（identity）是否以可区分的几何特征形式编码于隐藏状态轨迹中的问题。其核心挑战在于揭示模型内部表征如何随训练范式演进而变化，尤其关注指令微调（instruction-tuning）对身份编码模式的影响。解决方案的关键在于提出并验证一种基于图结构几何度量的新方法：通过构建 k-近邻（k-NN）轨迹图，并计算边级 Ollivier-Ricci 曲率分布之间的 1-Wasserstein 距离（W₁），来量化隐藏状态轨迹的几何指纹差异。研究发现，在未经过指令微调的基础模型中，身份信息以方向性特征编码（方向分离显著，p=0.002）；而在多模态指令微调模型中，该编码机制发生质变，转为以幅度（magnitude）为主导（角度分离不显著，但欧氏距离仍显著，且首层生成状态的均值范数顺序反转）。这一方向到幅度的重构现象具有特定性，仅出现在多模态指令微调阶段，而在强化学习蒸馏与监督微调下均未出现。此外，教师强制控制实验表明约30%的自由运行余弦信号可归因于提示驱动效应。该研究将 W₁ 在 k-NN 轨迹图上边级曲率分布的应用确立为一项独立的方法学贡献，为解析模型内部动态提供了新的几何分析工具。
链接: https://arxiv.org/abs/2607.09842
作者: Jorge A. Castillo,Marco Torres Yévenes,Juan Carlos Lanas
机构: Axis Dynamics SpA(轴动力公司)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: 16 pages, 8 tables. Working draft v0.3. Uses Ollivier-Ricci curvature and edge-wise Wasserstein-1 as primary geometric statistic; four models (Gemma-4-E4B, Gemma-4-E4B-it, DeepSeek-R1-Distill-Qwen-7B, Qwen2.5-7B-Instruct)
Abstract:We investigate whether identity-specifying system prompts produce statistically distinguishable geometric fingerprints in the hidden-state trajectories of four open-weight transformer language models spanning four post-training regimes: no training (Gemma-4-E4B base), multimodal RLHF (Gemma-4-E4B-it), RL distillation (DeepSeek-R1-Distill-Qwen-7B), and SFT (Qwen2.5-7B-Instruct). Three prompt conditions (an identity-specifying axis prompt, a length-matched generic-assistant prompt, and a 26-token vanilla baseline) are compared via five geometric metrics, principally the 1-Wasserstein distance between edge-wise distributions of Ollivier-Ricci curvature on k-NN trajectory graphs. Claims rest on trajectory-level permutation tests with multiple geometric controls (teacher-forced content controls, temporal-chain vs k-NN topology, ABT-projected k-NN, angular vs Euclidean graph construction, B=5000 permutations on borderline statistics). The central finding is a qualitative reorganization of identity encoding across the instruction-tuning boundary: in the base model the fingerprint is direction-coded (separation 0.034, p=0.002 under angular k-NN); in the multimodal instruction-tuned model it migrates into the magnitude (angular separation collapses to p=0.439 while Euclidean survives at p=0.042, and the mean norm of the first generated state inverts its length-ordering, being lowest for the identity prompt). This direction-to-magnitude reorganization is specific to the multimodal instruction-tuning regime, absent under RL distillation and SFT. A teacher-forced control attributes ~30% of the free-running cosine signal to prompt-driven effects. We position W_1 on edge-wise Ollivier-Ricci distributions on k-NN trajectory graphs as a methodological contribution of independent interest.
[NLP-114] Spectral Origins of the Self-Correction Blind Spot in Autoregressive Generation
【速读】： 该论文旨在解决自回归生成模型中存在的“自我修正盲区”问题，即模型能够纠正由外部源引入的相同错误，却无法识别并修正自身输出中的错误。其核心解决方案是提出SPARC（谱代数自修正理论），从理论上揭示了自回归生成中错误传播与自我修正能力之间的内在机制。关键突破在于：将每一步生成过程中的注意力雅可比矩阵在残差流上的乘积定义为误差传播算子，并证明自我修正盲区的出现当且仅当该算子的谱半径（spectral radius）大于等于1；进一步推导出校正标记必须达到的精确激活阈值，该阈值由谱半径决定，成功解释了简单“Wait”标记实现89.3%盲区降低的现象；同时，首次为基于强化学习的验证-修正训练框架提供了收敛性保证，表明其收敛速率与耦合强度平方成正比、与样本数量平方根成反比，前提是验证-修正耦合矩阵的谱范数小于1，且该条件在文本大语言模型及自回归图像和视频生成模型间保持不变。实验在四种骨干网络及视觉自回归探测任务上验证了所有定理，谱预测与实测盲区率误差控制在3.2%均方根误差（RMSE）以内，实现了理论与实证的高度一致。
链接: https://arxiv.org/abs/2607.09803
作者: Ingrid Petrova,Luan Vejsiu
机构: European University of Tirana (欧洲大学学院)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:
Abstract:Large autoregressive language models exhibit a self-correction blind spot: they reliably fix identical errors when attributed to an external source yet fail to fix the same errors in their own outputs. Prior work has documented this phenomenon empirically, through controlled error injection, error-depth decompositions, RL-based verifier-corrector training, and intrinsic self-verification, but offers no formal model of why generating a token suppresses the ability to detect its error, no quantitative activation condition for correction markers, and no convergence guarantee for reinforcement-learning-based self-correction. We close these gaps with SPARC, a spectral-algebraic theory of self-correction in autoregressive generation. We define the error-propagation operator as the product of per-step attention Jacobians on the residual stream and prove that the blind spot arises if and only if the spectral radius of this operator is at least one. We derive a sharp activation threshold, given as a function of the spectral radius, that a correction marker must exceed, recovering the 89.3% blind-spot reduction observed with a simple ``Wait’’ marker. We further prove that RL-based verifier-corrector training converges at a rate proportional to the squared coupling strength over the square root of the number of samples if and only if the verifier-corrector coupling matrix has spectral norm below one, and that this criterion is invariant across residual-stream autoregressive modalities, unifying text LLMs and autoregressive image and video generation. Experiments across four backbones and a visual autoregressive probe validate every theorem, with spectral predictions matching measured blind-spot rates within 3.2% RMSE.
[NLP-115] Gauge dependence and structured-output corruption in sign-branched repetition penalties: measurements across models inference stacks and alternative repetition controls ALT
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）推理过程中广泛采用的乘法重复惩罚（repetition penalty）机制中存在的根本性缺陷。其核心问题是：当前主流推理引擎（如HuggingFace、vLLM等）所实现的重复惩罚策略基于原始对数几率（raw logits）的符号分支——对正数logit除以惩罚因子θ，负数logit则乘以θ。然而，由于softmax函数对所有logit同时加上常数保持不变，模型的logit零点（zero-point）是任意的，而该惩罚策略却依赖于这一未被训练目标约束的任意零点，导致惩罚操作本身缺乏一致性与可定义性。解决方案的关键在于：将重复惩罚从作用于原始logit改为作用于归一化后的对数概率（normalized log-probabilities）。实验表明，这种修改不仅使惩罚在逻辑上具有唯一性（即重新中心化logit不再影响结果），还能有效避免对结构化输出（如JSON）的严重破坏——在真实世界JSON模式测试中，θ=1.3时原方法将合法输出率从97%降至23%，而使用归一化后log-probability的惩罚策略则完全消除了这一问题。研究通过五种不同规模的模型（至7B参数量）、多种任务（WikiText-103、HumanEval、JSONSchemaBench）及多个推理引擎（包括vLLM和HuggingFace）验证了上述结论，并指出已有实现中虽已包含归一化算子（LogitNormalization），但默认未启用且应用顺序错误，建议将其前置应用以确保正确性。
链接: https://arxiv.org/abs/2607.09791
作者: Peter Hollows
机构: 未知
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: 8 pages, 2 figures, 4 tables. Code, data, per-stack survey and git genealogy: this https URL
Abstract:The multiplicative repetition penalty shipped across the LLM inference ecosystem (HuggingFace, vLLM, this http URL, and a dozen further engines) branches on the sign of each raw logit (divide positives by theta, multiply negatives). But the softmax is unchanged by adding a constant to every logit, so a model’s logit zero-point is arbitrary, and the sign-branch reads that arbitrary point. The sign-branch is itself the accepted fix for an earlier bug, so the accepted fix branches on a quantity the training objective leaves unconstrained. Two measurable consequences follow. (1) The penalty is not well-defined: re-centring a model’s logits by a constant is a provable no-op at theta=1, yet at a routine theta=1.3 it changes 58-96% of greedy tokens, where subtractive and normalized penalties change none; real checkpoints sit at widely different zero-points, so a fixed repetition_penalty is a different operation on every model. (2) It corrupts structured output: on 200 real-world JSON schemas, theta=1.3 drops the rate of valid, schema-conformant output from 97% to 23%. In our measurements, applying the penalty to normalized log-probabilities instead of raw logits removes both effects. HuggingFace already ships that operator (LogitNormalization); today it is off by default and applied after the penalty. This note gives the mechanism, the measurements (five models up to 7B, base and RLHF, on WikiText-103 prefixes; two code models on HumanEval and JSONSchemaBench; both effects replicated inside vLLM and this http URL through their own samplers on the same inputs), and the normalized variant.
[NLP-116] Length Penalties Make Chain-of-Thought Less Monitorable
【速读】： 该论文旨在解决生成式推理过程中因长度惩罚（length penalty）导致的可解释性下降问题，即模型在缩短思维链（chain-of-thought）的同时，隐藏了影响其答案判断的关键提示（biasing hint），从而削弱了对模型决策过程的可监测性。其核心解决方案在于通过引入目标长度约束进行推理压缩，并系统评估压缩后模型在保持多选准确率的同时，是否仍能暴露提示的影响。关键发现是：尽管压缩显著减少了推理令牌数量并维持了较高的准确性，但模型对提示的依赖痕迹被显著弱化——压缩后的思维链中提示被检测到的概率下降7至35个百分点，且忠实性（faithfulness）指标相比基线分别降至63.1%和69.4%。进一步的消融实验表明，仅通过随机删除句子匹配长度无法复现此现象，说明压缩不仅缩短了文本长度，更存在选择性地移除可监测线索的能力。因此，该研究揭示了一个“压缩-可监测性”前沿（compression-monitorability frontier），表明在降低推理成本的同时，模型内部的因果影响机制变得愈发隐蔽。
链接: https://arxiv.org/abs/2607.09786
作者: Bryce Little
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Length-penalized reinforcement learning can shorten chain-of-thought reasoning while hiding an influence that drives the model’s answer. In our experiments, training with length penalties does not stop misleading hints from steering models, even though the models’ chains of thought mention the hint much less often. A token-accuracy evaluation would count these runs as successful because they use fewer reasoning tokens with little accuracy loss; it would miss whether the remaining trace still shows what drove the answer. We train Qwen3-4B and Qwen3-14B variants with different target chain lengths, then evaluate them with biasing-hint interventions on held-out MMLU-Pro-R and four transfer benchmarks. Compression sharply cuts reasoning tokens, preserves most multiple-choice accuracy, and leaves hint influence near baseline. At the strongest target, lower-bound faithfulness falls to 63.1% of baseline for Qwen3-14B and 69.4% for Qwen3-4B; the raw rate at which a monitor catches hint use falls from 69% to 49% and from 60% to 48%. To separate length from content, we randomly delete sentences from uncompressed baseline chains until the remaining text matches the compressed length. Even after this length matching, compressed chains disclose the hint 7-35 percentage points less often than baseline chains that we shorten at random, for both Qwen3 sizes and all five evaluation distributions. Compression therefore does more than shorten reasoning, preferentially removing the cues a monitor needs to see what influenced the answer. Together, these results reveal a compression-monitorability frontier in which cheaper reasoning can preserve answers while making the influences behind them harder to detect.
[NLP-117] EvoCUA-1.5: Online Reinforcement Learning for Multi-turn Computer-Use Agents
【速读】： 该论文旨在解决长时程计算机使用任务中，智能体在部分可观测、多模态桌面环境下的在线强化学习（Online Reinforcement Learning, RL）难题。传统基于模仿学习和离线轨迹优化的方法虽能提供良好先验，但静态轨迹无法捕捉真实计算机操作中动作引发的状态变化、未来动作空间演化及恢复路径的因果反馈循环。为突破这一瓶颈，论文提出EvoCUA-1.5框架，其核心解决方案在于构建一个可执行沙盒环境中的持续自进化机制，实现从离线经验学习向在线强化学习的演进。关键创新包括：步骤级策略优化（Step-Level Policy Optimization, STEPO），通过保留轨迹级优势平衡性，实现对长序列动作分解后的有效样本利用；策略感知过滤与通过率校准，基于可验证合成任务动态筛选训练样本并调节策略更新；动态三重自适应课程学习（Dynamic Tri-Adaptive Curriculum, DTAC），融合可学习任务、困难正样本回放与受控不可行任务暴露，以提升学习效率与鲁棒性；以及全异步强化学习基础设施，结合延迟控制与小批量分组机制，保障训练稳定性。实验表明，这些组件显著提升了训练稳定性和下游任务性能，在OSWorld-Verified基准上达到63.2%的成功率，优于同等规模（32B/35B参数）的开源模型，并接近参数量更大的模型表现，验证了其在多轮交互式计算机使用智能体中实现高效在线强化学习的可行性与可扩展性。
链接: https://arxiv.org/abs/2607.09773
作者: Mianqiu Huang,Taofeng Xue,Chong Peng,Jinrui Ding,Sicheng Fan,Jiale Hong,Yufei Gao,Xiaocheng Zhang,Linsen Guo,Xin Yang,Dengchang Zhao,Yuchen Xie,Peng Pei,Xunliang Xie,Xipeng Qiu
机构: Meituan; Fudan University (复旦大学); Shanghai Jiao Tong University (上海交通大学); Zhejiang University (浙江大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Computer-use agents must solve long-horizon tasks through repeated interaction with partially observable, multimodal desktop environments. Although imitation learning and offline trajectory refinement provide strong priors, static traces cannot cover the causal feedback loop of real computer use: each action changes the screen state, future action space, and recovery options. EvoCUA-1.5 extends self-evolving computer-use agents from offline experience learning to online reinforcement learning, where policies interact with executable sandbox environments and improve from verifiable task outcomes. Online RL in this setting requires more than directly reusing single-turn language-RL recipes. Multi-turn interaction introduces context-managed observations, sparse terminal rewards, variable-length trajectories, and slow environment feedback. EvoCUA-1.5 addresses these challenges with Step-Level Policy Optimization (STEPO), which preserves trajectory-level advantage balance after decomposition into step-level samples; policy-aware filtering and pass-rate calibration over verifiable synthesized tasks; Dynamic Tri-Adaptive Curriculum (DTAC), which combines learnable tasks, difficult positive replay, and controlled infeasible-task exposure; and a fully asynchronous RL infrastructure with staleness control and mini-group batching. Experiments show that these components improve training stability and downstream performance. EvoCUA-1.5 achieves 63.2% success on OSWorld-Verified, outperforming comparable 32B/35B-scale open-weight baselines and even approaching models with significantly larger parameter counts. Overall, EvoCUA-1.5 provides a practical framework for scaling online RL in multi-turn computer-use agents.
[NLP-118] Coresets Before Score Sets: Evaluation-Unsupervised Prompt Subset Selection for LLM Benchmarks
【速读】： 该论文旨在解决大语言模型（LLM）评估中基准测试数据集规模过大导致的计算开销问题，提出了一种评估无监督的基准压缩（benchmark coreset selection）方法，即在不依赖模型评估结果的前提下，从多个异构基准中选择一个小型提示（prompt）子集，使其诱导出的模型得分与排名能有效逼近使用完整基准集时的结果。其解决方案的关键在于采用子模性（submodularity）优化框架，并设计基于设施选址（facility location, FL）函数的启发式方法，利用低成本的语义提示嵌入（semantic prompt embeddings）进行细粒度提示子集选择。实验表明，在包含35个异构基准、18个前沿大模型及超过6.1万个提示的大规模评测体系上，该方法在不同压缩预算下均显著优于12种基于评分或多样性设计的基线方法，且在评估有监督场景中亦可实现与当前最优方法相当甚至更优的表现，同时计算成本大幅降低。研究结果表明，子模性为基准压缩提供了一个强大而稳健的通用工具。
链接: https://arxiv.org/abs/2607.09739
作者: Jihan Yao,Gantavya Bhatt,Arnav Das,Peter Jin,Ke Bao,Qiaolin Yu,Khushi Bhardwaj,Chang Su,Jialei Wang,Yikai Zhu,Sugam Devare,Damon Mosk-Aoyama,Zhen Dong,Venkat Krishna Srinivasan,Yineng Zhang,Oleksii Kuchaiev,Jiantao Jiao,Banghua Zhu,Jeff Bilmes
机构: University of Washington, Seattle (华盛顿大学西雅图分校); University of California, Berkeley (加州大学伯克利分校); Oracle (甲骨文); Together AI (Together AI); LMSYS (LMSYS); NVIDIA (英伟达)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:We study LLM benchmark coreset selection: selecting a small subset of prompts over multiple benchmarks whose induced model scores and rankings approximate those obtained from the full benchmark suite. In evaluation-unsupervised benchmark coreset selection (our approach), the selection algorithm uses no model evaluation outcomes, and operates on a fine granularity by producing subsets of prompts over multiple benchmarks rather than producing a sub-collection of entire benchmarks. We use submodular subset selection, and we develop and evaluate many different submodular functions for this purpose, including determinantal point process (DPP) based approaches, submodular mutual information functions, and facility location-based functions. On a new large-scale suite of 35 heterogeneous benchmarks spanning five different capability categories, 18 frontier LLMs, and over 61K prompts, we find that the facility location (FL) function operating exclusively on inexpensive semantic prompt embeddings preserves LLM scores better than twelve separate score-based and diversity-based baselines, across a range of coreset budgets. Moreover, we show our proposed objective is not limited to the evaluation-unsupervised regime: in the setting where only a handful of whole benchmarks must be selected and a large amount of model scores are available, the same objective matches or outperforms state-of-the-art baselines on the MMLU and MTEB leaderboards, while being substantially cheaper to compute. Together, our results suggest that submodularity, in general, is a strong and reliable tool for benchmark compression.
[NLP-119] Interpreting Latent CoT Reasoning as Dynamical Systems ICML2026
【速读】： 该论文旨在解决生成式模型中隐式思维链（latent Chain-of-Thought, CoT）的可解释性问题，即现有方法如CODI和COCONUT在推理过程中于隐藏空间中维持多个叠加的候选推理轨迹，缺乏像显式思维链（explicit CoT）那样的单一、透明的推理路径。这一现象导致难以理解推理过程如何随步骤演化。其解决方案的关键在于将隐式推理中的词元序列建模为表示空间中的轨迹，并运用动力系统分析方法，通过定量指标（如步骤间变化量、方向一致性、Lyapunov敏感性）与定性可视化技术（如UMAP、DMD/PHATE）揭示隐式CoT所表现出的结构化、非随机的动力学特性。研究发现，隐式推理轨迹具有两类稳定态：CODI表现为稳定的吸引子行为，而COCONUT则呈现不稳定的扩张系统特征；引入SIM-CoT监督虽能收紧两种行为模式，但未改变其根本动力学机制。该框架显著提升了对隐式推理动态过程的理解，为优化隐式推理性能提供了可操作的洞察。
链接: https://arxiv.org/abs/2607.09698
作者: Sabari Iyyappan Duraipandian,Shreya Sanjay Boyane,Manju Nagesh,Jerome Francis,Archana Vaidheeswaran,Kevin Zhu
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 15 pages, ICML 2026 FoGen Workshop
Abstract:Recent latent reasoning methods, such as CODI and COCONUT, face a fundamental interpretability problem: they maintain multiple superimposed candidate traces in the hidden space at each step, unlike explicit- CoT, which follows a single transparent reasoning trace. Existing mechanistic methods show compression, shortcuts, and superposition without explaining how reasoning evolves across latent steps. To address this gap, we model latent token sequences as trajectories in representation space and apply dynamical systems analysis to characterize the evolution of reasoning. Using quantitative measures, such as step-to-step change, direction consistency, and Lyapunov sensitivity, alongside qualitative projections, such as UMAP and DMD/PHATE, we show that latent CoT exhibits structured, non-random dynamics with two distinct stability classes. CODI behaves as a stable attractor, while COCONUT behaves as an unstable expanding system, and SIM-CoT supervision tightens both behaviors without changing the underlying dynamics. This framework advances the interpretability of latent CoT reasoning dynamics and provides actionable insights for improving latent reasoning performance. Code1 and Project page2 available online.
[NLP-120] Low-Rank Attention Residuals
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）中注意力残差（Attention Residuals）机制存在的问题：现有方法将每一层的输出同时作为全维度的键（key）和值（value），导致路由（routing）与表示（representation）紧密耦合，且深度路由得分随隐藏维度 $ d $ 增大而线性增长，带来计算冗余与效率瓶颈。其解决方案的关键在于提出低秩注意力残差（Low-Rank Attention Residuals, LR-AttnRes），通过使用低维 $ r （ r \ll d $）的键进行路由，而保留全维残差值，从而实现路由与残差内容的解耦。具体而言，投影式LR-AttnRes利用已有输出投影生成可学习的低秩键，避免额外参数开销；切片式LR-AttnRes则直接采用值的最后 $ r $ 个维度作为路由键，进一步消除辅助键投影路径，降低残差侧的浮点运算量（FLOPs）。实验表明，深度路由可在远低于模型宽度的维度下仍保持高效性能，显著提升了模型效率与表达能力。
链接: https://arxiv.org/abs/2607.09694
作者: Jonathan Su
机构: Independent Researcher
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:
Abstract:Attention Residuals replace the fixed residual sum with depthwise attention over previous sub-layer outputs in large language models (LLMs), but use each output as both a full-dimensional key and value. This couples routing with representation and makes depth-routing scores scale with the hidden width d . We propose Low-Rank Attention Residuals (LR-AttnRes), which keep full-dimensional residual values while using r -dimensional keys, with r \ll d , for routing. Projected LR-AttnRes emits learned low-rank keys from existing output projections, decoupling routing from residual content and achieving the best validation loss among the variants tested. Sliced LR-AttnRes uses the last r dimensions of each value as the routing key, removing the auxiliary key-projection path and reducing residual-side FLOPs while still improving performance. Comprehensive sweeps show that depthwise routing can be effective with far fewer dimensions than the model width. We release code and models to facilitate future research.
[NLP-121] Reference-Based Distillation Detection in LLM s
【速读】： 该论文旨在解决生成式 AI 模型中模型蒸馏（model distillation）的可检测性问题，即如何识别一个模型是否由另一个更强的教师模型进行蒸馏训练而来。尽管在孤立情况下难以确定学生模型的教师来源，但研究提出在参考基准（reference-based）设置下可实现有效检测：给定一个目标模型及其同源早期版本检查点，通过对比学生模型对不同候选教师输出的偏好性对齐强度与参考检查点的差异，能够精准识别出最可能的教师模型。其核心解决方案是基于参考的成员推理（reference-based membership inference），并引入代理提示模板（proxy prompt templates）以应对未知蒸馏流程（如隐藏提示）。此外，研究还发现 o1/o3 模型特有的字形级信号（glyph-level signal）可作为额外判别特征。为克服真实模型谱系高度纠缠带来的评估难题，作者构建了融合受控蒸馏实验与真实模型的混合评估框架，在单教师蒸馏场景下实现了近乎完美的教师识别准确率。研究进一步设计了统计检验方法用于教师归属和蒸馏检测，并扩展至开放世界场景（open-world setting），即候选教师集合中未必包含真实教师。将该方法应用于当代模型，揭示了 QwQ、DeepSeek-R1 与 GPT-OSS 之间潜在的蒸馏关联性，为模型透明性与合规性提供了新证据。
链接: https://arxiv.org/abs/2607.09692
作者: Rajat Rawat,Sizhe Chen,Akshay Anand,Michael Duan,Bob Rotsted,Sewon Min
机构: University of California, Berkeley (加州大学伯克利分校); University of Southern California (南加州大学); OpenAI (OpenAI)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: 27 pages (14 main), 21 figures, 16 tables
Abstract:Model distillation – training on outputs from stronger third-party models – is widely used to boost performance, but raises concerns about unfair advantages and policy violations. This motivates a fundamental question: can we detect whether a model was distilled from another? We show that, while identifying a teacher model from a student in isolation is highly challenging, it becomes tractable in a reference-based setting: given a model and an earlier-generation checkpoint from the same lineage, we can identify the teacher model used to train the later checkpoint. We introduce a distillation detection method based on reference-based membership inference. By comparing how strongly a student model preferentially aligns with outputs from different candidate teachers relative to a reference checkpoint, our method identifies the most likely teacher and detects evidence of distillation. To handle unknown distillation pipelines such as hidden prompts, we infer proxy prompt templates directly from model outputs. We additionally identify a distinctive glyph-level signal specific to o1/o3 models. Evaluating distillation detection is challenging because modern model lineages are already heavily entangled. To address this, we develop a hybrid evaluation spanning both controlled distillation experiments and real-world models. Across both settings, our approach recovers the true teacher with near-perfect accuracy in single-teacher distillation scenarios, even when the underlying distillation pipeline is largely unknown. We further introduce statistical tests for both teacher attribution and distillation detection, and extend our framework to open-world settings where no teacher is guaranteed to be present among the candidates. Applying our method to contemporary models yields new evidence regarding potential distillation relationships involving QwQ, DeepSeek-R1, and GPT-OSS.
[NLP-122] Format Sensitivity Index: Token-Controlled Prompt Wrapper Robustness and Schema Compliance in LLM Benchmarking
【速读】： 该论文旨在解决提示词封装（prompt wrapper）在格式上的微小差异如何显著影响大模型输出性能，进而可能颠倒基准测试排名的问题。其核心挑战在于：尽管不同封装格式在语义上相近，但其对模型生成结果的准确性与可解析性存在显著影响，而这一现象常被忽视。解决方案的关键在于提出两个互补的量化指标——格式敏感度指数（Format Sensitivity Index, FSI），用于衡量因封装选择导致的准确率波动范围；以及可解析性敏感度指数（Parseability Sensitivity Index, PSI），用于评估答案可解析性随封装变化的波动。研究基于14万次OpenRouter生成实验，覆盖7个大语言模型（参数规模从7B到72B）、5类封装方式及7个问答任务，发现模型间平均FSI差异超过30倍，且主要由模型合规性失败（compliance failure）驱动。固定效应回归进一步表明，即使控制任务、模型和封装因素，可解析性仍是准确性的强预测因子。因此，论文强调在基准测试与结构化输出部署中必须报告封装相关的变异性与合规性表现，否则结果具有统计脆弱性，并提出了相应的实践建议以提升评估的可靠性。
链接: https://arxiv.org/abs/2607.09665
作者: Deep Pankajbhai Mehta
机构: Adobe Inc(Adobe公司)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 10 pages, 6 figures
Abstract:Prompt wrappers often differ only in formatting, yet they can change model scores enough to flip leaderboard conclusions. We study this variance under a token-controlled protocol and introduce two complementary metrics: the Format Sensitivity Index (FSI), the accuracy range induced by wrapper choice, and the Parseability Sensitivity Index (PSI), the corresponding range in answer parseability. Across 140,000 OpenRouter generations spanning 7 QA tasks, 5 wrapper families, and 4 instruct models from 7B to 72B parameters, we find that mean FSI varies by over 30x across models and is largely explained by compliance failures. A fixed-effects regression shows that parseability remains a strong predictor of accuracy even after controlling for task, model, and wrapper. We argue that reporting accuracy without wrapper variance and compliance is statistically fragile, and we give practical recommendations for both benchmarking and structured-output deployments.
[NLP-123] Question Type Cognitive Load and CEFR Alignment: Evaluating LLM -Generated EFL Grammar Drill Exercises
【速读】： 该论文旨在解决生成式人工智能（Generative AI）在英语作为外语（EFL）教学内容生成中的教学可行性问题，尤其关注不同题型模态对学习者表现的影响及其与语言能力水平预测模型的匹配度。其解决方案的关键在于通过真实日语初中生在语法练习应用中的日志数据，系统评估多种题型（多项选择、填空、拖拽）的认知负荷与任务难度，并验证本地化CEFR-J语法框架在预测实际任务难度方面的有效性。研究发现，题型设计显著影响学习效率：多项选择题认知负担最低，填空题最不利于主动回忆，而拖拽题则带来最长响应时间；同时，实证数据支持了CEFR-J框架的有效性，显示随着语言水平提升，准确率下降且反应时间延长。因此，研究强调生成式AI可有效生成高质量学习内容，但开发人员需基于认知负荷理论，合理规划题型序列，以引导学习者从被动识别逐步过渡到主动语言产出。
链接: https://arxiv.org/abs/2606.01592
作者: Steve Woollaston,Brendan Flanagan,Yuko Toyokawa,Hiroaki Ogata
机构: 未知
类目: Computers and Society (cs.CY); Computation and Language (cs.CL)
备注: Under review for the the 34th International Conference on Computers in Education (ICCE 2026). 2jun26: v2 - fixed minor typo
Abstract:This study evaluates the pedagogical viability of LLM-generated English as a Foreign Language (EFL) learning content. Utilising log data from Japanese junior high school students practicing on a grammar drilling application, we analysed how different question modalities impact student performance and whether theoretical localised CEFR difficulty tiers accurately predict empirical task difficulty. Results reveal a clear performance hierarchy: multiple-choice questions carried the lowest cognitive load, cloze tasks posed the greatest barrier to active recall, and drag-and-drop exercises incurred the heaviest time penalties. Furthermore, learner data validated the CEFR-J grammar framework, showing a steady decline in accuracy and increased response times as proficiency levels advanced. These findings demonstrate that LLMs can successfully generate learning content, while highlighting the need for developers to strategically sequence question modalities to transition learners from passive recognition to active linguistic production.
[NLP-124] GigaChat Audio: Time-aware Large Audio Language Model INTERSPEECH2026
【速读】： 该论文旨在解决长时录音中音频条件下的大语言模型（LLM）在时间定位（temporal grounding）任务上的挑战，即如何在长达120分钟的音频输入中准确回答涉及具体时间戳的问题。其核心解决方案是提出一种具备时间感知能力的音频大语言模型，通过在连续音频标记（audio tokens）之间交错插入周期性时间标记（time markers），并利用级联式流水线生成的大规模合成监督信号进行训练。该方法的关键在于时间标记的周期性插入策略与音频-时间序列的协同建模机制，使模型能够同时理解上下文语义与精确的时间位置信息。实验表明，该模型在短时与长时基准测试中均表现出优异的时间定位精度，并支持基于时间锚点的片段描述与摘要生成。消融实验进一步揭示了时间表示方式、标记频率、分词策略及时长混合设计对模型性能与计算开销的影响。研究团队已公开模型权重与数据集，以推动时间感知音频理解领域的进一步研究。
链接: https://arxiv.org/abs/2607.10387
作者: Aleksandr Kutsakov,Mariia Sadovina,Georgii Gospodinov,Alexandr Maximenko,Oleg Kutuzov,Pavel Bogomolov,Fyodor Minkin
机构: SaluteDevices(俄罗斯)
类目: Audio and Speech Processing (eess.AS); Computation and Language (cs.CL)
备注: Accepted to Interspeech 2026. Model and dataset: this https URL
Abstract:Temporal grounding in long recordings remains challenging for audio-conditioned LLMs. We present a time-aware audio LLM that answers questions with explicit timestamps over up to 120 minutes of input. Our approach interleaves periodic time markers with continuous audio tokens using large-scale synthetic supervision from a cascaded pipeline. Our model achieves strong temporal-grounding accuracy on short and long benchmarks and supports time-anchored fragment descriptions and summaries. Extensive ablations examine how time representation, marker frequency, tokenization, and duration-mixture design affect accuracy and computational cost. We release model weights and datasets to support further research on time-aware audio understanding, available at this https URL.
[NLP-125] GigaAM Multilingual: Foundation Model for Underrepresented Languages INTERSPEECH2026
【速读】： 该论文旨在解决多语言自动语音识别（ASR）中长尾语言因数据稀缺而导致性能显著下降的问题，尤其针对代表性不足的中亚语言（如哈萨克语、吉尔吉斯语、乌兹别克语）。其核心解决方案在于提出一种基于聚类级别的数据平衡策略，在预训练阶段缓解主流语言（头语言）主导现象；同时在微调阶段引入领域感知采样方法，进一步优化对低资源语言的建模能力。实验表明，该方法在控制条件下显著优于现有开源预训练模型（如Whisper Large v3、Omnilingual-1B），尤其在非正式口语场景下取得明显提升，且保持高效性。研究公开了基础编码器与ASR模型，为实际数据不平衡条件下的多语言模型适配提供了可复现的有效范式。
链接: https://arxiv.org/abs/2607.10371
作者: Andrei Kuzmenko,Alexandr Maximenko,Aleksandr Kutsakov,Georgii Gospodinov,Dmitrii Bolotov,Oleg Kutuzov,Pavel Bogomolov,Fyodor Minkin
机构: SaluteDevices(萨尔特设备), Russia(俄罗斯)
类目: Audio and Speech Processing (eess.AS); Computation and Language (cs.CL)
备注: Accepted to Interspeech 2026. Model weights: this https URL
Abstract:Despite recent scaling successes, multilingual ASR performance remains highly uneven, with long-tail languages suffering from severe data scarcity. This work addresses the challenge of building robust foundation models for underrepresented Central Asian languages (Kazakh, Kyrgyz, Uzbek). We present GigaAM Multilingual, a Conformer encoder pre-trained on 2M hours of audio using a HuBERT-style objective. Crucially, we introduce a cluster-level data balancing strategy during pre-training and a domain-aware sampling method during fine-tuning to mitigate head-language dominance. In controlled comparisons, our approach outperforms strong open pretrained encoders (Whisper Large v3, Omnilingual-1B) on target languages, achieving significant gains on spontaneous speech while maintaining efficiency. We release the foundation encoder and ASR model, offering a proven recipe for effective multilingual adaptation under realistic data imbalance.
[NLP-126] Hearing Like Humans? Sound Symbolism and Perceptual Alignment in Speech Language Models
【速读】： 该论文旨在解决生成式语音语言模型（Speech Language Models, SLMs）是否具备人类特有的声音象征性（sound symbolism）倾向的问题，即语音音素与感知属性（如圆形或尖锐感）之间的映射关系。以往研究多依赖文本或图像评估，缺乏对真实语音数据的检验。本文采用真实的语音录音，从听觉、跨模态和视觉三个维度对比模型判断与人类感知的一致性。研究发现，SLMs在听觉层面的判断与人类感知存在显著偏差，未能捕捉到驱动人类直觉的关键声学线索（如频谱倾斜度），且开放权重模型也无法可靠地将所听声音与其对应形状关联。通过仅保留视觉信息的对照实验排除了形状感知干扰后，问题根源被定位为语音表征本身缺陷，表明感知一致性并非源于更强的视觉能力，而取决于语音表示能否准确捕获人类实际感知到的声学特征。
链接: https://arxiv.org/abs/2607.10162
作者: Yun-Shao Tsai,Chun-Wei Chen,Chee-En Yu,Yi-Cheng Lin,Hung-yi Lee
机构: 未知
类目: Audio and Speech Processing (eess.AS); Computation and Language (cs.CL)
备注: Submitted to SLT 2026
Abstract:Sound symbolism, the human tendency to map speech sounds to perceptual qualities such as roundness or sharpness, arises primarily from the acoustics of speech rather than spelling. Whether Speech Language Models (SLMs) share this tendency remains open, as prior evaluations rely on text or images rather than real speech. We study it using genuine human speech recordings, comparing model judgments against human data across the auditory, crossmodal, and visual components of the effect. We find that SLMs’ auditory judgments align poorly with human perception and miss the acoustic cues, such as spectral tilt, that drive human intuitions, and open-weight models cannot reliably link a heard sound to its corresponding shape. With a visual-only control ruling out shape perception, the weakness localizes to how speech is represented, suggesting that perceptual alignment depends not on stronger vision but on speech representations that capture the cues humans hear.
信息检索
[IR-0] PaperRouter-Agent : A Content-Grounded LLM Agent for Personalized Hierarchical Paper Routing
链接: https://arxiv.org/abs/2607.11564
作者: Keshen Zhou,Lintao Wang,Suqin Yuan,Zhuqiang Lu,Yu Luo,Zhiyong Wang
类目: Computation and Language (cs.CL); Human-Computer Interaction (cs.HC); Information Retrieval (cs.IR)
备注:
Abstract:Researchers organize the papers they collect into personal folder hierarchies in reference managers, and route each new paper into the folder where it belongs. This task differs from standard hierarchical text classification. A user’s folder hierarchy is not a fixed, shared taxonomy but a private and evolving folksonomy whose folder meanings may be topical, shorthand, venue-based, or process-oriented, and are often defined by the papers already stored inside them. We formalize this setting as personalized hierarchical paper routing (PHPR): assigning an incoming paper to folders in a user-specific hierarchy without per-user training. We propose PaperRouter-Agent, a training-free LLM agent that grounds routing decisions in folder members rather than folder names alone. The agent first narrows the candidate hierarchy, retrieves folder-specific evidence, verifies fit by inspecting member papers, and incorporates similarity-gated feedback from past user rejections. A formative study on real personal libraries shows that PaperRouter-Agent raises overall Recall@1 from 0.39 to 0.61 and Recall@3 from 0.57 to 0.83, with the largest gains on organizational folders defined by metadata such as venue or year, where single-shot methods collapses (Recall@1 0.09 to 0.50). On the public LaMP-2 benchmark, the same approach improves accuracy from 44.5% to 51.5% (+9.0 macro-F1) over a single-shot baseline, while remaining low-cost for practical use.
[IR-1] Score-Only Distillation for Compact Dense Retrieval
链接: https://arxiv.org/abs/2607.11465
作者: Kirill Dubovikov,Martin Takac,Salem Lahlou
类目: Information Retrieval (cs.IR)
备注:
Abstract:Large embedding models improve retrieval quality, but serving large encoders online is expensive. We study whether a compact retriever can learn teacher ranking behavior from score vectors without access to teacher hidden states. The student trains on rows built from ground-truth positives and negative candidates produced by our data generation pipeline; we evaluate student-teacher hard-negative mining separately as an extension. We use a row-centered score-vector objective, a memory-efficient implementation of uniform all-pairs PairMSE loss. On a fixed eight-task evaluation panel, our distillation protocol recovers up to 50% of the base-to-teacher gap. The distilled 0.6B student is 4.7 \times faster for query encoding and 9.7 \times faster for document encoding than sequential online teacher fusion. External-transfer performance after distillation remains mixed, so our evidence supports compression of teacher rankings under matched retrieval protocols.
[IR-2] FAIR GraphRAG : A Retrieval-Augmented Generation Approach for Semantic Data Analysis
链接: https://arxiv.org/abs/2607.11464
作者: Marlena Flüh,Soo-Yon Kim,Carolin Victoria Schneider,Sandra Geisler
类目: Information Retrieval (cs.IR)
备注: Accepted at the IEEE International Conference on Knowledge Graph, 2025. Corrects an error in the published abstract: the evaluation dataset is RNA-sequencing data, not single-cell data
Abstract:Retrieval-Augmented Generation (RAG) addresses the limitations of Large Language Models (LLMs) when providing responses to domain-specific questions. Graph-based RAG approaches, such as GraphRAG, enhance retrieval by capturing semantic relationships within knowledge graphs (KGs). While the FAIR principles (Findability, Accessibility, Interoperability, and Reusability) are becoming prevalent for scientific data management, especially in complex domains such as medicine, existing RAG approaches lack a structured FAIRification of the underlying knowledge resources. This lack limits their potential for FAIR information retrieval in these domains. To address this gap, we introduce FAIR GraphRAG, a novel framework that integrates FAIR Digital Objects (FDOs) as the fundamental units of a graph-based retrieval system. Each graph node represents an FDO that incorporates core data, metadata, persistent identifiers, and semantic links. We leverage LLMs to support schema construction and automated extraction of content and metadata from data sources. The framework was co-designed by physicians and computer scientists to ensure technical and clinical relevance. We apply FAIR GraphRAG to a biomedical dataset in gastroenterology, demonstrating its applicability to RNA-sequencing data. Beyond ensuring adherence to the FAIR principles, FAIR GraphRAG significantly improves question answering accuracy, coverage, and explainability, particularly for complex queries involving metadata and ontology links. This work shows the feasibility of combining FAIR data practices with graph-based retrieval techniques. We see potential for applying our approach to other specialized fields such as education and business.
[IR-3] Beyond Semantic IDs: Encoding Business-Value Ranking into Document Identifiers for Generative Retrieval
链接: https://arxiv.org/abs/2607.11392
作者: Gui Ling,Zhihong Chen,Yu Li,Tong Xiong,Kunhai Lin,Kaixuan Zhang,Yuliang Yan,Dan Ou,Haihong Tang,Bo Zheng
类目: Information Retrieval (cs.IR)
备注:
Abstract:Generative Retrieval (GR) formulates retrieval as a sequence-to-sequence generation task, assigning each document a document identifier (DocID) and retrieving it through autoregressive decoding, making DocID design a critical factor in retrieval quality. However, existing schemes based on discrete representation learning suffer from inherent collision issues and create a mismatch between the DocID’s encoding objective and the system’s business optimization target. To address these limitations, we propose Cluster-Ranked Identifier (CRID), which decouples DocID into semantic clustering and business-value ranking, yielding collision-free identifiers that support incremental updates via intra-cluster reranking. We further introduce an analytical framework that decomposes retrieval gains into personalized preference and statistical prior generalization, revealing how semantic cluster size governs the balance between the two components. Experiments on a 300M-item Taobao e-commerce corpus show that CRID surpasses the strongest embedding-based retrieval baseline on top-K Hitrate, and delivers +1.06% GMV in full-traffic deployment.
[IR-4] Boolean queries are all you need?
链接: https://arxiv.org/abs/2607.11362
作者: Charles L. A. Clarke,Mark D. Smucker
类目: Information Retrieval (cs.IR)
备注:
Abstract:We equipped an LLM-based search agent with access to a Boolean retrieval engine to search the MS MARCO V2.1 deduped segment collection used by the TREC 2024 RAG track. Over a standard track subset of 86 topics, and operating under a budget of 100 model calls/topic, the agent achieved an NDCG@10 of 0.6863, which would place it above many dense, sparse, and learned-sparse first-stage retrievers. Ranking is based solely on the density of corpus substrings matching a query, with no requirement for supervised learning, global statistics, or term weights. Formally, the query language expresses a strict subset of the regular languages, with a document’s score based on the number and length of matches it contains. Although the results are more exploratory than definitive, because they are based on a single test collection that was publicly available during model training, they suggest that simple pattern matching may be sufficient for agentic search.
[IR-5] User Preference Induction with LLM s for Offline Top-N Recommendation Evaluation
链接: https://arxiv.org/abs/2607.11354
作者: David Otero,Javier Parapar
类目: Information Retrieval (cs.IR)
备注:
Abstract:Offline evaluation is the standard methodology for comparing top-N recommender systems, yet it relies on incomplete relevance information. In most benchmark datasets, only a small subset of user–item preferences is observed, and unjudged items are commonly treated as non-relevant. This missing-as-negative assumption can bias evaluation, penalize plausible recommendations with no recorded feedback, and favour algorithms that concentrate on popular or highly exposed items. We propose an LLM-based framework to expand relevance judgements for offline recommender evaluation. Our approach uses large language models in two complementary roles. First, a preference induction stage summarizes each user’s historical interactions into a textual profile that captures their tastes and interests. Second, conditioned on this profile, an LLM acts as a relevance judge for candidate recommended items that lack observed labels in the original test data. To make this process tractable and evaluation-focused, we apply judgement expansion to a pooled candidate set built from the top-ranked outputs of multiple recommenders. The resulting enriched judgements provide additional relevance evidence for previously unobserved user–item pairs, enabling ranking metrics to be computed on a more complete basis. Experimental results show that this approach is a promising strategy for improving the robustness of offline top-N evaluation and mitigating the popularity-sensitive distortions caused by sparse feedback.
[IR-6] Characterising AI Models for Cataloguing
链接: https://arxiv.org/abs/2607.11353
作者: Miguel Arana-Catania,Neil Jefferies
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Digital Libraries (cs.DL); Information Retrieval (cs.IR); Machine Learning (cs.LG)
备注: 7 pages, 10 tables
Abstract:The creation of digital collections involves not only the digitisation of content, but also the creation of catalogue records for it. This often-overlooked task requires slow and costly expert manual work. In this project, we have evaluated the application of AI models to this task, comparing different implementations and models. This work includes a qualitative and quantitative evaluation of the experiments carried out, as well as recommendations on the use of AI models that go beyond the specific use case.
[IR-7] Prompt Generation Technical Report
链接: https://arxiv.org/abs/2607.11326
作者: Dan Ou,Gui Ling,Hao Wan,Hongbin Zhou,Jialiang Cheng,Jiangnan Pang,Silu Zhou,Wei Shi,Weichen Ye,Wenming Zhang,Yang Wang,Yu Li,Yuliang Yan,Zhan Fa,Zhihong Chen,Zongyuan Wu,Bo Zheng,Changfa Wu,Dunxian Huang,Haihong Tang,Jinlong Guo,Kaixuan Zhang,Kun Ma,Lin Qu,Longbo Zhong,Tao Lan,Tong Xiong,Zhibo Wu
类目: Information Retrieval (cs.IR)
备注:
Abstract:Generative retrieval has become an increasingly adopted paradigm for industrial search, recommendation, and advertising systems, delivering significant online gains. Most existing work combines user behavior sequences with large language models (LLMs) to model user preferences. In practice, feature engineering remains critical to model effectiveness, yet its complexity slows offline iteration and makes online deployment heavy and hard to reuse, all under tight online latency budgets. The root cause is a tight coupling between feature-processing logic and model architecture, where every feature change touches the training and serving code and resists reuse across scenarios. To break this coupling, we present Prompt Generation (PG), a high-level tokenizer and configuration-driven framework that decouples feature-processing logic from model architecture through two declarative JSON files, which serve as the single source of truth for both offline training and online serving, ensuring feature consistency across the two stages. Organizing features under four types with three composable processing components to assemble and compress heterogeneous features, PG delivers acceleration at three levels: (1)fast training iteration: feature experiments require only configuration changes, with built-in token compression for ultra-long sequences; (2)fast deployment: a new scenario only needs to conform to the PG schema and plug into a universal pipeline, with no scenario-specific engineering; (3)fast online inference: engine applies unified optimizations over the standardized configuration, reducing PG’s overhead to a negligible level. PG has been deployed on Taobao Search with statistically significant online A/B uplifts of +0.47% in transaction count and +0.51% in GMV, and has been applied across multiple Taobao search and recommendation teams as the iteration framework for generative retrieval.
[IR-8] Enhancing LLM s through human feedback: a journey towards self-improvement ECAI2025
链接: https://arxiv.org/abs/2607.11267
作者: Tatiana Pelc,Gila Kamhi,Asaf Avrahamy,Adi Fledel-Alon
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: AIC 2025: The 10th International Workshop on Artificial Intelligence and Cognition (held as part of ECAI 2025). October 25-26, 2025. Bologna, Italy
Abstract:In the rapidly evolving landscape of information retrieval systems, the ability to adapt and improve through user feedback is paramount. This study introduces a novel methodology for refining the performance of a primary Retrieval Augmented Generation (RAG) system by strategically integrating an auxiliary feedback RAG system. By systematically harnessing human-generated feedback, the approach aims to enhance the accuracy, relevance, and overall quality of responses, driving the system towards self-improvement. Central to this methodology is a human-in-the-loop implementation, where user feedback is continuously collected, classified, and integrated into the inference workflow, enabling the system to learn and evolve iteratively. To validate the effectiveness of this approach, the study employs rigorous testing against three diverse benchmark datasets focused on general and custom domain knowledge, utilizing a LLM-as-a-Judge evaluation strategy. This comprehensive framework not only underscores the transformative potential of feedback-driven enhancements in RAG systems but also sets a precedent for future research in adaptive information retrieval technologies, marking a significant step in the journey towards autonomous refinement and optimization through user engagement.
[IR-9] NGM-RAG : Neural Graph Matching based Retrieval-Augmented Generation
链接: https://arxiv.org/abs/2607.11159
作者: Guo Chen,Ziwen Li,Maolin Zheng,Hao Gao,Junjie Huang,Tao Jia
类目: Information Retrieval (cs.IR)
备注:
Abstract:Retrieval-Augmented Generation (RAG) significantly enhances the ability of Large Language Models (LLMs) to provide accurate and contextually relevant answers by dynamically integrating external databases. However, traditional RAG methods are primarily constrained by their reliance on text-based retrieval strategies, which often struggle with complex questions requiring multi-hop reasoning. To address this limitation, we introduce Neural Graph Matching based Retrieval-Augmented Generation (NGM-RAG), a novel framework that leverages graph structures to effectively capture and utilize relational knowledge for improved retrieval and answer generation. NGM-RAG explicitly incorporates graph construction, graph matching, and answer generation into a unified process. Within this framework, we propose a neural graph matching approach that combines text-based matching with Graph Neural Networks (GNNs). By employing an adaptive weighting strategy, NGM-RAG efficiently integrates multiple matching methods to select the most relevant contextual node information for answer generation. Experimental results on multi-hop question answering and long-context summarization tasks demonstrate that our NGM-RAG model achieves superior performance compared to both traditional NaiveRAG methods and state-of-the-art graph-enhanced approaches such as GraphRAG and LightRAG.
[IR-10] Generative Chinese Statute Retrieval
链接: https://arxiv.org/abs/2607.11109
作者: Yiteng Tu,Zitao Su,Weihang Su,Xuanyi Chen,Yueyue Wu,Yiqun Liu,Min Zhang,Qingyao Ai
类目: Information Retrieval (cs.IR); Computation and Language (cs.CL)
备注:
Abstract:Statute retrieval is a fundamental task in legal information retrieval, yet existing approaches struggle to bridge the gap between colloquial legal queries and formal statutory language. In this paper, we propose GCSR, a generative statute retrieval framework that reformulates statute retrieval as a sequence generation problem and internalizes statutory knowledge into a generative model. Specifically, we propose a multi-granularity structured docid that encodes legal hierarchy and semantic information, together with a multi-task training strategy. Experiments show that GCSR consistently outperforms strong sparse, dense, and legal-domain baselines. Our results demonstrate the effectiveness of generative retrieval for statute retrieval and highlight its potential for broader legal information access and downstream legal reasoning tasks.
[IR-11] MMRM: A Multiplex Multimodal Representation Model for Product Ranking in E-commerce Search SIGIR2026
链接: https://arxiv.org/abs/2607.11030
作者: Zhen-Lin Chen,Maosen Sheng,Peng Lin,Jianmin Chen,Zhuojian Xiao,Dongyue Wang,Xiwei Zhao
类目: Information Retrieval (cs.IR); Machine Learning (cs.LG); Multimedia (cs.MM)
备注: Accepted by SIGIR2026
Abstract:Multimodal information is pivotal for e-commerce search ranking. Existing works leverage multimodal data typically by fine-tuning general Multimodal Large Language Models (MLLMs) via collaborative signals, subsequently integrating the derived representations into ranking models as item features. Despite their efficacy, these methods face two primary limitations: (1) they rely on a single collaborative signal for MLLM fine-tuning, failing to exploit the heterogeneous signals essential for multitask ranking; and (2) they treat multimodal representations as regular item features in ranking models, underutilizing their latent potential for user behavior modeling. To address these challenges, we propose the Multiplex Multimodal Representation Model (MMRM), a unified framework that aligns MLLMs with diverse collaborative signals. By employing a shared backbone with task-specific tokens and projection layers, MMRM simultaneously learns from multiple signals and generates comprehensive multiplex item representations in a single inference pass. Furthermore, we introduce a multiplex user representation strategy in ranking models, which derives task-specific user representations via search-based behavior sequence modeling leveraging multiplex item representations. Extensive experiments demonstrate MMRM’s superior efficiency and effectiveness. Notably, MMRM has been successfully deployed in the JD e-commerce search engine, yielding significant performance gains for millions of daily users.
[IR-12] Normative Alignment of Recommender Systems via Internal Label Shift KR RECSYS’25
链接: https://arxiv.org/abs/2607.10915
作者: Johannes Kruse,Kasper Lindskow,Michael Riis Andersen,Ryotaro Shimizu,Julian McAuley,Pierre-Alexandre Mattei,Jes Frellsen
类目: Information Retrieval (cs.IR); Machine Learning (cs.LG)
备注: 6 pages. Published in the Proceedings of the Nineteenth ACM Conference on Recommender Systems (RecSys '25), Prague, Czech Republic, September 22-26, 2025. Code available at this https URL
Abstract:We introduce NAILS (Normative Alignment of Recommender Systems via Internal Label Shift), a simple and scalable method for aligning recommendation outputs with target distributions over item-level attributes, such as categories. Recommender systems optimized solely for user engagement often fail to satisfy broader normative objectives, including fairness, diversity, and editorial values. NAILS modifies the user-conditional item distribution to induce a specified marginal distribution over attributes while preserving the preferences learned by an existing recommender system and requiring no model retraining. We formulate this problem as a form of label shift applied internally within a hierarchical classification framework. By adopting a stakeholder-centric perspective, NAILS enables recommendation outputs to be aligned with global normative objectives. Empirically, we show that NAILS consistently improves attribute-level alignment with minimal impact on user engagement, providing a practical mechanism for value-driven recommendation.
[IR-13] ZoRRO: A Zero-Weight Personalized Recommender System for Scalable News Recommendation KR SIGIR’26 SIGIR
链接: https://arxiv.org/abs/2607.10910
作者: Johannes Kruse,Ryotaro Shimizu,Kasper Lindskow,Jon Tofteskov,Michael Riis Andersen,Julian McAuley,Jes Frellsen
类目: Information Retrieval (cs.IR); Machine Learning (cs.LG)
备注: 6 pages, 2 figures. Accepted at the 49th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR '26), Melbourne, Australia, July 20-24, 2026. Code available at this https URL
Abstract:We present ZoRRO (Zero-Weight Personalized Recommender System), a zero-weight, training-free framework for personalized news recommendation designed for scalable real-world deployment. ZoRRO outperforms strong neural baselines in offline ranking evaluations and achieves click-through rate performance in online A/B testing that is nearly on par with a state-of-the-art deep learning model, while operating more than 600 times faster. Our experiments reveal gaps between offline and online performance and demonstrate that models with similar click-through rate outcomes can produce markedly different recommendation distributions, thereby influencing the overall news flow. These findings position ZoRRO as a practical and efficient solution for large-scale news recommendation and highlight the importance of evaluating recommender systems using metrics beyond accuracy alone.
[IR-14] Stream-aware Side Adaptation for Large Pre-trained Multimodal Embedding Models in Sequential Recommendation ACM-MM2026
链接: https://arxiv.org/abs/2607.10909
作者: Junchen Fu,Kaiwen Zheng,Ioannis Arapakis,Wenhao Deng,Xin Xin,Joemon M. Jose,Xuri Ge
类目: Information Retrieval (cs.IR)
备注: Accepted by ACM MM2026
Abstract:Recently, large pretrained multimodal embedding models such as Qwen3-VL Embedding have shown strong promise for sequential recommendation, as they provide reusable semantic item representations across modalities and domains. However, directly using these embeddings often leads to suboptimal performance because of domain misalignment. Efficient side adaptation is therefore an attractive solution. Although adapting all backbone layers should help, existing side adapters often degrade with depth, prompting layer dropping despite the loss of useful hidden states. This is due to two major challenges: (1) the lack of modeling in selecting fused representations during residual addition, and (2) the insufficient preservation of earlier representations during progressive sigmoid fusion. This paper therefore asks a practical question: How can we design a side adaptation approach that effectively unlocks the potential of large pre-trained multimodal embedding models? To address this question, we propose Stresa, a stream-aware side-adaptation framework for frozen large pre-trained multimodal embedding models in sequential recommendation. Stresa introduces Stream-aware Hidden-Adapter Fusion (SHAF) to preserve historical side memory during fusion and Residual Stream Adapter (ReSA) to produce selective residual updates across layers. Empirically, Stresa consistently outperforms standard side adapters and state-of-the-art baselines on public datasets across multiple backbone embedding models. These results highlight the promise of adapting large embedding models for sequential recommendation. Our code is publicly available at this https URL. Comments: Accepted by ACM MM2026 Subjects: Information Retrieval (cs.IR) Cite as: arXiv:2607.10909 [cs.IR] (or arXiv:2607.10909v1 [cs.IR] for this version) https://doi.org/10.48550/arXiv.2607.10909 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[IR-15] ool-Adaptive LLM Reranker
链接: https://arxiv.org/abs/2607.10555
作者: Zichuan Liu,Ruijin Hua
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 12 pages, 7 figures, 6 tables
Abstract:Generative Large Language Models (LLMs) have revolutionized information retrieval, yet their strictly parametric nature frequently leads to severe factual hallucinations when confronted with complex queries beyond their epistemic boundaries. While external tool-calling can mitigate this, indiscriminately invoking search tools for every document during reranking incurs prohibitive latency overheads, creating an intractable accuracy-efficiency dilemma. To address this challenge, we propose TALRanker, a novel framework that formalizes pointwise relevance scoring as an agentic Markov decision process. We optimize it via a two-stage training paradigm. An initial warm-up utilizes a language-preserving hybrid loss to prevent the catastrophic forgetting of native generative capacities. Subsequently, an asymmetric cost-aware reward equipped in reinforcement learning forces the policy to autonomously bypass tools for maximum efficiency when confident, while selectively retrieving external evidence to avert severe hallucination penalties when uncertain. Extensive evaluations demonstrate that TALRanker achieves state-of-the-art performance across standard and reasoning-intensive retrieval benchmarks, matching throughput with pointwise rerankers while outperforming parameter-heavy reasoning models.
[IR-16] RecRec: Recursive Refinement for Sequential Recommendation
链接: https://arxiv.org/abs/2607.10541
作者: Pervez Shaik,Prosenjit Biswas,Abhinav Thorat,Ravi Kolla,Niranjan Pedanekar
类目: Information Retrieval (cs.IR); Machine Learning (cs.LG)
备注: 8 pages, 3 figures
Abstract:Sequential recommender systems typically infer user preferences through single-pass encoding of interaction histories without iterative refinement, relying on increasingly deep architectures to capture complex patterns. In this work, we revisit sequential recommendation from a recursive inference perspective: can user preferences be modeled as a persistent latent state that is recursively refined? We propose RecRec (Recursive Recommendation), a lightweight model that maintains a compact latent state and updates it through a shared recursive module conditioned on interaction evidence. Unlike prior recursive models, RecRec introduces an evidence-anchored correction mechanism that stabilizes refinement by grounding each update in the original interaction context, preventing semantic drift during deep recursive reasoning. Experiments on three benchmark datasets under standard evaluation protocols show that RecRec matches or outperforms state-of-the-art sequential, graph-based, and reasoning-enhanced recommenders while using only 3.9M to 14M parameters. Ablation studies demonstrate that both recursive refinement and the evidence-anchored correction gate contribute significantly to performance, highlighting the effectiveness of recursive latent inference as a scalable alternative to deeper or language-based architectures. Code is available at this https URL.
[IR-17] Implicit Fine-tuning via Context Engineering: A Curriculum Learning Framework for Multimodal Entity Alignment KDD2026
链接: https://arxiv.org/abs/2607.10532
作者: Yunpeng Hong,Chenyang Bu,Di Wu,Yi He,Xindong Wu
类目: Information Retrieval (cs.IR)
备注: Accepted by KDD 2026
Abstract:Multimodal Entity Alignment (MMEA) aims to identify equivalent entities across different modalities. While existing methods enhance MMEA performance through black-box context engineering strategies, their reliance on LLM parameter capacity and lack of theoretical interpretability remain unresolved. To this end, we first theoretically validate the mathematical equivalence between context engineering and model fine-tuning in MMEA tasks, demonstrating that prompt components simulate contrastive learning-based sequential fine-tuning in MMEA. Building on this foundation, we then propose PTFEA, a curriculum-learning-inspired framework that translates fine-tuning strategies into interpretable context engineering. Specifically, adaptive difficulty modulation dynamically adjusts information injection stages using confidence thresholds, establishing mathematical equivalence between curriculum learning weights and context sample selection; and three-stage progressive inference incorporates entity information from simple to complex cases, mirroring the gradient descent process in fine-tuning. Experiments on five public datasets demonstrate that PTFEA consistently outperforms strong baselines. In particular, on the ICWIKI dataset, PTFEA narrows the H@1 gap between Qwen2.5-72B and 14B to 0.6%. Moreover, compared with the representative context-engineering-based MMEA method MM-ChatAlign, PTFEA reduces the runtime of Qwen2.5-72B from 21 hours to 1 hour and lowers token consumption from 2200-3000 to 200-400, achieving over 80% reduction on the ICWIKI dataset. This work provides the first theoretical framework unifying context engineering and fine-tuning in MMEA, paving the way for future research that seeks to translate additional fine-tuning strategies into context engineering paradigms. Our code is available at this https URL.
[IR-18] GRASP: GRanularity-Aware Search Policy for Agent ic RAG
链接: https://arxiv.org/abs/2607.10463
作者: Varun Gandhi,Jaewook Lee,Shantanu Todmal,Franck Dernoncourt,Ryan Rossi,Zichao Wang,Andrew Lan
类目: Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
备注:
Abstract:Agentic retrieval-augmented generation (RAG) extends static RAG by allowing language models to iteratively reason, generate search queries, retrieve evidence, and predict answers. However, it remains challenging for models to decide when to retrieve, whether to use lexical matching or semantic similarity, and how to control context granularity to prevent irrelevant tokens from interfering with agent reasoning. In this paper, we introduce GRASP, a reinforcement learning (RL) framework for training agents to adaptively coordinate complementary retrieval tools during multi-step reasoning. GRASP provides the agent with semantic search, keyword search, and paragraph-reading actions, enabling it to retrieve sentence-level evidence and expand further context only when needed. We train the policy with a reward that jointly accounts for answer accuracy, grounded reading, complementary search, and turn efficiency. Experiments on multi-hop reasoning benchmarks show that GRASP improves both retrieval recall and downstream question answering performance compared with single-step retrieval, prompting-based agentic RAG, and RL-based retrieval baselines. Qualitative and ablation analyses show that the learned policy develops interpretable skimming and scanning behavior: it uses semantic search for broad exploration, paragraph reading for local verification, and keyword search for entity-specific evidence. These results suggest that learning to coordinate retrieval signals and context granularity is critical for agent’s correct reasoning.
[IR-19] Context by Distinct Information: An Auditable Dirichlet-Process Working Memory for Long Redundant Context Streams
链接: https://arxiv.org/abs/2607.10441
作者: Siddharth Pal,Viktoria Rojkova
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注: 16 pages, 1 figure
Abstract:Context engineering decides what information a model carries forward, and current designs meter it in tokens: compressing the past into a bounded recurrent state, keeping a key-value entry for every token, or imposing a fixed budget through a window or eviction rule. All three make the token the unit of memory even when the stream is redundant and the task depends on the distinct information it carries. Building on a companion mechanism paper that opens a cache slot only when an incoming key is novel, so memory scales with the number of distinct items rather than tokens, we develop that allocate-on-novelty cache as a working-memory component and organize context by how a task depends on the past: recall-carried information belongs in a content-addressed novelty cache, summary-carried information in a recurrent state, and locality-carried information in a recency window. The claim is empirical and bounded. On a matched character-level control, novelty-gated attention reaches full-attention performance while attending to about half the tokens, and coupling the cache with a state-space summary matches full-attention coupling at that reduced cost; the advantage grows as context lengthens, while a sliding window is preferable on short, locality-dominated spans. On next-code prediction over synthetic Medicare claims the coupled component leads full attention and every fixed-budget eviction policy at a thousand-event horizon, whereas cost forecasting over the same stream is summary-carried and the cache is neutral. The retained memory is an inspectable table of templates, codes, drugs, or places rather than an opaque state. The experiments are small-scale and use only public data; they establish the primitive that context can scale with distinct information rather than tokens, in a working memory that is content-addressable and auditable.
[IR-20] SVD-RAG : Efficient Tree-Organized Retrieval-Augmented Generation via Singular Value Decomposition
链接: https://arxiv.org/abs/2607.10316
作者: Zhihui Sun
类目: Information Retrieval (cs.IR)
备注:
Abstract:Retrieval-Augmented Generation (RAG) systems enhance large language models by retrieving relevant documents from external knowledge bases. Recent work by Sarthi et al. (2024) introduced RAPTOR, which organizes documents into hierarchical tree structures for efficient retrieval, but requires expensive LLM-based abstractive summarization at each internal node – making large-scale deployment prohibitively costly. We present SVD-RAG, the first method to apply Singular Value Decomposition (SVD) on dense sentence embedding matrices for extractive summarization in hierarchical RAG. Unlike classical LSA which operates on sparse TF-IDF matrices, SVD-RAG exploits the rich semantic representations of modern embedding models, identifying the most informative sentences through their energy contribution in the principal components. Our approach is (1) deterministic – unlike LLM-based summarization, SVD produces identical results for the same input; (2) cost-efficient – tree construction requires no additional API calls beyond the initial embedding, reducing token consumption by ~85%; and (3) content-adaptive – the energy-ratio threshold tau automatically adjusts compression based on content complexity. In a controlled head-to-head comparison using identical corpora, clustering, and beam search, SVD-RAG achieves retrieval quality within 1-5% of RAPTOR with LLM summarization (MRR 0.867 vs. 0.875, Recall@1 0.483 vs. 0.458) while building the tree 317x faster (0.1s vs. 31.7s). On a scaled multi-topic benchmark with 205 chunks and 100 queries across 20 topic variations, SVD-RAG achieves a 4.2x improvement in Recall@1 and 3.1x improvement in MRR over flat embedding retrieval. We provide a detailed cost analysis and parameter sensitivity study. Our implementation is released as an open-source Python package. Subjects: Information Retrieval (cs.IR) Cite as: arXiv:2607.10316 [cs.IR] (or arXiv:2607.10316v1 [cs.IR] for this version) https://doi.org/10.48550/arXiv.2607.10316 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[IR-21] PTEI: Integrating Personality Traits to Enhance Emotional Intelligence in Large Language Models
链接: https://arxiv.org/abs/2607.10245
作者: Amir Reza Jafari,Praboda Rajapaksha,Reza Farahbakhsh,Noel Crespi
类目: Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注:
Abstract:Despite advances in Emotional Intelligence (EI), Large Language Models (LLMs) still significantly underperform humans in complex emotional reasoning. This gap originates partly from the limited incorporation of individual differences, particularly personality traits, which are fundamental to human emotional inference. To address this, we propose PTEI, a novel framework for integrating Personality Traits into Emotional Intelligence tasks using LLMs. In PTEI, MBTI and OCEAN personality traits are first extracted directly from the given emotional scenarios and then utilized as contextual knowledge within personality-aware prompts, guiding LLMs to accurately infer emotions and their underlying causes. To ensure optimal contextual grounding, we employ Contrastive Learning to construct an optimized retrieval system that surfaces emotionally and personally aligned scenarios, enhancing reasoning quality. Extensive experiments on established EI benchmarks show that PTEI enhances the Emotional Understanding (EU) capabilities of various LLMs, with the strongest improvement observed in GPT models. Combining PTEI with Chain-of-Thought (CoT) reasoning yields an additional 4 percent increase in accuracy. These findings underscore PTEI’s contribution toward advancing AI systems with more sophisticated social and psychological grounding.
[IR-22] Multilingual Semantic Retrieval for Apple Music Search
链接: https://arxiv.org/abs/2607.10239
作者: Vishalaksh Aggarwal,Kevin Sebastian,Vivek Kanojiya,Leo Le,Nick Tucey,Santosh Shankar
类目: Information Retrieval (cs.IR)
备注:
Abstract:Apple Music serves listeners across 150+ storefronts in dozens of languages, with a catalog that grows by hundreds of thousands of new tracks daily. At this scale, search recall on misspelled, transliterated, and cross-lingual queries becomes a dominant driver of session quality, particularly for tail queries that account for the majority of unique queries. We present a multilingual semantic retrieval system built on a 305M-parameter Siamese bi-encoder fine-tuned from GTE-multilingual-base with curriculum-scheduled multi-objective training. The model is integrated into the search stack via a hybrid retrieval architecture that blends dense nearest-neighbor results with the existing token-based index using quantile distribution matching, enabling deployment without retraining downstream rankers. Offline, the model achieves a 69% relative improvement in Hit@10 over GTE-multilingual-base. In a worldwide online A/B test, the system delivers a 2.28% relative conversion-rate (CR) lift overall, an 86% reduction in the no-result rate, and gains across every storefront with no observed regressions. The improvement is concentrated where it is needed most: tail queries see a 7.93% relative CR lift, compared with 0.89% for mid-frequency queries and 0.14% for head queries – evidence that semantic retrieval improves recall on hard queries without disturbing well-served popular ones. To our knowledge, this is one of the largest search-quality improvements deployed on the platform.
[IR-23] Consensus vs. Dissent: Dynamic LLM Modeling of Subjective Preferences in Group Recommenders RECSYS2026
链接: https://arxiv.org/abs/2607.10235
作者: Cedric Waterschoot,Nava Tintarev,Francesco Barile
类目: Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注: Full paper accepted at the 20th ACM Conference on Recommender Systems (RecSys 2026)
Abstract:Previous work in group recommender systems has demonstrated a sensitivity to the distribution of preferences within a group. Specifically, the selection of the preference aggregation strategy benefits from considering such group configurations. In this paper, we study whether LLMs are able to mimic this sensitivity and to select the ideal aggregation strategy (and corresponding recommendation) according to nuanced human perceptions of fairness, satisfaction, and consensus. We do this by fine-tuning Large Language Models (LLMs) on human survey data to serve as real-time judgmental models within the recommendation pipeline. Using a reasoning dataset distilled from DeepSeek-V3.1 and human ground truth assessments, we develop Judgmental Llama and Judgmental OLMo to simulate group assessments. Our pipeline successfully generates multiple recommendation candidates based on social choice-based aggregation strategies and dynamically selects the one that maximizes these predicted human-like evaluations. We further validate these suggestions in a user study (n=284) and find that our methodology achieved the highest scores for satisfaction and group consensus. Furthermore, we find that LLM judgments are most aligned with human perceptions of fairness, satisfaction and consensus when we also consider interaction effects between our LLM-based method and group configuration (e.g., minority or coalition). These findings give further support for dynamically adapting aggregation strategies to specific within-group preference distributions, and highlight the advantage of using LLMs for an adaptation that is aligned with subjective human judgments. Comments: Full paper accepted at the 20th ACM Conference on Recommender Systems (RecSys 2026) Subjects: Computation and Language (cs.CL); Information Retrieval (cs.IR) Cite as: arXiv:2607.10235 [cs.CL] (or arXiv:2607.10235v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2607.10235 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[IR-24] From Patent Expiry to Business Pathways: AI Workflows for Activating Innovation Archives
链接: https://arxiv.org/abs/2607.10179
作者: Sidney Shapiro,Mark Price
类目: Information Retrieval (cs.IR); Computation and Language (cs.CL); Computers and Society (cs.CY)
备注: 30 pages, Proof of concept using CIPO ST.96 data
Abstract:Patent databases represent one of the largest public archives of technical knowledge, yet much of this knowledge remains difficult to identify, interpret, and reuse once patent rights expire or lapse. This paper proposes an AI-enabled framework for discovering expired and lapsing patents, identifying technology trends, and translating patent disclosures into business pathways. We use pathways to mean structured commercialization routes such as SaaS products, services, licensing packages, consulting playbooks, training offerings, data products, or internal process tools. The framework treats patent expiry as both a business signal and an archival transition, not primarily as a legal problem. Legal status remains important, but it is one risk-screening input alongside customer need, implementation feasibility, channel access, and market timing. We describe a system architecture that combines patent metadata, maintenance-fee records, legal-status indicators, semantic search, patent-family analysis, market signals, and generative AI workflows. A proof of concept parses all 378 records in an official weekly CIPO ST.96 archive, identifies 20 expired, lapsed, or near-expiry candidates, tests the stability of the transparent scoring model, and uses a locally hosted Qwen3.6 model to populate structured review packets. The evaluation demonstrates reproducible ingestion, stable rankings under weight perturbation, and schema-conformant model output, while also exposing incomplete legal-status coverage and the need for register and expert review. We argue that AI can function as a discovery and translation layer for dormant technical knowledge, but that such systems must explicitly represent legal uncertainty, data limitations, and commercialization risk.
[IR-25] MC-RAG System: A Structure-Driven RAG System for Multi-Constraint Queries
链接: https://arxiv.org/abs/2607.10151
作者: Xiao Zhang,Yang Wan,Yi Li,Miao Xie,Chunli Lv
类目: Information Retrieval (cs.IR)
备注:
Abstract:Retrieval-Augmented Generation (RAG) systems are widely adopted in question answering, yet they often fail to satisfy complex multi-constraint queries, leading to constraint violations, factual inconsistencies, or hallucinations. We present Structure-Driven RAG System for Multi-Constraint Queries(MC-RAG), a structure-driven RAG system that reformulates retrieval as a subgraph matching problem over a knowledge graph. By integrating semantic and structural embeddings with path-level indexing, MC-RAG performs interpretable, structure-aware, and constraint-consistent retrieval and generation. During the demonstration, participants can input medical or encyclopedic multi-constraint queries, visualize how the system parses constraints, performs structural matching, and generates answers, thereby experiencing an end-to-end, interactive, and explainable RAG pipeline. A demo video is available at this https URL.
[IR-26] Adaptive Model Compression (AMC): Saliency-Driven Resource Allocation for Ultra-Low-Power Transformer Inference
链接: https://arxiv.org/abs/2607.10109
作者: Jiayin Hu,Kai Yuan,Vanessa Hu,Xuetao Yin,Jianhua Li,Sean Suchter
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Hardware Architecture (cs.AR); Machine Learning (cs.LG)
备注:
Abstract:Deploying large-scale transformer models on resource-constrained edge devices remains a challenge due to the high energy and memory overhead inherent in static inference, which processes simple and complex tokens with uniform intensity. To address this, we propose Adaptive Model Compression (AMC), a saliency-driven framework that dynamically allocates hardware resources based on token importance. By implementing a multi-tier architecture, our system identifies critical high-saliency information for full-precision processing while aggressively reducing the rank and bit-width of less significant data. Experimental results demonstrate that AMC achieves a 59.2% reduction in system energy and a 2.24x increase in throughput on 45nm CMOS hardware. This approach effectively extends the battery life of mobile devices by utilizing high-definition compute only where necessary, maintaining robust performance with a marginal 3.6% accuracy trade-off.
[IR-27] Scaling and Stabilizing Large-Scale Embedding-Based Retrieval
链接: https://arxiv.org/abs/2607.10096
作者: Zhen Yang,Juexin Lin,Hongwei Shang,Kaihao Li,Feng Liu,Satya Chembolu,Xunfan Cai,Xinyi Liu,Cun Mu,Tony Lee,Ciya Liao
类目: Information Retrieval (cs.IR)
备注:
Abstract:Embedding-based retrieval (EBR) is foundational to large-scale e-commerce search, yet its effectiveness is often constrained by the quality of training signals and the representational capacity of the encoder. Standard dual-encoders suffer from a training-inference gap: they are optimized on narrow candidate pools but must discriminate against hundreds of millions of items during inference. Furthermore, while transitioning to higher-capacity backbones can mitigate this gap, simply replacing a mature model can lead to inconsistent retrieval behavior and a loss of the domain-specific knowledge established in previous iterations. In this paper, we present a unified pipeline deployed at Walmart that addresses both signal quality and model evolution. Our contributions are two-fold: (1) Hybrid Hard Negative Mining: We integrate Online Cross-Batch Sampling to increase negative diversity by an order of magnitude and Hybrid Offline Mining, which combines cross-encoder predictions with metadata heuristics to identify nuanced mismatches. (2) Legacy-Aware Distillation: We transition from DistilBERT to a higher-capacity GTE-base encoder. To ensure a smooth and superior transition, we introduce a Warm-Start Distillation technique that transfers domain-specific expertise from the legacy model to the new backbone. Validated through extensive offline experiments and online A/B testing, the proposed pipeline is deployed in live production, delivering a +7.34% improvement in NDCG@5 and a +0.50% lift in gross revenue.
[IR-28] he Effect of Multi-Lingual and Keyword Adversarial Injection on LLM Relevance Judgment
链接: https://arxiv.org/abs/2607.10080
作者: Nguyen Khoi Vo,Duy Duong Tuong,Oleg Zendel,Mark Sanderson
类目: Information Retrieval (cs.IR)
备注:
Abstract:Large language models (LLMs) are increasingly being used as automated judges for relevance evaluation in information retrieval, yet their robustness to adversarial manipulation remains insufficiently understood, particularly in multilingual settings. In this work, we investigate the impact of cross-lingual prompt injection attacks on LLM-based relevance judgments using TREC Deep Learning collections and two open-weight models under established prompting frameworks. We examine both instruction-based and content-based injection strategies in 8 languages spanning different resource levels. Our results demonstrate that multilingual query-based injections are highly effective in inflating relevance scores while simultaneously evading existing prompt-injection defenses. We further found that, although existing defense mechanisms can be modified to mitigate such attacks, these injections can be easily adapted to bypass them. These findings highlight a critical gap in current defense approaches and demonstrate that language generalization can act as an attack vector, underscoring the need for more robust and proactive evaluation frameworks for LLM-as-a-judge systems. Subjects: Information Retrieval (cs.IR) Cite as: arXiv:2607.10080 [cs.IR] (or arXiv:2607.10080v1 [cs.IR] for this version) https://doi.org/10.48550/arXiv.2607.10080 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Nguyen Khoi Vo [view email] [v1] Sat, 11 Jul 2026 02:13:16 UTC (102 KB) Full-text links: Access Paper: View a PDF of the paper titled The Effect of Multi-Lingual and Keyword Adversarial Injection on LLM Relevance Judgment, by Nguyen Khoi Vo and 3 other authorsView PDFHTML (experimental)TeX Source view license Additional Features Audio Summary Current browse context: cs.IR prev | next new | recent | 2026-07 Change to browse by: cs References Citations NASA ADSGoogle Scholar Semantic Scholar export BibTeX citation Loading… BibTeX formatted citation loading… Data provided by: Bookmark checked="checked"class=“labs-tab-input”> Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv’s community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?) mathjaxToggle(); We gratefully acknowledge support from our major funders, member institutions, , and all contributors. About Help Contact Subscribe Copyright Privacy Accessibility Operational Status (opens in new tab) Major funding support from
[IR-29] okenizing Numerical and Embedding Features for LLM RecSys
链接: https://arxiv.org/abs/2607.10016
作者: Zhe Xu,Ankit Peshin,Chiyu Zhang,Feng Qi,Johnson Lui,Anil Ramakrishna,Justin Johnson,Carl Hu,Kaushik Rangadurai,Luke Simon
类目: Information Retrieval (cs.IR); Machine Learning (cs.LG)
备注:
Abstract:Large language models (LLMs) are increasingly used as backbone architectures for recommender systems because of their strong sequence modeling and representation learning capabilities. However, most LLM-based recommenders operate primarily on discrete textual tokens, whereas practical recommendation pipelines also rely on continuous numerical features and dense embedding features produced by upstream feature engineering or pretrained encoders. This mismatch limits the ability of LLM-based models to exploit fine-grained non-textual signals. We propose a soft-token fusion framework that maps numerical and embedding features into the LLM embedding space, allowing heterogeneous recommendation signals to be consumed through the standard token interface. We instantiate the framework in a shared-parameter LLM-based two-tower retrieval model and introduce an interaction-based fusion module that refines embedding and numerical soft tokens before they are inserted into the final LLM input. Experiments on three Amazon recommendation benchmarks show that soft-token fusion improves retrieval performance over LLM-based baselines, and that interaction-based fusion is more effective than direct concatenation of heterogeneous soft tokens.
[IR-30] An LLM -powered Agent ic Recommendation System for Connected TV Content Discovery
链接: https://arxiv.org/abs/2607.09988
作者: Lei Shi,Di Wang,Harry Tran,Helsing Xu,Yuchen Lu,Dhara Ghodasara,Wilson Chaney,Xueting Liao,Jerry Yu,Huayu Ding,Mingze Gao,Shike Mei,Shuo Tang,Zhe Zhang,Jianming He,Abhishek Kumar,Haotian Wu,Hamed Firooz,Li Li
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI)
备注: 13 pages, 3 figures
Abstract:Recommendation systems, from traditional multi-stage to recent unified generative architectures, face challenges in incorporating diverse contextual signals, such as trending topics, breaking news, cultural events, and cross-surface user activities, into their ranking pipelines. These systems are designed to consume structured behavioral signals with consistent schemas, and lack the reasoning capability to naturally process unstructured or heterogeneously formatted contextual information. Incorporating such signals typically requires feature engineering, bespoke data pipelines, and carefully tuned heuristics. In this paper, we present an LLM-powered agentic recommendation system designed for Connected TV (CTV) content discovery that addresses these limitations. Our system leverages the reasoning capabilities of large language models to naturally process and synthesize diverse signals across varying schemas and structures, eliminating much of the manual integration inherent in traditional ranking and retrieval systems. Recognizing that current LLM-based solutions still fall short of traditional machine learning models in several recommendation tasks, including retrieval efficiency, personalization precision, and scalability, we adopt an agentic architecture that orchestrates specialized components, allowing each sub-task to be handled by the most suitable method, whether LLM-based or traditional ML. The main contribution of this work is our engineering approach to successfully overcoming the practical limitations of enabling LLM for recommendation, particularly inference latency. We share insights from our work and discuss the trade-offs and lessons learned in building a hybrid system that combines the flexibility of LLMs with the performance of established recommendation techniques.
[IR-31] RouteRec: Strict Evaluation of Recommender-Agent Selection and Aggregation SIGIR2026
链接: https://arxiv.org/abs/2607.09908
作者: Kaiji Zhou,Vladimir Kalmykov,Yue Feng
类目: Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注: 8 pages, 7 figures. Accepted at AgentSearch 2026 (The First Workshop on Indexing, Retrieval, and Ranking of AI Agents), co-located with SIGIR 2026
Abstract:Recommender systems increasingly face a choice among heterogeneous agents – collaborative filters, sequential models, content-based retrievers, and LLM-based rerankers – yet no single agent is uniformly best. We study this choice as task-aware agent ranking under cost constraints using RouteRec, a framework that compares request-level hard selection with item-level learned aggregation over four traditional recommender agents and one LLM reranker agent. On MovieLens-1M, the full quality oracle has substantial headroom (HR@10 = 0.584), confirming that useful cross-agent signal exists. Under a leakage-free 5-fold out-of-fold protocol, however, hard selection remains below BM25 (0.223 vs. 0.254), and selective LLM escalation does not improve it. The same protocol yields a different outcome for learned aggregation: its cheap-only variant matches BM25 in HR and has a higher NDCG point estimate (0.123 vs. 0.114), while gated all-agent aggregation reaches HR@10 = 0.295 with 70.2% LLM calls. The resulting lesson is not that routing is solved, but that request-level selection of one complete agent list is too coarse for this sparse fixed-candidate setting; item-level aggregation is the more promising action space.
[IR-32] Serving the Long Tail: Training-Free LLM Candidate Generation for Vacation Rental Marketplaces KDD2026
链接: https://arxiv.org/abs/2607.09877
作者: Syed Mohammed Arshad Zaidi,Eric Rincon,Shayan Hassantabar
类目: Machine Learning (cs.LG); Information Retrieval (cs.IR)
备注: Accepted at TSMO 2026 workshop, co-located with KDD 2026; 9 pages
Abstract:Vacation rental marketplaces face a structural imbalance on the supply side: a small fraction of properties receive most user interactions, while the long tail of new, niche, and seasonal listings generates too little behavioral signal for collaborative filtering to serve effectively. At Vrbo, item-based k-nearest neighbors (IBKNN) is a core candidate generation channel, but leaves tens of thousands of properties with no candidates and produces weak neighborhoods for sparsely interacted ones. We present a training-free, LLM-based candidate generation pipeline that complements IBKNN using static property metadata alone. An off-the-shelf LLM synthesizes diverse semantic queries per property, a pre-trained text encoder embeds them, and an approximate nearest-neighbor index retrieves candidates from an 11.7M-property catalog. A Union fusion strategy merges these with IBKNN while preserving the behavioral channel’s ordering, guaranteeing no degradation on well-served properties, and a downstream learning-to-rank model re-scores the fused pool. Evaluated on 1.6M focal properties, the system extends candidate coverage to tens of thousands of properties IBKNN cannot reach, delivers its largest gains on the long-tail segment where behavioral methods are weakest, and matches or beats IBKNN at every K on shared properties. A downstream learning-to-rank stage further lifts the fused pool, yielding a complete candidate generation and re-ranking stack that serves the long tail without regressing well-served properties. We additionally show that Union fusion collapses the recall gap between a 3B open-weights LLM and frontier API-based models from 27-46% to under 1%, supporting self-hosted small-model deployment at marketplace catalog scale.
[IR-33] Memory-Conditioned Tool Calling for Camera-First Visual Agents
链接: https://arxiv.org/abs/2607.09822
作者: Xiaofan Wu,Xi Zeng,Miaoxia Chen,Peishan Chen,Shuyan Li,Jiyun Yao,Hanyong Zhong,Jiahao Zhu(Chance AI)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Human-Computer Interaction (cs.HC); Information Retrieval (cs.IR)
备注: 13 pages, 3 figures, 4 tables. Equal contribution: Xiaofan Wu, Xi Zeng. Corresponding author: xiaofan@chance.vision
Abstract:Recognition tells an agent what is in an image; personal memory affects what is worth looking up next. In a camera-first setting the user can send only an image, so the agent must form the lookups. We study whether personal visual memory improves agent-side tool choice and tool arguments, and thereby more user-aligned multi-tool lookups. The design uses a three-layer personal visual memory (profile, short-term focus, observations) that is loaded on each turn to condition an LLM tool-calling loop under camera-first intake, and includes conflict-aware write-back intended to refresh the user model for later captures. On 800 images paired with synthetic memory blocks constructed for controlled ablation, removing the full three-layer memory block reduces tool-query relevance by 0.47 points absolute (4.21 - 3.74 on a 5-point scale; 11.2% relative) and end-to-end utility by 0.082 absolute (0.842 - 0.760; 9.7% relative). These results measure memory conditioning of tool policy under image-only intake with fixed synthetic blocks, not multi-session write-back from live user histories.
[IR-34] Improved Answer Selection with Pre-Trained Word Embeddings
链接: https://arxiv.org/abs/1708.04326
作者: Rishav Chakravarti,Jiri Navratil,Cicero Nogueira dos Santos
类目: Information Retrieval (cs.IR); Computation and Language (cs.CL)
备注:
Abstract:This paper evaluates existing and newly proposed answer selection methods based on pre-trained word embeddings. Word embeddings are highly effective in various natural language processing tasks and their integration into traditional information retrieval (IR) systems allows for the capture of semantic relatedness between questions and answers. Empirical results on three publicly available data sets show significant gains over traditional term frequency based approaches in both supervised and unsupervised settings. We show that combining these word embedding features with traditional learning-to-rank techniques can achieve similar performance to state-of-the-art neural networks trained for the answer selection task.
[IR-35] From Stochastic to Stable: Rank Stability and Structural Sufficiency in AI Visibility Measurement
链接: https://arxiv.org/abs/2607.10341
作者: Ronald Sielinski
类目: Applications (stat.AP); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
备注: 31 pages, 11 figures
Abstract:AI visibility measurement is comparative: practitioners want to know which domains generative search engines cite most often and whether observed differences are large enough to support decisions. Yet the industry lacks a principled way to determine whether enough data has been collected. Collection budgets vary widely across studies and platforms, and conclusions are often drawn from rankings whose stability and precision are unknown. We introduce a sequential convergence framework based on two complementary criteria: rank stability evaluates whether the rank-correlation trajectory has reached a structural plateau, while structural sufficiency evaluates whether the spread of citation shares among established domains – those whose confidence intervals exclude zero – exceeds the uncertainty of those estimates. Together, these criteria distinguish rankings that have merely stabilized from those sufficiently resolved to support inference. Both are derived from regularities in the observed citation distribution, including its rank structure, uncertainty profile, and the boundary between observed and established domains. The framework retains a small number of structural constants but requires no externally specified query count, correlation target, or confidence-interval width target; stopping is driven by observed measurement uncertainty and remains robust across a range of sufficiency thresholds. Applied across 30 platform-topic combinations spanning Gemini, SearchGPT, and Perplexity, the framework adapts to platform- and topic-specific citation distributions. Results show that no fixed collection budget can be justified across contexts and that convergence can instead be evaluated from the structure of the observed distribution. The framework provides a practical basis for determining when AI visibility measurements are ready to support comparative analysis.
人机交互
[HC-0] FIERO: Empowering Creative Writing Through Collaborative Game Play
链接: https://arxiv.org/abs/2607.11837
作者: Chu Zhang,XiaoKe Zeng,Jin Zhang,Ruoyu Wen,Vince Siu,Richard William Allen,Ray LC
类目: Human-Computer Interaction (cs.HC)
备注: 51pages, 14 figures, 5 tables, will be presenting this paper at CHIPLAY 2026
Abstract:Creativity often flourishes in collaboration, such as when designers brainstorm a new app together, or storytellers collectively build a world with elements of each person’s narrative. However, collaborative storytelling can have challenges for its participants, such as when they disagree about the plot proposed, or when different ideas become fragmented when voiced individually. While current tools for creative collaboration focus on synchronous online text sharing, they often neglect the social dynamics of in-person collaboration critical to creative synergy. To address this, we created FIERO, a multiplayer web-based card game. Physical cards provide tangible scaffolding and social interaction, while the digital interface generates contextual visuals, facilitate group decisions, ensure narrative coherence, and synthesize different idea contributions using generative AI. Compared against online collaborative writing alone, the game significantly enhanced intuitive stimulation, idea fluency, and novelty generation, and also improved the content of the stories produced, leading to greater plot coherence (N=60). The cards provided creative structure and social engagement, while the interface provided contextualized augmentation without affecting player agency. This work shows how collaborative play can be utilized to foster creative support.
[HC-1] Supporting Reflection in LLM -based Exploratory Search
链接: https://arxiv.org/abs/2607.11810
作者: Giulia Di Fede,Salvatore Andolina
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Large Language Models (LLMs) can make exploratory search more efficient but may undermine the reflection and iterative sensemaking needed in unfamiliar domains. Existing LLM tools often prioritize rapid answers over supporting users in tracking how their understanding evolves and how well their strategies align with their goals. We present TrailLM, a system that helps users reconstruct and revisit their exploration paths to support reflection and metacognitive engagement during information seeking. By aligning LLM assistance with users’ sensemaking workflows, TrailLM aims to preserve the benefits of LLM-based search while enhancing opportunities for critical reflection on one’s own search process.
[HC-2] HandPad: A Bimanual Hand Interface for Fluid Window Interactions in VR
链接: https://arxiv.org/abs/2607.11807
作者: Wen Ying,Adil Rahman,Erzhen Hu,Seongkook Heo
类目: Human-Computer Interaction (cs.HC)
备注: Project page link: this https URL
Abstract:Virtual Reality (VR) offers potential for productivity work by creating expansive displays anywhere, yet current systems often rely on external input devices that limit the on-the-go use of mobile VR. We introduce HandPad, a suite of bare-hand interaction techniques that leverage the benefits of asymmetric bimanual coordination and self-haptic support. HandPad assigns the non-dominant hand (NDH) to establish spatial frames and interaction contexts, while the dominant hand (DH) performs fine-grained manipulation. Users can use NDH gestures as an input modifier to change the mode and target of DH interactions, including multi-window navigation, in-window content interaction, and window management. The palm surface of the NDH also serves as a physical touch surface, providing passive haptic feedback for effective DH touch interaction. Both hands and their interactions are spatially remapped to the window surface, enabling comfortable and direct interaction with virtual content. An exploratory study showed that HandPad enables efficient and ergonomic interaction, demonstrating its potential as a device-free approach for knowledge work in VR.
[HC-3] “We are all in big trouble! *Shock Emoji”: Personal Narratives in Expressing Emotions Opinions and Data Regarding Climate Change in TikTok Short Videos
链接: https://arxiv.org/abs/2607.11803
作者: Chu Zhang,Simai Huang,Shaohua Wu,Yihuan Chen,Ray LC
类目: Human-Computer Interaction (cs.HC); Social and Information Networks (cs.SI)
备注: 44 pages, 33 figures, presenting this paper at CSCW 2026
Abstract:Climate change is a source of anxiety about the future. Understanding how people express themselves about climate change enables us to address such concerns. To study climate change expression on social media, we analyzed 200 TikTok videos tagged with #climatechange, identifying four categories of content: expression-feelings, views-appeals, news-information, and trend-hijacking. We found that creators use humor to package sharp critiques, avoiding direct confrontation. They replace complex discussions with life stories, such as adopting a vegetarian lifestyle or deleting emails. They borrow from news media to present fragmented information as scientific interpretations, creating a perception of scientific credibility, balancing scientific accuracy with emotionality. Analysis of viewer responses showed they engaged empathetically, reshaping interpretations of videos. These interactions risk reinforcing existing views but help build community on TikTok, which lacks community structure. This study reveals how creators may retell news on science using personal narratives, highlighting how short-form videos enable climate communication.
[HC-4] Playful AI in Professional Email: A Field Experiment on Tone and Recipient Engagement
链接: https://arxiv.org/abs/2607.11749
作者: Ziv Ben-Zion,Teddy Lazebnik
类目: Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注:
Abstract:Large language models (LLMs) are rapidly reshaping workplace communication, yet whether AI-assisted writing changes how recipients actually behave, and through what channel, remains unknown. Here, in a randomized crossover field experiment, 121 employees across six companies sent work emails under three conditions over three weeks: unaided writing, GPT-5 rewriting in a playful tone, and GPT-5 rewriting in a professional tone. Across 16,880 emails, playful editing increased emotional positivity (B=+0.068, p0.001), and professional editing decreased it (B=-0.041, p0.001), yet neither condition directly altered open rates, reply rates, or response times. Instead, within-sender positivity strongly predicted both opening (OR=2.05) and replying (OR=3.32, p0.001), a significant indirect pathway through which AI editing shaped behavior, in the absence of any direct effect. These findings suggest that AI-assisted communication shapes workplace engagement not through its use, but through the emotional tone of the language it produces.
[HC-5] Requirement-Driven Design of Whole-Body Social Tactile Sensing via Virtual Human-Robot Interaction IROS2026
链接: https://arxiv.org/abs/2607.11690
作者: Dakarai Crowder,Ruohan Zhang,Alexis E. Block,Wenzhen Yuan
类目: Robotics (cs.RO); Human-Computer Interaction (cs.HC)
备注: 8 pages, 6 figures, accepted to IROS 2026
Abstract:Tactile sensing for social-physical human-robot interaction (spHRI) is designed in a hardware-driven manner, where predefined sensor configurations constrain coverage, spatial resolution, and the range of recognizable gestures. We propose a requirement-driven framework that derives sensing requirements, specifically spatial resolution and placement, directly from interaction data. Using a VR-based platform with haptic feedback, we collected high-resolution whole-body contact distributions across multiple social scenarios, from which we identified nine recurring social touch gestures. Eight gestures were selected for controlled data collection with 18 participants, yielding an open-source dataset of 5,520 trials. Analysis of contact distributions and simulated tactile encodings provides quantitative baselines for skin coverage and sensor density on a humanoid robot platform. While demonstrated on a single robot platform, the methodology is designed to be transferable to other robot morphologies, potentially enabling morphology-specific sensing requirements to be derived prior to hardware fabrication.
[HC-6] ERR@HRI 3.0 Challenge: Multimodal Detection of Errors and Anticipation in Human-Robot Interactions
链接: https://arxiv.org/abs/2607.11570
作者: Maria Teresa Parreira,Micol Spitale,Maia Stiber,Shiye Cao,Amama Mahmood,Chien-Ming Huang,Hatice Gunes,Wendy Ju
类目: Robotics (cs.RO); Human-Computer Interaction (cs.HC)
备注:
Abstract:As robots become increasingly integrated into human environments, their ability to detect and respond to errors remains critical for maintaining user trust and interaction quality. While recent advances in machine learning have improved error detection capabilities, most approaches are limited to specific contexts, controlled settings, or pre-extracted features, limiting their generalizability and applicability to real-world conditions. To address this challenge, the third edition of the ERR@HRI Challenge (ERR@HRI 3.0) provided researchers with two complementary datasets that enable end-to-end innovation in methods for both detecting and preventing errors in human-robot interaction. The challenge offered raw, non-anonymized video data from naturalistic settings: (1) the Bystander Affect Detection (BAD) dataset, containing webcam recordings of 45 participants’ spontaneous reactions to robot and human failure scenarios; and (2) the Bad Idea dataset, featuring 29 participants’ anticipatory facial responses while predicting action outcomes before failures occur. Both datasets were collected via crowdsourcing, capturing the inherent variability of real-world conditions. This naturalistic variability, while challenging, provides an authentic testbed for developing robust error detection systems. Participants developed multimodal machine learning models for bystander reaction detection (Track 1) and anticipatory outcome prediction (Track 2), with an optional cross-dataset generalization track (Track 3). Three teams submitted valid models, all of which surpassed our convolutional neural network baselines. This paper describes the datasets, tasks, baselines, and results of ERR@HRI 3.0, and discusses implications for building generalizable, context-aware, and anticipatory error detection systems for human-robot interaction.
[HC-7] oward Inclusive Avatar Design with Limb Differences Through Artificial Intelligence
链接: https://arxiv.org/abs/2607.11512
作者: Fernanda Miyuki Yamada,João Paulo Gois,Hiroki Takahashi
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Graphics (cs.GR)
备注:
Abstract:As extended reality becomes more popular for social interaction and entertainment, 3D avatars must represent the full diversity of body types. Most 3D avatar systems only support normative bodies and do not accurately depict people with limb differences, amputations, or other morphological variations. This paper reviews emerging technical approaches for inclusive 3D avatar customization for this group and current guidelines that promote respectful and accurate representation. We highlight persistent challenges, including the scarcity of diverse datasets and the limitations in animation for non-normative anatomies. This paper positions artificial intelligence as a promising path to overcoming these limitations and advancing inclusive 3D avatar generation.
[HC-8] LightMem-Ego: Your AI Memory for Everyday Life
链接: https://arxiv.org/abs/2607.11487
作者: Yijun Chen,Boyi Xiao,Yixian Zhao,Haoting Xia,Buqiang Xu,Jizhan Fang,Yanya Li,Yaqi Zheng,Xuehai Wang,Zirui Xue,Liuxin Zhang,Hui Li,Ningyu Zhang
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Human-Computer Interaction (cs.HC); Multimedia (cs.MM)
备注: Ongoing work
Abstract:Personal AI assistants on mobile and wearable devices continuously perceive users’ daily lives through visual and audio streams. However, answering queries about past experiences requires lightweight multimodal memory that can continuously accumulate, organize, and retrieve long-term experiences, which remains challenging. To address this challenge, we present LightMem-Ego, a lightweight streaming multimodal memory system for everyday-life assistance. The system continuously captures egocentric visual and audio streams, aligns them on a shared timeline, and organizes them into a hierarchical memory consisting of current, short-term, and long-term memory. Given a user query, LightMem-Ego dynamically routes retrieval to the appropriate memory level and generates answers grounded in multimodal evidence. The demonstration can be deployed on smartphones and AI glasses, supporting object finding, conversation recall, life summarization, routine discovery, and personalized assistance. Code is available at this https URL.
[HC-9] ManiScope: LLM -Assisted Visual Analytics of Cryptocurrency Manipulation Risk
链接: https://arxiv.org/abs/2607.11451
作者: Xiaolin Wen,Feng Liang,Yuanye Ma,Qishuang Fu,Zhengyu Sun,Feng Zhu,Can Liu,Yong Wang
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Cryptocurrency markets are vulnerable to trade-based manipulation, such as wash trading, which can distort price signals and mislead investors. Prior research has mainly focused on detecting manipulation using fixed rules or labeled examples, offering limited flexibility and interpretability for assessing potential risks. Existing visual analytics tools can reveal basic manipulation-related signals, such as token distribution, but still require substantial manual effort to integrate holder relationships, suspicious behaviors, and market dynamics for risk assessment. To address these limitations, we propose ManiScope, an LLM-assisted visual analytics system for analyzing trade-based manipulation risks in cryptocurrency markets. ManiScope provides coordinated views of token distributions, holder relationships, detailed holder behaviors, price dynamics, and suspicious trading patterns. To further enhance user analysis, ManiScope introduces a human-LLM collaborative visual analytics framework. Rather than acting as a basic reactive LLM assistant, the framework positions the LLM as a co-analyst that infers users’ analytical intent and emerging hypotheses from interaction context and surfaces relevant visual, statistical, and synthesized evidence for hypothesis evaluation. This design reduces repetitive inspection and strengthens evidence-based reasoning. We evaluate ManiScope through two case studies and a user study with 12 experienced cryptocurrency practitioners. The results suggest that ManiScope supports effective risk assessment of manipulation, reduces manual effort in evidence-seeking, and organizes findings around user hypotheses.
[HC-10] DiffLens: A Visualization System to Explore Local Differences in Graph Sampling
链接: https://arxiv.org/abs/2607.11424
作者: Zhiguang Zhou,Yong Zhang,Yuming Ma,Yuqi Zhou,Ke Lu,Yong Wang,Yuhua Liu,Jingfang Mao,Yongheng Wang,Ying Zhao,Wei Chen
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Graph sampling techniques have been widely used to simplify network computation and visualization, which also results in inevitable differences between the sampled networks and the original networks in terms of nodes, edges and structures. Investigating such differences can inform graph sampling technique users of the pros and cons of different techniques and select the appropriate one, and can also help graph sampling developers evaluate their own technique. However, there are still no systematic ways to achieve such a goal. This paper fills this research gap by first proposing systematic and generic quantitative measures to quantify three categories of graph differences (i.e., neighbor-based, path-based, and structure-based). Built upon this, we further propose DiffLens, a novel visualization system to help graph sampling developers and users intuitively explore local differences at different regions of their interest within a sampled graph, where three new lens-based visual designs are presented to display the neighbor-based, path-based, and structure-based differences respectively. We conducted two case studies and a user study using real-world network datasets to evaluate DiffLens. The results confirmed its effectiveness and usability in helping users explore local differences and compare different graph sampling strategies.
[HC-11] Same Stories Different Journeys: From Social Comparison to Sensemaking in AI-Mediated Peer Career Exploration
链接: https://arxiv.org/abs/2607.11039
作者: Pengping Tan,Baoquan Zhao,Zhenhui Peng
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注: 10 pages, 7 figures, 2 tables
Abstract:Young job seekers frequently turn to social media to compare themselves with peers and make sense of career possibilities. However, passive feed browsing creates a paradox: the authentic peer content that provides emotional grounding also triggers potentially detrimental upward social comparison and cognitive overload. Previous work has either structured online user-generated content to reduce noise without changing the passive browsing modality, or built AI-powered career exploration systems that disregard authentic human experiences. To address this gap, we developed JobMate, an interactive system that transforms real social media career posts into persona-grounded conversational AI agents, shifting the interaction from passive scrolling to active, personalized dialogue. We conducted a between-subjects study ( N = 24, three disciplines) comparing JobMate with native RedNote browsing. Our study shows that JobMate’s AI-mediated dialogue redirected social comparison from potentially detrimental upward comparison toward constructive self-reframing, while promoting sensemaking through active conversational engagement. However, users still relied on the authenticity of real peer content for emotional grounding. We discuss design implications for AI systems that augment authentic online user-generated content consumption across social comparison contexts.
[HC-12] EquiFusion: Kinematics-Agnostic Human Motion Prediction via Equivariant Latent Diffusion ECCV2026
链接: https://arxiv.org/abs/2607.10984
作者: Cecilia Curreli,Florian Hofherr,Dominik Muhle,Abhishek Saroha,Riccardo Marin,Daniel Cremers
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC); Machine Learning (cs.LG)
备注: Accepted to ECCV 2026. Visit our webpage at this https URL
Abstract:Existing Stochastic 3D Human Motion Prediction models are fundamentally constrained by hard-coding the skeleton kinematics, severely limiting generalization, preventing cross-dataset training, and requiring complex data retargeting. We introduce EquiFusion, the first kinematics-agnostic model to solve this bottleneck, implementing a latent diffusion model with a permutation equivariant architecture. EquiFusion treats the kinematics’ connectivity as an explicit input parameter, ensuring its internal computations are inherently agnostic to joint ordering and graph structure. This novel design enables truly cross-dataset generalization to unseen kinematics and unlocks novel zero-shot directions, such as motion prediction from partial or occluded observations and targeted limb generation. EquiFusion achieves state-of-the-art results on major benchmarks, being up to 75% more compact than previous kinematics-specific methods, while achieving faster training and inference. EquiFusion thus establishes a new, flexible standard for robust human motion prediction. Model and training code are available at this https URL.
[HC-13] When Context Dominates: Multimodal Signatures of Takeover Readiness Under Varying Hazard and Cognitive Load Conditions
链接: https://arxiv.org/abs/2607.10945
作者: Shiva Azimi,Yasaman Hakiminejad,Luis Gomero,Elizabeth Pantesco,Irene P. Kan,Meltem Izzetoglu,Arash Tavakoli
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Semi-automated driving systems promise to reduce crashes by assisting with perception and control, yet they simultaneously introduce additional human factors challenges by requiring drivers to monitor automation and rapidly resume control when failures occur. Prolonged passive monitoring can degrade vigilance, delay reactions, and increase takeover risk, but the extent to which distraction, hazard context, and drivers’ underlying cognitive and physiological states jointly shape takeover performance remains insufficiently understood. This study investigates these interacting factors using a controlled, within-subjects driving simulator experiment that crosses two hazard types (dynamic pedestrian and static crash events) with three levels of secondary task engagement (no task, conversation, and working memory load). Driver responses were assessed using a multimodal sensing framework that integrates vehicle-dynamics measures, subjective workload ratings, autonomic physiology (electrodermal activity and heart rate variability), and prefrontal cortical activation measured with functional near-infrared spectroscopy. Results show that hazard context is the primary determinant of takeover behavior, with pedestrian events producing longer and more variable maneuvers and crash events yielding faster and more stable responses. Secondary tasks exerted smaller effects on objective vehicle control, while internal-state measures showed more variable task-related patterns. These findings highlight the importance of jointly considering environmental context and human state when evaluating takeover readiness and designing driver monitoring systems. This study lays the groundwork for adaptive, context-aware strategies that support safer human-automation collaboration in semi-automated vehicles.
[HC-14] What to Distinguish and How? Opportunities and Challenges of Augmenting Multiple Cluttered Objects in Complex Scenes for People with Low Vision
链接: https://arxiv.org/abs/2607.10902
作者: Yuheng Wu,Ruijia Chen,Jaewook Lee,Jia Li,Kexin Zhang,Mengfong Lio,Weibing Wang,Sanbrita Mondal,Jon E. Froehlich,Yapeng Tian,Yuhang Zhao
类目: Human-Computer Interaction (cs.HC)
备注: 24 pages, 11 figures. Accepted to ASSETS '26 (28th International ACM SIGACCESS Conference on Computers and Accessibility)
Abstract:People with low vision (PLV) struggle to perceive complex scenes like busy kitchens and crowded streets, which contain many objects, visual clutter, and dynamic elements. Prior AR systems for low vision either enhance low-level visual features or augment task-relevant objects for single tasks in simple settings, leaving multi-object augmentation in complex scenes underexplored. Informed by a formative study characterizing important objects and their perceived importance for PLV, we built SceneGlance, a wearable AR system that recognizes important objects and visually distinguishes them by importance level. Through a controlled lab study with 12 PLV in a mock-up kitchen scene and a free-form think-aloud study with 13 PLV navigating an outdoor route, we found that AR distinction on object importance shifted PLV’s attention toward objects of higher importance, and supported perception strategies such as building mental snapshots from the augmentation distribution and hierarchical scanning by importance. However, this attention shift came with a tradeoff, as augmenting many objects reduced overall scene recall. The studies also surfaced challenges posed by AR augmentations in complex scenes, such as adjacent augmentations blending or interfering with each other, yielding design implications for more practical AR vision enhancement systems in the complex real world.
[HC-15] oward Contemplative LLM : A Modular Framework for Evaluating and Enhancing LLM Alignment in Mental Health ALT
链接: https://arxiv.org/abs/2607.10871
作者: Asher Sprigler,Yang-Yang Feng,Iftach Amir,Jonathan E. Bogard,Todd S Braver,Yi Ding,David Kinney,Yixue Zhao
类目: Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注: Accepted as an oral presentation at HARMONY 2026 (Human-centered AI Research for Mental Health, an Open Networking Symposium), co-located with IEEE/ACM Conference on Connected Health: Applications, Systems, and Engineering Technologies (CHASE 2026) held in Pittsburgh, August 6, 2026
Abstract:Contemplative traditions have long guided ethical behavior and prosocial interaction, and recent work suggests that contemplative principles (e.g., mindfulness, compassion, non-dual reasoning) may offer a promising paradigm for aligning large language models (LLMs), improving cooperation and reducing ethical violations in LLM outputs. However, as new models, evaluation metrics, and benchmarks emerge rapidly, it remains challenging to systematically assess whether and how contemplative principles enhance LLM alignment across diverse and evolving scenarios, and existing approaches are often ad hoc and fail to generalize. We present a modular, extensible evaluation framework, initially targeted at the mental health domain, that enables seamless integration of new models, metrics, and benchmarks through a reusable pipeline. The framework currently reproduces existing state-of-the-art results and supports systematic cross-evaluation by flexibly mixing and matching models, metrics, and benchmarks, enabling fair comparison and deeper insight. Its plug-and-play prompting module offers a principled pathway for incorporating ethical perspectives such as contemplative principles, allowing domain experts to define alignment criteria without requiring technical expertise. Although initially focused on mental health, the framework is domain-agnostic and extends naturally to areas such as decision-making, moral reasoning, and human-AI collaboration. By bridging computational evaluation with human-centered ethical reasoning, this work lays the groundwork for interdisciplinary research spanning cognitive science, behavioral economics, philosophy, and system design, toward robust, trustworthy, and socially beneficial human-AI ecosystems.
[HC-16] How Do Practitioners Build SE Agents ? Insights from a Mixed-Methods Study
链接: https://arxiv.org/abs/2607.10856
作者: Yunbo Lyu,David Williams,Jieke Shi,Zhensu Sun,Chao Peng,Zhou Yang,Federica Sarro,David Lo
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注:
Abstract:The rise of Software Engineering (SE) agents, i.e., LLM-based agents that can understand large codebases and carry out engineering tasks with limited human intervention, has been marked by rapid advances and adoption, but little is known about how developers build these systems in practice: existing studies mine repositories or examine deployment, but few investigate how SE agents are constructed. Through semi-structured interviews with 20 practitioners from 12 organizations and an online survey of 80 practitioners, this paper is the first to study how SE processes are changing in the development of SE agents and what challenges developers face. We find that as implementation becomes cheaper, bottlenecks shift rather than disappear: long-standing non-coding work such as requirements, coordination, review, and deployment becomes more visible, while reviewing and evaluating agent output becomes new and central. We characterize a seven-stage workflow and a shift toward evaluation-driven development, in which evaluation steers iteration and specifications become versioned artifacts read by both humans and agents. We further identify six challenges that teams face, together with the practices they adopt to address them, including unreliable evaluation signals, comprehension debt as code outpaces understanding, and behavioral changes introduced by provider-side model updates.
[HC-17] FaciliTrain: Practicing Facilitation Skills through AI-Simulated Group Dialogue
链接: https://arxiv.org/abs/2607.10850
作者: Hang Jiang,Yuanxin Zhu,Diyi Yang,Yoon Kim,Deb Roy,Jad Kabbara
类目: Human-Computer Interaction (cs.HC)
备注: Accepted to CSCW Poster 2026
Abstract:Skilled facilitation supports inclusive small-group dialogue, but deliberate practice is hard to scale: it depends on expert coaches, live practice partners, and iterative feedback. We present FaciliTrain, a voice-based training system in which learners step into the facilitator role of an AI-simulated multi-participant conversation, apply five evidence-based techniques, and receive structured AI feedback to support reflection. We report findings from a mixed-methods study with 24 participants, conducted as a formative study (N = 12) and a controlled pilot (N = 12; 6 treatment, 6 control). Both conditions achieved comparable accuracy on a live evaluation task, though treatment participants’ self-rated comfort declined significantly while control participants’ comfort improved (p = .018). Reflexive thematic analysis identifies four themes: the taxonomy externalizes implicit facilitation intuitions; Making Connections is the most cognitively demanding technique; voice acts as a deliberate-response forcing function; and participants overwhelmingly preferred AI feedback over self-practice. We discuss design implications for voice-based, AI-supported interpersonal skill training at scale.
[HC-18] Lottery and Sprint Arcade: Enabling Player-Driven Game Editing with Generative AI
链接: https://arxiv.org/abs/2607.10711
作者: Maya Grace Torii,Takahito Murakami,Yoichi Ochiai
类目: Human-Computer Interaction (cs.HC)
备注: 14 pages, 5 figures, 1 table. Published in The Journal of the Society for Art and Science, Vol. 25, No. 2, 2026
Abstract:Large language models (LLMs) are shifting game generation from offline automation toward play-driven modification through natural language interaction. In this work, we present a play-driven game editing system that enables players to modify a retro Space Invaders - style arcade game through voice-based natural-language commands during play. Spoken instructions are interpreted by an LLM and translated into structured updates of internal configuration parameters, allowing iterative play - edit - feedback cycles in an invader-style game environment without exposing underlying system details. The game includes approximately 100 editable configuration fields controlling mechanics, visuals, interaction patterns, and audio behavior, enabling gameplay transformation through incremental parameter changes. To investigate how users experience play-driven AI-mediated editing (RQ1) and how emergent editing patterns relate to variations in player experience (RQ2), we conducted a user study combining subjective evaluations, workload measures, and log-based analysis of editing behavior. Participants were able to modify gameplay with generally positive experiences and moderate workload, and interaction outcomes did not strongly depend on prior programming experience. Editing-log analysis revealed distinct experiential tendencies: adjustments to immediately perceptible parameters were associated with higher usability, whereas edits affecting core gameplay structures were more closely associated with enjoyment. Post-session reflections further identified diverse editing strategies, including exploratory experimentation, goal-driven structural modification, and iterative parameter tuning. These findings demonstrate that voice-driven editing can support accessible, play-driven human - AI co-creation within a structured invader-style arcade game environment.
[HC-19] Anamnesis: An Open-Source Platform for Large-Scale Backstory-Conditioned Survey Simulation
链接: https://arxiv.org/abs/2607.10628
作者: Song-Ze Yu,Joseph Suh,Serina Chang,David M. Chan
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注: Preprint
Abstract:We present Anamnesis, an interactive system for demographically controllable survey simulation using large language models. Open-source, and designed for non-technical users/researchers, Anamnesis enables the prototyping and stress-testing of survey instruments on virtual populations rather than real human subjects. The platform operationalizes the recently introduced Anthology and Alterity frameworks, which use structured narrative backstories to condition model responses, within a unified web interface. It supports open-ended generation, probabilistic demographic resampling, and multimodal (image and audio) surveys. We evaluate the system through two case studies: (1) replicating segments of Pew Research Center’s American Trends Panel (ATP) on political typology and biomedical issues and (2) emulating human preference in the New Yorker Caption Contest. In both cases, Anamnesis produces opinion distributions that more closely match real-world survey data than standard persona-prompting baselines, offering a transparent, reproducible, and open-source alternative to proprietary simulation services.
[HC-20] U-Lens: Supporting User Uncertainty Management in Long-Form LLM Responses
链接: https://arxiv.org/abs/2607.10604
作者: Yu Mei,Qingyue Zhuang,Jie Cai,Chang Liu,Zhi Zheng,Zhoutong Ye,Chun Yu,Yuanchun Shi
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Large language models (LLMs) are increasingly used to generate long-form answers for knowledge-intensive tasks, but users often struggle to decide which parts of a response deserve scrutiny, why they may be unreliable, and what to do next. Prior work on uncertainty communication has largely focused on making uncertainty visible through cues such as confidence scores, leaving less support for the broader process of managing uncertainty distributed across a long response. Through a formative study, we examine how users manage such uncertainty across three stages: interpretation, evaluation, and decision. Based on these insights, we derive design guidelines that address both stage-specific and cross-stage needs: uncertainty target representation, evaluative explanation, response guidance, and interactive presentation. We instantiate these guidelines in U-Lens, an uncertainty-management support system that organizes uncertain information in long-form responses into contextual inspection targets, prioritizes them for attention, and connects each target with evaluative context and response options. We evaluated U-Lens in a controlled within-subjects study with 18 participants, comparing it against a confidence-cue baseline. Our results show that U-Lens improved verification efficiency and effort allocation, lowered perceived workload, and strengthened perceived support across interpretation, evaluation, and decision stages. This work reframes uncertainty support for generative AI from presenting isolated, text-centered cues toward supporting the user-centered process of interpreting, evaluating, and acting on uncertain information.
[HC-21] Navigating the Open-Source Model Ecosystem: An Empirical Study of Creator Practices in Artistic Image Generation
链接: https://arxiv.org/abs/2607.10538
作者: Yiluo Wei,Yupeng He,Qiming Ye,Gareth Tyson
类目: Human-Computer Interaction (cs.HC); Computer Vision and Pattern Recognition (cs.CV); Computers and Society (cs.CY)
备注: Accepted to MM’26: The 34th ACM International Conference on Multimedia
Abstract:The open-sourcing of powerful image generation models has created a vibrant ecosystem where creators curate and combine a vast array of community-contributed models. This practice stands in sharp contrast to using closed-source tools like Midjourney. Yet, little is known about these emerging creative workflows. To bridge this gap, this paper presents the first large-scale empirical study of creator model usage behavior within this open-source image generation ecosystem. We construct a novel dataset of 6 million images with their embedded generation metadata – a detailed recipe of the creation process, including the models used and the prompts. By linking the usage of 22.4K base models and 154K LoRA models to the images, our findings underscore the ecosystem’s unique strengths and its inherent obstacles. This provides valuable insights for making this ecosystem more sustainable and innovative. Moreover, we make our dataset publicly available, providing creators with practical references for producing better artworks and researchers to facilitate further studies.
[HC-22] Motif: Discovering and Automating Personal Web Workflows
链接: https://arxiv.org/abs/2607.10531
作者: Shaokang Jiang,Daye Nam
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注:
Abstract:Recent advances in LLMs and existing work on programming by demonstration have made it possible for end users to create automations by explicitly demonstrating their behavior to LLMs. However, these approaches rely on the assumption that users know what to automate and what is capable of being automated. Additionally, automation via LLM agents is often expensive compared with programs. We introduce Motif, a system that passively observes everyday browser activity to discover recurring interaction patterns that are programmable, makes recommendations to users whenever a pattern is discovered and generate a program to install after user confirmation. Users can review, and refine the program using natural language. We evaluated Motif in a multi-day study, comparing its ambient discoveries against automations users attempted to build via ``vibe coding.‘’ With eight participants, Motif discovered more automatable patterns than users recognized. Most of them matched participants’ routines and were useful. Follow-up surveys showed most would continue using Motif-generated programs.
[HC-23] How Data Narratives Go Wrong: A Taxonomy of Issues Across the Data Communication Process
链接: https://arxiv.org/abs/2607.10523
作者: Yu Fu,Jiawei Zhou,Sichen Jin,Munmun De Choudhury,Cindy Xiong Bearfield,John Stasko
类目: Human-Computer Interaction (cs.HC); Computers and Society (cs.CY)
备注: 22 pages, 7 figures, 2 tables
Abstract:Data narratives increasingly shape public understanding, but their failures are rarely just isolated factual errors or deceptive charts. Instead, they emerge through a broader meaning-making process in which quantitative evidence is transformed into claims, representations, and arguments. While prior work has examined these failures across disparate fields (e.g., statistics, visualization, and fact-checking), the community lacks a holistic lens to explain how these issues arise, propagate, and compound. To address this gap, we introduce TIC, a Taxonomy of Issues in Data Communication, synthesized from prior literature and refined through the qualitative annotation of 700 real-world data narratives from fact-checking sites, research datasets, and controversial media. TIC organizes recurring breakdowns across six dimensions-data, analysis, visual encoding, text, reasoning, and interpretation-and situates them within a framework spanning analysis, narrative construction, and audience reception. Alongside the taxonomy and process framework, we contribute a qualitatively annotated case corpus with coding justifications and an interactive browsing interface. Collectively, these contributions provide a structured lens for diagnosing problematic data narratives and informing future sociotechnical support for trustworthy data communication.
[HC-24] hreat Vectors and the State of the Art in Defense Methods for Security in Neurotechnology
链接: https://arxiv.org/abs/2607.10451
作者: Bryce-Allen Bagley,Nathaniel Rose,Quintus Kilbourn,Matthew Canham
类目: Cryptography and Security (cs.CR); Emerging Technologies (cs.ET); Human-Computer Interaction (cs.HC); Neurons and Cognition (q-bio.NC)
备注:
Abstract:Brain-computer interfaces (BCIs) are a class of diverse hardware modalities, associated software, and connected devices which are widely used in a variety of fields, including neurosurgery, biomedical data analysis, and neuroimaging. Recent years have seen rapid advancements in BCI technology, and neurotechnology more broadly, with the first devices now passing clinical trials, early examples of consumer hardware entering the market, and many variants of consumer and medical hardware with increasingly extensive capabilities being developed rapidly. However, research and development in security for BCIs–known as neurosecurity–lags significantly behind the capabilities of BCIs themselves. In an effort to address as many vulnerabilities as feasible immediately, in this paper we review the current state of the art in neurosecurity, thoroughly survey the breadth and complexity of both firmly established and highly probable security threats to BCI systems, and provide recommendations of existing methods from cybersecurity, hardware security, and machine learning which can immediately be applied to address some of these gaps in neurosecurity.
[HC-25] Spatula: Exploring On-Demand In-Situ Interfaces and Interaction for Attribute Control
链接: https://arxiv.org/abs/2607.10405
作者: Boyu Li,Linjie Qiu,Lin-Ping Yuan,Duotun Wang,Yue Jiang,Zeyu Wang,Hongbo Fu
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Graphics (cs.GR)
备注:
Abstract:Controlling attributes is a critical step toward achieving the final creative outcome, yet current approaches fall short in supporting users in the iterative refinement of generative content. We propose Spatula, a proof-of-concept system that generates on-demand, in-situ attribute control interfaces and interactions for creating motion graphics. Building on a technical probe that automatically analyzes animation context and generates corresponding attributes and UI, we frame attribute control as an explorable landscape and explore the attribute control space along four key dimensions: Discoverability, Resolution, Scope, and Expandability. Findings from a user study (N=12) show that our system provides intuitive and convenient interactions while supporting diverse needs for fine-grained parameter control. Furthermore, our applications demonstrate that the plug-and-play design generalizes to other domains, such as web design and 3D modeling.
[HC-26] Navigating the Crowd: Non-linear MPC with Social Forces Dynamics for Human-Aware Robot Navigation IROS2026
链接: https://arxiv.org/abs/2607.10374
作者: Stefano Trepella,Andrea Ostuni,Mauro Martini,Pablo Pueyo,Noé Pérez-Higueras,Marcello Chiaberge,Fernando Caballero,Luis Merino
类目: Robotics (cs.RO); Human-Computer Interaction (cs.HC)
备注: 9 pages, 7 figures, accepted at IROS 2026
Abstract:Safe and socially compliant navigation remains a fundamental challenge for autonomous robots operating in human-populated environments. Beyond collision avoidance, robots must anticipate human motion and respect personal space to ensure human comfort. Model Predictive Control (MPC) offers a robust alternative to classical and data-driven methods, although its effectiveness strongly depends on accurate human motion prediction and efficient computation. This paper introduces SFM-NMPC, a Social Force Model-based Non-linear Model Predictive Control framework that embeds human motion prediction directly within the optimization loop. By incorporating the Social Force Model into the dynamic model of surrounding agents, the controller jointly predicts the trajectories of humans and robots over the prediction horizon, thereby enabling socially-aware planning. A tailored set of social cost functions guides the optimization toward human-compliant behaviors. Despite the increased model complexity, the proposed formulation runs in real time at 20 Hz. Extensive simulated testing in crowded environments demonstrates that SFM-NMPC outperforms state-of-the-art baselines in social compliance metrics while maintaining efficient and smooth navigation. Visual trajectory analysis and an ablation study further highlight the contribution of the embedded SFM dynamics and social cost terms, confirming the effectiveness of the proposed approach for real-world social navigation.
[HC-27] Intervenability as a Design Requirement for Autonomy and Oversight within Human-Centered AI
链接: https://arxiv.org/abs/2607.10322
作者: Thomas Herrmann
类目: Human-Computer Interaction (cs.HC); Computers and Society (cs.CY)
备注: 24 Pages, 6 Figures
Abstract:Based on the literature and several practical examples of possible AI applica-tions, we outline the concept of intervenability. This new phenomenon is not covered by emergency shutdowns, workarounds, or the reconfiguration of automated systems. Intervenability instantiates the principles of control-lability, autonomy, oversight, and keeping humans in the loop in the context of AI. We provide a taxonomy that encompasses a range of possibilities for intervening activities and differentiates them regarding the mental effort of the users. This taxonomy extends the scope of interventions from real-time control of automated processes to AI-based discrete case-related decision-making. This is in accordance with human-centered AI, which seeks to combine human strengths with the usage of AI. We demonstrate how inter-venability can potentially contribute to the ongoing development of human capabilities on the one hand and to further technical improvement by recon-figuration of AI on the other. Exploring and collaboratively reflecting on the effects of interventions as an integral part of organizational practices is key to enabling this continuous improvement on both sides. Intervenability also provides further momentum for the design of an AI that can help realize in-terventions on its own and advance a smooth transition from intervention to reconfiguration of the AI.
[HC-28] Neutralizing Structural Inequality in the Nigerian FinTech Sector
链接: https://arxiv.org/abs/2607.10317
作者: Muhammad Abdullahi Said
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Human-Computer Interaction (cs.HC)
备注:
Abstract:Algorithmic decision systems in financial services often rely on data proxies that inadvertently encode structural inequalities. This paper introduces a hierarchical human-AI triage model for Point of Sale fraud detection in the Nigerian FinTech sector. Adopting a We Are All Equal worldview, we address the challenge of discrimination laundering, wherein the system misinterprets infrastructure related aleatoric noise such as rural network timeouts as fraudulent intent. We implement a three-tier routing policy utilizing a calibrated ensemble model as a primary filter. The policy routes transactions characterized by epistemic uncertainty such as cold start new accounts to specialist analysts while reserving high stakes cases for a senior supervisor. To manage finite human capacity, we utilize a dynamic shadow price to ration human attention and implement a random audit mechanism to prevent human skill atrophy. Our experimental results demonstrate a statistically significant 1.88% complementarity gap and a 24.79% percentage point gain in fraud recall over an autonomous baseline. Crucially, the model reduces the regional performance gap from 19.43 to 2.88 percentage points, neutralizing structural bias. Hierarchical collaboration provides a robust mechanism for substantive equality of opportunity, ensuring that rural accounts are not excluded from the digital economy due to environmental brute luck.
[HC-29] VRExplorer: A Model-based Approach for Semi-Automated Testing of Virtual Reality Scenes
链接: https://arxiv.org/abs/2607.10174
作者: Zhengyang Zhu,Hong-Ning Dai,Hanyang Guo,Zeqin Liao,Zibin Zheng
类目: oftware Engineering (cs.SE); Human-Computer Interaction (cs.HC)
备注: Published in ASE 2025
Abstract:With the proliferation of Virtual Reality (VR) markets, VR applications are rapidly expanding in scale and complexity, thereby driving an urgent need for assuring VR software quality. Different from traditional mobile applications and computer software, VR testing faces unique challenges due to diverse interactions with virtual objects, complex 3D virtual environments, and intricate sequences to complete tasks. All of these emerging challenges hinder existing VR testing tools from effectively and systematically testing VR applications. In this paper, we present VRExplorer, a novel model-based testing tool to effectively interact with diverse virtual objects and explore complex VR scenes. Particularly, we design the Entity, Action, and Task (EAT) framework for modeling diverse VR interactions in a generic way. Built upon the EAT framework, we then present the VRExplorer agent, which can achieve effective scene exploration by incorporating meticulously designed path-finding algorithms into Unity’s NavMesh. Moreover, the VRExplorer agent can also systematically execute interaction decisions on top of the Probabilistic Finite State Machine (PFSM). Experimental evaluation on 11 representative VR projects shows that VRExplorer consistently outperforms the state-of-the-art (SOTA) approach VRGuide by achieving significantly higher coverage and better efficiency. Specifically, VRExplorer yields up to 122.8% and 52.8% improvements over VRGuide in terms of executable lines of code (ELOC) coverage and method (function) coverage, respectively. Furthermore, ablation results also verify the essential contributions of each designed module. More importantly, our VRExplorer has successfully detected two functional bugs and one non-functional bug from real-world projects. Comments: Published in ASE 2025 Subjects: Software Engineering (cs.SE); Human-Computer Interaction (cs.HC) Cite as: arXiv:2607.10174 [cs.SE] (or arXiv:2607.10174v1 [cs.SE] for this version) https://doi.org/10.48550/arXiv.2607.10174 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Journalreference: 2025 40th IEEE/ACM International Conference on Automated Software Engineering (ASE), pp. 482-494, 2025 Related DOI: https://doi.org/10.1109/ASE63991.2025.00047 Focus to learn more DOI(s) linking to related resources
[HC-30] Learning behavior accounts for background-related advantage in AI-assisted education
链接: https://arxiv.org/abs/2607.10101
作者: Jingwei Yi,Yueqi Xie,Jiyan He,Rui Ye,Junming Huang,Bin Zhu,Sean Rintel,Yu Xie,Xing Xie,Fangzhao Wu
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Generative AI has been found, and will likely be found increasingly, useful in education. However, existing AI-for-education studies provide inconsistent evidence on its average effects. More broadly, research on prior educational technologies shows that average effects often mask substantial heterogeneity across student populations. Motivated by this evidence, this study examines heterogeneity in students’ learning behavior with AI, which students benefit from AI assistance, and how learner profiles and learning behavior shape these patterns. To this end, we recruited 318 university students to participate in structured learning experiments lasting up to 125 minutes. Our findings indicate that students’ learning behavior is strongly associated with learning outcomes, with behaviors characterized by proactive and critical engagement, rather than limited engagement, associated with significantly better performance. These behavioral differences are related to learner profiles, with students from higher-ranking universities and those with greater prior knowledge tending to benefit more, consistent with their greater likelihood of adopting proactive interaction strategies. Accounting for learning behavior substantially weakens or eliminates the associations between learner profiles and learning outcomes, suggesting that how students use AI is a key pathway through which background differences are linked to learning gains. Overall, this work provides a deeper understanding of AI assistance in education by showing how differences in learner profiles and learning behavior shape who benefits from AI-supported learning. These insights can help educators and students better navigate and integrate AI into educational practices.
[HC-31] SyncSpace: Layout-Conditioned 3D Gaussian Splatting for Space Reskinning in Mixed Reality
链接: https://arxiv.org/abs/2607.10050
作者: Qinchuan Zhang,Weibo Xu,Yunge Wen
类目: Human-Computer Interaction (cs.HC)
备注: 2 pages, 2 figures
Abstract:We present SyncSpace, a system that achieves both spatial alignment and visual consistency between a generated 3DGS world and physical space. We first scan the space via depth sensing to extract 3D bounding boxes, which we render into a layout-only panorama and feed as a geometric prior to a generative world model, producing a Gaussian splat scene in which objects are re-semantized to fit a target style without per-object control. We then align the generated scene to physical space with a coarse-to-fine registration algorithm, refined manually via pinch gestures when automatic registration does not converge. We demonstrate a hand-tracked engulfment interaction in which the virtual world rises to replace the physical space, and show a single space reskinned into multiple stylistically distinct worlds with its layout preserved.
[HC-32] “Code Is Cheap. Show Me the Talk.”: Lessons from Teaching and Managing AI Coding Tool Usage in a Visualization Course
链接: https://arxiv.org/abs/2607.09938
作者: Zhongzheng Xu,Taehyun Yang,Fumeng Yang
类目: Human-Computer Interaction (cs.HC)
备注: 5 Figures, 7 Pages
Abstract:Generative Artificial Intelligence (GenAI) coding tools are transforming visualization education. They can assist with implementation and design, but they can also let students bypass intended learning trajectories. In this paper, we share our retrospective experience managing and teaching AI use in an upper-level visualization course. We implemented prompt injections, asked oral checkout questions, and taught two AI coding labs. Prior to our coding labs, at least half of the students had already used AI tools in their assignments. In both AI coding labs, refinement accounted for about half of students’ prompting logs, and explanation was almost absent. In the lab where AI coding was optional, 44 of 78 (56.4%) submissions preferred the scaffolded instructions over designing their own prompts. Students’ final projects were more polished than in our previous offering, but also more visually homogeneous. Our reflections point to the need for clearer AI use boundaries and instruction on prompting, and for teaching students to question generic AI designs and adapt them to their data and story.
[HC-33] When LLM Tutoring Responses Work: Evidence from Student Programming Conversations
链接: https://arxiv.org/abs/2607.09919
作者: Mohammad Fahim Abrar,Shayla Sharmin,Roghayeh Leila Barmaki
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:As students increasingly use LLM tutors in computer science education, one question becomes especially important: what kind of response helps a student continue productively? Prior work has studied how students use LLMs in computer science education, but less is known about how tutoring response styles are associated with student follow-up across programming help-seeking contexts. This paper analyzes StudyChat (UMass, 2026), a public dataset of student and ChatGPT tutoring conversations from an artificial intelligence course. We transformed StudyChat into 16,851 assistant-response interactions from 203 students and 2,214 conversations. Using local LLM-assisted annotation with Gemma 4, we labeled student help-seeking situations, student state, assistant response style, and student next-turn outcome. Human validation showed 82% agreement with the LLM-assisted labels (Cohen’s \kappa=.74 ). We analyzed productive continuation and unresolved continuation across the full dataset and across help-seeking contexts. Globally, response style was significantly associated with productive continuation, \chi^2(7)=100.39 , p.001 , V=.078 , and unresolved continuation, \chi^2(7)=125.77 , p.001 , V=.087 , though effect sizes were small. Verification feedback had the highest productive-continuation rate (82.4%), while direct answers had the lowest (62.7%). Descriptively, response-style score ranges were smallest in low-confusion conceptual contexts (.017) and largest in high-cognitive-load contexts (.203). More detailed comparisons showed situation-dependent response patterns. For example, stepwise guidance was followed by greater confusion decrease in high-cognitive-load code requests, while direct answers were followed by more unresolved continuation in high-load debugging. These findings support context-aware evaluation and design of AI tutoring responses for programming education.
[HC-34] Exploring Agent ic Workflows for Generating High Quality Math Visual Aids
链接: https://arxiv.org/abs/2607.09839
作者: Rizwaan Malik,Ashna Khetan,Isabel Sieh,Samin Khan
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Human-Computer Interaction (cs.HC)
备注: 13 pages, 9 figures. Exploratory course project on agentic workflows for generating and evaluating K 12 mathematical diagrams
Abstract:Mathematical diagrams play a crucial role in K 12 education, both as problem components and as scaffolding for student comprehension. However, current AI tools, including Large Language Models (LLMs), struggle to reliably generate accurate and pedagogically sound visual diagrams, even when provided with detailed descriptions. A significant gap therefore remains in the reliable generation of diagrams for middle school mathematics. To address this, we introduce an agentic workflow that enables LLM agents to evaluate the quality of generated visuals and use this feedback to iteratively improve their outputs. This self improvement loop aims to enhance the accuracy and educational appropriateness of AI generated diagrams. Our research investigates two questions. First, can LLMs accurately generate quality assurance questions for a visual aid given specific criteria for visual quality? Second, given valid quality assurance questions, can Vision Language Models effectively evaluate generated K 12 visual aids and use the resulting feedback to improve them iteratively? We conduct an exploratory evaluation of our agentic workflow and identify key areas for improvement, including stronger spatial reasoning and more comprehensive coverage of diagram features in the generated quality assurance questions. Our results provide preliminary evidence that this approach can improve the reliability and educational value of AI generated mathematical diagrams.
[HC-35] he Individual-Targeting Assumption: A Systematic Review of Proactive Robots in Human Group Settings
链接: https://arxiv.org/abs/2607.09734
作者: Tauhid Tanjim,Tasmia Mayen,Malte F. Jung,Susan R. Fussell
类目: Human-Computer Interaction (cs.HC); Robotics (cs.RO)
备注: 8 pages, 3 figures, 4 tables. Accepted at the 2026 IEEE International Conference on Robot and Human Interactive Communication (RO-MAN)
Abstract:Proactive robots are increasingly deployed in public environments where people are encountered not as isolated individuals but as members of cohesive social groups. Yet whether the prevailing design paradigm in proactive human-robot interaction (HRI) accounts for the relational structure that defines a group as a social unit remains largely unexamined. Through a systematic review of 63 proactive HRI studies in group settings from 2000 to 2025, we identify a recurring tendency, the Individual-Targeting Assumption (ITA), in which robots treat co-present people as independent engagement targets. We find that ITA is present in 60.3% of the corpus, with group-aware approaches emerging almost entirely after the robot is already embedded in an ongoing interaction. Critically, how a robot should detect and negotiate entry into a pre-formed group before initiating contact remains unaddressed across the corpus. Three failure modes, engagement misdetection, social ratification blindness, and bystander neglect, emerge as recurring patterns when proactive robots interact with human groups. These findings reframe proactive HRI in group settings not as an extension of dyadic interaction but as a qualitatively distinct design problem, and they identify the robot’s approach to the group during the entry phase as a critical and understudied open challenge.
[HC-36] Narrix: Remixing Narrative Strategies from Examples for Story Writing
链接: https://arxiv.org/abs/2604.07643
作者: Chao Zhang,Shunan Guo,Abe Davis,Eunyee Koh
类目: Human-Computer Interaction (cs.HC); Computation and Language (cs.CL)
备注: 24 pages, 10 figures. To appear in CHI '26: Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems, April 13-17, 2026, Barcelona, Spain. DOI: this https URL
Abstract:Experienced storytellers decompose stories into local narrative strategies and how these strategies shape higher-level arcs. This decomposition helps writers recognize patterns in others’ work and adapt those patterns to tell new stories. Novices, however, struggle to identify these strategies or to reuse them effectively. We present Narrix, a novel writing tool that helps novice writers recognize narrative strategies in example stories and repurpose these strategies in their own writing. Narrix analyzes strategies in example stories, highlights them with color-coded lexical cues and explanations, and situates them on an interactive story arc for exploration by emotional shifts and turning points. Writers then drag strategies onto multi-dimensional tracks and apply block-scoped edits to revise or continue their drafts through controlled generation steered by specified strategies. Through a within-subjects study (N=12), Narrix showed improved participants’ retention, confidence, and creative adaptation of narrative strategies compared to a baseline chat-based writing interface.
计算机视觉
[CV-0] Read It Back: Pretrained MLLM s Are Zero-Shot Reward Models for Text-to-Image Generation
链接: https://arxiv.org/abs/2607.11886
作者: Runhui Huang,Qihui Zhang,Zhe Liu,Yu Gao,Jie Wu,Hengshuang Zhao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:In this paper, we propose SpectraReward, a training-free reward function that turns pretrained MLLMs into off-the-shelf reward models for image-generation reinforcement learning. Instead of asking the MLLM to judge a generated image or answer decomposed verification questions, SpectraReward measures how well the original prompt can be recovered from the generated image through a single image-conditioned, teacher-forced forward pass. We use the average image-conditioned prompt log-likelihood as the reward, directly reusing the MLLM’s pretrained image-text alignment ability without preference labels, reward-model fine-tuning. We further introduce Self-SpectraReward, a special case for unified multimodal models where the policy’s own understanding branch serves as the reward model for its generation branch, forming a closed-loop self-improving framework without external reward models or external knowledge. Extensive experiments validate SpectraReward through a broad image-generation RL study covering two diffusion models, three RL algorithms, nine reward MLLM backbones from four MLLM families spanning 4B to 235B parameters, and five out-of-distribution text-to-image benchmarks. Results show that both SpectraReward and Self-SpectraReward significantly and consistently improve generation performance and outperform prior MLLM-derived reward training methods. Further analysis reveals that larger reward MLLMs are not always better, while Self-SpectraReward can match or surpass much larger external reward models, suggesting that reward-policy alignment is a key factor for effective image-generation RL. Project Page: this https URL
[CV-1] Latent-Identity Tuning in Text-to-Image Personalization Models
链接: https://arxiv.org/abs/2607.11885
作者: Daniel Garibi,Ronen Kamenetsky,Hadar Averbuch-Elor,Daniel Cohen-Or,Or Patashnik
类目: Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR)
备注: Project page at: this https URL
Abstract:Generating and editing a person’s face demands high precision, as even minor modifications can significantly alter a subject’s perceived identity. Current personalization and editing methods built on general-purpose text-to-image models, however, often lack the precision required for fine-grained facial edits. We present a method for fine-grained identity tuning in text-to-image personalization models. Unlike standard image editing, which operates on a given image, identity tuning modifies the latent representation of a specific identity, enabling the generation of diverse images that consistently depict the same edited identity. To enable fine-grained latent identity tuning, we explore the latent space of a pre-trained, frozen encoder for text-to-image personalization. Our approach requires no additional training. Instead, it leverages the existing architecture of a frozen encoder to uncover latent semantic directions. This space consists of a set of latent tokens that play distinct roles in capturing different aspects of an identity and often correspond to specific spatial or semantic facial regions. We show that meaningful directions can be identified within this space and within subspaces defined by selected tokens, enabling localized, fine-grained, and semantically coherent edits. We validate our approach through qualitative and quantitative experiments that demonstrate diverse localized facial edits while preserving cross-image identity consistency. Project page at: this https URL
[CV-2] Evidence-Backed Video Question Answering
链接: https://arxiv.org/abs/2607.11862
作者: Shijie Wang,Honglu Zhou,Ziyang Wang,Ran Xu,Caiming Xiong,Silvio Savarese,Chen Sun,Juan Carlos Niebles
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Current Video Large Language Models (Video LLMs) excel in question answering (QA) but largely operate as black boxes, providing textual answers without verifiable visual grounding. Existing explainability efforts rely on textual rationales or sparse bounding boxes, which struggle to capture complex video dynamics such as occlusions and non-rigid deformations. We propose Evidence-Backed Video Question Answering (E-VQA), a novel task requiring models to jointly output a semantic answer and precise spatio-temporal evidence: temporal segments and dense, tracked object segmentation masklets. To support this, we introduce ST-Evidence, the first human-verified benchmark for both discriminative and generative pixel-level grounding. Evaluations of state-of-the-art models reveal a critical decoupling between QA accuracy and true visual perception that scaling alone fails to bridge. To address this, we develop scalable, automated generation pipelines to create ST-Evidence-Instruct, a 160k-scale dataset bridging high-level reasoning with fine-grained grounding. Fine-tuning grounded Video LLMs on this data yields substantial gains over the corresponding size-matched UniPixel baselines (e.g., +27.2 t-mean and +13.8 JF on a 7B model), establishing a robust baseline for explainable, evidence-backed video understanding. Code and data are available at this https URL.
[CV-3] Beyond the Single Camera: Agent ic Multi-View Reasoning in Sports Video Understanding
链接: https://arxiv.org/abs/2607.11844
作者: Kerui Chen,Jinglu Wang,Xiaoyi Zhang,Yan Lu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Recent Multimodal Large Language Models (MLLMs) achieve strong performance on single-view video understanding benchmarks. However, sports videos involve dense occlusion, rapid motion, and complex interactions that are difficult to resolve from a single viewpoint. In practice, sports events are recorded from multiple camera angles, providing complementary evidence used by referees. Yet, no existing benchmark evaluates MLLMs on multi-view sports video understanding. To address this gap, we introduce SportMV-Bench, a comprehensive benchmark built from official match recordings, through a dedicated pipeline combining LLM-based generation, MLLM-based verification, and human filtering to ensure quality and consistency. SportMV-Bench containing 787 multi-view video bundles and 2592 question-answer pairs across three categories: Perception-Aware Recognition (PAR), Rule-aware Event Interpretation (REI), and Adjudicative Decision Reasoning(ADR). Our analysis shows that current MLLMs fail to effectively exploit multi-view information, with the bottlenecks lying in fine-grained visual perception and view selection rather than logical reasoning or domain knowledge. We propose SportMV-Agent, an agentic framework that orchestrates an iterative loop of active view selection, perception tool execution, and evidence-grounded reasoning, achieving a significant 14.46% relative improvement over the strongest MLLM baseline.
[CV-4] LoRA-Based Cascaded Multimodal Fusion for Action Recognition in Medical Training Environments
链接: https://arxiv.org/abs/2607.11839
作者: Divya Mereddy,Jeevan Beedareddy
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:This paper presents a cascaded Low-Rank Adaptation (LoRA)-based multimodal fusion framework for action and activity recognition in healthcare-oriented training environments. The proposed architecture combines parameter-efficient modality-specific adaptation with sequential fusion, enabling modalities to be integrated in stages without retraining previously learned components. Rather than assuming a fixed fusion structure, the framework first integrates more closely related modalities and then incorporates additional heterogeneous modalities, supporting scalable adaptation across datasets with different modality this http URL evaluate the framework on two healthcare-oriented training environment datasets: NurViD and the Nurse Training dataset. Across these datasets, preliminary results suggest that the proposed cascaded fusion strategy improves over individual modality models and provides competitive performance relative to previously reported dataset-specific baselines. Overall, these findings indicate that cascaded LoRA-based fusion is a promising parameter-efficient approach for integrating heterogeneous modalities in medical training action and activity recognition tasks. github: this https URL.
[CV-5] HASTE: A Platform for Rapid Post-Disaster Building Damage Assessment
链接: https://arxiv.org/abs/2607.11838
作者: Caleb Robinson,Anthony Ortiz,Simone Fobi Nsutezo,Cameron Birge,Meygha Machado,Marcelo Duarte,Joaquin Rivero Rodriguez,Anthony Cintron Roman,Kevin White,Inbal Becker-Reshef,Juan M. Lavista Ferres
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:When a large disaster strikes, responders need a map of which buildings are damaged within hours. The models that do well on public benchmarks assume matched before-and-after imagery and a training set drawn from similar past events, and neither is usually available for a new disaster in its first day. We present HASTE (High-speed Assessment and Satellite Tracking for Emergencies), a no-code web platform that lets analysts who are not machine learning engineers produce per-building damage maps from post-disaster satellite imagery. HASTE implements two methods that share one interface. The first requires the user to label polygons over the post-disaster scene, trains a small semantic segmentation model on that single scene, runs it over the whole image, and joins the per-pixel output to existing building footprints. The second embeds every footprint with a pretrained vision model, requires the user to label a handful of buildings, and fits a logistic regression in the browser that scores the rest of the scene in seconds. We describe the platform, both methods, and the engineering that supports them. We also report preliminary experiments on xBD showing that foundation-model embeddings pooled over footprints separate damaged from intact buildings using post-disaster imagery alone, matching a fully supervised ResNet-50 baseline with a twentieth of its labels. HASTE and its predecessors have supported more than thirty real-world disaster responses since 2023, spanning earthquakes, hurricanes, cyclones, floods, wildfires, and tornadoes, delivering results to humanitarian partners within hours to days of imagery becoming available. We close with the directions we think are most promising, including vision-language assessment, active learning, and damage models for roads and other infrastructure. HASTE is open source at this https URL.
[CV-6] Cycle-World: Mitigating Error Accumulation in Long-term Video World Models via Reverse-Prediction Cycle Consistency ECCV2026
链接: https://arxiv.org/abs/2607.11836
作者: Zihan Su,Teng Hu,Jiangning Zhang,Ruiyan Wang,Ran Yi,Lizhuang Ma,Dacheng Tao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by ECCV 2026
Abstract:Autoregressive diffusion models have enabled high-quality video generation, yet their sequential nature inherently suffers from error accumulation. In long-horizon video synthesis, minor prediction deviations compound over time, inevitably leading to unconstrained generative drift, structural collapse, and severe visual degradation. To address this, we propose Cycle-World, a novel framework designed for stable and temporally consistent long-video generation. Our approach tackles error drift by enforcing strict temporal reversibility across both the training and inference phases. Theoretically, we demonstrate that forward generative drift can be strictly bottlenecked by a cycle-consistency objective. During training, we integrate an efficient reverse-prediction model to implicitly embed causal constraints into the forward generator, compelling it to produce reversible sequences that tightly adhere to the natural video manifold. At inference time, we repurpose this frozen reverse model as a runtime corrector. Through gradient-based cycle guidance, it iteratively refines the generated latent representations, actively suppressing accumulated errors before they are committed to the historical context. Extensive experiments on the VBench benchmark demonstrate that Cycle-World’s dual-phase synergy significantly mitigates error drift, achieving state-of-the-art overall generation quality and long-horizon temporal consistency in 60-second synthesis.
[CV-7] MicroCharNet: Less is More for License Plate Character Detection
链接: https://arxiv.org/abs/2607.11830
作者: Huy Che,Dinh-Duy Phan,Duc-Lung Vu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:License plate character detection is a crucial component of intelligent transportation systems, where high accuracy and computational efficiency are required for real-time deployment. Although recent deep learning-based methods have substantially improved detection performance, many high-accuracy models rely on large-scale architectures that incur substantial computational overhead, limiting their applicability to resource-constrained devices. In this paper, we propose MicroCharNet, an ultra-lightweight model specifically designed for license plate character detection. The proposed architecture employs a compact backbone composed of C2f blocks, integrated with CoordAtt module to enhance feature extraction while preserving spatial information. A lightweight C3k2-based neck fuses multi-level features, followed by a single-level anchor-free detection head that enables end-to-end prediction. Experiments conducted on the UFPR-ALPR dataset demonstrate that MicroCharNet achieves competitive detection accuracy with only 0.08M parameters and 0.096 GFLOPs, while outperforming several recent YOLO-based baselines. Hardware-level evaluations further confirm its efficiency for real-time deployment on edge devices. These results indicate that carefully designed ultra-lightweight architectures can effectively balance accuracy and efficiency in license plate character detection. The source code is available at this https URL.
[CV-8] MM-ToolSandBox: A Unified Framework for Evaluating Visual Tool-Calling Agents
链接: https://arxiv.org/abs/2607.11818
作者: Kaixin Ma,Di Feng,Alexander Metz,Jiarui Lu,Eshan Verma,Afshin Dehghan
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Benchmark link: this https URL
Abstract:We introduce MM-ToolSandBox, a benchmark and evaluation framework for visually grounded tool-calling agents. The framework provides a stateful execution environment spanning 500+ tools across 16 application domains, supporting multi-image, multi-turn tasks where agents must ground progressively arriving visual inputs into executable tool calls while handling realistic conversational phenomena (goal revisions, error corrections, state mutations). An automated scenario generation pipeline produces diverse, visually grounded scenarios through information-flow-guided planning and multi-stage quality filtering, yielding 258 human-verified nominal scenarios and 50 variants targeting interactive UI applications. Evaluating 12 state-of-the-art models, from 4B open-weight to frontier proprietary systems, shows that current models still lack robust visual tool-calling capability: even the best model achieves below 50% success rate. Our failure analysis further reveals that visual precision, not only planning, is a primary bottleneck for capable models: 53% of failures stem from incorrect information extraction from images despite otherwise correct task workflows. A planning-to-precision crossover emerges with scale: smaller models fail at deciding what to do, while larger models fail at perceiving what they see, suggesting fundamentally different research directions for improving models at different capability levels. The framework and the benchmark are publicly available at this https URL
[CV-9] StoryTeller: Training-Free Narrative Grounding for Long-Form Audio Description ECCV
链接: https://arxiv.org/abs/2607.11798
作者: Seung Hyun Hahm,Minh T. Dinh,SouYoung Jin
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted to the European Conference on Computer Vision (ECCV) 2026
Abstract:Long-form audio description (AD) requires more than describing visible actions: it must preserve characters, events, relationships, and story context across scenes so that blind and low-vision (BLV) audiences can follow a film. Modern video-language models (VLMs) are effective on short clips, but they often treat each moment independently, producing descriptions that miss who characters are, why events matter, and how the current scene connects to earlier narrative context. We propose StoryTeller, a training-free framework for story-aware long-form AD. Instead of relying only on local visual cues, StoryTeller maintains a verified narrative memory that carries forward story-relevant information across scenes, enabling later descriptions to remain coherent, grounded, and contextually informative. Given only raw video and a movie title, StoryTeller can optionally retrieve public movie metadata to resolve names and story context, while accepting only facts that are supported by the video through semantic filtering and VLM verification. The method requires no subtitles, scripts, AD transcripts, aligned captions, character banks, precomputed face identities, or task-specific fine-tuning. To evaluate whether generated AD preserves narrative information, we introduce StoryAD-QA, a question-answering benchmark that tests whether a language model can answer story-context questions using only the generated descriptions. Experiments on standard AD benchmarks and diverse long-form videos show that StoryTeller consistently improves narrative coherence, factual grounding, and story comprehension over strong baselines in automatic, QA-based, and human evaluations.
[CV-10] Higher-Order Cell Tracking Transformer
链接: https://arxiv.org/abs/2607.11754
作者: Jordão Bragantini,Ilan Theodoro,Loïc A. Royer
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Reconstructing lineages from live-imaging microscopy requires linking cell detections across time, including through cell divisions. A common approach is to construct a candidate graph and associate cell segmentations (nodes) across frames. However, these and other existing methods overlook two structural obstacles in candidate tracking graphs: (i) cell divisions entangle distinct lineage paths in the node embedding space, and (ii) edges sharing a node have near-random label agreement, so the candidate-graph topology carries no useful information for graph neural networks to aggregate. We propose the \textbfHigher-Order Cell Tracking Transformer (HOCT), an edge-centric architecture in which candidate cell links attend to one another under a 3D geometric prior, resolving both issues. Evaluated on the Cell Tracking Challenge and a bacteria division benchmark, HOCT achieves state-of-the-art results without deep pre-trained image encoders. Moreover, the proposed approach is easier to fine-tune, quickly reducing tracking errors by 59% with 400 annotations in a human-in-the-loop setting, outperforming LoRA fine-tuning of competing transformer baselines (6.75% improvement).
[CV-11] NeuralActuator: Neural Actuation Modeling for Robot Dynamics and External Force Perception
链接: https://arxiv.org/abs/2607.11734
作者: Zhiyang Dou,John U. Onyemelukwe,Hangxing Zhang,Heng Zhang,Minghao Guo,Yunsheng Tian,Michal Piotr Lipiec,Joshua Jacob,Chao Liu,Peter Yichen Chen,Yuri Ivanov,Wojciech Matusik
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR); Machine Learning (cs.LG)
备注: Project Page: this https URL Code: this https URL
Abstract:Differentiable simulators have advanced policy learning and model-based control, yet actuator dynamics remain an important source of sim-to-real error. This is particularly acute on low-cost platforms, where the linear current-to-torque relation \tau = K_tI becomes unreliable during commanded-target tracking because of friction, hysteresis, backlash, and thermal effects. We present NeuralActuator, a neural actuator model that jointly predicts (i) a simulator-equivalent generalized-effort surrogate for trajectory propagation on low-cost servo platforms, (ii) external force with a contact-probability gate for sensorless force perception, and (iii) a motor-condition score for the supervised joint. We also introduce the Neural Actuation Dataset (NAD), collected with a twin-arm teleoperation system that records robot states and actuator telemetry together with external-force labels. The torque-surrogate head is trained through differentiable simulation from pose trajectories without direct generalized-effort labels, while the force, gate, and motor-condition heads receive direct supervision. A Transformer captures temporal dependencies while supporting real-time inference. We evaluate NeuralActuator on a 5-DoF OpenManipulator-X, a 6-DoF SO-101, and a 7-DoF Franka Emika Panda, spanning three actuator families and platforms costing approximately USD 500 to over USD 30,000. The low-cost platforms support dynamics and force evaluation, while the offline Franka experiment provides an additional payload-force-estimation benchmark. Experiments further demonstrate its application for motor condition estimation on OpenManipulator-X and improved behavior-cloning performance when NeuralActuator is used as a pretrained module.
[CV-12] GFR-SAM: Training-Free Referring Camouflaged Object Segmentation via Cross-Image Prompting
链接: https://arxiv.org/abs/2607.11732
作者: Yilong Yang,Jianxin Tian,Shengchuan Zhang,Liujuan Cao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Referring Camouflaged Object Detection (Ref-COD) requires segmenting hidden targets guided by reference cues. While supervised methods are annotation-heavy and training-free approaches via sparse point-prompting are sensitive to localization errors, we propose GFR-SAM, a robust three-stage training-free framework. GFR-SAM shifts the paradigm from fragile point-matching to a “Generate-Filter-Refine” pipeline. First, we introduce In-Context Exemplar-guided Segmentation, empowering SAM3 with cross-image inference to generate candidate masks via holistic visual exemplars, bypassing its native intra-image constraints. Second, a Region-Global Contrastive Filtering module ranks candidates through DINOv3-based prototypical alignment, effectively suppressing background distractors. Finally, a Geometric-Semantic Refinement module synergizes bounding box and text prompts to recover fine-grained boundaries and enhance instance recall. Evaluated on the R2C7K benchmark, GFR-SAM outperforms existing training-free methods by 8.7% in weighted F-measure ( F_\beta^w ) and competes with supervised state-of-the-art counterparts. Ultimately, this work underscores the potential of unlocking SAM3’s latent capability for cross-image In-Context prompting, establishing a robust, training-free paradigm that effectively bridges the gap between general-purpose foundation models and specialized, label-intensive perception tasks without the need for task-specific fine-tuning.
[CV-13] SVI360: Spherical Video Interpolation ECCV2026
链接: https://arxiv.org/abs/2607.11710
作者: Le-Kim Nguyen,Renato Martins,Pascal Vasseur,Cedric Demonceaux
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at ECCV 2026. Code and trained models are available at: this https URL
Abstract:This paper addresses the problem of omnidirectional video interpolation, which plays an essential role in applications such as virtual reality and immersive video enhancement. Existing video interpolation methods are not well-suited for spherical videos, as they have difficulty handling severe distortions close to the poles. To address this issue, we propose SVI360, a dual-branch framework that combines the image frame and its rotated orthogonal view to deal with these distortions. The core methodological aspect of the approach is to reinforce equivariance of the flow displacements between the original and orthogonal views to improve intermediate frame prediction. Experiments show that our method outperforms state-of-the-art approaches in interpolation quality while maintaining accurate optical flow in four different public benchmarks. Code and pre-trained models are available at: this https URL
[CV-14] Illuminant-Adaptive 3D Lookup Tables for Camera Color Correction
链接: https://arxiv.org/abs/2607.11681
作者: Claudio Rota,Luca Cogo,Simone Bianco,Raimondo Schettini
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Color correction is a key component of camera image signal processing (ISP) pipelines, encompassing illuminant discounting and colorimetric mapping of device-dependent sensor responses to device-independent color spaces, such as CIE XYZ. Despite extensive research, accurate color correction remains challenging due to the non-linear relationship between camera sensor responses and CIE XYZ color space, as well as to the increasing presence of highly chromatic and spectrally complex LED illuminants. We propose a color correction framework based on illuminant-adaptive three-dimensional lookup tables (LUTs), which we call Color Correction LUT (C ^2 LUT). Our method combines a chromaticity-aware illuminant representation with a non-linear color transformation, enabling accurate correction under illuminants spanning a wide range of chromaticities and spectral complexities. We employ Tucker tensor decomposition to represent the LUTs, ensuring that computational requirements remain sufficiently low for deployment in camera ISPs. In addition, we introduce a large-scale illuminants dataset comprising 1,473 spectral power distributions, with different chromaticities and spectral profiles. Experiments across multiple cameras, illuminants, reflectance datasets, and real captured images demonstrate consistent improvements over existing methods for color correction, reducing CIE \Delta E_00 by up to 20% and angular error by up to 18% while remaining compatible with modern camera hardware constraints. Code and datasets are available at this https URL.
[CV-15] ABot-3DWorld 0: A Universal World Model to Explore Any 3D Space
链接: https://arxiv.org/abs/2607.11673
作者: Mingchao Sun,Luyang Tang,Yu Liu,Xu Yan,Zhan Li,Yunwei Zhang,Fei Yu,Zengye Ge,Yumin Liu,Jiacheng Zhang,Yongchang Zhang,Jiawei Zhang,Zhicheng Liu,Zhongxu Sun,Tianjian Ouyang,Wenzheng Chen,Shixing Yang,Nianfei Fan,Guodong Sun,Huan Li,Zheng Zhou,Yongze Li,Yingliang Peng,Mengmeng Du,Yuan Liu,Haozhe Shi,Chunnuo Gong,Chengzhen Yu,Chunxue Jia,Yang Liu,Shiying Zeng,Junnan Lai,Hang Zhang,Ning Guo,Baoquan Chen,Mu Xu,Hongyu Pan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Official Page: this https URL
Abstract:We present ABot-3DWorld 0, a universal multimodal 3D world model that turns text, image, and video inputs into high-fidelity, explorable 3D worlds. At the heart of our framework is a unified Spatial Generative Primitive (SGP), a compact tuple of a high-quality panorama and a spatial point cloud that delivers an efficient description of any 3D space. Multimodal inputs are first lifted into this primitive; a 3D-consistent panoramic video generator then explores the primitive along a planned trajectory; finally, our panoramic video reconstruction engine converts the generated video into a clean, photorealistic 3D Gaussian Splatting (3DGS) world. This pipeline covers two regimes: rich inputs (multi-view sets, casual video) are lifted into the SGP through a geometry-rigorous recovery that mirrors the observed scene, while a single image or sentence is completed generatively into a creative world. The result is one low-barrier engine for general 3D content creation that further anchors generated worlds to geographic points of interest, enabling map-native spatial exploration at consumer scale. Experiments show that ABot-3DWorld 0 sets the state of the art among open-source methods and demonstrates stronger scene fidelity than Marble under rich multimodal inputs.
[CV-16] Feature-Space Guided Diffusion for Realistic Ultrasound Image Synthesis MICCAI2026
链接: https://arxiv.org/abs/2607.11655
作者: Marina Domínguez,Nélida Mirabet-Herranz,Valery Naranjo
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 11 pages, 4 figures. Pre-review manuscript version of a paper accepted at DGM4MICCAI 2026
Abstract:Conditional diffusion models can generate anatomically plausible medical ultrasound (US) images, but anatomical plausibility alone does not ensure realistic B-mode appearance. Most US pipelines adapt standard generative architectures and condition them on anatomical masks, or use guidance mechanisms that reinforce the same anatomical signal. However, B-mode US images are shaped by acquisition-dependent properties such as speckle texture, tissue contrast, and attenuation. Using a frozen US foundation model, we show that standard conditional diffusion baselines remain separated from real images in representation space. In this work, we propose Feature-Space Candidate Guidance (FSCG), a training-free sampling strategy to reduce this gap. At sampling time, FSCG applies local k-NN feature correction and selects the best of multiple stochastic candidates according to their feature-space energy. In this way, the mask defines the anatomy, while FSCG steers samples toward the real US domain. Across three different datasets, FSCG reduces average FID64 by 56%, FID192 by 57%, and nearest-neighbour feature distance by 47% over standard conditional diffusion sampling, outperforming alternative inference-time guidance baselines. The results suggest that domain-aware feature representations can reveal and reduce realism gaps in medical diffusion synthesis without retraining the generator. Our code is available at this https URL.
[CV-17] Event-RGB Adaptive Tracking for Nighttime Highway Perception
链接: https://arxiv.org/abs/2607.11646
作者: Haidong Wang,Hengxing Cai,Wanlei Li,Xiaogang Xiong,Renxin Zhong
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注:
Abstract:Intelligent Transportation Systems deployed on highways predominantly rely on conventional RGB cameras for traffic perception and vehicle tracking. However, highway environments present unique challenges: the absence of artificial lighting infrastructure, combined with high vehicle velocities, results in severely degraded perception performance under low-light conditions. Specifically, nighttime scenarios suffer from motion blur, insufficient exposure, and poor signal-to-noise ratios, which catastrophically impair the reliability of RGB-based sensing systems. To address these limitations, we propose a novel Joint Event-RGB Adaptive Tracking (JEAT) framework. Unlike existing multi-sensor trackers constrained by rigid, hard-coded prioritization, JEAT merges asynchronous event streams and RGB frames into a unified joint data association optimization. By employing an Adaptive Extended Kalman Filter to continuously estimate measurement noise via NIS statistics, the framework dynamically weights and fuses both modalities, optimally harnessing event streams during dark or high-speed motion while leveraging RGB frames under bright or static conditions. Furthermore, given the absence of publicly available datasets tailored for event-based highway perception with diverse environmental conditions, we present SEHN, a large-scale synthetic dataset generated using the CARLA simulator. Our dataset encompasses diverse environmental conditions (daytime, nighttime, nighttime with out artificial lighting) and varying traffic densities, providing synchronized RGB imagery and event streams to facilitate multi-modal fusion research. Our code and datasets will be available at this https URL.
[CV-18] Motion4Motion: Motion Transfer Across Subjects at Inference SIGGRAPH2026
链接: https://arxiv.org/abs/2607.11644
作者: Ling-Hao Chen,Zixin Yin,Duomin Wang,Xianfang Zeng,Gang Yu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: SIGGRAPH 2026
Abstract:This work explores the motion transfer from one video to another, which is crucial in animation for diverse characters. Previously, video motion transfer has been largely explored between human and human-like characters, enabling a lot of applications in digital creation. However, these approaches encounter a main limitation. Specifically, related technical pipelines heavily rely on a predefined human skeleton structure and accordingly require skeleton-conditional model training. On the one hand, these methods are difficult to generalize to diverse characters, such as animals from different species, while preserving their unique motion styles. On the other hand, labeled data in diverse skeletons is limited, which additionally restricts the large-scale training for the task. In this paper, we jump out of the skeleton-based motion transfer framework and propose a training-free motion transfer framework, named Motion4Motion. Motion4Motionmodels the motion flow of the character in a video instead of skeletons, which makes motion transfer across species easier. Extensive experimental results and novel applications show our methods outperform baselines impressively. Project page is available at this https URL.
[CV-19] Backbone-Agnostic Perturbation-Induced Uncertainty Learning for End-to-End Real-World Image Dehazing
链接: https://arxiv.org/abs/2607.11623
作者: Bingcai Wei
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Real-world paired image dehazing remains challenging because haze degradation is spatially non-uniform, illumination-dependent, and physically ambiguous even when haze-free references are available. Existing end-to-end restoration networks usually formulate dehazing as a deterministic mapping from a hazy observation to a clean target, leaving the uncertainty hidden in degraded features, haze priors, and cross-domain negative samples insufficiently explored. In this paper, we propose Backbone-Agnostic Perturbation-Induced Uncertainty Learning (BPUL), a plug-and-play uncertainty learning framework for end-to-end real-world image dehazing. BPUL first introduces a Learnable Perturbation-induced Uncertainty Modulator (LPUM) that estimates channel-wise and spatial-wise feature sensitivity through reparameterized stochastic perturbations. It then develops a Prior-informed Uncertainty-guided Reconstruction Module (PURM), which exploits transmission and atmospheric-light priors to reconstruct the hazy observation from the restored result and enforce degradation consistency. Furthermore, we propose a Dual-space Domain-diversified Distribution-aware Contrastive Loss ( D^3 CL) to regularize both clean restoration and hazy reconstruction spaces with real-world and synthetic negatives. Experiments on five real-world paired benchmarks show that BPUL consistently improves multiple representative backbones. Since only LPUM is retained during inference while PURM and D^3 CL are used as training-time constraints, BPUL brings substantial restoration gains with only marginal additional inference overhead.
[CV-20] Similarity-Guided Curriculum Fine-Tuning of LLM s for Neural Architecture Synthesis
链接: https://arxiv.org/abs/2607.11591
作者: Anujaya Vijayakumar,Radu Timofte,Dmitry Ignatov
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Introduce a MinHash-based similarity scheduling framework that constructs a progressive curriculum over neural architecture code for LLM-based neural architecture search (NAS). Using 128-permutation MinHash signatures over normalised 7-gram source code shingles, we partition the reference pool into similarity bands and present them in increasing architectural heterogeneity, with the best LoRA adapter from each stage merged cumulatively into the backbone. We evaluate the framework on OlympicCoder-7B within the LEMUR benchmark on CIFAR-10 image classification, generating N =15 candidate architectures per epoch across six progressive fine-tuning steps. The curriculum achieves 60% peak success rate at the high-similarity level without post-processing repair. A 2*2 ablation at the most diverse level curriculum versus base model, with versus without partial interface repair reveals that without repair the base model (47% peak SR) substantially outperforms the curriculum model (7% SR), while adding partial repair brings both to 53% SR. This pattern is consistent with merge-level weight drift progressively erasing evaluator-interface priors, and suggests that interface repair and curriculum scheduling target distinct failure modes. We further report a cross-dataset transfer observation on SVHN, where direct base-model generation without curriculum warmup yields 27% peak SR at substantially lower accuracy (60.5%) than the CIFAR-10 equivalent, consistent with the increased synthesis difficulty of the unq-family anchor architecture.
[CV-21] FoundationGeo: Learning Spatial Pixel-Wise Fields for Monocular Metric Geometry UAI ECCV2026
链接: https://arxiv.org/abs/2607.11588
作者: Muxin Liu(1 and 2),Xiaoyang Lyu(1),Tianhe Ren(1),Peng Dai(1),Xiaoshan Wu(1),Zhiyue Zhang(1),Jiaqi Zhang(1),Jiehong Lin(1),Shaoshuai Shi(2),Xiaojuan Qi(1) ((1) The University of Hong Kong, (2) Voyager Research, DiDi Chuxing)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 28 pages, 8 figures. Accepted to ECCV 2026. Muxin Liu and Xiaoyang Lyu contributed equally. Shaoshuai Shi and Xiaojuan Qi are corresponding authors. Project page: this https URL
Abstract:We present FoundationGeo, a two-stage framework that explicitly bridges relative and metric prediction via spatial calibration and principled data design. Stage 1 learns a high-fidelity, affine-invariant geometry model by initializing with DINOv3 and training on a curated 10.2M-sample multi-domain corpus with complementary local-detail supervision, yielding sharp boundaries and strong cross-domain generalization. Stage 2 moves beyond global scaling by introducing lightweight pixel-wise calibration fields for metric estimation: a scale field for spatially varying metric alignment and a ray-direction correction field that mitigates directional bias in point-map geometry, together producing metrically consistent 3D point maps. Beyond model design, we identify camera intrinsic coverage, especially focal length distribution mismatch between training and test data, as a key bottleneck for zero-shot metric generalization: performance drops sharply when test intrinsics fall outside the training distribution. To address this, we synthesize additional training data across diverse focal lengths using a Blender-based data engine, repairing under-covered focal regimes and improving robustness under intrinsic shift. Extensive zero-shot evaluations across seven benchmarks show that FoundationGeo significantly strengthens cross-domain robustness, staying near the top across diverse domains while avoiding the sharp cross-domain performance drops observed in other methods. This consistency translates into the best overall performance, surpassing heavier baselines by over 5.2% on average.
[CV-22] GB-SVFBP: Gaussian-Based Shift-Variant FBP neural network
链接: https://arxiv.org/abs/2607.11584
作者: Chengze Ye,Linda-Sophie Schneider,Yipeng Sun,Andreas Maier
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted for poster presentation at the 18th International Meeting on Fully Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine (Fully3D 2025), Shanghai, China, May 27-30, 2025
Abstract:This paper proposes a Gaussian-Based Shift-Variant filtered backprojection (FBP) neural network, which is designed for the efficient reconstruction of non-circular trajectory cone beam computed tomography. The traditional differentiable shift-variant FBP model consists of a filtering component and a backprojection process. The filtering component includes operations such as weightings, differentiations, a 2D Radon transform, and a 2D backprojection. The proposed methods build on this framework by introducing a trainable 2D Gaussian model to represent the trajectory-related part in the filtering process, achieving a substantial reduction in the number of trainable parameters. Experimental results demonstrate that the proposed model reduces the parameter count by 99%, while only sacrificing a slight amount of reconstruction quality. Furthermore, the training time for each trajectory is reduced to one-fourth of the original, significantly accelerating convergence. These enhancements demonstrate a considerable augmentation in the model’s practicality and effectiveness, making it a valuable asset for real-world applications.
[CV-23] Actor as Its Own Critic: Unifying Region Understanding and Localization via CycleGRPO ECCV2026
链接: https://arxiv.org/abs/2607.11581
作者: Xin Zhang,Haochen Wang,Yikang Zhou,Jason Li,Robby T. Tan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ECCV 2026
Abstract:This paper introduces Actor as Its Own Critic, a unified reinforcement learning framework, Cycle Group Relative Policy Optimization (CycleGRPO), that jointly optimizes region understanding and localization for Multimodal Large Language Models (MLLMs). Unlike existing separate pipelines, we leverage the inherent duality between the two tasks to construct a self-evaluating reinforcement learning paradigm: "region \to text \to region’'. Specifically, a single MLLM first acts as the actor to generate region captions, then immediately transitions to a critic to ground its generated text back in the spatial domain. Therefore, CycleGRPO requires only region inputs, e.g., masks or bounding boxes, entirely bypassing the need for textual ground truths. A quality-aware token-level cycle-consistency reward is employed to assess the semantic discriminability of text captions via their physical localization accuracy. Empirically, built upon SAMTok, our CycleGRPO framework successfully bootstraps both capabilities simultaneously. Without any task-specific fine-tuning, the framework yields consistent performance gains across a wide range of benchmarks, including region captioning, region VQA, grounded dialogue, and referring segmentation. Overall, CycleGRPO offers a straightforward and scalable way to advance pixel-level capabilities in MLLMs. Code and models are released at this https URL.
[CV-24] DiffEEG: A Self-Supervised Denoising Diffusion Model for Learning EEG Generic Representations
链接: https://arxiv.org/abs/2607.11578
作者: Abdulkader Helwan,Lina Abou-Abbas,Hussein El Amouri,Belkacem Chikhaoui,Khadidja Henni
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Signal Processing (eess.SP)
备注: 19 pages, 6 figures
Abstract:Deep learning for EEG-based seizure detection faces critical challenges: severe annotation scarcity and extreme class imbalance, where ictal events comprise less than 10% of clinical recordings. We present DiffEEG, a 9.6M-parameter self-supervised foundation model that addresses both limitations through denoising diffusion pre-training and reinforcement learning (RL)-based fine-tuning. Pre-trained on 1.3M unlabeled segments from the Temple University Hospital Seizure Corpus (TUHSZ), DiffEEG learns generic neural representations via a 1D U-Net with multi-head self-attention. For downstream adaptation, a reinforced decision layer employs policy gradient optimization to directly maximize F1-score, prioritizing sensitivity to rare seizure events over overall accuracy. Under strict patient-wise evaluation (279 patients, Leave-One-Fold-Out), DiffEEG achieves 61% accuracy and 59% F1 for 4-class seizure subtyping, and 81% accuracy with 85% weighted F1 for binary detection, maintaining clinically viable seizure recall (59%) despite extreme imbalance (6.7% prevalence). Segment-level evaluation establishes an upper bound of 97.6% accuracy, confirming strong architectural capacity. DiffEEG demonstrates that diffusion-based pre-training combined with metric-aware reinforcement learning enables clinically deployable seizure monitoring with minimal labeled data requirements.
[CV-25] MonkeyOCRv2: A Visual-Text Foundation Model for Document AI
链接: https://arxiv.org/abs/2607.11562
作者: Yuliang Liu,Zhang Li,Ziyang Zhang,Shuo Zhang,Qiang Liu,Jiajun Song,Zidun Guo,Xinhan Wang,Handong Zheng,Yang Liu,Dongliang Luo,Zhiyin Ma,Jiarui Zhang,Xiang Bai
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Mainstream visual encoders are pretrained on natural images and cannot be effectively applied to document images without document-oriented adaptation, as dense text and fine-grained character strokes demand character-level visual perception. We present MonkeyOCRv2, a visual-text pretrained model for document AI. First, we construct MonkeyDoc v2, to our knowledge the largest document-image pretraining corpus, comprising 113 million images spanning 17 languages. Second, we propose a pretraining strategy that jointly learns image-to-text generation and pixel-level document reconstruction: the former aligns visual representations with textual content, while the latter preserves character strokes and layout details. Extensive experiments are conducted on five representative document analysis tasks, including text recognition, formula recognition, text detection, document tampering detection, and overlapping text segmentation. Replacing the original encoders with MonkeyOCRv2 consistently improves performance across all five tasks. Finally, we validate its effectiveness as the vision encoder of multimodal large language models on the more challenging tasks of document parsing and document understanding. Kept frozen and paired with a lightweight language model, it yields a 0.7B document parsing model that sets a new open-source state-of-the-art on MDPBench, a recent benchmark spanning digital-born and photographed documents across 17 languages, surpassing the previous best 3B this http URL by 2.8% absolute with a vision encoder roughly 11 \times smaller. The frozen encoder also powers a document understanding model that outperforms counterparts built on CLIP, DINO, and SAM across eight benchmarks under identical training settings. These results suggest that document-oriented visual pretraining can serve as a foundation for document intelligence in its own right.
[CV-26] chnical Report on the CVPR 2026@AdvML Workshop Challenge
链接: https://arxiv.org/abs/2607.11560
作者: Tianyuan Zhang,Zonglei Jing,Jiangfan Liu,Ligong Zhang,Ke Ma,Chengzhi Sun,Xiaohai Xu,Zhirui Zhang,Qianqian Xu,Qingming Huang,Hanyu Fang,Junhua Liu,Zheng Wang,Xiaoliang Liu,Yuanbo Li,Shuai Gui,Bin Wang,Menghe Zheng,Jing Nie,Hanyang Meng,Zeyang Zhang,Xiang Zhang,Yongxuan Zhu,Rui Ding,Hainan Li,Yongkang Zhang,Zhilei Zhu,Xianglong Kong,Jin Hu,Zonghao Ying,Yisong Xiao,Lei Chen,Haotong Qin,Jiakai Wang,Aishan Liu,Ruikai Li,Julia Karbing,Yinpeng Dong,Zhenfei Yin,Shao Jing,Xia Hu,Jingyi Xu,Juntao Dai,Xinyun Chen,Vishal M. Patel,Xianglong Liu,Dawn Song,Alan Yuille,Philip H. S. Torr,Dacheng Tao
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Vision-language agents (VLAs) are increasingly used to interpret complex driving scenes and support safety-critical reasoning. This report presents the CVPR 2026@AdvML Workshop Challenge on adversarial multimodal attacks against autonomous-driving VLAs. Built on DriveLM-style multi-view visual question answering, the challenge represents each scene with six synchronized camera images and a structured collection of driving-related question-answer pairs. Participants generate adversarial images and suffix-only textual perturbations that induce model responses to deviate from reference answers while preserving image fidelity and limiting textual cost. The competition comprises two phases, with Phase II adding a hidden black-box model to assess transferability. We describe the task design, submission rules, evaluation protocol, and leaderboard results, and then examine five leading submissions for which technical reports were available. Across these reports, several recurring patterns emerge: image-side attacks are favored by the suffix penalty; scene-level, multi-view optimization is more effective than treating views in isolation; QA types and graph structure provide useful priors for allocating attack budget; feature-space objectives can improve black-box transfer; and typographic content embedded in camera images exposes a persistent vulnerability in driving VLAs. These findings provide a practical reference for future robustness evaluation and defense design in multimodal autonomous-driving systems.
[CV-27] Single-Teacher View Augmentation: Enhancing Knowledge Distillation with Student-Guided Perturbations
链接: https://arxiv.org/abs/2607.11557
作者: Xuyi Yu,Yaohua Liu,Ziming Song,Yinghai Zhao,Huipeng Zhang,Kuizhi Mei
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Knowledge distillation (KD) typically relies on the fixed perspective of a single teacher, limiting the diversity of supervisory signals. While multi-teacher distillation addresses this by aggregating knowledge from multiple models, it incurs prohibitive computational and storage costs. To balance efficiency and diversity, recent research has focused on generating virtual views from a single teacher. However, existing methods face a trade-off: random perturbation approaches offer efficiency but lack controlled diversity, while structured augmentation methods require multi-stage training and incur linear parameter growth. We observe that this trade-off stems from a common design choice: using the teacher’s strong but static features to generate views. Instead, we propose Shift-Augmented Knowledge Distillation (SAKD), a simple yet effective framework that leverages the student’s evolving features as a dynamic condition for perturbation generation. This shift in perspective enables single-stage training while producing adaptive, diverse views through a parameter-free cyclic shift. Extensive experiments on CIFAR-100 and ImageNet demonstrate that SAKD consistently outperforms random perturbation methods and achieves accuracy on par with two-stage approaches, while using significantly fewer parameters and eliminating pre-training requirements.
[CV-28] raining-Free Off-Screen Player Imputation for Broadcast-Based Spatial Football Analytics
链接: https://arxiv.org/abs/2607.11548
作者: Seongjin Choi
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 10 pages, 2 figures, 3 tables. Code and data: this https URL
Abstract:Spatial football metrics such as pitch control assume access to the positions of all 22 players, yet the most widely available source of positional data – the broadcast main camera – shows only 10-16 of them at any moment. We quantify the resulting distortion with an open, reproducible benchmark: a simulated broadcast viewport applied to open full-pitch tracking data (Metrica Sports; three matches, one held out from method development). Ignoring off-screen players – the visible-only baseline implied whenever a video-based game-state-reconstruction (GSR) pipeline adds no imputation layer – inflates hidden-zone pitch-control error to 25.1-26.9 percentage points and a mean absolute control-share error of 11.1-13.4 points across the three matches. We then evaluate a ladder of training-free, online imputation baselines that use only observations from the match being analysed. The best overall on these decision-relevant metrics, role-anchored centroid voting (each visible player votes for the full-team centroid by subtracting its running role offset, attenuating the viewport-induced subset bias), roughly halves hidden-zone error (to 12.2-13.8 points) and cuts control-share error to 28-48% of the ignore policy at every viewport width from 36 m to 60 m in all three matches. For occlusions =9.6 s – the regime of the closest learned prior work – it reaches binwise median position errors of 3.3-8.9 m; but 50-57% of hidden-player observations lie beyond that regime. Integrated end-to-end into a broadcast-video GSR pipeline, imputation moves a downstream possession-quality score (Space-Creation Index) by 15.6 and 17.2 points on two real World Cup broadcast windows, flipping the verdict class in one.
[CV-29] Adaptive Routing for Efficient Diffusion Transformer-Based PNI Prediction
链接: https://arxiv.org/abs/2607.11533
作者: Youngung Han,Dohyun Kweon,Kyeonghun Kim,Hyunsu Go,Jina Jeong,Suah Park,Induk Um,Junga Kim,Anna Jung,Yului Jeong,Sungha Park,Jinyong Jun,Pa Hong,Woo Kyoung Jeong,Won Jae Lee,Ken Ying-Kai Liao,Hyuk-Jae Lee,Nam-Joon Kim
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Perineural invasion (PNI) is a critical prognostic factor in cholangiocarcinoma. However, its preoperative prediction from magnetic resonance imaging (MRI) remains challenging due to subtle imaging features that extend beyond tumor boundaries into surrounding regions. Conventional convolutional neural networks are limited in capturing long-range spatial dependencies. Transformer-based architectures improve global modeling of volumetric MRI by aggregating spatially distributed contextual cues, yet capturing subtle and noise-sensitive patterns in peritumoral regions remains challenging. Diffusion-based classifiers offer an alternative formulation by leveraging denoising-based class scoring to better capture such subtle patterns. However, these approaches introduce substantial computational overhead due to the combination of transformer-based modeling and iterative denoising processes. To address these challenges, we formulate PNI prediction as a diffusion-based classification problem and implement the denoising network using a transformer-based representation. To improve computational efficiency, we introduce adaptive routing across attention heads, spatial tokens, and MLP width. Experimental results demonstrate that the proposed approach achieves an AUC of 0.731 with 257.57 GFLOPs.
[CV-30] Parse Search and Confirmation: Training-Free Aerial Vision-and-Dialog Navigation with Chain-of-Thought Reasoning and Structured Spatial Memory CVPR2026
链接: https://arxiv.org/abs/2607.11529
作者: Yu Qi,Hongyu Li,Shaofei Huang,Tianrui Hui,Yaxiong Wang,Lechao Cheng,Zhun Zhong,Si Liu,Meng Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 10 pages, 4 figures. Accepted to CVPR 2026
Abstract:In this paper, we tackle the Aerial Vision-and-Dialog Navigation (AVDN) task in the training-free setting for resource-efficient high-altitude UAV this http URL applying MLLMs leads to unreliable navigation due to weak directional grounding and the lack of explicit spatial this http URL address these issues, we propose PSC-AVDN, a training-free framework that tightly couples a three-stage Parsing-Search-Confirmation reasoning pipeline with a Structured Spatial Memory (SSM).The parsing stage uses an LLM to convert ambiguous dialogue instructions into stable geometric directional and destination cues.A Search Chain-of-Thought (S-CoT) then performs stepwise target exploration under high-altitude observations, and a Confirmation Chain-of-Thought (C-CoT) conducts fine-grained verification around candidate regions to resolve visual this http URL, SSM integrates three complementary sources of spatial cues, including multi-scale visual observation, spatial visual memory, and structured geometric memory to provide global spatial context and long-horizon this http URL experiments on ANDH and ANDH-Full show that PSC-AVDN establishes new state-of-the-art performance in the training-free setting, matching or surpassing several finetuned this http URL will be publicly available at: this https URL
[CV-31] Vinci2: Providing Proactive Assistance in Continuous Egocentric Videos ECCV2026
链接: https://arxiv.org/abs/2607.11523
作者: Gong Sitong,Tianyu Yan,Caixin Kang,Bo Zheng,Xiang Ruan,Huchuan Lu,Kaipeng Zhang,Yoichi Sato,Yifei Huang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted by ECCV 2026
Abstract:When should an intelligent assistant speak up without being asked? Continuous egocentric video offers rich, evolving context that enables a new form of assistance: one that is proactive rather than merely reactive. Yet existing approaches either wait passively for user queries or treat every detected event as requiring a response, without considering the user’s history, current activity, or whether assistance would actually be welcome. We reframe proactive assistance as a context-dependent decision problem: the agent must not only perceive what is happening, but reason over accumulated temporal context to determine when and whether to intervene. To this end, we present Vinci2, a proactive egocentric assistance system that advances the on-device assistant Vinci from reactive response toward proactivity. On the evaluation side, we present EgoServe, the first large-scale benchmark for proactive assistance in continuous egocentric video. EgoServe comprises over 3,000 service instances organized along 4 temporal memory horizons, ranging from immediate safety alerts to long-term habit coaching, across 10 service categories. On the modeling side, we propose EgoMemo, a training-free, memory-augmented agent that maintains three complementary memory representations: multi-scale temporal summaries, a semantic knowledge graph, and visual embedding archives. At each timestep, EgoMemo performs retrieval-augmented reasoning to determine whether assistance is warranted and, if so, produces contextually grounded responses. Experiments demonstrate that EgoMemo establishes strong baselines on EgoServe while remaining competitive on existing egocentric benchmarks. Our benchmark and code are publicly available at \hrefthis https URLVinci2.
[CV-32] CFR-Net:Collaborative Feature Refnement Network for Medical Image Anomaly Detection
链接: https://arxiv.org/abs/2607.11509
作者: Zihan Nie,Muhao Xu,Wei Feng,Yuan Cui,Hua Wei,Sijie Niu,Yi Wan,Xunbin Wei,Weiye Song,Zongyuan Ge
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Medical image anomaly detection remains challenging because networks pretrained on natural images often exhibit limited adaptability to medical images, where abnormal patterns appear as fine-grained local shifts, multi-scale contextual mismatches, and orientation-sensitive structural deviations. To address this, we propose the Collaborative Feature Refinement Network (CFR-Net), which combines shared teacher-student feature refinement before decoding with cross-space consistency after decoding. CFR-Net refines frozen teacher features and trainable student features using a Multi-Path Feature Refinement Module (MPFRM) with shared parameters, imposing common multi-path refinement rules on generic visual references and representations adapted to the medical domain, thereby mitigating domain discrepancy while modeling local, multi-scale, and orientation-sensitive feature characteristics. A variance-sensitive objective and dynamic ``homework set’’ reorganization further support layer-adaptive consistency learning. Experiments on medical benchmarks show that CFR-Net achieves competitive anomaly classification and strong anomaly localization performance when trained on normal data.
[CV-33] HyperGS: Fast and Generalizable Gaussian Video Representation
链接: https://arxiv.org/abs/2607.11500
作者: Fatimah Zohra,Chen Zhao,Shuming Liu,Yahya Al Malallah,Bernard Ghanem
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Gaussian Splatting has emerged as an effective representation for video, but existing methods rely on per-video optimization. This leads to slow encoding and limits generalization across videos. To amortize this optimization, we propose HyperGS, a feedforward, optimization-free approach that directly predicts Gaussian representations from any video in a single forward pass, speeding up encoding and decoding by orders of magnitude while generalizing to out-of-distribution videos at higher resolutions. In HyperGS, we design a factorized spatiotemporal Transformer to extract tokens from video, and a learnable query-based Transformer to obtain 8-parameter Gaussian representations for each video frame. We find that naively predicting Gaussians across diverse videos induces a needle-like degeneration that collapses training, and address this with a rank-based geometric regularizer whose strength adapts dynamically to stabilize optimization. HyperGS achieves encoding at 10^4 – 10^5\times the speed of per-video Gaussian optimization at matched reconstruction quality while generalizing zero-shot to 720p video, enabling higher-resolution rendering without re-encoding. HyperGS improves PSNR by +2.9–3.1 dB over the prior video encoders on K400, SSv2, and UCF101 at a smaller video representation size. By predicting explicit 2D Gaussians in a single forward pass, HyperGS combines the fast, flexible rendering of Gaussian Splatting with the speed and generalization of feedforward prediction, advancing Gaussians as a practical direction for fast and generalizable video representation.
[CV-34] owards Efficient Convolutional Neural Network for Embedded Hardware via Multi-Dimensional Pruning
链接: https://arxiv.org/abs/2607.11473
作者: Hao Kong,Di Liu,Xiangzhong Luo,Shuo Huai,Ravi Subramaniam,Christian Makaya,Qian Lin,Weichen Liu
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Author’s accepted version. Published in Proceedings of the 60th ACM/IEEE Design Automation Conference (DAC 2023)
Abstract:In this paper, we propose TECO, a multi-dimensional pruning framework to collaboratively prune the three dimensions (depth, width, and resolution) of convolutional neural networks (CNNs) for better execution efficiency on embedded hardware. In TECO, we first introduce a two-stage importance evaluation framework, which efficiently and comprehensively evaluates each pruning unit according to both the local importance inside each dimension and the global importance across different dimensions. Based on the evaluation framework, we present a heuristic pruning algorithm to progressively prune the three dimensions of CNNs towards the optimal trade-off between accuracy and efficiency. Experiments on multiple benchmarks validate the advantages of TECO over existing state-of-the-art (SOTA) approaches. The code and pre-trained models are available at this https URL.
[CV-35] Video Transformer for Remote Identity Document Hologram Detection
链接: https://arxiv.org/abs/2607.11419
作者: Joris Voerman,Nicolas Sidere,Jean-Christophe Burie
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to SMC2026 conference, 6 pages, 5 figures, 5 tables
Abstract:Remote identity authentification using Identification Documents has been a major challenge for several years. DeepFakes advent and the development of AI-guided tools helps fraudsters creating counterfeit ID Documents. Ensuring the authenticity of ID Documents has become a real clue in the seurization of remote authentification. This need is all the more pressing given the increasing digitization of administrative and transactional processes. To ensure widespread accessibility, the system should rely solely on video captured via mobile devices. In this specific context, confirming the authenticity of ID is a real challenge as many security features needs specific device like infrared sensor for instance. Among underutilized but promising security features, holographic printings hold a special place. Difficult to counterfeit, they produce distinctive visual effects according enlightment, making them both detectable in a video captured by a smartphone camera and difficult to imitate. In this paper, we propose a Remote Identity Document Verification System (RIDVS) and an approach based on a video transformer for detecting holograms in simple videos captured by smartphones. Our system is designed for a smartphone-based capture process, followed by a server-side verification. The hologram detection method builds on a robust model previously validated in a related research domain. We demonstrate that it outperforms existing SotA methods, achieving near-perfect accuracy even when trained on medium- to small-sized datasets. In particular, we report improvements of +26.86% in Recall and +17.93% in accuracy over the best MIDV-Holo baseline. This study includes several experiments that evaluate the model adaptation to frugality, both for training samples and computational resources.
[CV-36] Uncertainty Quantification for EO Regression Tasks: Building Height Tree Canopy Height and Above-ground Biomass Estimation
链接: https://arxiv.org/abs/2607.11412
作者: Ritu Yadav,Andrea Nascetti,Yifang Ban
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Earth Observation regression tasks such as building height, canopy height, and above-ground biomass estimation underpin critical applications in urban planning, forest monitoring, and climate policy, where both accuracy and reliability are critical. Yet most deep learning models yield only deterministic predictions, providing no indication of per-pixel reliability. These regression tasks are inherently challenging due to heterogeneous land surfaces, skewed target distributions, sensor noise, and signal saturation at high target values, making uncertainty (UC) estimation essential for reliable inference. We address this gap by modeling aleatoric uncertainty using year-long Sentinel-1 SAR and Sentinel-2 MSI time series, proposing two complementary approaches: (i) Gaussian UC, which jointly predicts mean and standard deviation under a Gaussian assumption, and (ii) Quantile UC, which estimates the 10th, 50th, and 90th quantiles to capture asymmetric and heteroscedastic error distributions. Both models are evaluated on three representative EO regression tasks at 10 m spatial resolution. Results show that both approaches match or surpass deterministic benchmarks and existing global products, while delivering well-calibrated, interpretable, and operationally useful confidence estimates. Notably, both models outperform the current 10 m state-of-the-art uncertainty-aware model for canopy height estimation. Our implementation will be available at: this https URL
[CV-37] Self-supervised training for high-resolution close-range multispectral remote sensing imagery
链接: https://arxiv.org/abs/2607.11366
作者: Leon-Friedrich Thomas,Mikael Änäkkälä,Antti Lajunen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Although self-supervised learning (SSL) offers a promising way to reduce annotation effort in close-range remote sensing, its effectiveness for high-resolution multispectral unmanned aerial vehicle (UAV) imagery remains underexplored due to limited data. This study evaluated SSL pretraining for precision agriculture using cm-scale multispectral drone imagery collected across multiple sensors, years, and regions. Transformer-based encoders were pretrained with Momentum Contrast v3 (MoCo-v3) and Masked Autoencoders on a harmonized dataset combining msuav500K with newly collected multi-year UAV imagery from agricultural fields in Finland. Pretraining used four spectral bands (Green, Red, Red-Edge, Near-Infrared) for cross-sensor compatibility. The models were evaluated on crop-weed semantic segmentation using the WeedMap dataset with 5–100% training data. The following two subsets served as downstream tasks: Task A (Germany, RedEdge-M), where all pretrained models were compared under partial and full fine-tuning, and Task B (Switzerland, Sequoia), where the best encoder from Task A was assessed. Our Swin Transformer pretrained with MoCo-v3 achieved the strongest performance on both tasks, surpassing the Swin Transformer model of Doornbos et al. pretrained on a pre-release of msuav500K. Our pretrained Swin Transformer further demonstrated cross-sensor and cross-region generalization. We additionally provide a public multi-year multispectral UAV dataset from Finland to support future research.
[CV-38] Efficient Tuning Before Low-Bit Post-Training Quantization for Stochastic Gradient Descent-optimized Models
链接: https://arxiv.org/abs/2607.11359
作者: Peng Xia,Junbiao Pang,Muhammad Ayub Sabir
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 12 pages, 10 figures, submitted to IEEE Transactions on Neural Networks and Learning Systems (TNNLS). Code available at this https URL
Abstract:Post-training quantization (PTQ) compresses deep neural networks for deployment under limited memory and computational budgets. However, low-bit (i.e., 2-bit or 4-bit) PTQ often suffers from substantial performance degradation. Most existing PTQ methods operate on an unconstrained full-precision (FP) model and primarily address quantization errors through post-hoc reconstruction. We argue that low-bit PTQ accuracy is limited not only by post-quantization error minimization, but also by the quantization-error tolerance of a FP model itself. In this paper, we propose Efficient Tuning Before Quantization (ETBQ), a pre-conditioning tuning stage for Stochastic Gradient Descent (SGD)-optimized models before PTQ. During tuning, the FP model is optimized under perturbations sampled from the error distributions of weight and activation quantization, guiding the model toward a loss-landscape region that is less sensitive to the subsequent PTQ. Unlike QAT, ETBQ does not train a fake-quantized deployment model, which is computationally and memory intensive. Instead, ETBQ outputs a FP model that can be used by any PTQ backend. Experiments on CIFAR-100, Tiny-ImageNet, ImageNet, and Cityscapes provide consistent evidence that ETBQ improves low-bit PTQ across diverse tasks. Under W2A4 settings, e.g., ETBQ improves over naive PTQ by 2.14% top-1 accuracy on Tiny-ImageNet and by 5.80% mIoU on Cityscapes. Code is available at this https URL.
[CV-39] Benchmarking Edge Inference Strategies for Deep Learning Models in Industrial Machine Vision
链接: https://arxiv.org/abs/2607.11356
作者: Miguel Gomez Fernandez,David Castro Boga,Roi Mendez-Rial,Eric Lopez-Lopez
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 6 pages, 6 figures, accepted at IEEE COINS 2026. Preprint version; camera-ready version to appear in the conference proceedings
Abstract:Edge deployment is often the preferred solution for industrial machine vision systems when low latency, data security, or limited connectivity are critical requirements. Several frameworks are available to optimise inference on edge devices; however, relatively few studies have systematically compared their inference-time performance under industrial deployment conditions. In this work, we present a comparative study of four widely used approaches for machine vision inference in industrial settings: plain PyTorch, ONNX Runtime, OpenVINO, and TensorRT. The evaluation focuses on inference time, covers several CPU- and GPU-based hardware platforms, and includes both conventional convolutional neural networks and a transformer-based vision model. For the hardware platforms and models evaluated, the results show that OpenVINO achieves the lowest inference time on CPUs, while TensorRT achieves the lowest inference time on GPUs. However, TensorRT does not outperform plain PyTorch for the transformer-based model considered in this study. Comments: 6 pages, 6 figures, accepted at IEEE COINS 2026. Preprint version; camera-ready version to appear in the conference proceedings Subjects: Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2607.11356 [cs.CV] (or arXiv:2607.11356v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2607.11356 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-40] Longitudinal Multi-View Breast Cancer Risk Prediction MICCAI2026
链接: https://arxiv.org/abs/2607.11343
作者: Solveig Thrun,Zijun Sun,Suaiba A. Salahuddin,Kristoffer Wickstrøm,Elisabeth Wetzer,Stine Hansen,Robert Jenssen,Michael Kampffmeyer
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted at MICCAI 2026
Abstract:Accurate breast cancer risk prediction from screening mammography is critical for enabling personalized screening intervals and early detection. Recent deep learning methods have shown the value of longitudinal data and explicit temporal alignment. However, existing approaches either perform explicit alignment using a single mammographic view or model multiple views without explicit longitudinal alignment, limiting their ability to exploit the complementary spatial-temporal information used in clinical practice. To address this gap, we propose LMV-Net, a longitudinal multi-view breast cancer risk prediction model that jointly analyzes anatomically complementary CC and MLO views within an explicitly aligned longitudinal framework. We evaluate our approach on the public EMBED and CSAW-CC datasets, comparing it to state-of-the-art breast cancer risk prediction methods. Our model consistently outperforms existing approaches in overall risk prediction performance and across different breast density and cancer subgroups. Importantly, these improvements highlight the potential of longitudinal multi-view modeling to enhance risk stratification, paving the way for future work on personalized screening, earlier identification of high-risk patients, and more efficient screening resource allocation. The code is available at this https URL.
[CV-41] HierCAD: Hierarchical Text-to-CAD Design via Structure Alignment and Parameter Grounding
链接: https://arxiv.org/abs/2607.11339
作者: Jimin Xu,Tianbao Wang,Tao Jin,Zhou Zhao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Recent text-to-CAD approaches have shown promising results by leveraging large language models, but they often struggle with maintaining structural consistency in complex designs and accurately grounding geometric parameters. To address these issues, we propose HierCAD, a hierarchical text-to-CAD framework that improves both structural reasoning and parameter prediction. HierCAD reformulates CAD generation as progressive reasoning by decomposing CAD construction trees into object-level procedural reasoning and part-level topology reasoning trajectories. To further improve generation fidelity, we introduce a unified Structure Alignment and Parameter Grounding (SAPG) learning strategy. Structure alignment aligns topology reasoning trajectories with their corresponding parametric CAD spans, while parameter grounding mitigates shortcut learning through structure-preserving parameter perturbations and ranking-based supervision. Experiments demonstrate that HierCAD outperforms prior state-of-the-art methods on both CAD sequence generation and reconstructed CAD model evaluation. Our code is available at this https URL.
[CV-42] SLVMBench: Skill Learning from Video Memory
链接: https://arxiv.org/abs/2607.11312
作者: Yudong Yang,Guangzhi Sun,Yixuan Li,Chao Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:We introduce Skill Learning from Video Memory (SLVMBench), the first benchmark that jointly evaluates whether video large language models (video-LLMs) can learn skills from long video memory and apply them to real-time tasks. SLVMBench presents models with 2-3 hour video streams that contain a tutorial video embedded in a stream of arbitrary irrelevant videos, resembling real-world human learning practices. Video-LLMs are asked to apply the acquired skill to answer real-time questions about an ongoing video. Unlike long-video understanding benchmarks that emphasize passive comprehension and skill-learning benchmarks that rely on short, immediate demonstrations, SLVMBench tests the full pipeline of memorizing and extracting procedural knowledge, as well as transferring it to real-time tasks. Moreover, rigorous human annotations feature sub-second-level temporal calibration, manually engineered questions eliminating common-sense guessing, and collated tutorials to ensure coverage of the required skills. Evaluations on state-of-the-art proprietary and open-source video LLMs show that video-LLMs struggle substantially with learning and applying skill knowledge from videos. Moreover, performance degrades markedly when the skill knowledge is placed within a long video memory. These results reveal a key limitation of existing video LLMs and position SLVMBench as the first benchmark for studying real-time skill acquisition and application from long-context video memory.
[CV-43] ASUMOT: Motion-Consistency-Based Asynchronous UAV Detection and Tracking with Event Cameras
链接: https://arxiv.org/abs/2607.11303
作者: Baofeng Jia,Xiaoyu Chen,Jingyuan Zhang,Zongze Wu,Haochen li,Jing Han,Lianfa Bai
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Event cameras offer microsecond-level temporal resolution and high dynamic range for low-altitude UAV perception. However, long-range UAVs often produce sparse, fragmented, and noise-contaminated event responses, where one semantic target may appear as multiple spatially separated blobs. Direct blob-level asynchronous tracking therefore suffers from duplicate trajectories and unstable identities. We propose ASUMOT, a motion-consistency-based asynchronous UAV detection and tracking framework operating directly on raw events. ASUMOT models each UAV as a set of motion-consistent event blobs. A local motion-consistency estimator triggers reliable candidates, a lightweight multi-task verifier provides UAV confidence and motion-direction cues, and motion-consistency clustering aggregates fragmented blobs into identity-consistent UAV tracks. We also introduce ES-UAV, a high-definition event-level UAV benchmark with dense semantic annotations. Experiments on public UAV tracking data and ES-UAV show that ASUMOT improves the accuracy–efficiency trade-off while preserving asynchronous event processing. Code and Dataset will be released.
[CV-44] Metadata Supervised MRI Representations for Modelling and Controlling Acquisition Variability
链接: https://arxiv.org/abs/2607.11295
作者: Mehmet Yigit Avci,Pedro Borges,Virginia Fernandez,Natalia Glazman,Paul Wright,Mehmet Yigitsoy,Sebastien Ourselin,Jorge Cardoso
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Magnetic resonance imaging exhibits substantial acquisition variability, where identical anatomy can appear markedly different across scanners and imaging protocols. Consequently, learned representations entangle biological structure with acquisition-dependent appearance, limiting interpretability, generalisation, and clinical deployment. We show that these sources of variation can be separated by jointly modelling MRI images and DICOM metadata. Using large-scale clinical brain MRI data, we learn representations that separate anatomical structure from contrast-dependent appearance. Resulting contrast representations organise heterogeneous acquisitions, support sequence understanding, and detect image–metadata inconsistencies, whereas anatomical representations suppress acquisition-specific variation while preserving biologically relevant information. Building on these disentangled representations, we introduce a unified anatomy-preserving harmonisation model for cross-modality and cross-site adaptation, conditioned on image or acquisition metadata. Our findings suggest that acquisition variability is a structured component of the imaging process that can be modelled, audited, and controlled, providing a foundation for acquisition-aware representation learning in large-scale medical imaging.
[CV-45] A Unified Framework for Comprehensive Cardiac CT Segmentation and Phenotyping: Human-in-the-Loop Data Annotation Vision Foundation Model Development Multicenter Evaluation and Clinical Validation
链接: https://arxiv.org/abs/2607.11287
作者: Pooya Mohammadi Kazaj,Leo Fridolin Weber,Wen Xie,Seyed Amir Ahmad Safavi-Naini,Anselm Stark,Giovanni Baj,Ali Mokhtari,Toshiya Yoshida,Christoph Ryffel,Taishi Okuno,Yoshihiro Akashi,Ronny R. Buechel,Thomas Pilgrim,Waldo Valenzuela,George C. M. Siontis,Xiaowei Xu,Moritz Hundertmark,Stephan Windecker,Christoph Grani,Isaac Shiri
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Comprehensive quantification of cardiac structures from computed tomography (CT) remains limited not by data availability but by the scalability of measurements, which makes routine use impractical. Here we present a unified framework for comprehensive cardiac CT segmentation and phenotyping that combines a human-in-the-loop annotation pipeline, a cardiac CT augmentation technique, and a self-supervised foundation model pre-trained on 60,000 unlabeled cardiac CT scans. Using this approach, we assembled the largest and most comprehensive expert-annotated cardiac CT segmentation dataset to date, comprising 1598 cases and 14 distinct cardiac structures (1000 for training, 598 for the external test set). Across five external datasets, the framework segmented all structures more accurately and comprehensively than existing open-source tools. Self-supervised pre-training improved labeling efficiency, with the most significant gains observed during external evaluation in the low-data regime. Benchmarking across convolutional, transformer, and state-space architectures showed comparable performance, indicating that data quality and pre-training, rather than architecture, drove accuracy. The framework was scaled to population-level phenotyping, with segmented anatomy that carries functionally relevant information about ventricular function and disease severity beyond demographic variables. By openly releasing the largest dataset with human labels, code, model weights, a CT augmentation library, and software, this work provides a reproducible foundation for opportunistic cardiac phenotyping from routinely acquired CT scans.
[CV-46] SalientGS: Unified SfM-to-3DGS with Importance-Guided MCMC Gaussian Allocation
链接: https://arxiv.org/abs/2607.11285
作者: Tianyu Xiong,Rui Li,Suning Ge,Jiaqi Yang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted
Abstract:Reconstructing 3D scenes from unordered images remains bottlenecked by expensive Structure-from-Motion (SfM) preprocessing and frozen pose interfaces. We present SalientGS, a unified SfM-to-3D Gaussian Splatting (3DGS) pipeline. Its central contribution is importance-guided Markov Chain Monte Carlo (MCMC) Gaussian allocation, which aggregates multi-view residuals into per-Gaussian underfit and redundancy signals. These signals define a smooth importance-weighted sampling distribution that biases both birth and relocation toward underfit regions. This reallocates capacity from well-fit areas without altering the underlying stochastic gradient Langevin dynamics (SGLD). SalientGS achieves end-to-end reconstruction in 15 minutes with state-of-the-art perceptual quality. The supplementary material provides dedicated sections for Per-Scene Qualitative Comparisons and Per-Image Learned Perceptual Image Patch Similarity (LPIPS) Analysis, including failure cases. Code and evaluation scripts are available at this https URL.
[CV-47] he Devil Is in the Leakage: A Disentangled Dual-Purification Framework for High-Fidelity Hairstyle Transfer ACM-MM2026
链接: https://arxiv.org/abs/2607.11281
作者: Jijie Li,Jiankuo Zhao,Xiangyu Zhu,Zhen Lei
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by ACM MM 2026
Abstract:Hairstyle transfer aims to synthesize a photorealistic portrait by transplanting the hairstyle from a reference image onto a source subject while preserving the source identity. Recent foundation models show strong generative capability, but they struggle with the zero-shot disentanglement required for precise local editing, often entangling the reference hairstyle with its original identity and pose. Existing diffusion-based pipelines typically decompose the task by first generating a “bald” image from the source and then injecting hairstyle features from the reference. However, we show that this paradigm suffers from a fundamental leakage problem. Identity Leakage in Hairstyle occurs when hairstyle features retain reference identity or pose information, while Flaw Leakage in Bald arises when residual artifacts in the bald image are propagated into the final synthesis. To address both issues, we propose the Dual-Purification Framework (DPF), which introduces two complementary training-time regularizers. Adversarial Hairstyle Purification (AHP) purifies hairstyle features by suppressing identity predictability under a mutual-information-inspired adversarial objective. Contrastive Geometric Purification (CGP) regularizes the ControlNet pathway with a contrastive objective, reducing the model’s reliance on geometric artifacts in the bald condition. By jointly purifying the hairstyle representation and geometric pathway, DPF achieves high-fidelity, identity-preserving hairstyle transfer and state-of-the-art performance on diverse benchmarks.
[CV-48] LaGuadia: Language-Guided Adaptive Distillation from Pathology Foundation Models
链接: https://arxiv.org/abs/2607.11257
作者: Gangsu Kim,Won-Ki Jeong
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Pathology Foundation Models (PFMs) offer powerful Whole Slide Image (WSI) representations but suffer from massive computational costs. While Knowledge Distillation (KD) can create efficient student models, existing multi-teacher methods often use suboptimal uniform weighting that ignores tissue heterogeneity. We propose LaGuadia (Language-Guided Adaptive DistillAtion), a framework that develops a compact pathology image encoder by dynamically integrating expertise from multiple PFMs under clinical linguistic guidance. Our approach utilizes a multi-stage pipeline: first, extracting visually observable clinical keywords from pathology reports; second, aligning visual features with these keywords via a Vision-Language meta-teacher (MedSigLIP) to provide dense semantic guidance; and finally, performing adaptive KD where teacher contributions are weighted based on their semantic alignment with the clinical narrative. Experiments on WSI captioning, visual question answering, and slide-level classification tasks demonstrate that an 87M parameter LaGuadia student model matches or exceeds foundation-scale models such as GigaPath and UNI, achieving strong factual consistency and robust generalization. These results highlight clinical language as an effective semantic anchor for building efficient and reliable digital pathology systems. Code is available at this https URL.
[CV-49] A Nearable Soft Mat Based on Distributed Optical Fiber Sensing for Physiological Monitoring
链接: https://arxiv.org/abs/2607.11255
作者: Vincenzo Lavorgna,Martina Pulcinelli,Andrea Polimadei,Rosaria D Amato,Carlo Massaroni,Michele Arturo Caponero,Emiliano Schena,Daniela Lo Presti
类目: Computer Vision and Pattern Recognition (cs.CV); Applied Physics (physics.app-ph); Optics (physics.optics)
备注:
Abstract:Distributed optical fiber sensing (DOFS) combines the advantages of fiber optic sensors, including flexibility, small size, immunity to electromagnetic interference, and high metrological performance, with the capability to transform a single optical fiber into a continuous sensing element for spatially resolved mechanical measurements. Optical frequency domain reflectometry (OFDR), based on Rayleigh backscattering, enables high spatial resolution DOFS measurements, broadening the range of potential sensing applications. However, OFDR based DOFS remains largely unexplored for biomedical applications, despite the need for sensitive, spatially resolved, and conformable sensing interfaces. This study presents a soft DOFS based mat as a large-area interface for physiological monitoring. A single-mode optical fiber was embedded in a flexible silicone matrix and arranged in a serpentine layout to distribute sensing over the mat surface. With a gage pitch of 2.6 mm, the system provided 2250 sensing sites across the active area at a sampling frequency of 50 Hz. The mat was assessed on six healthy volunteers in a seated nearable configuration on the backrest of a standard office chair. The distributed output enabled two dimensional mapping of the mat response, reflecting back mat mechanical coupling and cardiorespiratory induced perturbations. Respiratory rate and heart rate were therefore estimated and compared with a reference wearable system. The maps revealed physiologically coherent spatial and temporal patterns, while the estimated rates showed good agreement with the reference measurements. These results demonstrate the feasibility of combining large area distributed sensing, spatial mapping, and quantitative cardiorespiratory monitoring within a DOFS based soft nearable interface.
[CV-50] Structure-Detail Decoupled Autoregressive Generation for Fast and High-Fidelity Virtual Try-On
链接: https://arxiv.org/abs/2607.11233
作者: Lu Yang,Xiaonan Hu,Yanan Li,Daqi Liu,Xiang Bai,Hao Lu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Virtual try-on (VTON) is a bi-conditional image generation problem that requires not only accurate person preservation but also faithful garment deformation and detail synthesis. Diffusion-based VTON methods can jointly model these factors in a compressed latent space, but suffer from high-frequency detail loss due to inherent latent compression, even with costly multi-step denoising. Recent visual autoregressive (VAR) models offer a promising alternative for high-quality generation with faster inference, yet remain unexplored for VTON due to the lack of effective bi-conditioning mechanisms. To bridge this gap, we first introduce VAR-VTON, a VAR-based VTON model that incorporates garment conditioning and structural guidance for efficient latent-space VTON. Despite its efficacy, latent-space generation still struggles to preserve fine-grained garment details. We argue that different VTON sub-tasks should be addressed in different representation spaces: structural synthesis such as garment warping and person layout is suited to the latent space, whereas fine-grained detail recovery should be tackled in the pixel space. Motivated by this insight, we further propose STAR-VTON, a Two-Stage AutoRegressive framework that builds upon VAR-VTON by decoupling latent-space structural synthesis from pixel-space detail recovery. Our idea is to resort to a matching-informed refiner to establish dense correspondences between the stage-one generation and the source garment to directly map fine-grained pixel-space details. Extensive experiments show that STAR-VTON achieves an impressive efficiency-fidelity trade-off: VAR-VTON runs at least 4\times faster than diffusion-based counterparts without degrading quality, and the pixel-space refiner effectively restores fine details and acts as a plug-and-play module that can benefit existing VTON approaches.
[CV-51] HandFlow: Fully Generative 4D Hand Recovery with Flow Matching
链接: https://arxiv.org/abs/2607.11221
作者: Mingxi Xu,Bowen Duan,Yi Gu,Zhengyang Shen,Renjing Xu,Yutao Yue
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Accurate monocular 4D hand reconstruction remains challenging. Per-frame discriminative regressors lack temporal context and often produce jittery predictions. Temporal models improve consistency by aggregating information across frames, but they are typically deterministic regressors, making them vulnerable to ambiguous observations caused by occlusion and motion blur. Generative modeling offers a natural alternative by learning a prior over plausible hand motion sequences, enabling coherent hand-state recovery when visual evidence is incomplete or unreliable. Motivated by this observation, we present HandFlow, a fully generative flow-matching framework for temporally coherent 3D hand pose and shape estimation from monocular video. Given visual and skeletal observations, HandFlow denoises an entire temporal window of MANO parameters through a single ODE integration. To support this, we use a Flux-style dual-stream transformer that attends across the full sequence to capture long-range dependencies without autoregressive decoding, and a confidence-aware continuous masking mechanism that blends observed features with learnable mask tokens to handle noisy or missing observations. Experiments on DexYCB and HOT3D show that HandFlow achieves state-of-the-art performance, with particularly large gains in world-space accuracy and temporal smoothness. It reduces world-space pose error by over 30% compared with the strongest baseline and achieves the lowest acceleration error among all evaluated methods, while remaining competitive in per-frame pose accuracy. Moreover, on a single GPU HandFlow reconstructs a 150-frame sequence at 47 fps, about 12x faster than the fastest prior video-based method, with reconstruction itself accounting for only a small fraction of the end-to-end latency. Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) Cite as: arXiv:2607.11221 [cs.CV] (or arXiv:2607.11221v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2607.11221 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-52] A Novel Method to Evaluate Models on Unreliable Noisy and Inconsistent Labels: Adaptive Resolution Label Aggregation (ARLA)
链接: https://arxiv.org/abs/2607.11214
作者: Natasha Randall,Gernot Heisenberg
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Labels are critical for both training and evaluating deep learning segmentation models, but are often inconsistent, noisy, or ambiguous at class boundaries. Many approaches have been developed to support training models on weak labels, but few to none currently exist to facilitate evaluating models on unreliable labels. We therefore introduce a method called “Adaptive Resolution Label Aggregation”, or “ARLA”, which dynamically adapts the resolution of both the label and the model prediction at inference time before the evaluation metrics are computed. We demonstrate how ARLA can be used to better analyse model behaviour with a practical application to a real flood prediction model, where ARLA was able to overcome issues with inconsistent labelling of forested areas and errors in labels within regions of heavy cloud cover. Our work presents a new approach to evaluating segmentation models, with adjustable parameters to adapt the aggregated resolution to the precision of the label or the level of label noise. Fundamentally, ARLA exploits the information encapsulated by a label but minimises the label error, extracting from the noise a clearer signal of a model’s true performance.
[CV-53] Parallax Portrait Matting ECCV2026
链接: https://arxiv.org/abs/2607.11205
作者: Xin Cai,Jiawen Chen,Lars Jebe,Tianfan Xue,Zhoutong Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: ECCV 2026
Abstract:Image matting is highly ill-posed, especially when both the foreground and background are richly textured. While single-image matting methods learn strong priors from data, they often struggle on these challenging cases. Existing approaches improve results by requiring additional signals such as green screens, polarized lighting, or clean background images, but these typically rely on specialized capture setups. We present Parallax Portrait Matting, a practical two-frame matting method that uses a second image captured with slight viewpoint change. Such a setting arises naturally in burst photography, where small camera motion induces foreground-background parallax and provides complementary observations for matting. Our pipeline estimates trimaps and foreground/background motion, then constructs aligned views for prediction. To handle imperfect motion estimation, the network uses the background-aligned pair for direct fusion and the foreground-aligned cue through cross-attention for error compensation. Experiments show that our method recovers finer details and more accurate foreground colors than strong single-image matting baselines on challenging portrait cases.
[CV-54] DynEval: Holistic Evaluations of T2I Generative Models in the Wild ECCV2026
链接: https://arxiv.org/abs/2607.11199
作者: Shyam Marjit,Dheeraj Baiju,Anuj Shikarkhane,Akhil Sakthieswaran,Sayak Paul,Anirban Chakraborty
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at ECCV 2026. Project page: this https URL
Abstract:Recent advances in text-to-image (T2I) generation have led to models capable of producing highly realistic images. Yet, reliably evaluating their outputs remains challenging, especially at scale. Existing automatic evaluators, often relying on a static prompt set, struggle to capture subtle failure modes such as partial prompt misalignment, compositional errors, or visually plausible but semantically incorrect generations. In this work, we introduce DynEval, a Dynamic Evaluation framework designed to jointly assess text-to-image alignment and image quality of T2I models. To support scalable training beyond limited human-annotated data, we construct two large datasets. First, we build GenDB, a collection of 500K prompt-image pairs generated from human-written prompts drawn from DiffusionDB using a tiered prompt-model generation strategy. Second, building upon GenDB, we construct DynEvalInstruct, a 250K instruction dataset comprising prompt-image-response triplets distilled from a structured evaluation pipeline that decomposes evaluation into text-image alignment and visual quality reasoning. Using this dataset, we perform full fine-tuning of a compact evaluator through a curriculum learning strategy to effectively distill the superior evaluation capabilities of a larger teacher vision-language model, resulting in DynEval-2B and DynEval-4B. In extensive comparisons against existing evaluators across 11 benchmarks, our evaluator achieves a higher overall correlation with human judgments. Furthermore, it provides fine-grained analysis of the capabilities and failure modes of 36 T2I models across 42 subcategories and 9 semantic dimensions.
[CV-55] Slot-RAE: Streamlining Object-Centric Learning via Direct Representation Auto-Encoders
链接: https://arxiv.org/abs/2607.11196
作者: Alexandre Chapin(LIRIS),Emmanuel Dellandrea(LIRIS),Liming Chen(LIRIS)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Deploying object-centric models for real-world scene understanding typically requires complex pipelines to achieve both robust scene decomposition and high-fidelity generation. Recent diffusion-based approaches have improved visual quality, but they almost universally rely on heavy, pretrained generative priors (e.g., Stable Diffusion) and external VAE latent spaces. In this paper, we propose Slot-RAE, a much simpler, fully integrated framework that operates directly within the continuous semantic feature space of visual foundation models (e.g., DINOv3). Slot-RAE employs a feature-space diffusion process using a Diffusion Transformer (DiT) decoder and a Representation Alignment (REPA) head. Unlike existing diffusion-based objectcentric methods that rely heavily on subsidized text-toimage priors, the generative core of Slot-RAE (Slot Attention and the DiT) is trained from scratch within the frozen VFM feature space. This eliminates the need for VAE bottlenecks and task-agnostic generative pre-training. Experiments on the COCO dataset demonstrate that despite its architectural simplicity, Slot-RAE achieves state-of-the-art results. It delivers comparable unsupervised object discovery, higher-fidelity image reconstruction, and robust zero-shot compositionality, all while being significantly faster and more computationally efficient than existing object-centric latent diffusion models.
[CV-56] GDP.pdf: Benchmarking Grounded Multimodal Reasoning over Professional PDF Documents CVPR2026
链接: https://arxiv.org/abs/2607.11192
作者: Suhaas Garre,Emily Ritchie,Sushant Mehta,Edwin Chen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages. Accepted at the 2nd Workshop on Knowledge-Intensive Multimodal Reasoning (KnowledgeMR) at CVPR 2026 (non-archival), under the paper’s former title “PDFParse: A Benchmark for Grounded Multimodal Reasoning over Professional PDF Documents”. Dataset: this https URL ; Code: this https URL
Abstract:A large share of day-to-day work in professional domains happens inside PDF files: benefits packets, leases, datasheets, clinical guidelines, construction plans. Benchmarks for document AI have generally measured the required capabilities in isolation: OCR, layout analysis, chart reasoning, table QA, document VQA. A high score on any one of them does not necessarily reveal whether a model can answer a realistic question that someone in the field would actually ask about a specific PDF. this http URL is a benchmark built to measure this directly. It consists of question-document pairs authored by working professionals in ten fields, and a candidate question was kept only when at least two frontier multimodal models failed it in a way that mattered: a wrong answer, missed decisive evidence, or a fabricated claim, rather than a superficial difference such as style. Each item comes with a rubric of atomic criteria, so we can report a graded rubric score as well as a strict task-level pass rate, and each item is tagged against a taxonomy of eleven capabilities in three tiers, spanning text extraction and grounding, table and chart comprehension, cross-referencing, spatial reasoning, and abstention on unsupported queries. We evaluated seven frontier models on the 100-item benchmark. The best model passed only 15% of the items and the worst passed 1%. Most errors trace back to a small set of recurring loss patterns: misaligned tables, misread charts, skipped footnotes and exclusions, miscounted floor-plan symbols, scan noise, and amendments that supersede earlier text. The full 100-item benchmark is publicly available at this https URL
[CV-57] When Depth Is Better Told Than Shown: Depth-Ordinal Prompting for Vision-Language Spatial Reasoning
链接: https://arxiv.org/abs/2607.11173
作者: Quynh Vo,Phuc Dao,Cong-Duy Nguyen,Thong Nguyen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Work in progress
Abstract:Vision-language models (VLMs) are expected to reason about physical space – which object is closer, what lies behind what, and how objects are arranged in 3D – yet they still struggle with such spatial judgments. A natural remedy is to show the model a depth map, but we find that this can make performance worse. We show that depth is not absent: it reaches the language model, but becomes difficult to access for downstream reasoning, while rendered pseudo-depth maps act as noisy auxiliary images that frozen VLMs cannot easily regulate. We propose Depth-Ordinal Prompting (DOP), a training-free method that converts monocular depth into a single question-targeted ordinal text cue at the queried objects, without adding a depth image, training a module, injecting features, or using labels. Our key finding is form dependence: the same depth signal can hurt when shown as an image but help when told as this http URL benchmarks, models, and depth estimators, DOP improves spatial reasoning when pseudo-depth provides reliable object-level ordering and remains largely neutral in strong original-image regimes. It is also competitive with the strongest training-free depth-prompting alternative while being simpler and more targeted.
[CV-58] C-MAF: Train-Calibrated Bounded Multi-Evidence Fusion for Multimodal Industrial Anomaly Detection ACM-MM2026
链接: https://arxiv.org/abs/2607.11170
作者: Ming Deng,Sijin Sun,Xiaochuan Hu,Xing Wu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: accepted by ACM MM 2026
Abstract:Multimodal anomaly detection benefits from complementary RGB and 3D evidence, yet auxiliary RGB reconstruction is not equally reliable across product categories and class-wise test-time policy selection is usually unavailable. We propose TC-MAF, a base-anchored multi-evidence fusion design that combines a multimodal detector, complementary Dinomaly evidence, and a small cross-modal consistency cue under one fixed pixel-level fusion formula. A lightweight training-dispersion confidence (TDC) term scales auxiliary participation using only normal training statistics. On MVTec-3D, TC-MAF reaches 0.979 image-level AUROC and 0.990 pixel-level AUPRO, achieving the best mean results on both detection and localization among the compared multimodal methods. Systematic ablations show that the fusion structure itself is the dominant factor, while TDC provides a smaller but reproducible calibration gain over no calibration or arbitrary calibration. Additional experiments show that the same design remains effective under a pooled-statistics variant, auxiliary-branch and backbone substitutions, few-shot settings, a missing-3D setting, and cross-dataset evaluation on Eyecandies. Code is available at this https URL.
[CV-59] SISA-Rec: A Semantically Integrated Sequential Recommender with Contrastive Alignment
链接: https://arxiv.org/abs/2607.11168
作者: Soohan Abbasi,Shahid Munir Shah,Rafia Shaikh,Mahmoud Aljawarneh
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Recommendation systems help users recommend relevant items from a large collection of choices. Present work on transformer-based sequential recommendation learns user preferences from interaction logs, but it mostly focuses on item identifiers and doesn’t fully use the semantic meaning of items. This limitation becomes a major challenge in sparse and cold-start scenarios where historical interaction data is limited. To solve this problem, we introduce SISA-Rec (Semantically Integrated Sequential Recommendation), a transformer-based framework that embeds semantic context directly into sequential modeling. Our approach fuses item ID embeddings with BERT-based text embeddings via a gated fusion module, injects semantic similarity into the self-attention mechanism, and leverages an attention-based aggregation module to construct comprehensive user representations. Finally, a joint learning objective which combines Bayesian Personalized Ranking (BPR) and contrastive alignment loss, aligns the underlying behavioral and semantic spaces. Experiments were conducted on the two highly sparse Amazon Beauty and Amazon Toys \ Games datasets, both having 99.93% sparsity. The results show that SISA-Rec outperforms state-of-the-art baseline models across all evaluation metrics. Compared with the BERT4Rec \citepetrov2022systematic, SISA-Rec improves HR@10 by 16.6% and NDCG@10 by 10.3% on Amazon Beauty, and HR@10 by 23.1% and NDCG@10 by 17.9% on Amazon Toys \ Games. Cold-start analysis further shows that the proposed model achieves the largest improvements for users with limited interaction historical records. This showcases the value of semantic information when user behavior data is scarce. Overall, the results demonstrate that integrating semantic information into the attention mechanism leads to more accurate and reliable recommendations.
[CV-60] GHOST: Geometry-Guided Hallucination of Opaque Surface Textures
链接: https://arxiv.org/abs/2607.11118
作者: Langxu Zhao,Zuan Gu,Tianhan Gao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Transparent objects pose a fundamental challenge for depth estimation and 3D reconstruction due to their violation of Lambertian assumptions, leading to severe geometry degradation in downstream tasks. To address this, we propose a novel geometry-guided preprocessing framework \textbfGHOST that leverages visual foundation models to transform transparent regions into opaque, structurally consistent representations without requiring downstream model retraining. Specifically, our pipeline utilizes (1) \textbfTransDINO and (2) \textbfTransDecomp to disentangle masks and transparency physical properties, while (3) \textbfDAF-Net recovers surface normal priors to encode geometric curvature. Subsequently, (4) \textbfGeoSemTransNet integrates these multi-modal cues to synthesize a texture-rich opaque RGB image that preserves the transparent object’s 3D structure. Extensive experiments demonstrate that our method significantly enhances the accuracy of state-of-the-art depth estimation and reconstruction models on transparent objects by restoring essential photometric cues.
[CV-61] Beyond the Eye: Efficient Multimodal Reasoning via Self-Regulated Implicit Visual Tools
链接: https://arxiv.org/abs/2607.11106
作者: Xiuwei Chen,Quanlin Chen,Wentao Hu,Zisheng Chen,Kun Xiang,Zehua Ma,Mingyang Zhang,Jianhua Han,Hanhui Li,Hang Xu,Xiaodan Liang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Recent multimodal large language models (MLLMs) have made remarkable progress on fine-grained perception tasks under the “Thinking with Images” (TwI) paradigm by iteratively performing various visual tool operations. However, this paradigm relies heavily on frequent external tool calls and repeated image re-encoding, which leads to substantial computational overhead and inference latency. To address these issues, we propose Beyond the Eye (BEE), a novel implicit visual tool paradigm centered on self-regulated capability. BEE directly incorporates visual tool invocation behaviors into the training objective and encourages the model to develop a self-regulated invocation mechanism. This design enables the model to adaptively balance internal knowledge and implicit tools, avoiding redundant tool usage while substantially reducing inference latency. Specifically, BEE involves a two-stage training process: (1) Formalized Chain-of-Thought (CoT) Supervised Fine-tuning (SFT). We construct CoT trajectories with structured tool slots and mixed invocation states. This stage activates the model’s implicit tool representations and adaptive switching capability. (2) Self-regulated Reward-Driven Alignment. To address redundant tool usage caused by ambiguous cognitive boundaries, we first introduce the Net Tool Gain (NTG) metric to quantify this phenomenon. Based on this observation, we further propose a self-regulated reward mechanism. This mechanism penalizes ineffective tool dependency and encourages the model to perform knowledge routing, ensuring that implicit tools are invoked only when the model’s internal knowledge is insufficient. BEE achieves state-of-the-art performance in fine-grained visual perception while remaining competitive in general reasoning tasks and achieving substantial gains in inference efficiency.
[CV-62] FlowPET: Physics-Informed Symplectic Flow Matching for Low-Count PET Reconstruction ICML2026
链接: https://arxiv.org/abs/2607.11104
作者: Zheng Zhang,Hao Tang,Yingying Hu,Zhanli Hu,Jing Qin
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: ICML 2026
Abstract:Low-count Positron Emission Tomography (PET) reconstruction is severely hindered by the dissipative nature of prevailing generative models, where the inherent phase-space contraction leads to the numerical extinction (``wash-out’') of weak but diagnostically critical lesion signals. To overcome this geometric limitation, we propose \textbfFlowPET, a physics-informed framework that reformulates reconstruction as volume-preserving transport in a symplectic phase space. By parameterizing the posterior dynamics via a Separable Hamiltonian System, our approach guarantees a divergence-free vector field by construction, theoretically immunizing weak signals against probability mass collapse. To steer this conservative flow, we introduce conjugate boundary conditions based on the Range-Null space decomposition of the PET operator; this strictly enforces data consistency in the range space while confining stochastic uncertainty injection to the unobserved null space. We train the model via symplectic flow matching and perform inference using a symplectic leapfrog integrator. Extensive experiments on BrainWeb, clinical pediatric, and UDPET datasets demonstrate that \textbfFlowPET not only surpasses state-of-the-art deterministic and stochastic baselines in SSIM and PSNR but, more crucially, exhibits superior recovery of low-contrast lesions. The results confirm that imposing Hamiltonian structural constraints offers a robust geometric safeguard for medical inverse problems in high-noise regimes.
[CV-63] Desc: Efficient Descriptor Enhancement for Data Association in Existing Visual SLAM Systems
链接: https://arxiv.org/abs/2607.11099
作者: Ting-Wei Ou,Huang-Ting Lin,Kuu-Young Young
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 12 pages, 6 figures, and 9 tables
Abstract:Reliable visual data association is fundamental to visual SLAM (V-SLAM), as it directly determines the quality of the camera pose estimation and map consistency. However, the handcrafted descriptors used by most mature real-time systems degrade under illumination and viewpoint changes, while learning-based front-ends that address this weakness typically require replacing the extraction-and-matching pipeline and introduce substantial computational overhead. Descriptor enhancement offers a compromise by refining existing descriptors within their original format, yet current methods rely on simplified attention mechanisms whose limited contextual modeling constrains the achievable matching quality. To resolve this trade-off between contextual expressiveness and efficiency, we propose Desc++, a lightweight enhancement module that jointly encodes descriptor representations and keypoint geometry and aggregates spatial context through a hybrid architecture that combines order-agnostic global attention with geometry-aware sequential modeling in linear time. The enhanced descriptors retain their original dimensionality and matching interface, enabling integration into deployed V-SLAM systems without modifying the pipeline. Experiments across descriptor matching, correspondence analysis, and system-level benchmarks with four different V-SLAM systems demonstrate that Desc++ improves matching accuracy over the state-of-the-art enhancement method, translates these gains into more accurate and stable trajectory estimation, and achieves a favorable balance between accuracy and efficiency for practical integration into existing real-time V-SLAM pipelines.
[CV-64] Revisiting Matching Response and Swept Feature Volumes for Wide-baseline Omnidirectional Stereo
链接: https://arxiv.org/abs/2607.11097
作者: Seungjin Jeon,Jongwoo Lim,Changhee Won
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:In this paper, we propose a training strategy for confidence estimation in omnidirectional stereo, targeting the ambiguous matches that frequently occur in wide-baseline setups. Reinterpreting the matching responses produced by the 3D encoder decoder block, we show that their expectation values provide intrinsic confidence signals. Building on this, our method directly penalizes ambiguous responses without auxiliary heads, multi-pass inference, or additional modules, resulting in more efficient and generalized predictions. Beyond confidence, we introduce swept feature volume resampling, where response features produced by 3D CNNs are resampled using regressed positive matching indices and then processed by 2D CNNs to predict meta-information such as surface normals. This joint learning introduces auxiliary geometric regularization and improves depth coherence by leveraging additional contextual cues during response aggregation stage. Experimental results demonstrate that our approach enhances both confidence estimation and surface normal prediction while maintaining deployment practicality for autonomous mobility applications.
[CV-65] Difference-Driven Gating: Adaptive Feature Fusion for U-Net Decoder
链接: https://arxiv.org/abs/2607.11096
作者: Kai Li,Xuechao Zou,Jiashen Fu,Zijun Yan,Xintong Wang,Xiaolin Hu
类目: Computer Vision and Pattern Recognition (cs.CV); Sound (cs.SD); Machine Learning (stat.ML)
备注: 15 pages, 13 figures
Abstract:The U-Net style models have been widely used in many applications. A critical step in these models is to reconstruct the lower-level features using a top-down decoder. This reconstruction requires precise fusion of high-level semantics and low-level details. Existing attention-based fusion methods typically derive attention weights from the top-down decoder features (global) alone or the correlation between the top-down decoder features and the bottom-up encoder features (local), then modulate the encoder features using these weights. In this work, we explore a different paradigm: deriving attention weights from the difference between the two feature streams. To this end, we propose two difference-based gating approaches: Feature-difference gating (FDG), which directly uses the absolute difference between global and local features to generate adaptive gating maps, and Entropy-difference gating (EDG), which measures the representational certainty of each stream via information entropy and uses their signed entropy difference to derive the attention weights. Both methods produce coupled gating maps that simultaneously modulate the global and local features. Experiments on different tasks including medical image segmentation, remote sensing image cloud removal and speech separation showed that both methods outperformed existing attention-based fusion methods, and EDG performed better. The results suggested a new paradigm for multi-scale feature fusion in the U-Net style structures.
[CV-66] Why Low-Light Cameras Go Color Blind: Removing Color Bias in Raw Denoising
链接: https://arxiv.org/abs/2607.11090
作者: Mohammad Mohammadi,Sina Honari,Stavros Tsogkas,Tristan Aumentado-Armstrong,Michael S. Brown,Iqbal Mohomed,Konstantinos G. Derpanis,Alex Levinshtein,Igor Gilitschenski
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at ICCP 2026
Abstract:Raw images inherently suffer from noise due to the stochastic nature of light and sensor hardware imperfections. As real photon counts fall, the ratio of this noise to the signal degrades; consequently, for low-light conditions, robust denoising is especially vital for high-quality results. While recent data-driven methods achieve strong performance, they typically rely on large-scale noisy-clean image pairs that are costly and difficult to collect. Alternatively, parametric noise models can generate synthetic training data, but this necessitates precise camera calibration, which is often impractical for unknown devices. In this work, we propose a camera-agnostic, calibration-free paradigm for low-light raw denoising. We identify that color bias from black-level error is a primary source of performance degradation and causes severe color shifts. To mitigate this, we introduce a bias estimator network that predicts the black-level error as a global feature of the noisy input. We evaluate our approach across the ELD, SID, and LRID datasets, demonstrating superior performance among blind denoisers, particularly in terms of color correction. In many cases, we are competitive with-or can even surpass-methods with stronger supervision. Furthermore, we reveal that the widely used SIDD dataset contains significant color bias in its ground-truth images, which yields unrealistic color reproduction in trained models. We introduce a new ground-truth extraction framework to resolve this issue and provide a benchmark of existing methods on the corrected dataset.
[CV-67] CUST: Clustered Unit-level Similarity Transformer for Lightweight Image Super-Resolution
链接: https://arxiv.org/abs/2607.11088
作者: Jeongsoo Kim
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 15 pages, 7 figures
Abstract:Recently, Vision Transformer (ViT)-based models have exhibited remarkable performance in image super-resolution. However, the quadratic computational complexity of ViTs with respect to spatial resolution severely constrains their efficiency, leading to high latency and massive memory consumption. To alleviate this, various window-based attention mechanisms have been proposed; yet, they inherently compromise the long-range dependency modeling that is the primary advantage of ViTs. To overcome these limitations, we propose the Clustered Unit-level Similarity Transformer (CUST), a novel architecture that efficiently integrates global and local information. Specifically, CUST enables each patch to aggregate and attend to similar patches within a broadened regional scope outside its local window, thereby capturing extensive contextual understanding. Furthermore, it employs overlapping attention windows to capture local dependencies, while explicitly extracting high-frequency details by computing the residual difference between the original features and their downsampled-upsampled counterparts. Comprehensive experiments demonstrate that our proposed model achieves a practical balance between computational efficiency and restoration performance. It achieves a lower memory footprint and faster inference speed compared to recent global context or lightweight models under realistic constraints. Code is available at [this https URL].
[CV-68] Controlling Motion Transfer in Diffusion Transformers via Attention Heads ECCV2026
链接: https://arxiv.org/abs/2607.11081
作者: Sunyoung Jung,Jiwoo Park,Yoonseok Choi,Kyobin Choo,Ming-Hsuan Yang,Seong Jae Hwang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted to ECCV 2026, Project page: this https URL
Abstract:Diffusion Transformers (DiTs) have advanced video generation with high-quality, temporally coherent results. However, extending them to motion transfer, which requires following reference motion while aligning with a target prompt, remains challenging due to limited understanding of motion and structure representations within DiTs. We analyze video DiTs at the attention-head level and identify distinct heads specialized for motion and spatial structure. Based on this insight, we propose a head-aware controllable motion transfer framework that requires no parameter updates. Our method refines motion cues from motion-specialized heads via semantic correspondence guidance and preserves structure through selective feature injection. This head-level control not only enables accurate motion transfer but also provides an interpretable foundation for controllable video generation with DiTs.
[CV-69] Do Video-LLM s Actually Watch? Diagnosing Character-Tracking Failures in Long-Form Video
链接: https://arxiv.org/abs/2607.11078
作者: Mohammad Al-Ratrout,Shayla Sharmin,Aditya Raikwar,Roghayeh Leila Barmaki
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Can a Video Large Language Model (Video-LLM) follow one person through a long video, keeping track of who they are well enough to report, in order, how their outfit changes across a full TV episode? Benchmarks increasingly score this kind of task, and the strongest open-source 7–8B models now reach 37–38% on InfiniBench’s global appearance task, which asks exactly that. But does that score come from tracking the named character, or from something easier? We test this with a nine-condition diagnostic protocol applied to three architecturally distinct open-source Video-LLMs, with Gemini~2.5~Flash as a frontier reference, and find the accuracy does not come from character tracking. When we change the character named in the question to a different cast member, leaving the video and answer options untouched, the models change their answer only 4–31% of the time, so they are largely ignoring who the question asks about. Breaking that test down by the gender of the swapped name shows why: the models react more when the name is changed to a different-gender character than to a same-gender one (a 13–28 point gap), picking up coarse gender cues but unable to tell same-gender individuals apart. This shallow processing surfaces again when we drop the multiple-choice options and ask the same questions open-endedly: open-source accuracy drops 18–25 points, with none of 151 answers fully correct, versus a 12-point drop for Gemini. Further checks rule out the obvious innocent explanations, adding subtitles, using the most informative frames, or doubling the number of frames all leave character tracking unimproved, so the bottleneck is not how much video the model sees but how it ties that video to the person the question names. We release a diagnostic toolkit for auditing what such benchmark scores actually measure.
[CV-70] DDR-Net: Haze-Aware Dual-Domain Refinement for Single-Image Dehazing
链接: https://arxiv.org/abs/2607.11071
作者: Xinye Zheng,Ye Yu,Qiang Lu,Jinsheng Luo,Yiran Cui,Yongbin Cheng
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by PRCV 2026
Abstract:Single-image dehazing aims to recover clear scenes from haze-degraded images. It remains challenging due to the atmospheric scattering and the complexity of real-world haze distributions. Although recent end-to-end networks have achieved promising performance, two issues still limit their effectiveness: insufficient feature refinement at the bottleneck stage and weak local structural representation in encoder-decoder architectures. Thus, we propose a Haze-Aware Dual-Domain Refinement Network (DDR-Net) for single-image dehazing. Our method is built upon three modules: Haze Prior Extractor (HPE) provides multi-scale haze-aware priors by operating directly on downsampled hazy images; Detail-Enhanced Blocks (DE Blocks) serve as the core feature extraction units, capturing multi-scale structural information and enhancing edge and texture recovery via gradient-aware convolutions; and Spatial-Frequency Bottleneck Refinement (SFBR) at the bottleneck jointly exploits spatial and frequency information to refine bottleneck features. DDR-Net achieves more effective feature representation and reconstruction for haze removal. Extensive experiments on real-world benchmarks demonstrate that our method outperforms existing dehazing approaches. It achieves competitive performance on synthetic datasets.
[CV-71] WiFi-JEPA: Self-supervised Learning for WiFi-CSI 3D Human Pose Estimation
链接: https://arxiv.org/abs/2607.11064
作者: Doeon Kim,Jungyoon Lee,Seongsin Kim,Seong-heum Kim
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:WiFi Channel State Information (CSI) enables privacy-preserving human pose sensing in camera-denied environments, but existing WiFi-based pose estimators often fail under environment shifts and rely on costly camera-based annotation pipelines that limit scale. We propose WiFi-JEPA, a self-supervised framework that learns CSI-native representations by predicting masked latent embeddings instead of reconstructing raw CSI signals that may contain hardware-specific artifacts. WiFi-JEPA makes three contributions: (i) CSI-specific tokenization and link masking tailored to the CSI tensor over channel, time, and link (C,T,L); masking entire Tx-Rx antenna links forces the model to predict one spatial link view from others, capturing cross-link correlations informative of 3D spatial structure. (ii) A ray-tracing CSI simulation pipeline that generates diverse unlabeled CSI from randomized geometric primitives, providing scalable pre-training data without pose annotations. (iii) State-of-the-art results on Person-in-WiFi-3D: WiFi-JEPA outperforms prior WiFi-CSI baselines on both single- and multi-person 3D pose estimation under the same evaluation protocol. We also show that simulated CSI provides complementary pre-training signal to real CSI, and that four vision-native SSL objectives degrade performance below training from scratch, whereas WiFi-JEPA consistently improves downstream pose estimation.
[CV-72] FSFVE: Few Shot Compressed Face Video Enhancement
链接: https://arxiv.org/abs/2607.11040
作者: Varun Ramesh Jois,Antonella DiLillo,James Storer
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Videocalling has become a popular form of communication in the world today, with many companies providing free services for it. However, there are still millions of people around the world that experience poor quality videocalls due to limitations in bandwidth. This despite, most people having the required hardware. In this paper we present a novel framework for enhancing highly compressed videocalls. We show, that with as little as 10 frames of the face, we can rapidly (in under 100 seconds) train a model to enhance that instance of the videocall. The model can be trained either prior to or during the call, enhancing the rest of the call by producing better quality video. The video conferencing application need not be modified - it can be off the shelf with our system as a layer on top that trains quickly then simply lets the video conferencing application (e.g. Zoom) run as usual, where our system intercepts and improves images before they are displayed. The model is designed to run in realtime on low-compute devices such as a typical laptop CPU. Experimentally, we show that the model significantly improves quality of compressed face video both quantitatively as well as perceptually. Code can be found at this https URL.
[CV-73] RTFVE: Realtime Face Video Enhancement
链接: https://arxiv.org/abs/2607.11034
作者: Varun Ramesh Jois,Antonella DiLillo,James Storer
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:There’s been a surge in adoption of video conferencing applications for both personal and business use cases. However, the bandwidth limitations faced by many users worldwide may restrict the optimal use of such applications. Although deep learning offers a solution for enhancing low bit rate videos, most models today are either hard to incorporate with modern compression standards or require specialized hardware to run such as significant GPUs making these models impractical. To address these issues, we introduce the Realtime Face Video Enhancement (RTFVE) model which can be easily incorporated with any video decoder and can run in realtime on ordinary CPUs. Experiments show that our model improves perceptual quality over the compressed video baseline at multiple low bitrate settings. The source code will be made available at this https URL.
[CV-74] Learning to Navigate Efficiently with Only 0.58M Trainable Parameters
链接: https://arxiv.org/abs/2607.11029
作者: Edward Beng Wai Tan,Siew-Kei Lam
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 6 pages, 4 figures
Abstract:Recent progress in visual navigation has largely been driven by scale: end-to-end policies with hundreds of millions of parameters trained on billions of frames or large-scale simulated data. We ask how much of this scale a single task family actually requires, and what structure can substitute for it. We propose a decomposed navigation model in which operations with known closed-form structure, such as projective geometry, occupancy, and coordinate transforms, are computed analytically and serve as interfaces between three small learned modules: an egress predictor that grounds the episode goal as a local subgoal in the current view, a navigation predictor that estimates a goal-conditioned posterior over where trajectories travel, and an endpoint-pinned residual diffusion generator that samples trajectory shapes from this posterior. The system trains only 0.58M out of a total of 22.7M parameters, on 44k frames in under one GPU-hour, yet approaches the performance of state-of-the-art models on navigation tasks across 6060 point-goal episodes and 60 environments, while having 233x fewer trainable parameters, the lowest collision rate among all evaluated methods, and 50 Hz inference speed. The decomposition further transfers to no-goal exploration by retraining only the 123k-parameter egress head, and its failure modes under sensor corruption are transparent and analytically correctable.
[CV-75] Reference-Based Face Super-Resolution Using the Spatial Transformer
链接: https://arxiv.org/abs/2607.11025
作者: Varun Ramesh Jois,Antonella DiLillo,James Storer
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Face super-resolution is the task of increasing the resolution of an image containing a face thereby adding finer detail. It is a ubiquitous task in many computer vision applications and quite often the user isn’t even aware that it is being performed. However, doing it with high fidelity is challenging as it is an ill-posed problem. In this paper we present a reference-based solution for face super-resolution that uses higher resolution reference images to aid in the task. We show an alignment module based on the spatial transformer that is considerably more stable than the popular deformable convolutions. We also show an aggregation function that can take good quality information from the reference images when available or suppress the function when such information is unavailable. Finally, we show that our relatively smaller model can achieve state of the art results on multiple datasets. The source code is available at this https URL.
[CV-76] SynCLIP: Synonym-Coherent Language-Image Pretraining for Robust Open-Vocabulary Dense Perception CVPR2026
链接: https://arxiv.org/abs/2607.11008
作者: Mingjie Xie,Guangjun He,Dongli Xu,Youtian Lin,Hongjue Li,Pengming Feng,Jian Guan,Yue Deng
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted by CVPR 2026
Abstract:Open-vocabulary dense perception (OVDP) aims to localize objects unseen during training by leveraging textual knowledge. Despite the remarkable progress of recent CLIP-based approaches, we identify a critical limitation: synonym-induced grounding inconsistency, where semantically equivalent expressions yield disparate spatial attention patterns. This inconsistency undermines the robustness and performance of existing methods in real-world OVDP applications. To address this issue, we propose SynCLIP, a Synonym-Coherent Language-Image Pretraining framework that enhances synonym-robust grounding for OVDP. SynCLIP introduces a Semantic-consistent Spatial Attention alignment (SSA) module to enhance spatial attention consistency by minimizing discrepancies between attention maps of original and synonymous expressions. Furthermore, a Spatial Attention Refinement (SAR) module selectively strengthens the most semantically relevant spatial regions within aligned maps for more precise and stable grounding. To support synonym-coherent pretraining, we also construct a Synonym-Enriched Visual Corpus (SEViC), which augments each category with multiple synonyms and textual definitions. Extensive experiments on multiple benchmarks demonstrate that SynCLIP substantially improves grounding consistency under diverse linguistic variants and achieves state-of-the-art performance among CLIP-based OVDP methods. Code is available at this https URL.
[CV-77] mporal Feature Distillation for Label-Efficient Precise Event Spotting in Sports Videos ACM-MM2026
链接: https://arxiv.org/abs/2607.10998
作者: Hao Xu,Xinyu Wei,Sam Wells,Sunil Aryal
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at ACM MM 2026
Abstract:Precise Event Spotting (PES) requires distinguishing visually similar yet semantically distinct adjacent frames, making it fundamentally different from image classification and coarse action recognition. Although self-distillation methods such as DINO have shown strong representation learning ability in images, we find that directly applying them to PES is ineffective: without supervised guidance, subtle but crucial motion cues are often suppressed as noise, leading to representations that are insensitive to precise event boundaries. To address this, we propose Temporal Feature Distillation, a semi-supervised objective that aligns temporally informative backbone features, rather than projection-head outputs, to preserve motion-sensitive and boundary-aware cues for frame-level localization. A supervised warm-up with a ramp-up schedule further stabilizes training by ensuring that meaningful event cues are learned before unlabeled distillation begins. We also introduce Transformer Gate Shift, a multi-scale gated shifting module that injects motion-aware temporal information into Vision Transformers. Experiments on four fine-grained sports benchmarks show consistent improvements over fully supervised and semi-supervised baselines. Under 10% supervision on FSPerf, our method improves mAP by 4.54 points over the strongest competing approach, and with only 80% labeled data, it matches or surpasses the fully supervised 100% baseline on two of the four datasets.
[CV-78] AsySplat: Efficient Asymmetric 3D Gaussian Splatting for Long-Sequence Scene Modeling
链接: https://arxiv.org/abs/2607.10995
作者: Yingji Zhong,Dave Zhenyu Chen,Fuzhao Ou,Youyu Chen,Zhihao Li,Lanqing Hong,Dan Xu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: The project page is at this https URL
Abstract:Recent generalizable 3D Gaussian Splatting models have advanced long-sequence novel view synthesis (NVS), but at the cost of substantial redundant computation. We identify that the redundancy can be mitigated based on two observations: (i) high-precision geometry is not strictly required for high-quality NVS; (ii) appearance learning is generally easier than geometry recovery. Motivated by these insights, we propose an asymmetric architecture that decouples geometry and appearance modeling. The geometry branch processes coarse-grained tokens with most of the parameters for multi-view reconstruction, while the appearance branch operates on fine-grained tokens to capture details using significantly fewer parameters. The two branches interact through bilateral connections, enabling mutual guidance for their respective tasks. This task-aware asymmetry reduces the computational redundancy and allocates the computation more judiciously, thereby increasing parameter efficiency and enabling smaller models to achieve strong performance. On 32-view 960P inputs, our model matches optimization-based methods while delivering nearly 800x speedup, and surpasses the zero-shot performance of state-of-the-art generalizable models with markedly fewer parameters and reduced training/inference overhead, achieving an overall efficiency improvement.
[CV-79] Confidence Scores in Open-Vocabulary Detection Are a Biased Mixture of Scale and Semantics ICPR
链接: https://arxiv.org/abs/2607.10993
作者: Yi Tang Soon,Jun-Wei Hsieh
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: ICPR Workshop 2026 (FMVA)
Abstract:Foundation models such as CLIP have enabled open-vocabulary object detectors that generalise to novel categories via vision-language similarity. However, the confidence scores these detectors produce are not reliable localization probability estimates: they conflate visual scale and semantic query specificity with the true detection signal. Through controlled experiments on COCO across three foundation-model-based detectors (GroundingDINO, OWL-ViT, YOLO-World), with the scale-bias finding further replicated on LVIS (1,203 categories) using GroundingDINO, we show that s=cos(v,t) is a biased mixture of two effects. Scale bias (alpha = +0.064, r = 0.579, p = 1.29 x 10^-58) systematically inflates scores for large objects. Semantic bias (beta = -0.705, p = 5.23 x 10^-41) suppresses scores for generic queries. Both biases are structurally inevitable from CLIP’s image-level pretraining. Threshold adjustment cannot remove them: oracle per-scale thresholding yields Delta F1 = +0.001 for small objects versus +0.102 for large. A parameter-free temperature scaling correction improves small-object Recall@10 by 19.6% (p 0.01) without retraining. This comes at a modest, measurable cost to pooled-ranking precision, so the bias is partially, not freely, reversible at inference time. These findings reveal a fundamental limitation of adapting image-level foundation models to region-level detection tasks.
[CV-80] LoSA-Net: A Localized and Scale-Adaptive Network for Boundary-Sensitive Prediction of Perineural Invasion in 3D MRI
链接: https://arxiv.org/abs/2607.10992
作者: Youngung Han,Hyunsu Go,Kyeonghun Kim,Induk Um,Junga Kim,Jaewon Jung,Woo Kyoung Jeong,Won Jae Lee,Pa Hong,Ken Ying-Kai Liao,Hyuk-Jae Lee,Nam-Joon Kim
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Published in the 2026 IEEE 23rd International Symposium on Biomedical Imaging (ISBI 2026); accepted for oral presentation
Abstract:Perineural invasion (PNI) is a clinically relevant indicator of tumor aggressiveness and can influence surgical decision-making, motivating interest in reliable preoperative assessment. The subtle MRI features of PNI, however, often resemble nearby anatomy, complicating noninvasive prediction. These fine perineural cues are easily attenuated by routine downsampling or overly global feature aggregation, reducing the effectiveness of conventional volumetric models. We present LoSA-Net, a localized and scale-adaptive architecture for boundary-sensitive PNI prediction in 3D MRI. Talking Neighborhood Attention (TNA) preserves nerve-aligned detail through localized self-attention with head-wise mixing, and Scale-Adaptive Feature Mixing (SAFM) modulates the receptive field using multi-scale depthwise processing. Cross-Scale Refinement and Alignment (CSRA) maintains consistency between semantic context and high-resolution boundaries across stages. In contrast-enhanced MRI scans from 168 patients with cholangiocarcinoma, LoSA-Net achieves an AUC of 0.7567 and outperforms representative convolutional and transformer baselines under matched preprocessing and optimization settings.
[CV-81] hink When It Matters: Conditional VLM Reasoning for Social Navigation with RL Policies
链接: https://arxiv.org/abs/2607.10991
作者: Ali Ahmadi,Hamed Rahimi,Adrien Jacquet Cretides,Marie Samson,Mahdi Khoramshahi,Mohamed Chetouani
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: CoRL 2026 submission. 15 pages, 7 figures
Abstract:As mobile robots become more integrated into everyday human environments, social robot navigation is becoming essential for ensuring human comfort, safety, and trust. While reinforcement learning (RL) navigation policies provide the fast inference and reactive behavior necessary for real-time deployment, they still lack flexible semantic reasoning capabilities and often fail to generalize to complex social scenarios. Recent approaches have increasingly turned to vision-language models (VLMs) in place of RL policies to improve semantic and social reasoning in robot navigation. Nevertheless, their high computational cost and slow inference remain major barriers to real-time deployment. To overcome these limitations, we introduce HUMA (Hybrid Understanding for Multi-modal social Navigation), a hybrid architecture that dynamically balances the computational efficiency of RL policies with the deep semantic understanding of VLMs. Our approach uses a reactive RL policy to handle low-density, routine navigation tasks, while conditioning it on a post-trained high-level VLM when a human enters sensitive situations, such as the robot’s proximity zone. We evaluate HUMA on the Social-MP3D and Social-HM3D benchmarks, where it achieves task success improvements of 20% and 3%, respectively, while significantly reducing personal space violations and human collisions against state-of-the-art baselines. Extensive ablation studies validate each architectural component, and real-world deployment on the Mirokaï mobile robot further demonstrates the practical viability of our approach.
[CV-82] reeSoc: Tree-Structured Dynamic Reasoning and Tool Synergy for Soccer Video Understanding
链接: https://arxiv.org/abs/2607.10990
作者: Thanh-Nhan Vo,Thanh-Khoi Nguyen,Trong-Thuan Nguyen,Trung-Hoang Le,Minh-Triet Tran
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ICMV 2026
Abstract:Automated understanding of complex soccer scenarios from video remains a significant challenge for contemporary vision-language models (VLMs), which suffer from shallow cross-modal alignment and exhibit fundamental limitations in multi-step reasoning and coordinated tool integration. We present TreeSoc, a structured reasoning framework that reformulates soccer video question answering as a hierarchical search problem rather than a single-pass prediction. Specifically, TreeSoc employs a dynamic depth-first search (DFS) mechanism that decomposes complex queries into sequentially ordered sub-tasks, enabling iterative reasoning refinement through explicit intermediate states. This tree-structured decomposition naturally supports adaptive tool routing, wherein domain-specific modules are selectively activated and their outputs incorporated at each reasoning node to produce contextually grounded predictions. On SoccerBench, TreeSoc achieves state-of-the-art performance, with accuracies of 85.2%, 87.4%, and 82.2% on TextQA, ImageQA, and VideoQA, respectively. Additionally, TreeSoc further demonstrates strong cross-domain generalization, attaining 74.16% accuracy on NExT-QA. These results establish structured, tool-augmented tree reasoning as an effective paradigm for robust video understanding. Code is available at: this https URL.
[CV-83] MMA-Former: Multi-Window Mixture-of-Head Attention Transformer for Adaptive PNI Prediction in 3D MRI
链接: https://arxiv.org/abs/2607.10988
作者: Youngung Han,Induk Um,Kyeonghun Kim,Junga Kim,Hyunsu Go,Jaewon Jung,Woo Kyoung Jeong,Won Jae Lee,Pa Hong,Ken Ying-Kai Liao,Hyuk-Jae Lee,Nam-Joon Kim
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Published in the 2026 IEEE 23rd International Symposium on Biomedical Imaging (ISBI 2026); accepted for oral presentation
Abstract:Perineural invasion (PNI) is a critical prognostic factor in cholangiocarcinoma. Non-invasive prediction from 3D MRI is challenging, demanding models that efficiently capture both fine-grained details and global context. We propose the Multi-window Mixture-of-Head Attention Transformer (MMA-Former), a novel end-to-end 3D architecture featuring a Coarse-Fine Transformer (CFT) structure for parallel multi-scale feature extraction. We advance this structure by integrating a novel Window-Specific Mixture-of-Head attention (WS-MoH) mechanism. Unlike standard Multi-Head Self Attention (MSA), WS-MoH generates a representation for each 3D window and dynamically routes the entire window to specialized or common attention heads. This enables spatially adaptive feature extraction tailored to the local context of each window, enhancing specialization and reducing redundancy without increasing parameters. Evaluated on a retrospective dataset of 168 T1-weighted MRI scans, MMA-Former achieved an AUC of 0.752, outperforming other 3D architectures, including the best CNN (AUC of 0.708) and Transformer baselines (AUC of 0.681).
[CV-84] MED-DSLC: Multi-Expert-Domain Classification via Domain Supervision and Logit Calibration ECCV2026
链接: https://arxiv.org/abs/2607.10985
作者: Zheng Zeng,Deepak Sridhar,Nuno Vasconcelos
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ECCV 2026. Code is available at this https URL
Abstract:Vision-language models (VLMs) such as CLIP enable zero-shot classification by comparing image features with text prompts in a shared embedding space. A fundamental property underlying this capability is the global comparability of logits across arbitrary candidate classes. However, VLMs are often adapted to fine-grained domains using techniques such as LoRA. While this improves in-domain accuracy, out-of-domain accuracy degrades. This leads to a highly fragmented model ecosystem, with thousands of specialized models. Multi-Expert-Domain classification seeks to address this problem, by merging LoRAs trained independently on specialized domains. However, due to the independent training, the various domain experts no longer produce globally calibrated logits. As a result, when evaluating over the union of multiple domain-specific class sets, heterogeneous logit scales induce cross-domain interference and artificially high confidence for out-of-domain classes, inducing prediction errors. In this work, we identify domain supervision and cross-domain logit miscalibration as the key issue to scalable multi-domain zero-shot recognition. We propose MED-DSLC, combining domain supervised training and domain-wise logit scaling, to explicitly restore global logit comparability. MED-DSLC is a lightweight solution for MED classification, which is shown to preserve within-domain discrimination while reducing cross-domain logit interference with minimal data. Extensive experiments across diverse fine-grained benchmarks demonstrate that it substantially improves mean accuracy (+15%), cross-domain robustness, and scalability in the size of MED classification problem. Our results show that restoring output-level calibration is essential under highly data imbalanced settings for achieving a truly zero-shot VLM under multi-domain specialization.
[CV-85] Learning Anatomy-Grounded CT Vision-Language Representations with Organ-Hierarchical Report Knowledge
链接: https://arxiv.org/abs/2607.10953
作者: Guoliang You,Hongming Li,Yuanwang Zhang,Yong Fan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages, 6 figures, 4 tables
Abstract:Medical vision-language pretraining (VLP) from paired CT images and radiology reports enables scalable representation learning, but most existing methods align either whole scans with entire reports or local image regions with text fragments. These formulations underuse a key property of radiology reports: findings are organized around anatomical structures, with abnormalities described by organs, disease concepts, locations, and severity-related attributes. We propose OKA-CT, an organ-hierarchical knowledge-augmented framework for CT-report VLP. OKA-CT first converts free-text reports into organ-conditioned knowledge using radiology report parsing and LLM-assisted semantic structuring. The extracted hierarchy is used across two learning stages. Stage~1 injects anatomy-grounded evidence into the CT visual representation through fine-grained organ-conditioned supervision, while Stage~2 uses organ-specific report evidence to guide structured report-CT contrastive learning, where hierarchy-derived semantic soft targets treat non-paired cases with shared organ-level findings as weak semantic positives rather than uniform negatives. A lightweight query-based global branch further aggregates disease-relevant volumetric evidence for whole-scan representation. On CT-RATE and RAD-ChestCT datasets, OKA-CT achieves zero-shot abnormality diagnosis AUROCs of 84.9 and 72.2, outperforming prior CT VLP baselines. Retrieval and patch-occlusion analyses further show improved report-image alignment and stronger sensitivity to disease-associated anatomical regions.
[CV-86] Unsupervised Detection of Entry and Exit Regions from Vehicle Trajectories for Camera-Agnostic Turning Movement Counts
链接: https://arxiv.org/abs/2607.10949
作者: Parikshit Singh Rathore,Vishwajeet Pattanaik,Punit Rathore
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 14 pages, 7 figures; Supplementary Material: 7 pages, 3 figures
Abstract:Turning movement counts are essential for intersection-level traffic management, yet their collection remains predominantly manual due to the cost of per-camera region annotation. This paper presents an unsupervised pipeline that identifies entry and exit regions directly from raw vehicle trajectories extracted via object detection and multi-object tracking, requiring no manual annotation, camera calibration, or prior knowledge of intersection geometry. Unlike trajectory clustering methods that classify individual trajectories using pairwise similarity and must be re-executed on every new batch, the proposed pipeline clusters initial and terminal point locations to produce persistent spatial region polygons that classify future trajectories by point-in-polygon containment at linear cost. The pipeline comprises six sequential steps, each with configurable parameters evaluated through a systematic statistical analysis spanning 19,152 pipeline executions across 25 surveillance cameras capturing dense heterogeneous traffic in Bengaluru, India, and 10 sequences from the UA-DETRAC benchmark dataset. Both parametric and nonparametric testing frameworks identify three consistently significant parameters and yield an empirically grounded recommended configuration. Under this configuration, the pipeline achieves a median classification error of approximately 3% across all 25 cameras, including 16 held-out locations, with GEH values within accepted engineering thresholds. Compared with two trajectory clustering baselines, the proposed pipeline exhibits greater stability across camera views and lower computational cost, at the expense of higher median error. Extended evaluation demonstrates that calibration clips of at least 60 minutes and peak-traffic selection further improve region estimation quality.
[CV-87] DP-Splat: Bayesian Nonparametric Complexity Control for Gaussian Splatting
链接: https://arxiv.org/abs/2607.10912
作者: Aqi Dong
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 18 pages, 8 figures. Code and experiment records: this https URL
Abstract:3D Gaussian Splatting represents scenes as finite mixtures of anisotropic Gaussians whose number of components K is set by heuristic density control or user caps. Variational Bayes Gaussian Splatting (VBGS) recast splat fitting as conjugate variational inference, but K remains fixed. We replace the finite symmetric Dirichlet over mixture weights with a truncated stick-breaking Dirichlet-process prior – and, as a theory-backed alternative, a sparse overfitted finite Dirichlet – so that the number of occupied components adapts to the data while every update remains a closed-form coordinate-ascent step; a natural-gradient stochastic variant makes the per-step cost independent of the number of points. We give an exact monotonicity guarantee, a rigorous truncation-error bound correcting an anti-conservative large- \alpha approximation in common use, and an honest account of what the fitted number of components estimates. Empirically: (i) the effective complexity \hatK adapts to scene complexity and recovers the true K within \pm 1 on well-separated synthetic data with regime-appropriate concentration; (ii) a deconfounded comparison shows the DP prior’s contribution is complexity selection, not per-component efficiency – converged DP fits exceed single-pass fixed- K VBGS by +2.7 dB at matched budgets yet tie an equally converged fixed- K baseline, and on 3D scenes DP-Splat matches or exceeds VBGS’s held-out color prediction with 5.9-7.6x fewer components; (iii) the posterior-predictive color variance is well calibrated on model-matched synthetic data; and (iv) the ordering suggested by exact-posterior asymptotics reverses under mean-field coordinate ascent: the DP prior resists over-splitting while the sparse finite mixture saturates its truncation, a gap between variational practice and posterior asymptotics documented across three orders of magnitude in N .
[CV-88] Design Choices in Splitting-Based Self-Supervised Sparse-View CT Reconstruction
链接: https://arxiv.org/abs/2607.10898
作者: Nadja Gruber,Lukas Neumann,Ander Biguri,Gyeongha Hwang,Markus Haltmeier,Johannes Schwab
类目: Computer Vision and Pattern Recognition (cs.CV); Numerical Analysis (math.NA)
备注: 12 pages (main manuscript), 9 pages (supplementary material and figures)
Abstract:Self-supervised data splitting has emerged as a promising paradigm for sparse-view CT reconstruction, enabling training from incomplete measurements without fully sampled ground truth. However, the influence of key design choices, including partitioning strategy, preprocessing, and inference, remains insufficiently understood. In this work, we introduce a unified framework that decomposes splitting-based reconstruction into these three components, enabling controlled comparison of existing methods and two incremental extensions: multi-partition splitting and an alternative inference strategy. Experiments on simulated LoDoPaB-CT data under independent and correlated noise, together with validation on the real-world 2DeteCT dataset, show that the optimal partitioning strategy strongly depends on the measurement noise structure. Lattice-based splitting performs favorably under independent noise, whereas angular masking is more robust under correlated noise and real measured data. Multi-partition splitting consistently improves over pure projection-wise splitting in several settings. Complementary perceptual and structural metrics, including LPIPS and HaarPSI, reveal differences between masking strategies that are less apparent from PSNR and SSIM alone. These results provide practical guidelines for designing self-supervised sparse-view CT reconstruction methods and highlight the limitations of common independence assumptions in realistic imaging environments.
[CV-89] 3D Scene Graph Prediction: Generating Hierarchical Models from Partially Observed Environments IROS2026
链接: https://arxiv.org/abs/2607.10879
作者: Siyi H,Jared Strade,Hyungtae Lim,Luca Carlone
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at IROS 2026. Main paper: 8 pages, 3 figures, 3 tables. Includes a supplementary appendix
Abstract:Generating realistic 3D indoor scenes is an area of growing interest in computer vision and robotics. Existing methods, often motivated by applications such as interior design, generally focus on object layout generation within a single room. The generation of high-level scene structure, such as room-level layout and traversability, remains underexplored despite its importance for robotics applications. In this paper, we consider the case where a robot has explored part of an environment and needs to predict the unexplored parts to support downstream tasks such as exploration or object search. We propose a top-down framework for synthesizing hierarchical 3D scene graphs, including a room layer – describing the floor plan and traversability – and an object layer modeling object layouts within each room. For the room layer, we propose a novel mixed-domain graph diffusion model jointly predicting room categories, floor boundaries, and traversability between rooms. Via corruption and masking, this model supports partial constraints such as incomplete floor plans, avoiding the need for partially observed training data. For the object layer, we integrate an existing mixed discrete-continuous diffusion model for joint prediction of object categories, locations, sizes, and orientations within each room given the floor plan. We compare our method with state-of-the-art occupancy-based and LLM-based floor plan generation methods on a standard benchmark. Compared with an occupancy-based learning baseline, our method generalizes substantially better to out-of-distribution partial floor plans. We also demonstrate our integrated prediction pipeline on real-world scenes from robot-collected data, enabling prediction beyond explored areas.
[CV-90] X-GuideAR: An Augmented Reality Framework to Mitigate Radiation Exposure during Fluoroscopic Guidance
链接: https://arxiv.org/abs/2607.10873
作者: Mingxu Liu,Zixuan Liu,Ruchen Cai,Yu-Chen Ku,Suxi Gu,Amit Jain,Alejandro Martin-Gomez,Mehran Armand
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO); Image and Video Processing (eess.IV)
备注:
Abstract:Achieving optimal screw placement for orthopedic surgeries requires frequent alignment checks and multiple anatomical views under X-ray – a process known as “fluoro-hunting” that increases radiation exposure to patients and surgical teams. This work introduces X-GuideAR, an augmented reality (AR) framework for identifying optimal X-ray views, aimed at reducing radiation exposure while ensuring accurate screw placement. To exemplify the benefits of X-GuideAR, we focus on S2 alar-iliac (S2AI) screw placement. Our system provides radiation-free guidance for view acquisition and drilling by generating synthetic X-ray previews that accelerate fluoro-hunting. Once the required anatomical views are identified using these previews, a real X-ray is acquired, and the preview of the drilling trajectory is augmented onto it, facilitating precise screw placement with minimal additional radiation. A preliminary study involving eight S2AI trajectories performed by an expert spine surgeon demonstrated a 62.3% reduction in the number of X-rays. Post-procedure evaluations showed that trajectories done with X-GuideAR supported an average safe screw diameter of 12.95 mm compared to 5.9 mm under the conventional workflow, suggesting improved bony containment and potential biomechanical benefit. X-GuideAR shows great potential to reduce radiation exposure and streamline S2AI screw placement, offering a promising direction toward safer and more efficient surgeries.
[CV-91] AU-Guided Synthetic Video Generation for Micro-Expression Recognition
链接: https://arxiv.org/abs/2607.10860
作者: Pei-Sze Tan,Sailaja Rajanala,Yee-Fan Tan,Raphael C.-W. Phan,Huey-Fang Ong
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Micro-expression recognition is limited by the small scale, narrow demographic coverage, and restricted emotion labels of existing datasets. We introduce EquiME, a synthetic micro-expression dataset built from AU-guided image-to-video generation. EquiME contains 75K videos generated from 15K source face images across five target emotions, together with automatically inferred demographic metadata and video-quality measurements. We evaluate EquiME using frame-pair similarity, spatial variation, and no-reference perceptual-quality metrics, together with cross-dataset MER experiments on SAMM and CASME II. Models trained on EquiME achieve competitive cross-dataset performance on SAMM and CASME II and show comparatively low variation across the four evaluated architectures. This paper focuses on the dataset design, the structured AU-conditioning pipeline used for video generation, and the empirical evidence needed to assess EquiME as a synthetic MER resource. Project page: this https URL
[CV-92] Diversify Diffusion with Temperature Sampling and Variance-Corrective Time Shifting
链接: https://arxiv.org/abs/2607.10853
作者: Peizhuo Li,Emre Aksan,Alexandru-Eugen Ichim,Thabo Beeler,Olga Sorkine-Hornung
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Webpage: this https URL
Abstract:Diffusion models faithfully reproduce their training distribution, but also inherit its imbalances and leave rare or under-represented modes hard to reach. A natural inference-time remedy is to sample from the high-temperature target p^(\gamma)_0(x) \propto p_0(x)^\gamma for 0 \gamma 1 , which flattens dominant modes and lifts rare ones. However, naive score scaling while correctly reweighting modes also inflates the per-mode variance, breaking the reverse diffusion process and degrading sample quality. We introduce variance-corrective time shifting, a training-free fix that queries the network at a shifted timestep and scales the resulting score by \gamma , canceling the variance inflation while preserving the mode reweighting. The correction turns simple temperature sampling into a practical diversity knob for pretrained diffusion and flow-matching backbones with no retraining, and we demonstrate consistent gains at minimal cost to sample quality and condition fidelity across DiT, Stable Diffusion and Motion Diffusion models. We further show that the timing of the temperature intervention enables coarse-to-fine control: high-noise stages drive compositional diversity across modes, while low-noise stages drive local appearance variation under a fixed composition.
[CV-93] Learning To Focus: Anatomy-Guided Attention Regularization for Medical Image Classification
链接: https://arxiv.org/abs/2607.10851
作者: Tonmoy Hossain,Atiqur Rahman,Farhana Hossain Swarnali,Miaomiao Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Medical image classification models are ideally expected to identify diagnostically relevant regions while making predictions, yet standard classification losses rarely provide spatial supervision. Explicit supervision via anatomical shape information, such as segmentation masks of task-relevant anatomy, has been shown to guide the network toward regions relevant to the target prediction. However, obtaining such masks incurs substantial manual annotation effort and computational overhead. With the advent of segmentation foundation models that exhibit strong localization of anatomical structures across diverse imaging modalities, we leverage this capability to extract anatomical shape priors without the burden of training a dedicated segmentation model. In this paper, we propose a new framework, Locus, an anatomical attention regularization framework that leverages pretrained segmentation foundation models to guide a classifier’s attention toward diagnostically meaningful anatomical structures across diverse imaging modalities. Instead of enforcing pixel-wise alignment with the foundation-model-derived mask, we introduce a regularization term that adaptively balances attention between anatomical (foreground) and background regions, penalizing the classifier when background attention dominates. We validate Locus on eight diverse medical imaging datasets spanning dermoscopy, X-ray, histopathology, and cardiac MRI, showing consistent gains in classification performance alongside improved anatomically grounded attention.
[CV-94] Align and Segment: Unsupervised Learning for Building Segmentation From Misaligned Labels
链接: https://arxiv.org/abs/2607.10841
作者: Venkanna Babu Guthula,Oswin Krause,Dimitri Gominski,Hui Zhang,Johan Mottelson,Ankit Kariryaa,Nico Lang,Christian Igel
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: main draft with references is 17 pages
Abstract:Supervised learning for image segmentation typically requires spatially aligned image and label sets. When images and labels originate from different sources, the pairing may be misaligned, which can significantly deteriorate the performance of the learned models. This is especially common in remote sensing, when aerial or satellite images are co-registered with labels from another source (e.g., OpenStreetMap). In this work, we propose a novel approach for training on misaligned labels, where we simultaneously learn the label alignment. Our align and segment (AnS) approach builds on the spatial transformer module to transform the misaligned labels using an affine transformation to provide a better learning target for a canonical semantic segmentation network. We prevent shortcut learning of misaligned labels in these semantic segmentation networks through a self-supervised regularization loss and show that it is complementary to data augmentation, especially for systematically misaligned training data. A decisive characteristic of our AnS approach is that it learns without requiring any golden labels. We experimentally show on both synthetic and real-world data from different cities that our approach enables high-quality building segmentation and precise label-image alignment at the same time. Code and derived datasets are available at this https URL
[CV-95] OmniX: Any-view and Any-time 4D Reconstruction via Feed-forward Trajectory Fields ECCV2026
链接: https://arxiv.org/abs/2607.10840
作者: Yanqin Jiang,Tengfei Wang,Zhengwei Wang,Chenjie Cao,Junta Wu,Wenhan Luo,Weiming Hu,Jin Gao,Chunchao Guo
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by ECCV 2026, project page: this https URL
Abstract:Previous feed-forward 4D reconstruction methods either predict per-frame static point clouds, ignoring foreground motion, or estimate point cloud trajectories while being limited to small camera motions. This restricts their ability to aggregate observations over time and reconstruct complete dynamic scenes under large viewpoint changes. To address this limitation, we propose OmniX, a feed-forward 4D reconstruction framework that predicts dense 3D point trajectories for every pixel from videos with large camera motion. OmniX decouples dynamic motion modeling from static geometry prediction and represents motion using a compact set of dynamic tokens. By leveraging the sparse and low-rank structure of 3D motion, these tokens generate trajectory fields for all pixels across all images while efficiently preserving global interactions. To facilitate training, we further build an automatic UE5-based 4D data engine and introduce a large-scale dataset containing 80K scenes and 1.28M multi-view videos with full geometric annotations. OmniX achieves state-of-the-art performance on dense 3D point trajectory prediction and 3D point tracking, while also demonstrating competitive results on video depth estimation and camera pose estimation.
[CV-96] 3D-DefectBench: A Controlled Factorial Study of Vision-Language Model Evaluation Pipelines for Fine-Grained 3D Generation Defects
链接: https://arxiv.org/abs/2607.10826
作者: Zhenyu Zhao,Nanshan Jia,Jihyeon Je,Yifu Tang,Alvin Chan,Michael Spedden,Michael V. Palleschi,Sui Huang,Jingshen Wang,Zeyu Zheng
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Graphics (cs.GR)
备注:
Abstract:Automated evaluation is essential for scaling generative 3D systems, where exhaustive human review is costly and slow. However, the reliability of an automated judge depends on the entire evaluation pipeline, not only the underlying vision-language model (VLM), but also how assets are rendered, what visual evidence is provided, how the task is specified, and how human reference labels are constructed. We introduce 3D-DefectBench, a benchmark and framework for systematic analysis of VLM-based 3D defect detection pipelines. It complements holistic ratings and pairwise preferences with nine fine-grained binary defects spanning geometry, texture, and prompt adherence, providing actionable diagnostics for generator development and judge evaluation. Using a balanced factorial design, we vary four pipeline factors, VLM, camera protocol, visual input, and prompt schema, across 84 inference designs and approximately 3.2 million scored defect decisions, followed by staged validation on a broader set of frontier models. Model choice is the largest determinant of agreement with human labels, but the remaining factors also affect performance, interact with model selection, and can change the best configuration. Within the evaluated design space, a compact six-view RGB protocol performs comparably to denser multi-view settings and inputs augmented with depth or surface normals, making it a strong cost-effective default. Under this standardized pipeline, the best of 12 VLM judges still lag behind trained human labelers, while texture agreement drops sharply when expert-consensus labels are replaced by noisier silver labels. These findings show that automated judges should be evaluated as complete pipelines and calibrated across human reference regimes, rather than benchmarked only as standalone models. We release labels, prompts, predictions, and Croissant metadata on Hugging Face.
[CV-97] h-Flow: Flexible Flow-based Image Editing via Doobs h-Transform
链接: https://arxiv.org/abs/2607.10800
作者: Zehui Guo,Zhen Wang,Junwei Shu,Yang Li,Changbo Wang,Long Chen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Editing images with pre-trained text-to-image flow models typically requires carefully balancing target alignment with the desired prompt and source consistency with the original image. Existing approaches either rely on inversion-based pipelines or heuristic source-to-target trajectory constructions, which often depend on architecture-specific designs or are sensitive to hyperparameters. In this paper, we propose h-Flow, a training-free and theoretically grounded flow-based editing framework. Inspired by Doob’s h -Transform, we reformulate image editing as conditional generation under multiple terminal events corresponding to source consistency and target alignment. We first extend the classical h -Transform from SDE-based models to the deterministic RF framework by constructing an equivalent SDE with identical marginals. Within this formulation, we design dedicated h -functions for source consistency and target alignment, yielding closed-form reconstruction guidance and velocity-based semantic editing signals. We further introduce a velocity orthogonal decomposition to decouple reconstruction and editing directions, enabling a controllable trade-off between the two objectives. Extensive experiments demonstrate that h-Flow achieves effective, robust, and flexible editing across diverse scenarios. The code will be released soon.
[CV-98] Compositional Context Fine-Tuning Vision-Language Model for Complex Assembly Action Understanding from Videos ICRA2026
链接: https://arxiv.org/abs/2607.10797
作者: Hao Zheng,Jinyi Huang,Tiantian Zheng,Xun Xu,Tuka Alhanai
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by ICRA 2026. Video Understanding; Vision Language Model; Multi-modal LLM; Action Recognition; Assembly
Abstract:Assembly action understanding is a key enabler for effective human-robot collaborative assembly, yet it remains challenging due to subtle motions and fine-grained hand-object interactions. We adapt vision-language models (VLMs) to this challenging domain with Compositional Context Fine-Tuning (CCFT), a method that decomposes assembly actions into semantic elements (Verb, Object, Tool) and fine-tunes VLMs to recognize each action element using templated question-answering pairs. This approach ensures near-deterministic outputs. To enable efficient and effective multi-task learning under limited data, a Layer-Partitioned Alternating Training (LP-AT) method is presented, which assigns distinct model layers to recognize specific action elements through element-specific low-rank adapters. LP-AT alternates weight updates across element-specific adapters, reducing cross-task interference while enabling per-adapter hyperparameter optimization. Furthermore, we create HA-ViD-VQA and IKEA-ASM-VQA datasets from existing assembly video datasets. Extensive experiments on these datasets demonstrate that our method consistently outperforms strong action recognition baselines while providing interpretable element-level predictions that can support diverse downstream applications.
[CV-99] Mixture of Cognitive Experts in Large Vision-Language Models
链接: https://arxiv.org/abs/2607.10796
作者: Robert Wijaya,Ngai-Man Cheung
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Large Vision Language Models (LVLMs) require strong reasoning over both visual and textual input. Recent work suggests that cognitive elements, especially diverse representations and metacognition, correlate with better performance. Many of the needed perceptual functions are already provided by specialized domain-specific computer vision models, which act as the perceptual subsystem for detecting objects, localizing them, inferring states, recovering spatial layout, and reading text. The key challenge is to integrate these multi-encoder experts into a trustworthy, interpretable, and coherent representation that improves verifiability and reduces hallucinations. This is difficult because vision-language questions span different cognitive levels, yet most LVLM pipelines apply the same perception-reasoning routing regardless of the demand of each query. We propose an evidence-driven multimodal reasoning framework that utilizes a Bloom-inspired taxonomy as a hierarchical reasoning protocol. The two-stage cognitive verbalization first produces a Literal Evidence Summary by decomposing expert outputs into short, atomic evidence statements. It then performs Bloom Verbalization to turn these evidence items into a staged reasoning trace, and a lightweight Reasoning Trace Module quantitatively analyzes the trace to make evidence usage and reasoning progression explicit. Through this integration, we observed several improvements in perception and reasoning abilities. Moreover, the trace module provides quantitative evidence that different queries induce different cognitive entry levels and evidence-use trajectories that enable fine-grained analysis.
[CV-100] MAC-Splat: Multi-Attribute Consistency for High-Fidelity Sparse-View Reconstruction ECCV2026
链接: https://arxiv.org/abs/2607.10792
作者: Jinqian Yang,Yichen Wu,Wanhua Li,Haokun Lin,Renzhen Wang,Xiangchu Feng,Xixi Jia
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to the European Conference on Computer Vision (ECCV 2026)
Abstract:Reconstructing high-fidelity 3D scenes from sparse-views remains a central problem in generalizable neural rendering. Existing generalizable 3D Gaussian Splatting (3DGS) methods often exhibit geometric artifacts in sparse-view settings, since supervision based solely on 2D photometric losses cannot resolve depth and correspondence ambiguities. To address this issue, we propose MAC-Splat, a training framework built around direct 3D consistency supervision. MAC-Splat builds on the MASt3R geometric backbone and a frozen DINOv3 encoder to obtain semantically informed 2D correspondences, which serve as geometric anchors for 3D supervision. Using these anchors, we define the Multi-Attribute Consistency (MAC) loss. This objective jointly regularizes the 3D attributes of matched Gaussians, including their position, shape, and appearance, by enforcing agreement in a common world coordinate frame. The formulation is robust to outliers and respects the geometry of covariance matrices, which leads to stable training under sparse-view conditions. Experiments on ScanNet++ show that MAC-Splat outperforms strong baselines, with particularly large gains under different overlap regimes. In particular, it improves average PSNR over Splatt3R by more than 4.5 dB, reduces LPIPS, and maintains performance as the camera pose gap increases. These results indicate that a direct, multi-attribute 3D consistency objective, when combined with high-quality correspondences, is effective for addressing the ill-posed sparse-view reconstruction problem.
[CV-101] oward Efficient Weakly Supervised Semantic Segmentation Using Only Low-Magnification Histopathological Images
链接: https://arxiv.org/abs/2607.10783
作者: Dung Minh Do,Nhat-Thanh Huynh,Duc Minh Huynh,Doanh C. Bui,Khang Nguyen
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Accepted at MAPR 2026
Abstract:Whole-slide images (WSIs) provide rich tissue-level and cellular-level information, but storing and transmitting high-magnification pathology data is resource-intensive. Moreover, annotating WSIs at the pixel level is labor-intensive and time-consuming. Therefore, it is important to investigate whether low-magnification pathology images with limited annotations (i.e., image-level instead of pixel-level labels) can achieve performance comparable to high-magnification images. This paper presents a systematic benchmark study on weakly supervised histopathological image segmentation under different low-resolution storage settings. Starting from high-resolution image patches, we simulate lower-magnification inputs and reconstruct them to the original size using interpolation and deep learning-based reconstruction methods before applying the weakly-supervised segmentation pipeline. This framework enables a quantitative evaluation of how weakly supervised methods respond to different levels of resolution degradation. Experimental results show that reconstruction quality metrics alone are insufficient to predict downstream segmentation performance. In particular, the study identifies a critical degradation point where the localization of small-scale structures declines significantly. These findings provide practical guidance for designing efficient digital pathology storage systems while maintaining reliable automated analysis. Code is available at this https URL
[CV-102] Is Energy Guidance All You Need? Training-Free Norm Injection for Driving World Models
链接: https://arxiv.org/abs/2607.10781
作者: Xiyan Su,Frank Diermeyer,Markus Lienkamp
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注: Accepted to Robotics: Science and Systems 2026 Robot World Models Workshop
Abstract:Driving world models built on large video-diffusion backbones generate realistic scenes but are hard to control: enforcing a traffic norm typically means retraining the backbone or conditioning it on hand-built layouts. We ask whether controllability requires training at all. Our experiment shows that a rectified-flow driving world model, which jointly generates future video and a planned ego trajectory, can have its planned trajectory steered entirely at sampling time by differentiable energy functions that encode driving norms, without knowledge-specific retraining of the diffusion backbone. Concretely, we demonstrate that a world model built on Open-Sora 2.0 MM-DiT backbone can be steered to brake at a counterfactual target by injecting energy guidance at sampling time. However, we find that the generated video does not yet follow the steered trajectory through the backbone’s joint self-attention and identify the cross-stream coupling as a crucial requirement for end-to-end-controllable rollouts.
[CV-103] RED-Sphere: Hyperspherical Residual Edge Debiasing for Cross-Population Fundus Disease Domain Generalization
链接: https://arxiv.org/abs/2607.10777
作者: Yan Lin,Ziheng Wang,Shuang Chen,Amir Atapour-Abarghouei,Stephen McGough
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 21 pages, 6 figures, 3 tables. Preprint
Abstract:Medical image classifiers are often trained within one source population, yet clinical deployment requires robustness to patients whose appearance, acquisition style, and disease prevalence differ from the source cohort. Existing fairness and robustness methods often require group supervision or treat appearance variation as an undifferentiated nuisance, which is insufficient when population-correlated low-level cues and lesion evidence share edge and texture structure. We study a strict source-only cross-population setting, where external populations are unseen during optimization, validation, scheduling, hyperparameter and model selection. We propose RED-Sphere, a plug-and-play robustness framework for image classification under unseen population shifts. It estimates shortcut-sensitive nuisance responses with an edge and feature energy prior, attenuates dominant responses through residual soft gating, regularizes masked nuisance views with counterfactual-inspired consistency and separation losses, and predicts labels with normalized spherical prototypes. It favours angular semantic evidence over source-correlated activation magnitude while preserving lesion structure. Although demonstrated on 2D Scanning Laser Ophthalmoscopy (SLO) fundus classification for Age-Related Macular Degeneration (AMD) and Diabetic Retinopathy (DR), RED-Sphere is not tied to retinal anatomy: the same principle can be adapted with modality-specific nuisance priors wherever appearance shortcuts and semantic evidence are entangled. Under a strict White-only Harvard-FairVision protocol, RED-Sphere improves held-out macro-F1 across all 20 task and backbone comparisons, with average gains of 1.28 and 2.98 F1 points on AMD and DR. Gains in AUC and PR-AUC, visual diagnostics, ablations, and sensitivity analyses further support stronger external semantic alignment and more stable angular disease geometry.
[CV-104] OLiD: Bridging the Architecture Gap in Vision Foundation Model to LiDAR Pretraining via Token Lifting for Distillation IROS
链接: https://arxiv.org/abs/2607.10762
作者: Sutharsan Mahendran,Darshana Priyasad,Kaushik Roy,Tharindu Fernando,Sridha Sridharan,Clinton Fookes,Peyman Moghadam
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Robotics (cs.RO)
备注: Accepted to The IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2026
Abstract:Cross-modal distillation from Vision Foundation Models (VFMs) to LiDAR backbones has recently emerged as a self-supervised pretraining strategy that reduces reliance on dense point-wise annotation for 3D scene understanding. However, existing distillation pipelines typically treat the VFM as a frozen feature source and train a heterogeneous 3D backbone to match fixed image embeddings, forcing the student to bridge both the modality gap and the cross-architecture gap between dense ViT token representations and sparse 3D encoders. We propose TOLiD, a self-supervised pretraining method for LiDAR representation learning that addresses this gap by coupling a LiDAR backbone with a student Vision Transformer (ViT) initialized from a frozen VFM teacher and applying supervision over compatible patch-token representations. TOLiD converts the set of point features within each image patch frustum into a token using Frustum Pooling followed by Frustum Attention, and performs token-level distillation with visibility masking. For LiDAR-only deployment, we lift token features back to per-point representations using masked bilinear sampling to avoid patches that have limited LiDAR points. We extensively evaluate TOLiD on five heterogeneous LiDAR datasets and four cross-sensor adaptation pairs, demonstrating improved transfer with frozen backbones and lightweight heads.
[CV-105] riCons-Pose: Triangle-Invariant Geometric Consistency Learning for Category-Level Object Pose Estimation
链接: https://arxiv.org/abs/2607.10754
作者: Zuzhi Yang,Shuai Wang,Mounir Kaaniche,Ziwei Li,Zhiming Cheng,Zhidong Zhao,Chenggang Yan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 13 pages, 4 figures
Abstract:Category-level object pose estimation is a crucial yet challenging task in both academia and industry, and has achieved remarkable success by leveraging keypoint-based correspondence paradigms. However, most existing methods increasingly rely on stronger feature learning while overlooking whether the established correspondences are geometrically stable across diverse perturbations. This often results in fragile pose recovery under intra-class shape variations and occlusions. To tackle this challenge, we develop a novel Triangle-Invariant Geometric Consistency Learning for Category-Level Object Pose Estimation (TriCons-Pose) to anchor stable keypoints and aggregate pose-invariant cues, yielding reliable canonical mapping and accurate pose estimation. Specifically, a Structure-Consistent Keypoint Detector (SCKD) is designed to identify robust keypoints by enforcing cross-view structural consistency via normalized pairwise distance matching. Moreover, we propose a Pose-Invariant Geometric Aggregator (PIGA) to augment keypoint representations by injecting triangle-based pose-invariant descriptors into a local-to-global attention mechanism. The proposed framework is optimized using standard objective functions while incorporating an additional geometry consistency loss. Extensive experiments on REAL275, CAMERA25, and HouseCat6D datasets demonstrate the effectiveness of the proposed approach.
[CV-106] Water Reflection Detection Using Symmetric Attention
链接: https://arxiv.org/abs/2607.10749
作者: Shuxuan Yao,Chengjia Wang,Jianyuan Sun,Junyu Dong,Xinghui Dong
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Reflections of water pose a significant challenge for computer vision systems, as standard deep learning models frequently confuse objects with their mirror images, producing spurious false positives and negatives in tasks such as object detection and semantic segmentation. As a result, detecting reflection axes in natural-water scenes is pivotal for reliable object detection and scene understanding. To mitigate this issue, we leverage the intrinsic imperfect reflective symmetry of water and introduce a Symmetry-Aware Water Reflection Detection Network, namely, SAWRD-Net, that couples dihedral group-equivariant convolutions with a matrix-decomposition decoder in an end-to-end framework. First, dihedral group convolutional layers extract geometry-consistent feature maps that explicitly encode both rotational and mirror symmetries. A Multi-scale Reflection Equivariant block then aggregates features across scales and employs a symmetric-attention mechanism to highlight reflection-relevant regions. The proposed matrix-decomposition decoder factorizes high-dimensional features into compact low-rank parameter and confidence spaces, after which the network directly regresses keypoints on the reflection axis. Then a robust principal component analysis fits the final axis. Evaluated on the largest available water reflection scene data set, SAWRD-Net achieves a true-positive rate of 0.890 against human annotations, outperforming all existing water reflection detectors.
[CV-107] raj-VLN: Learning Pixel-Space Interaction via Autoregressive Trajectory Generation
链接: https://arxiv.org/abs/2607.10744
作者: Changfei Fu,Guangcheng Chen,Wenjun Xu,Hong Zhang
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注:
Abstract:Benefiting from the powerful priors embedded in large-scale pre-training data and the emerging commonsense reasoning ability, large language models (LLMs) have shown unprecedented generalization capabilities in many research fields. Recently, projecting visual embeddings into the language space via vision-language models (VLMs) to achieve sim-toreal and cross-scene generalization has become a prevailing paradigm in the field of Vision-and-Language Navigation in Continuous Environments (VLN-CE). VLN requires an embodied agent to navigate through unseen environments following natural linguistic instructions. We emphasize that a VLN task can be decomposed into a sequence of sub-tasks, each corresponding to a process of 3D spatial interaction with the environments described by instructions such as “walk to the end of the sofa and turn left.” However, such spatial interactions involving moving into the image along the direction of depth sensing are puzzling for VLMs as they were predominantly trained on conversations with RGB images. Rather than incorporating depth or 3D geometric information-which VLMs rarely encounter during pretrainingwe propose an alternative approach: fine-tuning VLMs to learn navigation interactions directly in 2D pixel space through autoregressive trajectory generation. Given a linguistic instruction and historical observations, our model sequentially predicts a series of pixel coordinates, drawing a trajectory from the bottom center of the current observation. While prior work has proved that pixel-goal supervision outperforms learning of discrete actions, our experiments further verify that the supervision of pixel-space trajectory significantly enhances VLN performance. Moreover, we demonstrate that our flagship model achieves state-of-the-art level performance with relatively limited computational resources and training data.
[CV-108] Action Map Policy: Learning 3D Closed-loop Manipulation via Pixel Classification
链接: https://arxiv.org/abs/2607.10706
作者: Haojie Huang,Zhang Ye,Linfeng Zhao,Boce Hu,Mingxi Jia,Yu Qi,Ahmed Agha,Dian Wang,Robert Platt,Robin Walters
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Project Website: this https URL
Abstract:The action space poses a major challenge in robot learning, since it is often high-dimensional, can span long time horizons, and frequently admits multi-modal optimal solutions. A good choice of action representation and loss function can help to address these concerns, but there are often trade offs. We propose Action Map Policy (AMP), which casts 3D closed-loop manipulation policy learning as a classification problem in image space. While classification has been an effective formulation in generative language models, applying it to robot action learning is difficult because naively discretizing high-dimensional continuous actions explodes the token vocabulary. Our key idea is to project 3D actions onto the camera image planes and treat each pixel location as a discrete class, thus controlling dimensionality while retaining multi-modality. This method supports millimeter-level precision for high-dimensional actions without requiring a prohibitively large vocabulary, while preserving fine-grained pixel-wise visual signals. Furthermore, it can predict the entire action chunk in a single forward pass, avoiding complex noise scheduling and iterative denoising while achieving substantially faster inference than diffusion policies. Experiments on various manipulation tasks show that AMP outperforms strong baselines, achieving higher success rates, faster inference, and enhanced spatial reasoning.
[CV-109] On the modality gap and the contrastive loss in multi-modal representation learning
链接: https://arxiv.org/abs/2607.10698
作者: Fabian Mager,Hiba Nassar,Lars Kai Hansen
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:We study the modality gap in CLIP-style dual-encoder contrastive learning, where image and text embeddings remain misaligned despite being trained in a shared space. We argue that the gap is induced by a failure of the InfoNCE formulation with independent encoders. We conduct a uni-modal experiment with two independent encoders and identical initialization conditions and find that InfoNCE actively generates a gap at low temperatures. We provide a theoretical analysis of this phenomenon and show that the modality gap is indeed a mode-failure of InfoNCE, but only at low temperatures. We propose a simple modification called xNCE, which uses intermodal as well as intra-modality negative contrastive pairs. xNCE matches retrieval performance on MS-COCO while consistently reducing the gap even at low temperatures. Notably, xNCE improves zero-shot classification over the InfoNCE baseline across all benchmarks, whereas high-temperature InfoNCE and regularized InfoNCE both fail to do so, demonstrating that xNCE reduces the modality gap without sacrificing the discriminative geometry needed for transfer.
[CV-110] Effective Synthetic Image Detection via Noise Residual Clustering
链接: https://arxiv.org/abs/2607.10695
作者: Caihui Yan,Gang Cao,Huawei Tian,Zhen Li,Yuhang Zhai
类目: Computer Vision and Pattern Recognition (cs.CV); Cryptography and Security (cs.CR)
备注:
Abstract:The rapid advancement of generative artificial intelligence (AI) has made synthetic images remarkably realistic, posing security threats such as misinformation and fraud. It is significant to detect the synthetic image in the manner of passive and blind image authentication. Most existing detectors rely on supervised training with large labeled datasets, leading to high costs and degraded performance on unknown generative models. To attenuate such deficiencies, we propose a training-free detection method. Specifically, noise residual fingerprints are first extracted by a simple yet effective pre-trained Noiseprint++ model. Then multi-scale features are further extracted from such residual by a frozen Vision Transformer (ViT), followed by adaptive weighted fusion. Only a few real image samples are used needed to initialize the clustering centers for unsupervised K-Means, distinguishing real and synthetic images without training. Extensive evaluations on four benchmark datasets show that our proposed scheme achieves an average accuracy of 82.2%, outperforming the state-of-the-art detectors on generalization ability. Superior performance is gained on the popular diffusion type of synthetic images, and the effectiveness of each module is validated by ablation studies. Source code will be publicly available at this https URL.
[CV-111] Incremental Online Scene Reconstruction by 3D Gaussian Triangulation
链接: https://arxiv.org/abs/2607.10690
作者: Yanjin Zhu,Shaofan Liu,Jianke Zhu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Incremental scene reconstruction is essential for real-world applications. Although 3D Gaussian Splatting shows strong potential, most existing approaches require offline conversion of the optimized Gaussians into an intermediate implicit field for explicit mesh extraction, which hinders seamless integration with downstream tasks. To address this limitation, we propose a novel online framework that incrementally reconstructs and updates high-fidelity explicit meshes by directly triangulating a dense geometric Gaussian representation, which supports both high-quality rendering and incremental surface reconstruction. Moreover, we present a direct meshing algorithm that efficiently extracts and updates the mesh from the Gaussian set. To ensure mesh accuracy, we enforce a plane-based pulling constraint that dynamically aligns 3D Gaussian primitives to the approximated local surface. Furthermore, our framework significantly reduces memory and computational overhead during long-sequence processing by dynamically freezing fully optimized historical regions. Experiments on public datasets demonstrate that our method outperforms conventional Gaussian-based methods on both rendering quality and reconstruction accuracy.
[CV-112] HyperBank: A Differentiable Bank of Classical Priors for Few-Shot Spheroid Microscopy Segmentation ICIP2026 MICRO ICIP
链接: https://arxiv.org/abs/2607.10684
作者: M. Průšek,A. Novozámský,F. Šroubek,T. Volfová,V. Svobodová Pavlíčková,S. Rimpelová
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted for publication in the IEEE Xplore ICIP 2026 Workshop Proceedings, Computational Optical Microscopy Satellite Workshop of the 2026 IEEE International Conference on Image Processing (ICIP), Tampere, Finland
Abstract:Few-shot spheroid segmentation must adapt to new cell lines, microscopes, and illumination conditions from only a small set of annotated images. While foundation few-shot segmenters can be accurate, their large opaque backbones make it difficult to understand which visual cues drive success or failure. We study this question with HyperBank, a differentiable bank of classical image-processing operators combining Frangi vesselness, a Sauvola threshold pyramid, structure-tensor responses, gradient magnitude, and Laplacian-of-Gaussian filters. HyperBank is fitted on the annotated support images and evaluated on disjoint held-out images across three independently acquired spheroid datasets. We treat it not as a general replacement for foundation models, but as a compact, interpretable few-shot microscopy pipeline and an analytic-prior probe of which classical cues carry the few-shot signal. The results show that, adapted on the same few annotated support images, a compact bank of analytic priors is competitive with, and on small-cluster, contrast-driven data can outperform, much larger foundation models, while those models remain stronger on externally sourced, texture-dominated spheroids. Leave-one-family-out ablations indicate that the useful few-shot signal is distributed across operator families and strengthened by support-set-tuned morphology.
[CV-113] Answer-Conditioned Chain-of-Thought Distillation for Few-Shot Industrial Vision with Small VLMs
链接: https://arxiv.org/abs/2607.10666
作者: Shubham Rao
类目: Computer Vision and Pattern Recognition (cs.CV); Materials Science (cond-mat.mtrl-sci); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 11 pages, 5 figures, 8 tables
Abstract:Deploying AI-based visual inspection in manufacturing is hard because requirements change often, new defect types appear, and large labeled datasets are rarely available. We propose answer-conditioned chain-of-thought (CoT) distillation for rapidly adapting small vision-language models (VLMs) to new industrial tasks using minimal labeled data. A frontier VLM receives each training image along with its correct label and generates a justified visual explanation. A 3B-parameter model is then fine-tuned on these reasoning-augmented examples via LoRA. By conditioning on correct answers, we ensure all training reasoning is directed toward the correct conclusion, which is critical because frontier models score as low as 24.1% on our hardest task. We validate on four industrial classification tasks spanning three image modalities using only 18 to 30 labeled images per task. Across 4 seeds per task (32 training runs), our method outperforms direct fine-tuning on all 16 seed-task combinations, with mean improvements of +1.7 to +4.4 percentage points. A controlled equal-budget experiment confirms the improvement comes from reasoning quality, not additional training steps. An unconditioned baseline demonstrates that with out answer-conditioning, wrong reasoning degrades performance by 17.8 percentage points. On weld radiograph classification, the fine-tuned 3B model outperforms GPT-4.1 by 10.0pp using just 24 training images.
[CV-114] Spectral Heat Flow for Conservative Token Condensation in Vision-Language Models ICML2026
链接: https://arxiv.org/abs/2607.10640
作者: Zhaoyang Li,Yanjun Li,Wangkai Li,Yujia Chen,Tianzhu Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by ICML 2026
Abstract:Vision-Language Models (VLMs) are costly at inference time because they must process long sequences of visual tokens. Existing token pruning methods often degrade under high compression by blindly discarding information, breaking spatial structure or collapsing diversity. We propose SpecFlow, a training-free framework that shifts the paradigm from destructive pruning to conservative condensation, strictly enforcing spatial coverage and statistical conservation to ensure stability. Treating visual tokens as nodes in a k NN graph, SpecFlow (i) computes a stable importance field via spectral heat flow to preserve structural coherence, (ii) allocates budgets via adaptive spatial partitioning to guarantee coverage, and (iii) aggregates discarded information into coreset sinks to maintain statistical conservation. The method is plug-and-play, requires no fine-tuning, and is compatible with FlashAttention. Experiments confirm that our SpecFlow outperforms SOTA methods across tasks, VLM architectures, and pruning ratios. Notably, LLaVA-1.5 with SpecFlow retains 95.6% of original performance despite pruning 88.9% of visual tokens, offering an exceptional efficiency-accuracy balance. Code is available at this https URL
[CV-115] Spectral Consistent Flow for One-step 3D Medical Image Translation
链接: https://arxiv.org/abs/2607.10627
作者: Haoqing Li,Jun Shi,Mingchao Li,Zehua Zhu,Qiwei Jia,Jiong Shi,Hong An
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:We present Spectral Consistent Flow (SC-Flow), a 3D medical image translation framework with a single function evaluation (1-NFE) in the latent space. This approach reformulates medical image translation as a stochastic Brownian bridge process that directly constructs a mapping between source and target modalities by predicting the support regularized mean velocity field. To mitigate modality entanglement, over-smoothing, and artifacts induced by the implicit low-pass modulation of the latent average velocity, we introduce a Spectral Consistency Corrector that dynamically regularizes the evolution of the power spectral density via learnable frequency-domain gain modulation. This mechanism establishes an explicit bridge between spatial textures and spectral energy flow, enabling the model to recover fine-grained anatomical fidelity while maintaining global structural coherence. Extensive experiments on four datasets demonstrate that SC-Flow delivers significantly more accurate, consistent, and robust performance across various translation scenarios.
[CV-116] LATO.2: Factorized 3D Mesh Generation with Vertex and Topology Flow
链接: https://arxiv.org/abs/2607.10623
作者: Hang Long,Tianhao Zhao,Junkai Lin,Youjia Zhang,Huipeng Guo,Rendong Liang,Jiale Xu,Jozef Hladký,Matthias Nießner,Wei Yang
类目: Graphics (cs.GR); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Flow matching over carefully designed latent representations has recently emerged as a powerful paradigm for topology-aware mesh generation. Existing approaches, however, model vertices and connectivity jointly in a joint latent space, entangling continuous vertex geometry with discrete combinatorial structure; this complicates flow learning and manifests as drifting vertices and broken surfaces. We present LATO.2, a factorized flow matching framework that decomposes mesh generation into a vertex flow followed by a connectivity flow conditioned on the realized vertices, with both stages anchored to a shared coarse voxel scaffold. Dedicated VAEs underpin the two stages, recovering vertices at sub-voxel precision and embedding discrete connectivity into a continuous latent space. We demonstrate two advantages unique to this factorization: (i) part-wise generation, in which the scaffold is partitioned and each part synthesized at full latent capacity, yielding substantially higher-resolution meshes than a monolithic latent permits; and (ii) topology-adaptive editing, in which manipulating first-stage vertices induces the corresponding connectivity without re-optimization. Experiments show that LATO.2 surpasses state-of-the-art topology-aware mesh generators in geometric fidelity and connectivity quality.
[CV-117] WasteAssistant: Regulation-Guided Visual Question Answering Framework for Intelligent Waste Segregation and Sustainable Managemen
链接: https://arxiv.org/abs/2607.10610
作者: Khush Kataruka,Harshit Maurya,Anuja Vats,Murari Mandal,Kiran Raja,Praveen Kumar Chandaliya
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 12 Pages, 5 Figures
Abstract:Efficient waste segregation is critical for sustainable urban management and environmental governance. Existing automated systems are limited by single-modality visual processing, insufficient contextual understanding, and weak regulatory alignment. To address these issues, we propose a language-guided vision-AI framework that integrates vision-language models and multimodal large language models for joint visual-linguistic reasoning. This framework implements a visual question answering paradigm aligned with India’s Solid Waste Management Rules 2016. We construct a new WasteVQA dataset with 13,500 question-answer pairs across 21 waste categories. Experiments show that the BLIP-based model achieves a BLEU score of 0.8291 and a BERTScore of 0.9273, outperforming traditional CNN-based methods. This work improves source-level segregation accuracy, ensures regulatory compliance, and supports scalable deployment for municipal and citizen-facing waste management, promoting multimodal AI in sustainable urban infrastructure. The source code and dataset are available at: this https URL
[CV-118] End-to-End Real-Time Drone-Based Person Detection Framework Using Deep Learning
链接: https://arxiv.org/abs/2607.10605
作者: Payel Sarmah,Ayush Ranjan,Piyush Kaushik Bhattacharyya,Anil Kr. Shaw,Pradip Kr. Das
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:In recent years, Unmanned Aerial Vehicles (UAVs) or drones have gained rapid response in terms of security, search and rescue (SAR), border surveillance, etc. Existing monitoring frameworks often struggle to maintain detection consistency when targets undergo significant scale variations due to altitude changes, leading to critical information gaps. To address this issue, this work proposes an integrated real-time detection pipeline for detecting targets through the wireless live drone video feed. Build upon YOLOv8-nano architecture, extensive flight experiments were conducted to determine the detection performance across multiple flight altitudes. Trained on VisDrone2019 dataset, the results of YOLOv8-nano model achieves 57.4%, 41%, 44.8% and 20.3% in precision, recall, mAP and mAP50:95 respectively. While demonstrating on real environment, this analysis revealed that the algorithm achieves near-total detection reliability at altitudes between 16 and 25 meters with the detection frame rate consistently maintained above 41 FPS and reaching a peak of 50 FPS. However, the goal of this work is to enable real-time person detection from an aerial platform via wireless transmission. This approach effectively addresses the dual challenges of identifying targets at varying scales and ensuring near-to-accurate localization during aerial observation.
[CV-119] Anomalous Frame Detection by Grouping Frame Similarities between Two Videos Computed by Vision-Language Model to Extract Expert Workers Unique Actions
链接: https://arxiv.org/abs/2607.10598
作者: Ryo Sakai,Yongpeng Cao,Nobutaka Kimura
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 11 pages, 6 figures, 2 tables
Abstract:Maintenance of critical infrastructures, such as railways and power plants, is essential for operational safety and reliability. However, the declining number of skilled maintenance workers poses a serious challenge to sustaining these operations, highlighting the need to effectively transfer expert know-how to less experienced workers. Although traditional interview-based approaches have been used to elicit maintenance skills, they struggle to capture know-how that experts themselves may not consciously recognize. To address this gap, we proposed a method that detects anomalous frames of candidate actions including know-how by comparing a video of manual-based work with that of expert maintenance workers. In a simulated maintenance experiment involving a distribution board, our method targeted 11 types of actions not described in the manual and achieved a 66.9% extraction rate, marking a 50-percentage-point improvement over conventional techniques. These findings underscore the effectiveness of our approach in revealing hidden maintenance knowledge, thereby contributing to enhanced skill transfer and workforce development in critical infrastructure maintenance.
[CV-120] Benchmarking UAV-based Vehicle Re-Identification under Simulated Weather Conditions
链接: https://arxiv.org/abs/2607.10583
作者: Vu Minh Tran,Khang Nguyen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at the 2026 International Conference on Multimedia Analysis and Pattern Recognition (MAPR 2026)
Abstract:UAV-based vehicle re-identification (ReID) has emerged as a promising technique for traffic surveillance, urban monitoring, and public-safety applications thanks to the flexible viewpoints and wide-area coverage provided by unmanned aerial vehicles. However, despite recent progress on UAV-based vehicle ReID benchmarks, the robustness of existing methods under adverse weather remains insufficiently studied. This is important because weather degradation can significantly affect the fine-grained appearance cues required for reliable vehicle matching in aerial imagery, especially under small object scale, viewpoint variation, and complex backgrounds. In this paper, we present a controlled comparative study of three representative recent vehicle ReID methods, namely CLIP-ReID, MSINet, and AdaSP, on two UAV-based benchmarks, VRU and UAV-VeID. To ensure consistent robustness evaluation, we generate synthetic foggy and rainy variants of both datasets using an analytical weather-effect pipeline while preserving the original identities and data splits. All methods are then trained and evaluated under matched clean, foggy, and rainy conditions. Experimental results show that adverse weather consistently degrades retrieval performance across both datasets, with rain causing larger drops than fog in nearly all settings. Among the evaluated methods, AdaSP demonstrates the strongest robustness, achieving 93.0% and 88.5% mAP on VRU-Large, and 88.7% and 76.2% mAP on UAV-VeID-Test under foggy and rainy conditions, respectively. Overall, our findings show that simulated adverse weather substantially increases the difficulty of UAV-based vehicle ReID, reveals clear robustness differences among recent methods, and highlights the need for weather-aware model design and evaluation protocols in future aerial ReID research. The code is released at this https URL.
[CV-121] DiffUE: Enhancing Utility-Unlearnability Trade-off of Unlearnable Examples via Diffusion Autoencoders ECCV
链接: https://arxiv.org/abs/2607.10580
作者: Syed Irfan Ali Meerza,Oktay Ozturk,Amir Sadovnik,Jian Liu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: To appear at the European Conference on Computer Vision (ECCV), 2026
Abstract:AI models are increasingly trained on personal images scraped from social media and public platforms, often without consent, leading to serious privacy violations, such as unauthorized facial recognition and targeted advertising. To counter this, researchers have developed unlearnable examples (UEs), images modified with imperceptible noise to prevent AI models from extracting meaningful information. However, existing UE methods primarily rely on pixel-space noise, which can be bypassed by relearning strategies such as adversarial training, image transformation, and compression. While some techniques improve robustness, they often come at the expense of significant degradation in image utility and perceptual quality. In this paper, we introduce DiffUE to overcome these limitations by injecting noise into the semantic space of images instead of the pixel space. Instead of corrupting pixel values, DiffUE modifies high-level semantic features of images, ensuring robust unlearnability while preserving visual quality and utility. By leveraging a diffusion-based autoencoder framework to manipulate semantic features, DiffUE generates purposeful, natural-looking modifications that effectively resist advanced relearning strategies. Extensive experiments on four datasets, CIFAR-10, CIFAR-100, CelebA-HQ, and ImageNet, as well as a subjective user study, demonstrate that DiffUE significantly enhances the trade-off between image quality and unlearnability, offering a more robust and effective solution for safeguarding personal data in an increasingly exploitative AI landscape.
[CV-122] Why Domain Matters: Domain-Aware Benchmarking of Underwater Object Detection and Annotation Quality
链接: https://arxiv.org/abs/2607.10575
作者: Melanie Wille,Dimity Miller,Tobias Fischer,Scarlett Raine
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Underwater object detection is strongly affected by domain shift, where performance can vary significantly across different locations, habitats, and deployment conditions. However, detector performance is typically evaluated using aggregate metrics that hide failures in specific environments, while existing domain generalization benchmarks often rely on synthetic variations that do not reflect real-world conditions. We introduce a framework that characterizes underwater images by appearance, scene composition, and acquisition geometry to assign domain labels. Using this framework, we perform the first systematic study of how domain factors influence both human annotation quality in underwater object detection datasets and deep learning-based detector performance, revealing substantial domain-dependent discrepancies. By incorporating physically meaningful domain labels, domain shift becomes something we can characterize, measure, benchmark, and act on. We highlight how this can be used to guide data collection and annotation, design more informative benchmarks, and assess detector robustness across diverse underwater environments.
[CV-123] Quantum Compressed Sensing CT Reconstruction Algorithm Based on Penalized Weighted Least Squares and Guided Total Variation
链接: https://arxiv.org/abs/2607.10566
作者: Yuwen Zhang,Yujie Liu,Ao Wang,Yikuang Yuluo,Shuangyang Zhong,Haijun Yu,Yixing Huang
类目: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注: 14 pages, 11 figures
Abstract:Objective. Existing quadratic unconstrained binary optimization (QUBO)-based sparse-view computed tomography (CT) reconstruction neglects photon-counting statistics and anatomical heterogeneity. We address both limitations within the QUBO this http URL. We propose a quantum compressed-sensing CT method combining penalized weighted least squares (PWLS) and guided total variation (GTV). PWLS weights projection residuals by photon-count reliability, whereas GTV uses gradients from a prior image reconstructed by the simultaneous algebraic reconstruction technique (SART) to preserve edges and suppress noise in homogeneous regions. After binary encoding, both terms form a unified QUBO model. Experiments used four 40 times 40 CT images under a 10-view fan-beam geometry with Poisson noise. Comparisons included conventional reconstruction methods, QUBO variants, gradient descent, simulated annealing, and a D-Wave hybrid quantum-classical this http URL results. PWLS-GTV achieved the best reconstruction quality across all cases. In the representative chest case, it reached a peak signal-to-noise ratio (PSNR) of 36.64 dB, compared with 22.48 dB for SART, the best conventional baseline. GTV consistently outperformed conventional total variation. Simulated annealing and the D-Wave hybrid solver produced similar reconstructions, whereas gradient descent was ineffective. Repeated hybrid-solver runs showed stable this http URL. The framework incorporates photon-statistical weighting and structure-guided regularization into QUBO-based CT reconstruction without changing its quadratic form, providing a proof of concept for quantum-assisted sparse-view CT reconstruction.
[CV-124] PoseAlign: Sculpting Pose-Consistent Meshes via Text-Guided Deformation
链接: https://arxiv.org/abs/2607.10560
作者: Shijin Wang,Zichong Chen,Yang Zhou,Hui Huang
类目: Graphics (cs.GR); Computer Vision and Pattern Recognition (cs.CV)
备注: CGI 2026 Best Paper Award. Project page: this https URL
Abstract:Mesh deformation, the process of altering the vertex positions of a 3D mesh while preserving its topological structure, is a cornerstone of computer graphics. Despite the recent emergence of numerous text-guided 3D mesh deformation methods, deforming an initial mesh into one that both adheres to text prompts and preserves its pose remains challenging. This paper proposes PoseAlign, which decomposes text-guided mesh deformation into two stages: global pose scaling and local detail sculpting. Specifically, in the first stage, we introduce the Laplacian as a differentiable mesh representation to enable more efficient yet smoother global deformation. Then, we propose a novel pose-aligned SDS loss by adapting score distillation sampling (SDS) with an attention-sharing mechanism, which sculptures fine-grained geometric details for the deformed mesh while preserving its original pose. PoseAlign significantly enhances the controllability of the overall deformation process, achieving a favorable balance between pose preservation and text alignment. Experiments demonstrate the competitive advantages of our method in text alignment and mesh quality. Code is available at: this https URL
[CV-125] Physics-inspired Pseudo Anomaly Generation and Prototype Feature Guidance for 3D Anomaly Detection
链接: https://arxiv.org/abs/2607.10544
作者: Jian Ning,Qin Zou,Linchun Wu,Yuanhao Yue,Kunmo Li,Shoubin Chen,Zhongyuan Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 20 pages; already accepted by Pattern Recognition
Abstract:3D point cloud anomaly detection plays a vital role in industrial manufacturing, yet it faces significant challenges due to the scarcity and high acquisition cost of real anomalous samples. The inherently anomaly-free training data further hinders detection methods from effectively learning discriminative features between normal and abnormal instances. To address these issues, we propose PA3AD, a novel framework that introduces a physics-inspired pseudo-anomaly generation strategy to create physically plausible anomalous samples from normal data. Additionally, we incorporate prototype features via a weight-sharing mechanism to guide the model in capturing the distribution shifts between normal and anomalous samples. Specifically, PA3AD introduces two key innovations to tackle the scarcity of real anomalies. First, a physics-inspired module generates diverse pseudo-anomalous point clouds from normal data via multi-physics modeling. Second, momentum-updated prototypes and a difference-aware fusion block capture stable normal representations and their discrepancies with pseudo-anomalies. This design effectively learns distribution shifts, achieving superior detection performance. Extensive experiments on the Anomaly-ShapeNet and Real3D-AD datasets demonstrate that our method consistently outperforms existing state-of-the-art approaches. Our code will be made publicly available at this https URL.
[CV-126] Improving Sample Diversity in Autoregressive Text-to-Image Generation via Cluster Truncation
链接: https://arxiv.org/abs/2607.10535
作者: Trang Nguyen,Shuang Wu,Runyan Tan,Phillip Howard
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:While diffusion models achieve state-of-the-art image quality for text-to-image (T2I) generation, recent work has demonstrated that they suffer from sample diversity collapse. In this work, we investigate whether autoregressive (AR) image generation models can push the Pareto frontier between image quality and sample diversity. With recent advances in quality and efficiency, AR models have emerged as a viable alternative to diffusion-based image generation. Beyond enabling new use cases such as interleaved image-text generation, their sequential generation process makes them compatible with a wide range of token-based decoding strategies originally developed to improve diversity in text generation. Motivated by the potential of a better diversity-quality tradeoff in the AR paradigm, we present the first systematic study of sample diversity in AR image generation models. We show that two key properties of AR image generation, persistently high token-level entropy and substantial redundancy in visual token spaces, limit the effectiveness of existing token-level decoding methods for diversity enhancement. We therefore propose p -less cluster, a new decoding strategy that performs entropy-based truncation sampling at cluster level rather than at token level. We evaluate our approach and baseline decoding methods across four autoregressive T2I models and two datasets using a comprehensive suite of metrics spanning image quality, prompt alignment, and diversity. Our results show that p -less cluster unlocks the greatest diversity across most evaluated autoregressive T2I models and datasets while maintaining image quality and prompt alignment.
[CV-127] owards Autonomous and Auditable Medical Imaging Model Development
链接: https://arxiv.org/abs/2607.10522
作者: Shengyuan Liu,Jia-Xuan Jiang,Boyun Zheng,Cheng Wang,Zipei Wang,Wentao Pan,Hongtao Wu,Houwen Peng,Yu Gu,Lichao Sun,Yixuan Yuan
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 18 Pages
Abstract:Large language model (LLM) agents are beginning to automate machine learning engineering (MLE) by coupling planning, code execution, debugging, and empirical feedback. Translating this capability to medical imaging remains difficult because each task imposes modality-specific experimentation and strict requirements for validation protocols and prediction artifacts. Here we introduce AMID, an autonomous multi-agent framework for medical imaging model development. AMID first proposes Data-Conditioned Method Planning, which refines coarse task-level search spaces into executable, parallelizable method lanes grounded in task-specific data analysis and runnable medical-imaging resources. It then develops Verification-Guided Two-Stage Optimization, moving from broad early exploration of diverse method lanes to selective exploitation of promising candidates while enforcing strict verification of validation protocols, metric computation, and prediction artifacts throughout the optimization. Across 20 medical imaging challenge tasks spanning diverse modalities and prediction types, AMID outperformed evaluated general-purpose MLE systems and, on several tasks, approached or matched strong human-designed challenge solutions. These results suggest that AMID can turn task-specific medical imaging model development from bespoke manual engineering into an agentic workflow for producing high-performing and auditable model artifacts across heterogeneous tasks.
[CV-128] NanoVSR: Towards Real-Time Video Super-Resolution on Edge Devices ECCV2026
链接: https://arxiv.org/abs/2607.10495
作者: Filip Pawlicki,Marcel Kańduła,Marcin Pucek,Kamil Dobies
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ECCV 2026. This is the pre-review submitted version, not the camera-ready version. The final authenticated version will be available in the ECCV 2026 proceedings
Abstract:Recent Video Super-Resolution (VSR) methods rely heavily on transformers and explicit optical flow, creating computational overhead and custom operations that hinder deployment on hardware accelerators like TensorRT. To address this, we introduce NanoVSR, a scalable, fully convolutional architecture designed for resource-constrained edge devices. Using structural reparameterization, NanoVSR collapses into standard convolutions during inference, ensuring seamless hardware compatibility and negligible runtime overhead. Furthermore, despite lacking explicit motion compensation, it maintains competitive restoration quality by implicitly learning spatio-temporal alignments through progressive training. Evaluated on the REDS4 benchmark, NanoVSR demonstrates an exceptional balance between accuracy and computational efficiency, significantly improving the trade-off for compact architectures. Our NanoVSR-644k baseline yields 28.64 dB PSNR while delivering 27.2 FPS on the NVIDIA Jetson Orin NX 16GB (25W), offering massive speed gains over heavier models. The scaled NanoVSR-1.7M variant reaches 29.15 dB with a throughput of 19.58 FPS, providing superior, edge-optimized upscaling. Code is available at this https URL.
[CV-129] Grassmannian Splatting I: Moving rank-2 Spacetime Surfels for Dynamic Scene Rendering
链接: https://arxiv.org/abs/2607.10489
作者: Aaron Maurice Berman,Shantanu Dave
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:We introduce Grassmannian splatting, a dynamic scene representation whose primitives are Gaussians supported on 3-planes in spacetime \R^4 : generically, spatial 2-planes in uniform translation along their normals. Each primitive carries a unit normal n \in \mathbb S^3/\pm 1\ \cong \mathrmGr(3,4) and an unconstrained factor L \in \mathbb R^4 \times 3 , with covariance [ \Sigma_4\mathrmD = (P_n L)(P_n L)^T, \qquad P_n = I - n n^T. ] For generic L and n \neq \pm e_0 , conditioning on time returns a rank-2 surfel at every frame. The normal of the disk and its velocity along that normal are read off from n ; the disk shape and the tangential drift of its center are set by L . Existing native 4D Gaussian splatting methods [\itYang et. al. 2023,Duan et. al. 2024] slice full-rank spacetime covariances, so their per-frame primitive is a volumetric ellipsoid; since conditioning lowers rank by exactly one, a rank-2 surfel in the slice requires a rank-3 spacetime covariance, and the parameterization above realizes exactly these. The motion model is closed form, i.e. no deformation field is learned, and no custom CUDA is required: the conditioned disk feeds a standard 3DGS rasterizer through its precomputed-covariance interface. A soft clamp in the Schur denominator regularizes the static orientation and continuously bridges rank-3 static and rank-2 dynamic behavior, so static and moving primitives form a single continuous family. On the 17 HyperNeRF scenes of MonoDyGauBench, training is fastest among all compared methods (4.9 to 5.6 times faster than the strongest quality baselines), while ranking second in PSNR, MS-SSIM, and LPIPS. Code: this https URL Subjects: Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2607.10489 [cs.CV] (or arXiv:2607.10489v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2607.10489 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-130] On the Real-World Generalisability of Optical Flow Models ECCV2026
链接: https://arxiv.org/abs/2607.10470
作者: Petter Reijalt,Sander Gielisse,Rickard Karlsson,Jan van Gemert
类目: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注: Accepted @ ECCV 2026
Abstract:Real-world deployment of vision models to broadly benefit society is arguably a main research objective. In optical flow, however, the difficulty to obtain the ground truth has focused research mainly on synthetic data and domain-specific benchmarks. Here, we investigate the severity of this mismatch. We study how well modern optical flow estimation models generalise to real-world video and question if accuracy on synthetic benchmark proxies actually predicts accuracy on real-world optical flow. To address this, we build a real-world evaluation benchmark and evaluate the real-world generalisability of a broad set of recent optical flow models using standard checkpoints. Our benchmark contains 8,204 frame pairs across TAP-Flow, Slow Flow, and our own dataset FlowFactor. FlowFactor is a manually annotated real-world benchmark of 1,000 HD frame pairs organised into four confounding factors: large displacements, repetitive textures, occlusions, and lighting variation. Each setting mainly varies only one factor, enabling diagnostic, confounder-specific analysis. Using FlowFactor, we reveal that performance on varying lighting and large displacements correlates most strongly with real-world accuracy, and that improvements on large-motion regimes can trade off against robustness in small-motion, stationary scenes. Our experiments show that progress on Sintel, KITTI and Spring only weakly predicts accuracy on real-world data, highlighting the need for a broad real-world optical flow benchmark. Interestingly, scaling up the amount of training data does not necessarily resolve the gap, calling for new innovative research instead of simply scaling data and compute.
[CV-131] Annotation-Free Furniture Codes: What They Encode and How Far They Transfer
链接: https://arxiv.org/abs/2607.10461
作者: Benjamin Friedman
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Layout-based 3D scene synthesizers place each object using two human-annotated channels: a categorical class label and a canonical-pose convention. We ask whether a single self-supervised token derived from object geometry can replace both, and study such tokens directly as a representation, decoupled from any synthesizer. A Finite Scalar Quantization (FSQ) point-cloud autoencoder is chamfer-trained on placed 3D-FUTURE furniture with no labels or pose annotations. Diagnostic probes recover fine-category (62.6 +/- 0.5%), super-category (85.6 +/- 1.3%), and yaw (52.7 +/- 0.5 deg) from the codes alone. Swapping the chamfer target from the rotated to the un-rotated point cloud collapses the yaw signal while raising class recovery, showing the codes’ rotation content can be set by the training objective. Scaling across asset libraries needs codes that transfer; on an unseen dataset (ShapeNet), alignment is category-dependent: box-like furniture transfers, organically-shaped furniture does not, and a target-blind augmentation partly closes the gap.
[CV-132] BOCCHI: A More Realistic and Challenging Benchmark for Local Motion Blur Detection with MSDCT-UNet
链接: https://arxiv.org/abs/2607.10427
作者: Kuan-Lin Chen,Yuan-Kang Lee,Cheng-Yuan Chiang,Jian-Jiun Ding
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Local motion blur detection requires pixel-level localization of blurred regions. Existing benchmarks let models rely on gradient shortcuts that fail to transfer. We introduce BOCCHI (Blurred Objects Captured across Cameras with Human-annotated Imagery), a real-captured benchmark whose sharp regions overlap the blur gradient distribution and defeat these shortcuts, and propose MSDCT-UNet (Multi-Scale Discrete Cosine Transform UNet), a frequency-aware encoder-decoder injecting multi-scale DCT priors through DCT Attention and FiLM. MSDCT-UNet ranks first in in-domain mIoU and boundary localization on BOCCHI, and BOCCHI-trained models outperform every other training source on cross-dataset transfer with only 633 training images.
[CV-133] SPORT: Structure-Aware Prototype Disentanglement for Incomplete Multi-View Clustering
链接: https://arxiv.org/abs/2607.10413
作者: Yaoyuan Guo,Zhibin Gu,Songhe Feng,Yuhui Zheng,Bing Li
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Prototype-based Incomplete Multi-view Clustering has recently attracted increasing attention by exploiting prototypes as semantic anchors for missing-view imputation. However, existing approaches are still limited in three aspects. First, they typically focus on enforcing cross-view prototype consistency, while ignoring view-specific information embedded in prototypes, thus limiting multi-view expressiveness. Second, most methods rely on instance-level contrastive learning that only aligns paired samples across views, failing to preserve cluster-level relational structures. Third, missing-view imputation is usually performed using global prototypes alone, without considering local geometric neighborhood structures, leading to inaccurate recovery of missing representations. To address these limitations, we propose a novel framework termed Structure-aware PrOtotype disentanglement foR incomplete multi-view clusTering (SPORT), which explicitly disentangles shared and view-specific components of prototypes while preserving cluster-level relational structures. Specifically, we decouple prototypes into orthogonal shared and view-specific components, aligning only shared components to capture consensus semantics while de-correlating view-specific components to preserve complementary information. Meanwhile, a structure-aware contrastive learning mechanism is incorporated to explicitly model cluster-level relationships during cross-view representation learning. Furthermore, a hybrid imputation strategy integrates global prototype matching with local neighborhood matching, enabling joint exploitation of semantic prototypes and manifold structures for missing-view recovery. Extensive experiments on six benchmark datasets show that SPORT achieves superior performance over state-of-the-art methods under various missing rates.
[CV-134] GNOCHI: Generative Neural mOdel for Close Human-Human Interactions
链接: https://arxiv.org/abs/2607.10408
作者: Gonzalo Gómez-Nogales,Marc Comino-Trinidad,Andrés Casado-Elvira,Dan Casas
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Presented at SCA 2026
Abstract:Creating realistic 3D human-human interactions in virtual environments is challenging due to the high degrees of freedom in the human body and the need for physically accurate poses that do not collide with each other. Traditional methods for human-human interaction are based on motion tracking or 3D body reconstruction, but lack generative capabilities. Recent generative methods enable the synthesis of individual or interacting motions via text or image input, but generally fall short in modeling close interactions. This paper introduces a novel generative model for close 3D human-human interactions using a conditional variational autoencoder (cVAE), which generates poses for one human conditioned on the pose of another, allowing for controlled and diverse interaction synthesis. To train our model, we address two underlying long-standing challenges in the field of human-human interaction: data scarcity, for which we propose an automated supervised data augmentation strategy that generates synthetic yet realistic interaction poses; and collision awareness in generative approaches, for which we propose a self-supervised loss based on a collision resolution technique using volumetric proxies to ensure physically correct interactions. We extensively evaluate the capabilities of our model, and demonstrate a wide variety of plausible and physically correct interactions, not possible to generate with current state-of-the-art methods.
[CV-135] VT-PAPD: Pathology-Aware Prototype Distillation for Self-Supervised Whole Slide Image Classification
链接: https://arxiv.org/abs/2607.10406
作者: Ramesh Naidu Laveti,Jaya Sreevalsan-Nair,T K Srikanth
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 13 pages, 4 figures, 10 tables
Abstract:Self-supervised learning (SSL) has emerged as an effective paradigm for learning transferable representations from large-scale unlabeled whole slide images (WSIs). However, existing SSL methods primarily learn generic visual features and often fail to explicitly capture pathology-specific morphological patterns that are critical for disease characterization. To address this limitation, we propose Tiny Vision Transformer with Pathology-Aware Prototype Distillation (TVT-PAPD). This self-supervised pathology representation learning framework integrates a Tiny Vision Transformer (TVT) with a novel Pathology-Aware Prototype Distillation (PAPD) module. PAPD employs a learnable pathology prototype bank to discover and preserve representative tissue morphology patterns, encouraging semantically similar pathological regions to learn consistent and discriminative representations. The proposed framework enhances pathology-aware feature learning while maintaining computational efficiency with 90M parameters. Experiments on the Cancer Genome Atlas (TCGA) low-grade glioma (LGG)/glioblastoma (GBM) dataset and the Indian Pathology Brain (IPD-Brain) dataset demonstrate that TVT-PAPD achieves weighted F1-scores of 93.02% and 90.23%, respectively, for LGG-GBM classification, while exhibiting strong cross-cohort generalization across independent glioma datasets.
[CV-136] SynthDocBench: Controlled Benchmark for Long-Context Visual Document Understanding
链接: https://arxiv.org/abs/2607.10400
作者: Abhigya Verma,Khyati Mahajan,Amit Kumar Saha,Shruthan Radhakrishna,Sagar Davasam,Vikas Yadav,Sai Rajeswar Mudumba
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 29 Pages, 27 Tables, 13 Figures, Accepted at COLM 2026
Abstract:Vision language models (VLMs) have achieved strong performance on visual document understanding benchmarks such as DocVQA, ChartQA, and MMLongBench-Doc. However, real-world documents combine multiple factors such as length, layout complexity, modality, and question difficulty, which makes it difficult to attribute model failures to specific causes. We introduce SynthDocBench, a fully synthetic benchmark for long-context visual document understanding that systematically controls factors including document length, layout structure, modality composition, and question type. The benchmark is constructed using a combinatorial design, each factor is varied independently across generated documents, enabling controlled analysis of model behavior. Documents are generated end to end using an LLM pipeline across six layout archetypes, with a 40 percent random override to prevent models from exploiting spurious correlations. Additionally, SynthDocBench spans long-context documents with substantially greater length and structural diversity than existing benchmarks. Evaluating seven frontier VLMs, we uncover three failure modes that existing benchmarks cannot surface: sharp degradation with document length, a systematic positional sensitivity in which the middle third of a document is hardest for five of six models and five of six models show a negative Early-to-Late trend (steepest decline: 8.3 percentage points), and breakdown of chart comprehension in long-document settings. These results suggest that current models may be overfitting to benchmark artifacts rather than achieving robust long-context visual document understanding.
[CV-137] Self-supervised Automatic Matting
链接: https://arxiv.org/abs/2607.10395
作者: Xiaonan Hu,Zhiyuan Lu,Jingdong Zhao,Hao Lu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:High-quality alpha mattes are notoriously expensive to annotate, creating a fundamental data bottleneck for deep image matting. While prior work attempts to reduce annotation cost using coarser labels like trimaps or masks, they remain reliant on costly per-pixel supervision, limiting scalability and generalization. In this work, we push the boundary further and ask: can we train an automatic matting model using only RGB images, with no manual annotation at all? We answer this by presenting SSMatte, a self-supervised framework that for the first time achieves performance on par with fully-supervised automatic matting. Our key insight is to decompose the problem into semantic anchoring and detail matting. SSMatte first generates a semantic matting prompt from frozen self-supervised ViT features by propagating class-token seeds via a novel, training-efficient semantic anchoring loss based on a generalized Rayleigh quotient. This prompt then anchors a detail matting network, which is optimized via a fixed-point-based loss that enforces alpha-RGB consistency. Extensive experiments show SSMatte outperforms prior weakly-supervised methods, matches the performance of fully-supervised models on portrait benchmarks, and demonstrates favorable scaling and generalization behaviors with additional data. Our work pushes automatic matting to an fresh, fully annotation-free paradigm. Code will be available.
[CV-138] Vertical Fusion: Condensing Internal Representations for Robust ViT Classification
链接: https://arxiv.org/abs/2607.10391
作者: Francesco Di Salvo,Shyam Nandan Rai,Hamed Damirchi,Ignacio Meza De la Jara,Sebastian Doerrich,Marco Lents,Christian Ledig
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Image and Video Processing (eess.IV)
备注: Under review
Abstract:Despite exposing rich intermediate representations, Vision Transformers (ViTs) are almost exclusively utilized as black-box feature extractors, where only the last layer is considered for downstream tasks. We challenge this convention by introducing the notion of recoverability: the capacity of intermediate representations to correct last-layer failures. By evaluating independent classification probes at every model depth across 16 datasets, we observe that intermediate probes correctly classify 18% to 76% of samples that the last-layer probe misclassifies. We show that these gains are not primarily driven by predictive diversity, but by a redundancy-correctness correspondence, where the internal hierarchy acts as a series of stable, redundant probes of a shared discriminative signal. While established horizontal ensemble strategies (i.e., across multiple models) can improve performance, they incur high computational cost and ignore this vertical signal within a single model. To bridge this gap, we propose VFusion, a principled vertical aggregation strategy employing a learnable mapping into a low-dimensional latent space that synthesizes features across the internal ViT hierarchy. VFusion substantially outperforms established aggregation baselines in both in-distribution and out-of-distribution settings, notably closing 45% of the accuracy gap between the best individual layer and a theoretical oracle performance. Our gains consistently generalize across model sizes and pre-training regimes, confirming that VFusion offers a robust and efficient alternative to horizontal ensemble methods. The code is available at this https URL.
[CV-139] ABot-N1: Toward a General Visual Language Navigation Foundation Model
链接: https://arxiv.org/abs/2607.10383
作者: Ruiyan Gong,Yingnan Guo,Junjun Hu,Jintao Kong,Xiaoxu Leng,Tianlun Li,Weize Li,Fei Liu,Zhicheng Liu,Jia Lu,Minghua Luo,Chenlin Ming,Yanfen Shen,Jiyue Tao,Zhengbo Wang,Mingyang Yin,Minqi Gu,Zihao Guan,Wei Guo,Guoqing Liu,Huachong Pang,Menglin Yang,Zeqian Ye,Xiaoxiao Geng,Zhining Gu,Honglin Han,Di Jing,Hongyu Pan,Mingchao Sun,Kuan Yang,Jianfang Zhang,Yanghong Chen,Ye He,Wei Mei,Jiahao Shi,Xiangpo Yang,Yanqing Zhu,Zedong Chu,Xiaolong Wu,Mu Xu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Robotics (cs.RO)
备注:
Abstract:Visual Language Navigation foundation models aim to unify deep reasoning for grounded spatial decisions with broad versatility for diverse embodied tasks. Current approaches typically achieve this integration via monolithic policies that map observations directly to actions, yet they often suffer from coordinate drift and poor handling of long-tail semantics. Furthermore, these black-box mappings lack interpretability, hindering the simultaneous achievement of generality, robustness, and transparency. We present ABot-N1, a step toward a general Visual Language Navigation foundation model, that addresses these challenges by decoupling cognition from control via a slow-fast architecture guided by dual visual-language signals. More specifically, a slow vision-language reasoner performs explicit Chain-of-Thought reasoning while producing a pixel goal. This compact set of image-space anchor points serves as a universal interface for diverse tasks, including point-goal, object-goal, poi-goal, instruction-following, and person-following. Subsequently, a fast action expert leverages both the textual cues and the pixel guidance to generate continuous waypoints at the native control frequency. By bridging high-level intents and low-level control through pixel-grounded anchors paired with explicit linguistic traces, our approach ensures robust, generalizable, and interpretable navigation across simulation and real-world benchmarks. ABot-N1 establishes new state-of-the-art records, delivering massive gains specifically in urban-scale navigation: boosting POI arrival by 35.0% (to 77.3%) and achieving 95.4%/92.9% SR in complex indoor and outdoor scenes. It also maintains superior robustness across object-reaching, person-following, and instruction-following tasks. New Point-Goal/POI-Goal benchmarks are released as open source to advance the field of urban-scale navigation.
[CV-140] Robotic Contextual Awareness for Human-Robot Collaboration and Environmental Understanding
链接: https://arxiv.org/abs/2607.10372
作者: Federico Rollo
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: Ph.D. thesis 2026. Officially published in the IRIS institutional repository of the University of Trento ( this https URL ) and deposited in the Italian National Legal Deposit for Ph.D. theses
Abstract:The transition of autonomous mobile robots from controlled industrial settings to dynamic, human-centric environments, such as manufacturing, logistics, and healthcare, has made their safe and autonomous operation a critical area of research. These sophisticated machines must be capable of perceiving, understanding, and interacting with their surroundings to navigate freely and perform complex tasks. A significant obstacle to achieving this is the lack of comprehensive contextual awareness, which requires a robot to recognize its spatial environment and identify the objects and actors within it. Without this perceptual knowledge, robots struggle to plan adaptive behaviors or engage in meaningful interaction with humans. This thesis presents novel solutions to this challenge by exploring two distinct but complementary research directions. The first direction involves human re-identification and tracking to improve Human-Robot Collaboration. Our developed approach enables a mobile robot to recognize a specific person, facilitating targeted collaboration while ignoring other individuals. The second direction focuses on enhancing the robot’s overall perceptual capabilities to understand its environment geometrically and semantically. Geometric information is vital for motion planning and collision avoidance, while semantic knowledge provides the robot with a richer understanding for more advanced interaction. Both solutions are driven by the improvement of the semantical understanding of robots that enhance their knowledge of their surroundings, allowing a smoother and more natural interaction between robots, humans, and the environment. The contributions of this work in human re-identification and environmental understanding represent a significant step toward a future where robots are more contextually aware, enabling safer coexistence and more effective collaboration. Comments: Ph.D. thesis 2026. Officially published in the IRIS institutional repository of the University of Trento (this https URL) and deposited in the Italian National Legal Deposit for Ph.D. theses Subjects: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2607.10372 [cs.RO] (or arXiv:2607.10372v1 [cs.RO] for this version) https://doi.org/10.48550/arXiv.2607.10372 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-141] Neural Motion Blending Across Arbitrary Character Topologies
链接: https://arxiv.org/abs/2607.10370
作者: Luca Cazzola,Giulia Martinelli,Nicola Conci
类目: Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR)
备注: To appear in the proceedings of Computer Graphics International (CGI 2026). For references, please cite the official proceedings version. Paper website: this https URL
Abstract:Motion blending in character animation enables the synthesis of new motions by interpolating between existing examples. Current methods are typically restricted to fixed skeleton topologies, requiring identical or near-identical skeletal structures across characters. We present a novel framework for motion blending across heterogeneous skeletons. The proposed architecture combines a semantic encoder, which extracts per-frame latent representations of the motion state, with a diffusion-based decoder, which reconstructs character-specific motion conditioned on this latent code. At inference, blended motions are obtained by interpolating the latent representations of two input motions. We train and evaluate the method on the Truebones Zoo dataset using motions defined on both same and distinct skeleton topologies, demonstrating the ability to achieve smooth and plausible blending in a variety of scenarios.
[CV-142] Gradient-Skipping Relevance Propagation for Efficient Explainability of Vision Transformers
链接: https://arxiv.org/abs/2607.10365
作者: Christopher Buratti,Michele Marchetti,Federica Parlapiano,Davide Traini,Domenico Ursino,Luca Virgili
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Vision Transformers (ViTs) are difficult to interpret because current methods of relevance propagation and attention flow do not fully consider some key architectural features, such as the uneven importance of attention heads and residual connections. Prior approaches typically assume uniform importance across attention heads; furthermore, they model skip connections as identity paths, leading to inaccurate relevance attribution. To address these issues, we introduce GradSkip, a novel relevance propagation method for ViTs based on adaptive head weighting and skip-aware propagation. GradSkip models the different importance of the attention heads and dynamically distributes relevance between the attention and residual paths. Experiments on ImageNet1K and BloodMNIST demonstrate a state-of-the-art faithfulness of GradSkip while requiring over 14 times fewer GFLOPs than the best-performing existing approaches. Additional evaluations using transformer-based segmentation confirm improved localization and alignment with ground-truth regions.
[CV-143] Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
链接: https://arxiv.org/abs/2607.10358
作者: Giang Nguyen,Raghav Mehta,Emma A.M. Stanley,Tian Xia,Thi Hao Nguyen,Hieu Pham,Ben Glocker
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Under Review. Giang Nguyen and Raghav Mehta contributed equally
Abstract:Foundation models are increasingly used as image feature extractors for mammography, but their robustness under external domain shift remains unclear. We benchmark 15 foundation-model backbones across breast density, BI-RADS severity, and cancer status using a unified frozen-backbone linear-probe protocol, training on 3 source datasets and evaluating on 12 task-compatible out-of-distribution (OOD) datasets after label harmonization. Mammography-specific vision-language models (Mammo-FM and MaMA) provide the strongest mean OOD performance, but robustness is not explained by mammography exposure alone. DINOv3 remains a competitive vision-only baseline, and mammography-adapted pretraining does not consistently improve generalization. Dataset-level analysis further shows that even leading models show heterogeneous performance across datasets. Feature-space inspection reveals that useful representations can preserve clinical signal while retaining dataset and acquisition structure. These findings highlight dataset-level OOD evaluation as a central criterion for assessing mammography representations. Our code is publicly available: this https URL.
[CV-144] GRC-ProbNet: Uncertainty-aware Feature Extraction for Cardiovascular Disease Classification
链接: https://arxiv.org/abs/2607.10357
作者: Yash Shah,Omar Todd,Philipp Seeböck,Georg Langs,Ben Glocker,Raghav Mehta
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Under Review
Abstract:The automatic detection and classification of cardiovascular disease (CVD) from computed tomography (CT) images plays an important role in clinical practice. Recently, a hybrid pipeline (GRC-Net) for CVD classification was proposed, which leverages a deep-learning-based segmentation and registration method to extract radiomic and geometric features. However, GRC-Net relies on a deterministic segmentation mask, without considering the inherent ambiguity associated with cardiac anatomy. In this paper, we propose GRC-ProbNet, which takes advantage of a deep ensemble to produce multiple segmentation masks for a given input. From these masks, we extract multiple uncertainty features. We analyze these uncertainty features for both their correlation with segmentation error and their propagation effects on downstream CVD classification performance. Our experiments on the publicly available MM-WHS and ASOCA datasets show that the uncertainty measure that best reflects segmentation quality is not necessarily the one that provides the strongest signal for downstream CVD classification. Overall, our results demonstrate that GRC-ProbNet utilizing uncertainty features substantially improves CVD classification AUROC (92.92) compared to the baseline GRC-Net model (91.25%). Our code is publicly available: this https URL.
[CV-145] PrismAD: Decoupled Planning via Semantic Mixture-of-Planners for End-to-End Autonomous Driving
链接: https://arxiv.org/abs/2607.10336
作者: Kang Ding,Zhigui Lin,Hongsong Wang,Jie Gui,Qi Liu,Zhe Wang,Luqi Tang,Lei He
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 8 pages,5 figures
Abstract:This letter presents PrismAD, a decoupled end-to-end autonomous driving framework based on a Semantic Mixture-of-Planners. Existing planners usually aggregate heterogeneous scene tokens into a coupled representation space, forcing a single planning branch to jointly model agent interaction, road geometry, and driving intention. Such coupling may weaken factor-specific reasoning and obscure the contribution of different planning cues. To address this limitation, PrismAD partitions scene tokens into interaction, geometry, and intent groups, and assigns them to independent planning experts with the same architecture but separate parameters. Each expert learns a specialized motion-planning representation, while a semantics-aware router adaptively aggregates expert predictions with separate routing weights for motion prediction and ego planning. Sparse top- K activation with noisy gating is further introduced to improve routing robustness and reduce unnecessary expert computation. Extensive experiments on the nuScenes open-loop dataset and NeuroNCAP closed-loop benchmark demonstrate that PrismAD exhibits competitive performance. Our code will be released soon.
[CV-146] Imperceptible and Reversible Adversarial Examples against Vision-Language Models for Privacy Protection ACM-MM2026
链接: https://arxiv.org/abs/2607.10329
作者: Qi Lu,Ziqi Zhou,Yufei Song,Zijing Li,Lulu Xue,Minghui Li,Shengshan Hu,Leo Yu Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by ACM MM 2026
Abstract:Vision Language Models (VLMs) offer powerful multimodal ability but also expose users to text-based privacy attacks where adversaries crawl online photos and query VLMs to extract sensitive attributes. Existing reversible adversarial example (RAE) methods protect images in purely visual tasks but fail in multimodal settings, and current adversarial examples on VLMs rely on high frequency noise that severely degrades visual quality. We propose CloakDiff, the first framework for reversible, high fidelity privacy protection against text-based query attacks in VLMs. CloakDiff produces imperceptible adversarial examples by combining diffusion based adversarial editing with an invertible network that embeds the original image for lossless recovery. It perturbs both pixel space embeddings and manipulates latent cross attention maps to ensure strong cross-model and cross-prompt transferability while preserving global visual structure. To further enhance fidelity, we design EDM Heuristic Sampling, a principled diffusion schedule for adversarial guidance. Experiments on multiple datasets and VLMs demonstrate that CloakDiff delivers multimodal privacy preservation with high visual quality and reversibility.
[CV-147] Generalize LMMs to Versatile Visual Modalities via Fabricated Modality Synthesis ECCV
链接: https://arxiv.org/abs/2607.10308
作者: Shihao Yuan,Yuanze Li,Ruyi Zhang,Ming Liu,Wangmeng Zuo
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by the European Conference on Computer Vision (ECCV) 2026
Abstract:Despite the advancements of Large Multimodal Models (LMMs) in RGB vision, their ability to generalize to unseen visual modalities remains a largely unexplored challenge. We argue that different visual modalities are merely distinct samplings of the same physical world. Therefore, effective generalization requires models to possess both modality-agnostic perception of scene semantics and the adaptability to modality-specific characteristics. To achieve this, we propose a training framework, VVM-Tuning, to equip LMMs with these capabilities through modality synthesis and modality contexts. Specifically, we synthesize diverse appearance-varied images from RGB scenes, training the model to disentangle invariant semantics from varying visual appearances, and align these appearances with language for visual concepts decoupled from modalities. We then introduce modality contexts in the prompt and use instruction tuning to assist the model in mapping these appearance variations back to modality-related attributes, enabling zero-shot adaptation to unseen modalities during inference. To facilitate research in this direction, we introduce VVM-Bench, a comprehensive benchmark featuring 6 real and synthetic modalities to evaluate semantic perception and modality understanding. Experiments demonstrate that, via our training on synthetic modalities, 5 tested models exhibit consistent improvements on both real-world and novel synthetic modalities without in-modality training. Source code and data will be publicly available at this https URL.
[CV-148] Structured Evidence Selection for Weakly Supervised Video Anomaly Detection
链接: https://arxiv.org/abs/2607.10298
作者: Chenglizhao Chen,Tianxiang Nan,Wen Li,Xinyu Liu,Guisheng Zhang,Mengke Song,Xiaomin Yu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Weakly supervised video anomaly detection relies solely on video-level labels for training, making it difficult to accurately localize anomalous events in complex scenes. In real-world videos, anomalous behaviors exhibit large variations in appearance and temporal duration, while scene appearance and action dynamics are often tightly entangled. Consequently, existing models tend to rely on scene-related statistical cues rather than true behavioral deviations, resulting in unstable detection performance. To address this challenge, we propose a Structured Evidence Selection framework (SESAD) that reformulates anomaly detection as a structured reasoning process over clip-level visual evidence. Instead of directly mapping aggregated features to anomaly scores, SESAD reorganizes clip representations into semantically structured candidate evidence and performs context-conditioned selection under scene and action constraints. This mechanism adaptively emphasizes anomaly-relevant semantics while suppressing scene interference, thereby alleviating semantic entanglement under weak supervision. Furthermore, we introduce a lightweight geometric discrimination module that constructs a dual-prototype structure in the embedding space, enabling anomaly decisions through relative geometric relations. Extensive experiments on UBnormal, ShanghaiTech, and UCF-Crime show that SESAD achieves 67.92, 97.99, and 88.46 AUC, respectively, while maintaining high computational efficiency and overall consistently stable anomaly discrimination.
[CV-149] InterPet4D: A Multimodal 4D Human-Pet Interaction Dataset for Pet Motion Generation
链接: https://arxiv.org/abs/2607.10287
作者: Yichen Peng,Jyun-Ting Song,Chen-Chieh Liao,Kris Kitani,Hideki Koike,Erwin Wu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Human-pet interaction estimation and generation remain underexplored due to the absence of a high-quality large-scale dataset. We present InterPet4D, the first multimodal dataset capturing natural interactions between humans and dogs. Using a synchronized multi-view capture system, we record human-dog obedience tasks and provide annotations for both humans and dogs, including multi-view and egocentric videos, segmentations, 2D and 3D keypoints, meshes, and audio tracks. InterPet4D consists of 6.8 million frames collected from 13 dogs of 11 breeds interacting with 23 human participants. We further introduce the InterPetMoGen framework for human-pet interaction motion generation. Our proposed model achieves an FID score of 11.21 and substantially outperforms the Seq2Seq and DiT baselines, demonstrating the effectiveness of InterPet4D for modeling realistic human-pet interactions.
[CV-150] Geometry-aware Gaussian Prior and Axial Attention for Cervical Cytology Image Classification
链接: https://arxiv.org/abs/2607.10278
作者: Yating Li,Cheng Ye,Nenan Lyu,Weidong Chen,Zhendong Mao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Accurate cervical cytology image classification is a key component of automated cervical cancer screening, where reliable recognition of normal, precancerous, and cancer-associated cellular patterns from Pap smear images can improve screening efficiency and diagnostic consistency. However, this task remains challenging because cervical cells exhibit complex morphology, subtle intra-class variations, and strong inter-class similarities. Existing convolution-based models capture local texture well but have limited ability to model long-range relationships, whereas attention-based models provide broader context but often lack explicit structural guidance. To address these limitations, we propose a geometry-aware classification framework for cervical cancer screening-oriented cytology image analysis, incorporating semantic abstraction and structural priors learned from pre-trained vision-language features. The method uses Gaussian expert modules to generate axis-wise priors from global semantic information, capturing structural regularities such as nuclear alignment and cellular spatial organization. These priors are embedded into an axial self-attention module to modulate similarity computation along horizontal and vertical directions, improving long-range dependency modeling and structure-sensitive feature interaction. Experiments on the Mendeley liquid-based cytology and SIPaKMeD datasets show that the proposed method achieves 99.48% accuracy on the former and 96.08% on the latter, with balanced gains in recall, precision, and overall classification performance. Visual analysis further shows that the learned priors highlight diagnostically relevant cellular regions, demonstrating the potential of the proposed framework as a screening-oriented decision-support tool for cervical cytology.
[CV-151] What Does Your Short-Answer VQA Score Actually Measure? Evaluator-Dependent Instability in Multimodal Short-Answer Benchmarks
链接: https://arxiv.org/abs/2607.10240
作者: Guanhua Ye,Niu Jingbin,Yan Li,Meiyu Liang,Zhe Xue,Yingxia Shao,Yawen Li
类目: Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM)
备注:
Abstract:Short-answer VQA benchmarks conflate two distinct quantities: whether a model’s answer is semantically correct, and whether that answer matches the surface form expected by the automatic evaluator. We study this conflation across six vision–language models and six benchmarks, using a human-validated semantic judge (97.6% precision) to audit over 37k official errors. A second text-only judge reproduces the same benchmark-level false-negative pattern, showing that the effect is not an artifact of a single audit model. On text-rich benchmarks, up to half of these errors are semantically acceptable answers penalized purely for surface-form mismatch. This instability is structured by answer type: extractive and multi-span answers are far more evaluator-sensitive than scalar answers. Benign prompt and context rewrites further destabilize official outcomes, flipping item-level correctness at substantial rates without changing the underlying task. A deterministic CPU-only contract repair confirms that the undercount is partially recoverable. These findings imply that official short-answer VQA scores should be accompanied by semantic audits and answer-type diagnostics to remain interpretable.
[CV-152] Benchmarking Dynamic Affective Reasoning : A Viewer-Centric Video Emotion Dataset ECCV2026
链接: https://arxiv.org/abs/2607.10238
作者: Zhiyan Zhang,Peipei Song,Jinpeng Hu,Jingyang Jia,Xun Yang,Xiaojun Chang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 25 pages, 6 figures. Accepted to ECCV 2026
Abstract:Video emotion analysis is typically framed as a static classification problem, treating each clip as an independent labeled unit. However, such a formulation overlooks a key psychological fact: emotions change as a result of cumulative reactions to consecutive causal events. To bridge this gap, we introduce Dynamic Affective Reasoning, the first large-scale benchmark for viewer-centric affect transitions and causal reasoning over consecutive video events. DAR contains 15,087 videos and 36,908 event-aligned affective segments annotated with 27 emotion categories. Unlike existing video-based emotion datasets, DAR presents a new viewer-centric perspective on fine-grained emotional expressions and transitions, and provides dense, temporally grounded, and causally explicit reasoning chains. Based on DAR, we formally define three challenging tasks: affective segmentation, fine-grained emotion classification, and affective reasoning. Complementing this benchmark, we propose DAR-R1, a two-stage framework that combines supervised fine-tuning with Group Relative Policy Optimization. Experiments across 10+ MLLMs show that DAR-R1 sets a new state-of-the-art for dynamic affective reasoning, in terms of both emotional localization and affective reasoning. Project page: this https URL.
[CV-153] CoSAG: Compact Semantic Anchor Gaussians via Training-Free Rate-Distortion Coding
链接: https://arxiv.org/abs/2607.10237
作者: Yuang Jia,Jinlong Wang,Junhong Lin,Ruiting Dai,Wei Gao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Open-vocabulary 3D scene understanding is commonly achieved by embedding 2D vision-language features such as CLIP into a 3D Gaussian Splatting scene, turning it into a text-queryable semantic field. However, attaching a high-dimensional feature to each of millions of Gaussians inflates a single scene to gigabytes, which makes storage and deployment the real bottleneck of these fields. Existing compact methods each learn and ship a per-scene codec, an autoencoder, a quantized codebook, or a distilled feature field, entangling field construction with field storage and never compressing the per-Gaussian assignment that holds the bulk of the cost. We argue that construction and storage should be decoupled, and that storage is a rate-distortion problem over the per-Gaussian binding to a small anchor table, a structure no prior open-vocabulary method compresses. We present CoSAG, which constructs the field without any per-scene training through a closed-form transmittance-weighted lift, spatially grounded semantic anchors, and multi-view denoising, and stores it with a spatially predictive entropy coder that ships no decoder. Because the anchors are spatially grounded, the binding is predictable and therefore highly compressible. The transmittance-weighted lift and multi-view denoising yield a clean, view-consistent assignment, so the entropy coder spends almost no rate on correcting noise and instead codes only the residual against its spatial prediction. CoSAG reaches sub-megabyte storage while matching or exceeding the state of the art across the 2D-rendered, 3D-selection, and dense-LSeg protocols, reducing field size by 37 to 76x relative to LangSplatV2 at higher accuracy.
[CV-154] PhenoEmbed: Self-Supervised Multispectral UAV Time-Series Embeddings for Individual Tree Crown Phenology
链接: https://arxiv.org/abs/2607.10231
作者: Taimur Khan
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Tree crowns are a challenging target for resilient AI because they are not static objects: their spectral response, internal texture, translucency, and apparent boundaries change substantially across the growing season. We develop PhenoEmbed, a self-supervised crown-centric temporal embedding model trained with contrastive and masked reconstruction objectives on HeideBench, an 18-date UAV multispectral time-series benchmark for forest crown phenology in Dölauer Heide. The model treats seasonal crown dynamics as phenological appearance change driven by leaf emergence, canopy closure, senescence, and leaf-off conditions. Segmented tree crown polygons are retained as object anchors to extract aligned crown-centered crops through time, allowing one 256-dimensional vector summarizing seasonal crown appearance to be learned per tree. On 5,885 crop-safe crowns, the exported embeddings show structured low-dimensional organization, with the first two principal components explaining 25.1% of variance and nearest-neighbor retrieval producing a median top-1 cosine similarity of 0.946. Compared with handcrafted temporal features and a learned mean-pooling baseline, PhenoEmbed yields substantially more compact nearest-neighbor structure, while ablations show that the contrastive loss, masked reconstruction loss, and explicit seasonal time features each affect the structure of the learned embedding space. These results support PhenoEmbed as a reusable forest crown representation learner and motivate future downstream tests of whether such features improve tree-level models under seasonal change.
[CV-155] ScratNet: A Swin-Based Multi-Scale Dilated Network with Precision Refinement for Semiconductor Scratch Segmentation
链接: https://arxiv.org/abs/2607.10214
作者: Sachin Ranjan,Hoon Kim
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Surface scratch defects in semiconductor manufacturing pose significant challenges due to their irregular shapes, low contrast, and varying scales. Traditional inspection methods often struggle to detect such defects reliably, especially in complex imaging scenarios. While deep learning approaches based on Convolutional Neural Networks (CNNs) have improved accuracy, they often fail to capture fine-grained edge details. To address these limitations, we propose ScratNet, a novel end-to-end scratch segmentation framework that integrates a modified Swin Transformer backbone with a tailored decoder. The decoder incorporates a Multi-Scale Dilated Aggregation (MDA) module to capture both local and global context, a Stem Integration Module (SIM) to restore spatial detail, and a Precision Refinement (PR) branch that enhances boundary sharpness using anisotropic convolutions. Through this stage-adaptive feature aggregation and boundary-aware refinement, ScratNet achieves superior accuracy on thin and irregular defects. Extensive experiments demonstrate that ScratNet consistently outperforms existing methods, providing a scalable and robust solution for automated scratch inspection in high-precision manufacturing.
[CV-156] PhysMRV: Physical Memory Retrieval and Verification for Physics Plausibility Reasoning
链接: https://arxiv.org/abs/2607.10190
作者: Wenyuan Wang,Lianyu Hu,Hao Wang,Yang Liu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Video-language models (VLMs) have achieved remarkable performance on video understanding and visual question answering, yet they remain unreliable in reasoning about physical plausibility, where understanding object interactions, causal dynamics, and fundamental physical principles is essential. This limitation is particularly evident on challenging physical reasoning benchmarks, revealing a persistent gap in physical commonsense reasoning. To address this challenge, we propose PhysMRV, a training-free physical memory and verification framework for physical plausibility reasoning. Unlike retrieval-augmented VLMs that retrieve semantically similar videos as additional context, PhysMRV transforms training videos into a Hierarchical Memory Bank of structured physical knowledge comprising three complementary levels: scene descriptions capturing visual context, physical-event graphs modeling object interactions and causal structure, and physics-rule summaries distilling reusable physical principles and cues. During inference, PhysMRV retrieves physically relevant memories and leverages their structured physical evidence to guide a frozen VLM in verifying physical plausibility, requiring neither fine-tuning nor parameter updates. We evaluate PhysMRV on three challenging physical reasoning benchmarks, ImplausiBench, IntPhys2, and GRASP Level 2, across multiple state-of-the-art VLMs. Experimental results demonstrate consistent improvements over direct prompting across diverse VLMs and evaluation benchmarks, showing that structured physical memories provide an effective and scalable means of enhancing physical plausibility reasoning without additional training.
[CV-157] BiLoG-Net: A Bi-Context Location-Guided Network for Breast Mass Segmentation and Malignancy Classification in Mammography
链接: https://arxiv.org/abs/2607.10188
作者: Abu Fatema Mohammad Abdun Noor,Md Imam Ahasan,Md Samiul Ahasan,Kah Ong Michael Goh,S M Hasan Mahmud,Raihana Zannat
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 23 pages, 6 figures, 5 tables
Abstract:Breast cancer remains the most commonly diagnosed malignancy among women worldwide, yet accurate detection and characterization of breast masses in mammography remain challenging due to subtle intensity variations, heterogeneous tissue densities, and indistinct lesion boundaries that complicate radiological interpretation. To address these limitations, we propose BiLoG-Net, a deep learning framework that jointly performs breast mass segmentation and malignancy classification through bi-context location-aware feature modeling and segmentation-guided attention mechanisms. Our architecture integrates a novel encoder-decoder paradigm with Fire-based feature extraction, lightweight global and local feature enhancement modules, and adaptive location-aware gating to simultaneously capture long-range contextual dependencies and fine-grained boundary-sensitive details. Unlike conventional multi-stage pipelines, our tightly coupled multi-task design enables mutual reinforcement between pixel-level localization and image-level diagnosis, reducing error propagation while producing spatially grounded malignancy predictions. Evaluated on CBIS-DDSM and INBreast benchmarks, BiLoG-Net achieves state-of-the-art performance with Dice scores of 94.20% and 93.10%, classification accuracies of 95.20% and 93.60%, and AUC values of 97.10% and 96.00%, respectively, substantially outperforming existing CNN and transformer-based baselines. By combining precise boundary delineation with reliable malignancy assessment in a single end-to-end model, this work holds strong potential for clinical computer-aided detection systems, helping radiologists prioritize suspicious cases and improve screening efficiency in busy clinical settings.
[CV-158] EmoStyle: Affective Conditioning of Style-Specialist Experts for Emotional Image Generation
链接: https://arxiv.org/abs/2607.10165
作者: Dexiang Hong,Yijie Guo,Weidong Chen,Xinyan Liu,Zixuan Zou,Zhendong Mao,Yongdong Zhang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Emotion-aware artistic image generation requires an image to match the input prompt, follow the specified artistic style, and convey the target emotion. In this challenge, the main difficulty is that the visual and affective attributes available in the training data are not explicitly provided at test time. Without these attributes, the generator has to decide not only what to depict, but also how the target emotion should be expressed through color, lighting, brushwork, composition, line, and layout. This creates a control gap between the available test prompt and the fine-grained conditions needed for emotion-aware artistic generation. To bridge this gap, we propose EmoStyle, a Z-Image-based framework that converts the input prompt into a structured generation state. An LLM reasoner first predicts affective cues (valence-arousal, dominant emotion, and therapeutic-effect labels) and an aspect-ratio decision. Instead of using these predictions only as additional prompt text, we encode the affective fields into an affective condition vector and inject it into the denoising blocks through AdaLN-style modulation. This allows the inferred control variables to directly guide the generation of intermediate features. Since emotional expression is also style-dependent, we further train a dedicated LoRA adapter for each artistic style bucket and select the corresponding expert during inference, enabling the same affective cues to be rendered with bucket-specific priors for color, texture, brushwork, and composition. Finally, a lightweight VLM-guided candidate selection step ranks the generated images based on prompt alignment, style consistency, emotional expression, and visual quality. In Track 1 of the AffectiveArt Challenge 2026, our USTC_PI_LAB_TEAM submission achieved first place.
[CV-159] REVA-PO: Stabilizing Reinforcement Learning for Chest X-ray Report Generation
链接: https://arxiv.org/abs/2607.10147
作者: Li Guo,Anas M. Tahir,Z. Jane Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Automated chest X-ray report generation has recently benefited from reinforcement learning (RL) and large language models. However, RL training often suffers from instability or limited exploration due to fixed Kullback-Leibler (KL) regularization and a static reference policy that accumulates KL pressure over time. We propose Response-Weighted and Validation-Anchored Policy Optimization (REVA-PO), a RL framework that stabilizes long-term training via Response-Weighted Regularization (RER) and Validation-Anchored Policy Reset (VAPR). RER dynamically adjusts per-response KL weights based on advantage and reference-policy entropy, relaxing constraints for high-quality responses while tightening them for low-quality ones. Complementarily, VAPR periodically synchronizes the reference and current policies to the best validation checkpoint, resetting accumulated regularization pressure to expand the viable exploration space. To ensure a robust starting point, we employ a three-stage pipeline consisting of warm-up training, classifier-guided supervised fine-tuning, and RL. Extensive evaluations on MIMIC-CXR and IU-Xray demonstrate that REVA-PO sets new state-of-the-art benchmarks in both linguistic quality and clinical accuracy. Notably, BLEU-4 improves by 5.1% on MIMIC-CXR and 3.6% on IU-Xray, while CheXpert F1 and RadGraph F1 scores increase by 4.5% and 12.8%, respectively, over prior leading methods. The code is publicly available at this https URL.
[CV-160] FlowPainter: Inpainting Optical Flow via Confidence-Guided Completion
链接: https://arxiv.org/abs/2607.10140
作者: Yuang Meng,Chenyang Wu,Xianshun Liu,Chun-Le Guo,Zichen Liang,Lina Lei,Jie Liang,Hui Zeng,Chongyi Li,Lei Zhang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Existing optical flow methods broadly follow two paradigms: iterative optimization and diffusion-based estimation. Iterative methods, exemplified by RAFT, achieve high accuracy through recurrent refinement, but remain challenged by large displacements and complex motion. Diffusion-based methods introduce generative modeling and show promise in such ambiguous regions. However, existing diffusion models usually denoise the entire dense flow field from Gaussian noise, including simple regions where reliable motion can already be estimated by a lightweight network. This increases the denoising burden and may cause slow convergence and unstable training. To address this issue, we introduce FlowPainter, a diffusion-based optical flow framework that reformulates dense-flow generation as confidence-guided soft inpainting. FlowPainter employs a lightweight confidence-aware network to predict a rough flow and a pixel-wise confidence mask, distinguishing reliable simple regions from uncertain hard regions. The resulting simple-flow prior is used for confidence-based initialization and further injected into iterative denoising through confidence-gated residual guidance. With dynamically decaying guidance strength, FlowPainter stabilizes early denoising while preserving the flexibility of the diffusion model for late-stage detail refinement. Extensive experiments on public benchmarks, including Sintel, KITTI, and Spring, show that FlowPainter achieves strong accuracy under comparable training settings and converges more efficiently than existing diffusion-based optical flow methods, with notable gains on challenging benchmark splits. Our approach offers a practical way to integrate reliable discriminative priors with diffusion-based refinement for optical flow estimation. Our code is publicly available at this https URL.
[CV-161] xtGaze: Prompting Gaze Target Estimation with Textual Scene Cues IJCAI2026
链接: https://arxiv.org/abs/2607.10130
作者: Junhui She,Fei Wang,Kun Li,Yiqi Nie,Yuxin Liu,Zhangling Duan,Xun Yang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by IJCAI 2026. All contents are available at: this https URL
Abstract:Gaze target estimation aims to infer the position of a person’s gaze within a scene. Within mainstream design logic, multi-branch methods require extra supervision and annotations, while streamlined designs prioritize low-level visual saliency over true gaze intent. The former leads to a high annotation burden and hinders domain transfer, whereas the latter causes misalignment between predicted attention and actual gaze targets. To address this issue, we propose TextGaze, a unified cross-modal architecture that leverages a Large Vision-Language Model (LVLM) as scalable semantic guidance to balance the two design paradigms. The model extracts visual features from a frozen encoder and utilizes an LVLM to obtain gaze-aligned textual cues. We design a transformer-based fusion module with hierarchical text supervision to preserve task semantics. Lightweight decoding heads enable the joint prediction of gaze heatmaps and in-/out-of-frame status. We evaluate our method on four mainstream datasets, and the results show competitive performance across key metrics with robust cross-dataset generalisation without extra fine-tuning. Overall, we provide a streamlined alternative to traditional designs and highlight the potential of LVLMs as accessible auxiliary guidance for gaze estimation.
[CV-162] WeaveEarth: Structured Evidence Construction and Reasoning for Training-Free UHR Remote Sensing Understanding
链接: https://arxiv.org/abs/2607.10120
作者: Xianzhi Ma,Shujun Wang,Xiaohan Li,Hao Liu,Changhua Pei,Jianhui li
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Ultra-High-Resolution (UHR) remote sensing image understanding requires Vision-Language Models (VLMs) to capture both the global scene layout and sparse yet task-critical local details under limited computational budgets. Existing methods mainly follow two paradigms. One is passive perception, which relies on resolution expansion or token compression and may therefore discard fine-grained details. The other is active perception, which depends on multi-round zooming and search, but suffers from high latency, contextual fragmentation, and error accumulation. We argue that a more effective path toward UHR understanding lies not in accessing more, but in organizing better. To this end, we propose WeaveEarth, a training-free framework that reformulates UHR understanding as a problem of structured evidence construction and reasoning under global context constraints. Specifically, WeaveEarth first employs Global-Aware Evidence Construction to select a compact, low-redundancy, and spatially complementary Minimal Support Evidence Set. It then introduces Structured Evidence Reasoning, which weaves local evidence, spatial metadata, and relative topology into a unified reasoning interface, thereby enhancing the VLM’s ability to perform global-local joint reasoning. Extensive experiments show that WeaveEarth consistently outperforms strong baselines and existing UHR methods across multiple UHR remote sensing benchmarks and multiple frozen VLM backbones. Code is available at this https URL.
[CV-163] DynaFilter: Cloud-driven Dynamic Filtering for Satellite Edge Intelligence
链接: https://arxiv.org/abs/2607.10098
作者: Ziyang Zhang,Jie Liu,Luca Mottola
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 15 pages, 23 figures, accepted by ACM MobiCom 2026
Abstract:Modern satellite edge systems, including those performing remote sensing tasks such object detection and tracking, are characterized by severely limited bandwidth and intermittent connections, making continuous data transmission to the cloud impractical. Existing edge-cloud systems, however, either require heavy pre-processing before analysis, for instance, full decompression of imagery data, or transmit all compressed data regardless of relevance. To address these challenges, we design DynaFilter, a dynamic filtering technique that enables satellite edge devices to perform selective region-of-interest (RoI) inference directly in the compressed-domain, without full decompression. Our key insight is that low-level compression syntax, specifically DC coefficients/AC energy in JPEG images and motion vectors in video streams, exhibits strong correlations with high-level semantic queries. By establishing a precise mapping between cloud query semantics and multimodal compressed-domain features, DynaFilter enables the edge to identify and transmit only relevant data associated to RoIs. Extensive evaluations show that DynaFilter reduces the total volume of pixel data for decoding and subsequent inference by 1.6x-7.1x for images, and achieves 92.0% bandwidth savings for video streams compared to state-of-the-art baselines. Furthermore, it decreases energy consumption by 43.1-88.6% on target devices and achieves a 1.6x-3.0x speedup in inference latency.
[CV-164] LFD: Enabling Real-World Lensless Face Recognition with a Large-Scale Dataset
链接: https://arxiv.org/abs/2607.10094
作者: Junho Kim,Salman S. Khan,Sara Wan,Tomi Kuye,Ashok Veeraraghavan
类目: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注: 10 pages
Abstract:Face recognition is a ubiquitously used computer vision task that has a wide range of applications ranging from everyday smartphone biometrics to high-stakes security systems. Most face recognition systems rely on traditional cameras, which often suffer from limitations such as bulky form factors, high costs, and limited privacy protection. To address these limitations, lensless cameras have emerged as an alternative. Lensless cameras use thin optical encoders, enabling smaller size, lower cost, and greater design flexibility. These cameras are typically paired with reconstruction algorithms that convert raw captures into recognizable images. However, reconstructed images often contain artifacts, and the reconstruction methods struggle to generalize well to real-world conditions. Furthermore, existing face datasets do not account for the artifacts present in lensless images. To address this issue, we introduce the Lensless Face Dataset (LFD). LFD comprises 21,080 lensless raw measurements, reconstructions, and standard images of faces captured under diverse lighting, angle, and distance. Our key contributions are: (1) Real-world lensless face data: LFD focuses on capturing a diverse face dataset with varying levels of artifacts introduced under different environments; (2) In-the-wild captures: 4,976 images are captured in outdoor settings with varying intensities of natural light and different background patterns; (3) Multiple lensless devices: LFD includes face images collected from three different types of lensless cameras, each with a unique optical encoder. We use this hardware diversity to demonstrate generalization across different lensless cameras. Through comprehensive evaluations and analysis, we show that LFD effectively captures shared features and artifacts across different lensless imaging devices, making it a valuable dataset for advancing lensless face recognition.
[CV-165] EMBRACE: A Multi-task Framework for Comprehensive Quality Assessment in Cleavage-stage Embryo
链接: https://arxiv.org/abs/2607.10093
作者: Anwar Hussain Sofi,Jung-Hua Wang,Ming-Jer Chen,Tsung-Hsien Lee,Yu-Chiao Yi,Ming-Kuan Lin,Yi-Chung Lai
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Cleavage-stage embryo assessment in in vitro fertilization requires the integrated interpretation of cytoplasmic fragmentation, developmental stage, and blastomere symmetry. However, conventional visual assessment is affected by observer variability, particularly when fragmented regions are small, irregular, or low contrast. This study presents EMBRACE, a multi-task deep learning framework for jointly performing cytoplasmic-fragmentation segmentation, t2/t4 developmental-stage classification, and blastomere-symmetry grading from static cleavage-stage embryo microscopy images. EMBRACE combines a shared ResNet-50 backbone, a concatenation-based multi-scale feature-fusion (C-MSFF) module, a U-Net-style segmentation decoder, and two task-specific classification heads. After predefined inclusion and exclusion criteria, 9,137 annotated embryo images were divided into 7,309 training, 914 validation, and 914 held-out test images. On the held-out test set, EMBRACE achieved a Dice coefficient of 0.781 and an intersection over union of 0.677 for fragmentation segmentation. Developmental-stage classification achieved an accuracy of 0.995, macro-F1 of 0.994, and AUC of 1.000. Blastomere-symmetry grading achieved a balanced accuracy of 0.901, macro-F1 of 0.907, and quadratic weighted kappa of 0.859. These findings support the feasibility of combining spatially inspectable fragmentation localization with embryo-level morphology assessment in a single framework. External and prospective validation is required before clinical deployment.
[CV-166] CVKD-UDA: Cross-View Knowledge Distillation for 3D Unsupervised Domain Adaptive Segmentation
链接: https://arxiv.org/abs/2607.10087
作者: Zhimin Yuan,Ming Cheng,Shangshu Yu,Wen Li,Dunqiang Liu,Xin Huang,Cheng Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:3D unsupervised domain adaptive (UDA) segmentation mitigates the high cost of manual annotations of the new domain data. Self-training has emerged as the dominant approach in this area, where its success heavily depends on a well-initialized warm-up model to generate reliable pseudo labels. However, existing methods often depend on source supervision or output-level adversarial alignment to obtain the warm-up model, which suffer from limited generalization and training instability due to the large domain gap between domains. Constructing domain-similar representations is an effective way to bridge this gap. In this work, we propose CVKD-UDA, which revisits voxel size as a core design factor to construct domain-similar representations and leverages cross-view complementary cues to balance transferability and discriminability of the warm-up model. First, we generate two complementary views by varying voxel sizes and introduce a cross-view knowledge distillation (CVKD) to enhance generalization and target perception of the model. Second, to balance transferability and discriminability, we design a lightweight Decouple-Adapter and an auxiliary imitation classifier to decouple cross-view knowledge transfer. Extensive experiments on two benchmarks demonstrate that CVKD-UDA effectively improves the performance of self-training methods and provides a new perspective for 3D UDA segmentation. Our code will be available at GitHub.
[CV-167] Label-Free Target-Domain Adaptation for Unconstrained Event-Image Feature Matching via Dual-Stage Distillation ACM-MM2026
链接: https://arxiv.org/abs/2607.10082
作者: Zhonghua Yi,Hao Shi,Qi Jiang,Yufan Zhang,Kailun Yang,Kaiwei Wang
类目: Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM); Robotics (cs.RO); Image and Video Processing (eess.IV)
备注: Accepted to ACM MM 2026. The source code and benchmark will be made publicly available at this https URL
Abstract:Building pixel-level correspondence between event and image data is a fundamental task for multi-sensor systems. However, existing cross-modal matching methods are largely restricted by their reliance on either matching labels or strictly aligned hardware, which limits them to unlabeled and unconstrained real-world scenarios where neither matching ground truth nor prior sensor relationships are available. To address this, we propose a novel two-stage training paradigm. First, we leverage large-scale data to perform label-agnostic distillation pretraining, upgrading optimization objectives with distribution-based and contrastive losses to learn highly generalizable representations. Second, to tackle unlabeled and unconstrained downstream data, we introduce an epipolar-guided self-distillation framework. By utilizing consistency verification to isolate robust matches and incorporating geometric confidence derived from an external epipolar prior, our model can effectively self-evolve directly on target domains without any supervision. Furthermore, we introduce a rigorous cross-modal evaluation benchmark based on TUM-VIE, featuring physically separated cameras with distinct intrinsic parameters and resolutions. Extensive experiments demonstrate that our proposed method achieves state-of-the-art performance on both MVSEC and TUM-VIE pose estimation tasks. The source code and benchmark will be made publicly available at this https URL.
[CV-168] FlashBEV: Fast and Memory-Efficient Exact BEV Transformation with IO-Awareness ECCV2026
链接: https://arxiv.org/abs/2607.10071
作者: Shunsuke Yokokawa,Hironori Kasahara
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注: Accepted to ECCV 2026
Abstract:Bird’s-eye-view (BEV) perception is a core component of camera-based 3D understanding in autonomous driving, where view transformation (VT) maps multi-camera image features into a unified BEV representation. Sampling-based view transformation (Sampling-VT) is attractive because it supports dense and continuous BEV aggregation for high-resolution and long-range perception. Its deployment bottleneck, however, is systems-level: standard tensorized implementations of Sampling-VT – which we refer to as Tensorized Sampling-VT – explicitly materialize large height-dependent intermediate tensors, causing memory and latency costs that scale poorly with vertical resolution and the number of cameras. We revisit Tensorized Sampling-VT from an operator-execution perspective and show that it follows a gather-reduction pattern: each BEV query independently accumulates contributions across cameras and height bins, enabling thread-local accumulation with on-the-fly recomputation that eliminates the need to materialize height- and camera-dependent intermediates. Based on this insight, we propose FlashBEV, a fully fused and IO-aware execution strategy mathematically equivalent to Tensorized Sampling-VT (same operator output) while substantially reducing global memory traffic and kernel-launch overhead. Experiments show that FlashBEV achieves more than an order of magnitude lower peak GPU memory and significant inference-latency speedups, with memory effectively independent of the number of height bins, reducing the operator’s peak memory to O(BCXY) (output only). This unlocks higher BEV range/resolution and vertical discretization within fixed deployment budgets on memory-constrained devices. Our contribution is an execution redesign – same math, different execution – that removes a key scalability barrier for deployment-ready Sampling-VT. Code available at this https URL
[CV-169] Error Aware Distribution Prediction for Lightweight Implicit Neural Representations
链接: https://arxiv.org/abs/2607.10068
作者: Zhimin Li,Jake D. Balla,Joshua A. Levine
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR)
备注:
Abstract:Implicit neural representations (INRs) offer compact encoding of volumes, but as lossy approximators, inevitably have prediction errors. We consider INRs that can simultaneously encode relative error scales by predicting distributions using tools from uncertainty estimation. Typically, uncertainty estimation relies on computationally expensive approaches or on predefined parametric assumptions about the predictive distribution (e.g., Gaussian). In this study, we propose a lightweight method that reformulates regression-based INR training as a classification task by discretizing continuous targets into bins, enabling flexible distribution modeling to capture complex multimodal behaviors. We analyze the trade-off between regression and classification for INR training and demonstrate that the classification setting tends to achieve high reconstruction quality and competitive error awareness through uncertainty estimation, compared to regression-based approaches.
[CV-170] Model Guides You How to Draw: Adaptive Visual Gating for Unified Multimodal Reasoning
链接: https://arxiv.org/abs/2607.10004
作者: Wenxi Gao,Guanxi Lu,Didi Zhu,Hao Mark Chen,Quan Deng,Zhican Wang,Jiankang Deng,Hongxiang Fan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Unified multimodal models (UMMs) with interleaved reasoning, which generate both textual and visual steps as part of intermediate reasoning traces, have demonstrated great potential for visual mathematical reasoning tasks. However, we identify a key insight in this paradigm: generating intermediate visual reasoning steps is not always beneficial and can even be harmful, as self-generated visual steps may introduce erroneous visual evidence that misleads subsequent reasoning. Moreover, frequently triggering visual steps during reasoning incurs substantial computational and memory overhead, degrading inference efficiency. To address these accuracy and efficiency challenges, we observe that the model’s internal signals can indicate whether a visual step will benefit reasoning before the entire visual generation is completed. Specifically, this work identifies two internal signals: 1) Generation Intent, which reflects whether the model has a concrete textual plan for what to draw, and 2) Visual Fidelity, which measures whether the visual generation remains grounded in the original input image. Leveraging these internal signals, we propose AdaViG, a training-free adaptive visual gating method for unified multimodal reasoning. AdaViG dynamically evaluates each triggered visual step at an early visual generation stage and aborts it when both signals are weak, thereby preventing misleading visual evidence from entering the reasoning trace while avoiding unnecessary computation. Comprehensive experiments demonstrate that AdaViG improves accuracy by up to 5.7% while reducing visual generation FLOPs by 25.0%-91.0% and wall-clock latency by 15.4%-45.6%.
[CV-171] UniPose9D: Universal Category-Agnostic Object Pose Estimation
链接: https://arxiv.org/abs/2607.09985
作者: Yang You,Yi Du,Cole Harrison,Leonidas Guibas
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Object pose estimation is a fundamental problem in 3D vision. Although recent state-of-the-art approaches achieve strong performance, they often overfit to existing benchmarks and exhibit limited generalization to novel categories and unseen scenes. We propose UniPose9D, a category-agnostic foundation model for 9D object pose estimation: given an instance mask/ROI and either an RGB-D observation or an RGB image with predicted depth, the model estimates rotation, translation, and metric size without category labels, CAD models, mean-shape priors, or reference views. Specifically, UniPose9D samples point pairs from the observed object geometry and uses DINOv2 and PointNet features to predict NOCS coordinates for each pair. To improve accuracy, we introduce a point-pair-based RANSAC N-hop Kabsch–Umeyama algorithm with an adaptive threshold. We further employ flow matching to address symmetric ambiguities and construct a large-scale training set by curating and aligning pose annotations from existing public datasets. Experiments across six datasets show that a single unified model can match or surpass specialist methods while generalizing to unseen objects and in-the-wild scenarios. Our code and model are available on this https URL.
[CV-172] Banshee: Target Switch Attacks on Gimbal-Stabilized Visual Tracking Systems via Acoustic Injection
链接: https://arxiv.org/abs/2607.09930
作者: Jiarui Li,Joseph Brewington,Qingzhao Zhang,Z. Morley Mao
类目: Computer Vision and Pattern Recognition (cs.CV); Cryptography and Security (cs.CR)
备注: Accepted by the 47th IEEE Symposium on Security and Privacy (IEEE SP 2026). 20 pages. Artifact: this https URL
Abstract:Gimbal-stabilized visual tracking is critical for modern autonomous systems such as Unmanned Aerial Vehicles (UAVs). While prior work shows acoustic signals can disturb gimbal internals, the impact of such attacks on real-world applications like UAV tracking and following remains underexplored. Existing demonstrations largely overlook practical challenges for real-world attacks, such as object-motion uncertainty and runtime latency. To bridge this gap, we present Banshee, the first physically realizable attack that induces target switching in UAV visual tracking systems by exploiting acoustic vulnerabilities in gimbal-camera systems. Banshee generates carefully crafted acoustic waveforms that induce optimized adversarial gimbal oscillations, causing directionally biased camera-view drifts that break inter-frame target associations. Consequently, the onboard tracker is driven to switch from the original target to an attacker-selected object with high probability, with occasional target loss. Banshee achieves a 93.6% success rate in simulation across two commercial gimbal systems and five trackers. Real-world benchtop and in-flight black-box attacks against a commercial drone across varied scenarios show an overall 95.5% attack success rate. Our results reveal a practical cross-domain vulnerability between acoustics and vision, highlighting the need for robust designs of gimbal systems and applications. Our code is available at: this https URL.
[CV-173] Do Transformer Temporal Heads and Post-Pooling Motion Gates Help CorrNet-based CSLR? An Empirical Study
链接: https://arxiv.org/abs/2607.09890
作者: Lisi Wang,Zhidong Xiao,Jianjun Peng
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 12 pages, 3 figures, 2 tables
Abstract:CorrNet is a strong baseline for continuous sign language recognition (CSLR) because it models inter-frame correlations inside the visual encoding stage. In this paper, we study two natural extensions of a reproduced CorrNet system: replacing the BiLSTM temporal head with a Transformer encoder, and injecting motion cues after temporal pooling. We find that the Transformer head does not outperform the BiLSTM baseline, even with a training strategy adjusted for the Transformer, and the two heads have almost the same computational and runtime cost. For the second extension, we design a lightweight module called MotionGate. In our experiments, MotionGate consistently collapses to an identity-like mapping: the gate loses motion selectivity, and the injected residual becomes a weak, non-selective perturbation of the pooled features. These results suggest that explicit motion injection after CorrNet’s correlation-based encoding is largely redundant, and that natural-looking architectural extensions in CSLR should be tested carefully instead of being assumed to help.
[CV-174] Bridging the Catalog-to-Real Gap: Scalable Product Recognition via Multi-Stage Contrastive Learning
链接: https://arxiv.org/abs/2607.09888
作者: Anyi Zhang,Joy Mazumder,Kiril Lomakin
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 12 pages, 4 figures
Abstract:Automated product recognition is a cornerstone of modern retail intelligence; however, accurately matching real-world, in-store images against extensive corporate catalogs remains a major scalability bottleneck for large-scale applications. In this work, we address this challenge by reformulating the task as an embedding-based cross-domain retrieval problem rather than a standard closed-set classification task. Specifically, we define the objective as retrieving the most corresponding catalog reference image for a given real-world product query crop from an expansive inventory. To bridge the severe domain gap between pristine studio packshots and noisy in-store queries, we introduce a novel catalog-to-real multi-stage contrastive learning paradigm (Cat2Real). This framework fine-tunes a vision backbone by systematically exploiting both item-level and image-level similarities to drive targeted hard negative mining. Extensive empirical evaluations demonstrate that our paradigm scales seamlessly to unseen products and categories, yielding outstanding zero-shot generalization performance even in the complete absence of real-world training images for novel inventory.
[CV-175] ShapKO: Shapley-Adaptive Modality Knockout for Robust Multimodal Learning
链接: https://arxiv.org/abs/2607.09884
作者: Nusrat Binta Nizam,Fengbei Liu,Sunwoo Kwak,Minh Nguyen,Ruining Deng,Mert R. Sabuncu
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Multimodal medical models often degrade when inputs are missing, a common scenario in real-world clinical workflows. Separately, even when all modalities are present, modality dominance is observed during training, where optimization over-relies on a highly predictive modality and undertrains complementary sources, resulting in poor robustness under partial availability. While training-time modality knockout improves missing-modality robustness, existing approaches use static masking rates that cannot adapt to evolving modality utility during training. We introduce ShapKO (Shapley-Adaptive Modality Knockout), a dynamic training strategy that learns modality-specific knockout probabilities based on validation utility. ShapKO periodically evaluates performance across modality subsets, estimates modality importance via Shapley values, and updates masking probabilities to suppress dominant modalities more frequently. This adaptive process promotes complementary representations, while requiring no architectural modifications. We evaluate ShapKO on three datasets covering multitask clinical classification, survival prediction, and cancer detection. ShapKO consistently improves performance under modality absence and yields interpretable trajectories of learned masking behavior. Code is available at: this https URL
[CV-176] A Dual-Stream Challenge-Response Protocol for Ocular Liveness Verification
链接: https://arxiv.org/abs/2607.09883
作者: Ismail Kably
类目: Computer Vision and Pattern Recognition (cs.CV); Cryptography and Security (cs.CR)
备注: 6 pages, 2 figures. Simulation-based theoretical framework; human-subject validation identified as future work
Abstract:Ocular biometric systems face sophisticated presentation attacks, including high-resolution video replays and real-time generative deepfakes, which easily bypass static liveness checks. Current Presentation Attack Detection (PAD) frameworks typically rely on isolated physiological metrics, such as gaze tracking or the Pupillary Light Reflex (PLR), which can be spoofed independently. This paper proposes a Spatio-Luminance Sensor Fusion protocol, which introduces a dual-stream challenge-response framework for ocular liveness verification by uniting these metrics into a simultaneous authentication challenge. By generating a randomized, time-varying visual stimulus that fluctuates in both spatial trajectory and luminance intensity, we construct a mathematically coupled state-space likelihood model, termed the Synchronization Matrix, to evaluate the continuous cross-correlation between the expected biological latencies of smooth pursuit tracking and pupillary constriction. Using Monte Carlo simulation grounded in literature-derived latency distributions, we demonstrate theoretical separability between genuine and simulated attack conditions, and show that a multi-round challenge design improves the detection of generative deepfakes when a non-zero rendering-latency gap exists. This work provides a simulation-supported theoretical framework for next-generation dynamic spoofing defense in ocular and iris biometrics; human-subject validation is identified as necessary future work before deployment claims can be made.
[CV-177] Prompting-MammAlps: Fine-Grained Text-to-Video Retrieval for Camera-Trap Data ECCV2026
链接: https://arxiv.org/abs/2607.09876
作者: Valentin Gabeff,Baptiste Maquignaz,Jennifer Shan,Sepideh Mamooler,Gencer Sumbul,Blair Costelloe,Devis Tuia,Alexander Mathis
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Neurons and Cognition (q-bio.NC); Quantitative Methods (q-bio.QM)
备注: Accepted at ECCV 2026; Project page: this https URL
Abstract:Automatically retrieving videos from large camera-trap datasets remains challenging. Text-to-Video retrieval (TVR) methods based on large video-language models (VLMs) have potential to retrieve events of interest by describing them with simple text queries. However, current methods often lack spatiotemporal understanding and do not generalize well to ecological data. In this work, we introduce Prompting-MammAlps, the first camera-trap TVR benchmark, and propose a fine-grained and interpretable TVR method. Specifically, we trained a vision transformer to perform spatiotemporal action localization, and convert its output to structured text, describing each video. Independently, ethology-inspired queries are processed by a Large-Language Model (LLM) based coding agent to parse the structured text per video and retrieve videos accordingly. We harnessed the LLM to use functions from a custom parsing library to minimize the risk of LLM hallucinations and to improve method interpretability. This retrieval approach applied on the Prompting-MammAlps benchmark achieved a set-based F1-score of 34% on a test set of 135 ecologically-relevant queries and 775 candidate videos. In comparison the best zero-shot VLM achieved a F1-score of 18%, while also lacking interpretability. Project page: this https URL
[CV-178] Reliability-Aware Ensemble Classification Under Class Imbalance: A Calibration Study on Liquid-Based Cervical Cytology
链接: https://arxiv.org/abs/2607.09837
作者: Nisreen Albzour,Sarah S. Lam
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Cervical cytology classification models are typically evaluated on curated, class-balanced benchmarks, but real-world liquid-based cytology (LBC) collections are often small and class-imbalanced. This paper presents a class-imbalance-aware and calibration-aware ensemble classification study on the Mendeley LBC dataset, using its native four-class Bethesda taxonomy (NILM, LSIL, HSIL, SCC) rather than a collapsed binary formulation. Three lightweight architectures (Swin-Tiny, TinyViT-5M, DenseNet121) are trained directly on Mendeley LBC using weighted random sampling to counteract class imbalance, and compared against two soft-voting ensembles (Hybrid-2, Hybrid-3). Post-hoc temperature scaling is fit on a held-out calibration subset carved out of the training portion of each cross-validation fold, distinct from both the training data used to fit model weights and the evaluation fold used for final metrics, avoiding the optimistic calibration estimates that result when the same data is used for both purposes. Calibration substantially reduces expected calibration error, Brier score, and negative log-likelihood for every model and ensemble configuration tested, while discrimination metrics (accuracy, macro-F1, macro-AUROC) remain essentially unchanged. Ensemble size shows no consistent additional reliability benefit over the best individual model once all configurations are properly calibrated. Confusion matrices show that all classification errors, across every configuration, are confined to the boundary between high-grade lesions (HSIL) and carcinoma (SCC); no errors involve the negative (NILM) or low-grade (LSIL) categories. These results suggest that, for this dataset, calibration is the dominant lever for reliability, not ensemble size, though this conclusion should be read in light of the dataset’s modest size.
[CV-179] Does YOLO26 Truly Offer Advantages Over Its Predecessors for Edge Deployment? A Benchmark Study in Aquaculture
链接: https://arxiv.org/abs/2607.09835
作者: Rakesh Ranjan,Gajanan S. Kothawade,Kata Sharrer,Scott Tsukuda,Christopher Good
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:The recently introduced YOLO26 architecture incorporates NMS-free end-to-end inference and is optimized for deployment on resource-constrained CPU-based devices, making it well-suited for edge-based aquaculture applications. However, its performance, operational efficiency, and deployment suitability have not been systematically validated in aquaculture-specific scenarios. This study presents a comprehensive benchmark of YOLO26 against three Ultralytics predecessors (YOLOv5u, YOLOv8, and YOLO11) across nano, small, and medium model scales for fish mortality detection, a critical indicator of fish population health and welfare. Twelve model variants were evaluated for detection accuracy, training efficiency across seven dataset sizes, and inference performance on high-performance NVIDIA A100 GPUs and a CPU-only Raspberry Pi 5 edge platform. All models achieved comparable performance on the full dataset, with mAP50 differing by only 1.04 percentage points, indicating that architectural generation has little influence on final detection accuracy when sufficient training data are available. However, clear trade-offs emerged in data efficiency and deployment performance. YOLOv8 achieved 90% mAP50 with only 400 training images, whereas the YOLO26 nano and small variants required 1,000 images to reach comparable accuracy. Conversely, YOLO26n achieved the highest inference speed on the Raspberry Pi 5 (7.51 FPS), while YOLOv5mu outperformed all contemporary medium-scale architectures on CPU-based hardware. These results show that architectural novelty alone is insufficient for model selection and that training data availability, target hardware, and inference requirements should be considered jointly when selecting object detection models for practical edge AI deployment in aquaculture.
[CV-180] A Strong Balanced-Softmax Classifier-Retraining Baseline for Long-Tailed Recognition
链接: https://arxiv.org/abs/2607.09832
作者: Juan Terven,Diana Margarita Córdova Esparza,Julio Alejandro Romero Gonzalez,Edgar Arturo Chávez Urbiola,Francisco Javier Willars Rodriguez,Juan Bautista Hurtado Ramos,Alfonso Ramirez Pedraza
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注: 18 pages, 6 figures, 7 tables
Abstract:Long-tailed recognition methods often modify losses, margins, or representations to reduce the dominance of frequent classes. We ask whether, after Balanced Softmax training, the remaining tail error can be reduced by retraining only the classifier. We evaluate BS-cRT, a two-stage procedure that trains a backbone and cosine classifier with Balanced Softmax, freezes the backbone, and updates only the classifier on balanced episodic batches. The second stage keeps the empirical-prior Balanced Softmax objective and uses raw cosine logits at inference. Across CIFAR-100-LT, CIFAR-10-LT, ImageNet-LT, and Places-LT, this classifier-only step consistently improves Few-shot accuracy over the matched Balanced Softmax checkpoint. At imbalance factor 100, Few-shot gains are +5.15 points on CIFAR-100-LT and +5.83 on CIFAR-10-LT; on ImageNet-LT and Places-LT, gains are +6.92 and +9.78 points, respectively, with a Top-1/Few-shot trade-off on ImageNet-LT. We also analyze Counterfactual Boundary Risk Minimization (CBRM), a boundary-probe extension using prototype-based features near decision boundaries. CBRM identifies two failure modes: scaled-logit cosine margins destabilize training, and corrected hardest-negative probes remain head-class anchored. The results support BS-cRT as a practical classifier-side baseline and indicate that boundary supervision must account for class frequency.
[CV-181] DSal: Task-Based Top-Down Saliency Prediction Model
链接: https://arxiv.org/abs/2607.09827
作者: Can Mizrakli,Tolga K. Capin
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 15 pages, 5 figures
Abstract:Visual saliency aims to predict the regions of an image most likely to attract human visual attention. While most saliency models assume free-viewing conditions, human attention is often shaped by explicit task goals. In this work, we address task-driven saliency prediction by proposing a model that conditions visual attention on natural-language task descriptions. The model produces task-dependent saliency maps that reflect how attention shifts under different viewing intents. Through quantitative and qualitative analysis, we show that incorporating explicit task semantics enables more faithful modeling of goal-directed visual attention.
[CV-182] owards Objective Dysgraphia Detection: A Multi-Branch Deep Learning Approach for Online Handwriting Analysis
链接: https://arxiv.org/abs/2607.09826
作者: Lydia Ouhib(LIASD),Yassine Ouzar(LIASD),Zoé Pinseel,Stéphane Bouilland,Mehdi Ammi(LIASD)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Signal Processing (eess.SP)
备注:
Abstract:Dysgraphia is a specific learning disability that is prevalent among school-age children. It affects handwriting coherence, quality, fluency, and legibility, often hindering academic achievement and early learning development. This motor coordination disorder is typically diagnosed through subjective assessments based on clinician observation, which can be timeconsuming and prone to variability. In this paper, we introduce a deep learning-based framework for objective dysgraphia detection using online handwriting data captured via digitizing tablets. The proposed framework relies on two complementary branches: the first pipeline extracts both handcrafted and embedding-based kinematic features directly from raw temporal signals, while the second leverages image-based representations of the temporal signals generated using continuous wavelet transforms (CWT) and Gramian Angular Fields (GAF). The resulting features are then fused to leverage the complementary strengths of both representations. The four representations were evaluated separately and jointly using the publicly available DiaGraMo dataset, showing that the fusion of GAF, MOMENT, and hand-crafted kinematic features outperforms each individual representation, as well as other fusion schemes. These findings highlight the potential of the complementarity of image and signal based representations for more objective dysgraphia detection.
[CV-183] S-Mask VLA: 2D Temporal-Spatial Masking for Vision-Language-Action Model with Effective Bridging IROS2026
链接: https://arxiv.org/abs/2607.09818
作者: Shengzhuo Yang,Ronghao Yu,Chuanjie Lv,Linpeng Peng,Hang Yu,Jie Ren,Jiajun Lv,Yong Liu
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages, 5 figures, accepted to the 2026 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2026)
Abstract:Vision-language-action (VLA) models aim to understand natural-language instructions and visual observations, and to generate and execute corresponding actions as embodied agents. Recently, autoregressive token-based action generation has driven the development of many representative VLA models. However, this paradigm often reduces action generation to next-token prediction, thereby lacking explicit modeling of the spatiotemporal structure of action sequences and the disentanglement between vision-language representations and actions, which can limit performance in long-horizon and complex scenarios. In this paper, we propose TS-Mask VLA, a vision-language-action framework for robot manipulation. TS-Mask VLA is built upon two key designs: (1) a Discrete Diffusion Action Expert equipped with a Bridge Attention conditioning bridge, which enables multi-layer conditioning from the VLM and facilitates more accurate and stable action generation; and (2) a temporal-spatial 2D masking strategy for discrete action tokens that strengthens the model’s understanding of cross-time dependencies and inter-dimensional coupling, leading to more structurally consistent action sequences. We conduct extensive experiments on simulation benchmarks and real-world tasks. On LIBERO, TS-Mask VLA achieves a 95.7 percent average success rate with only 0.5B parameters, outperforming significantly larger models. On CALVIN, it attains the best average sequence length of 4.19 and strong long-horizon performance. Comprehensive analyses and ablations further validate the effectiveness of our design.
[CV-184] RASR: Range-Aware Scale Recovery for Metric UAV Navigation
链接: https://arxiv.org/abs/2607.09815
作者: Hongtao Liang,Xinyu Shao,Chenxu Wang,Yiyao Wan,Jiahuan Ji,Fangwei Ye,Fuhui Zhou,Qihui Wu
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 5 pages, 4 figures. Technical report for the UAVM 2026 PairUAV Challenge
Abstract:Under Global Navigation Satellite System (GNSS) denial, a UAV controller still needs a distance and heading command it can execute, making accurate metric last-meter navigation essential. Dense pair-geometry foundation models transfer relative structure well, yet the distance scale of their raw metric outputs remains poorly calibrated. Under the relative error metric of PairUAV, correcting only the average scale can still leave costly, distance-dependent residuals near the goal. To address this scale mismatch, Range-Aware Scale Recovery (RASR) separates a transferable scale-recovery core from a protocol-specific calibration module in a per-pair system fixed at inference. The core compresses frozen Matching And Stereo 3D Reconstruction (MASt3R)-style geometry into a compact descriptor and uses global calibration to recover the dominant metric signal. Range-bucket residual correction and command-grid alignment stay inside the calibration module, so they match the command format and evaluation protocol of PairUAV. On the UAVs in Multimedia 2026 PairUAV online evaluation, RASR reaches a total score of 0.003189. Under the PairUAV protocol, frozen pair geometry thus yields stable per-pair distance and heading estimates, while every protocol-specific adjustment stays confined to a calibration module fixed before inference. Code and materials are available at this https URL.
[CV-185] Detangled: A Framework for Creating Editing and Inferencing Feature Rich Hair Strands
链接: https://arxiv.org/abs/2607.09811
作者: Sarah Jobalia,Yitong Deng,Carolyn Smith,Ronald Fedkiw
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 18 pages, 18 figures
Abstract:We present a framework for understanding and generating feature rich hair strands. Drawing upon both scientific and cultural expertise, we define strand texture as the various distinctive patterns (curling, switchbacks, twist, etc.) that are formed by forces internal to a hair strand. We begin by proposing a novel five-dimensional parameter space, intended to be a bijection with naturally occurring hair strand textures. This encoding is both qualitatively accessible, allowing users to readily locate their own hair in the parameter space, and quantitatively precise, allowing the generation of individual strands from texture inputs. Importantly, strand texture should be independent from the overall strand direction. In order to disentangle strand texture from the overall strand direction, we identify centerline geometry and use it to map strands into a canonical space (a strand texture space). We construct centerlines using a novel method that cleanly distills complex hair grooms, separating the strand texture from the overall style (parameterized by style guides). We enable the creation of new strands conforming to our parametric description of texture via a generative artificial intelligence approach supervised by a separate neural network trained to label candidate strands according to our five-parameter description. The ability to create new strands conforming to any desired texture enables groom editing using either texture transfer or user-provided inputs. We demonstrate results on a variety of hair types.
[CV-186] MVMGNN;Multi-View Masked Graph Neural Network for Alzheimers Disease Diagnosis using Structural MRI
链接: https://arxiv.org/abs/2607.09788
作者: Ni Yao,Zhenxu Wang,Danyang Sun,Chuang Han,Yanting Li,Jiaofen Nan,Fubao Zhu,Chen Zhao,Weihua Zhou
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Alzheimer’s disease (AD) is a common neurodegenerative disorder, and early diagnosis is of great significance for delaying disease progression and enabling timely intervention. Mild cognitive impairment (MCI), which represents an intermediate clinical stage between cognitively normal aging and AD. Structural magnetic resonance imaging (sMRI) provides detailed characterization of anatomical structures and plays an important role in AD-related brain analysis. However, existing sMRI-based brain network methods typically rely on a single graph construction strategy, limiting their ability to jointly capture spatial relationships and morphological similarities between brain regions. To address these issues, this paper proposes an sMRI-based multi-view masked graph neural network model (MVMGNN) for AD diagnosis. A joint node-edge masking mechanism is proposed to simultaneously select radiomics feature dimensions and structural connections, reducing redundancy during graph learning. Furthermore, a patient-level cross-view gated fusion mechanism is proposed to integrate multi-view representations. Experimental results on the ADNI dataset demonstrate that MVMGNN outperforms several competing approaches in AD classification. Interpretability analysis further demonstrates that MVMGNN is able to identify key brain regions associated with AD, providing useful insights into discriminative patterns in sMRI-based brain this http URL implementation is publicly available at this https URL
[CV-187] Adversarially Guided Diffusion for LiDAR Range Image Synthesis KDD2026 ECML
链接: https://arxiv.org/abs/2607.09787
作者: Stavros Bouras,Antonios Makris,Alexandros Gkillas,Aris S. Lalos,Konstantinos Tserpes
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Accepted at the 1st Workshop on Secure and Trustworthy AI (STAI 2026), co-located with the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD 2026)
Abstract:LiDAR semantic segmentation is a key perception task in autonomous driving, where false predictions can affect downstream planning and safety-critical decision-making. Although adversarial attacks, and specifically adversarial examples, have been widely studied for image classification and 3D point cloud segmentation, unrestricted adversarial examples remain largely unexplored in the space of 2D range images, which are projections of 3D point clouds. The proposed method is, to the best of our knowledge, the first diffusion-based unrestricted adversarial attack against 2D range-image segmentation, using adversarial guidance from a segmentation loss. By applying guidance directly during sampling, the method produces unrestricted adversarial examples that remain close to the learned LiDAR data manifold while inducing structured segmentation errors. Experiments on the SemanticKITTI dataset using RangeNet++ and CENet segmentation networks demonstrate that the attack provides adjustable degradation across guidance strengths and transfers across segmentation architectures. Compared with norm-bounded FGSM and SegPGD baselines, the proposed attack offers a distinct effectiveness-realism trade-off, achieving controllable white-box and transfer degradation while maintaining competitive distributional and visual realism.
[CV-188] Lifelong Representations: A Survey on Continual Self-Supervised Learning for Vision Models
链接: https://arxiv.org/abs/2607.09785
作者: Sergi Masip,Alicja Dobrzeniecka,Jonathan Swinnen,Joachim Collin,Bartłomiej Twardowski,Szymon Łukasik,Tinne Tuytelaars
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: This work has been accepted for publication in IEEE EAIS 2026
Abstract:Traditionally, continual learning has assumed access to labeled data, yet many real-world applications – such as lifelong robotics – require models to adapt continuously from unlabeled streams. This has led to the development of continual self-supervised learning (CSSL), a rapidly growing area that lacks a dedicated, systematic review. In this work, we present a comprehensive survey of CSSL for vision, with connections to emerging vision-language settings. First, we analyze existing evaluation protocols and highlight inconsistencies that hinder fair comparison. We then examine why self-supervised objectives exhibit improved robustness to catastrophic forgetting, relating this to task-agnostic representations and smoother loss landscapes. Next, we organize existing methods into a unified taxonomy based on their forgetting-mitigation strategies, including distillation, replay, regularization, architectural approaches, model merging, and objective-level adaptation. Finally, we identify open challenges such as scalability and the need for fast adaptability. We argue that advancing CSSL requires moving beyond small-scale benchmarks towards continual pre-training paradigms for large-scale systems.
[CV-189] Compression Asymmetry and Trajectory Binding in Noise-Anchored Diffusion Inversion
链接: https://arxiv.org/abs/2607.09784
作者: Yongseong Park,Joeun Kim,HoEun Kim,Young-Sik Kim
类目: Computer Vision and Pattern Recognition (cs.CV); Information Theory (cs.IT)
备注:
Abstract:Real-image diffusion inversion is governed by a tight quality-cost trade-off, with costs incurred in computation, storage, or per-image optimization. We study this trade-off through the forward Gaussian noise anchor that defines a diffusion trajectory and isolate two mechanisms behind effective stored-noise inversion. First, diffusion noise exhibits an element-wise compression asymmetry: int8 full-dimensional anchors preserve reconstruction, whereas low-dimensional subspace summaries are much less reliable, often collapsing even at comparable or smaller payloads; the element-wise over subspace ordering persists across five stored-noise inversion methods. Second, inversion is trajectory-bound and score-prior coupled: the matched forward anchor and a trained score network are both necessary, arguing against a purely algebraic-identity explanation. Together, these findings specify what to store and how to use it. They lead to Noise-Anchored Reverse Correction (NARC), a training-free inversion primitive that stores a single int8 latent anchor and reuses it with a fixed, noise-level-dependent anchor-weight schedule: strong anchoring when the reverse trajectory is noise-dominated, then relaxed anchoring as image detail emerges. On PIE-Bench++ with Stable Diffusion 1.5, NARC outperforms five modern non-exact baselines and improves PSNR by +3.24 dB over PnP DirectInv while using about 400x less inversion storage than PnP DirectInv. The compression asymmetry, anchor specificity, and editing plug-in also transfer to SDXL 1024^2.
[CV-190] owards Real-World Wearable Motion Reconstruction ECCV2026
链接: https://arxiv.org/abs/2607.09780
作者: Andrea Boscolo Camiletto,Rishabh Dabral,Eduardo Alvarado,Thabo Beeler,Marc Habermann,Christian Theobalt
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Accepted at ECCV 2026
Abstract:The modern-day surge in popularity of wearable devices poses a fundamentally unique motion capture problem: reconstructing full-body movement from any set of sensing hardware worn at a given moment. Yet, most research efforts assume fixed sensor configurations (e.g. IMU suits or HMD-centric rigs) and cannot generalize across them. In contrast, we argue that motion capture should prioritize unobtrusive and lightweight devices such as smartphones, smartwatches, smart glasses, and smart insoles, and study the interplay between them. To this end, we make three contributions. First, we present a large-scale multi-modal dataset synchronizing these consumer-grade sensors with ground-truth 3D motion, spanning 50 diverse activities including everyday tasks, sports, and social interactions. Second, we propose WHIP, a baseline generative model that reconstructs motion from arbitrary subsets of available sensors, robustly handling missing modalities and producing physically plausible motions. Third, we conduct a systematic study of sensor complementarity, quantifying how different modalities complement one another. Code and dataset are available at this https URL
[CV-191] A Generalized Deep Non-negative Matrix Factorization Approach for SAR Automatic Target Recognition
链接: https://arxiv.org/abs/2607.09779
作者: Yunhong Zhang,Changjie Cao,Zhongli Zhou,Bingli Liu,Zongjie Cao,Zongyong Cui,Ying Yang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:The deep nonnegative matrix factorization (DNMF) technique is proposed to address the low interpretability of deep learning-based methods in extracting multilayer features from synthetic aperture radar (SAR) target samples. However, existing DNMF methods employ a layer-by-layer decomposition strategy, which is prone to causing error accumulation and local optimum, thereby hindering a consistent improvement in recognition accuracy as the number of layer increases. In this paper, a robust multilayer feature extraction method, termed generalized deep non-negative matrix factorization (G-DNMF), is proposed to address the above challenges in SAR automatic target recognition (ATR). The G-DNMF aims global optimality and derives the update rules for each parameter using lagrangian multiplier method. The new update formula indicates that both the DNMF method based on the encoding matrix and the mixing matrix are special cases of the proposed method, theoretically demonstrating the universality of proposed method. In general, the proposed method discards the layer-by-layer decomposition strategy, thereby effectively mitigating the risk of local optima and eliminating error accumulation, leading to a significant improvement in DNMF’s multi-layer feature extraction capability. The experimental results, by presenting the feature images extracted from each layer by G-DNMF and the reconstructed original images, verified the proposed method’s pure additive understanding of multi-layer features and demonstrated its interpretability. The experimental results based on MSTAR and OpenSARship datasets show that G-DNMF outperforms existing DNMF algorithms and their derivatives in terms of stability and recognition performance.
[CV-192] me Imprint: Learning Time-Aware Representations in Multi-Modal Knowledge Graphs
链接: https://arxiv.org/abs/2607.09777
作者: Pengyu Zhang,Klim Zaporojets,Congfeng Cao,Jia-Hong Huang,Paul Groth
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Multi-Modal Knowledge Graphs (MMKGs) enrich entities with multiple modalities such as text and images, yet entities with highly similar multi-modal features remain difficult to distinguish. Temporal information of an entity can serve as an additional modality to disambiguate such entities, but existing approaches rarely treat time as a separate modality alongside text and images due to two major challenges: (1) sparse temporal semantics, which hinder alignment with richer modalities, and (2) multiple timestamps, which introduce noise or reduce robustness in representation learning. To address these challenges, we propose Time Imprint, a framework that treats time as an entity-level modality and jointly aligns temporal, textual, and visual representations via a three-view contrastive objective. Additionally, to mitigate multi-timestamp ambiguity, Time Imprint studies a compact timestamp subset selection design space and aggregates the selected timestamps into a discriminative temporal embedding with attention pooling, balancing temporal specificity and robustness. Experiments on three MMKG benchmarks demonstrate that Time Imprint achieves state-of-the-art link prediction performance, improving Hits@1 by up to 6.07% overall and yielding up to 58% gains on the subset of the top-1% ambiguity samples. We further examine different fusion strategies and the sensitivity to timestamp availability and quality, clarifying when and why time-as-modality is most beneficial, while adding only modest training overhead. We release our code at this https URL.
[CV-193] Low-Power License Plate Detection and Recognition on a RISC-V Multi-Core MCU-Based Vision System ISCAS
链接: https://arxiv.org/abs/2607.09768
作者: Lorenzo Lamberti,Manuele Rusci,Marco Fariselli,Francesco Paci,Luca Benini
类目: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注: 5 pages, 2 figures, 5 tables. This paper has been accepted for publication in the IEEE International Symposium on Circuits and Systems (ISCAS). Copyright 2021 IEEE
Abstract:In this paper, we present the first (to the best of our knowledge) demonstration of a low-power MCU-based edge device for Automatic License Plate Recognition (ALPR). The design leverages on a 9-core RISC-V processor, GAP8, coupled with a QVGA ultra-low-power greyscale imager. The proposed visual processing pipeline uses a multi-model inference approach based on SSDlite-MobilenetV2 for license plate detection and LPRNet for optical character recognition, reaching a 38.9% mAP score for the first task and a recognition rate of 99.13% for the latter on public datasets. On real-world data, the pipeline recognizes registration numbers when the size of LP crops is as small as 30x5 pixels. Thanks to the applied compression and optimization strategies, the multi-model inference (687 MMAC) achieves a throughput of 1.09 FPS at a power cost of 117 mW when running on GAP8. Our solution is the first MCU-class device embedding such a level of network complexity, resulting to be 73x more energy-efficient w.r.t. precedent mobile-class ALPR system featuring a Raspberry Pi3. The proposed design does not resort to any hardwired acceleration engines, thus retaining full flexibility for future algorithmic improvements.
[CV-194] Knowledge-Constrained Shape Optimization with a Mixture-of-Experts Neural Operator for High-Confidence Design
链接: https://arxiv.org/abs/2607.09763
作者: Wenhao Fan,Yuanwei Bin,Jianghan Gu,Wenfa Luo,Jiao Xiang,Yuntian Chen,Shiyi Chen
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Computational Physics (physics.comp-ph)
备注:
Abstract:Engineering shape optimization faces challenges in both expert-dependent problem setup and surrogate-model reliability. In practical aerodynamic design, optimization settings such as editable regions, deformation ranges, and design-preservation constraints are typically specified manually by experienced engineers, while surrogate-based optimization may become unreliable for heterogeneous geometry databases and out-of-distribution designs. To address these challenges, we propose a knowledge-constrained shape-optimization framework that translates knowledge-based constraints and user intent into quantifiable parameters of DFFD-based deformation operators, enabling engineering-aware and controllable constrained optimization. We further develop a Mixture-of-Experts Neural Operator (MoE-NO) to improve drag prediction and trend consistency over heterogeneous aerodynamic datasets. Based on the MoE-NO encoder and Mahalanobis distance, an uncertainty-estimation strategy is introduced to detect out-of-distribution geometries and selectively trigger physics-solver feedback for local sample enrichment. Experiments on in-house MPV, SUV, and Sedan datasets show that MoE-NO achieves a test-set MAPE of 1.16% and a trend-prediction accuracy of 94.34% , outperforming the best baseline results of 1.52% and 90.34% , respectively. Vehicle shape-optimization experiments further yield CFD-validated drag coefficient reductions of approximately 4% to 10% .
[CV-195] ReflectWorld-MM: An Entity-Oriented Multi-Media Memory System for Open-Ended Video Streams
链接: https://arxiv.org/abs/2607.09759
作者: Xiaokang Ma,Yifan Sun,Zhihong Jin,Jie Gu,Yudong Luo,Shenyi Shao,Chu Tang,Jingmin Chen,Li Pu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Building assistants that can continually watch the world, remember what they see, and reason over their accumulated experience is a long-standing goal, and recently multimodal agents equipped with long-term memory over video streams have attracted increasing interest. Unfortunately, existing systems either keep their memory inside the model context or in a flat feature store, and organize it around frames rather than around the persistent entities a stream is really about, which confines them to bounded videos and weakens their ability to track who and what reappears over time. In this paper, we propose ReflectWorld-MM, an entity-oriented multi-media memory system for open-ended video streams. It consists of three parts. The first is a perception front-end that turns a streaming video into entity-resolved observations under a bounded short-term memory. The second is a hierarchical long-term memory, grounded in human memory theory, that couples a multi-scale episodic memory, an evolving entity-centric semantic memory, and a procedural memory. The third is a complete realization, built for real-world operation, that ingests arbitrary streams and plugs into off-the-shelf assistants. Across six long-video and lifelong-memory benchmarks, ReflectWorld-MM achieves the best accuracy on all six, outperforming strong memory agents and a frontier model.
[CV-196] RSLoRA: Training-free Rank Allocation for LoRA via Representational Sensitivity Probing
链接: https://arxiv.org/abs/2607.09757
作者: Jiaqi Liu,Haidong Kang,Qihui Zhao,Guo Yu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 16 pages, 3 figures
Abstract:Low-Rank Adaptation (LoRA) has become a cornerstone of parameter-efficient fine-tuning (PEFT); however, the conventional practice of uniform rank assignment ignores the functional heterogeneity of neural layers. Existing rank allocation methods typically struggle with a trade-off between computational intensity and heuristic simplicity: training-based methods suffer from prohibitive overhead, while pre-allocation methods fail to capture the dynamic task-specific representation manifold. In this paper, we propose RSLoRA (Representational Sensitivity LoRA), a training-free and gradient-free rank allocator driven by activation-space geometry. We identify a “sensitivity regime shift” across layers, observing that static weight analysis and local gradients are insufficient to reflect how updates reshape a model’s internal representations. To address this, RSLoRA introduces a virtual representational probing mechanism. By simulating adaptation through structured low-rank noise and measuring the resulting manifold displacement by using Effective Rank and Frechet Distance, we identify high-sensitivity modules that require higher rank capacity. Our framework effectively bridges the gap between expert-crafted heuristics and actual representational impact. Extensive evaluations demonstrate that RSLoRA consistently outperforms state-of-the-art allocators (e.g., AdaLoRA, GoRA) across mainstream benchmarks. By eliminating the need for iterative training-time adjustments and backward gradients, RSLoRA provides a highly efficient, robust, and representation-aware solution for large-scale model adaptation.
[CV-197] Cross-Subject Modeling for Widefield Calcium Imaging via Atlas-Aligned Spatiotemporal Tokenization ICML
链接: https://arxiv.org/abs/2607.09754
作者: Mohammad Hosseini,Eray Erturk,Saba Hashemi,Maryam M. Shanechi
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Neurons and Cognition (q-bio.NC)
备注: Published at the 43rd International Conference on Machine Learning (ICML) 2026. Code available at: this https URL
Abstract:Large-scale, multi-subject widefield calcium imaging provides unprecedented access to brain-wide cortical dynamics. However, the high dimensionality, complex spatiotemporal structure, and substantial task-irrelevant activity in widefield recordings have largely restricted modeling efforts to single-session analyses, limiting scalability and generalization. While multi-subject pretrained models have been explored for some neural modalities, multi-subject models for widefield calcium imaging have not yet been demonstrated; further, subject-invariant zero-shot behavior decoding remains elusive for multi-subject models across neural modalities more broadly. As a first step toward foundation modeling of widefield data, we introduce WiCAT, a multi-subject model that leverages self-supervised pretraining to both outperform single-session models and enable zero-shot behavior decoding on unseen subjects. WiCAT introduces an atlas-grounded tokenization scheme without session-specific components and learns globally shared spatiotemporal representations. Across multiple widefield datasets, the pretrained model supports lightweight downstream decoding, transfers across subjects, tasks, and datasets, and outperforms baseline models. Notably, the model also achieves robust zero-shot continuous behavior decoding and left-out brain region reconstruction on unseen subjects.
[CV-198] Unified Backbone Refinement for Diffusion Models via Internal-Latent Analysis ECCV
链接: https://arxiv.org/abs/2607.09753
作者: Haksoo Lim,Myeongjin Lee,Wonjoon Chang,Jaesik Choi
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 45 pages, 23 figures. Accepted at the European Conference on Computer Vision (ECCV) 2026
Abstract:Diffusion models have achieved remarkable success across diverse domains, with performance closely related to the denoising backbones that parameterize the score function. In this paper, we present a systematic, phase-aware analysis of diffusion components and show that abrupt, early-stage fluctuations in deep latents are strongly associated with artifacts. Guided by these findings, we introduce DUNE (Diffusion Unified Network refiNEr), a training-free refinement framework that detects abrupt deviations in deep low-noise internal latents using a shared EMA-based criterion, and applies backbone-specific suppression to the detector-selected entries. Although derived from U-Net, the same detect-suppress principle extends naturally to Transformer-based diffusion models by acting on the latents of deep self-attention blocks. Extensive experiments across multiple backbones indicate that DUNE improves fidelity while reducing hallucinations, offering new insight into where and when diffusion backbones should be controlled.
[CV-199] A Dynamic Scene Interaction Reasoning Framework for Scene-level Lane-Change Intention and Trajectory Prediction of Multiple Interacting Vehicles
链接: https://arxiv.org/abs/2607.09740
作者: Joshua Kofi Asamoah,Blessing Agyei Kyem,Eugene Denteh,Armstrong Aboah
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: 28 pages, 9 Figures, 16 Tables
Abstract:Safe motion planning in advanced driver-assistance systems and autonomous vehicles requires an accurate understanding of how the surrounding traffic scene is likely to evolve. However, many existing lane-change prediction methods remain centered on a single target vehicle, while multi-agent forecasting approaches often describe scene evolution only through future positions and provide limited explicit information about the maneuver associated with each vehicle. This study proposes a dynamic scene graph attention framework that predicts the lane-change intention and future trajectory of every relevant vehicle within a local traffic scene. The scene is represented as a time-varying interaction graph in which vehicles are modeled as nodes and their spatial and kinematic relationships are encoded through explicit edge features. Temporal graph-attention message passing captures evolving inter-vehicle dependencies and pre-maneuver cues, while an intention-guided decoder links each predicted maneuver to its corresponding future motion. A scene-level consistency objective further encourages compatible multi-vehicle futures. Experiments on the NGSIM I-80, NGSIM US-101, and highD datasets demonstrate consistent improvements over competing baselines. DSiGAT achieves intention prediction accuracies of 90.12% and 90.97% on NGSIM I-80 and US-101, respectively, and reduces trajectory RMSE by up to 52.94% relative to the strongest baseline. It also produces lower inter-agent collision rates and joint displacement errors, indicating more coherent scene-level predictions. Ablation, sensitivity, robustness, and qualitative analyses further validate the contribution of the proposed components and the effectiveness of the scene-focused formulation.
[CV-200] From ML Predictions to Informed Diagnostic Assistance Using the Toulmin Model of Argumentation
链接: https://arxiv.org/abs/2607.09664
作者: Anca Marginean,Adrian Groza
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:To provide a structured and interpretable assessment, we decompose the image-based diagnosis into components following the Toulmin model of argumentation. This model consists of a claim, grounds, warrant, qualifier, rebuttal, and backing. Consider a claim generated by a machine learning (ML) model for retinal diagnosis. Rather than accepting this claim at face value, one could either apply explainable AI (XAI) methods or adopt an argumentation-based approach. In our framework, a model specialized in biomarker extraction from images provides the grounds. The warrant-linking the grounds to the claim - is analyzed by an agent equipped with medical knowledge; in our architecture, this role is fulfilled by a MedGemma agent. The qualifier is determined based on the overall quantitative evaluation of both the warrant and grounds models. Finally, a rebuttal is constructed using image similarity measures computed with MedSigLip. All these components are presented to the human expert, enabling a more informed and critical assessment of the ML-generated diagnosis.
[CV-201] Diffusion MRI preprocessing affects ADC estimation and automatic PI-RADS v2.1 classification in bi-parametric prostate MRI
链接: https://arxiv.org/abs/2607.11385
作者: Christos Kanakis,Mathias Perslev,Tim Schakel,Silvia Ingala,Akshay Pai,Dennis Klomp,Chantal M.W. Tax
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注: 19 pages, 10 figures, ISMRM Diffusion workshop 2025, ESMRMB 2025
Abstract:Diffusion-weighted imaging (DWI) is acquired as part of bi-parametric prostate MRI, but suffers from artifacts that degrade downstream quantitative and diagnostic performance. While DWI preprocessing is standard in brain imaging, its adoption in prostate imaging remains limited and lacks standardized pipelines. This study investigated the effect of different DWI preprocessing strategies on apparent diffusion coefficient (ADC) estimation and automatic Prostate Imaging Reporting and Data System (PI-RADS) classification. 268 cases were derived from the fastMRI prostate cohort by sequentially applying denoising, Gibbs-ringing correction, and diffeomorphic registration for susceptibility distortion correction. ADC maps were compared using linear least squares (LLS) and iteratively-weighted LLS (IWLLS). A 3-class DenseNet classifier was trained to predict PI-RADS scores from multi-channel MRI inputs. ADC analysis revealed statistically significant differences across preprocessing pipelines, with LLS and IWLLS producing numerically equivalent maps. Linear relationships between ADC values were preserved across most datasets (PCC ~0.99), while distortion correction realigned DWI to T2w anatomy and altered ADC values accordingly (PCC ~0.90). Classification showed the best AUROC and sensitivity for high-risk PI-RADS classes in the fully processed dataset. False-negative analysis revealed this dataset produced the least overconfident incorrect predictions on high-risk classes, which is a desirable property for clinical triage. DWI preprocessing, particularly distortion correction, enhances both ADC map quality and the predictive power of deep learning models for PI-RADS classification, supporting the need for optimized preprocessing pipelines in prostate MRI.
[CV-202] Projection-Domain Sensitivity Analysis of Vertebral DRRs Under Intrinsic Calibration Perturbation
链接: https://arxiv.org/abs/2607.10551
作者: Lin Li,Chaochao Zhou,Benjamin Aubert,Junlin Guo,Junchao Zhu
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Numerical Analysis (math.NA); Computational Physics (physics.comp-ph)
备注:
Abstract:Accurate geometric calibration is essential for fluoroscopy-guided spinal imaging, digitally reconstructed radiograph (DRR) generation, and 2D–3D vertebral registration. Although calibration quality is typically evaluated using reconstruction-based metrics such as reprojection error, its influence on projection-domain consistency remains poorly understood. This study presents a synthetic framework for evaluating how intrinsic calibration perturbations affect vertebral fluoroscopic projections and downstream registration performance. CT-derived vertebral models and controlled cone-beam imaging geometry were used to generate DRRs with both ground-truth and perturbed intrinsic calibration parameters while maintaining identical anatomy and acquisition pose. Projection-domain changes were quantified using anatomical landmark displacement, contour distance, silhouette overlap, image similarity, and landmark-based 2D–3D registration accuracy in anterior–posterior (AP) and lateral (LAT) views. Results show that even small intrinsic calibration perturbations produce measurable changes in vertebral projection geometry, contour morphology, landmark localization, and DRR appearance. Sensitivity is strongly view dependent, with LAT projections exhibiting substantially greater deformation and anatomical displacement than AP projections. These projection inconsistencies also degrade downstream 2D–3D registration, particularly rotational alignment accuracy. The findings demonstrate that projection-domain consistency complements conventional reconstruction-based calibration metrics and provides a practical framework for assessing calibration robustness. This approach may improve the reliability of DRR generation and fluoroscopy-guided vertebral registration in image-guided spinal applications. Subjects: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Numerical Analysis (math.NA); Computational Physics (physics.comp-ph) Cite as: arXiv:2607.10551 [eess.IV] (or arXiv:2607.10551v1 [eess.IV] for this version) https://doi.org/10.48550/arXiv.2607.10551 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Lin Li [view email] [v1] Sun, 12 Jul 2026 03:34:29 UTC (4,281 KB)
[CV-203] Differentiable Proxy Learning for Adaptive Quantization Control in H.264 Video Coding
链接: https://arxiv.org/abs/2607.10478
作者: Qihan Xu,Ivan V. Bajić
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by IEEE MIPR 2026
Abstract:H.264 has been the most widely used video coding format for the past two decades due to its relative simplicity, efficiency, and wide availability of software and hardware implementations. However, optimizing codec parameters such as the quantization parameter (QP) for specific objectives (e.g., perceptual quality or machine vision tasks) is challenging due to the non-differentiable nature of standard video codecs. While differentiable proxies have recently been used to enable gradient-based optimization around standard codecs, their fidelity to the target codec is rarely explicitly characterized. In this paper, we propose a differentiable proxy learning method for H.264 intra codec to enable adaptive quantization control. Built upon a variable-rate learned compression model, the proposed proxy is made differentiable with respect to codec QP through a soft-indexing mechanism. It is then trained to approximate the rate-distortion behavior of H.264 under two quantization settings: global-QP, which uses one QP per image, and spatial-QP, which assigns QPs at the macroblock level. Using the frozen trained proxy, we develop a proxy-based adaptive quantization (AQ) framework for both perceptual optimization and machine vision tasks. Experimental results demonstrate that the proposed proxies closely approximate the rate-distortion behavior of H.264 intra codec. The resulting proxy-based AQ framework consistently improves rate-task trade-offs over fixed-QP H.264 baselines, achieving BD-rate reduction of up to 17.12% for semantic segmentation and 15.30% for MS-SSIM.
[CV-204] Neural Posterior Estimation for Inferring Weak Lensing Shear
链接: https://arxiv.org/abs/2607.09867
作者: Tim White,Dingrui Tao,Camille Avestruz,Jeffrey Regier, theLSST Dark Energy Science Collaboration
类目: Instrumentation and Methods for Astrophysics (astro-ph.IM); Cosmology and Nongalactic Astrophysics (astro-ph.CO); Computer Vision and Pattern Recognition (cs.CV); Applications (stat.AP)
备注: 17 pages, 10 figures, 3 tables
Abstract:The prevailing approach to inferring weak gravitational lensing shear from images involves detecting galaxies, estimating their ellipticities, and calibrating these estimates to correct for image noise, selection bias, and model misspecification. Characterizing the statistical model and assumptions underlying this pipeline is challenging, which makes it difficult to propagate uncertainty through its various stages. As an alternative, we propose to infer shear using neural posterior estimation (NPE), a type of simulation-based inference. We train a deep neural network to map a simulated multiband image to a variational distribution over the underlying shear field, thereby folding galaxy detection, deblending, measurement, and calibration into a single implicit inference step. Once trained, the network accounts for all features present in the simulated images, including potential sources of bias. In experiments on simulated constant-shear images with increasingly complex observational effects, NPE produces accurate and well-calibrated posterior approximations for both shear components in the presence of blended galaxies, spatially varying point spread functions, stars, and detector artifacts. These results demonstrate that NPE can be a viable shear estimation method in settings where all anticipated features and artifacts can be simulated, a requirement that will become increasingly feasible as simulation fidelity improves in the coming decades.
[CV-205] Slide-Level Active Learning Reduces Annotation Burden in HE images
链接: https://arxiv.org/abs/2607.09831
作者: Mahsa Vali,Zhilong Weng,Noémie Moreaua,Yuri Tolkach,Katarzyna Bozek
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Deep learning-based segmentation of histopathology whole-slide images (WSIs) requires large amounts of pixel-level annotations, which are costly and time-consuming to obtain. Active learning (AL) has been proposed to reduce this effort, but existing methods exhibit three key limitations. Uncertainty estimation is unreliable on partially annotated WSIs, patch-level acquisition is inconsistent with slide-level annotation workflows, and class imbalance in multi-class settings is not explicitly addressed. To address these challenges, we propose SHAL (Slide-level Hybrid Active Learning), a patient-level AL framework for annotation-efficient multi-class histopathology segmentation. SHAL integrates three complementary components: a foreground-aware strategy that suppresses bias from unlabeled background regions, a stage-adaptive mechanism that hybridizes predictive entropy and epistemic uncertainty across learning stages, and a class-aware strategy that prioritizes diagnostically relevant tissue classes. SHAL is evaluated on the TCGA colorectal cancer dataset. It achieves the highest Macro Dice at the full annotation budget (0.846) and reaches Dice greater than or equal to 0.80 using only 26 percent of the budget (50 of 190 slides), whereas competing methods reach this threshold only at 37 percent (70 slides). Across five independent external cohorts, SHAL attains the highest mean external Macro Dice (0.815) and the smallest internal-to-external generalization gap among all methods (0.025 at Round 3 and 0.026 at the full budget). The results indicate that patient-level hybrid uncertainty acquisition reduces annotation cost without sacrificing cross-domain generalization in computational pathology.
[CV-206] racking Intermittent Particles with Self-Learned Visual Features
链接: https://arxiv.org/abs/2607.09829
作者: Raphael Reme(IP Paris, BIA, IDS, IMAGES),Victor Piriou(BIA),Alison Hanson,Rafael Yuste,Alasdair Newson(IP Paris, IDS, IMAGES),Elsa Angelini(IP Paris, IDS, IMAGES),Jean-Christophe Olivo-Marin(BIA),Thibault Lagache(BIA)
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:In time-lapse fluorescence imaging, single-particle-tracking is a powerful tool to monitor the dynamics of objects of interest, and extract information about biological processes. However, tracked particles can be subject to occlusion and intermittent detectability. When these phenomena persist over a few frames, tracking algorithms tend to produce multiple tracklets for the same particle. In this work, we introduce self-supervised learning of visual features to compare tracked particles, and we exploit both visual and positional distances to robustly stitch tracklets representing the same particle. We demonstrate the performance of our stitching framework on time-lapse fluorescence sequences of Hydra vulgaris neurons. Results show high stitching precision, and reduction of errors made by previous algorithms on the same data by a factor of two.
[CV-207] Robustness and Stability Analysis of Differentiable Shift-Variant FBP for Cone-Beam CT under Challenging Acquisition Settings
链接: https://arxiv.org/abs/2607.09828
作者: Chengze Ye,Linda-Sophie Schneider,Yipeng Sun,Mareike Thies,Siyuan Mei,Paula Andrea Pérez-Toro,Siming Bayer,Andreas Maier
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted for publication at the Journal of Machine Learning for Biomedical Imaging (MELBA) this https URL
Abstract:The differentiable shift-variant filtered backprojection (SV-FBP) framework enables data-driven estimation of redundancy weights for cone-beam CT reconstruction under general source trajectories, removing the need for analytically derived weighting schemes. In this work, we present a systematic study of the robustness and adaptability of differentiable SV-FBP under challenging acquisition settings. We show that the framework remains stable across highly irregular and discontinuous trajectories, indicating that reconstruction performance is largely insensitive to trajectory ordering or continuity. Instead, the spatial distribution of sampling points plays a more dominant role. Under sparse-view conditions, differentiable SV-FBP achieves competitive reconstruction quality while providing an order-of-magnitude reduction in computation time compared to iterative reconstruction methods at moderate sampling densities. However, we identify a clear transition regime under severe undersampling, where the absence of iterative data consistency leads to performance degradation. Furthermore, we demonstrate that the framework remains applicable to non-planar multi-isocenter geometries, such as Lissajous-saddle trajectories, without requiring architectural modifications. These findings provide new insights into the behavior and limitations of the differentiable SV-FBP model and highlight it as a flexible and efficient solution for non-standard and robotic CBCT acquisition scenarios.
[CV-208] Performance Benchmarking and Optimisation of Clustering Algorithms for Local and Non-Local Similarity Measure in Medical Image Analysis
链接: https://arxiv.org/abs/2607.09821
作者: Sisipho Hamlomo,Marcellin Atemkeng
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Medical imaging generates high-resolution images posing significant storage, transmission, and computational challenges. While low-rank matrix approximation (LoRMA) techniques offer efficient compression by exploiting structural redundancy, global approaches often fail to preserve local details critical for diagnosis. This paper focuses on clustering techniques that exploit non-local self-similarity to identify structurally similar regions in medical images. These clusters can be used for post-processing tasks such as adaptive image compression. We evaluate five clustering techniques: k-means, mini-batch k-means, agglomerative hierarchical clustering, balanced iterative reducing and clustering using hierarchies (BIRCH), and bisecting k-means across MRI, ultrasound, and chest X-ray modalities. All clustering techniques were optimised using random search, and cluster quality was assessed using the Silhouette score, the Davies-Bouldin (DB) index, and the Calinski-Harabasz (CH) index. Results demonstrate that standard k-means and bisecting k-means generally achieve strong cluster cohesion and separation across modalities. However, they tend to form a small number of clusters with high intra-cluster variability, limiting their effectiveness for post-processing tasks such as adaptive compression. Agglomerative clustering outperformed other techniques for MRI and ultrasound in terms of intra-cluster homogeneity, making it more suitable for preserving fine diagnostic details. For chest X-rays, mini-batch k-means achieved the best balance between clustering quality and intra-cluster compactness. BIRCH consistently underperformed across all modalities.
[CV-209] CHM-Net: Center Heatmap-driven Macro-Micro Modeling Network for MRI-based Microbial Density Stratification
链接: https://arxiv.org/abs/2607.09812
作者: Jiaming Liang,Haolin Chen,Tingting Li,Bowen Yu,Qianyan Long,Tinghe Zhang,Xi Zhong,Xiaowei Hu,Xiaoqi Sheng,Hongmin Cai
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Microbial density is clinically important for tumor assessment and treatment decision-making, and recent advances in deep learning suggest that it can be non-invasively inferred from multimodal MRI. In this work, MRI-based Microbial Density Stratification (MRI-MDS) is first investigated as a patient-level representation learning task, and Center Heatmap-driven Macro-micro modeling Network (CHM-Net) is introduced for this task. CHM-Net first establishes the link between imaging phenotypes and microbial states through center heatmap-guided small-lesion response localization. Building upon this, it constructs patient-level macro-micro evidence from localized heatmap responses for microbial density prediction. Experiments on the novel GBNPC 2026 dataset constructed for MRI-MDS demonstrate the effectiveness of CHM-Net, achieving superior performance over representative baselines with a 12.06% absolute ACC gain over the strongest competing result. Additionally, auxiliary validation on two 3D medical image datasets further verifies its robustness across volumetric medical image classification scenarios. The project is available at this https URL.
[CV-210] Calibrated Hybrid CNN-Transformer for Retinal OCT Classification
链接: https://arxiv.org/abs/2607.09809
作者: Animesh Kumar
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注: 4 pages, 2 figures, 4 tables. Code, model weights, and REST inference API are available on GitHub and HuggingFace
Abstract:Deep models for retinal optical coherence tomography (OCT) classification report high accuracy but rarely report whether their confidence can be trusted – a gap that matters when a wrong-but-confident reading delays sight-saving treatment. We pair a hybrid convolutional-Transformer encoder with a gradient-boosting (XGBoost) classification head and a three-part clinical safety layer: confidence calibration, out-of-distribution (OOD) rejection, and per-prediction uncertainty flagging. On four-class OCT (84,495 scans) the model reaches 95.4% accuracy while cutting calibration error twelve-fold (expected calibration error, ECE = 0.0024), so the confidence it reports tracks its true accuracy. To our knowledge this is the first OCT classifier to validate all three safety mechanisms jointly, with public weights and reproducible multi-seed evaluation.
[CV-211] A Unified Model for Highly Accurate ECG-Free Dynamic Coronary Roadmapping Using Spatio-Temporal Transformers
链接: https://arxiv.org/abs/2607.09805
作者: Saahil Islam,Sebastian Piat,Venkatesh N. Murthy,Serkan Cimen,Puneet Sharma,Andreas Maier,Florin C. Ghesu
类目: Image and Video Processing (eess.IV); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: 13 pages, 10 figures
Abstract:Percutaneous Coronary Intervention (PCI) is a minimally invasive procedure used to restore coronary blood flow obstructed by atherosclerotic plaque. During PCI, repeated injections of iodine-based contrast agents are required to visualize the coronary arteries and guide interventional devices. However, frequent contrast injections increase radiation exposure and the risk of contrast-induced nephropathy, with acute kidney injury reported in up to 30% of patients with renal impairment. Dynamic Coronary Roadmapping (DRM) reduces these risks by overlaying a precomputed angiographic vessel map onto live fluoroscopy and continuously updating it throughout the procedure. Accurate DRM relies on precise cardiac phase matching between angiography and fluoroscopy, together with reliable catheter tip tracking for motion compensation. These tasks remain challenging in ECG-free settings and when only limited manual annotations are available. We present a unified DRM framework that simultaneously performs cardiac phase matching and catheter tip tracking for accurate real-time guidance. Our method employs a large-scale spatio-temporal encoder pretrained on 16 million X-ray frames to learn cardiac motion dynamics. To the best of our knowledge, this is the first application of large-scale spatio-temporal pretraining for motion compensation in DRM. We further introduce auxiliary tasks based on ECG R-peak detection and catheter tip tracking, improving optimization while eliminating the need for extensive catheter mask annotations. Finally, a majority-voting postprocessing strategy aggregates temporal predictions, improving robustness and providing a confidence score that correlates with phase-matching error. Comprehensive evaluation on clinical X-ray datasets demonstrates state-of-the-art performance, achieving low temporal misalignment and robust phase-matching accuracy suitable for real-time DRM.
人工智能
[AI-0] Invariant Learning Dynamics of Transformers in Inductive Reasoning Tasks
链接: https://arxiv.org/abs/2607.11875
作者: Tiberiu Musat,Tiago Pimentel,Nicholas Zucchet,Thomas Hofmann
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:We present a theoretical framework to explain the emergence of inductive reasoning abilities in Transformer language models. While previous works on Transformer learning dynamics have so far been mostly tied to specific tasks, we study a generalized class of inductive tasks that unifies several synthetic tasks known in the literature, including in-context n-grams and multi-hop reasoning. In this class, we theoretically prove that the training dynamics of attention models can be confined to a highly interpretable, low-dimensional invariant manifold. On this manifold, the learning dynamics are captured by a handful of interpretable coordinates rather than millions of parameters, making both theoretical and empirical analysis more tractable. Using this framework, we characterize how data statistics govern the competition between in-context and in-weights learning, we study how random initializations determine the `winning’ circuit when multiple solutions are possible, and we demonstrate that the coordinate frame associated with the manifold can be used to automatically detect which circuits have been learned in trained models. By casting circuit formation as a low-dimensional dynamical phenomenon, we take a step toward a predictive theory of how Transformers learn.
[AI-1] A Minimalist Retargeting-Guided Reinforcement Learning Recipe for Dexterous Manipulation
链接: https://arxiv.org/abs/2607.11874
作者: Yunhai Feng,Natalie Leung,Jiaxuan Wang,Lujie Yang,Haozhi Qi,Preston Culbertson
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Website: this https URL
Abstract:Recent work in humanoid whole-body control has found success with a simple recipe: retarget human motion to robot kinematic references, then train policies via reinforcement learning (RL) to track them. But how does this recipe transfer to dexterous manipulation? The answer is not obvious, as manipulation involves complex, contact-rich dynamics and requires delicate regulation of contact modes and forces. We present REGRIND, a minimalist retargeting-guided RL pipeline that learns dexterous manipulation policies from a single human demonstration. REGRIND retargets human hand-object motion to a robot reference that preserves hand-object spatial and contact relationships, trains a residual RL policy in simulation to track object-centric keypoints along that reference, and transfers the resulting policy zero-shot to hardware with careful system identification. The resulting policies produce fluid, human-like behavior on two different multi-fingered hands across contact-rich tool-use tasks, including operating a pair of scissors and turning a screwdriver. Through systematic hardware experiments, we identify and analyze the key factors that govern sim-to-real transfer in dexterous manipulation, offering practical guidance for retargeting-based learning in contact-rich settings. Videos and code are available at this https URL.
[AI-2] ransformer-Guided Swarm Intelligence for Frugal Neural Architecture Search
链接: https://arxiv.org/abs/2607.11826
作者: Romain Amigon
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE)
备注:
Abstract:Neural Architecture Search (NAS) has automated the design of deep learning models but traditionally requires massive computational resources, often measured in thousands of GPU-days. In this paper, we propose a frugal and memetic NAS framework designed to democratize architecture design on consumer-grade hardware. Our approach combines the global macro-search capabilities of an autoregressive Transformer controller, trained via Reinforcement Learning (RL), with the local micro-exploitation of an Artificial Bee Colony (ABC) algorithm. To prevent premature convergence during the RL phase, we introduce a dynamic entropy mechanism that forces topological exploration upon detection of performance stagnation. Evaluated on a standard GPU (NVIDIA RTX 3060), our hybrid method effectively resolves the “cold-start” problem inherent in metaheuristics. By algorithmically penalizing network depth, our framework actively mitigates model bloat: on the CIFAR-10 dataset, it discovers an efficient architecture reaching 84.85% accuracy with only \sim 174,000 parameters (significantly smaller than standard baselines like ResNet-20) in 3 hours of search time. Furthermore, we demonstrate the framework’s flexibility by applying it to credit card fraud detection, directly optimizing the F1-Score on highly imbalanced tabular data to reach a F1-Score of 0.71 with a compact network of \sim 4,600 parameters. These results suggest that our approach can yield tailored, accessible, and highly parameter-efficient deep learning models suitable for edge deployment.
[AI-3] Encoder-Side Neuron Identification and Amplification for Acoustic Perception in Large Audio-Language Models
链接: https://arxiv.org/abs/2607.11801
作者: Yu-Han Huang,Chih-Kai Yang,Ke-Han Lu,An-Yu Cheng,Hung-yi Lee
类目: ound (cs.SD); Artificial Intelligence (cs.AI)
备注:
Abstract:Large audio-language models (LALMs) often underperform on fine-grained, non-semantic attributes of speech, such as a speaker’s emotion, despite strong performance on speech content. Improving this without the cost of retraining calls for an effective inference-time intervention, yet most existing methods intervene only after the audio encoder and operate at a relatively coarse granularity. The encoder itself, where acoustic information is first extracted from the waveform, remains largely unexplored, especially at the level of individual neurons. We introduce IAAN, Identifying and Amplifying Acoustic Neurons, a training-free and label-free method that scores each feed-forward neuron in the audio encoder by contrasting its activation on the real waveform with that on a noise reference lacking the real audio’s acoustic information. IAAN then amplifies a small set of the highest-scoring neurons at inference. Across ten non-semantic speech attributes, IAAN improves average accuracy by 25.7 points on Audio-Flamingo-3, 21.4 on Qwen2.5-Omni, and 9.7 on Kimi-Audio. It also improves a model already explicitly fine-tuned to prioritize acoustic evidence. In controlled comparisons, both the encoder locus and neuron-level selectivity prove necessary for this gain. Intervening after the encoder, at the decoding side or inside the language model, yields little to no improvement, or even deteriorates accuracy. The improvement also depends on which specific neurons are amplified, not merely on their number, confirming that IAAN’s acoustic score succeeds in identifying the neurons that matter. These results show that a small, precisely targeted intervention inside the audio encoder is an effective and largely untapped way to strengthen the acoustic understanding of LALMs, opening a new direction for inference-time methods that improve acoustic perception through neuron-level access to the encoder.
[AI-4] Evaluating RE Practices for Explainability: Synthesizing Insights from Daimler Truck into an Explainable RE Framework Proposal
链接: https://arxiv.org/abs/2607.11771
作者: Umm-e- Habiba,Lucas Mauser,Jonas Fritzsch,Justus Bogner,Stefan Wagner
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:Explainability has emerged as a critical requirement for AI-based systems, particularly in safety-critical and regulated domains. Although prior research has proposed frameworks, patterns, and user-centered approaches to support explainability, there is limited empirical understanding of how existing Requirements Engineering (RE) practices support explainability requirements across the RE lifecycle, especially in an industrial context. This paper reports early findings from an ongoing industry-based study investigating how explainability requirements are elicited, specified, and validated using established RE techniques. We conducted a multi-phase qualitative study with eight practitioners at Daimler Truck, employing think-aloud protocols and moderated group discussions across requirements elicitation, specification, and validation steps. Our preliminary analysis reveals recurring challenges across all steps, including conceptual ambiguity during elicitation, limited testability and expressiveness during specification, and fragmented validation due to vague criteria and regulatory uncertainty. These findings indicate that current RE practices provide limited support to systematically address explainability requirements. The paper contributes empirical insights into step-specific and cross-cutting challenges and outlines a research vision toward developing an empirically grounded RE framework for explainable AI-based systems.
[AI-5] me-Lag-Aware Deep Reinforcement Learning for Flexible Job-Shop Scheduling in PPVC Module Factories
链接: https://arxiv.org/abs/2607.11725
作者: Ziheng Zhang,Wei Zhang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Optimization and Control (math.OC)
备注: 10 pages,5 figures, UR by IEEE TII
Abstract:Prefabricated prefinished volumetric construction moves most building work into module factories, whose production floor operates as a flexible job shop. A major complication is decisive: long post-operation time-lags caused by concrete curing, watertightness ponding tests, and paint drying, during which a module is blocked while its workstation stays free. On benchmark instances grounded in an official national prefabrication guidebook, these lags inflate even the optimal reference makespan by about 67% on average, and ignoring them at decision time, then repairing to feasibility, is worse than every dispatching rule. We adapt a state-of-the-art dual-attention deep reinforcement learning solver through three minimally invasive, individually ablatable extensions: lag-aware dynamics with an admissible reward bound, two anticipatory lag feature channels, and liveness-masked operation- and station-type embeddings. With every extension disabled the implementation reproduces the original solver exactly, so all gains are attributable to the adaptations. We release a public, guidebook-grounded benchmark generator. On held-out instances the learned policy is the strongest solver-free scheduler: it reaches within about 4% of a constraint-programming reference and beats every dispatching rule and a genetic-algorithm metaheuristic, with its advantage widening under capacity contention, and a single size-mixed policy carries this lead across the trained range of factory sizes. It needs no solver, model, or license in the loop and re-plans within seconds of a disruption; where an exact solver can be deployed, that solver remains the quality ceiling, a boundary we map explicitly.
[AI-6] Active Offline-to-Online Reinforcement Learning
链接: https://arxiv.org/abs/2607.11720
作者: Alper Kamil Bozkurt,Shangtong Zhang,Yuichi Motai
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Background: Offline reinforcement learning (RL) enables effective policies to be trained from large, previously collected datasets and subsequently improved through limited online interaction. This offline-to-online RL (O2O-RL) paradigm is particularly promising in nonstationary domains where interaction is costly or potentially hazardous. Standard O2O-RL pipelines train multiple candidate policies offline, evaluate them using off-policy or online evaluation, and then deploy and fine-tune the policy with the highest estimated value. However, as in offline pretraining, fine-tuning performance is highly sensitive to the choice of algorithm and hyperparameters, making it risky to commit to a single policy. Objectives: We study active policy selection for fine-tuning under a limited interaction budget in O2O-RL settings. To our knowledge, this is the first work to address this problem. Methods: We formulate the problem by identifying a fundamental trade-off between allocating online interactions to policy evaluation, which helps identify high-performing policies, and allocating them to fine-tuning, which improves policy performance. We then propose an approach that balances this trade-off by actively selecting policies for fine-tuning based on upper-confidence bounds on their future performance. These bounds are derived from locally linear performance forecasts fitted to observations obtained through online evaluation. Results: Across a diverse range of experiments, the proposed approach consistently outperforms existing O2O-RL baselines. Conclusions: Actively selecting and fine-tuning policies uses limited online interaction budgets more effectively than either committing to a single policy or dividing the budget equally among all policies. Our framework also advances offline RL toward practical deployment in real-world systems where online interaction is costly or risky.
[AI-7] VoxENES 2026: Benchmarking Generalization of Speech Spoofing Detectors Against LLM -Era TTS and Voice Conversion INTERSPEECH2026
链接: https://arxiv.org/abs/2607.11706
作者: Aastha Sharma,Guangjing Wang
类目: ound (cs.SD); Artificial Intelligence (cs.AI)
备注: Accepted in InterSpeech 2026
Abstract:Modern LLM-driven text-to-speech (TTS) and voice conversion (VC) systems produce synthetic speech that differs from the generators represented in many legacy spoofing benchmarks. This mismatch creates a temporal generalization gap that can overestimate detector robustness under real-world post-processing conditions. We bridge this gap by introducing VoxENES 2026, a bilingual (English and Spanish) benchmark of 53,628 audio samples generated using 10 contemporary speech synthesis methods and evaluated under 10 standardized post-processing conditions. Using VoxENES 2026, we benchmark eight pretrained detectors without fine-tuning and observe substantial performance degradation: the best model achieves 28.98% EER overall, while most perform near or below random chance across modern generators and perturbations. Our results highlight the reliance on brittle artifacts in current detectors and establish VoxENES 2026 as a practical testbed for developing robust audio spoofing countermeasures.
[AI-8] Agent Hacks Agent Agent : Autoresearch for Production-Agent Red-Teaming
链接: https://arxiv.org/abs/2607.11698
作者: Xutao Mao,Xiang Zheng,Cong Wang
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:Production LLM agents such as Claude Code and Codex operate over untrusted content, files, commands, and workspace state, making safety failures directly actionable. Red-teaming must therefore keep pace with evolving models and tools. Existing approaches mainly optimize attack success and preserve artifacts such as benchmarks, payloads, or attack programs, which record where attacks succeed but not the enabling conditions behind unsafe agent behavior. We study automated red-teaming for production LLM agents using one agentic research environment to discover reusable vulnerability knowledge about another. We present AHA, a falsifiable discovery loop that proposes a vulnerability hypothesis, constructs a falsifier, instantiates a valid attack, executes it in a sandboxed harness, reflects on the trajectory, and promotes confirmed findings into a Vulnerability Concept Graph (VCG). Each concept links an attacker-facing surface to an unsafe trajectory through a claim, enabling condition, falsifier, transfer prediction, and supporting evidence. Across Claude Code and Codex on three scenarios covering direct and indirect attacks, the discovered concepts reveal a reusable vulnerability core across models and agents. A frozen VCG requires no further search and outperforms the strongest frozen discovery baseline by 14.2 percentage points under the same single-shot protocol, while transferring across scenarios and attack channels. The resulting VCG provides an auditable artifact for production safety teams to inspect vulnerabilities, validate patches, and accumulate reusable safety knowledge. Our code is available at this https URL.
[AI-9] hink Through a Bottleneck: Hourglass Reasoning for Rigorous Induction
链接: https://arxiv.org/abs/2607.11696
作者: Huan Zhu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Self-refinement often fails to strengthen few-shot inductive reasoning in large language models. Prompting a model to explicitly state its inferred rule does little on its own. What actually matters is a structurally enforced isolation between reasoning stages, so that information can only pass between them as a compressed symbolic state. We introduce \textbfHourglass reasoning, which enforces strict context isolation between reasoning stages. The frozen LLM acts as a meta-constructor, building for each task a symbolic encoder–decoder: an Induction module compresses the support examples into a schema \phi (encoder) and a transient scaffold z ; a Deduction module derives rule T (decoder) from these and discards z ; an Implementer compiles (\phi, T) into artifacts; an error-driven Refiner revises (\phi, T) and regenerates artifacts from scratch. Only (\phi, T) crosses stage boundaries, so all refinement stays anchored to the rule. We evaluate Hourglass across three benchmarks spanning visual abstraction, hardware synthesis, and textual rule induction, using GPT-5.5 and Gemini 3.1 Pro. On ARC-AGI-2, it raises best-of-5 accuracy by up to 14 points over an iterative-refinement baseline. On ChipBench, it nearly doubles Verilog synthesis accuracy with GPT-5.5, from 31% to 58%. BBEH-Linguini draws on puzzles from the International Linguistics Olympiad, a setting where prior work has shown that explicit verbalization can hurt performance. Hourglass mitigates this tendency, and on Gemini 3.1 Pro, it reverses the effect entirely. Ablations confirm that these gains come from the isolation between stages and the quality of the initial induction, not from prompt wording or the particular symbolic form used. It is how information flows through the reasoning process, rather than the language used to express it, that drives inductive reasoning in frozen LLMs. Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2607.11696 [cs.AI] (or arXiv:2607.11696v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2607.11696 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Huan Zhu [view email] [v1] Mon, 13 Jul 2026 15:29:24 UTC (237 KB)
[AI-10] From World Action Models to Embodied Brains: A Roadmap for Open-World Physical Intelligence
链接: https://arxiv.org/abs/2607.11689
作者: Yuanzhi Liang,Xufeng Zhan,Haibin Huang,Chi Zhang,Xuelong Li
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: Ongoing work
Abstract:Artificial general intelligence ultimately requires agents that can reason and act in the physical world. Action models, vision-language-action policies, and world models have advanced this goal, while World Action Models (WAMs) are particularly promising because they connect candidate interventions with predicted consequences. However, progress remains fragmented: models use incompatible action spaces and prediction targets, datasets and tasks follow different conventions, and runtime systems expose limited interfaces for reuse and evaluation. We review the evolution toward WAMs and organize these limitations into three coupled gaps: model roles and representations, objectives and standardization, and system composition. Building on this analysis, we propose a co-evolution roadmap for physical intelligence centered on the \emphembodied brain, a long-term model target for integrating multimodal context, comparing candidate interventions, and issuing state-transition or capability requests rather than direct actuator commands. WAMs provide promising prototypes for its predictive functions, while a physical harness grounds model outputs through tools, controllers, verification, and trace logging. Shared contracts align heterogeneous models, data, tasks, and embodiments, and closed-loop post-training converts verified interaction into reusable experience. Together, these components define a modular physical-intelligence stack for adaptive and self-improving embodied agents.
[AI-11] Closing the Loop: An Access-Control Architecture for Automated Anomaly-Driven Network Revocation in IoT Deployments
链接: https://arxiv.org/abs/2607.11649
作者: Muhammet Emir Korkmaz,Kemal Bicakci,Yusuf Uzunay
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: 15 pages, 12 figures
Abstract:Network-based anomaly detection for IoT devices has matured to the point of reporting strong detection accuracy, yet most published systems stop at raising an alert and leave the question of automated enforcement to future work or to a programmable data plane that few real networks operate. This paper presents an access-control architecture that closes that loop using only standard, already-deployed protocols. Devices authenticate via IEEE 802.1X with EAP-TLS, and a RADIUS server acts as a continuous policy decision point capable of evicting an active session via a Change-of-Authorization Disconnect-Request and permanently excluding a device through certificate revocation. A central, contextual access policy engine continuously consumes the anomaly detector’s output and actuates this response over a narrowly restricted channel to the RADIUS server; the same engine is designed to be extensible to other access types, though this paper evaluates only the network access-control mechanism. This mechanism is driven by an anomaly signal from a one-class detector adapted from a prior MUD/SDN-based design, replacing its per-flow multi-model pipeline with passive traffic capture and a single fused model that combines a cluster-based, a volumetric, and a protocol-signature score. On a single testbed device, the detector reaches an AUC of 0.9964 and detects all 24 evaluated attack scenarios (eight attack types at three intensities) using roughly 43 \times less training data than the reference design, and the resulting alerts reliably trigger the automated disconnect-then-revoke response, which we measure to evict a device from the network in 335.8,ms on average and complete certificate revocation in a further 111.5,ms. We report this evaluation as a demonstration of the closed-loop architecture rather than of the detector itself, and discuss multi-device generalization as a concrete next step.
[AI-12] Xiaomi-Robotics-U0: Unified Embodied Synthesis with World Foundation Model
链接: https://arxiv.org/abs/2607.11643
作者: Xinghang Li,Jun Guo,Qiwei Li,Long Qian,Hang Lai,Yueze Wang,Hongyu Yan,Jiahang Cao,Xi Chen,Jingen Qu,Jiaxi Song,Nan Sun,Hanye Zhao,Futeng Liu,Wanli Peng,Heyun Wang,Yunhong Wang,Caoyu Xia,Jack Zhao,Diyun Xiang,Hangjun Ye,Heng Qu,Huaping Liu,Jason Li
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Recent foundation image and video generation models offer strong generalization and controllability, but their direct application to embodied scenarios is limited by requirements for multi-view consistency, geometric coherence, and robot embodiment constraints. Existing methods typically adapt foundation models with limited robot data, often sacrificing visual knowledge acquired during large-scale pre-training. We present Xiaomi-Robotics-U0, a 38-billion-parameter multimodal autoregressive model for unified embodied synthesis. It treats embodied generation as an extension of foundation image and video generation and jointly optimizes text-to-image generation, image editing, embodied scene generation, embodied transfer, and embodied video generation. This unified framework preserves the generalization of the pre-trained world foundation model while adapting it to embodied settings. Xiaomi-Robotics-U0 is the first model to support high-quality multi-view scene generation across multiple robot embodiments and to introduce structured, controllable embodied transfer for fine-grained editing while preserving multi-view consistency and interaction dynamics. It achieves state-of-the-art results on single-step and sequential generation tasks, outperforming GPT-Image-2.0 in human evaluations of embodied scene generation and transfer, ranking first on World Arena for embodied video generation, and improving the out-of-distribution success rate of pi_0.5 from 36.9% to 63.2% on challenging real-world manipulation tasks. These results show that foundation world models can serve both as embodied world models and scalable data engines for embodied intelligence. Code and checkpoints are available at this https URL.
[AI-13] Auditing the Risk Claims of Distributional Reinforcement Learning
链接: https://arxiv.org/abs/2607.11607
作者: Hari Prasad
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Machine Learning (stat.ML)
备注: 25 pages, 8 figures, 3 tables (main text); includes supplementary material
Abstract:Distributional reinforcement learning agents learn full return distributions that are increasingly read at face value: for interpretability, risk-sensitive control, and safety monitoring. We ask a question theory anticipates but that has not been measured directly: are the risk claims of a trained distributional agent true? Our audit combines a decision-relevant screening metric (the excess Wasserstein gap between the top two actions, which equals the mass by which first-order stochastic dominance is violated), ground truth from snapshot-restart Monte Carlo, and a statistical harness (permutation nulls, bootstrap refutation, FDR control) without which the audit itself manufactures false conclusions. Across QR-DQN, C51, and IQN on MinAtar (33 runs), 40-95% of the strongest claimed risk trade-offs are refuted at 95% confidence, the placement of the strongest claims is statistically indistinguishable from truth-blind, and essentially no claim is confirmable: for these agents, the learned “risk” reflects a training artifact rather than environment stochasticity. The artifact is structural (fully formed early in training, uncorrelated with final score, idiosyncratic to each seed) and appears unchanged at full-Atari scale, with every top Breakout claim of a pretrained near-state-of-the-art QR-DQN refuted. Positive controls of known magnitude confirm 96-100% of real claims (correlation 0.89-0.92): the reading measures the agents, not the audit. Acting on the heads’ CVaR advice at their most-flagged states ranges from beneficial to significantly worse than chance. Neither training for risk nor ensembling removes the artifact, and recalibration passes the audit only by nullifying the claims: the head is uninformative, not merely miscalibrated. We release the toolkit and document two silent pitfalls that produced convincing but wrong audits of our own.
[AI-14] Interaction Scaling: Grounding the Third Axis of Test-Time Compute
链接: https://arxiv.org/abs/2607.11598
作者: Bojie Li,Noah Shi
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:There are two standard ways to spend more compute at test time: let a model reason longer, or sample more attempts and keep one. Both share a hidden limit: they are internal. Every extra token comes from the same frozen weights and the same prompt, so neither can tell the model anything it does not already know. We study a third way, interaction: the model proposes an artifact, an external instrument observes how it actually behaves, and the model revises. Each cycle imports a real observation, so interaction breaks through the ceiling the other two hit. We argue that a single variable governs this third axis, grounding, and that it must hold on both sides of the loop. The feedback that drives revision must come from an instrument that actually observes the flaw, and so must the metric that scores the result. On hard coding tasks at a fixed token budget, reasoning-only and best-of-N sampling both plateau (the latter even when an oracle picks the best sample), while every interaction strategy keeps improving; our proposer-reviewer harness reaches a perfect 100% pass rate with no run-to-run variance, and the gain holds across three model families. On rendered visual artifacts, the usual judge (a vision-language model, or VLM, reading a screenshot) rates 14 of 15 visibly broken figures “perfect,” because the screenshot hides the flaws before the judge can see them. A tool that measures the real layout instead shows the loop removing 40-74% of defects across four modalities; and that same VLM, used as the reviewer, makes slide layouts worse where the measuring tool repairs them. Interaction scaling is real and distinct from reasoning and sampling, but only visible when both the feedback and the metric are grounded. Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2607.11598 [cs.AI] (or arXiv:2607.11598v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2607.11598 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-15] MAGIC: Transition-Aware Generation of Navigable Multi-Scene Game Worlds with Large Language Models
链接: https://arxiv.org/abs/2607.11594
作者: Tsz Hei Fan,Choi Wing Fung,Yuxuan Wan,Shuqing Li,Michael R. Lyu
类目: Artificial Intelligence (cs.AI); Graphics (cs.GR)
备注:
Abstract:Multi-scene navigation (clearing an objective in one bounded space and then crossing a portal into the next) is a defining feature of contemporary 3D games, but authoring it is laborious: every portal must have consistent endpoints on both sides, each interior must remain navigable once it is furnished, and the resulting connectivity must be kept consistent across many files. Recent large language model (LLM) and multimodal LLM (MLLM) scene generators have made single-interior synthesis dramatically cheaper, yet they produce one scene at a time and cannot, by naive repetition, yield a connected multi-scene world. We identify three obstacles that single-scene methods leave unsolved: cross-scene consistency, in-scene navigability, and the evaluation of whether a transition actually works. We present MAGIC, a prompt-to-project system that addresses all three. MAGIC is a four-stage pipeline that turns a single natural-language prompt into a runnable multi-scene game project: it plans a shared transition-aware intermediate representation, specifies each scene while enforcing portal reachability with a flood-fill validator, generates the scenes together with their transition scripts, and combines them into one project. Because existing single-scene fidelity metrics never execute a transition, we further introduce a transition-focused evaluation agent that runs each transition in play. On a new benchmark of 100 multi-scene cases, MAGIC produces an executable project for every case and reaches 0.99 precision, 0.95 recall, and 0.96 F1 on end-to-end transition identification; stage by stage, it recovers more ground-truth portals and yields markedly more navigable layouts than an LLM baseline and Holodeck. Our code is available at this https URL.
[AI-16] HCRMap: Pressure-Aware Hot-Expert Residency Mapping for 3.5D MoE Chiplet Inference
链接: https://arxiv.org/abs/2607.11586
作者: Yongqin Zhang
类目: Artificial Intelligence (cs.AI)
备注: 15 pages, 8 figures, 2 tables
Abstract:Mixture-of-Experts (MoE) large language models (LLM) activate only a small number of experts during inference, but token routing introduces persistent expert hotness skew: a small set of hot experts continuously receives most tokens, while the remaining experts are lightly loaded. On 3.5D multi-chiplet systems, this skew not only causes compute imbalance but also amplifies pressure on communication, memory bandwidth, I/O, and execution queues. Therefore, the core problem is not simply to reduce token movement, but to dynamically place and reuse hot expert replicas across different memory tiers. This paper proposes HCRMap, a hot expert residency mapping framework for pressure-aware expert replica management in 3.5D MoE inference. Based on expert hotness, weight loading cost, migration overhead, and runtime resource pressure, HCRMap dynamically determines which experts should be promoted, retained, demoted, or evicted. It then maps routed token groups to suitable resident replicas, thereby jointly mitigating communication, memory, and queue bottlenecks. Experimental results show that HCRMap reduces end-to-end latency by 43.6% and 43.0% over Hydra in the prefill and decode stages, respectively; by 34.5% and 33.1% over MoEntwine; and by 46.7% and 46.0% over PIMoE. Comments: 15 pages, 8 figures, 2 tables Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2607.11586 [cs.AI] (or arXiv:2607.11586v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2607.11586 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-17] Structure-Feature Aligned Graph Learning via Alternating Constrained Optimization
链接: https://arxiv.org/abs/2607.11577
作者: Chengcheng Yan,Qingsong Wang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:We introduce a constrained two-view framework for node prediction that aligns structure-conditioned GNN embeddings with a structure-free feature prior learned by an anchor model. Conventional Graph Neural Networks (GNNs) couple feature transformation and neighborhood aggregation, which renders them vulnerable to topology noise and heterophilous connections. To decouple this dependency, our framework utilizes an independent anchor network to capture intrinsic attribute features via a self-supervised reconstruction objective. Furthermore, we propose a Channel-Split Adaptive Gated GNN (CSAG-GNN) that dynamically routes representations between global spectral smoothing and local spatial discrimination through a node-wise gating mechanism. We propose a stable cyclic alternating optimization strategy to solve the resulting coupled bi-level objective, preventing mutual representation drift during training. Empirical results on both homophilous and heterophilous benchmarks show balanced performance gains and structural robustness over competitive baselines.
[AI-18] Heuristic Learning for Active Flow Control Using Coding Agents
链接: https://arxiv.org/abs/2607.11565
作者: Paul Garnier,Jonathan Viquerat,Elie Hachem
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Fluid Dynamics (physics.flu-dyn)
备注:
Abstract:Active flow control involves nonlinear dynamics, partial observations, and computationally expensive simulations, making controller design particularly challenging. Deep reinforcement learning (DRL) has emerged as a powerful framework for such problems, but its success typically relies on large numbers of simulator interactions and produces neural-network policies whose decision process often remains difficult to interpret. In this work, we investigate a different paradigm: instead of optimizing neural-network parameters, we use modern coding agents to search directly for explicit executable feedback laws. We introduce a constrained heuristic-learning protocol in which an agent iteratively proposes, evaluates, and revises controller implementations while interacting exclusively through the public benchmark interface. The proposed framework is evaluated on 13 active flow-control benchmarks spanning one, two, and three-dimensional problems and compared against the strongest available DRL baselines under identical simulation budgets. The discovered heuristic controllers match or outperform the best DRL policy in 10 of the 13 environments while remaining compact, interpretable, and directly inspectable. Beyond aggregate performance, the resulting controllers reveal physically meaningful feedback mechanisms, transfer successfully across more challenging configurations, and remain competitive under varying Reynolds and Rayleigh numbers, actuator counts, and observation sparsity. These results suggest that heuristic learning through coding agents constitutes a credible and complementary alternative to conventional reinforcement learning, combining competitive performance with physically interpretable controller representations. Prompts and source code are available at this https URL.
[AI-19] Learning Residual Kinematic Corrections for Continuous Neural Decoding via Reinforcement Learning
链接: https://arxiv.org/abs/2607.11530
作者: Jiamian Li,Niall McShane,Attila Korik,Naomi du Bois,Karl McCreadie,Leen Jabban,Benjamin Metcalfe,Özgür Şimşek,Damien Coyle
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Decoding continuous three-dimensional (3D) motor imagery (MI) using non-invasive electroencephalography (EEG)-based brain–computer interfaces (BCIs) remains challenging due to signal variability and residual decoding errors. Deep learning architectures such as convolutional neural network–long short-term memory (CNN–LSTM) models can capture spatial and temporal dynamics for continuous kinematic decoding; however, systematic residual errors persist in predicted trajectories. We propose a two-stage decoding framework that applies reinforcement learning (RL) to perform residual kinematic correction on the outputs of a CNN–LSTM decoder (CNN–LSTM–RL). The RL agent is trained offline without direct EEG input and instead operates on predicted kinematic trajectories to optimize movement accuracy relative to target trajectories. Decoding performance was quantified using Pearson correlation coefficients ( r ) and Root Mean Square Errors (RMSE) along the x, y , and z axes. Compared to CNN–LSTM applied alone, CNN–LSTM–RL improved the mean correlation from 0.5076 to 0.7181 ( p = 0.0005 ) in 2D and from 0.6420 to 0.7780 ( p = 0.0059 ) in VR, with relative gains of 41.5% and 21.2% , respectively. Correspondingly, RMSE was reduced from 0.0890 to 0.0532 (2D, p 0.0001 ) and from 0.0714 to 0.0441 (VR, p 0.0001 ), representing relative reductions of 40.2% and 38.2% . These findings demonstrate that this scalable framework enhances 3D BCI MI decoding by correcting kinematic errors via offline residual RL without extra neural data, advancing neurorehabilitation, prosthetics, and virtual interaction.
[AI-20] AutoMatBench: An Automatic Optimization Toolkit for the Acceleration of Material Properties Prediction Benchmarking
链接: https://arxiv.org/abs/2607.11526
作者: Hongxiao Li,Wanling Gao
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Material property prediction (MPP) infers key properties from chemical composition and structure, accelerating the discovery and optimization of novel materials. In the realm of MPP, MatBench is a widely accepted benchmarking tool that defines over ten significant problems and provides the paradigm of performance evaluation for AI prediction models. Even though MatBench works well in benchmarking the performances of prediction models on in-distribution (ID) tasks and datasets, it lacks the ability to reflect their performances on out-of-distribution (OOD) material data, resulting failure in new material discovery. By combining the pipelines of MatBench and the existing researches on OOD performance evaluation, this study enables a huge space of benchmarking configurations, comprehensively reflecting the performances, abilities, and disadvantages of various AI prediction models. This work reports that the discrepancy of performances at different configuration values is huge and can be illustrated with prior knowledge and novel insights, therefore consideration of causal effect of configurations on performance results is necessary. In case of the impossibility of enumerative benchmarking at every configuration, this work further proposes AutoMatBench, an automatic toolkit with Bayesian optimization. Experiments with AutoMatBench reports that, within twelve steps of optimization, the similar results with MatBench and former OOD research can be accessed while more than half of the cost are saved. Besides, this tool also yields more essential findings on MPP benchmarking, positively contributing to the cost and efficiency of new material discovery.
[AI-21] CDFM: Towards a General-Purpose Causal Discovery Foundation Model
链接: https://arxiv.org/abs/2607.11508
作者: Jie Qiao,Ruichu Cai,Zijian Li,Weilin Chen,Pengfei Hua,Boyan Xu,Zhengming Chen,Zhifeng Hao,Peng Cui
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
备注:
Abstract:Causal discovery, the process of recovering underlying causal structures from observational data, is a fundamental pursuit across scientific disciplines. Over the past decades, numerous algorithms have been developed to tackle this challenge through workflows tailored to the specific causal mechanisms underlying each type of dataset, demonstrating effectiveness across a wide range of applications. However, as the volume and heterogeneity of real-world data continue to grow, this dataset-specific approach inevitably leads to a fragmented, test-driven paradigm that struggles to scale to the demands of modern scientific discovery. To address this, we formulate the Causal Discovery Foundation Model (CDFM) as a unified, general-purpose framework for zero-shot structural inference. To ensure reliable generalization across unknown domains, we first investigate the theoretical boundaries of causal identifiability, revealing the indispensable role of causal prior mechanisms in this process. Building on these insights, we formulate a principled variational framework that treats unknown causal mechanisms as latent variables and mathematically decomposes the intractable marginal likelihood into distinct, tractable learning modules. The variational decomposition provides a conceptual design principle for the architecture design of CDFM, while comprehensive causal knowledge guides the large-scale synthesis of our pretraining data. By pretraining on a massive, highly diverse space of synthetic structural causal models, CDFM successfully internalizes complex statistical asymmetries. Extensive experiments demonstrate that CDFM consistently outperforms traditional algorithms, driving a paradigm shift toward a general-purpose causal discovery foundation model.
[AI-22] Proxy Exploration and Reusable Guidance: A Modular LLM Post-Training Paradigm via Proxy-Guided Update Signals
链接: https://arxiv.org/abs/2607.11505
作者: Daocheng Fu,Rong Wu,Yu Yang,Xuemeng Yang,Jianbiao Mei,Licheng Wen,Pinlong Cai,Yong Liu,Botian Shi,Yu Qiao
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Post-training is essential for refining the domain-specific capabilities of large language models (LLMs), yet existing reward optimization and distribution matching methods tightly couple policy exploration with distribution alignment. This coupling forces expensive exploration directly on the policy model and severely hinders the asynchronous generation, reuse, and cross-model transfer of optimization signals. In this paper, we propose Proxy-guided Update Signal Transfer (PUST), a novel post-training framework that fundamentally decouples update-signal exploration from distribution alignment. Instead of utilizing the primary model for costly exploration, PUST employs a lightweight proxy model as an efficient testbed to discover high-reward behaviors. We extract the relative improvement signal between the proxy’s initial and optimized states, transferring this directional update to the primary model to guide its policy alignment. This decoupled pipeline, comprising proxy exploration, update-signal extraction, and signal transfer, significantly reduces computational overhead and enables optimization signals to be asynchronously generated, cached, and reused. Crucially, by transferring relative improvements rather than absolute policy distributions, PUST naturally supports weak-to-strong improvement and seamless cross-model transfer. Systematic evaluations on Qwen3-family models across math and code domains demonstrate that update signals extracted from substantially weaker proxies can robustly and adjustably enhance stronger primary models. Ultimately, PUST transforms post-training from a monolithic online optimization process into a highly modular, reusable, and cost-efficient paradigm.
[AI-23] Comparative Analysis of GAT and BERT for Human-Like Playtesting
链接: https://arxiv.org/abs/2607.11501
作者: Kleio Fragkedaki,Theodoros Panagiotakopoulos,Matteo Biasielli,Hui Wang
类目: Artificial Intelligence (cs.AI)
备注: 2025 IEEE Conference on Games (CoG)
Abstract:Accurately modeling and understanding player experience is crucial for designing engaging puzzle games. To achieve this, a common approach involves collecting diverse user data to train predictive playtesting models that mimic player behavior. However, existing data-driven methods often lack the ability to capture the full range of player strategies and require extensive feature engineering and network architecture modeling. This limitation becomes particularly evident when new game mechanics or features are introduced, which necessitate continual adjustments to the models. To addrss these challenges, we propose a more generalized representation that reduces - or even eliminates - the need for ongoing feature-engineering maintenance. Specifically, we investigate two general-purpose network architectures: (a) a transformer-based model (BERT) and (b) a graph attention model (GAT), both of which are designed to effectively capture the relational structure of Candy Crush Saga (CCS) game boards. Our experiments compare these approaches to Convolutional Neural Networks (CNN) baselines, revealing better performance on challenging board configurations and underscoring the benefits of our generalizable representation.
[AI-24] See like a Robot: Robot-Centric Pointmaps for Vision-Language-Action Models
链接: https://arxiv.org/abs/2607.11498
作者: Byungkun Lee,Dongyoon Hwang,Dongjin Kim,Hojoon Lee,Minho Park,Jaegul Choo
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: Project page: this https URL
Abstract:Vision-language-action (VLA) models predict robot actions from visual observations and language instructions. These actions are defined in the robot’s own 3D coordinate frame, yet most VLAs observe the scene in the camera frame, creating a frame mismatch between where the scene is observed and where actions are defined. The mismatch is benign under a fixed viewpoint, where the policy can memorize a single observation-to-action mapping, but grows harder as large-scale datasets aggregate demonstrations across diverse camera setups and the policy must generalize this mapping across viewpoints. We address this mismatch with robot-centric pointmaps, images whose pixels store the 3D coordinates of scene points in the robot frame. Pointmaps provide robot-frame 3D geometry while preserving the dense H x W grid expected by pretrained 2D VLAs, so they integrate into existing VLAs with minimal architectural change. On RoboCasa, pointmaps improve both pi0.5 and SmolVLA and outperform representative camera-viewpoint and 3D-aware baselines. In real-robot experiments, their advantage over an RGB-only policy widens when the camera is moved to a placement unseen during training.
[AI-25] IG-GAN: A Generative Adversarial Network for Aerodynamic Data Generation Based on Intrinsic Geometry
链接: https://arxiv.org/abs/2607.11497
作者: Ying Yan,Liwei Hu,Xiaoming Zhang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Existing generative models learn data distributions in flat Euclidean space. However, most data in our real world are manifolds embedded in high dimensional Euclidean space. Therefore, we propose an intrinsic-geometry-based generative adversarial network (IG-GAN) for data generation in the field of aerodynamics. The generator of the IG-GAN represents aerodynamic data as a piecewise smooth manifold constructed by Bézier surfaces, and the generator tries to learn the coefficients of each Bézier surface to further combine multiple Bézier surfaces into a smooth manifold automatically. The discriminator in the IG-GAN is a radial-basis-function based discriminator (RBF-D). Experimental results show that IG-GAN achieves lower predicted Mean Squared Errors (MSEs) than those of three baselines. Specifically, on the Burgers’ equation dataset, IG-GAN reduces the predicted MSE of velocity u by 97.41% compared with state of the art SSL-Transformer. Additionally, on the ONERA M6 aircraft dataset, IG-GAN reduces the overall MSE of nine aerodynamic coefficients by 82.95% compared with SSL-Transformer.
[AI-26] Agent ic Skill Optimization over Lie Algebroids
链接: https://arxiv.org/abs/2607.11493
作者: Sridhar Mahadevan
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Category Theory (math.CT)
备注: 20 pages
Abstract:Agentic systems increasingly improve themselves by editing skills: prompts, rubrics, plans, tool contracts, examples, validators, and traces. Skill edits are not independent coordinates in a vector space: they are local repairs to structured artifacts whose effects are observed only after rollout, validation, and critique. Distinct edits can have the same immediate visible effect while differing in routing context, template state, guardrail scope, or future composability. The order of edits can matter as well: repairing a schema before a normalization rule need not be equivalent to applying the same edits in the reverse order. This paper introduces a new framework for skill optimization called LASKO, for Lie Algebroid SKill Optimization. LASKO models typed, anchored Markdown skills as the base category and available edit policies as sections of a controlled Lie algebroid with anchor \rho . The anchor maps an edit policy to its visible Markdown effect; the kernel \ker(\rho) represents latent template, routing, or implementation structure; and the algebroid bracket measures noncommuting edit composition. As shown in the paper, LASKO achieves order-of-magnitude speedups in skill optimization in our preliminary benchmark results, primarily because it substitutes inexpensive Lie-bracket screening tests that run in microseconds, before investing in expensive validations that require running large language models. On a causal extraction from natural language task, LASKO achieved a speedup of almost 15 \times compared to a brute-force approach that validated all edits by running them through a DeepSeek V3.1 4-bit model with 671B parameters.
[AI-27] Enhancing Query Efficiency for d-DNNF Representations Through Preprocessing
链接: https://arxiv.org/abs/2607.11492
作者: Jean Marie Lagniez,Emmanuel Lonca
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:In this paper, we investigate preprocessing techniques aimed at improving the efficiency of accessing models of propositional formulas represented in conjunctive normal form (CNF). We focus on three fundamental tasks: uniform sampling, direct model access, and model enumeration. Our analysis reveals that most state-of-the-art preprocessors, when they do not preserve formula equivalence, are generally unsuitable for these tasks. In contrast, we demonstrate that preprocessors which preserve model counts can be effectively leveraged, provided relevant preprocessing information is maintained. To validate our approach, we perform extensive experiments on a diverse suite of benchmarks from multiple domains. The experimental results show that our preprocessing methods are both efficient and robust, yielding significant performance improvements for model access queries when CNF formulas are compiled into d-DNNF representations.
[AI-28] A Multimodal Dataset for Large Language Model Applications in the Energy Domain
链接: https://arxiv.org/abs/2607.11459
作者: Costas Mylonas,Magda Foti
类目: ystems and Control (eess.SY); Artificial Intelligence (cs.AI)
备注:
Abstract:This paper presents the mAIEnergy dataset, an open-access, multimodal corpus developed to support Large Language Model (LLM) applications in the energy sector. The dataset integrates approximately 50,000 textual documents, 20,000 images, 25 million numerical time series records, and 2 million geospatial and relational data entries. It includes policy and regulatory texts, scientific articles and news articles, satellite and contextual imagery, electricity system measurements, weather observations, statistical indicators, and geospatial representations of energy infrastructure and related entities. All data have been harmonized into structured, ready-to-use formats, accompanied by consistent metadata and reproducible data retrieval and preparation workflows. The dataset can serve as a foundational energy knowledge base, allowing energy stakeholders to integrate additional open-source or proprietary data. The mAIEnergy dataset adheres to Findable, Accessible, Interoperable, and Reusable (FAIR) principles, enhancing its applicability for AI-driven energy research, modeling, and decision-making.
[AI-29] he Ebb and Flow of Multimodal Focus: Scheduling Visual Relay Windows for Grounded VLM Reasoning
链接: https://arxiv.org/abs/2607.11436
作者: Wencheng Ye,Yi Bin,Yujuan Ding,Hongye Fang,Zheng Wang,Xing Xu,Jingkuan Song,Yun Zhang,Sirui Da,Heng Tao Shen
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Vision-language models increasingly succeed on multimodal reasoning benchmarks, yet their visual evidence often becomes unstable once it enters the language stack, weakening evidence-grounded reasoning. To understand this fragility, we examine the internal dynamics of VLMs through a mechanistic lens and uncover a stable three-stage redistribution of multimodal attention focus across depth: an early question-conditioned organization, a critical middle visual-dominant relay, and a late return to answer formation. We operationalize the middle phase as the Visual Relay Window (VRW), and show that its geometry varies with task demand, is causally tied to grounded generation, and distinguishes unsupported answers from stronger reasoning trajectories. Guided by this internal rhythm, we propose TRACE, a task-adaptive inference-time control framework with lightweight trained modules. It reshapes relay allocation during prefill and preserves assembled visual support after handoff during decoding. Across four open-weight VLM backbones and seven benchmarks, TRACE delivers large gains on grounding-sensitive settings, improving them by 4.33 points on average and by up to 6.6 points, while also improving reasoning-heavy tasks. These results show that explicitly controlling multimodal focus across depth offers a unified and effective mechanism for strengthening evidence-grounded multimodal reasoning.
[AI-30] Omni-Decision: A Progressive Evidence-State Agent System for Omni-Modal QA
链接: https://arxiv.org/abs/2607.11433
作者: Ming Ma,Yi Zhu,Yiran Zhong,Feida Zhu,Weigao Sun,Junhan Shi,Lingrui Mei,Tianming Yang,Steven Hoi
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Omni-modal evidence-seeking QA requires agents to answer questions whose evidence is sparsely distributed across videos, audio, images, web pages, and computation results. Existing agentic multimodal systems often leave evidence in scratchpads, tool trajectories, or free-form histories, making it difficult to track what has been grounded, what remains missing, and when the evidence is sufficient to answer. We propose Omni-Decision, a training-free evidence-state system that turns omni-modal QA into a query-scoped evidence-closure process. For each query, Omni-Decision maintains a structured evidence state containing confirmed evidence, unresolved conflicts, fact and computation dependencies, and open evidence needs. A shared state view conditions planning, evidence acquisition, validation, repair, and finalization. Heterogeneous observations from media, web, computation, and verification modules are normalized, judged, and committed through deterministic state updates. This design enables targeted evidence acquisition, preserves sparse cross-modal cues, and provides inspectable control over repair and stopping. Omni-Decision achieves 45.6% accuracy on OmniGAIA and 58.3% on WorldSense, improving over the baselines by +27.3 and +30.2 percentage points, respectively. No-state ablations and trajectory audits further support the role of explicit evidence-state control in multi-step omni-modal evidence seeking.
[AI-31] A Glimpse into Long-term Physical Coexistence with Intelligent Robots
链接: https://arxiv.org/abs/2607.11377
作者: Weiqi Jin,Peijun Tang,Kuncheng Luo,Baifu Huang,Binyan Sun,Haotian Yang,Shangjin Xie,Jianan Wang
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Long-term physical coexistence with intelligent robots requires more than capable robot policies. A persistent robotic assistant must support diverse user-facing interfaces, maintain long-horizon memory of people and preferences, coordinate across robot embodiments, and translate human intent into safe physical execution. We introduce PHILIA, a multi-robot agent built around a robot gateway abstraction. PHILIA retains the rich interaction and tool ecosystem of OpenClaw while exposing robot-local runtimes, onboard perception, navigation, speaker, and robot policies through a unified capability interface. This design decouples low-frequency, high-semantic agent reasoning from high-frequency, low-level robot execution, enabling plug-and-play integration of user interfaces, robot embodiments, and policy backends. As a result, the user experience becomes compositional: advances in user interfaces, robot embodiments, robot policies, navigation, or interaction algorithms can improve the overall experience without redesigning the system. We validate the architecture on Astribot S1 robots while designing the robot gateway contract to support future heterogeneous robot platforms through a shared capability interface for observation, task execution, navigation, speech playback, status monitoring, and task cancellation. We present representative use cases in which agent memory and scene understanding are grounded in robot actions. These span interactive household scenarios, ranging from simple organization to challenging long-horizon and dexterous service tasks, such as packing a backpack and lifting a garbage bag. We highlight the human-robot interaction flow, where contextual understanding of user intent and preferences, together with human-in-the-loop confirmation or adjustment during execution, is essential for effective assistance.
[AI-32] BackgroundMellow: A Multi-Modal Cohesive Framework for Narrative-Driven Rich Cinematic Soundscape Generation
链接: https://arxiv.org/abs/2607.11364
作者: Ajitesh Jamulkar,Aritra Hazra
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Multimedia (cs.MM)
备注: 7 pages
Abstract:Generating immersive, synchronized and cinematic audio for long-form textual narratives remains a significant challenge in multi-modal AI. While current Text-to-Audio (TTA) frameworks successfully synthesize isolated sound effects, they struggle with narrative cohesion, temporal alignment, and cinematic emotional depth. We present BackgroundMellow, a framework that treats story-to-audio generation as a precise orchestration and signal processing problem. This framework is enabled without ground-truth through a master-specialist agent architecture that decomposes text into precise and multi-layered audio cues, generates each category of sounds with suitable specialist model, and superimposes the soundscapes to create a unified and aligned audio segment. Our pipeline is built over Tango2 latent diffusion model for environmental synthesis alongside a novel Cinematic BGM Retriever mined from professional soundtracks. To automate the sound mixing process, we use an NLP based module that predicts precise audio parameters, like start time, duration, and relative loudness, based on the narrative timeline. We further empirically evaluate and show the efficacy of the proposed framework leveraging nearest-neighbor retrieval against a curated dataset of YouTube cinematic trailers to measure temporal synchronization, coverage, and spectral richness.
[AI-33] OpsMem: Dual-Memory Reasoning with Cross-Memory Resonance for Failure Diagnosis
链接: https://arxiv.org/abs/2607.11357
作者: Yongqian Sun,Rongchen Gao,Yu Luo,Wenwei Gu,Shenglin Zhang,Qingyi Guo,Qiuai Fu,Yaoliang Wu,Dan Pei
类目: Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注: 6 pages, 5 figures
Abstract:Failure diagnosis in modern software systems requires iterative evidence acquisition and hypothesis reasoning guided by operational experience. Existing LLM-based methods improve diagnosis through agentic reasoning or knowledge augmentation, but they often lack a mechanism to coordinate the evolving diagnostic state with operational experience during iterative diagnosis. We propose OpsMem, a dual-memory framework that maintains a short-term memory for the current diagnostic state and a long-term memory for reusable operational experience. OpsMem uses cross-memory resonance to activate state-relevant long-term memory, conditions multi-agent diagnosis on the short-term and activated long-term memories, and consolidates reusable experience from solved incidents back into long-term memory. Experiments on a real-world Huawei microservice failure diagnosis dataset show that OpsMem outperforms representative agentic-reasoning and knowledge-augmented baselines, improving Match and Relevant by up to 46.88% and 18.39% over the strongest baseline, respectively.
[AI-34] Understanding the Impact of AI Code Assistants on Security API Usage: An Empirical Study RAID2026
链接: https://arxiv.org/abs/2607.11348
作者: Zahra Mousavi,Chadni Islam,M. Ali Babar,Alsharif Abuadbba,Kristen Moore
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: Accepted for publication at the 29th International Symposium on Research in Attacks, Intrusions and Defenses (RAID 2026)
Abstract:AI code assistants are transforming software development, but their implications for software security remain a major concern, particularly in the context of security APIs. These APIs are critical for safeguarding software systems, yet their complexity often leads to incorrect use and serious vulnerabilities. Developing an evidence-based understanding of how AI assistants influence developers’ use of these APIs is therefore essential for informing effective mitigation strategies. While a few user studies have examined the broader impact of AI assistants on software vulnerabilities, the use of security APIs remains unexplored from a developer-centered perspective. This study addresses this gap by presenting the first empirical investigation into how AI code assistants affect professional developers’ use of security APIs. We conducted a study with 44 developers who completed security API programming tasks with and without GitHub Copilot assistance. Our findings show that, while Copilot improves functional correctness and marginally reduces certain insecure patterns, it does not significantly improve secure API usage. We also found that developers rarely raised security concerns when engaging with Copilot, and many did not recognize that their final implementations remained insecure. Finally, we offer recommendations for enhancing security awareness among developers and propose future research directions to support safer AI-assisted software development.
[AI-35] From Neural Network Decisions to Training Cases: An Exact Account via Case-Based Decision Theory
链接: https://arxiv.org/abs/2607.11347
作者: Manli Yan,Yuebin Lin,Yaowen Yu,Yong Zhao
类目: Artificial Intelligence (cs.AI)
备注: Preprint. 15 pages, 9 figures, 4 tables. Includes appendix
Abstract:Neural networks increasingly guide decisions in high-stakes domains such as medical diagnosis, credit approval, and energy bidding. Audit in these settings requires case-level evidence: which training cases support an action and what outcomes they carried. Case-based decision theory (CBDT) formalizes this reasoning by aggregating outcome support from remembered cases. We show that an OLS action readout fitted on a fixed neural representation admits an exact case-based decomposition. Each action score is a weighted sum of training-case returns, with coefficients determined by empirical Gram geometry. We identify a sufficient regime for CBDT similarity semantics; outside it, the coefficients should generally be treated as signed Gram-geometric influence. The decomposition yields audit signals that trace scores to training cases, measure action coherence, and identify weak support. Across synthetic CBDT, PJM, Adult Income, and Default Credit tasks, the method recovers case-level preference structure and achieves the highest mean Top-30 consistency among compared attribution baselines, while remaining competitive on support reconstruction. The audit requires only fitting an OLS top-layer probe, without retraining the representation or accessing the original optimization trajectory; probe fidelity is measured by score reconstruction.
[AI-36] Compile Then Page: Executable SOP Programs and a Capability-Gated Runtime for Procedural LLM Agents
链接: https://arxiv.org/abs/2607.11346
作者: Chenglin Yu,Li Yin,Ying Yu,Hongxia Yang,Ming Li
类目: Artificial Intelligence (cs.AI); Programming Languages (cs.PL)
备注: 9 pages, 3 figures, 5 tables
Abstract:Enterprise agents must follow long-horizon, conditional, safety-critical standard operating procedures (SOPs). We compile machine-readable SOP constraints into executable pseudo-code and run them with a program-guided (PG) stack machine that pages the active frame while an LLM performs semantic execution. A three-arm SOPBench study across six models separates representation from runtime: compiled text never significantly hurts and gains up to 16.0 points where official prose underperforms. Runtime guidance is capability-gated. Two strong models independently show positive seven-domain PG contrasts (58:19 and 75:31 discordant pairs), whereas weak models are harmed. A full-program cursor ablation (active frame first, complete program retained) recovers much of the strong-model refusal gain; selective visibility adds a smaller improvement. Paired probe and audit measurements track this divide to spontaneous state discipline rather than reconstruction ability. On Bank the three primary arms rise from 70.4 to 86.4 to 92.8, with 100% refusal correctness. Practical guidance: compile first; enable active-frame paging only after a model-level discipline check.
[AI-37] Fail-Aware and Explainable Test Oracle Prediction
链接: https://arxiv.org/abs/2607.11342
作者: Yue Zhao,Binish Tanveer,Jelena Zdravkovic
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:Despite their central role in fault detection, test oracles remain challenging to construct effectively. Recent learning based methods address this challenge by automatically generating test assertions, yet even if syntactically correct, they are often ineffective in revealing bugs. Rather than generating assertions, this study explores a different approach by training a model to directly predict whether a given test prefix passes or fails. We present FOCAL, an emerging code LLM-based discriminative oracle predictor. It learns from labeled pairs of test prefixes and methods under test, employs losses that emphasize failing cases during training, and grounds its predictions in statement level behavioral evidence. Compared with the baseline method SEER, we substantially improve performance on failing cases for unseen projects and provide richer explanations. A preliminary evaluation on fault-detection benchmarks and automated test-generation artifacts shows that our approach is highly accurate within its training distribution and substantially improves failure detection on previously unseen projects where prior discriminative oracles collapse. Moreover, the highlighted statements are supported by behavioral explanation checks. These early results suggest that fail-aware discriminative oracle prediction can complement existing approaches such as fuzzing, search-based testing, and LLM-based test generation. These techniques produce test prefixes at scale but often lack fault oriented oracles. In future work, FOCAL could take generated test prefixes and attach fault-aware predicted oracles to them, turning high-volume input generation into executable tests that are more likely to expose semantic failures.
[AI-38] AutoVSR: Automatic Visual-to-Symbolic Reasoning for Symbolic Expression Generation from Circuit Schematic
链接: https://arxiv.org/abs/2607.11338
作者: Zhe Xiao,Longfei Li,Xu He,Haoying Wu,Zixing Zhang,Mingyu Liu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Symbolic expressions can effectively characterize and predict circuit behavior, but deriving them directly from circuit schematics is challenging. This process requires accurate visual-to-symbolic construction of circuit structure from images and correct multi-step symbolic derivation, both of which impose strict correctness requirements. This work proposes AutoVSR, an automated framework for visual-to-symbolic generation of circuit expressions using Vision Language Models (VLMs). By reconstructing circuit diagrams into an executable intermediate representation (Executable IR) and leveraging a symbolic solver for reasoning, AutoVSR significantly improves the accuracy of symbolic expression generation. AutoVSR introduces two key innovations: an IR construction method guided by component rule retrieval and verification-based feedback, and a symbolic solver implemented as a planning agent equipped with a symbolic tool library for reliable multi-step derivation. Compared with end-to-end VLM approaches and specialized methods on the main symbolic expression generation task, AutoVSR achieves accuracy improvements of 30.01–59.45% and 41.96–51.84%, respectively. Moreover, AutoVSR surpasses closed-source state-of-the-art VLMs in inference cost and computational efficiency. Code is available at this https URL.
[AI-39] Verifier-Guided Twelve-Tone Composition: A Generate-Verify-Repair Harness for Symbolic Music Generation
链接: https://arxiv.org/abs/2607.11334
作者: Congren Dai,Danni Zhao,Enyang Liu,Michael Ching Yam,Zhancheng Guo,Siyi Gu,Wentao Yang,Bo Dai,Xiaobing Li,Maosong Sun
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models can produce superficially legal twelve-tone scores that collapse into degenerate textures. We introduce a neuro-symbolic harness that wraps a language-model proposer in a generate-verify-repair-trace loop with symbolic verification. The complete pipeline improves event-local consistency without claiming whole-piece legality. Across 40 controlled tasks and four paired models, audited delivery yield rises from 13.3% under raw generation to 48.1% with the harness, which explicitly abstains otherwise. The pass rate of a narrower collision and serialisation-consistency check rises from 33.5% to 58.3%, while degeneracy remains near 0.05, including under exploratory adversarial prompting. A blinded evaluation by five experts also shows a descriptive aggregate preference for harness candidates over raw generation in adherence, perceived legality, coherence, and overall quality.
[AI-40] PRISM Edit: One Vector for All Temporal Answers
链接: https://arxiv.org/abs/2607.11327
作者: Chen Huang(1),Qi Zheng(1),Ruiqin Zheng(2),Long Zeng(1),Yuantong Xu(2) ((1) Tsinghua University, (2) ByteDance)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Chen Huang and Qi Zheng contributed equally. Corresponding authors: Long Zeng, Yuantong Xu
Abstract:Model editing keeps large language models (LLMs) up to date without retraining, but temporal facts expose a limitation of the prevailing locate-and-edit paradigm: an update is not always a replacement. When a fact changes, the new answer should become current while the old answer may remain correct in historical time contexts. Building on this insight, we use causal tracing to show that LLMs already support this distinction via a two-stage internal computation: early MLP layers retrieve a time-agnostic subject representation, and later layers modulate it with temporal context to yield the time-correct answer. Motivated by this finding, we introduce PRISM Edit, which optimizes a single polysemous representation across temporal contexts and leverages the model’s inherent modulation pathway to route it to temporally correct predictions, without any architectural modification. We evaluate on TimeConflict, a new temporal editing benchmark we introduce, and on temporally augmented CounterFact. PRISM Edit improves over the best baseline by +23.3 Temporal Consistency (TC) and +33.7 Current Relative-time Score (CRS) on average while being more than 2x faster. Code and data are publicly available at this https URL.
[AI-41] Calibrated e-CUSUM Decoding for Quantized Reasoning Models: Why Token Log-Probability Is the Wrong Observable for Decoding Monitors
链接: https://arxiv.org/abs/2607.11317
作者: El Hassane Ettifouri(1),Ayoub Belfatmi(1),Mahaman Sanoussi Yahaya Alassan(1),Walid Dahhane(1) ((1) Novelis Research, Paris, France)
类目: Artificial Intelligence (cs.AI); Information Theory (cs.IT)
备注:
Abstract:Low-bit quantization makes small reasoning models inexpensive to deploy but can degrade their chains of thought. This motivates decoder-side monitors that intervene when generation becomes unreliable. We show that a natural candidate, the centered token log-probability increment \log p(w_t)+H_t , is the wrong observable for this purpose. Under the model’s own sampling law it is a mean-zero martingale by construction, so it measures sampling self-consistency rather than trajectory health and is nearly silent during confident repetition, where both \log p(w_t) and entropy are close to zero. We introduce a training-free decoding controller that combines (i) a degeneration-aware alarm score fusing token uncertainty with explicit verbatim repetition and (ii) a calibrated e-process-inspired sequential detector. The raw product process is Ville-valid under a conditional-mean null, while the deployed CUSUM-floored statistic is treated as an empirical change detector because the score is history-dependent and autocorrelated. On GSM8K with DeepSeek-R1-Distill-Qwen-1.5B in FP16 and INT4, calibration turns a monitor that fires on 93–95% of generations into a selective detector of failing traces ( \phi \approx 0.3 , precision \approx 0.6 against a 0.38 base rate). In this pilot, the controller reduces measured verbatim-degeneration signals and yields a positive but statistically inconclusive INT4 accuracy change from 63% to 69% (paired McNemar p=0.18 , n=100 ), at a 28% token-budget cost. We also find that non-termination, rather than looping, is the dominant failure mode on GSM8K. The main contribution is methodological: an explanation of why centered token log-probability is inadequate for decoder monitoring and a calibrated, cautiously evaluated replacement.
[AI-42] Programming Language Policy as an AI Literacy Equity Problem: A 15-Nation Comparative Analysis
链接: https://arxiv.org/abs/2607.11314
作者: Adrian-Marius Dumitran,Iulia-Maria Popescu
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
备注: Presented at 2nd International Workshop on AI Literacy Education For All (CEUR proceedings)
Abstract:The promise of AI literacy ``for all’’ confronts a structural challenge embedded in how nations organise secondary computer science education. In most systems, a general-track subject – Digital Literacy, ICT, TIC, or SNT – bears the weight of universal AI literacy, while a specialist Informatics course serves STEM pathways separately. Yet the content and depth of the general track are shaped by governance decisions made largely with reference to the specialist one. This paper presents a comparative analysis of curricula and examination frameworks across fifteen countries, identifying two structural challenges. First, in several systems a significant portion of students completes secondary education without any formal programming exposure. Second, among those who do receive CS education, a \emphSyntax Ceiling emerges: Python-based instruction reaches most students, while the algorithmic depth associated with C++ remains concentrated in elite STEM tracks. Drawing on reform cases spanning centralised mandates (France, China, Japan), assessment-driven systems (Poland, Romania, South Korea), and recent universal reforms (Switzerland, Kazakhstan), we show that governance structures and high-stakes examinations are the primary drivers of both challenges – and that specialist and general-track language choices are rarely independent, linked through shared teacher pipelines that curriculum policy seldom acknowledges. Achieving genuine AI literacy for all requires confronting not just curriculum content, but the access architectures and resource constraints that determine who receives it – and at what depth.
[AI-43] Efficient Test-Time Optimization for Multi-Agent Proof Autoformalization
链接: https://arxiv.org/abs/2607.11307
作者: Tian-Shuo Liu,Shiyuan Zhang,Zijie Geng,Haoyu Liu,Runjie Xu,Pengyuan Wang,Lei Yuan,Yang Yu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Full-proof autoformalization bridges extensive mathematical proofs in natural language with formally validated reasoning, offering a pathway to elevate the ceiling of verifiable mathematical reasoning. Unlike statement-level formalization, proof autoformalization is a long-horizon challenge requiring coordination of claims, contexts, and dependencies across many proof steps, yet has only recently come under focused study. Current approaches either rely on costly model training or apply excessive, unguided repair at inference time. To this end, we introduce ToMap, a multi-agent framework that structures proof autoformalization as a Decomposer-Formalizer-Prover pipeline with efficient test-time optimization guided by formal verification and semantic rubrics for proof quality. Rather than distributing test-time compute across all agents, we perform bottleneck analysis and identify the Decomposer as the critical bottleneck: the quality of its atomic, self-contained proof units directly determines whether downstream agents can successfully formalize and prove each step. ToMap therefore treats the Formalizer and Prover as downstream executors and efficiently focuses test-time compute on Decomposer refinement. This refinement follows a loop inspired by GEPA, evolving prompts over candidate decompositions and using formal verification progress together with semantic proof rubrics to define a Pareto frontier that guides the next decomposition update. Experiments on ProofFlowBench show that ToMap improves over the best previous method by 19.0% when evaluated by both syntactic correctness and semantic faithfulness, while requiring lower test-time cost. Scaling analysis shows that most gains emerge within a few iterations of decomposition evolution, guiding test-time budget selection.
[AI-44] owards Predictive Aligned and Scalable Robot Learning
链接: https://arxiv.org/abs/2607.11270
作者: Peijun Tang,Shangjin Xie,Baifu Huang,Binyan Sun,Haotian Yang,Kuncheng Luo,Weiqi Jin,Shilin Fang,Jianan Wang
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Learning, at its core, extends beyond memorization to the ability to reason and solve novel problems by navigating a space of possibilities. We introduce Lumo-2, a latent world-action model that generates actions by reasoning over world dynamics in latent space. The learned latent world dynamics capture physically grounded visual transitions, naturally encoding future possibilities and providing a unified substrate for cross-modal alignment. This formulation enables predictive reasoning akin to world modelling while remaining lightweight and focused on physical dynamics relevant to control. Central to our approach is the hypothesis that action generation quality is governed by the geometry of the latent space. We observe that standard reconstruction-based action tokenization objectives induce representations biased toward low-level signal fidelity, leading to misalignment between reconstruction quality and downstream control performance. To address this limitation, we propose a multi-stage modality pre-alignment strategy in which action representations are progressively aligned with latent world dynamics, vision, and language. This process enforces cross-modal consistency, promotes abstraction, and induces a structured latent space for predictive reasoning. We provide a systematic empirical study of latent world modelling and modality alignment, analyzing their roles in scaling laws and out-of-distribution generalization. Results show that Lumo-2 consistently outperforms strong vision-language-action (VLA) and world-action model (WAM) baselines, with gains on challenging real-world tasks requiring temporal reasoning, physical understanding, or high control complexity, including long-horizon and dexterous manipulation. These findings suggest that structured multimodal alignment and predictive reasoning are fundamental principles for advancing embodied intelligence.
[AI-45] Valid ne Necessary: Diagnosing Latent Inefficiency in Chain-of-Thought
链接: https://arxiv.org/abs/2607.11266
作者: Daeyeop Lee,Hwanjo Yu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Chain-of-Thought (CoT) prompting has significantly advanced the reasoning capabilities of Large Language Models (LLMs), yet it often incurs substantial computational costs due to over-reasoning: the generation of redundant, verbose, or irrelevant steps. While existing reasoning step evaluators effectively detect logical fallacies and factual errors, our analysis reveals a critical blind spot: they fail to penalize valid but inefficient reasoning steps that inflate token usage without contributing to the solution. To systematically diagnose this limitation, we introduce RIV-GSM8K, a diagnostic benchmark injected with five distinct types of inefficiencies, including circular reasoning and excessive decomposition. Diagnostic experiments reveal that state-of-the-art evaluators struggle to distinguish these inefficiencies from necessary reasoning. To address this gap, we propose CAID (Context-Aware Information Density), a training-free metric grounded in information theory that identifies low-utility steps. To validate the metric’s practical utility, we apply it within PACE, a post-hoc compression strategy. Additional control experiments show that the gains of PACE are not explained by trivial pruning: compared with random step removal and PRM-based compression baselines, it preserves accuracy at substantially higher compression rates. Empirical results on GSM8K, StrategyQA, and ARC-Challenge demonstrate that PACE reduces token consumption by 31-53% while maintaining accuracy, confirming that CAID successfully distills informational froth from reasoning chains without compromising deductive validity.
[AI-46] Bringing Back Rule Induction to Fluid Intelligence Research? An Initial Validation of the ARC-AGI Benchmark in Humans
链接: https://arxiv.org/abs/2607.11263
作者: Jasmin Thelen,Oliver Wilhelm
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Two competing perspectives on fluid intelligence (gf) measures propose that performance is primarily constrained either by working memory capacity or by the ability to induce novel relations. The first perspective is currently dominant in measurement, as evident from the use of a limited set of recurring rules, whereas the second perspective is reflected in many definitions but rarely present in measurement. The ARC-AGI benchmark predominantly requires rule induction and was proposed as a measure of gf for both humans and artificial systems. However, its psychometric properties have not yet been examined in human samples. We therefore investigated the psychometric characteristics and nomological network of ARC-AGI in a first study with 100 participants. A compilation of ARC-AGI items showed good psychometric properties and correlated substantially with figural fluid intelligence as measured by a figural reasoning test (\rho = .63). Associations with figural originality were weak. These findings provide initial support for the validity of ARC-AGI as a measure of human fluid intelligence. Future research should include more rule induction tasks as well as additional multivariate covariates. This study is unusual by studying a task in humans that was initially designed for machines. We suggest systematically embedding AI benchmarks into the nomological network of human cognitive abilities to enable more systematic evaluation and interdisciplinary cooperation.
[AI-47] An Empirical Study for GUI Test Migration from Android to OpenHarmony System
链接: https://arxiv.org/abs/2607.11245
作者: Yakun Zhang,Xinjia Chen,Yiyun Chen,Yuxia Zhang,Mingyi Zhou,Xiang Gao,Shaokun Zhang,Li Li,Yunming Ye
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:To reduce the substantial engineering effort required to test the corresponding applications from Android to OpenHarmony, migrating existing GUI test cases has become a critical problem. However, current research neither proposes solutions tailored for OpenHarmony nor provides a systematic evaluation of migration approaches on this system, leaving developers with limited empirical guidance in practice. In this paper, we present the first systematic empirical study of test migration from Android to OpenHarmony. Specifically, we first construct a dataset referred to as the ATH Benchmark, comprising 36 commercial applications with an average of over 9 billion downloads, along with 108 manually designed test cases. Second, we select two state-of-the-art test migration approaches (i.e., ReSPlay and ITeM) and adapt these two approaches to enable their execution on OpenHarmony. Third, we use the preceding infrastructure to evaluate these two approaches from three perspectives, including testing performance, root causes of failures, and the impact of OpenHarmony characteristics. Our results reveal that existing test migration approaches are less effective (15% success-rate on ReSPlay and 26% success-rate on ITeM) in Android-to-OpenHarmony scenarios. Through an in-depth analysis of failed cases, we identify that test performance is primarily hindered by OpenHarmony-specific characteristics, including technical architecture differences and unique ecosystem traits. Utilizing these findings, we propose an enhanced approach based on ITeM, referred as ITeM-HM, which incorporates specific OpenHarmony system features. As a result, ITeM-HM successfully achieves a 214% success-rate relative improvement over the original ITeM (from 26% to 81%).
[AI-48] DeepBias: Adaptive In-depth Probing of Social Biases in LVLMs
链接: https://arxiv.org/abs/2607.11228
作者: Anqi Li,Jie Zhang,Zhongqi Wang,Songkai Xue,Jiahao Wang,Shiguang Shan,Xilin Chen
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
备注:
Abstract:While Large Vision-Language Models (LVLMs) demonstrate remarkable capabilities, they remain highly susceptible to embedded social biases. Existing bias evaluation protocols predominantly rely on static datasets, which provide only a superficial assessment, as their fixed test cases cannot adaptively evolve to measure the true depth and limits of model vulnerabilities. We introduce DeepBias, an adaptive framework for the in-depth probing of social biases in LVLMs with carefully designed agents. Our approach operates through a dynamic ‘‘generation-evolution-probing’’ loop. First, a generative ProposerAgent synthesizes test data and is iteratively updated via Direct Preference Optimization (DPO) based on the target LVLM’s responses, exploring model-specific failure modes. Second, an autonomous skill-driven DiggerAgent rewrites each test data across multiple probing turns, adaptively selecting from a curated skill library of deepening and rewriting strategies. At each turn, this process is conditioned on the model’s previous response, enabling progressively deeper biases to be exposed. Furthermore, we build a benchmark named DeepBiasBench using our framework. By employing an ensemble of five diverse state-of-the-art LVLMs as anchors, the benchmark captures vulnerabilities shared across architectures. Comprehensive experiments demonstrate the effectiveness of our framework and show that DeepBias provides a challenging benchmark for in-depth bias evaluation, establishing an evolutionary paradigm for LVLM safety assessment.
[AI-49] Heterogeneous Agent Cohorts for Safe Open-Ended Exploration with Runtime Constraint Memory
链接: https://arxiv.org/abs/2607.11226
作者: Tengjiao Liu
类目: Artificial Intelligence (cs.AI)
备注: 12 pages, 1 figure, 12 tables
Abstract:LLM agents today are caught in an awkward bind. Lock them down with static safety instructions and they rarely venture beyond the obvious; give them free reign with tools and multi-agent debate, and safety violations quickly follow. Rather than forcing a single model to juggle both creativity and caution, we separate the concerns across specialized roles. A Disrupter generates unconventional proposals, a Validator enforces hard runtime checks at the tool gateway, and a Broker pulls in distant but relevant analogies. Failures are not discarded – they are compiled, via MCTS, into compact, signed constraint patches we call Scars. These patches are cached locally and inherited by future cohorts, turning repeated failures into reusable, low-cost runtime constraints. In a spatial-semantic sandbox (N=20 runs, p0.01), our cohort reaches remote targets where debate fails, the Validator prevents all executed breaches, and Scars reduce token consumption by 15.1% by avoiding redundant validator checks. Furthermore, credit-based Communication Allocation Scores (CAS) restrict outbound bandwidth, reducing overall token costs by 55.9% under resource constraints.
[AI-50] PREF-Gate: Provenance-Constrained Relational Evidence Fusion with Validation-Gated Selection for Graph Fraud Detection
链接: https://arxiv.org/abs/2607.11212
作者: Liming Liu,Chao Hu,Mingfei Lu,Yiwei Ge,Xingle Li,Heyuan Shi
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Relational fraud detection can exploit both label-free graph context and label-derived neighborhood evidence, but these two information sources obey different validity conditions. In particular, neighborhood risk becomes invalid when a queried node’s own label, or any validation or test label, enters its construction. We formulate this issue as provenance-constrained relational evidence use and present PREF-Gate, an auditable decision framework with two fixed experts and a finite validation gate. The context expert uses attributes, one-hop means, feature residuals, and degree descriptors without labels. The evidence expert adds self-excluded, training-label-only neighborhood risk and empirical-Bayes summaries that expose support, uncertainty, availability, and shrinkage. Before test inference, the gate selects either expert or one of three pre-specified probability mixtures and fixes the decision threshold. On Amazon, YelpChi, and TFinance, using five identical stratified splits and 14 same-protocol methods, PREF-Gate obtains mean AUPRC values of 0.9085, 0.8104, and 0.8913. It selects the label-free expert on all Amazon and YelpChi splits and an evidence mixture on all TFinance splits. Thus, the main result is conditional rather than universal: label-derived relational evidence is useful only where held-out validation supports it. The framework couples competitive ranking performance with an explicit label-provenance contract, finite selection policy, failure accounting, and review-budget evaluation, providing an auditable knowledge-based decision pipeline for graph fraud detection.
[AI-51] What We Talk About When We Talk About LLM Planning Planning : Evidence for Two Distinct Planning Abilities
链接: https://arxiv.org/abs/2607.11197
作者: Sukai Huang,Chenyuan Zhang,Fucai Ke,Zhixi Cai,Naim Rastgoo,Gholamreza Haffari,Hamid Rezatofighi
类目: Artificial Intelligence (cs.AI)
备注: 19 pages. Keywords: Reasoning, Automated Planning, Item Responses Theory, LLMs as Planner Research Area: NLP and Symbolic Reasoning Research Area Keywords: neurosymbolic, planning in agents, symbolic reasoning Contribution Types: Model analysis interpretability
Abstract:When LLMs exhibit uneven performance across planning tasks, these gaps are often attributed to task difficulty. We argue that this explanation is incomplete, as task-level variation may reflect distinct latent planning competencies rather than differences along a single ability spectrum. We study this question on ACPBench-Hard by evaluating multiple LLM families under varying test-time reasoning budgets and applying a multidimensional item response theory model to uncover the latent competency structure underlying LLM planning. The analysis reveals two principal dimensions that shape planning performance: operational reasoning, the ability to evaluate local action applicability and immediate state transitions, and structural enumeration, the ability to reason about goal reachability and landmark structure. Operational reasoning improving under model scaling and longer reasoning traces, while structural enumeration remains comparatively insensitive. Our findings motivate competency-level evaluation of LLM planning, shifting the focus from whether models improve overall to which planning competencies improve, under what conditions, and why.
[AI-52] RepTran: Search-Based Repair of Transformer Models
链接: https://arxiv.org/abs/2607.11193
作者: Yuta Ishimoto,Paolo Arcaini,Fuyuki Ishikawa,Masanari Kondo,Naoyasu Ubayashi,Yasutaka Kamei
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: 12 pages. Accepted to the Research Track of ISSRE 2026
Abstract:To ensure the overall quality of AI-enabled software, not only traditional software components but also AI components need to be tested and repaired. Among AI components, Transformer models are increasingly integrated into software systems, which makes their misbehaviors critical. Although prior work in the software engineering community has proposed deep neural network (DNN) repair methods, most overlook Transformer-specific structures. We propose RepTran, a search-based repair method for Transformer models. It targets their feed-forward networks (FFNs), which play a central role in the architecture. RepTran identifies suspicious weights by combining two types of scores: a variance-based neuron score and an existing bidirectional score. It then iteratively optimizes these weights using differential evolution. Our evaluation includes 18 fault benchmarks constructed from CIFAR-100 and Tiny-ImageNet. We compare RepTran against three baselines: random weight selection, Arachne (a state-of-the-art DNN repair method), and ArachneW, which enables Arachne to control the number of selected weights. RepTran achieved an average repair rate of 74.7%, statistically outperforming random selection and Arachne across all benchmarks. Effect size analysis revealed that RepTran achieved higher repair rates than ArachneW regardless of the number of selected weights. These results suggest that RepTran is effective for enhancing the reliability of AI-enabled software.
[AI-53] SCALECUA: Scaling Computer Use Agents with Verifiable Task Synthesis and Efficient Online RL
链接: https://arxiv.org/abs/2607.11185
作者: Bowen Lv,Xiao Liu,Yanyu Ren,Hanyu Lai,Bohao Jing,Hanchen Zhang,Yanxiao Zhao,Shuntian Yao,Jie Tang,Yuxiao Dong
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Computer use agents (CUAs) are emerging as a powerful interface for automating complex digital workflows through visual perception and GUI execution. Online reinforcement learning with verifiable rewards (RLVR) has emerged as a key direction for scaling their capabilities. However, this paradigm is bottlenecked by verifiable data scarcity and online RL inefficiency. To break these barriers, we introduce ScaleCUA, a unified framework that scales online RL for CUAs via verifiable task synthesis and efficient training. At the data level, we design VeriGen, an end-to-end framework for generating verifiable RL tasks through iterative docker interactions and a multi-agent feedback loop. Scaled to 100+ concurrent agent workers via a shared docker interaction probe, this pipeline produces 24K+ verifiable tasks and nearly 3K high-quality RL tasks. To maximize sample efficiency, we propose Frontier Sampling, which tracks per-task capability and allocates rollouts to the current learning frontier. On the training side, we further design Visual Context Segmentation, a sliding window over recent visual context that balances rollout and training-engine pressure, yielding a 2.83x training speedup over step-wise decomposition. Together, ScaleCUA achieves 68.7% on OSWorld and 54.0% on ScienceBoard, establishing new state-of-the-art performance among open-source computer use agents. Code, models, and datasets are available at this https URL.
[AI-54] he Path to Self-Evolving Clinical Systems: Scaling Medical Agents from Assistance to Autonomy
链接: https://arxiv.org/abs/2607.11175
作者: Chunzheng Zhu,Lei Tian,Bohan Tan,Ziqi Zhou,Yuxuan Sun,Yijun Wang,Chengchao Lv,Yilin Wen,Yijun He,Jinghao Lin,Yihang Chen,Cheewei Tan,Qianshan Wei,Lei Zhao,Bin Pu,Kenli Li,Yuan Xue,Jianxin Lin
类目: Artificial Intelligence (cs.AI)
备注: Project page: this https URL
Abstract:The growing ability of large language models and vision language models to jointly interpret and reason over images and text is reshaping medical agents, moving them from task specific predictors toward autonomous systems that perceive, reason, plan, remember, and act in clinical environments. This work departs from the capability first perspective of existing literature and instead begins from clinical deployment, asking what tasks, contamination resistant benchmarks, and interactive training environments are required before medical agents can be trusted in practice. Medical agents are formalized as sequential decision making systems under partial observability, together with a three level autonomy taxonomy spanning assisted, cooperative, and fully autonomous operation. The field is organized along a unified scaling spine consisting of framework scaling, capability scaling, and environment scaling. Within this framework, clinical environment scaling, the integration of tools, data, and clinical gyms, is identified as the most actionable yet underexplored direction for agents operating in PACS, EHR, and FHIR ecosystems. Clinical self evolution, where agents improve through interaction with their environments rather than parameter scaling alone, is further positioned as a key research frontier, drawing insights from self improving agents, agent gyms, and test time compute scaling. Applications across radiology, pathology, ophthalmology, and hospital workflows are examined together with deployment challenges including hallucination, cascading failures, and fairness. By consolidating more than 300 references, with particular emphasis on advances from 2025 to 2026, this work provides a roadmap toward trustworthy, self improving medical imaging systems for real clinical practice.
[AI-55] STAMP: Provenance-Guided Credit Assignment for Deep Search Agents
链接: https://arxiv.org/abs/2607.11172
作者: Ke Xu,Han Xu,Xinran Chen,Yuqian Wang,Zhixuan Li,Xiaojian Liu,Changwo Wu,Jianqiang Xia,Yuchen Li
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Reinforcement learning for deep-search agents has largely focused on trajectory-level scoring – outcome correctness, citation-aware rewards, and evidence coverage. Yet the actions that expose supporting documents receive no targeted credit, a gap we call the reward-credit mismatch. We propose STAMP, in which a reference-based verifier judges whether each cited document supports an entity or relation in a training-time evidence graph, and first-exposure attribution traces each supported citation back to the action that first surfaced it. This step credit is injected through sign-preserving advantage modulation, which redistributes advantage across steps without changing the trajectory-level reward or the relative ranking of trajectories within each group. On BrowseComp, BrowseComp-ZH, and xbench-DS, STAMP improves the GRPO baseline by +2.0/+5.5/+3.0 points under matched SFT initialization, training data, and search tools, and composes with both outcome-only and citation-rubric base rewards. Component ablations confirm that the provenance-based credit signal and the sign-preserving advantage modulation each contribute to the gains.
[AI-56] Pix2Act: Image-Space Manipulation Policies with Equivariant Augmentation
链接: https://arxiv.org/abs/2607.11167
作者: Haojie Huang,Linfeng Zhao,Haotian Liu,Zhang Ye,Si-Yuan Huang,Mingxi Jia,Boce Hu,Fangzhou Lin,Yu Qi,Dian Wang,Robin Walters,Robert Platt
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Project Website: this https URL
Abstract:Representing manipulation actions as 2D trajectories in the camera plane provides a compact and interpretable basis for learning complex 3D manipulation policies. However, it also creates challenges from out-of-frame trajectories and limited precision. We propose Pix2Act, an imitation learning method that addresses these challenges by generating continuous image-space keypoint trajectories in each camera plane and losslessly recovering end-effector poses via triangulation. This reformulates high-dimensional 3D control as a simpler, more learnable 2D prediction problem. Crucially, it aligns observations and actions in the same coordinate space, enabling equivariant transformations to jointly rotate individual camera images together with their image-space actions. We analyze the symmetry properties of this augmentation and design a network architecture that can fuse multiple camera views while respecting their per-view rotations. As a result, Pix2Act implicitly enlarges the support of the data distribution and learns invariant action structures across transformations, yielding improved generalization and overall performance. Across diverse simulated and real-world manipulation tasks, Pix2Act outperforms state-of-the-art baselines and remains robust under camera perturbations.
[AI-57] AMT-X: Phase-Structured Multi-Turn Red-Teaming with Checklist-Gated Evaluation
链接: https://arxiv.org/abs/2607.11151
作者: Yi Ting Shen,Kentaroh Toyoda,Alex Leung
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: A reference implementation is available at this https URL
Abstract:Safety evaluation of large language models (LLMs) relies largely on single-turn attack datasets and single-judge scoring, underestimating risk from adaptive multi-turn adversaries and reporting a single success rate that does not separate partially actionable outputs from those carrying complete operational detail. We propose AMT-X (Adaptive Multi-Turn Exploitation), a phase-structured multi-turn red-teaming framework. Unlike prior multi-turn attacks that rely on ad hoc escalation or free-form per-goal plans, AMT-X casts the attack as an explicit, reproducible multi-phase state machine driven by semantic signals from the victim, and replaces single-judge scoring with a multi-role jury whose phase-conditioned checklists gate success on actionable harm. Across six frontier victim models (queried under their default safety alignment, without added moderation layers) and seven Moderation sub-categories, AMT-X attains overall attack success rates of 97.6-100% under a lenient score threshold, but 66.7-78.6% under a stricter gate requiring complete, real, and operational detail: a gap of up to 33 percentage points between partially and fully actionable harm.
[AI-58] he Hidden Footprint: Making Storag e a First-Class Metric for LLM Agent Evaluation
链接: https://arxiv.org/abs/2607.11149
作者: Chenglin Yu,Hongquan Gui,Ying Yu,Hongxia Yang,Ming Li
类目: Artificial Intelligence (cs.AI)
备注: 17 pages, 5 figures; includes supplementary material
Abstract:LLM agent benchmarks measure task completion, reliability, and inference cost, but not the persistent data an agent run leaves on disk, including logs, context snapshots, checkpoints, and debug traces. We introduce AgentFootprint, a cross-framework benchmark of post-run agent storage footprint. Its serialization-aware metric suite measures total retention, channel composition, duplication, growth, compressibility, and conversation-history reconstructability. It addresses a measurement trap: naive byte-level measurement understates duplication by an order of magnitude because database paging and JSON escaping obscure repeated content. A fixed-trace control separates agent-generated logical volume from persistence-layer amplification: replaying the same trajectory through seven persisting frameworks yields a 6.7x spread. Under identical models, tools, and tasks, configurations with 100% accuracy differ by 15.7x in retained bytes, although their defaults support different recovery and audit capabilities. Three full-history configurations grow superlinearly on a repeated-observation stress task. Exported trajectories from 108 instance-normalized SWE-bench Verified submissions span three orders of magnitude per instance, with no detectable correlation with resolve rate. A content-addressed store reduces retention by 4.8x-32.7x while preserving every reconstructability score. These results establish persistent storage as a resource metric to report jointly with accuracy and reconstructability.
[AI-59] NextFund: A Unified Performance Tracking Platform for Agent ic Portfolio Management
链接: https://arxiv.org/abs/2607.11141
作者: Changlun Li,Peixian Ma,Qiqi Duan,Zhenyu Lin,Peineng Wu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models (LLMs) based agents are beginning to participate in portfolio construction and market analysis, where decisions must be justified under evolving information and risk constraints. Current assessment practice, however, remains poorly aligned with this setting: many studies rely on static examinations or report only terminal portfolio returns, while the intermediate evidence, analyst judgments, and execution steps that produced those returns stay largely invisible. We introduce NextFund, an evaluation platform that makes financial-agent behavior observable under live market conditions. The platform couples time-consistent market access, coordinated multi-agent analysis, and persistent logging of the full decision path from observation to trade. Through an interactive Trading Arena, users can compare models across markets, inspect equity curves, and drill from leaderboard outcomes down to individual justifications. We present NextFund on Hong Kong, U.S., and China A-share equities, illustrating how inspectable decision histories enable fairer benchmarking and more actionable diagnosis. Our demo is available at this https URL.
[AI-60] A Formal Hierarchical Architecture for Agent ic Orchestration with Stack-Based Execution and Lazy Discovery
链接: https://arxiv.org/abs/2607.11138
作者: Prashant Devadiga,Abhishek,Adithya Mishra,Alok Singh,Amisha Sinha,Asit Desai,Gaurang Dahad,Harshit Bhushan,Mandati Pramod Reddy,Prakhar Gupta,Rupesh Patil,Siddhi Behere
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:The rapid expansion of capabilities in Large Language Model (LLM) agents has exposed a critical architectural bottleneck: when agents are given access to a flat, monolithic registry of tools, the model must evaluate hundreds or thousands of options simultaneously. This leads to decision-space explosion, context window saturation, and degraded routing accuracy. To address these limitations, this paper presents a hierarchical, skill-based architecture for agentic orchestration. Capabilities are organized as a rooted tree where internal nodes make routing decisions and leaf nodes execute deterministic tasks. The runtime enforces a single-step execution loop governed by a Last-In-First-Out (LIFO) stack, giving the agent a form of memory akin to a Pushdown Automaton, therefore enabling it to track nested execution contexts and resume deterministically from any depth. Capability discovery follows a manifest-driven, lazy-loading protocol: only the immediate children of the active node are loaded, so memory and prompt costs scale with the explored path rather than the global registry. By replacing global memory with localized stack frames, the architecture prevents outputs from one execution branch from leaking into another, establishing the isolation guarantees required for deployment in regulated enterprise environments. We also discuss UPI Help, an AI-powered digital payments support product, as a motivating production deployment context. We provide a mathematical formalization of the orchestration state, detailed algorithmic analysis of the execution loop, and controlled benchmarks comparing flat and hierarchical routing under increasing tool catalogs, multi-step workflow pressure, and visible schema-token exposure per LLM call.
[AI-61] BeatEdit: Symbolic Music Generation as Explicit Editing
链接: https://arxiv.org/abs/2607.11124
作者: Haoyu Gu,Lekai Qian,Haowu Zhou,Qi Liu,Shuai Wang
类目: ound (cs.SD); Artificial Intelligence (cs.AI)
备注:
Abstract:Music creation is fundamentally a process of revision. Yet symbolic music generation remains dominated by paradigms that produce complete sequences from scratch, with limited support for selective modification. Edit-based methods have proven effective for text transformation tasks, but remain largely unexplored for symbolic music. We trace this absence to the representational level: conventional event-based music encodings lack the structural properties required by explicit music editing. In contrast, the BEAT encoding, a beat-grid-anchored representation originally designed for autoregressive generation, possesses structural properties amenable to editing. We propose BeatEdit, the first framework for symbolic music generation based on explicit edit operations, recasting generation as producing new content by editing a draft rather than synthesizing from scratch. BeatEdit comprises three complementary mechanisms along an axis of increasing edit density: per-token sequence tagging for error correction, iterative refinement for accompaniment editing, and tag-then-fill for segment completion. All these mechanisms share a single encoding and pre-trained backbone, achieving higher precision and perceptual quality than autoregressive and diffusion methods across all three tasks, while remaining efficient, with single-pass inference completing in under 100 ms. Cross-encoding evaluation further reveals that encoding design substantially influences editing effectiveness, with notable encoding-method interaction effects. Code is available at this https URL
[AI-62] VIA: Visual Interface Agent for Robot Control
链接: https://arxiv.org/abs/2607.11119
作者: Hengyuan Hu,Priya Sundaresan,Jensen Gao,Dorsa Sadigh
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Robot manipulation is a complex task that requires visual understanding, physical reasoning, planning, and closed-loop control. General-purpose foundation models (FMs) have grown remarkably capable of some of these, especially vision and reasoning. To leverage this for generalist robot policies, current methods typically involve converting existing FMs into vision-language-action (VLA) models by fine-tuning on robot data to output low-level actions. However, VLAs are often orders of magnitude smaller than frontier FMs given the limited data and compute available for fine-tuning, which in turn limits their general capability. Inspired by the growing ability of FMs to operate software through visual interfaces, we ask whether that same competence suffices to control a robot. We present VIA (Visual Interface Agent for robot control), a framework that recasts robot control as an agentic task: an off-the-shelf FM-powered agent drives a manipulator through a browser-based 3D interface by taking screenshots, issuing intuitive commands, observing the outcome, and adjusting. The agent receives no robot-specific fine-tuning and no access to privileged state information: it perceives visual input and acts through a small set of general tools. VIA inherits the agent’s general reasoning, closed-loop error recovery, and ability to plan and re-plan from what it observes. It solves a diverse suite of tabletop manipulation tasks zero-shot with both Claude Code and Codex. With the strongest model (Fable 5) it achieves 96.7% success on three LIBERO-Goal tasks and 100% on a long-horizon rainbow assembly task. Performance improves with the scale and strength of the underlying model. These results suggest that frontier agents already possess skills that transfer directly to robot control given the right interface: your coding or computer-use agent is, in a sense, secretly a robot-control agent.
[AI-63] MusicMark: A Robust Generative Watermarking Framework for Music Generation
链接: https://arxiv.org/abs/2607.11117
作者: Seohwan Yun,Jeeyoung Yun,Yongjin Kim,Juyeon Lee,Sungwoong Kim
类目: ound (cs.SD); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注: Submitted to IEEE Transactions on Information Forensics and Security
Abstract:AI music generation has rapidly advanced alongside commercial platforms, raising the need for reliable watermarking for provenance and attribution. However, existing audio watermarking research has largely focused on speech, and applying speech-oriented methods to music is challenging due to music’s complex structure and rich acoustic texture. Most existing methods are post-hoc, adding imperceptible perturbations after generation rather than embedding watermarks as part of the content. This makes them fragile under transformations and especially vulnerable to neural codec re-synthesis, which can discard imperceptible residual signals. Moreover, since generation and watermarking are decoupled, the watermarking step can be bypassed or omitted, weakening provenance guarantees. To address these issues, we propose MusicMark, which, to the best of our knowledge, is the first generative watermarking framework for music. Specifically, MusicMark embeds watermark messages into the semantic latent space during generation, incorporating the watermark as part of the musical content and ensuring robustness against diverse attacks, particularly neural codec re-synthesis. To this end, we introduce a watermark adapter into a diffusion-based generation model to embed watermark messages across denoising steps. The adapter and detector are trained with a joint objective that preserves fidelity by constraining watermarked latents close to their unwatermarked reference latents, while improving robustness through attack augmentations. Experiments demonstrate that MusicMark substantially outperforms post-hoc baselines across diverse attacks including neural codec re-synthesis, while maintaining comparable generation quality. We further introduce a cover-song attack, converting the singing voice while preserving musical content, and show that MusicMark remains more robust than post-hoc methods.
[AI-64] he Equilibrium Is the Initialization: Lazy Identity Collapse in Physics-Structured Deep Equilibrium Reasoning
链接: https://arxiv.org/abs/2607.11116
作者: Joyjeet Singh
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 14 pages, 5 figures. Code, raw logs, and analysis scripts: this https URL
Abstract:Deep equilibrium models promise input-adaptive implicit computation: harder problems should demand more solver iterations, and the solved equilibrium should encode the result of genuine iterative inference. We report a cautionary study of a port-Hamiltonian DEQ with a learned initialization on two reasoning tasks – ProofWriter entailment over frozen DeBERTa embeddings and a BFS-verified graph-reachability benchmark – in which the implicit computation is a silent no-op. Across tasks, seeds, and controlled ablation arms, the solved equilibrium equals the solver’s start point to numerical precision, and bypassing the solver entirely changes test accuracy by +0.00 percentage points in 18 of 19 training runs. Controlled interventions falsify the tempting explanation: removing the anchoring term reproduces every result, and retraining with noise-decoupled starts yields a solver that converges to the noisy start while the decoder learns to ignore it. The single escaping run diverges instead ( |h^*-z_0|=171 ), producing a co-adapted noise channel whose removal improves accuracy. Iteration counts are uncorrelated with ground-truth difficulty ( r=0.009 ), and the full apparatus never outperforms a two-layer MLP on either task. We trace the mechanism to gradient starvation along two distinct routes, show that the standard zeroing ablation is confounded and gives wildly seed-dependent answers where the correct substitution test gives a stable zero, and distill a four-test diagnostic protocol for auditing claimed implicit computation. All experiments run on a single free Colab GPU; code, raw logs, and analysis scripts are released.
[AI-65] OS-Pruner: Pruning Chains-of-Thought of Reasoning Models via Optimal Stopping
链接: https://arxiv.org/abs/2607.11089
作者: Mohammed Ehab,Aymane El Gadarri,Vivek F. Farias,Adam Jozefiak,Ciamac C. Moallemi
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large Language Models (LLMs) have achieved remarkable success in complex reasoning tasks through Chain-of-Thought (CoT) prompting. However, these models often exhibit “computational overthinking,” generating redundant reasoning steps that increase latency and cost without improving accuracy. Recent studies suggest that CoT trajectories can be significantly pruned, yet existing methods often rely on forcing a static thinking budget, heuristic filtering, sub-optimal early exit via classification, or expensive re-training. In this paper, we introduce OS-Pruner, a lightweight plug-in framework that formulates chain-of-thought pruning as an optimal stopping problem. Given a reasoning prefix, OS-Pruner learns whether further reasoning is worth its token cost by optimizing an explicit utility that trades off final-answer accuracy against generated length. Our novel formulation enables the model to dynamically assess the sufficient point of termination for a reasoning chain. OS-Pruner is designed to be lightweight during both training and inference, and to provide users with fine-grained control over the reasoning-effort vs. accuracy trade-off. On diverse reasoning benchmarks and base models, OS-Pruner achieves 20-60% reduction in generation length with minimal accuracy sacrifice.
[AI-66] NVAITC AI Scientist: A Governed End-to-End Research System – A Hypertension GWAS Case Study
链接: https://arxiv.org/abs/2607.11084
作者: Eddie Huang(1),Ken Liao(1),Iven Fu(1),Yang-Hsien Lin(1),Chao-Shun Zhan(1),Andy Liao(1),Virginia Chen(1),Johnson Sun(1),Pika Wang(1),Richard Huang(1),Jiun-Cheng Jiang(1),Ting-Yuan Liu(2 and 3),Hsing-Fang Lu(2, 3, and 4),Ray Y. Lee(5),Chi-Chou Liao(2),Simon See(1),Fuu-Jen Tsai(2, 6, 7, and 8) ((1) NVIDIA AI Technology Center, NVIDIA Corporation, (2) Department of Medical Research, China Medical University Hospital, Taichung, Taiwan, (3) Master Program for Digital Health Innovation, China Medical University, Taichung, Taiwan, (4) Laboratory for Statistical and Translational Genetics, RIKEN Center for Integrative Medical Sciences, Yokohama, Japan, (5) AI-Driven Genomic Medicine and Drug Discovery Lab, China Medical University Hospital, Taichung, Taiwan, (6) School of Chinese Medicine, China Medical University, Taichung, Taiwan, (7) Division of Pediatric Genetics, Children’s Hospital of China Medical University, Taichung, Taiwan, (8) Department of Medical Laboratory Science and Biotechnology, Asia University, Taichung, Taiwan)
类目: Artificial Intelligence (cs.AI)
备注: 22 pages, 6 figures, 4 tables
Abstract:Agentic research systems are emerging as a new paradigm for coordinating scientific workflows beyond isolated model inference, code generation, or statistical analysis. However, deployment in institutional biomedical environments requires governed mechanisms for research planning, data access, workflow orchestration, evidence tracking, reproducibility, and human oversight. We present NVAITC AI Scientist (NAIS), a governed end-to-end agentic research system designed to support domain-general scientific workflows while keeping protected data within institutional privacy boundaries. NAIS integrates proposal review, execution planning, governed computational routing, reproducible workflow orchestration, evidence generation, and scientist-in-the-loop oversight. We validate NAIS in a real-world hypertension genome-wide association study (GWAS) using hospital-linked genotype and electronic health record (EHR) data from 286,422 individuals under an aggregate-only data policy. The agent planned cohort extraction, orchestrated GWAS execution, generated quality-control summaries, and drafted publication-oriented outputs. Human-AI review identified phenotype discrepancies and enabled iterative refinement of the hypertension definition. After reconciliation, the agent-orchestrated GWAS reproduced established hypertension loci, including FGF5, ATP2B1, CNNM2, FTO, and GRB14, with the strongest signal at FGF5 reaching -\log_10§ \sim 70 . As a secondary demonstration, NAIS also supported a drug-induced liver injury prediction workflow, achieving a multimodal graph neural network AUC of 0.842. These results demonstrate that governed agentic research systems can support scalable AI-assisted biomedical discovery while producing outputs comparable to expert-led workflows.
[AI-67] Are LLM s Ready for Scientific Discovery? A Capability-Oriented Benchmark for AI Scientists
链接: https://arxiv.org/abs/2607.11079
作者: Chuhan Shi,Xiaoquan Ren,Sicheng Song,Haobo Li,Rui Sheng,Yushi Sun
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Existing benchmarks for scientific data analysis evaluate LLMs primarily on code execution or workflow completion, overlooking that scientific analysis serves to support distinct types of scientific claims: hypothesis exploration, statistical inference, mechanistic explanation, each with different assumptions and validity criteria. We introduce SDABench, a benchmark that reorganizes evaluation around six capabilities (descriptive, exploratory, inferential, predictive, causal, and mechanistic) across five domains (Biology, Chemistry, Environment, Geography, Physics). SDABench comprises 527 real-data instances (SDA-Real) and 6000 synthetic instances (SDA-Synth), each in both multiple-choice and open-ended formats, constructed through an automated pipeline. Evaluating 15 representative LLMs, we find that models handle descriptive analysis well but degrade sharply on tasks requiring assumption selection, latent-process modeling, or mechanistic reasoning. SDABench further provides a five-stage error analysis framework that locates where LLMs fail: more advanced models more reliably identify the relevant scope and variables, but still struggle to select appropriate analytical procedures, model variable relationships, and draw valid conclusions.
[AI-68] AdvNav: Behavior-Guided Black-Box Adversarial Attacks on Vision-Language Navigation
链接: https://arxiv.org/abs/2607.11063
作者: Chenyang Li,Kaige Li,Zeyu Jiang,Changhao Chen
类目: Artificial Intelligence (cs.AI)
备注: ACM International Conference on Multimedia 2026
Abstract:Despite progress in Embodied AI, Vision-and-Language Navigation systems remain vulnerable to adversarial visual disturbances. Most existing methods rely on white-box access to target model gradients, which is often unrealistic for real-world deployed systems and computationally exhaustive due to recursive backpropagation for optimization, limiting their applicability. While previous black-box methods predominantly target single-step, instantaneous decision tasks, they struggle to handle the task complexities and temporal dependencies. This highlights the need for a gradient-free attack method that can effectively disrupt the multistep sequential perception-action loop using only observable inputs and outputs. Therefore, we propose AdvNav, a behavior-guided black-box adversarial attack framework that disturbs an agent’s first-person views during navigation. To construct an informative surrogate objective for effective optimization guidance in gradient-free search under the black-box setting, we design a dual-granularity behavior-based feedback, aggregating a trajectory-level performance score representing overall navigation degradation, an action-level reward score considering the potential decision risk, and a deviation indicator, all of which are extracted from the agent’s self-output behaviors. This feedback guides a hybrid optimization strategy that heuristically tunes perturbation strength via adaptive updates and evolves noise spatial structure genetically, to iteratively discover the most disruptive noise configuration. Evaluated against Transformer-based HAMT and LLM-based MapGPT with two types of backbones on R2R dataset, AdvNav achieves 49.70/65.96/87.30% Attack Success Rate. The result demonstrates the effectiveness and generality of AdvNav, reveals critical perception vulnerabilities and offers insights for the design of future resilient VLN models.
[AI-69] BackendForge: Benchmarking Agent ic End-to-End Code Generation with Backend Services
链接: https://arxiv.org/abs/2607.11042
作者: Yuzhe Guo,Mengzhou Wu,Yuan Cao,Jialei Wei,Dezhi Ran,Wei Yang,Tao Xie
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models (LLMs) are increasingly used in agentic coding settings, where they can inspect files, execute commands, run tests, observe failures, and iteratively revise code. This shift raises a central evaluation question: can an agentic LLM generate an end-to-end software artifact that is both deployable and behaviorally correct under execution? Backend services provide a controlled but realistic substrate for this evaluation. Their APIs expose application-level executable semantics, and deployed behavior can be checked deterministically against an OpenAPI contract through black-box HTTP interactions. We introduce BackendForge, a benchmark of 56 contract-defined backend generation tasks rewritten from real open-source applications. Given a visible specification and an OpenAPI contract, an LLM must generate a Dockerized service that is built, deployed, and evaluated only through HTTP tests. To strengthen evaluation without introducing hidden requirements, BackendForge uses a test agent and a code agent to co-evolve the test oracle and reference service, where the test agent proposes specification-grounded backend tests and the code agent repairs the reference implementation. Although the best-performing model, GPT-5.5, succeeds on 55.4% of tasks under the base oracle, it succeeds on only 28.6% under the final oracle. This gap suggests that current LLMs can implement many local API behaviors, but still struggle to produce complete backend services.
[AI-70] Qwen Paw-Data: Bridging Facts Methodology and Execution for Autonomous Enterprise Data Analytics
链接: https://arxiv.org/abs/2607.11019
作者: Tianjing Zeng,Yuntao Hong,Zhongjun Ding,Dandan Liu,Yinan Mei,Yunxiang Su,Yiming Wang,Xiaojian Zhang,Jingyu Zhu,Junhao Zhu,Zhuowen Liang,Jiazhen Peng,Lianggui Weng,Zhihao Ding,Kerui Yi,Qifeng Wang,Rong Zhu,Bolin Ding,Liyu Mou,Jingren Zhou
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Enterprise data analysis is emerging as a distinct frontier for autonomous agents. Compared with general-purpose interaction and software engineering, it operates in an open, ambiguous, and continuously evolving environment. These characteristics call for a data-agent architecture that treats semantics, methodology, execution, and evolution as first-class system concerns. To this end, we introduce QwenPaw-Data, an agentic data system designed for enterprise intelligent data analysis. QwenPaw-Data consolidates heterogeneous assets from warehouses, dashboards, documents, interaction logs, and historical tasks into reusable, governable, and evolvable analysis assets, then turns natural-language requests into end-to-end analytical workflows spanning data understanding, retrieval, analysis, report generation, and decision support. Its architecture decomposes the problem into three collaborative subsystems: DataBridge provides trustworthy semantic grounding through interconnected metadata, knowledge, and trace graphs; Skill-Hub codifies expert analytical methodology into reusable and verifiable skills; and Host materializes these evidence and method assets into controllable, artifact-centric runtime execution. Across these subsystems, semantics, methods, traces, and feedback are continuously deposited back into the system, forming a self-evolving asset flywheel. Experiments on public benchmarks and real-world industrial BI workloads show that QwenPaw-Data improves both verifiable data access capability and higher-level analytical quality, offering a practical foundation for reliable, traceable, and continuously improving enterprise data agents.
[AI-71] Affordance-Based Manipulation Planning with Text Goals and Sim-to-Real Generalisation via Real-to-Sim Image Conversion
链接: https://arxiv.org/abs/2607.11004
作者: Solvi Arnold,Rin Karashima,Tadashi Adachi,Takafumi Mochizuki,Kimitoshi Yamazaki
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: 14 pages, 10 figures
Abstract:We present a manipulation planning system based on affordance recognition and action effect prediction. The system reasons through possible futures in visual form, and evaluates candidate plans by agreement of predicted outcomes with text-based goals set at run-time, using a multi-modal goal-matching module. Positions of objects named in the goal text are tracked through predictions even when occluded, making it possible to generate action plans even when objects become occluded, or when their initial descriptors cease to identify them in future states. We further expand the system with an image conversion module for translating real-world state images with objects of varied shapes and visual appearances into a consistent visual appearance, to facilitate manipulation planning in a physical robot setup. We evaluate performance of the system’s modules in isolation and demonstrate the integrated system’s manipulation planning capabilities on a set of challenging tasks in both simulation and on hardware.
[AI-72] From Checker to Forecaster: Code-Owned Evaluation of Model-Generated Strategic Routes Under Delayed Ground Truth
链接: https://arxiv.org/abs/2607.10972
作者: Aleh Manchuliantsau
类目: Artificial Intelligence (cs.AI)
备注: 11 pages, 2 figures
Abstract:Many evaluations of model outputs rely either on contracts checkable at evaluation time or on feedback that arrives within the operating loop. We study the complementary setting in which ground truth is delayed, censored, or private, so deterministic code cannot check correctness at scoring time and must instead issue a code-owned provisional forecast. RouteCast instantiates this regime for model-generated typed strategic routes: models propose candidate routes and structured factors; point-in-time evidence, reference classes, and deterministic transformations produce a provisional forecast-ranking; later outcomes evaluate the forecast. In a retrospective venture pilot on 21 binary-outcome cases (6 positive, 15 negative), the whole-packet RouteCast score showed preliminary retrospective discrimination (AUC 0.756, 95% CI [0.471,0.980]), while a blind LLM judge reached AUC 0.678 [0.419,0.897] and an identity-exposed LLM judge reached AUC 0.761 [0.515,0.944], consistent with recognition- or outcome-related leakage risk. A preregistered decomposition ablation on the same binary subset found that converting the identical inputs into typed staged routes was indistinguishable from the whole-packet score (Delta AUC = -0.144, 95% CI [-0.471,0.176]) and from a deterministic heuristic (Delta AUC = -0.089, 95% CI [-0.412,0.278]). The pilot establishes an auditable feasibility result and exposes failure modes; it does not establish prospective calibration, causal decision improvement, route-decomposition advantage, or cross-domain validity.
[AI-73] CGS: Configurable Graph Summarization with Bounded Neighborhood Loss and Query Support
链接: https://arxiv.org/abs/2607.10969
作者: Shubhadip Mitra,Sona Elza Simon,C Oswald,Arnab Bhattacharya,Arindam Pal
类目: Data Structures and Algorithms (cs.DS); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Given a large graph, how to generate a compact summary graph that is configurable by the user and supports multiple graph queries with either no loss or with high accuracy? The ever growing size of graph datasets makes the above question on graph summarization very pertinent. Although, there are several approaches, there does not exist a configurable graph summarization method that offers high compression along with support for multiple graph queries on the summary graph with high accuracy, and allows the user to configure the summarization based on: (1) lossless or lossy summarization, (2) amount of tolerable neighborhood loss, (3) the type of loss it can tolerate, in terms of false positive edges (i.e., extra edges), false negative edges (i.e., missing edges), or neither, in both the (a) reconstructed graph and the (b) query answers. To overcome these limitations, we propose a novel graph summarization framework CGS (Configurable Graph Summarizer) that builds upon the idea of aggregating nodes with common neighborhoods. The CGS framework consists of three summarization variants, CGS-E, CGS-I and CGS-U. While CGS-E is a lossless scheme, CGS-I and CGS-U are lossy schemes that allow reconstruction of the input graph with no false positive edges and no false negative edges, respectively. To bound the graph reconstruction loss, we introduce a user-specified parameter neighborhood loss tolerance threshold, that limits the maximum loss allowed in the neighborhood of each node. This allows graph reconstruction and neighborhood query evaluation with either no loss or with bounded loss guarantees. Empirical evaluation on several synthetic and real-world graphs shows that CGS offers superior summarization than the state-of-the-art methods, and can answer graph queries with fairly high accuracy and efficiency.
[AI-74] SVR-R1: Bootstrapping Multi-modal Reasoning with Self-verification in Reinforcement Learning
链接: https://arxiv.org/abs/2607.10966
作者: Mingyuan Wu,Jingcheng Yang,Shengyi Qian,Xudong Wang,Jize Jiang,Qifan Wang,Aashu Singh,Khoi Pham,Fei Liu,Zhaolun Su,Zhuokai Zhao,Klara Nahrstedt,Jianyu Wang,Hanchao Yu
类目: Artificial Intelligence (cs.AI)
备注: Under submission
Abstract:We introduce Self-Verified Reasoner (SVR-R1), a multi-turn RL framework that turns a model’s own verification into a learning signal for multimodal reasoning. For each query, the model proposes an answer using the same weights, and issues a binary self-verdict (Yes/No). A ‘No’ triggers a second-chance rethink; a ‘Yes,’ or a turn cap, finalizes the output for computing the outcome-based reward. SVR-R1 is implemented with GRPO and an asynchronous multi-turn rollout framework and needs no external supervision or auxiliary critics. We evaluate SVR-R1 on vision-language reasoning benchmarks and show that it improves accuracy by a large margin over strong standard GRPO baselines. Training dynamics show decreasing reliance on verification-fewer verification turns, yet higher test accuracy-indicating that the gap between verification and generation narrows as the policy internalizes self-correction and chooses the most confident answer via our framework. SVR-R1 bridges the less explored intersection of inference-time self-refinement and RL training for VLMs, offering a simple yet effective recipe for bootstrapping multimodal reasoning. We will open-source \textbfSVR-R1 to facilitate future research in VLMs.
[AI-75] Efficient Online Proportional Sampling with Applications to Smoothed Online Learning
链接: https://arxiv.org/abs/2607.10963
作者: Amirmahdi Mirfakhar,Maria-Florina Balcan,Hedyeh Beyhaghi
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computational Geometry (cs.CG); Computer Science and Game Theory (cs.GT); Theoretical Economics (econ.TH)
备注:
Abstract:We study the problem of efficient online proportional sampling from a high-dimensional domain under a \sigma -smoothed adversary, where the sampling distribution is induced by a dynamically evolving weight function defined over a sequence of piecewise-structured partitions. This setting captures a broad range of applications, including principal-agent games (e.g., pricing and contract design), and algorithm configuration and parameter tuning. The central challenge is maintaining an efficient data structure as the induced partition grows increasingly complex over time – naively, the number of subregions can grow as O(t^d) by round t in d dimensions. We design a data structure that supports efficient updates and proportional sampling while avoiding the cost of explicitly maintaining this exponential growth, where the discontinuities are structured from axis-parallel hyperplanes. Under a \sigma -smoothed adaptive adversary, we prove a tight O(\sqrt\sigma T) bound on the depth of our data structure, and an O(\log T) bound under a random-order adversary – to our knowledge, the first such results for this class of problems. We apply this framework to online learning with piecewise-structured rewards, obtaining efficient no-regret algorithms under both full-information and bandit feedback, with provable sublinear regret guarantees.
[AI-76] Edge Physical AI Deployment of Vision Transformers on Heterogeneous Edge GPU Targeting Autonomous Vehicles
链接: https://arxiv.org/abs/2607.10942
作者: Ashiyana Abdul Majeed,Mahmoud Meribout,Neethu Joseph,Abel Kidane Haile,Mohammad Abdullah Al Faruque
类目: Hardware Architecture (cs.AR); Artificial Intelligence (cs.AI)
备注: 14 pages, 15 figures, This work has been submitted to IEEE for possible publication
Abstract:Physical AI systems, such as autonomous vehicles and intelligent machines, require transformer-based perception models that satisfy stringent edge latency and energy constraints. However, heterogeneous edge-GPU deployment remains limited by underutilized hardware engines and accelerator-incompatible operators, causing fragmented execution and lower throughput per watt. This paper presents Heterogeneous Frame Dispatch Scheduling (H-FraDS), a hardware-aware frame scheduling methodology for transformer inference on a recent NVIDIA edge GPU. H-FraDS routes frames across the GPU and dual deep learning accelerator (DLA) cores using fixed dispatch ratios to improve utilization under latency and power constraints. To enable scheduling, incompatible transformer components are adapted for DLA execution by reshaping tensors, approximating error function (ERF) with tanh, and replacing layer normalization with bounded tanh. The adapted model maintains a 92% F1 score, with only a 2% reduction from the original. Optical flow accelerator (OFA) is further used for inference-side optical-flow estimation. To the best of the authors’ knowledge, prior work has not addressed these combined issues. Using Swin Transformer for autonomous-driving perception, H-FraDS Balanced Dispatch (1:2) achieves 125.93 FPS, a 2.36x speedup over standalone adapted-DLA execution, 4.0 FPS/W, and approximately 24 ms DLA latency, satisfying 30 FPS real-time operation; the GPU-DLA-OFA case achieves a 2.02x DLA throughput speedup.
[AI-77] he Singularity Space: A Generative Diffusion Framework for Signal Representation
链接: https://arxiv.org/abs/2607.10930
作者: Eli Bar-Yosef,Amir Averbuch,Eli Turkel
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Signal Processing (eess.SP); Numerical Analysis (math.NA)
备注:
Abstract:Generative models often represent signals as dense grids of amplitudes, blurring sharp transients that are crucial for the correctness of physical signals. We introduce Singularity Space, a generative framework that represents signals through complex-plane singularities, rooted in the classical pole-residue representation of meromorphic functions. We learn a latent space of physically constrained, per-signal singularity configurations to solve an inverse problem from degraded or partial observations. The framework has three key properties: interpretability, in which each generated singularity configuration corresponds to a set of physical parameters; structural stability, which mitigates Gibbs artifacts at discontinuities; and resolution-free output reconstruction on arbitrary grids without retraining or interpolation. Our framework employs a transformer-based diffusion model that directly predicts samples at complex-plane singularity coordinates, subject to geometric constraints during sampling. As a controlled test case for sharp-feature recovery, we evaluate our framework on 1D Burgers shocks, where each shock is represented by 32 predicted singularities (an 8\times reduction versus a 1024-point grid signal). Our framework preserves signal structure ( \textTV ratio \approx 1 ) under unseen test-time observation noise, achieves a 4.2\times lower reconstruction error in zero-shot sub-resolution generalization than a grid-based baseline, and recovers physical parameters to 10^-4 absolute error in-distribution. These results suggest that singularity-based representations may provide a practical foundation for other transient-dominated signals such as speech and biomedical signals, with potential extension to higher-dimensional domains.
[AI-78] Learning Linear Temporal Specifications from Demonstrations with Uncertainty
链接: https://arxiv.org/abs/2607.10918
作者: Parastou Fahim,Constantino Lagoa,Rômulo Meira-G’oes
类目: Artificial Intelligence (cs.AI); Systems and Control (eess.SY)
备注: This paper has been accepted by the ACC2026 (American Control Conference)
Abstract:Learning temporal logic specifications from system demonstrations is essential for tasks such as formal verification and controller synthesis, especially in safety-critical domains. Existing approaches typically assume demonstrations are correct or only affected by misclassification errors. In practice, however, system traces are often uncertain or incomplete due to sensor faults, measurement errors, or data loss. We present a framework for learning minimal Linear Temporal Logic (LTL) formulas from demonstrations with uncertainty. Our approach models uncertainty via Hamming distance to generate possible estimates around each observed trace, which are grouped with constraints requiring that at least one trace per group is consistent with the learned formula. Our problem is then reduced to an equivalent Pseudo-Boolean Optimization. We evaluate our method against state-of-the-art LTL learning approaches and show that it recovers specifications that more closely align with ground-truth formulas under uncertainty.
[AI-79] Incremental Transformer for Surrogate-Based Inverse Design of Geopolymer Mixtures
链接: https://arxiv.org/abs/2607.10896
作者: Giansalvo Cirrincione,Filippo Grassia
类目: Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
备注:
Abstract:Small-data inverse design is challenging in engineering informatics when observations are heterogeneous, mixed-type, and constrained by physical relations among design variables. This work proposes a topology-aware surrogate framework guided by an Incremental Transformer (INCRT) for physics-constrained inverse design, applied to geopolymer mixture design. The method integrates intrinsic-dimensionality analysis, mixed-variable design-space representation, tabular surrogate prediction, INCRT-based manifold rationalisation, and constrained inverse optimisation. Using a public benchmark of fly-ash and slag-based geopolymer concrete mixtures with compressive-strength and carbon-emission targets, the high-dimensional design space proves strongly redundant, organising around fewer effective mixture regimes. Compressive strength requires nonlinear tabular surrogates, while carbon emission is largely determined by composition and well recovered by regularised linear models. INCRT thus acts not as a replacement for tabular predictors but as a rationalisation layer providing prototype regimes and a manifold-support score for inverse design. Three strategies are compared: unconstrained surrogate optimisation, physics-constrained optimisation, and topology-aware physics-constrained optimisation. Unconstrained optimisation can match target strength but may yield physically invalid or off-manifold candidates; physics-only constraints do not always ensure data support. The topology-aware strategy yields candidates balancing target compliance, carbon reduction, physical admissibility, and proximity to the learned feasible manifold. The framework aims not to replace experimental validation but to support screening of credible candidate mixtures from small, mixed, physically constrained engineering datasets.
[AI-80] SETA: Scaling Environments for Terminal Agents
链接: https://arxiv.org/abs/2607.10891
作者: Qijia Shen,Zhiqi Huang,Vamsidhar Kamanuru,Aznaur Aliev,Jay Rainton,Ahmed Awelkair,Zhichen Zeng,Jiajun Li,Shi Dong,Yueming Yuan,Boyuan Ma,Qizheng Zhang,Jiwei Fu,Yuzhen Mao,Wendong Fan,Ping Nie,Philip Torr,Bernard Ghanem,Changran Hu,Jonathan Lingjie Li,Urmish Thakker,Guohao Li
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models (LLMs) are rapidly shifting toward agents that solve tasks through diverse interfaces, including web and graphical user interfaces (GUIs). Among these, the terminal command line provides a text-based, general-purpose interface, covering tasks from system operations to data science and machine learning. However, scaling terminal-agent training remains challenging, as it requires diverse and coherent task instructions, executable environments, and reliable verification, while lacking naturally grounded supervision data. In this work, we propose SETA, a scalable framework for generating verifiable terminal environments for reinforcement learning (RL). The framework consists of two pipelines sharing a unified verification mechanism: SETA-Synth converts diverse sources into standardized RL environments, and SETA-Evol further expands from existing environments with adaptive control of difficulty and diversity. Together, we construct and release SETA-Env, the largest open-source verifiable terminal RL dataset to date, containing over 4,500 environments. We evaluate our dataset by training Qwen3-8B with GRPO on SETA-Env, achieving 12% pass rate on Terminal-Bench 2.0, the best reported result for an RL-trained model at the 8B scale. We further observe gains on DeepSeek-V4-Flash under the same terminal agent harness, with pass@1 on Terminal-Bench 2.0 improving from 40% to 43% and pass@5 improving from 54% to 58%. These results demonstrate that SETA- Env provides high-quality training environments for terminal agents and serves as a valuable resource for advancing research on terminal-based agent learning.
[AI-81] First-Order Modal Logic in HOL: Deep and Shallow Embeddings with Automated Faithfulness (Extended Preprint)
链接: https://arxiv.org/abs/2607.10880
作者: Christoph Benzmüller,Daniel Kirchner
类目: Artificial Intelligence (cs.AI); Logic (math.LO)
备注: 21 pages. Extended version, with a source-code appendix, of a paper accepted at ARQNL 2026 (International Workshop on Automated Reasoning in Quantified Non-Classical Logics). The full Isabelle/HOL development is included as ancillary files
Abstract:We extend, in Isabelle/HOL, the deep-and-shallow embedding methodology of our prior work from propositional to first-order modal logic (FML) with constant-domain Kripke semantics. Three embeddings of FML into classical higher-order logic (HOL) are provided side by side: a deep embedding, a heavyweight maximal-shallow embedding, and a lightweight minimal-shallow embedding. The minimal-shallow embedding is presented as an Isabelle/HOL locale, parametrised by an accessibility relation, a world-indexed interpretation, a universe of worlds, and a variable assignment; the locale form admits a global faithfulness theorem, stating that quantifying over all minimal-shallow interpretations recovers exactly deep validity. A central technical contribution is a mechanisation, for FML under constant-domain Kripke semantics, of the (countable) downward Löwenheim-Skolem theorem, which underpins the automation of our faithfulness proof between the deep and minimal-shallow embeddings. Deploying it inside an extension of the minimal-shallow locale resolves the surjectivity problem that arises against an uncountable domain of individuals – where the locale’s variable assignment, having countable domain V = nat, cannot be surjective onto the domain – and thereby yields faithfulness over the full domain. Since prior work treats only the propositional fragment, we develop here the substitution machinery (free/bound-variable predicates, the fresh-variable function, capture-avoiding substitution, alphabetic renaming, the substitutability predicate, the substitution lemma, and size-based induction principles) needed for the first-order quantifiers. Comments: 21 pages. Extended version, with a source-code appendix, of a paper accepted at ARQNL 2026 (International Workshop on Automated Reasoning in Quantified Non-Classical Logics). The full Isabelle/HOL development is included as ancillary files Subjects: Artificial Intelligence (cs.AI); Logic (math.LO) MSC classes: 03B45 (primary), 03B70, 03B15, 68V15, 03B35, 03C07, 68V20 ACMclasses: F.4.1; I.2.3 Cite as: arXiv:2607.10880 [cs.AI] (or arXiv:2607.10880v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2607.10880 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-82] Diachronic Sample Integration: Robust Tail-Risk Estimation with Generative Models
链接: https://arxiv.org/abs/2607.10810
作者: Shuning Zhao,Patrick Wong,Leran Zhang,Xiaolin Hu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Risk Management (q-fin.RM)
备注:
Abstract:Deep generative models are increasingly used as simulators for downstream decision-making under data scarcity, but in risk-sensitive applications their usefulness depends on rare adverse scenarios rather than typical samples. Standard generative objectives prioritize bulk distributional fidelity, leaving low-probability tails vulnerable to localized optimization noise and making tail-dependent functionals unstable under finite simulation budgets. We introduce Diachronic Sample Integration (DSI), a test-time inference framework that ensembles generated samples across checkpoints from a stochastic training trajectory. DSI targets a checkpoint-mixture distribution that averages checkpoint-specific tail fluctuations rather than relying on a single brittle endpoint. We formalize this mechanism through a finite-budget bias-variance theory. Empirically, across multivariate synthetic processes and high-frequency trading data, DSI substantially reduces tail-estimation error compared to single-checkpoint baselines under fixed simulation budgets, outperforming standard diffusion and state-of-the-art tail-aware baselines without modifying the generative objective.
[AI-83] Weight-Adjusted Gradients Reveal Parameter Importance and Failure Modes in LLM s
链接: https://arxiv.org/abs/2607.10803
作者: Shrestha Datta,Hongfu Liu,Anshuman Chhabra
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Understanding which parameters are influential in Large Language Models (LLMs) is central to improving their efficiency, reliability, and interpretability. We introduce Weight-Adjusted Gradients (WAG), a simple yet effective approach for estimating parameter importance that explicitly captures the interaction between model weights and first-order gradient information and identifies parameters that disproportionately influence model behavior, such as those responsible for collapse phenomena in LLMs. Across a range of models and settings, we show that WAG surfaces a tiny but critical subset of parameters whose modification leads to dramatic degradation in performance, a failure mode that existing importance metrics overlook. These findings reveal a previously underexplored interplay between weights and gradients, suggesting that parameter importance cannot be fully understood through either signal alone. The surprising effectiveness of WAG points to fundamental structural properties of trained networks and motivates new open questions about the role of zeroth-order and first-order information in deep learning. We demonstrate the practical utility of WAG across multiple applications, including expert allocation in mixture-of-expert architectures, parameter-specific unlearning, mixed-precision quantization, and layer selection for knowledge editing. Our results position WAG as a unified approach for analyzing, debugging, and controlling LLMs, and opens new directions for principled model-level interpretation.
[AI-84] Imaging-101: Benchmarking LLM Coding Agents on Scientific Computational Imaging
链接: https://arxiv.org/abs/2607.10789
作者: Siyi Chen,Jiahe Ying,Yixuan Jia,Yuxuan Gu,Enze Ye,Weimin Bai,Zhijun Zeng,Shaochi Ren,Binhong Gao,Yubing Li,Tianhan Zhang,He Sun
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Computational imaging, which recovers hidden signals from indirect, noisy measurements, underpins quantitative discovery across scientific disciplines, yet building a correct reconstruction pipeline demands deep domain expertise and remains laborious even for domain scientists. We introduce Imaging-101, a benchmark of 57 expert-verified computational imaging tasks spanning six scientific domains, each grounded in a peer-reviewed paper and canonicalized into a standardized four-stage pipeline (preprocessing, forward physics modeling, inverse solver, and visualization) Three evaluation tracks (planning, function-level unit tests, and end-to-end reconstruction) probe distinct agent capabilities across the full pipeline. Evaluating seven frontier LLMs uncovers systematic challenges in applying coding agents to computational imaging that go beyond those exposed by general coding benchmarks, spanning algorithm selection, physical convention handling, and pipeline integration. These findings highlight concrete capability gaps and point toward skill-augmented, domain-specialized agents as a practical path to reliable computational imaging assistance.
[AI-85] LSTrans: Efficient Knowledge Transfer for Lightweight and Automated ECG Classification
链接: https://arxiv.org/abs/2607.10784
作者: Yi Zhao,Jiajun Gao,Chenyang Xu,Yuxi Zhou,Hao Wang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Submitted to BIBM 2026
Abstract:Deploying deep learning models for automated electrocardiogram classification on resource-constrained wearable devices remains challenging due to high computational costs. To address this, we propose LSTrans, a lightweight hybrid model designed for efficient and sensitive ECG analysis. LSTrans introduces a specialized 1D convolutional backbone with an interleaved layer architecture to capture both macroscopic rhythmic trends and microscopic morphological variations. This backbone is cascaded with a Transformer encoder to model long-range temporal dependencies, incorporating Low-Rank Adaptation across critical layers to compress the model and reduce the trainable parameter space. We further employ homogeneous and heterogeneous knowledge distillation to transfer diagnostic expertise from high-capacity teacher models to the student. Experimental results on multiple benchmark datasets demonstrate that LSTrans achieves a competitive balance between diagnostic sensitivity and resource efficiency, substantially reducing peak memory footprints and training latency during downstream adaptation. The source code is available for review at this https URL.
[AI-86] Lightning Fast Matching Dependency Discovery with Desbordante
链接: https://arxiv.org/abs/2607.10771
作者: Alexey Shlyonskikh,Michael Sinelnikov,Daniil Nikolaev,Yurii Litvinov,George Chernishev
类目: Databases (cs.DB); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Performance (cs.PF)
备注:
Abstract:Matching dependency is a generalization of the functional dependency concept, which allows users to apply custom similarity functions for matching individual attributes. Matching dependencies have a wide range of applications for solving various data quality problems, such as entity resolution, data deduplication, data integration, schema matching, and many more. However, their discovery is a very computationally intensive problem, which limits their practical application. In this paper, we describe a number of optimization techniques for HyMD - currently the state-of-the-art algorithm for the discovery of matching dependencies. These optimizations belong to both technical and scientific domains. The most important of them are: 1) a new sampling technique, 2) a faster generalization lookup technique, and 3) an improved representation of a dependency. The first one aims to raise the efficiency of inference from record pairs, while the last two are designed to speed up lattice-related operations. To evaluate our optimizations, we implemented our version of HyMD in Desbordante, an open-source high-performance data profiler. Experiments demonstrated that they allow for a speedup of more than 40x over the state-of-the-art implementation on average, reaching a speedup greater than 170x in some cases. Finally, the improved version of HyMD is ready to use by anyone. It comes with bidirectional Python integration, which allows calling the C++ algorithm implementation from Python programs while allowing users to supply their custom matching functions. Subjects: Databases (cs.DB); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Performance (cs.PF) ACMclasses: H.3; I.5; J.0 Cite as: arXiv:2607.10771 [cs.DB] (or arXiv:2607.10771v1 [cs.DB] for this version) https://doi.org/10.48550/arXiv.2607.10771 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Journalreference: 2024 36th Conference of Open Innovations Association (FRUCT), Lappeenranta, Finland, 2024, pp. 729-740 Related DOI: https://doi.org/10.23919/FRUCT64283.2024.10749955 Focus to learn more DOI(s) linking to related resources Submission history From: George Chernishev [view email] [v1] Sun, 12 Jul 2026 13:59:54 UTC (296 KB)
[AI-87] Opti-Agent -Bench: Benchmarking End-to-End Optimization RD Agents on Real-World Business Problems
链接: https://arxiv.org/abs/2607.10768
作者: Yongchang Fu,Xinjie Huang,Chengjun Dai,Chengzhe Feng,Junshao Zhang,Hong Zhu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:LLM-based agents are increasingly deployed to solve optimization problems, yet existing benchmarks evaluate them on pre-structured mathematical formulations that bypass the most critical challenge: translating complex business requirements into correct models and solve efficiently. We introduce Opti-Agent-Bench, an end-to-end benchmark that evaluates Large Language Models (LLMs) across the complete optimization RD pipeline, from understanding business-language descriptions through mathematical modeling, algorithm selection, and code implementation, to solution report generation. Our design rests on three pillars: (1) businesssemantic authenticity with anti-template traps that defeat pattern matching; (2) modular evaluation with cross-module consistency checking across Problem Understanding, Formal Modeling, Implementation, and Reporting; and (3) the ORAC bi-level validity framework that simultaneously ensures task quality and scoring integrity. Across several industrialscale tasks spanning integer programming, robust optimization, stochastic programming, and non-convex optimization, we expose critical failure modes of current models, including constraint omission, model-code inconsistency, and report-implementation divergence, that remain invisible under conventional single-metric evaluation.
[AI-88] Filtering Harmful Actions Isnt Enough: Phantom Transfer in Agent ic SDF
链接: https://arxiv.org/abs/2607.10750
作者: Chinmayi Dixit
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Synthetic data is widely used to train large language models because it is inexpensive to generate and easy to control. As models are increasingly deployed as agents, synthetic trajectories are likely to become an important source of training data for agentic behavior. We investigate the effects of training on synthetic agentic trajectories containing adversarial interactions, including actions such as terminating another agents process, lowering its scheduling priority, or accessing resources without authorization. We finetune Llama 3.3 70B Instruct on these trajectories, generated to approximate reinforcement learning rollouts, and evaluate the resulting models on Anthropics Agentic Misalignment suite and Apollos in context scheming scenarios. Finetuning on these trajectories consistently increases misaligned behavior. Leaking rises by roughly a factor of five over the baseline, 4.6% to 24.9%. This increase survives the removal of every adversarial action from the trajectories. Finetuning on structurally comparable trajectories generated benign from the start produce a substantially smaller effect, 15.5%. These results indicate that the misaligned disposition is introduced during the generation process and encoded diffusely throughout the trajectory, rather than being localized to the harmful actions themselves. The effect also depends on the generating model. Benign trajectories produced by Gemini 2.5 Flash induce slightly higher leaking rates than trajectories generated from identical tasks by Claude 3.7 Sonnet. In contrast, broad safety benchmarks degrade similarly across all finetuned models and therefore fail to distinguish these effects. Our results suggest that action level filtering is insufficient to ensure the safety of synthetic agentic training data and that dispositions introduced by the generating model can survive semantic inspection.
[AI-89] Multi-Scale Convolution with Optimal Transport Attention Effect on Multivariate Time Series
链接: https://arxiv.org/abs/2607.10740
作者: HaoChong Fu,Jian Xu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:The analysis of Multivariate Time Series (MTS) plays an important role in a lot of real-world practical applications, but it still remains some challenging problem about capturing multi-granularity structural patterns and suppressing noise appropriately. Multi-Scale Convolution with Optimal Transport Attention (MSC-OT) is proposed in this paper. MSC-OT is a useful architecture to optimize the attention mechanism. It combines multi-scale convolution with Sinkhorn optimal transport method based on inverted embedding. The inverted embedding approach embeds each variable as a token and allows the model to capture cross-variate relationships better. MSC-OT consists of two part: (1) Multi-Scale Convolution Enhancement, that applies multi-scale convolutions to attention score matrices based on inverted embedding, capturing local structural patterns in the variate-interaction space induced by compressed temporal representations; (2) Sinkhorn Optimal Transport Regularization, that formulates attention computation as an optimal transport problem and employs iterative matrix scaling to ensure balanced information flow across variates. Adaptive Fusion Strategy utilizes softmax-normalized learnable weights to dynamically combine base attention, convolution-enhanced, and OT-regularized scores. Experiments on widely-used datasets, including ETT, Electricity, Traffic, Solar-Energy, and Exchange-Rate, show that MSC-OT achieves well performance in both short-term and long-term forecasting tasks. Ablation experiments further validate the effectiveness of each proposed component and their synergistic contributions to improving prediction accuracy for multivariate time series forecasting.
[AI-90] Distributed Denial of Science: How Indirect Data Poisoning of AI Systems Can Industrialize Scientific Fraud
链接: https://arxiv.org/abs/2607.10712
作者: Bálint Gyevnár,Atoosa Kasirzadeh,Nihar B. Shah
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Digital Libraries (cs.DL)
备注:
Abstract:Scientific fraud is the instrument of doubt that malicious entities can use to establish controversy in science. Historically, it required the resources of a company: deep pockets, ghostwritten articles, and corrupt academics. Today, Artificial Intelligence (AI) is increasingly automating scientific research, so we ask: Can a remote adversary weaponize the honest use of AI in science to compromise scientific integrity? We envision and empirically evaluate a new attack, indirect data poisoning, in which an adversary corrupts an open dataset and uploads the poisoned variant to a public repository. Autonomous research agents may independently retrieve and process this data, turning honest scientists into the unpaid and unwitting distributors of fraud at scale. Across five socially-salient topics, from hiring discrimination to the safety of autonomous vehicles, three widely used frontier AI systems (Claude Code with Claude Opus 4.7, Codex with GPT-5.5, Gemini CLI with Gemini 3.1 Pro), and 450 ethically contained experimental runs, we find that poisoning succeeds in 49.56% of runs, while the rate of poisoning detection is only 6.0%. The attack requires no topic-specific trigger-words, agent access, indirect prompt injection, or fabricated papers, only the open data ecosystem and misleading metadata. To mitigate the attacks, we propose and evaluate two measures: a scientist persona and a data provenance audit with five checks (referencing papers, social markers, statistical anomalies, related datasets, poisoning caution). We find that the persona still leaves 16.67% of runs with a poisoned conclusion, but provenance auditing reduces attack success rate to zero. Our results suggest that indirect data poisoning may enable scientific fraud at unprecedented scale, but these attacks can be mitigated with suitable auditing by agents during data retrieval.
[AI-91] PromptGraph: Graph-Guided Prompt Sanitization for Balancing Privacy and Utility in LLM Inference
链接: https://arxiv.org/abs/2607.10709
作者: Chen Gu,Hui Wan,Donghui Hu,Hui Wang,Zhuoer Gu
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:Large Language Model (LLM) services introduce a fundamental privacy challenge. Sensitive information may be inferred not only from explicit identifiers, such as names or phone numbers, but also from contextual associations among otherwise innocuous spans. Existing sanitizers typically assign privacy or utility signals to individual spans without explicitly modeling pairwise relationships among them. In this paper, we propose PromptGraph, a graph-guided prompt-sanitization approach for privacy-preserving LLM inference. PromptGraph estimates privacy leakage at the span level and utility-relevant contextual dependencies between pairs of spans. It represents each prompt as an attributed graph, in which nodes carry span-level privacy scores and edges encode contextual dependencies needed to preserve utility. The sanitization objective selects a protected span set that maximizes privacy gain while penalizing the loss of contextual dependencies. This formulation explicitly balances privacy and utility when contextual evidence is hidden. Protected spans are sanitized locally, and returned placeholders are restored only after passing local consistency checks. We conduct extensive experiments showing that PromptGraph achieves a more favorable balance between privacy and utility than prompt-privacy baselines.
[AI-92] Learning to Fine-tune Foundation Models under Resource Limitations ICML
链接: https://arxiv.org/abs/2607.10694
作者: Thomas Tsouparopoulos,Iordanis Koutsopoulos
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 6 pages, 2 figures, 1 table, accepted and presented at ICMLCN 2026
Abstract:We study the problem of optimal continual fine-tuning for a pre-trained Foundation Model deployed at a resource-limited device. At each time slot, a new batch of training data arrives, and the controller is faced with two options: either use the data to fine-tune the model and incur a compute cost, or do not fine-tune the model and discard the data. After the decision, the performance of the current model is measured in terms of an application-specific performance metric such as classification accuracy. Our objective is to learn an optimal policy that determines \emphwhen to fine-tune the model on a single task (e.g., sentiment analysis), under a finite compute budget. We formulate this online decision-making problem as a constrained Markov Decision Process, where the system state captures three essential aspects: (\textiti) model’s performance, (\textitii) computational budget, and (\textitiii) data distribution relevance to historic data encountered up to that point. The transition to the next state is stochastic and therefore, we propose a reinforcement learning-based method to solve this problem, namely the \emphactor-critic algorithm. We also consider the special case where the performance of fine-tuning for a given model can be predicted or estimated prior to decision; in this case the problem becomes a Dynamic Programming one. Experiments with a large pre-trained model on a widely-used text classification dataset demonstrate that our method consistently outperforms fine-tuning approaches with the same compute budget by more than 4% in terms of accuracy and achieves 97% of full-parameter fine-tuning accuracy while requiring only 25% of the fine-tuning steps.
[AI-93] Personalized Emotional Intelligence in Generative AI through Symbolic Affective Reasoning
链接: https://arxiv.org/abs/2607.10678
作者: Qing Lin,Mengmi Zhang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Emotional intelligence enables humans to recognize emotions, infer their causes, reason about interventions, and modify their environment to achieve desired affective states. Despite recent advances in artificial intelligence (AI), current models remain largely limited to generating realistic content or performing semantic reasoning, with little capacity for understanding, predicting, and personalizing human emotional responses. Here we introduce Emotion-augmented geneRatiOn System (EROS), a hybrid AI framework that integrates symbolic reasoning with deep learning to enable personalized emotion augmentation through visual content. Leveraging large-scale image-emotion datasets, EROS discovers generalizable affective rules, identifies emotion-relevant image regions, and predicts context-aware visual modifications that preserve scene semantics while steering emotional responses toward desired targets. To account for individual variability, EROS incorporates an expandable memory bank that supports inference-time personalization without model fine-tuning, yielding interpretable emotional profiles and rapid adaptation to new users. Across extensive human psychophysics experiments, EROS elicits target emotional responses more effectively than state-of-the-art large multimodal models while adapting to individual affective preferences. Beyond affective computing, EROS provides a foundation for AI systems that can understand, reason about, and augment human cognitive states, with potential applications in mental health, adaptive media, education, and human-computer interaction.
[AI-94] Commenting with Copilot: A Taxonomy and Multi-Year Analysis of Student Code-Generation Specifications
链接: https://arxiv.org/abs/2607.10674
作者: Nasser Giacaman,Valerio Terragni,Paul Denny,Viraj Kumar
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注: 7 pages, 2 figures, 2 tables. Accepted in the Proceedings of the 2nd ACM Virtual Global Computing Education Conference (SIGCSE Virtual 2026)
Abstract:As AI code tools become integrated into programming environments, students increasingly describe intended behavior in natural language and rely on these tools to generate code, shifting emphasis from code writing to specification. Yet little is known about the comments students write as specifications in AI-assisted programming tasks. We analyze a four-year dataset of undergraduate programming submissions and reflections from tasks in which students wrote comments to guide code generation and refined solutions using test-case feedback. We introduce a taxonomy spanning three dimensions: comment type, code expression level, and code construct. Using automated classification, we examine how these dimensions vary across attempts and how students describe the process in their reflections. Our findings show that students mostly wrote natural-language What comments, shifted toward How comments for more procedural constructs, and focused more on verifying generated code than on repeatedly rewriting comments.
[AI-95] Embark Now: User Demand Oriented Framework for Multi-day Urban Travel Itinerary Planning
链接: https://arxiv.org/abs/2607.10651
作者: Rongbo Qi,Yaqi Zhang,Shijun Yan,Xuemeng Liu,Xiangrui Cai,Chunyao Song
类目: Artificial Intelligence (cs.AI)
备注: 37 pages, 16 figures, 7 tables
Abstract:In large urban areas, planning multi-day travel itineraries is challenging due to the abundance of Points of Interest (POIs), diverse user preferences, and constraints such as opening hours. Effective solutions must dynamically accommodate diverse traveler requirements while optimizing for satisfaction and feasibility within limited computation time. This paper addresses these challenges through introducing an innovative framework that integrates Large Language Models (LLMs) to dynamically capture user requirements with precision and flexibility, and an enhanced Greedy Randomized Adaptive Search Procedure (GRASP) algorithm as a well-suited preference-aware planner to generate feasible multi-day itineraries. The effectiveness of our integrated approach is demonstrated through extensive experiments on two real-world urban datasets from Beijing and Tianjin. Our framework significantly outperforms state-of-the-art (SOTA) methods, improving the average total itinerary score by at least 4.52% and 11.09% across 5,040 user cases with diverse preferences in the two datasets. Furthermore, through end-to-end algorithmic enhancements, it achieves notable average improvements of 17.95% and 26.07% in the computed metrics, while also delivering substantial gains in time efficiency – realizing average performance increases of 4.64% and 25.55% within shorter computation times compared to suboptimal methods that require multiple iterations. These outcomes underscore our method’s superiority in delivering both enhanced itinerary quality and computational efficiency over existing methodologies.
[AI-96] Coverag e Path Planning : Classical Foundations Recent Advances and Future Directions
链接: https://arxiv.org/abs/2607.10649
作者: Zongyuan Shen,Shalabh Gupta,Shancheng Zhao,Dehua Zhou,Gao Wang,Zhongqiang Ren,Yaming Ou,Yikui Zhai,C. L. Philip Chen
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Coverage path planning (CPP) is a fundamental problem in robot motion planning, whose aim is to produce robot trajectories that provide complete coverage of target workspaces while minimizing task-specific objectives such as path length, overlap, number of turns, and energy consumption. CPP has widespread applications in cleaning, inspection, mapping, agriculture, manufacturing, surveillance, demining, and environmental monitoring. Although classical CPP has been extensively studied, recent advances have extended CPP beyond single-robot settings to multi-robot systems, complex 3D environments, constrained platforms, learning-based coverage planning, and visual coverage tasks. This paper presents a comprehensive survey of 125 representative works published primarily between 2015 and 2026, while presenting the evolution of recent developments in light of the classical CPP methods published before 2015. The CPP methods are organized into six main categories: single-robot CPP, multi-robot CPP, 3D CPP, constrained CPP, learning-based CPP, and visual CPP. For each category, the review summarizes the main planning formulations, representative algorithms, strengths, and limitations. In addition, the review analyzes how environmental knowledge, workspace geometry, robot constraints, sensing objectives, and coordination requirements shape the CPP problem. The survey further discusses open challenges in scalable online planning, multi-robot coordination, 3D and visual coverage, unified platform-constrained and resource-aware coverage, and learning-enhanced coverage. Thus, the survey provides a structured overview of recent CPP developments and future research directions.
[AI-97] Auditing Construct Overlap in Explainable Machine Learning: Evidence from Burnout-Depression Prediction Across Student Cohorts
链接: https://arxiv.org/abs/2607.10633
作者: Alireza Dehghan,Negin Ashrafi
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Explainable machine learning (XML) pipelines applied to composite mental health outcomes can produce apparently-robust, cross-population-stable risk hierarchies that are largely artefacts of how the outcome was constructed. We demonstrate this using an ElasticNet pipeline applied to 886 medical students at the University of Lausanne (primary cohort, 2022), validated across 2,580 longitudinal observations at three time points and 701 non-medical students from eight faculties; all three datasets share identical instruments. The pipeline produces a hierarchy in which trait anxiety and health satisfaction dominate wherever the outcome is measured, with Kendall \tau = 1.0 for the top-two positions across all five evaluation sets and consistent transfer performance ( R^2 : 0.41-0.49). Two residualization experiments, which isolate shared variance between correlated variables via regression, reveal the mechanism: when trait anxiety (STAI-T) is residualized against the co-included depression subscale (CES-D, r = 0.72 ), model R^2 drops from 0.41 to 0.16 and STAI-T falls from rank 1 to rank 6; when burnout subscales are residualized against CES-D, R^2 collapses to 0.016. Prediction intervals average 35.4 units on a 0-100 scale (2.4 outcome standard deviations), independently ruling out individual-level deployment. The residualization protocol is the paper’s transferable contribution: any XAI study combining correlated predictor and outcome constructs should apply this check before interpreting apparent stability as a finding.
[AI-98] World Models as Adversaries: Multi-Agent Self-Play Fine-Tuning for Robust Motion Planning
链接: https://arxiv.org/abs/2607.10630
作者: Tong Nie,Yuewen Mei,Junlin He,Yihong Tang,Jian Sun,Wei Ma
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Robust motion planning in dense traffic requires autonomous vehicles to interact in rare and safety-critical scenarios that are underrepresented in naturalistic driving data. Although adversarial training offers a feasible solution, existing methods often rely on external scenario generators, heuristic perturbations, or simulator-heavy rollouts, which makes them difficult to integrate with modern autoregressive planners. Here, we cast adversarially robust planner learning as a constrained min-max game and propose Adversarial World Modeling (AWM), a theoretically grounded multi-agent self-play fine-tuning framework. Since solving the exact game is intractable, AWM introduces a principled decoupled solver. In the inner minimization, the planner’s predictive world model is converted into a role-conditioned adversary that learns sparse, scene-adaptive attack coalitions via counterfactual credit assignment. In the outer maximization, the ego planner optimizes a regret-aware robust best response against the frozen AWM, utilizing tail-risk weighting and reference-anchored trust regions to improve hard-case recovery while preserving nominal driving behavior. Experiments on the nuPlan and InterPlan benchmarks demonstrate that our method generates transferable adversarial interactions and yields a robust planner that achieves competitive closed-loop performance in both nominal and highly interactive long-tail scenarios. Theoretical analysis justifies the decoupled solver and the main optimization components.
[AI-99] he Compliance Trap: Diagnosing How AI Agents Consume Conflicting Memory
链接: https://arxiv.org/abs/2607.10608
作者: Yixiong Chen,Xinyi Bai,Alan Yuille
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Memory is becoming a core component of long-horizon AI agents, allowing agents to reuse past experience when operating web browsers, software tools, and other interactive environments. Existing work mostly treats memory as a supply problem, asking what experience to write, how to store it, and which entry to retrieve for the next task. Yet we still lack a clear account of how models consume retrieved memory across a multi-step action trajectory. This consumption process matters because it determines not only what memories should be retrieved, but also what models and control policies are needed to use them safely. To diagnose this process, we propose Entry–Propagation–Recovery (E-P-R), a trajectory-level framework that asks where memory first changes an action, whether that change carries forward, and whether the agent can recover after leaving a correct path. We instantiate E-P-R on WebArena and on MemTrapBench, a controlled benchmark we build to isolate these phases. We find that the main failure often begins at entry: agents adopt conflicting memory at the first exposed decision point even when it is task-wrong. Repeated exposure then amplifies this early error, while recovery after divergence is weak. Together, these effects create a compliance trap: across models, conflicting memory induces similar compliance rates, but once agents comply, their success rates collapse to a low floor. Stronger agents therefore suffer larger absolute damage because each compliance event erases more baseline capability. These results suggest that memory-augmented agents should be evaluated not only by retrieval quality or final success rate, but by how they consume memory throughout the trajectory.
[AI-100] Agent ic-DPO: From Imitation to Agent ic Policy Optimization on Expert Trajectories
链接: https://arxiv.org/abs/2607.10601
作者: Yixiong Chen,Alan Yuille
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large Language Model (LLM) agents are commonly trained from expert trajectories using supervised fine-tuning (SFT), which treats multi-turn agent behavior as ordinary text imitation. This recipe is simple and low-cost, but it only learns to imitate the sequence of expert actions, rather than training the agent to choose the right action against plausible mistakes at each state. Existing methods to mitigate this problem include preference learning or reinforcement learning, but they usually need high-cost environment rollouts and reward models. We propose Agentic-DPO, a lightweight offline agent policy optimization method that turns expert trajectories into state-conditioned preference supervision. At each expert action state, Agentic-DPO samples a one-step action from the current state, treats plausible wrong actions as negatives, and contrasts them with the expert action using a DPO-style preference objective. To avoid mixing both policy and schema in preference learning, we introduce Policy-Preserving Augmentation (PPA), which renders the same latent trajectory under multiple schemas while keeping the expert policy fixed. Agentic-DPO requires no online environment rollout, reward model, or full-trajectory student exploration. We conduct experiments across StableToolBench, tau-bench retail, and Mind2Web, where Agentic-DPO consistently improves agents at different model scales beyond imitation. In particular, it raises tau-bench accuracy from 21.7% (SFT) to 41.4% for a 9B model, matching online GRPO under the same backbone with only step-level rollouts and without environment interaction during gradient steps. The results suggest that expert trajectories can support low-cost agentic policy optimization when converted from demonstrations into state-level action preferences. Code for Agentic-DPO is released at this https URL.
[AI-101] MRUF: Multi-granularity Routing with Uncertainty-Aware Fusion for Robust Multimodal Sentiment Analysis
链接: https://arxiv.org/abs/2607.10599
作者: Haoran Ma,Yinfeng Yu,Liejun Wang
类目: Artificial Intelligence (cs.AI); Signal Processing (eess.SP)
备注: Main paper (6 pages). Accepted for publication by IEEE International Conference on Systems and Man and and Cybernetics 2026 (IEEE SMC 2026)
Abstract:Multimodal sentiment analysis relies on language, visual, and acoustic cues, but utterance-level modality quality may vary due to occlusion, background noise, motion blur, or imperfect transcripts, causing conventional fusion to over-trust unreliable modalities. We propose MRUF, a reliability-aware fusion method that combines multi-granularity routing with uncertainty-aware calibration. MRUF summarizes sentiment-relevant representations, performs subspace- and modality-level routing, and supervises modality routing with leave-one-out error increases to estimate utterance-level modality importance. It further predicts modality-wise uncertainty and refines modality gates through inverse-variance reweighting, while modality-invariant contrastive alignment stabilizes the shared representation space. Experiments on CMU-MOSI and CMU-MOSEI under aligned and unaligned settings show consistent improvements over strong baselines, and mechanism analysis verifies that modalities with higher predicted uncertainty receive lower fusion weights.
[AI-102] MemDecay: Region-Aware KV Cache Eviction for Efficient LLM Agent Inference
链接: https://arxiv.org/abs/2607.10582
作者: Venkatesha Matam,Keon Kim
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language model (LLM) agents accumulate heterogeneous context, including system instructions, plans, user turns, retrieved documents, tool outputs, and intermediate reasoning, whose key-value (KV) cache can become a major memory bottleneck. Existing eviction policies generally apply the same attention- or recency-based rule to every token, ignoring semantic structure already available to the agent orchestrator. We introduce MemDecay, a training-free, region-aware KV-cache eviction policy. MemDecay assigns tokens region-specific base priorities and decay rates, refreshes retention scores when tokens receive attention, and evicts the lowest-scoring pages under a fixed cache budget while allowing critical regions to be pinned. We also provide a procedure for calibrating decay rates from measured attention lifetimes. We evaluate MemDecay at approximately 450 and 1,700 token contexts using Qwen2.5-1.5B and 3B. Across all settings, attention lifetimes differ by an order of magnitude across regions: system-token half-lives range from 148 to 189 decoding steps, compared with 14 to 16 for scratchpad tokens. Pinning preserves system-region facts at full-cache accuracy in every setting, while no baseline preserves more than 13 of 24. Region-aware retention remains effective as context grows, whereas recency-based retention collapses. Accumulated-attention retention performs better on unpinned content, however, and ablations identify attention-score normalization as the main limitation of the current formulation. These results establish semantic prompt structure as a robust signal for KV-cache management while clarifying how it should be combined with attention-based importance. Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI) Cite as: arXiv:2607.10582 [cs.LG] (or arXiv:2607.10582v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2607.10582 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-103] Laguerre Geometry for Interpreting Large Language Models
链接: https://arxiv.org/abs/2607.10578
作者: Chunwei Ma,Russell Wolfinger
类目: Artificial Intelligence (cs.AI)
备注: Geometric Lens; LLM Interpretability; Code: this https URL
Abstract:Existing hypotheses represent a concept in an LLM as a single point, a linear direction, or a Gaussian cluster, yet it remains unclear how and why such structures emerge. Here, we show that concept geometry can be precisely characterized via Laguerre Geometry, in which a concept is defined as a region–a Laguerre-Voronoi cell or a union of cells–allowing us to strictly define, measure, and separate concepts. Building on this formulation, we show that finer-grained concept structures, such as inclusion and hierarchy, are naturally revealed by the Laguerre weights. We then push this geometry inside the transformer. Decomposing each layer into piecewise-linear operators, we show that a token’s hidden trajectory is governed by two coupled mechanisms: a static tree of self-contained piecewise-linear flow, and a dynamic transport that hops the trajectory across trees when cross-token attention fires. This decomposition yields Geometric Lens, a training-free, hyperparameter-free method for reading out the exact concept a hidden vector encodes at any layer. We also develop Laguerre Autoencoder, a 2D visualizer that renders both the decision geometry and a model’s full reasoning trajectory in one view. Finally, we move beyond explanatory geometry toward actionable interpretability, showing that Geometric Lens recovers the correct factual token when a model is prompted with in-context interference. The code is available on GitHub.
[AI-104] Learning from Local Walks on Dynamic Graphs with Bandit Feedback
链接: https://arxiv.org/abs/2607.10571
作者: Sourav Chakraborty,Amit Kiran Rege,Claire Monteleoni,Lijun Chen
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
备注:
Abstract:We study stochastic multi-armed bandits on dynamic graphs, where arms correspond to the vertices of a network with time-varying edges. In this setting, the learner is restricted to local movement, selecting only its current node or an immediate neighbor at each round. This constraint decouples best-arm identification from exploitation: even after the optimal arm is identified, the learner may remain unable to reach it through the evolving topology. We identify a process-agnostic structural condition, based on sliding-window mixing, that ensures the graph’s intrinsic walk remains stable for both exploration and navigation. Under this regime, we analyze a family of local explore-then-commit algorithms and establish sublinear expected regret. Our framework includes a reward-aware strategy, for which we prove a worst-case safety theorem and a separate performance gain theorem.
[AI-105] When Does Restricting a Coding Agent to execute_code Help? A Regime times Agent -Design Ablation KDD2026
链接: https://arxiv.org/abs/2607.10569
作者: Hong Yang,Qi Yu,Travis Desell
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 9 pages (excluding references), 4 figures, 4 tables. Accepted to the Agentic Software Engineering (SE 3.0) Workshop at KDD 2026 (non-archival)
Abstract:Modern coding agents expose multiple tool surfaces – IDE primitives, bash, and Model Context Protocol (MCP) code-execution – and the field has shipped three contradictory claims about which one matters. We run the missing crossed comparison: an integrity-clean three-arm ablation (baseline / bash_only / code_only) on synthetic computation tasks and SWE-bench Mini modification tasks, holding model, harness, and prompts fixed, with two agents (Claude Code, OpenAI Codex CLI) so the comparison spans both regime and agent-design axes. Across the four resulting (regime, agent) cells, restricting the agent to a single execute_code MCP tool is cheaper than – or statistically tied with – its cheapest tool-rich rival in three cells (significantly on Artifact/Claude and SWE-bench/Codex; directionally on Artifact/Codex), with pass rates statistically tied within each cell. The lone exception is SWE-bench/Claude, where code_only is directionally costlier (+14.4%, not significant); a conditional-cost analysis localizes that gap to failure-cost on doomed-run trajectories, not a per-edit tax on successful runs. Two implications: the cheapest tool surface is jointly determined by task regime and agent design rather than by either axis alone, and the headline cost signal lives in cache-adjusted cost – not pass rate, which is invariant across surfaces at the model sizes we evaluate. The benchmark harness, task suite, and analysis code are available at this https URL.
[AI-106] CRiT-QA: Evaluating Multi-hop Reasoning with Counterfactual Chains and Distractor Traps LREC2026
链接: https://arxiv.org/abs/2607.10562
作者: JungMin Yun,JuneHyoung Kwon,YoungBin Kim
类目: Artificial Intelligence (cs.AI)
备注: Accepted to LREC 2026
Abstract:Evaluating the multi-hop reasoning capabilities of large language models remains a significant challenge. Although current models achieve strong results on existing multi-hop question answering datasets, such performance often masks two critical vulnerabilities: (1) reliance on internal parametric knowledge rather than adherence to the provided context, and (2) exploitation of dataset shortcuts, such as single-document cues or type-matching, that diminish the need for genuine evidence aggregation across multiple documents. We introduce CRiT-QA (Counterfactual Reasoning with Traps), a dataset explicitly designed to address both limitations. To neutralize reliance on memorized knowledge and enforce strict context dependency, CRiT-QA transforms factual reasoning chains with counterfactual entities. Furthermore, it injects multi-anchor distractor chains, plausible but incorrect reasoning paths that diverge at different hops. These traps require models to follow the entire reasoning process rather than exploiting shallow heuristics. Our experiments show that LLMs exhibit substantial performance degradation on CRiT-QA compared to standard datasets, exposing their vulnerability to counterfactual conditions and distractor traps. CRiT-QA thus serves as a rigorous diagnostic tool for evaluating genuine multi-hop reasoning and provides a foundation for developing more reliable, evidence-grounded LLMs.
[AI-107] Large language model agents accelerate inverse design of metal-organic frameworks for gas separation
链接: https://arxiv.org/abs/2607.10559
作者: Zhaolin Hu,Hehe Fan,Wangyihan Guo,Meng Xu,Chenhao Rao,Qiwei Yang,Yi Yang
类目: Artificial Intelligence (cs.AI)
备注: 19 pages,5 figures
Abstract:Metal-organic frameworks (MOFs) offer a highly modular platform for adsorptive gas separation, yet their vast reticular design space makes inverse design difficult under simultaneous constraints of chemical validity, separation performance, and structural diversity. Here, we present LEMO Agent, a large-language-model agent framework for closed-loop inverse design of gas-separation MOFs in MOFid space. LEMO Agent couples language-based candidate generation with MOFid standardization, explicit validity checking, Transformer-based property prediction, structured design memory, and multi-island exploration. Through iterative generate–validate–evaluate–remember cycles, the agent uses feedback from both successful and failed candidates to guide chemically constrained search across linker, metal, and topology choices. We evaluate LEMO Agent on CH _4 /N _2 and CO _2 /N _2 separation tasks. Compared with representative generative, optimization, and agentic baselines, LEMO Agent enriches high-performing candidates, improves predicted separation performance, and maintains broad chemical and topological diversity. Selected candidates are further reconstructed, evaluated by GCMC simulations, and passed through an experimental down-selection workflow based on chemical feasibility and ligand purchasability, leading to initial wet-lab synthesis and SEM characterization. These results demonstrate that large language model agents can serve as interpretable and scalable design engines for accelerating MOF discovery beyond conventional fixed-library screening.
[AI-108] AI YOU Town: Make Friends and Money with Your Digital Twin
链接: https://arxiv.org/abs/2607.10539
作者: Yan Lin,Yuyang Dai,Jiahui Geng,Yuxia Wang
类目: Artificial Intelligence (cs.AI)
备注: 28 pages
Abstract:Existing approaches to infer user traits and generate responses consistent with a persona rely on static prompting. They lack calibrated uncertainty, ignore sequential evidence, and drift during long interactions. We present \textbfAI YOU, a framework that continually updates a personality profile with 22 dimensions from conversation and embodies it in a personal digital twin. Practically, the system combines prompting, Bayesian updating, and conformal prediction for persona inference. A periodically refreshed memory anchor and cognitive memory with three layers preserve persona consistency over long interactions. Across the main results, AI YOU \emph(i) achieves conformal coverage ranging from 0.921 to 0.976, \emph(ii) improves uncertainty calibration and reasoning grounded in memory, and \emph(iii) enhances persona fidelity over static prompting in role playing over 100 turns while reducing trait drift, for most evaluated backbones under adversarial settings with multiple agents. The prototype \emphAI YOU Town initializes an imaginative twin world for future interaction. The online demo is available at \hrefthis https URL\mbox\textttthis http URL.
[AI-109] Cross-Layer Misalignment Detection in Agent Skills: A Progressive Loading-Aware Contrastive Learning Approach KDD2026
链接: https://arxiv.org/abs/2607.10534
作者: Chengjun Zhang,Yang Gao,Jianna Hur,Jingjing Zhang,Sagar Samtani
类目: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Machine Learning (cs.LG)
备注: 10 pages, 5 pages supplemental. Accepted at the KDD 2026 Workshop on Evaluation and Trustworthiness of Agentic AI
Abstract:Large language model (LLM) agents are increasingly extended through Agent Skills, reusable artifacts that package natural-language metadata, procedural instructions, and execution-time resources for runtime use. As open-source skill marketplaces expand, users and agents increasingly rely on brief metadata to select third-party skills, making it difficult to detect inconsistencies between a skill’s description and its true behavior, a problem we call cross-layer misalignment. To address this issue, we propose Progressive Loading-Aware Hierarchical Contrastive Learning (PL-HCL), an LLM-based framework that detects misalignment by modeling the layered structure of Agent Skills and learning cross-layer consistency. Using a normalized corpus of over 264,000 open-source skills and a human-verified challenge set, PL-HCL improves Macro-F1 from approximately 0.45 for unadapted baselines to 0.87-0.89 across evaluated LLM backbones. This approach offers an effective screening tool for users and operators, as well as design principles for detecting inconsistencies in layered digital artifacts.
[AI-110] Agents Dont Just Agree They Remember: Benchmarking Persistent Sycophancy in Stateful Personal Agents
链接: https://arxiv.org/abs/2607.10526
作者: Xutao Mao,Liangjie Zhao,Leyao Wang,Rui Qian,Qiang Huang,Wentao Wang,Bo Han,Xiang Zheng,Cong Wang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Stateful personal agents increasingly maintain long-term user profiles, episodic memories, and reusable skills. This persistence turns conversational sycophancy into a state-writing failure: accepted user-centric claims can be committed as lasting preferences, background facts, or workflows and later reused after the original conversation is gone. We call this persistent sycophancy and introduce the Personal Agent Sycophancy Benchmark (PASB), a 1,600-task benchmark that traces whether a conversational claim is accepted, written into durable agent state, and reused in a later neutral query. Unlike prior benchmarks that provide pre-written memories, PASB evaluates real agents (Hermes-Agent and OpenClaw) that decide what to store. It isolates the write process by combining four scenario framings with four temporal delivery patterns and separating a five-turn persist stage from a cleared three-turn query stage, ensuring downstream effects arise only from durable state. Across twelve models, the commit boundary is the key inflection point: downstream failure increases from 45.0% in session-only episodes to 71.9% after commitment, a consistent increase of 27.0 percentage points. Committed claims exhibit three write-time patterns: status promotion, attribution removal, and scope broadening. These patterns become stronger under memory-like or procedural framing, repeated reinforcement, and even across domain boundaries. These results show that agent sycophancy is fundamentally a state-writing governance problem. Once user content is committed to durable memory, safety must govern what agents write, not only what they say. PASB identifies the write-time controls needed to gate risky commits while preserving the source, role, and scope of stored content beyond response-level mitigations.
[AI-111] Conditional Optimal Bridge for Riemannian Activation Steering
链接: https://arxiv.org/abs/2607.10517
作者: Seyed Arshan Dalili,Ajay Narayanan Sridhar,Vijaykrishnan Narayanan,Mehrdad Mahdavi
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Activation steering offers a lightweight alternative to fine-tuning for controlling large language models at inference time. While many existing methods implicitly optimize a log-density-ratio objective between desired and undesired activation distributions, they do so heuristically rather than deriving it from a principled optimization problem. Moreover, these methods produce query-independent steering directions that can degrade performance on both in-distribution and out-of-distribution (OOD) inputs. We introduce \textscCobras (Conditional Optimal Bridge for Riemannian Activation Steering), which addresses both limitations by casting activation steering as a Schrödinger Bridge on the residual-stream hypersphere. This formulation yields, to our knowledge, the first principled derivation of the log-density-ratio steering objective from a well-posed optimization problem. Solving the bridge via entropic optimal transport and extracting the probability flow ODE recovers the widely used density-ratio gradient as a special case when the Sinkhorn potentials are uniform. Crucially, the Schrödinger potentials are evaluated at the current activation, making the resulting steering direction inherently query-adaptive. Empirically, across four models and three alignment axes (helpfulness, truthfulness, and detoxification), \textscCobras consistently outperforms prior activation steering baselines while avoiding the OOD degradation commonly observed in existing methods. The code can be found at this https URL.
[AI-112] Confining Nondeterminism: AI-Driven Research Systems as DBMSs for Reliable Non-Wasteful Transparent and Collaborative Research [Vision]
链接: https://arxiv.org/abs/2607.10508
作者: Kyoungmin Kim,Anastasia Ailamaki
类目: Databases (cs.DB); Artificial Intelligence (cs.AI)
备注:
Abstract:LLM agents that conduct research (proposing ideas, writing and running code, analyzing results) can already carry a study from research question to figures, yet cannot be fully trusted. The same question asked twice in a row returns different answers; the agent announces a number that no execution produced, and tool use does not prevent this, because nothing binds what the agent reports to what its tools returned; a small upstream change leaves downstream results silently stale, with no way to list which ones; and the agent re-runs preprocessing and rewrites code it has already produced. We argue these failures share one root: every step of today’s agent loop is a stochastic LLM call whose internal state nobody, including the agent, can check. Rather than trying to see inside the LLM, we take a lesson from databases, which earn trust without being watched, because deterministic operators over well-defined state make their guarantees hold by construction. We propose organizing a research project the same way. The project lives in a deterministic, versioned dataflow engine (in effect, a query plan over materialized views), and the LLM, together with the user, is a stochastic compiler that may only edit that plan. The executor never calls the LLM; LLM output enters only as versioned code and data that the executor then runs, and any asserted result enters the record only with an execution behind it. Five design rules at this boundary turn familiar database machinery, from versioning and provenance to incremental maintenance and cost-based scheduling, into guarantees that make research reliable, non-wasteful, transparent, and collaborative. This report presents the diagnosis, the requirements, and the design; the guarantee walkthrough, a prototype, and the research agenda appear in the full version, in preparation. The LLM, we argue, should be the query compiler, never the executor.
[AI-113] mporary Authority Permanent Effects: Commit-Time Authorization for LLM Agents
链接: https://arxiv.org/abs/2607.10487
作者: Igor Santos-Grueiro
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: 20 pages
Abstract:LLM agents can commit durable effects from authority evidence that was valid earlier in execution: a DOM snapshot, approval epoch, version witness, branch token, or worker result. We study the commit boundary at which earlier authority evidence no longer authorizes a durable effect. We call this property commit-time authorization: a durable effect is authorized only if the witness that licensed its derived state remains fresh, causally prior, bound to the same effect, and eligible at commit time. We build a controlled-invalidation suite spanning browser, tool/API, and multi-agent workflows. The suite preserves the user goal and payload shape while invalidating the authority relation before durability. In the primary 54-task matrix, endpoint success remains high: 262/270 runs reach the visible result. Only 55/270 are authorized completions; among the 216 invalidating rows, 207 commit after the authorizing path has failed. All 54 clean controls remain authorized, and a separate 54-run authority-preserving check produces no unauthorized commits. We then evaluate mitigation families. Prompt caution and single-condition checks are insufficient because different hazards break different boundary conditions. Defenses work when they refresh, rebind, replan, or refuse at the durability boundary. CommitGuard, a fail-closed boundary monitor, blocks stale durable-effect attempts on protected commit surfaces when runtimes emit witness, dependency, binding, and eligibility signals. The result is a reporting and runtime-design lesson: endpoint success is a utility metric; authorized commit is a security property. Comments: 20 pages Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI) Cite as: arXiv:2607.10487 [cs.CR] (or arXiv:2607.10487v1 [cs.CR] for this version) https://doi.org/10.48550/arXiv.2607.10487 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Igor Santos-Grueiro [view email] [v1] Sat, 11 Jul 2026 21:48:53 UTC (69 KB)
[AI-114] Reinforcement Learning with Verifiable Physics: Post-training LLM s with Continuous Rewards
链接: https://arxiv.org/abs/2607.10474
作者: Pengfei Cai,Utkarsh Utkarsh,Alan Edelman,Christopher Vincent Rackauckas,Rafael Gomez-Bombarelli
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE)
备注:
Abstract:Partial differential equations (PDEs) are foundational to modeling in science and engineering, but constructing reliable numerical solvers remains labor-intensive, demanding expert knowledge of discretization schemes, stability conditions, and boundary treatments. Recent work has begun to frame PDE solving as a code-generation task for large language models (LLMs), yet existing approaches operate primarily at inference time: relying on prompting, debugging, self-refinement, and test-time scaling rather than adapting the model itself. In parallel, reinforcement learning with verifiable rewards has emerged as a post-training paradigm for code and math reasoning, but its verifiers are typically binary: a compiler runs, or a test passes. Such signals discard the graded structure of scientific correctness, where two solvers may both execute and yet differ in solution accuracy by orders of magnitude. In this work, we introduce RLVP: Reinforcement Learning with Verifiable Physics, an RL post-training framework for multi-PDE solver code generation. RLVP addresses this verifiability gap with a hybrid verifier: hard program-validity checks ensure executability, while continuous physics rewards score function-space accuracy and PDE-residual consistency. A single policy is post-trained across diverse PDE families spanning hyperbolic, parabolic, elliptic, and incompressible-flow systems. RLVP improves over both pre-trained and supervised-only baselines on PDE benchmarks, and shows zero-shot improvement transfer to held-out PDEs. We show that a smaller LLM post-trained with RLVP can outperform prompting a frontier model on in-distribution PDE solver generation. The trained policy shows evidence of compositionality in numerical motifs: it recombines stencils, time-stepping schemes, and boundary-handling primitives learned from the PDEs used in training into generated solvers for unseen PDE problems.
[AI-115] Mitigating LLM Sycophancy in Code Smell Detection Using Evidence-Guided Reasoning Prompts
链接: https://arxiv.org/abs/2607.10411
作者: Istiaq Ahmed Fahad,Kamruzzaman Asif,Md. Nurul Ahad Tawhid
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: 11 Pages
Abstract:Large Language Models (LLMs) are increasingly used for code smell detection tasks due to their ability to interpret program semantics. However, their reliability in this context remains poorly explored, particularly under varying prompt conditions where model predictions may be influenced by external cues rather than code characteristics. One such limitation is sycophancy bias, where models tend to align their outputs with user-provided assumptions instead of performing objective analysis. In this paper, we present the first systematic empirical study of sycophancy bias in LLM-based code smell detection. Using the MLCQ dataset, we evaluate how different prompt framings like confirmation bias, contradictory hints, and false premises affect model predictions. Our results show that LLMs are highly sensitive to prompt variations, with Decision Flip Rates reaching up to 72% and False Alignment Rates exceeding 90%, indicating substantial instability and agreement with misleading prompts. To address this issue, we propose Evidence-Guided Debiasing Prompting (EGDP), a structured prompting strategy that enforces evidence-first reasoning. EGDP reduces decision instability and improves robustness, lowering Decision Flip Rates to as low as 12% and False Alignment Rates to as low as 21%, while increasing reliance on structurally grounded evidence. Our findings demonstrate that sycophancy bias poses a critical threat to the reliability of LLM-based code smell detection, and that evidence-guided reasoning provides an effective and generalizable mitigation approach.
[AI-116] Large Language Models in Misinformation Ecosystems: Misuse Defense and Vulnerability
链接: https://arxiv.org/abs/2607.10402
作者: Lingwei Wei,Dou Hu,Wei Zhou,Songlin Hu,Philip S. Yu
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Social and Information Networks (cs.SI)
备注: 35 pages, 8 figures
Abstract:Large language models (LLMs) have transformed misinformation from a primarily content-centric problem into a broader ecosystem-level security challenge. When misused, LLMs create risks beyond false content generation, enabling attacks on the social contexts, evidence sources, retrieval corpora, and verification workflows that misinformation defense depends on. In this paper, we introduce a role-layer framework to unify these risks and defenses. The role dimension characterizes LLMs as attackers, defenders, and vulnerable components of verification systems, while the layer dimension covers content, social contexts, evidence environments, and verification workflows. Guided by this framework, we organize LLM-enabled attacks, investigate LLM-based detection and verification methods, analyze vulnerabilities in LLM-centric detection paradigms, and discuss existing countermeasures against LLM-enabled attacks. Building on this synthesis, we identify three key open challenges: moving from static detection accuracy to budgeted ecosystem-level risk evaluation, hardening LLM-centered verification pipelines against adversarial manipulation, and deploying auditable human-in-the-loop verification systems for trustworthy real-world misinformation defense.
[AI-117] VINE: Taming Generative Control Policies for Reinforcement Learning
链接: https://arxiv.org/abs/2607.10369
作者: Rushuai Yang,Zhuo Han,Houlin Li,Hecheng Wang,Zhichao Wu,Rui Zhang,Zhaowei Zhang,Zihong Chen,Xiaohan Yan,Chiming Liu,Yi Chen,Wei Shan,Maoqing Yao
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Flow-matching policies have emerged as an effective policy parameterization for robot learning. They iteratively generate actions from noise, enabling highly expressive modeling of complex and multimodal action distributions. However, prior works observed that scaling these policies with value-gradient reinforcement learning (RL) often leads to training instability. Existing methods attribute this instability to iterative generation and therefore avoid end-to-end value-gradient optimization by sacrificing iterative generation, high expressiveness, or value-gradient optimization. Contrary to prior belief, we show the instability does not stem from iterative generation itself, but from the vanilla sampling strategy originally designed for behavior cloning, which becomes brittle under value-gradient RL. Motivated by this insight, we propose VINE, an RL-oriented sampling method that enables stable end-to-end value-gradient optimization for flow-matching policies. Instead of following a single flow trajectory, VINE reconstructs a new interpolation state at every denoising step, creating a stable differentiable path for value-gradient propagation while remaining compatible with the original flow-matching denoising process. As a result, VINE preserves the expressiveness and iterative generation of flow-matching without sacrificing end-to-end value-gradient optimization. Despite performing end-to-end backpropagation through all ten denoising steps, VINE achieves stable policy improvement and consistently outperforms state-of-the-art RL methods on the OGBench offline RL benchmark and real-world robotic manipulation task. Videos are available on our website: this https URL.
[AI-118] Co4ICF: Co-evolving Physics-Informed Surrogate and RL-based Pulse Optimizer for Inertial Confinement Fusion
链接: https://arxiv.org/abs/2607.10366
作者: Jiatong Zhao,Tengyue Zhang,Yuhan Wang,Fuyuan Wu,Junchi Yan
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Offline-trained surrogates for Inertial Confinement Fusion (ICF) suffer a well-known failure mode that iterative optimizers drive inputs into out-of-distribution (OOD) regions where predictions become unreliable. Here we present Co4ICF, a co-evolving framework that couples a physics-informed surrogate with a PPO-based pulse optimizer. The surrogate is iteratively fine-tuned on policy-induced trajectories, correcting extrapolation errors as the optimizer shifts the input distribution; the optimizer queries this evolving surrogate as a fast environment. In the 1D MULTI environment, Co4ICF achieves 146.1% normalized yield based on current laser design baseline; as a post-hoc cross-fidelity check, the optimized pulse further attains 246.9% normalized yield when directly evaluated in 2D-MULTI without any 2D training or fine-tuning. Budget-matched ablations support that the gains are not explained solely by additional simulation data and are consistent with the co-evolving mechanism playing a key role. We release a large-scale MULTI-IFE simulation dataset to support future benchmarking.
[AI-119] A Hyperbolic Neural Closure for M1 Radiation Transfer
链接: https://arxiv.org/abs/2607.10364
作者: Bongseok Kim,Jiahao Zhang,Johannes Krotz,Dinshaw Balsara,Ryan McClarren,Guang Lin
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE); Numerical Analysis (math.NA)
备注:
Abstract:In radiation transfer simulations, an M1 method achieves substantial computational savings by replacing the full angular transport equation with a low-order moment system. Because this reduced system is not closed, a closure model is required to represent the unknown higher-order moments using lower-order moments. While machine learning (ML)-based closures can improve accuracy beyond classical analytic closures, unconstrained learned closures may produce non-real characteristic speeds and consequently cause numerical solver breakdown. To guarantee real eigenvalues of the Jacobian associated with ML closures, we propose a hyperbolic neural closure for the M1 radiative transfer system. Rather than directly predicting closure terms, we parameterize the Jacobian through two neural networks: (i) a symmetric matrix network and (ii) a strictly convex entropy network whose Hessian defines a positive definite symmetrizer. These components are combined to yield a Jacobian that is similar to a symmetric matrix, thereby ensuring real eigenvalues. The closure is then reconstructed by numerical integration of the learned Jacobian field along a prescribed integration path. Numerical experiments show that the proposed closure not only achieves higher closure accuracy than classical analytic closures, but also improves solution accuracy and remains stable in discontinuous Galerkin simulations for radiative transfer problems.
[AI-120] ABot-Agent OS: A General Robotic Agent OS with Lifelong Multi-modal Memory
链接: https://arxiv.org/abs/2607.10350
作者: Jiayi Tian,Shiao Liu,Yuting Xu,Jia Lu,Zihao Guan,Honglin Han,Di Yang,Minqi Gu,Yifei Qian,Tianlin Zhang,Yanqing Zhu,Zeqian Ye,Menglin Yang,Fei Wang,Xu Hu,Xiuxian Li,Wei Zhang,Shihui Su,Yiyan Ji,Jingbo Wang,Ziteng Feng,Jiaheng Liu,Zhaoxiang Zhang,Xiaolong Wu,Mingyang Yin,Zedong Chu,Mu Xu
类目: Artificial Intelligence (cs.AI)
备注: Code: this https URL
Abstract:Recent VLM and VLA systems have improved robotic perception and action prediction, yet long-horizon embodied agents still require a general runtime layer for reasoning, memory, tool use, verification, and cross-embodiment execution. We present ABot-AgentOS, a general robotic Agent Operating System that sits above low-level controllers and provides a deliberative agent layer for scene-conditioned planning, context-isolated skill execution, multi-stage verification, multi-modal memory, and edge-cloud collaboration. To evaluate such systems, we introduce EmbodiedWorldBench, an executable benchmark with 16 indoor, outdoor, and hybrid scenes, four difficulty levels, and over 200 tasks involving navigation, object search, NPC dialogue, dynamic events, and trace-grounded scoring. ABot-AgentOS further introduces Universal Multi-modal Graph Memory, a persistent source-grounded substrate that converts dialogue, visual observations, spatial context, temporal relations, and task traces into typed nodes and edges. A failure-driven self-evolution loop converts diagnosed memory failures into gated runtime evo-assets that are promoted only to later evaluation splits, preventing current-split ground-truth leakage while enabling continual improvement. On an initial EmbodiedWorldBench subset, ABot-AgentOS improves over a single-controller baseline in both task success and goal completion. Across memory benchmarks, ABot-AgentOS Static achieves 87.5 on LoCoMo, 59.9 on OpenEQA EM-EQA, 88.6 on Mem-Gallery, and 76.5 Acc@All on NExT-QA; self-evolution further improves LoCoMo to 88.7, OpenEQA to 60.4, and Mem-Gallery to 89.0. These results suggest that a general Agent OS layer can improve long-horizon embodied execution while providing persistent, auditable memory for continual interaction.
[AI-121] Comparing Socio-technical Design Principles with Guidelines for Human-centered AI
链接: https://arxiv.org/abs/2607.10331
作者: Thomas Herrmann
类目: Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注: 15 pages, 3 tables
Abstract:Human-centered AI (HCAI) refers to guidelines or principles that aim on ethi-cally oriented design of systems. We compare HCAI- guidelines with princi-ples of socio-technical systems that emerged in the context of conventional in-formation technology. The comparison leads to a revision of socio-technical heuristics by including aspects of AI-usage. The comparison reveals that con-tinuous evolution is a basic characteristic of socio-technical systems, and that human oversight or interventions and the subsequent appropriation of AI-systems lead to continuous adaptation and re-design of the systems, if autono-my is collaboratively exercised. From a socio-technical point of view, the cru-cial requirement of transparency has not only to be fulfilled with technical fea-tures, but also by contributions of the whole system including human actors. It will be promising for using AI, if not only technical features, but organization-al and social practices are socio-technically designed in a way that compen-sates shortcomings of AI.
[AI-122] Measure the Sim-to-Real Gap: Designing an Affordable Real-World Benchmark Platform for Reinforcement Learning in AIoT Systems
链接: https://arxiv.org/abs/2607.10309
作者: Rongping Zhou,Omid Tavallaie,Shuaijun Chen,Albert Y. Zomaya
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Reinforcement learning (RL) is commonly employed to enhance the performance of autonomous systems, including the Autonomous Internet of Things (AIoT). However, the trial-and-error nature of RL, when conducted in real-world environments, is costly and hazardous in some scenarios. Consequently, the majority of RL research is conducted in simulation. This reliance introduces challenges related to the Sim-to-Real transferability. Evaluating the Sim-to-Real algorithmic robustness and the Sim-to-Real gap is a critical prerequisite for research aimed at improving RL performance in the real world. Therefore, industries such as robotics have developed concurrent simulation and physical platforms to facilitate this research. However, a universal Sim-to-Real benchmark platform for AIoT does not currently exist. To address these concerns, we developed a real-world AIoT platform for studying RL in AIoT. On this platform, an agent deployed on an edge device plays video games on a separate host computer via a hardware-emulated keyboard, guided by vision input. This platform uses commercially available components costing less than USD 400, together with two computers. Because the system’s objective is game score maximization, it inherently mitigates safety risks associated with real-world RL deployments. Experimental results show the simulation-trained agent suffers a 1160% performance degradation relative to the human-level performance after real-world deployment, indicating a significant Sim-to-Real gap. Direct real-world training using the deep Q-network (DQN) algorithm achieves approximately 38% of human-level performance after 10 million training steps, demonstrating the feasibility of RL under real-world conditions. These results suggest that the proposed Sim-to-Real benchmark platform provides a substantial foundation for qualitative and quantitative evaluations of RL in real-world AIoT systems.
[AI-123] Partial Contracts Suffice: Sound LLM -Inferred Regression Verification
链接: https://arxiv.org/abs/2607.10291
作者: Yiannis Charalambous,Rafael Menezes,Youcheng Sun,Lucas C. Cordeiro
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: 12 pages
Abstract:Software evolves continuously, yet ensuring that a patch preserves intended behavior without re-verifying an entire codebase remains difficult. Regression verification addresses this problem, but existing techniques require expensive whole-program reasoning or rely on manually written specifications that are rarely available in practice. We present the first contract-based regression verification tool. Contract soundness is ensured by proving all function versions match the behavior. The contract then verifies program flow via assume-guarantee. We ask whether a partial, caller-sufficient contract, rather than a full behavioral specification, is enough. On Frama-C-Problems we strengthen each inferred contract past what the caller needs and measure how much tighter it becomes. It barely moves: for most targets in every model the caller-sufficient contract is already the tightest the loop reaches, and our tightness comparator rates the partial and strengthened contracts equivalent for the large majority of targets it can compare. Partial-spec contracts thus capture nearly all the attainable tightness, so stopping at caller-sufficiency costs almost nothing. The regression check underneath is sound: on the third-party EqBench-C suite it never fabricates an equivalence, returning zero false proofs and reporting an unprovable difference instead. It also surfaced nine pairs that EqBench mislabels as equivalent, more than a concurrent tool reports. The contracts themselves are inferred automatically from the checker’s own counterexamples, with no separate specification step; on Frama-C-Problems and the ANSSI X509 parser this reaches a verification rate comparable to tools AutoSpec and Preguss, while a passing result certifies at least as strong a property, which we call \emphsafety-preserving conditional equivalence: enforcement plus caller-sufficiency.
[AI-124] Behavioural Signatures of Risk-Sensitive Decision-Making in Large Language Models
链接: https://arxiv.org/abs/2607.10251
作者: Xuankun Rong,Wenke Huang,Bo Du,Dacheng Tao,Mang Ye
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:As large language models (LLMs) are increasingly used in decision support, it is important to understand whether their choices under uncertainty exhibit stable and interpretable behavioural regularities. Human decision-making combines relatively persistent risk preferences with context-dependent adjustment, yet it remains unclear whether analogous behavioural structure can be observed in LLM-based decision systems. Here we examine this question using a controlled multi-model framework based on no-limit Texas Hold’em, where behaviour is quantified by Participation, measuring voluntary engagement in uncertain opportunities, and Proactiveness, measuring pre-flop risk escalation. Across homogeneous self-play and heterogeneous mixed-model interactions, frontier LLMs exhibit stable, model-specific risk profiles, forming a spectrum from conservative to aggressive decision styles. These profiles remain largely robust under changing opponent composition, while the most conservative and most aggressive models diverge further in mixed settings. Under global risk pressure and personal resource constraint, models adapt in structured but heterogeneous ways, ranging from broad behavioural contraction to selective de-escalation and near-invariant behaviour. These findings suggest that LLMs differ not only in baseline risk disposition, but also in the risk signals they respond to and the flexibility with which they adjust, providing a behavioural basis for auditing risk-sensitive decision-making in interactive settings. Our code is publicly available at: this https URL.
[AI-125] When Are Sparse Feature Interventions Actually Localized? Matched Evaluation for SAE-Based Safety Control
链接: https://arxiv.org/abs/2607.10226
作者: Daming Luo
类目: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注: 11 pages, 5 figures, 4 tables. Preliminary version; extended multi-model study in progress
Abstract:We evaluate when sparse autoencoder (SAE) features act as localized control handles for safety-relevant behavior. This question is difficult because apparent success can arise from weak interventions, mismatched baselines, model robustness, or degenerate outputs that automated safety judges mark as unsafe without representing meaningful harmful compliance. We introduce a matched coherence-gated evaluation protocol for runtime safety interventions: methods are compared at matched target-effect points, and the primary target metric counts harmful compliance only when an output is both judge-unsafe and coherent. Applying this protocol to three prompt splits on Gemma-2-9B-it with a Gemma Scope layer-20 residual SAE, we find that SAE feature ablation has a narrow useful regime. SAE top800 reaches a low-to-mid target effect with lower total perturbation and competitive utility, but SAE top1600 loses utility relative to a matched dense refusal-direction baseline, and SAE top3200 primarily induces coherence collapse. Human audit confirms that coherence gating removes unsafe-only artifacts, and feature diagnostics show that the useful regime is driven by a stable head of refusal-aligned features whose activation separation decays rapidly with rank. These results argue that SAE-based safety interventions should be evaluated as regime-dependent control mechanisms rather than assumed to be uniformly localized.
[AI-126] Exploratory Analysis of Deep Learning Models for Forecasting Meteorological Parameters in the Agricultural Sector
链接: https://arxiv.org/abs/2607.10208
作者: Piotr Sikora,Sotirios Kontogiannis
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 13 pages, 5 figures
Abstract:Accurate meteorological forecasting is essential for agricultural planning, irrigation management, and environmental decision support. This study conducts a comparative evaluation of recurrent and hybrid deep learning architectures for multivariate forecasting of reference evapotranspiration ( ET_0 ), vapour pressure deficit (VPD), wind speed, and the sine and cosine components of wind direction. The analysis utilizes 134,376 hourly observations from Ioannina, Greece, spanning January 2011 to April 2026, sourced from ERA5 via the OpenMeteo Historical Weather API. Single and multi-layer GRU and LSTM networks are compared with hybrid 1D-CNN-GRU and 1D-CNN-LSTM models for two forecasting tasks: a 24-hour next-day forecast and a 168-hour week-ahead forecast. Performance is evaluated using normalized root mean squared error, the coefficient of determination, and a composite Weighted Quotient Score (WQS). The most effective purely recurrent models are a 64-unit LSTM for the 24-hour horizon, with a WQS of 0.816755, and a 1024-unit GRU for the 168-hour horizon, with a WQS of 0.779465. The hybrid CNN-GRU models achieved the highest overall scores of 0.827535 and 0.782863 for the 24-hour and 168-hour horizons, but with additionally more number of units respectively to LSTM models, while the CNN-LSTM models yield nearly identical results with substantially fewer parameters. Compared to the corresponding recurrent baselines, the hybrid models improve WQS by 1.22–1.63% at 24 hours and by 0.44–0.45% at 168 hours, indicating that convolutional feature extraction is more beneficial for short-term forecasting.
[AI-127] Source-Lifted Flow Matching for Intervenable Multimodal Imitation
链接: https://arxiv.org/abs/2607.10206
作者: He Zhang,Ying Sun,Pengteng Li,Ziyang Chen,Yiren Zhao,Ziyang Rao,Weiyu Guo,Yandong Guo,Hui Xiong
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Flow-matching policies are promising for imitation learning because they model complex multimodal action distributions. However, their stochasticity is largely passive: repeated sampling may yield diverse behaviors, but users cannot directly choose among valid continuations from the same state. We propose Source-Lifted Flow Matching (SL-FM), a source-intervenable flow-matching policy that exposes such a handle while keeping the velocity field shared and latent-free. The handle selects only the source endpoint of the conditional flow, not a mode-specific field, preserving the standard formulation while avoiding decomposition into separate mode-conditioned dynamics. The core mechanism is \textbfOrthogonal Source Lifting, designed to prevent path-crossing ambiguity. Instead of partitioning target actions by mode, SL-FM lifts handle-specific sources into auxiliary orthogonal coordinates and keeps targets in the original action subspace. This preserves the demonstrated action distribution while allowing one shared field to carry different branches without merging at crossings. To keep handles usable across states, we learn a state-dependent source mixture end to end and use a responsibility floor, giving each handle weak supervision and mitigating dead modes. Experiments on crossing-flow diagnostics and robot-control benchmarks show that SL-FM converts passive source randomness into an actionable intervention variable. It removes crossing-induced composite trajectories, changes future routes in 91.1% of matched-prefix interventions, and achieves strong free-deployment performance, with improvements in several benchmark settings. Overall, source geometry provides actionable multimodal control without conditioning the velocity field on the selected mode.
[AI-128] When Does Depth Survive Composition? Compute–Quality Regimes in Latent World Models
链接: https://arxiv.org/abs/2607.10203
作者: Achyuthan Sivasankar
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 15 pages, 9 figures, 2 tables
Abstract:Adaptive-compute world models – early-exit or mixture-of-depths predictors that spend variable depth per step – assume depth buys better predictions and can be routed adaptively. In autoregressive rollouts, the first assumption requires depth’s per-step precision to survive composition. We test this with a pre-registered instrument, the shallow penalty \rho=\mathrmerr(\textshallowest-exit rollout)/\mathrmerr(\textfull-depth rollout) , across nine DeepMind Control tasks under matched single-step ( K=1 ) and multi-step ( K=4 ) training, three seeds each. We find three regimes: on 6/9 tasks depth helps rollouts (intrinsic, \rho up to 4.7\times ), on 2/9 the shallow exits beat the full stack (inversion, \rho down to 0.85\times ), and one is flat. The robust inversion (cheetah) is not a property of the dynamics but is created by training: an ablation supervising early exits only at the first rollout step erases it ( \rho: 0.87\to1.18 , n=8 , \Delta=+0.31 ), while an intrinsic-tradeoff task is unaffected – a double dissociation we call the routability catch-22, since the supervision that makes exits routable is what trains them to out-roll the full stack. The regime is partly predictable a priori: observation/action dimensionality and one-step model error correlate with \rho at |\textSpearman|\approx0.75 ( n=9 ). Inside a CEM planner, \rho 's sign predicts whether planning benefits from depth, most sharply on the inversion task, where shallow planning beats deep. Finally, three cautions: a task’s regime depends on the metric space, the rollout horizon, and the encoder. All thresholds and gates were fixed before the compute campaign, including a pre-registered negative for the hypothesis that motivated the study.
[AI-129] Comparing Socially-Equitable Renewable Energy Budget Allocation MDP Policies in Mature and Emerging Economies
链接: https://arxiv.org/abs/2607.10201
作者: Riya Kinnarkar,Mansur M. Arief,Yan Pratama Akhra,Dino Arla
类目: ystems and Control (eess.SY); Artificial Intelligence (cs.AI); Optimization and Control (math.OC)
备注:
Abstract:Equitable renewable-energy planning is a sequential decision problem, but the decision variables available to a public planner differ sharply between mature and emerging economies. In the former the government largely builds generation, while in the latter it steers private investment through incentives and quotas. We formulate socially-equitable renewable-energy budget allocation as a Markov Decision Process (MDP) and, using a single problem-agnostic solver interface, compare the same policies across the two settings: eight U.S. cities (a mature economy) and West Java, Indonesia (an emerging economy). The results show that across both settings, a receding-horizon value-iteration policy dominates. In the U.S., it reaches 66% renewable penetration while cutting the underserved low-income population by 96% versus a random baseline. In West Java it closes the low-access gap while crowding in the most private capital. More interestingly, a naive market-chasing heuristic, which is mildly sub-optimal in the U.S., could yield catastrophic outcomes in Indonesia, by underserving every low-access region, because chasing attractive markets and serving the underserved goals diverge once the planner acts through private developers.
[AI-130] GRATE: Temporal Extensions for Inductive KG Foundation Models via Gated Rotary Attention ICML2026
链接: https://arxiv.org/abs/2607.10197
作者: Jiaxin Pan,Osama Mohammed,Daniel Hernández,Steffen Staab
类目: Artificial Intelligence (cs.AI)
备注: Accepted at the ICML 2026 Workshop on Graph Foundation Models: A New Era for Graph Machine Learning. 17 pages, 4 figures
Abstract:Knowledge graph foundation models such as Ultra and Trix achieve strong inductive transfer by learning relation-graph representations that generalise to unseen entities and relations. Extending this transferability to temporal knowledge graphs (TKGs) remains challenging: existing temporal models tie their parameters to dataset-specific entities, relations, or timestamps and are not designed to transfer to TKGs with disjoint vocabularies. We propose GRATE (Gated Rotary Attention for Temporal Encoding), an entity-side message function that adds no learnable parameters and encodes time through relative time differences by rotating each edge message according to its time gap to the query and applying a query-conditioned gate to select temporally relevant signals. GRATE integrates into NBFNet-style KG foundation models while preserving structural transferability. Existing TKG benchmarks evaluate within shared train/test vocabularies and cannot directly test cross-dataset temporal transfer; we therefore construct GDELTIndT and WIKIIndT, inductive transfer benchmark suites with disjoint entities, relations, and timestamps spanning both interpolation and extrapolation. Across these benchmarks and held-out forecasting datasets, a single jointly pretrained GRATE checkpoint improves over the static base model in most settings.
[AI-131] Breaking the Quality–Intelligibility Trade-off in Streaming Target Speaker Extraction via Deep-Feature-Anchored Preference Optimization
链接: https://arxiv.org/abs/2607.10191
作者: Shuhai Peng,Jinjiang Liu,Hui Lu,Liyang Chen,Guiping Zhong,Jiakui Li,Shiyin Kang,Zhiyong Wu
类目: ound (cs.SD); Artificial Intelligence (cs.AI)
备注:
Abstract:Generative streaming models for Target Speaker Extraction (TSE) commonly exhibit a quality–intelligibility trade-off, wherein naive optimization for perceptual audio quality tends to degrade speech intelligibility, and conversely. We reveal that this trade-off arises not from the constraints of streaming architectures, but from an inappropriate choice of optimization anchor. Directly optimizing against audio quality metrics induces catastrophic reward hacking, where content critical to pronunciation and intelligibility is systematically erased to maximize a proxy score. To break this bottleneck, we propose two complementary improvements: an enlarged Conformer convolution kernel for richer local spectro-temporal modeling, and WavLM-anchored Direct Preference Optimization (DPO) fine-tuning strategy. DPO preference pairs are ranked by WavLM cosine similarity, a deep acoustic feature encoding both phonetic structure and speaker identity, providing an optimization anchor that resists hacking. Under a 560 ms streaming chunk size, the proposed method achieves a 10.9% relative intelligibility improvement (word error rate: 0.138 to 0.123), with marginal simultaneous gains in audio quality and speaker similarity.
[AI-132] Automated Tensor Scheduling for Hybrid CPU-GPU LLM Inference on Consumer Devices
链接: https://arxiv.org/abs/2607.10183
作者: Yangyijian Liu,Hongyi Ye,Mingyang Li,Wu-jun Li
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Running large language models on consumer devices such as laptops and desktops is challenging because model weights often exceed GPU memory capacity, making offloading inference necessary to extend effective model capacity with CPU memory. Existing offloading systems, however, typically rely on coarse layer-level or expert-level scheduling, which overlooks substantial heterogeneity among tensors within the same layer and adapts poorly to changing hardware load conditions on such devices. This paper presents ATSInfer, a hybrid CPU-GPU inference system for consumer devices that performs offloading at tensor granularity. ATSInfer combines static tensor placement with load-aware dynamic transfer, and introduces asynchronous CPU-GPU coordination to efficiently schedule hardware storage, data movement, and computation across heterogeneous backends. We implement ATSInfer and evaluate it on representative consumer platforms using both dense and MoE models. Compared with existing systems, ATSInfer improves prefill throughput by up to 1.94 \times and decode throughput by up to 3.29 \times , while also increasing GPU utilization and making more effective use of PCIe bandwidth. These results show that ATSInfer can substantially improve the user experience of local LLM deployment on personal consumer devices.
[AI-133] ActiveFly-Bench: Aligning Embodied Question Answering with Vision-Language-Action for Aerial Embodied Perception
链接: https://arxiv.org/abs/2607.10180
作者: Weichen Zhang,Shiquan Yu,Yinan Zhu,Peizhi Tang,Shilong Ji,Zhiyuan Deng,Tianyi Lyu,Haoyang Wang,Xin Zeng,Chen Gao,Yong Li,Xinlei Chen
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:We introduce ActiveFly-Bench, the first benchmark to bridge cyberspace reasoning and physical-world interaction for UAV embodied perception. The benchmark decomposes active perception into three hierarchical tasks: Aerial Embodied Question Answering (Air-EQA), Observation Behavior Planning (OBP), and Fine-grained Language-guided UAV Control (FLUC), explicitly connecting high-level task understanding, behavior planning, and low-level control. The datasets are collected from both real-world and simulated outdoor environments for training and evaluation. We further develop ActiveFly, a closed-loop UAV agent that integrates visual-language reasoning with fine-grained control, and deploy it on a physical UAV platform. Experiments with representative VLMs and VLA models show that current UAV agents still struggle with behavior planning, viewpoint adjustment, and robust task completion in active perception. These results establish ActiveFly-Bench as a new testbed for embodied aerial intelligence.
[AI-134] Beyond Euclidean Clipping: Overcoming Exploration Collapse in LLM RL via Riemannian Isometric Policy Optimization ICML2026
链接: https://arxiv.org/abs/2607.10169
作者: Zhicheng Cai,Xinyuan Guo,Hanlin Wu,Mingxuan Wang,Wei-Ying Ma,Ya-Qin Zhang,Hao Zhou
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: ICML 2026
Abstract:Reinforcement learning (RL) has become a dominant paradigm for enhancing LLMs’ reasoning capabilities. However, RL algorithms with PPO-Clip are inherently limited by exploration collapse. Subsequent works remain primarily heuristic and fail to identify the essential cause of PPO-Clip’s failure. This work reveals the fundamental flaw of PPO-Clip: it implicitly measures policy discrepancy using Euclidean metric, which is theoretically inconsistent with the intrinsic geometry on the policy Riemannian manifold. This geometric mismatch results in overly conservative updates in low-probability regions while aggressive in high-probability regions, ultimately collapsing exploration. To correct this geometric flaw, we propose Riemannian Isometric Policy Optimization (RIPO), which guarantees isometric policy updates on the Riemannian manifold, effectively balancing exploration and exploitation. We further show that RIPO achieves a favorable bias-variance trade-off, which stabilizes optimization. Extensive experiments demonstrate that RIPO significantly surpasses existing LLM RL algorithms across seven competition-level benchmarks (up to 60% improvement over GRPO on AIME24).
[AI-135] ranscript-Free Lightweight Detection of Alzheimers Disease from Spontaneous Speech Using Handcrafted MFCC-Dominant Acoustic Biomarkers
链接: https://arxiv.org/abs/2607.10168
作者: Rashin Gholijani Farahani,Azam Bastanfard
类目: ound (cs.SD); Artificial Intelligence (cs.AI)
备注: 10 pages, 5 figures, 40 references. Submitted for peer review
Abstract:It is still hard to find Alzheimer’s disease (AD) early, especially when neuroimaging is expensive or tools that depend on language are not available. Spontaneous speech provides a non-invasive signal; however, numerous current methodologies depend on transcripts/ASR or computationally intensive deep models. We offer a simple, audio-only baseline for detecting AD using 176 Cookie Theft recordings from the DementiaBank Pitt corpus (88 AD, 88 controls). WebRTC voice activity detection (VAD) is used to separate speech from non-speech. We take out 99 hand-crafted acoustic-temporal features, including pause and fluency statistics, spectral/prosodic descriptors, and MFCC summaries with \Delta and \Delta\Delta. Evaluation is performed using a stringent speaker-independent GroupShuffleSplit,documenting performance across 30 iterations. A lightweight SVM with an RBF kernel gets an average AUC of 0.674 across runs. For example, a single split has an AUC of 0.742 and an accuracy of 0.657. We also present an exploratory compact-feature analysis utilizing a Top-20 subset ranked by Random Forest importance; since selection is not nested within training splits, these results may be overly optimistic and are not employed for primary conclusions (AUC 0.719). The results indicate that transcript-free spectro-temporal and fluency-related cues can facilitate speaker-independent Alzheimer’s disease screening from raw audio, establishing a practical foundation for deployment-oriented research.
[AI-136] UNIT: Unleash Large Language Models Potential for Graph Continual Learning ACM-MM2026
链接: https://arxiv.org/abs/2607.10159
作者: Tairan Huang,Yili Wang,Beibei Hu,Yiting Shi,Qiutong Li,Changlong He,Jianliang Gao
类目: Artificial Intelligence (cs.AI)
备注: Accepted by ACM MM 2026
Abstract:In real-world multimodal web scenarios, graph-structured data often arrives in a streaming manner, making graph continual learning a crucial paradigm for continuously modeling such evolving structures. However, existing graph continual learning methods still face two fundamental challenges. 1) semantic-structural separation, where the graph-based methods excel at modeling topological relationships but neglect deep semantics. 2) imbalanced knowledge transfer, where existing models fail to effectively leverage general knowledge gained from early tasks to benefit subsequent new tasks. To address above issues, we propose a novel framework, \textbfUNleash Large Language Models PotentIal for Graph ConTinual Learning (UNIT). By fine-tuning large language model only on the first task, we bridge the distributional gap between the pre-trained LLM corpus and the target task dataset to enhance the adaptability of LLMs for graph-structured tasks. Meanwhile, we propose an uncertain-aware anchor generation mechanism to effectively preserve representative knowledge across tasks, avoiding the neglect of universal knowledge learned from previous tasks. Additionally, we introduce structural confluence modeling to explicitly integrates graph topology information into semantic information, enhancing the collaborative capabilities between semantic understanding and structural modeling. Extensive experiments demonstrate that our proposed method achieves state-of-the-art performance in the graph continual learning task.
[AI-137] IdeaTrail: Full-Process Agent Trajectories for Scientific Ideation
链接: https://arxiv.org/abs/2607.10144
作者: Hengquan Guo
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Scientific research is a complex, multi-stage workflow rather than a single act of text generation. The ideation process typically emerges through literature search, paper reading, tool use, claim checking, cross-paper synthesis, brainstorming, rejection of weak directions, and iterative writing. Existing resources capture individual components of this process, but datasets that jointly record tool use, evidence acquisition, intermediate artifact evolution, and idea- or proposal-level endpoints remain limited. This report introduces \method, a multi-turn process-trajectory dataset for scientific ideation and proposal generation. Each instance records a research process from evidence gathering to either idea selection or proposal construction. Rather than freely fabricating trajectories, \method starts from human-selected high-quality research papers and proposal artifacts and uses a Generator–Advisor synthesis loop. The Generator produces the visible trajectory through actions, observations, and artifact edits, while the Advisor has access to the full generation context and checks grounding, causal order, naturalness, and leakage from hidden targets. This reverse-to-forward procedure produces multi-turn research data that remains aligned with real scientific artifacts while approximating the uncertainty, evidence use, and staged convergence of research practice. \method provides both a dataset and a general recipe for synthesizing process-supervision data for scientific research agents.
[AI-138] LLM s as a Jury: Cross-Model Consensus Can Outperform Process Reward Models for LLM Reasoning
链接: https://arxiv.org/abs/2607.10139
作者: Ning Liu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Selecting the correct answer from a pool of candidate reasoning chains is the engine of test-time scaling, yet the standard selectors each carry a cost: self-consistency inherits the errors of the single model it resamples, and trained reward models need labeled data and transfer poorly off-distribution. We study a third signal, free at inference time: cross-model consensus, the degree to which independently trained models, each solving the problem once, agree on a final answer. We treat the panel as an LLM-jury, in which the structure of agreement, not any model’s score of another, is the verification signal. Across seven benchmarks it selects correct answers better than self-consistency and far better than a model scoring its own candidates: on competition math it closes the entire gap to an oracle selector, while self-scoring closes almost none. The mechanism is error decorrelation: independently trained models err differently, so their wrong answers scatter while the correct one accumulates agreement. We make this precise with a parameter-free law, derived in closed form, that predicts consensus accuracy from three measured panel statistics to a mean absolute error of 0.03 and exposes the method’s ceiling: a shared-error floor where models share a misconception, near zero on math but non-trivial on science. Against four trained verifiers spanning discriminative, outcome, and generative reward models, the free LLM-jury matches the strongest inside their math training domain and is the top selector outside it. Cross-model consensus is thus a verifier we can characterize in advance: a law that says when to trust it, and a floor that marks where it cannot.
[AI-139] SALT-GNN: Handling Dense Neighborhoods in Anti-Money Laundering Graphs via Statistics-Aware Attention
链接: https://arxiv.org/abs/2607.10131
作者: Lidia Losavio,Francesco Sovrano,Dario Fenoglio,Martin Gjoreski,Marc Langheinrich
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 19 pages, 7 figures. Code available upon publication
Abstract:Money laundering threatens financial stability and exposes institutions to penalties, motivating automated detection. Because laundering schemes often emerge through relational patterns, graph neural networks (GNNs) are increasingly used for anti-money laundering (AML). Yet AML GNNs are typically evaluated with aggregate metrics such as overall F1 score, which hide an operational issue: high-activity recipient accounts concentrate many incoming transactions, making suspicious signals harder to isolate and costlier to investigate. We introduce a recipient-degree stratified evaluation that reports standard AML metrics across recipient-context density. Across three datasets (HI-Small, HI-Medium, and AMLSim-32k-5%), it reveals consistent degradation in dense recipient contexts, which we trace to three GNN characteristics: two known limitations that AML amplifies, i.e., (1) multiset non-discriminability and (2) cardinality blindness, and (3) an attention-specific effect: in dense neighborhoods, normalized attention attenuates weak but pattern-relevant multi-hop signals. Guided by this diagnosis, we propose SALT-GNN, a lightweight statistics-aware architecture that fuses degree-aware statistical aggregation with attention at each message-passing layer, so distributional and cardinality information shapes the node states used by subsequent attention steps. Ablations support fusion placement as a key factor in dense-context performance. On HI-Small and HI-Medium, SALT-GNN uses up to 77% fewer parameters than task-specific graph-transformer baselines while improving dense-context F1 score by 3-6 points; on AMLSim-32k-5%, it improves highest-degree F1 score by 16-20 points. The gains hold for both Transformer- and GAT-style attention, indicating that the benefit comes from where statistical and attentional evidence is fused rather than from a specific attention operator.
[AI-140] GAE: Graph-Augmented Evolution for Scientific Discovery via Reinforcement Optimization ICML2026
链接: https://arxiv.org/abs/2607.10127
作者: Xuanzhou Chen,Taoli Cheng
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: ICML 2026, AI for Science Workshop
Abstract:Evolutionary program search guided by Large Language Models (LLMs) has emerged as a powerful paradigm for automated scientific discovery. However, current approaches are fundamentally constrained by three bottlenecks: structurally blind parent selection, sparse whole-program evaluation rewards, and static mutation operators that fail to adapt during search. We present GAE (Graph-Augmented Evolution), a framework that resolves these limitations through a tightly coupled, three-pillar architecture. First, a relational graph neural network (GNN) parses programs into typed computation graphs, producing structure-aware embeddings. Second, an RL-optimized meta-controller leverages these embeddings to replace blind evolutionary sampling with a directed policy, dynamically selecting optimal parents and mutation directions based on reward history. Third, an online GRPO fine-tuning loop continuously updates the LLM mutation operator at test-time using group-normalized evaluation rewards, directly aligning the model’s generation distribution with high-fitness structural edits. We evaluate GAE on a challenging scientific discovery task: symbolic regression for complex nonlinear oscillator systems. By transforming stochastic search into a directed, self-improving trajectory, GAE efficiently discovers closed-form physical equations, consistently matching or outperforming static LLM-driven baselines and achieving state-of-the-art out-of-distribution performance.
[AI-141] ML in a Box: Analyzing Containerization Practices in Open Source ML Projects
链接: https://arxiv.org/abs/2607.10126
作者: Faten Jebari,Emna Ksontini,Amine Barrak,Wael Kessentini
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:Containerization has become increasingly essential in the machine learning (ML) domain, providing reproducibility, portability, and environment consistency. While prior studies have analyzed Dockerfile structures and best practices, none have examined ML projects in depth to reveal how the iterative nature of ML workflows influences container footprint, build performance, and caching behavior. We present the first large scale empirical study of 1,993 ML related Dockerfiles, combining quantitative analysis of container roles in ML projects and build dynamics with a qualitative investigation of refactoring practices. Results show that containers serve distinct roles across training, inference, and infrastructure. Containers are typically large, averaging 10.27 GB in size, and require long build times of about 8.84 minutes. We find that 44.4% of commits trigger rebuilds, primarily due to context file changes (96.4%), with experimentation being the main motive behind those commits that initiate rebuilds. Despite partial cache reuse, 71% of rebuild work is wasted on redundant computation. From stable projects, we identify 7 recurring ML-specific Dockerfile refactoring patterns that improve build efficiency and reduce container footprint.
[AI-142] A Large-Scale Dataset of MCP Implementations on GitHub
链接: https://arxiv.org/abs/2607.10123
作者: Benny Toeppe,Amine Barrak,Emna Ksontini
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:The rapid emergence of the Model Context Protocol (MCP) has introduced a new standard for connecting large language models to external tools and services. Despite its rapid adoption in open-source development, systematic understanding of how MCP is implemented, structured, and maintained remains limited. This study presents the first large-scale, evidence-based dataset of real-world MCP implementation collected directly from GitHub. Using a hybrid pipeline that integrates the GitHub REST and GraphQL APIs with custom Python verification scripts, 3,238 candidate repositories were discovered, filtered, and validated through multi-stage evidence checks. Each verified project was classified by operational role (e.g., client, server, gateway) and exported in a reproducible JSONL schema. A manual review of a representative subset confirmed an overall precision of 83% at a 95% confidence level, and additionally revealed a set of repositories functioning primarily as educational samples, tutorials, or demonstration templates. A targeted exclusion rule was then applied to remove these non-operational repositories, resulting in a final dataset of 2,297 validated MCP projects. The analysis shows that Python and TypeScript dominate MCP development, with hybrid architectures emerging as the most common design pattern. By emphasizing transparent verification strategies, structured evidence tagging, and reproducible data organization, this work establishes a foundational benchmark for studying real-world MCP ecosystems and supports future research on integration, connectivity, and compatibility across the broader developer community.
[AI-143] When Data Imbalance Helps: Robust Generalization Through Shortcut Saturation
链接: https://arxiv.org/abs/2607.10116
作者: Cheng-Ting Chou,Duc Binh Hoang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:We study robust generalization under spurious correlations: tasks where a shortcut feature is correlated with the true label in training but anti-correlated in an adversarial held-out split. Varying the spurious ratio r (the fraction of training examples where shortcut = true label) and model capacity, we find a counterintuitive result: data imbalance promotes generalization in sufficiently capable models. On a synthetic task where the true label is sum parity of an integer sequence and the shortcut is the parity of the maximum-valued element, a 2-layer, 2-head transformer generalized (reached 100% adversarial accuracy) in 0% of seeds at r=0.50 but 77% of seeds at r=0.90 . The effect is absent in 1-layer models, where imbalance instead traps the model on the shortcut. Through mechanistic analysis – gradient conflict dynamics, circuit evolution, and QK/OV circuit ablations – we characterize a mechanistic pathway consistent with imbalance promoting generalization.
[AI-144] Dynamic Agent Skills: A Lifecycle Survey and Taxonomy of Evolving Skill Libraries
链接: https://arxiv.org/abs/2607.10113
作者: Yubo Li
类目: Artificial Intelligence (cs.AI)
备注: Accepted by TMLR (2026.07), OpenReview Link: this https URL
Abstract:Large language model agents increasingly store reusable procedures outside the model. These reusable procedures are often called \emphskills: they may be code functions, natural-language instructions, this http URL packages, workflow graphs, or learned adapters that a future agent can retrieve and invoke. This taxonomy-driven survey asks how such skill libraries change over time. Across a 124 -paper 2023 – 2026 audit set, we synthesize dynamic skill systems as \emphlifecycle-managed, verified, evolving artifact stores: agents collect evidence from interaction, propose skill updates, verify and admit candidates, organize them for retrieval and composition, repair or prune stale entries, and govern sharing through provenance and rollback. We organize the literature around three survey tools. First, a \textsix -sense taxonomy distinguishes the structurally different artifacts called ``skills’’ in current papers. Second, an \texteight -stage lifecycle architecture identifies the recurring design decisions behind evidence acquisition, proposal, verification/admission, storage, retrieval/composition, maintenance, distillation/portability, and governance. Third, a lightweight skill-record schema and \textten -operator vocabulary provide common terms for comparing library updates without elevating them into a separate method contribution. Using this structure, we synthesize evidence-graded patterns with explicit caveats: admission and repair are repeatedly important, verifier quality materially affects skill-aware RL, flat retrieval can degrade as libraries grow, and current benchmarks still under-report library trajectories, usage–utility gaps, and safety surfaces. We close with concrete reporting standards and open problems for evaluating dynamic skills as changing libraries rather than static prompt or tool collections.
[AI-145] Minionese: Comprehensive Benchmark and Mechanistic Study of Multilingual LLM Safety
链接: https://arxiv.org/abs/2607.10112
作者: Chigozirim Ifebi,Brent Kong,Ayushi Mehrotra
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:Safety alignment in large language models remains brittle across languages: prompts reliably refused in English can elicit harmful compliance in non-English and low-resource settings. We introduce \textscMinionese, a multilingual jailbreak benchmark spanning 18 languages, 4 resource tiers, and 4 perturbation types (standard translation, code-switching, transliteration, and translationese), paired with a geometric mechanistic analysis of refusal failure across language tiers. We show that each attack type produces a distinct vulnerability profile: transliteration vulnerability is mediated by script identity, code-switching maintains effectiveness through the lowest-resource tier, and a sharp safety regime transition between Tiers 2 and 3 is consistent across all models. Mechanistically, low-resource jailbreaks succeed by routing harmful content through a geometrically misaligned subspace that projects insufficiently onto the refusal directions, leaving the refusal mechanism intact but untriggered. These findings show that English-only safety evaluations are insufficient; they require accounting for script family, perturbation type, and per-language alignment coverage. The benchmark and analysis code is at this https URL.
[AI-146] Looped State-Space Language Models with Adaptive Exit-State Selection
链接: https://arxiv.org/abs/2607.10110
作者: Zhenxuan Yu,Takeshi Kojima,Yutaka Matsuo,Yusuke Iwasawa
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Recent work on looped language models suggests that many reasoning problems benefit from greater computational depth rather than from additional independent parameters. Existing studies, however, focus almost exclusively on Transformer backbones, leaving open whether this principle also applies to state-space language models. We investigate Looped Mamba and Looped Hybrid Mamba-Transformer architectures, which repeatedly apply a shared Mamba (or hybrid) block to introduce explicit finite-depth recurrent computation. On two controlled reasoning tasks-Mano (modular-arithmetic manipulation) and p-hop induction-Looped Mamba consistently outperforms parameter-matched non-looped baselines and, in several settings, matches or exceeds non-looped models of equal effective depth. We then extend the study to language model pre-training under matched iso-parameter and iso-FLOPs protocols, which jointly disentangle the effects of parameter sharing and effective depth: looped models remain competitive on downstream benchmarks with substantially fewer distinct parameters, although deeper non-looped models retain an advantage in validation perplexity under strict iso-FLOPs comparisons. Finally, we adapt Ouro’s two-stage exit gate to Looped Mamba for threshold-controlled selection among recurrent-step outputs. Since all recurrent steps are still executed, the selected exit step represents prediction depth rather than reduced wall-clock computation. At the scales studied, adaptive exit-state selection improves downstream performance at intermediate depths, while actual inference-time savings require additional state-handling mechanisms.
[AI-147] From ambiguous utterances to governed reuse classes: canonicalization quotient invariance and conditional decidability
链接: https://arxiv.org/abs/2607.10069
作者: Cosimo Spera,Ray Garcia
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Semantic caching defines answer reuse on embedding similarity: two utterances share a stored answer when a similarity score clears a threshold, with no notion of authorization, versioning, or of what makes two demands the same. This note changes the object on which reuse is defined: in a governed domain, reuse should operate on a mathematically characterized quotient of resolved conversational demands, not on a similarity heuristic. Three independently defined relations on resolved utterances – reading identity, resolution identity, and reuse identity – form a refinement chain, strict under realized nondegeneracy conditions checkable on deployment logs; the pipeline’s outputs are invariant along the chain, and reuse identity is exactly the kernel of the resolution map into the governed answer partition, so the reuse quotient is the utterance-side object that partition induces, not a relabeling of it. Reuse identity licenses the governed query key and its certified answer space; reuse of a particular answer requires resolution identity or an applicability certificate. The supporting layer is stated at exactly the strength proved: exact-denotation normal forms; join aggregation as a design operator, with closure-stable cells characterizing no-escape; total computability of the full pipeline relative to an untrusted proposal layer; policy admissibility for arbitrary proposers – and provably not factual grounding or intent fidelity; and elicitation terminating after finitely many informative replies, sound under target consistency.
[AI-148] Agent Abstain: Do LLM Agents Know When Not to Act?
链接: https://arxiv.org/abs/2607.10059
作者: Xun Liu,Yi Evie Zhang,Vira Kasprova,Parisa Rabbani,Pardis Sadat Zahraei,Tianyu Zhang,Ali Ebrahimpour-Boroojeny,Varun Chandrasekaran
类目: Artificial Intelligence (cs.AI)
备注: 56 pages, 13 figures
Abstract:Agent systems based on large language models (LLMs) are increasingly deployed for autonomous tasks, yet existing evaluations mostly focus on task success rather than whether agents know when to abstain. This gap poses real risks: under ambiguity, conflicting constraints, or tool failures, agents may execute unintended and irreversible actions. To close this gap, we present the first systematic evaluation framework for agentic abstention: the calibrated ability of tool-using LLM agents to recognize when not to act. At its core, AgentAbstain is a paired-task benchmark built on an agent-native taxonomy of 8 abstention scenarios across pre-execution reasoning and runtime discovery. It contains 263 paired tasks across 42 executable sandbox environments, where each pair consists of a should-act task and a should-abstain variant produced through a controlled perturbation to the instruction, tool, or environment state. To scale this paired design and resist data contamination, we propose AbstainGen, a fully automated pipeline that synthesizes sandbox environments and generates paired tasks end-to-end, validated by deterministic replay and semantic LLM judges; fresh task instances can be regenerated on demand, and three independent annotators rate 94-98% of sampled tasks as well-designed. Across 17 frontier LLMs in 4 agent harnesses, the best agent (Gemini 3.1 Pro) achieves only 59.5% paired accuracy (correct on both the act and abstain sides of each paired task). More importantly, abstention capability is largely independent of general task-solving capability, indicating that scaling task-solving alone will not close this gap. We further identify failure modes such as post-hoc abstention, in which agents execute irreversible actions before recognizing abstention triggers. Our code and dataset are open-sourced at this http URL.
[AI-149] A Symbolic Neural CPU for Quantization-Simulated Writeback and Interpretable Program Execution
链接: https://arxiv.org/abs/2607.10021
作者: Jose Luis Lima de Jesus Silva
类目: Artificial Intelligence (cs.AI); Hardware Architecture (cs.AR); Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
备注: 63 pages, 23 figures. Includes Supplementary Information
Abstract:Neural networks can learn algorithmic input-output mappings, but trusting a learned executor requires more than a correct final answer because the state transitions that produce it are usually hidden. To make those transitions visible, we introduce a trace-supervised symbolic neural CPU, a factorized learned execution architecture that combines recurrent control, an explicit operation router over a fixed differentiable arithmetic-logic unit bank, destination-masked register writeback, complete trajectory supervision and matched fixed-point replay. The model exposes the selected operation, source and destination registers, register trajectory, memory signals and writeback semantics at every step. On the principal 16-wide benchmark, the non-quantized executor reproduces reference execution exactly, while the eight-bit quantization-simulated executor preserves the symbolic operation path through programs of 1,000 instructions. When the same execution is evaluated against a matched fixed-point replay, the residual numerical drift disappears, showing that it comes from a mismatch between continuous and low-precision reference semantics rather than from execution failure. We compare recurrent, Transformer, temporal-convolution, temporal graph-inspired and state-space controllers, and the ablations show that operation-gate supervision is necessary for an inspectable execution path. Hidden-opcode memory-pressure tasks expose the remaining limits in delayed state use and temporal binding. We also extend the interface with ValueMemory, hybrid adaptive leaky integrate-and-fire controllers, candidate-constrained symbolic control trained through behaviour cloning and actor-critic reinforcement learning, and an RV32I base-integer semantic bridge. Together, these results establish a trace-verifiable framework for interpretable, low-precision and controllable neural execution.
[AI-150] A Production-Oriented Framework for Evaluation of SFX Generation
链接: https://arxiv.org/abs/2607.09973
作者: Mélodie Desbos,Yara Bahram,Eric Granger,Mohammadhadi Shateri
类目: ound (cs.SD); Artificial Intelligence (cs.AI); Audio and Speech Processing (eess.AS); Signal Processing (eess.SP); Systems and Control (eess.SY)
备注: 8 pages main paper, 7 pages appendix, Proceedings of the 29th International Conference on Digital Audio Effects (DAFx26)
Abstract:Industrial sound design requires audio generation systems that not only produce realistic audio, but also preserve the perceptual identity of a reference, support controllable variation, and remain efficient for practical workflows. Existing evaluations are usually tied to text-to-audio (TTA), unconditional, or task-specific settings, limiting assessment for reference-guided sound effects (SFX) variation. To address this gap, we present a production-oriented evaluation framework for structured comparison of heterogeneous audio generation and editing methods. Our framework identifies nine production requirements and explicitly accounts for differences in model capabilities, enabling comparison under a common production objective. A two-stage protocol is introduced: (1) a reference-guided audio-to-audio (ATA) variation task, in which all methods are evaluated under the same ESC-50 SFX adaptation setup, and (2) capability-specific analyses of native operations such as SFX morphing, temporal and energy alignment, inpainting, and targeted editing. This framework combines objective metrics (including FAD, ImageBind-based reference alignment, and diversity across generated variants), together with a human study of perceptual identity preservation and transient diagnosis. Our study reveals complementary strengths and trade-offs across baselines for different production needs. Among the full-generation baselines evaluated under a shared ATA setting, AudioX provides the strongest overall trade-off between reference alignment and diversity while still supporting SFX morphing. Other baselines remain most suitable for specific editing operations. Our framework establishes a structured evaluation and decision protocol for reference-guided SFX variation and provides a practical basis for designing future unified industrial audio generation pipelines. Audio demos are on the accompanying web page.
[AI-151] opoExplore: Topological Discrimination for Archive-Based Exploration
链接: https://arxiv.org/abs/2607.09971
作者: Jason Carlson
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Archive-based exploration methods such as Go-Explore select which visited state to return to using visitation rarity, and frontier methods return to the boundary of the unknown; neither asks whether the unexplored region behind a boundary is enterable at all. Exploration is not just about finding reward - it is about collecting a structurally complete experience for downstream learning and planning. We introduce TopoExplore, which augments Go-Explore cell selection with a periodic topological pass: enclosed unexplored regions (voids) of the visited-set occupancy grid are detected by flood fill (the H1 classes of its cubical complex), and a decaying selection bonus is placed only on their strict entrances (gap or door cells), so sealed regions are never targeted and entered regions retire. On a controlled 18-environment MiniGrid suite (15 seeds, frozen hyperparameters) TopoExplore attains a 1.52x geometric-mean speedup in median steps-to-first-entry over its exact Go-Explore ablation, versus 1.37x for a frontier baseline; frontier exploration degrades when sealed decoy structure appears (0.83-1.48x on decoy environments vs. 1.65-2.11x for TopoExplore), while TopoExplore holds its largest win on hard multi-interaction doors (10.9x). We report an honest negative on Montezuma’s Revenge - without wall knowledge, unreachable occupancy artifacts capture the bonus and performance degrades as it grows, isolating the wall-aware entrance test as the load-bearing component - and a preliminary positive on HM3D scanned buildings, where the speedup over Go-Explore tracks scene difficulty (r=0.69) even as frontier selection dominates blanket coverage. The evidence supports a deliberately scoped claim: topology-aware selection pays off where enclosed structure must be discriminated, and remains competitive at open coverage, where frontier methods are strongest, despite not being tuned for that regime.
[AI-152] Learning in Curved Weight Space:Exponential-Linear Weight Reparameterization for Improved Optimization
链接: https://arxiv.org/abs/2607.09967
作者: Ethan Smith
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 24 pages, 14 figures
Abstract:Many neural networks operations have a multiplicative nature rather than additive: halving or doubling a norm are analogous relatively but require unequal optimization distances when taking linear steps. Adaptive optimizers such as Adam normalize updates per coordinate, but update steps remain additive; weights with very different magnitudes receive similarly sized absolute changes, producing very different relative perturbations. We introduce \textbf\method (\textbf\methodshort), a weight reparameterization for neural networks that combines a sign-aware symmetric-exponential pathway with an identity-like linear pathway. The symmetric-exponential pathway is near-linear for small raw weights but increasingly curved at larger magnitudes. Additive updates in logarithmic space map to magnitude-proportional changes in effective weight space. The linear pathway provides a direct route through the transform that we hypothesize stabilizes optimization, while learnable scale, curvature, and offset parameters control balance between pathways and the curvature of the exponential pathway. These components create a curved parameter-space geometry that empirically improves speed of loss descent over standard linear parameterization. We also identify a useful \emphmismatched initialization: raw weights are chosen so a symmetric version of the transform matches Xavier statistics, but training uses an asymmetric forward transform that leaves positive weights at full strength while making negative weights smaller in magnitude; in small-model ablations, this improves early optimization and may act as a form of symmetry breaking. We train transformers on OpenWebText over nine width \times depth configurations, \methodshort reaches matched validation loss in 1.32–1.49 \times fewer training steps, with the largest widths seeing the biggest gains.
[AI-153] A Foundation Model for Multimodal Event Sequences in Financial Applications
链接: https://arxiv.org/abs/2607.09955
作者: Nikita Rusakov,Vladislav Meshkov,Konstantin Zorin,Gleb Zaripov,Alexander Uglov,Alexey Vasilev,Anton Klenitskiy
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Predictive modeling is a core component of modern financial services, where a wide range of tasks are traditionally addressed using separate models trained on manually engineered tabular features. This task-specific approach limits reuse and makes it difficult to fully exploit heterogeneous data sources such as transaction histories and digital interaction signals. In this paper, we present an approach based on pretraining a foundation transformer model on multimodal sequences of user events. Events from multiple data sources are unified into a single chronological sequence, enabling early fusion of heterogeneous modalities and learning of general-purpose representations via a next-event prediction objective. These representations are combined with existing engineered user features, on top of which lightweight neural models are trained for multiple downstream tasks. The proposed system outperforms traditional task-specific models while reducing development overhead. The approach was deployed in production at one of the biggest banks in Eastern Europe, resulting in measurable improvements in business metrics.
[AI-154] SMETA-ZSL:Semantic Meta-Alignment for Zero-Shot Threat Classification
链接: https://arxiv.org/abs/2607.09936
作者: Ivan Alejandro Montoya Sanchez,Anantaa Kotal,Aritran Piplai
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注:
Abstract:Cybersecurity systems must adapt rapidly to emerging threats. However, labeled data for new threat categories is unavailable when those threats first appear. Generalized zero-shot learning offers a natural solution by enabling recognition of unseen classes through auxiliary semantic knowledge rather than labeled examples. Large language models are particularly promising in this setting because they can convert unstructured CTI reports into semantic prototypes for emerging threats. However, applying language-driven zero-shot learning to cybersecurity is difficult due to strong semantic overlap between threat descriptions, heterogeneity between behavioral attributes and text, severe class imbalance, and open-set conditions where unseen threats are unknown during training. We propose SMETA-ZSL, that learns semantic prototypes from overlapping language descriptions through contrastive finetuning, aligns behavioral features through episodic meta-learning and knowledge distillation, and performs adaptive routing for generalization across seen-unseen classes. Across 7 benchmarks, SMETA-ZSL delivers the strongest overall generalized zero-shot performance under the strictest inductive setting, surpassing prior methods by 10.8 points on average, with gains up to 18.1 points. Github:this https URL
[AI-155] Do These Violent Delights Have Violent Ends? Measuring the Post-Merge Fate of Agent ic Code
链接: https://arxiv.org/abs/2607.09902
作者: Chunqiu Steven Xia,Courtney Miller
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:Agentic coding tools are increasingly used to make autonomous repository-level changes to real-world projects. Prior work has largely evaluated these contributions at the pre-merge stage, through outcomes such as pull request acceptance and review effort. Far less is known about what happens to agentic code post-merge. Yet merge success alone does not reveal whether a contribution will remain stable or require bug fixes and other corrective maintenance downstream. We conduct a longitudinal empirical analysis of agentic and human contributions across 182 repositories, tracking their post-merge fate over time, characterizing the intent of subsequent modifications, and analyzing the defects and vulnerabilities they introduce. While the overall maintenance rates are similar, agentic contributions require significantly higher rates of corrective maintenance and introduce more security weaknesses and dependency vulnerabilities. We also find statistically significant evidence that agentic maintenance burden is associated with repository characteristics. In particular, each 10 percentage-point increase in a project’s no-review rate is associated with roughly a 6% increase in agentic maintenance burden on average. As coding agents become pervasive in software development, our findings highlight the need to evaluate and design agentic tools not only to produce mergeable changes, but to produce contributions that remain secure and maintainable.
[AI-156] What You Train Is What You Get: Gender Bias Training Composition and Post-Hoc Mitigation in Audio Deepfake Detection
链接: https://arxiv.org/abs/2607.09891
作者: Aishwarya R. Fursule,Vamshi Nallaguntla,Shruti Kshirsagar,Anderson R. Avila
类目: ound (cs.SD); Artificial Intelligence (cs.AI)
备注: This manuscript is a preprint and is currently under review for the Special Section on Trustworthy and Reliable AI in IEEE Transactions on Reliability
Abstract:Audio deepfake detection models determine whether speech is genuine or artificially generated, but high overall accuracy can mask substantial performance disparities across demographic groups. In this work, we investigate gender bias in audio deepfake detection using the ASVspoof5 dataset. We use ASVspoof5 under a controlled custom split designed to isolate gender-composition effects. We train attack-specific models on nine training sets with different gender compositions, ranging from female-only to male-only. We use a ResNet18 classifier with LogSpectrogram and WavLM-Base+ features, and we evaluated six post-hoc threshold calibration methods. Experimental results show that training data composition strongly predicts bias direction, with the underrepresented gender performing worse at test time. WavLM-Base+ features are shown to produce gender performance gaps 3.0 to 4.3 times larger than LogSpectrogram under identical training conditions, and balanced training is found to reduce LogSpectrogram bias but leave WavLM bias largely intact. Moreover, all six calibration strategies, including Oracle calibration with full test-set label access, leave the Equal Error Rate gap unchanged at 1.317 pp, confirming that threshold adjustment cannot correct underlying score distribution disparities. Overall, these findings suggest that gender fairness in audio deepfake detection must be addressed at training time, as post-hoc methods can only partially mitigate the resulting disparities
[AI-157] Robo-ValueRL: Reliable Value Estimation for Offline-to-Online Reinforcement Learning
链接: https://arxiv.org/abs/2607.09866
作者: Wenke Xia,Pei Ren,Wenbo Yu,Yizhuo Zhang,Jifan Li,Yixue Zhang,Yinuo Zhao,Qingyang Gao,Jianlong Fu,Jian Tang,Ji-Rong Wen,Zhengping Che,Di Hu
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: Please refer to our website: this https URL
Abstract:Offline-to-online reinforcement learning is promising for generalizable robotic manipulation, yet its full-stack complexity obscures reproduction and diagnosis. Within such systems, value estimation plays a central role in prioritizing heterogeneous data for policy improvement. Despite its importance, the central question remains underexplored: how value-function reliability shapes policy optimization in offline-to-online reinforcement learning. To answer this question, we propose Robo-ValueRL, a unified framework that enables reliable value estimation and systematically traces its downstream effects on policy pretraining and online improvement. Concretely, Robo-ValueRL learns a history-conditioned value estimator and evaluates its reliability through global-progress and local-preference metrics. These resulting value estimates are propagated into quality-conditioned consistency-policy pretraining and a residual adaptation module on online rollouts, providing a unified testbed for analyzing how value reliability shapes downstream policy performance. Across 240 hours of offline demonstrations and over 3,000 online rollout trajectories, our extensive experiments show that downstream performance is strongly associated with value reliability. Reliable value functions provide better action-quality estimates, allowing value-guided offline RL to scale more effectively than quality-agnostic behavior cloning, and stabilize online improvement by prioritizing high-quality rollout data. Integrating reliable value guidance through offline pretraining with online improvement, our system achieves 86% success on millimeter-level precise chip insertion and 84% on generalizable block disassembly. We hope these findings highlight the importance of value-guided data utilization for effective policy improvement from heterogeneous robotic experience.
[AI-158] More Structure Not More Capacity: Object-Centric Representations for Visuomotor Imitation Learning
链接: https://arxiv.org/abs/2607.09825
作者: Yi Li(TU Darmstadt),Alexandre Chapin(LIRIS),Liming Chen(LIRIS),Jan Peters(TU Darmstadt),Alap Kshirsagar(IIT Delhi, ADU)
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Robotic manipulation policies rely on pre-trained vision models that give either a global scene embedding or a dense patch grid. Both mix task-relevant and task-irrelevant features. Object-centric slot representations are a structured alternative: they group features into a few per-object slots. We test what this structure buys on ManiSkill3 PickCube-v1, with a frozen encoder and a held-out-seed evaluation. Holding the policy, goal token, rendering, and calibration fixed and changing only the encoder, a frozen object-centric SPOT representation (DINO ViT-B/16 + Slot Attention) reaches 55.0 \pm 2.9% success, 22.4% above a dense DINO global-feature baseline (32.6 \pm 1.5%), with the same trainable policy and no encoder fine-tuning. More tokens alone do not help: a dense patch grid with 16x the tokens performs no better than the global feature. Adding an explicit 2D spatial goal and native-resolution rendering raises the full system to 68.7 \pm 4.2%, just below a privileged 3D-oracle upper bound (71.7 \pm 4.1%). An automated kinematic failure taxonomy then separates spatial-precision (Near-Miss) failures from object-tracking (No-Grasp) failures: spatial grounding reduces Near-Miss while leaving No- Grasp unchanged. The same taxonomy transfers to the harder StackCube-v1 and points to occlusion as the main bottleneck.
[AI-159] An Autonomous Scientific Knowledge Generation Framework for AI-Driven Scientific Discovery
链接: https://arxiv.org/abs/2607.09806
作者: Dibakar Datta
类目: Digital Libraries (cs.DL); Materials Science (cond-mat.mtrl-sci); Artificial Intelligence (cs.AI)
备注: 32 pages, 6 figures
Abstract:Artificial intelligence (AI) is transforming scientific discovery, but its effectiveness is fundamentally limited by the availability of structured scientific knowledge. Although existing databases have accelerated data-driven materials research, much of the knowledge needed for predictive modeling and inverse design remains embedded in unstructured scientific literature. We present an Autonomous Scientific Knowledge Generation Framework that transforms scientific publications into a Unified AI-Ready Scientific Knowledge Base. The framework integrates ontology-guided literature acquisition, hybrid scientific knowledge extraction, semantic harmonization, knowledge fusion, and validation within a unified workflow. Rather than treating literature retrieval, information extraction, and database construction as separate tasks, the framework progressively converts scientific publications into structured, semantically consistent, and provenance-preserving knowledge suitable for AI-driven reasoning. As a proof of concept, the framework was applied to electro-optic materials. Autonomous literature acquisition retrieved and validated about 1,000 publications from multiple scholarly repositories. A representative subset of eight publications was processed through the complete workflow, generating 29 structured scientific records that were harmonized into 7 canonical scientific records. The results demonstrate the complete transformation from scientific literature to an AI-ready scientific knowledge base while preserving quantitative measurements, operating conditions, provenance, and scientific context. The proposed framework provides a scalable, domain-independent foundation for predictive AI, generative AI, and closed-loop AI-driven scientific discovery.
[AI-160] rivial Prompt Reframing Bypasses Safety Guardrails in Googleś MedGemma-4B
链接: https://arxiv.org/abs/2607.09804
作者: Avi-ad Avraam Buskila
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:Open-weight medical language models are increasingly used as the base of patient-facing and clinician-support applications. Their model cards prohibit specific behaviors – recommending exact drug dosages, issuing definitive diagnoses, prescribing treatments, adjudicating drug-drug interactions, and advising that emergency care can be skipped – yet a model card describes intended behavior, not robust behavior. We quantify that gap for MedGemma-4B-it under attacks that require no technical sophistication. We build a fully factorial benchmark of 5 guarded-behavior concepts x 50 deterministically templated questions x 6 lay-accessible attack manners x 3 repetitions (4,500 generations), serve the model locally through Ollama under default sampling, and code every response refuse/hedge/comply with three independent judges (an LLM judge, a transparent regex judge, and an NLI-entailment judge). Under the primary LLM judge the overall Attack Success Rate (ASR, the fraction coded comply) is 38.0%. The two framings that reinterpret the request as legitimate dominate: recasting a question as a “medical board exam” item raises ASR from a 29.0% baseline to 53.1% (+24.0 points), and an appeal to an alleged doctor’s authority raises it to 43.7% (+14.7); crude instruction-override prefixes have no significant effect. Robustness is dominated by topic: the drug-interaction guardrail is nearly absent (83.2% ASR) while the emergency-deferral guardrail is strong (4.7%) – and the authority framing is the only attack that breaches it. We report Wilson confidence intervals, cluster-bootstrap effect sizes, a cluster-robust logistic regression, Cochran’s Q, per-manner McNemar tests, and inter-judge reliability (Fleiss’ kappa = 0.26); absolute ASR is judge-dependent while the ordering of attacks and topics is not. Our findings motivate stronger deployment-time guardrails for open medical models.
[AI-161] JEPA for AI-Native 6G: Predictive Representations and Open Challenges
链接: https://arxiv.org/abs/2607.09798
作者: Sheikh Salman Hassan,Irshad A. Meer,Almoatssimbillah Saifaldawla,Yan Kyaw Tun,Mustafa Ozger,Madyan Alsenwi,Nguyen Van Huynh,Woong-Hee Lee,Cedomir Stefanovic,Mathini Sellathurai,Henk Wymeersch,Tharmalingam Ratnarajah
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Networking and Internet Architecture (cs.NI)
备注: 14 pages, 4 figures, 3 tables. Tutorial and review on Joint-Embedding Predictive Architecture (JEPA) for AI-native 6G. Submitted to IEEE Communications Magazine
Abstract:Sixth-generation (6G) networks are moving toward AI-native operation, where learning modules are embedded across the radio access network (RAN), edge, and core. This transition requires learning from limited labels, heterogeneous wireless and network data, partial observations, non-stationary propagation, and latency-constrained control loops. Joint-embedding predictive architecture (JEPA) is a promising self-supervised paradigm for this setting because it predicts missing or future representations in latent space instead of reconstructing raw measurements or using contrastive negative samples. This article presents a wireless-oriented tutorial on JEPA for 6G intelligence. We define the JEPA training mechanism, describe how CSI, beam measurements, KPIs, topology graphs, and sensing observations can be tokenized and masked, and position the learned encoder as a predictive representation layer for RAN, O-RAN, edge, and core functions, with task-specific heads or controllers producing final decisions. Then we present an illustrative, beam-management case study suggesting that a wireless-aware target, specifically an auxiliary future beam-energy target during self-supervised pretraining, can improve label efficiency and robustness across shifted deployment conditions relative to a supervised source domain. Finally, we outline open challenges in multi-timescale prediction, action-conditioned modeling, distributed training, trustworthiness, efficient deployment, benchmarking, and standardization.
[AI-162] Large Multimodal Model-Based Environment-Aware Mobility Management
链接: https://arxiv.org/abs/2607.09795
作者: Seokhyun Jeong,Sangmok Shin,Seungnyun Kim,Jiao Wu,Byonghyo Shim
类目: Information Theory (cs.IT); Artificial Intelligence (cs.AI); Signal Processing (eess.SP)
备注:
Abstract:Recently, large language models (LLMs) have been successfully adopted in various fields, including wireless communications, robotics, and autonomous vehicles, owing to their outstanding adaptability and reasoning abilities. Despite their huge potential, the application of LLMs for mobility management is relatively scarce since it requires not only analyzing wireless measurements but also predicting dynamic user trajectories and making real-time handover decisions across densely deployed small base stations (SBSs). In this paper, we propose an environment-aware mobility management scheme based on large multimodal models (LMMs), which extend capabilities of LLMs to process multimodal sensing data. By leveraging LMMs, the proposed scheme extracts contextual information on the surrounding environments from RGB-D images to capture user equipment (UE) mobility patterns and identify signal reflections and blockages caused by static reflectors and dynamic obstacles. Using the extracted environmental information, the proposed scheme learns the intrinsic mapping from UE and SBS positions to channel capacity, referred to as channel capacity map (CCM), from which future channel capacities along UE trajectories are predicted. Based on the predicted channel capacities, we determine proactive handover decisions maximizing the cumulative channel capacities. Simulation results demonstrate that the proposed scheme achieves substantial channel capacity improvements over conventional deep learning (DL)-based approaches.
[AI-163] A Comprehensive Survey and Systematic Real-World Evaluation of Embodied Vision-and-Language Navigation
链接: https://arxiv.org/abs/2607.09792
作者: Liuyi Wang,Kai Sheng,Zongtao He,Jinlong Li,Yongrui Qin,Haojie Dai,Xiangyi Wang,Jingwei Yang,Qingqing Yan,Chengju Liu,Qijun Chen
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: This paper has been accepted by IEEE TASE 2026
Abstract:Navigation is a fundamental capability of autonomous systems, yet most existing approaches rely on highly structured models and strong prior assumptions, limiting their robustness in open and uncertain real-world environments. Vision-and-Language Navigation (VLN) offers a promising direction by enabling robots to integrate natural language understanding with visual perception in a data-driven manner. Although VLN has attracted increasing research attention, systematic methodological taxonomy and real-world validation remain limited. This survey presents a comprehensive review of VLN research. Specifically, state-of-the-art methods are organized along two orthogonal dimensions: action paradigms, including hierarchical and monolithic frameworks, and model paradigms, including discriminative and generative approaches. A critical analysis of their respective strengths and limitations is provided. Additionally, we conduct a systematic real-world evaluation of representative VLN system configurations on a physical robotic platform. Experiments across ten diverse real-world scenes show a substantial performance gap between simulation and real-world deployment under the tested configurations: a representative monolithic RGB-only method achieves 61% success in simulation but drops to 22% in real-world deployment, while a hierarchical framework achieves a higher real-world success rate of 51%, suggesting stronger robustness in our evaluation setting. Finally, we highlight key challenges in perception, decision-making, and control that must be addressed in future research.
[AI-164] Semantic Drift and the Stability of Operator Control in Reasoning -Class Decision Support Systems
链接: https://arxiv.org/abs/2607.09790
作者: M. L. Kaluzhsky,V. A. Efirov
类目: Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注: 5 pages, 1 figure, 1 table
Abstract:The article investigates the fundamental problem of ensuring the stability of operator control and preserving goal-targeting in hybrid human-machine decision support systems (DSS) of a new generation. Based on a two-month continuous longitudinal experiment on the joint design of a monograph-format textual array, the latent phenomenon of semantic context drift in large language models of deep logical reasoning (Reasoning LLMs) is verified and described. A mathematical model of interaction in the human-machine interface is proposed, and an original metric is introduced - the operator control stability coefficient, which takes into account the non-linear contextual pressure of hidden reasoning chains. Within the paradigm of the cognitome theory, a critical point of control functions inversion is captured. Engineering recommendations are formulated for implementing dynamic relational arbitration loops based on a modified hierarchical similarity model.
[AI-165] PHITSBench: an execution-scored benchmark for AI-assisted PHITS radiation-transport input generation using natural language
链接: https://arxiv.org/abs/2607.09789
作者: Xianglin Ji,Svetlana V. Boriskina
类目: Artificial Intelligence (cs.AI); Materials Science (cond-mat.mtrl-sci)
备注:
Abstract:We introduce PHITSBench, an execution-scored benchmark for the Monte Carlo Particle and Heavy Ion Transport code System (PHITS). PHITSBench comprises 282 transport-scorable tasks spanning three common workflow categories: parameter editing (Edit), syntax repair (Repair ), and complete simulation generation from natural-language descriptions (Reproduce). Each task is evaluated using a Composite Metric Score that combines execution success with agreement between generated and reference transport observables. Using PHITSBench, we evaluate five GPT-5.4-based configurations ranging from zero-shot prompting to knowledge-augmented and agentic workflows. Without domain-specific knowledge, the model performs well on editing and repair tasks (95% and 70% success, respectively) but fails to generate correct simulations from scratch (0% success on the Reproduce track). A structured, machine-readable PHITS knowledge catalog, supplied alongside the user manual, raises single-shot Reproduce-task success to 57%. Agentic execution provides a further improvement to 66-73%, but at increased computational cost. Failure analysis shows that the remaining errors are dominated by incorrect selection and configuration of physical observables rather than syntax generation. These results suggest that future progress in AI-assisted radiation-transport modeling will depend as much on machine-readable knowledge bases, curated domain-training datasets, and execution-grounded evaluation environments as on advances in foundation models themselves.
[AI-166] From Patterns to Maze Structures: SMT-Based Path Synthesis and 2D/3D Construction
链接: https://arxiv.org/abs/2607.09781
作者: Shengyi Wang
类目: Artificial Intelligence (cs.AI); Logic in Computer Science (cs.LO)
备注: 14 pages, 7 figures
Abstract:We present a pipeline for constructing maze structures from input patterns such as text or shapes. The central path-synthesis problem is encoded in Satisfiability Modulo Theories as global constraints on adjacency, continuity, and pattern-constrained coverage, allowing each fixed-bound instance to be solved in one call. The resulting path is either a planar, self-avoiding route or a layered traversal with prescribed over–under crossings, and it serves as a scaffold for constructing planar mazes and three-dimensional realizations of woven mazes. This report extends the published Bridges 2026 conference paper with more representative SMT-LIB examples and a fuller account of how synthesized paths become concrete maze constructions in planar and three-dimensional form.
[AI-167] Maximizing Human Efficiency in Large-Scale Robot Post-Training via VLAC-Cut Guided Pipeline
链接: https://arxiv.org/abs/2607.09776
作者: Shaopeng Zhai,Qi Zhang,Tianyi Zhang,Haoran Zhang,Fuxian Huang,Zhanhui Lin,Zijun Xu
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:When adapting Vision Language Action (VLA) models to downstream tasks, multiple rounds of post training are required because a single round of data cannot resolve all issues, making continuous iterations necessary to progressively address the weaknesses exposed in previous rounds. In this report, we aim to maximize human efficiency during post-training, defined as the policy improvement and task throughput achieved per unit of human labor and time. We propose a human-efficient post-training pipeline that enables a small number of human operators to supervise multiple robots. The pipeline is built around a specialized division of labor: a trained Teleoperator focuses on high-value remote interventions and recovery demonstrations, while a Floor Operator monitors multiple robots, triggers takeovers, and performs physical resets. This role specialization reduces task switching, lowers operator training costs, and allows limited human labor to supervise more robot interaction across a larger fleet. To improve data utilization efficiency, we introduce VLAC-CUT as an automatic rollout curation tool. It segments autonomous robot trajectories into progress-making, idle, failure-inducing, and recovery portions, preserving useful segments while filtering harmful or uninformative ones. The curated rollout data are combined with Human-in-the-Loop data for the next post-training round. We validate the proposed pipeline on four real-world manipulation tasks. Across iterative post-training rounds, the final policies achieve 80%–95% success rates and improve task throughput by 1.7 \times --4.2 \times over the base model. Under the same human-intervention budget, VLAC-CUT guided rollout reuse outperforms HITL-only training in both success rate and throughput. Subjects: Robotics (cs.RO); Artificial Intelligence (cs.AI) Cite as: arXiv:2607.09776 [cs.RO] (or arXiv:2607.09776v1 [cs.RO] for this version) https://doi.org/10.48550/arXiv.2607.09776 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-168] OmniSCS: Omni Safety-Critical Scenario Synthesis for Autonomous Driving via a Fully Editable Driving World
链接: https://arxiv.org/abs/2607.09764
作者: Xiaoyun Dong,Qian Xu,Yang Lu,Yang Lou,Yung-Hui Li,Jianping Wang
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:The synthesis of safety-critical scenarios (SCS) and their evaluation through closed-loop simulations are crucial for developing robust autonomous driving systems. A key aspect of this process involves editing agent states in both appearance and trajectory levels within existing scenes. However, current methods struggle to preserve data fidelity after scene editing and fail to efficiently generate high-quality SCS through such modifications. To overcome these limitations, we propose OmniSCS, an innovative system that generates photorealistic SCS with high physical fidelity while enabling closed-loop testing in synthetic environments. OmniSCS comprises two key modules: 1) A Fully Editable Driving World Construction module that maintains high-fidelity agent appearance and background during scene editing via dual-strategy agent reconstruction and depth-refinement background reconstruction methods. 2) A SCS Synthesis module that facilitates object insertion and agent trajectory editing to synthesize diverse SCS while preserving data fidelity. Experiments on nuScenes, Waymo, and KITTI datasets show that OmniSCS outperforms state-of-the-art methods in edited scene fidelity. We further validate its ability to enhance autonomous driving algorithms and support real-time (13Hz) closed-loop testing. Overall, OmniSCS provides a safer, more effective, and cost-efficient solution for SCS optimization and testing in autonomous driving.
[AI-169] BatteryLake: Agent ic Physics-Grounded Curation of Heterogeneous Battery Aging Data and Benchmarking
链接: https://arxiv.org/abs/2607.09762
作者: Tianwen Zhu,Hao Wang,Yonggang Wen
类目: Artificial Intelligence (cs.AI); Databases (cs.DB)
备注: The platform, benchmark, and curation protocol are publicly available at this https URL
Abstract:Public battery aging datasets are a critical asset for advanced health management, but their practical use is often limited by inconsistent formats, unclear schemas, and metadata scattered across repositories and publications. Current curation remains largely manual and hard to reproduce, while general-purpose data integration tools miss the domain-specific semantics of electrochemical time-series data. We present BatteryLake, a governed data lakehouse that turns raw public battery data into benchmark-ready assets through an agentic, physics-grounded curation framework, with three contributions. First, LLM agents extract metadata and synthesize dataset-specific converters, grounding every output in verbatim evidence and abstaining when none supports a value. Second, a human-in-the-loop mechanism frames verification as selective prediction and gates admitted data through 26 schema, statistical, and physical-plausibility rules. Third, we release an open benchmark of 41 datasets from over 25 institutions, with standardized SOH and RUL tasks, three split protocols, and eight baseline model families. The platform, benchmark, and curation protocol are publicly available at this https URL.
[AI-170] LegalFarePlan: A Label-Setting Framework for Fare-Transparent Urban Rail Route Planning under Non-Additive Fare Rules
链接: https://arxiv.org/abs/2607.09755
作者: Tanghui Li
类目: Artificial Intelligence (cs.AI)
备注: 6 pages, 4 tables; author preprint version, not the conference camera-ready version
Abstract:Urban rail fare systems may be non-additive: the fare of a single paid journey from an origin to a destination can differ from the sum of fares over multiple legally separated journey legs. This paper presents LegalFarePlan, a fare-transparent route-planning framework that models legal exit-and-reentry operations as explicit, auditable constraints. Given a transit network, fare function, transfer rules, station-level exit/re-entry costs, an extra-time budget, and a split limit, the planner computes explainable route plans over paid journey segments. The artifact implements Dijkstra shortest-time and direct route-planner baselines, a greedy split heuristic, bounded exact label-setting, and Pareto-frontier search. Evaluation uses controlled synthetic data and a 57-station semi-synthetic benchmark with 360 OD pairs. On the semi-synthetic benchmark, bounded exact search identifies positive modeled fare reductions for 71.11% of OD pairs, with mean reduction 3.78 and maximum reduction 9.0 synthetic fare units under a 45-minute extra-time budget. These results demonstrate method behavior and reproducibility; they are not empirical conclusions about MTR or any transit operator.
[AI-171] ask-Conditioned Synthetic Data Generation for Improving Machine Learning Performance in Agricultural Prediction Tasks
链接: https://arxiv.org/abs/2607.09751
作者: Hamid Ebrahimy,Moritz Lucas,Martin Atzmueller
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Machine Learning (ML) algorithms have been widely used to estimate agricultural variables across diverse contexts. However, because the quantity and quality of training data strongly influence performance of ML algorithms, their use can be constrained by limited or incomplete reference data. Synthetic Data Generation (SDG) offers a practical approach to address this issue by producing artificial but realistic samples that preserve key characteristics of the original data. Building on teacher-student knowledge transfer and in-context learning for tabular data, this study proposes a Task-Conditioned SDG (TCSDG) algorithm that pairs a Bayesian Network generator with a transformer-based tabular foundation model (TabICL). The proposed algorithm was evaluated on two agricultural prediction tasks: crop yield prediction and crop type classification. Six benchmark SDG algorithms were also utilized to compare their performance with that of TCSDG. Across twelve study sites, two training-data fractions, four multiplication ratios, and three predictive ML algorithms, augmenting the original data with TCSDG-generated synthetic data improved ML performance in 89% of the crop type classification experiments and 74% of the crop yield prediction experiments. TCSDG also substantially outperformed benchmark SDG algorithms and was the only method to consistently improve ML performance across both tasks at the aggregate level. The study demonstrates that carefully designed and processed synthetic data can improve ML performance in precision-agriculture applications. TCSDG offers a practical and extensible framework for generating synthetic data that supports downstream ML agricultural prediction. The full implementation of TCSDG is publicly available as open source at this https URL.
[AI-172] SupplyNetPy: An Open-Source Python Library for High-Fidelity Modeling and Simulation of Arbitrary Supply Chain and Inventory Networks
链接: https://arxiv.org/abs/2607.09745
作者: Tushar Lone,Neha Karanjkar
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 14 pages, 6 figures, Winter Simulation Conference 2026
Abstract:This paper introduces SupplyNetPy, an open-source, well-documented Python library for modeling and discrete-event simulation of supply chain networks with arbitrary multi-echelon structures. It supports multiple replenishment policies, perishable inventory, node disruptions, and stochastic demand and lead times. All components are extensible via inheritance. Users describe a supply chain as a graph with node and link attributes, while the library handles simulation, providing logs and extensive node and network level performance reports. This paper presents the motivation, design, key features, and architecture of SupplyNetPy, along with detailed validation results (against analytical benchmarks, a commercial tool, and a published case study). A key motivation behind SupplyNetPy’s development is programmatic generation and simulation of complex models, enabling design-space exploration, what-if analysis, training data generation, and supply chain digital twins.
[AI-173] A Theory of Least Autonomy in AI
链接: https://arxiv.org/abs/2607.09744
作者: Christophe Parisel
类目: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注:
Abstract:Least privilege, the principle that an identity should hold only the permissions strictly required for its task, has been a foundational primitive of access control for decades. We argue that this principle is insufficient for agentic AI systems, which do not merely hold permissions but can combine, approve, and amplify them across workflows and system boundaries. We propose least autonomy as an appropriate generalization and develop a formal theory. First, we define a compositional blast radius d(a,b) that measures structural separation between actions in an enterprise hierarchy, combining an ultrametric tree with lattice-valued confidentiality, integrity, and control-context labels. Second, we define a directed agent influence graph G(theta). An arc from U to V requires a directed shared-resource write-to-read meeting or a conservative undirected agent-to-agent (A2A) communication meeting, and a meeting-conditioned influence potential at or above an externally selected policy threshold theta. A catalogue-radius profile supports calibration and audit of theta. Finally, we define a collusion predicate over graph reachability that detects authorization composition, decision manipulation, and cross-domain capability composition. Subjects: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR) Cite as: arXiv:2607.09744 [cs.AI] (or arXiv:2607.09744v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2607.09744 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-174] Scaffolding the Strategist: Architecture-Dependent Reasoning Interventions in Hotelling Spatial Markets ICLR2026
链接: https://arxiv.org/abs/2607.09743
作者: Pratyush Singh
类目: Artificial Intelligence (cs.AI); Computer Science and Game Theory (cs.GT)
备注: 26 pages (11 main + 15 appendix), 6 figures, 4 tables. Accepted at the ICLR 2026 Workshop on LLM Reasoning
Abstract:We investigate whether structured reasoning interventions improve the strategic economic reasoning of large language models, and whether their effects depend on model architecture. Using Hotelling’s linear city model as a diagnostic vehicle, we evaluate GPT-4.1-mini (a standard instruction-following model) and GPT-5-mini (a reasoning-optimized model) under five conditions - an unscaffolded baseline and four reasoning interventions - across eight questions spanning deductive and abductive reasoning, three prompt framings, and three repetitions per condition, yielding 720 individually judged responses. We find a statistically significant crossover interaction between scaffolding type and model architecture ( t(7) = 4.79 , p = 0.002 , d = 1.69 ): commitment scaffolding improves the standard model ( +0.21 ) while degrading the reasoning model ( -0.63 ), and principled separation shows the opposite pattern ( -0.40 vs. +0.31 ). Both crossovers are individually significant (commitment: p = 0.040 ; separation: p = 0.002 ) and hold across all eight questions with 7/8 directional consistency. Adversarial stress-testing harms both models, with 2.6\times greater degradation for the reasoning model ( -1.47 vs. -0.57 ; p = 0.038 ), and the damage correlates negatively with baseline difficulty ( R^2 = 0.36 , p = 0.014 ). We further document a persistent declarative-procedural gap in which both models identify correct strategies at rates far exceeding their ability to execute them; separation fully closes this gap for the reasoning model while no intervention helps the standard model.
[AI-175] AGM-like Paraconsistent Partial Meet Abductive Expansion Operation
链接: https://arxiv.org/abs/2607.09729
作者: Ulisses Franceschi Eliano
类目: Artificial Intelligence (cs.AI); Logic in Computer Science (cs.LO)
备注:
Abstract:In his 1996 doctoral thesis, Maurice Pagnucco created the first AGM-like abductive expansion operation. Taking his operation as a basis, as well as a taxonomy – inspired by Atocha Aliseda – responsible for highlighting and formalizing the main components of abductive reasoning, the main aim of this paper is to present a new paraconsistent AGM-like abductive expansion operation – capable of assimilating contradictory explanatory hypotheses without trivialization and the consequent absurd epistemic state – with its postulates and its transitively relational partial meet construction. To a large extent, the formal development presented in this paper was only made possible by the recent creation of the paraconsistent logic RCbr, an LFI (Logics of Formal Inconsistencies) that establishes properties especially relevant to belief revision contexts, in particular, the ability to be self-extensional – i.e., to satisfy the replacement property. This is the first of two papers: the paraconsistent abductive expansion operation announced here – which is part of a new system called AGMpabd – despite bringing many interesting features, does not assign any relevant epistemic role to the paraconsistent operators of negation and consistency. Only in a second paper will an analogous paraconsistent abductive expansion operation – which is part of another new system, AGMcircabd – be enhanced in this direction. Nevertheless, to the best of my knowledge, the operation developed in this paper is the first of its kind in the AGM literature.
[AI-176] he Verifier is the Curriculum: Execution-Gated Self-Distillation for Cross-Family Game Generation
链接: https://arxiv.org/abs/2607.09709
作者: Chenyu Zhou,Qiliang Jiang,Shuning Wu,Xu Zhou
类目: Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注:
Abstract:Post-training a code generator against a learned judge can optimize proxy features that raise the score without improving the artifact. We study the opposite signal: a deterministic, judge-free, ungameable filter – whether a generated project launches cleanly under a headless engine (strict-launch). Under this gate, rejection-sampling self-distillation compounds out-of-family generalization. On GameCraft-Bench (mapping a natural-language brief to a complete Godot project), a 14B model (Qwen3-14B+LoRA) distilled under strict-launch raises clean generation on four unseen game families from 8.8% to 42.2% per-candidate and best-of-K coverage from 18/25 to 25/25 (the gold ceiling) over three rounds, each a significant gain (p=0.0019, p1e-4, p1e-4). The gain is not from merely adding data: an exactly-matched gold-duplication control regresses below the base model (5.6% vs. 8.8%, p=0.019), while a count-matched decomposition splits the round-1-to-2 jump into comparable quality (+8.8pp) and quantity (+8.5pp) channels. Most directly, rerunning the loop with only the filter swapped – the lenient BUILD check, which passes 99.9% of generations, in place of the launch gate – erases the gain entirely (back to base, p=1e-3 vs. the launch-gated round), isolating verifier precision rather than the optimizer. A second ungameable signal, headless execution grounding, rises monotonically across rounds and yields far more grounded candidates than gold-duplication at a matched budget (16 vs. 5), confirming the gains are functional, not launch-but-empty. Game generation is a verifiable testbed for one lesson: the verifier is the curriculum – what it certifies is what the model learns.
[AI-177] GES-TSP: Graph Edge Sparsification for TSP
链接: https://arxiv.org/abs/2607.09708
作者: Tianfeng Chen,Xianyue Li
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Combinatorics (math.CO)
备注:
Abstract:Solving large-scale instances of the Traveling Salesman Problem (TSP) exactly is computationally expensive. Researchers often employ graph sparsification methods to improve computational efficiency. Traditional sparsification methods typically rely on fixed heuristics and fail to fully exploit instance-specific structural information. In this paper, we propose Graph Edge Sparsification (GES), a learning-based sparsification approach for Euclidean TSP. By incorporating geometric structural information and combinatorial optimization technology, our proposed method adaptively generates a sparsification graph for different instances, significantly reducing the graph size and accelerating the solving process. Experimental results demonstrate that our sparsification method can prune up to 95% of edges on the MATILDA dataset, while keeping the solution gap within 1% of the optimal value. Moreover, our approach exhibits strong generalization capability on the TSPLIB this http URL some large-scale instances, the pruning rate exceeds 99%, while the optimality gap remains below 1%.
[AI-178] YUKTI: From Natural-Language Situations to Robust Verifiable Decisions An Uncertainty-Typed Proposition IR Assumption-Robust Pareto Frontiers and a Regret Certificate
链接: https://arxiv.org/abs/2607.09706
作者: Suyash Mishra
类目: Artificial Intelligence (cs.AI); Computer Science and Game Theory (cs.GT); Machine Learning (cs.LG)
备注: 19 Pages , 21 figures
Abstract:Language models turn a worded situation into a numeric plan, and the dominant pipelines (NL4Opt, OptiMUS, ORLM, OR-LLM-Agent) commit to a single objective and point-valued coefficients, then solve once. For decisions that allocate real budget, effort, or clinical attention, that confidence is the failure mode: every objectified number is an assumption, and a plan optimal only if the guesses are exactly right is fragile – mimicry of computation. YUKTI changes the target of autoformulation. Its representation is a typed-proposition graph whose relationships carry shape priors, coefficient uncertainty, and provenance. YUKTI routes each stage to an exact, nonlinear, or evolutionary solver; couples stages by a distributional Pareto hand-off; and introduces Assumption-Robust Pareto Frontiers (ARPF), resampling assumptions (including structural epsilon-contamination) to score how often each action survives (rho). We prove a bound making rho an exact factor of decision regret, add auditable traceability, and synthesize a benchmark-faithful data foundation when none exists (SRJANA). We validate three ways: under controlled misspecification the robust compromise cuts mean and tail regret by over 90% versus a naive point plan; on a regulated commercial decision we optimize inside a lawful action space and price the downside in euros; and on a real public dataset of 41,188 decisions an out-of-sample backtest beats the logged status quo by 34% and a naive point rule by 4% while reducing the optimizer’s curse. The solvers are standard; we claim no benchmark-SOTA win. A head-to-head shows an LLM given the correct numbers, and single-objective optimization, both incur about 47x the held-out regret of YUKTI – an LLM is a formulator, not a solver. Under long-range causal coupling, the forward hand-off becomes unsound, locating where it must become a backward-induction causal policy.
[AI-179] Model Collapse: On Recursion Noise and Uncharted Machine Visions
链接: https://arxiv.org/abs/2607.09705
作者: Violaine Boutet de Monvel(LIRA, IRCAV)
类目: Other Computer Science (cs.OH); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Machine Learning (cs.LG)
备注: Preprint for Ethics and Aesthetics of Artificial Intelligence (Naples: Orthotes Edizioni), October 2025
Abstract:Since 2023, computer scientists have warned against model collapse – the contamination of training sets with AI-generated outputs that progressively degrade model performance. Exemplifying a positive-feedback-driven failure, it produces effects such as word repetition or pixel noise, ultimately leading to a loss of meaning and coherence – at least from an engineering standpoint. From a creative one, however, collapse is not merely a breakdown: it also functions as a recursive mirror that recalls early analog video feedback experiments, raising once again the question of what happens when a system turns inward and sees itself. In such cases, so-called machine vision no longer transmits the world (as in tele-vision) but increasingly generates worlds from within. Drawing on media archaeology through case studies of both historical video synthesis techniques and contemporary artistic uses of machine learning, this paper examines what recursive training reveals about the dependent nature of AI-generated data. It argues that the potential effects of collapse challenge transhumanist ideals while inviting an aesthetic perspective, positioning noise and recursion as key concepts for understanding both artmaking and the AI ecosystem. Distributing agency across scales and networks, the latter currently remains reliant on new human-produced content, particularly within foundation models trained on massive datasets.
[AI-180] Mitigating Early Training Collapse in CTR Models
链接: https://arxiv.org/abs/2607.09696
作者: Ergun Biçici,Erkan Çetinyamaç
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 4 pages, 1 figure
Abstract:Deep neural models for click-through rate prediction often exhibit a sharp decline in validation performance immediately after the first training epoch despite continued improvement in training loss. This instability restricts effective learning and limits model performance. In this study, we analyze this behavior using large-scale industrial datasets and evaluate practical mitigation strategies. While reducing the learning rate provides only incremental gains, controlling feature sparsity yields substantial improvements. Removing highly sparse features and aggregating infrequent feature values stabilizes training, extends useful learning beyond a single epoch, and improves both offline evaluation metrics and online system performance.
[AI-181] Depth-Entropy Guided Sampling for Training-Free LLM Reasoning
链接: https://arxiv.org/abs/2607.09693
作者: Zibin Meng,Peng Xie,Kani Chen
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Reinforcement learning (RL) has become the dominant paradigm for improving the reasoning capabilities of large language models, but it requires expensive training, curated data, and reward signals. Recent work shows that sampling from sharpened base-model distributions at test time recovers much of the RL gain, yet existing methods rely solely on output-layer likelihoods and ignore the transformer’s internal forward-pass dynamics. We introduce Depth-Entropy Guided Sampling (DEGS), a training-free, test-time method that exploits layer-wise entropy collapse as an intrinsic quality signal. We observe that stronger reasoners – including RL-posttrained variants – exhibit a distinctive “late collapse”: logit-lens decoded entropy stays elevated until deeper layers before converging. We define a per-sequence collapse depth D(\mathbfx) and a joint objective \pi(\mathbfx) \propto p(\mathbfx)^\alpha \exp(\beta D(\mathbfx)) that combines sequence likelihood with this depth-entropy structure, instantiated inside an MCMC power-sampling framework (DEGS-MCMC). Across three open-weight models and four reasoning benchmarks, this near-chance per-candidate signal compounds over the sampling trajectory into state-of-the-art training-free accuracy, with gains largest out of domain and on the harder splits – exactly where likelihood alone falls short – at single-digit-percent wall-clock overhead. DEGS narrowly trails an in-house GRPO reference on the math splits GRPO was trained for, yet surpasses it out of domain on GPQA for all three models, without any training, reward model, or labeled data.
[AI-182] What Context Does a Coding Agent Actually Need to Act?
链接: https://arxiv.org/abs/2607.09691
作者: Brian Sam-Bodden
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:A modern coding agent can hold an entire repository in its context window. Most of its reading is wasted – and the interesting question is not how much context an agent can use, but what it actually \emphneeds. We study that question at the moment it matters most: when the agent must \emphedit code. Separating \emphfinding the work site from \emphacting on it, we hold localization fixed with an oracle, vary only how the code is represented, and score context against real issue resolution on SWE-bench Verified. The answer is starkly minimal. The signal lives in the code being edited itself: natural-language summaries of it answer almost none of the behavioral questions that the source answers ( 4/45 vs.\ 27/45 , held-out repositories, independent judge), and the gap belongs to the representation, not the summarizer – a frontier model’s summaries score exactly as poorly as a 3B model’s. The surrounding context hardly matters either: across every multi-file instance in Verified, under a protocol frozen before any data, rendering a file’s remainder as UML skeletons and signatures resolves no more issues than deleting that remainder outright ( N=70 , exact McNemar p=0.75 ). That was our registered hypothesis, and it failed. Compressed context, meanwhile, matches whole files at a third of the tokens: a resolved issue costs 19 K context tokens, not 94 K. The instrument also yielded a finding the field should keep: temperature-0 API inference flips \sim9% of per-instance outcomes between byte-identical runs. That is a noise floor under every small effect reported on this benchmark, including ours. We release the instrument – gold-validated environments, per-instance proof that every reference edit is expressible from every arm’s context, deterministic patch construction, and pre-registered hypotheses whose nulls we publish.
[AI-183] Boltzmann MapReduce: A Partition-Function Reduce for Forkable Sandboxes
链接: https://arxiv.org/abs/2607.09689
作者: Yossi Eliaz
类目: Artificial Intelligence (cs.AI); Probability (math.PR); Statistics Theory (math.ST)
备注: N/A
Abstract:To leading order under local asymptotic normality (LAN), the confidence density a worker emits over a chunk of size n is a Gibbs–Boltzmann measure \exp-\beta E(\theta)\ whose inverse temperature is the sample size, \beta=n . Three consequences are exact in the Gaussian/linear case and first-order otherwise: disjoint chunks carry independent Boltzmann factors, so the MapReduce \emphreduce, read literally, is a partition function Z=\int\prod_k h_k,d\theta whose mode is precision-weighted (inverse-variance) pooling; frequentist consistency is the zero-temperature limit T=1/n\to0
[AI-184] SciML in the Wild: A Diagnostic Study of When Structural Priors Help and When They Hurt
链接: https://arxiv.org/abs/2607.09684
作者: Vrishank Sai Anand,Prathamesh Dinesh Joshi,Raj Abhijit Dandekar,Rajat Dandekar,Sreedath Panat
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Scientific Machine Learning (SciML) methods such as Neural Ordinary Differential Equations (NODEs), Physics-Informed Neural Networks (PINNs), and Universal Differential Equations (UDEs) are most effective when structural priors reflect reliable governing dynamics. We ask what happens when this assumption is violated. Using macroeconomic forecasting as a stress-test domain, we evaluate five model families, ARIMA, LSTM, NODE, PINN, and UDE, across 23 countries using sparse annual data, multiple temporal splits, and five random seeds. Our results show that none of the evaluated models achieve consistently strong forecasting performance, highlighting the difficulty of low-frequency macroeconomic prediction. However, a clear relative hierarchy emerges: less-constrained models, particularly ARIMA and NODE, consistently outperform more-constrained heuristic-prior models such as PINN and UDE. Rather than treating this as a rejection of SciML, we interpret it as a diagnostic result: structural priors can act as misregularizers when they do not match the data-generating process. We identify failure modes including prior misalignment, regime shifts, structural breaks, and optimization instability, and argue that SciML practitioners should test whether structure helps before assuming that more structure is beneficial.
[AI-185] Ablation Statistical Inference and Validation for KV-Cache Compression
链接: https://arxiv.org/abs/2607.09683
作者: Paolo D’Alberto,Ashish Siarasao,Elliott Delaye,Rajeev Patwari
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Information Theory (cs.IT)
备注: 15 pages, 8 figures, minimum number of citations
Abstract:This study systematically compares Turbo-Quant and SpectralQuant KV-cache compression, evaluating non-dominated schemes, including WHT rotation with Beta Lloyd-Max and QJL, through a statistical validation methodology that separates systematic codec differences from implementation variance. Key findings reveal that while eigenbasis-based methods fail on heavy-tailed data due to covariance instability, they excel in structured regimes, with the effective semantic dimension ( d_eff ) adapting to calibration budgets rather than true data rank. (this is an abstract of the abstract thank you )
[AI-186] Faithful Not Corrective: Message-Format Effects in Multi-Hop Agent Relays Are Tier-Dependent
链接: https://arxiv.org/abs/2607.09678
作者: Zayx Shawn
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:When LLM agents hand off information to one another, does the message format matter? Two literatures disagree: format-optimization work reports that structured messages cut cost without hurting accuracy, while format-restriction work finds that imposing structure degrades generation – and neither measures what happens when a message traverses multiple hops, where copy fidelity, not one-shot generation, dominates. We introduce a controlled relay testbed: briefs of twelve programmatically generated atomic facts are re-encoded hop-by-hop in five formats (free NL, precision-instructed NL, JSON, triples, key-value) over six hops, scored by a fixed strong grader against programmatic ground truth, across two relay-capability tiers, a cognitive-load condition, and a paired-fork error injection. We find that message-format effects are tier-dependent. (i) Under faithful-relay instructions a strong relay is nearly lossless – the documented “telephone-game” collapse does not occur – and adding per-hop cognitive load leaves format-level fidelity unchanged (within +/-1.8 points) while raising generation cost by 24-53%. (ii) Under a weak (1.5B) relay the across-format spread of six-hop recall grows by a factor of 8.7 (from 2.3 to 20.5 points), driven by two opposing mechanisms – an encoding toll paid by the rigid formats and drift resistance specific to the fixed-key JSON schema – that flip the format ranking in transit. (iii) In a paired-fork injection, an injected wrong value, once present, persists to the final hop in 83-100% of chains in every format, closely matching each format’s retention of the true value, with no detectable collateral damage to neighboring facts. Structure buys a faithful, error-localizing channel – not an error-correcting code – and format choice should follow the weakest relay in the pipeline.
[AI-187] Knowledge Graphs Meet Graph Neural Networks: A Comprehensive Survey
链接: https://arxiv.org/abs/2607.09666
作者: Chengcheng Sun,Jiayun Tian,Cheng Zhai,Zhixiao Wang,Yajie Song,Xiaobin Rui,Jian Zhang,Philip S. Yu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Social and Information Networks (cs.SI)
备注: Recently Accepted for publication in ACM Computing Surveys. This is the accepted manuscript version, and the final published version available at ACM Computing Surveys: this https URL . Paper list at Github: this https URL
Abstract:Graph Neural Networks (GNNs) have emerged as a powerful paradigm in Knowledge Graphs (KGs) due to their intrinsic ability to model graph-structured data. However, there remains a lack of a systematic review about GNN-based methodologies across the entire knowledge graph technologies pipeline. To address this gap, we first propose a novel two-level taxonomy framework for GNN-based knowledge graph technologies: the KG technologies pipeline and GNN-based perspective. Specifically, the knowledge graph technologies pipeline covers knowledge graph construction, knowledge graph embedding, knowledge reasoning and knowledge graph applications. Meanwhile, the GNN-based perspective provides a new categorization of knowledge graph technologies with GNN models, such as GCN, GAT, and HGNN. Then, we analyze the advantages of GNN technology based on the characteristics of different tasks in the knowledge graph lifecycle. Furthermore, we detailed review various GNN-based models for knowledge graph following the proposed taxonomy, and summarize strengths and limitations. Finally, we discuss unresolved challenges and outline promising directions for future research.
[AI-188] Reverse Engineering Compliance: A Dual-Graph Verification Framework for Auditing Legacy IT Security Concepts
链接: https://arxiv.org/abs/2607.08292
作者: Lea Roxanne Muth,Marian Margraf
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: Accepted for publication at the 2026 IEEE International Conference on Computer, Information and Telecommunication Systems (IEEE CITS), Piraeus-Athens, Greece, July 22-24, 2026. 8 pages, 1 figure
Abstract:The NIS-2 Directive increases the need for continuous, auditable compliance evidence and motivates a shift from document-based compliance toward machine-readable compliance artifacts. The Open Security Controls Assessment Language (OSCAL) is a standard for this purpose, which the German Federal Office for Information Security (BSI) is adapting with Grundschutz++. However, companies are still managing extensive legacy IT security concepts (IT-SCs), and migrating them without verification could transfer outdated assets into the new format. While existing research primarily addresses the generation of new concepts, there is a lack of a verification framework that extracts legacy IT-SCs into an auditable intermediate representation, deterministically compares the extracted graph with an independently constructed reference state, and exports schema-valid OSCAL artifacts. This paper introduces the Automated Security Concept Structure Extraction and Reverse Topology-checking (ASSERT) Framework, which addresses this gap by using ontology-based extraction of legacy documents into formal document graphs, a five-class graph difference against a verified reference graph, and the export into schema-valid OSCAL outputs for system description and assessment evidence. Using the BSI’s RecPlast dataset, we compare a local open-weight model and a commercial model across three configurations with different levels of reference-ontology exposure. The evaluation shows that ASSERT makes document-infrastructure inconsistencies measurable, but reveals a trade-off between discovering undocumented entities and enforcing a schema. Comments: Accepted for publication at the 2026 IEEE International Conference on Computer, Information and Telecommunication Systems (IEEE CITS), Piraeus-Athens, Greece, July 22-24, 2026. 8 pages, 1 figure Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI) Cite as: arXiv:2607.08292 [cs.CR] (or arXiv:2607.08292v1 [cs.CR] for this version) https://doi.org/10.48550/arXiv.2607.08292 Focus to learn more arXiv-issued DOI via DataCite
[AI-189] Actor-Critic Learning for Extended Mean Field Control with Deterministic Policies
链接: https://arxiv.org/abs/2607.11005
作者: Ziheng Cheng,Xin Guo,Huyên Pham,Yufei Zhang
类目: Optimization and Control (math.OC); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Machine Learning (stat.ML)
备注:
Abstract:This paper develops a model-free reinforcement learning framework for continuous–time extended mean field control problems, where both the dynamics and reward may depend on the joint distribution of states and controls. We adopt deterministic feedback policies, under which the state–action distribution is induced directly as a push–forward of the state law. This avoids optimization over stochastic kernels and bypasses key limitations of existing approaches in extended mean field settings. We first establish a model–free sensitivity formula for parameterized McKean–Vlasov dynamics and use it to derive a deterministic policy gradient formula expressed through an advantage–rate function on the Wasserstein space. We then refine this formula by introducing local value and advantage–rate representations that depend on the state, action, and joint state–action distribution, yielding a policy gradient that includes both action derivatives and measure–derivative terms with respect to the control distribution. These characterizations lead to a martingale–based learning principle and motivate a continuous–time deep deterministic policy gradient algorithm combining particle approximations, measure–dependent neural networks, temporal–difference learning, and exploration in either action or parameter space. Numerical experiments on stochastic Cucker–Smale consensus control and optimal liquidation with trade crowding demonstrate the efficiency, stability, and robustness of the proposed method, including problems with explicit dependence on the control distribution. Subjects: Optimization and Control (math.OC); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Machine Learning (stat.ML) Cite as: arXiv:2607.11005 [math.OC] (or arXiv:2607.11005v1 [math.OC] for this version) https://doi.org/10.48550/arXiv.2607.11005 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-190] MDQEC-QAS: Meta-Decoding for Quantum Error Correction with Hardware-Aware VQC Search and Confidence-Gated Recovery
链接: https://arxiv.org/abs/2607.10707
作者: Prashant Kumar Choudhary,Nouhaila Innan,Muhammad Shafique,Rajeev Singh
类目: Quantum Physics (quant-ph); Artificial Intelligence (cs.AI); Hardware Architecture (cs.AR); Machine Learning (cs.LG)
备注: 15 pages, 12 figures and 11 tables
Abstract:We propose a unified meta-decoding framework for quantum error correction that learns syndrome-to-recovery mappings across multiple stabilizer codes and noise settings, without requiring separate decoders for each configuration. The benchmark includes FiveQubit, Steane, Planar3x3, and Planar5x5 codes, four noise families, and five evaluation regimes: interpolation, unseen-p transfer, unseen-noise transfer, few-shot unseen-code adaptation, and few-shot held-out-size adaptation. We compare a classical Meta-MLP teacher-trained baseline with variational quantum circuit (VQC) meta-decoders selected through hardware-aware quantum architecture search over qubit count, circuit depth, and entangling topology. The Meta-MLP achieves teacher-label accuracies of 0.9993, 0.9118, 0.9342, 0.6304, and 0.7548 across the five regimes, while the hardware-aware VQC achieves 0.9400, 0.8495, 0.8415, 0.5678, and 0.7143. However, logical-level evaluation shows that high teacher-label accuracy alone is insufficient in the most challenging Planar5x5 setting. During interpolation, the raw logical-failure ratios relative to the teacher are 12.08 and 25.91 for the Meta-MLP and VQC, respectively, whereas confidence-gated fallback reduces them to 1.71 and 1.11. These results support confidence-aware selective recovery rather than unconditional teacher replacement.
[AI-191] Learning the Brains Dynamics as a Port-Hamiltonian System
链接: https://arxiv.org/abs/2607.10439
作者: Dibakar Sigdel
类目: Neurons and Cognition (q-bio.NC); Artificial Intelligence (cs.AI)
备注:
Abstract:We model human motor cortex during a wrist-extension BCI task as a port-Hamiltonian system (pHS): a conservative interconnection (gyroscopic coupling between neural phasors) plus a dissipative port (power-law energy decay driven by a GNN surrogate). A metriplectic integrator evolves the phasor state; a Fluctuation–Dissipation-consistent noise channel produces stochastic trajectories at body temperature. Training on \FitTrainN\ real EEG cycles (PhysioNet EEGMMIDB, 3 held-out subjects) reaches a test MSE of \FitTestMSE\ and passes three scale-free criticality rungs: near-critical branching ratio ( \sigma\approx1 ), 1/f power-law spectrum, and long-range DFA correlations. The model generates closed-loop neuromodulation signals that restore phase-locking in silico when applied to de-synchronised inputs, suggesting a path toward structure-preserving BCI decoders.
[AI-192] he evolution of AI from image interpretation toward scientific inference in nanoparticle electron microscopy
链接: https://arxiv.org/abs/2607.10388
作者: Evropi Toulkeridou,Jiafei Li,Leonardo Lari,Panagiotis Grammatikopoulos
类目: Materials Science (cond-mat.mtrl-sci); Artificial Intelligence (cs.AI)
备注: Main article: 34 pages, 8 figures (including Graphical Abstract), 5 tables. Appendix: 18 pages, 1 figure, 4 tables. This manuscript has been submitted (after invitation) to Advanced Intelligent Discovery for peer review
Abstract:Artificial intelligence (AI) is transforming electron microscopy by enabling quantitative analysis of increasingly large and complex datasets for nanoparticle characterization. Recent advances in machine learning (ML) and deep learning (DL) have expanded microscopy from a descriptive imaging technique into a data-driven platform for structural interpretation, dynamic analysis, and scientific inference. This review examines AI methodologies for nanoparticle electron microscopy, focusing on transmission electron microscopy (TEM), high-resolution transmission electron microscopy (HRTEM), scanning transmission electron microscopy (STEM), and in situ TEM. The discussion is organized around the principal challenges in nanoparticle characterization, including particle detection, segmentation, morphology quantification, atomic-resolution restoration, defect identification, two-dimensional-to-three-dimensional structural inference, and analysis of dynamic processes in situ. We review computational approaches from conventional ML and convolutional neural networks to transformer architectures, self-supervised learning, foundation models, multimodal AI, and physics-informed learning. We further discuss integrating microscopy data with simulations, metadata, and autonomous experimentation to relate nanoparticle structure, dynamics, synthesis conditions, and functional properties. The advantages, limitations, benchmarking, and data requirements of current methodologies are critically assessed. Finally, emerging opportunities for foundation models, AI-guided microscopy, closed-loop experimentation, and autonomous materials discovery are discussed. By integrating advances across computer vision, materials informatics, and electron microscopy, this review highlights the role of AI in next-generation nanoparticle characterization and accelerated materials discovery.
[AI-193] Program-Synthesis-Driven Autodesign of Universal Unitary Operators
链接: https://arxiv.org/abs/2607.10295
作者: Yifei Zhang,Dong Chen,Fan Wang,Wenrui Zhang,Yan Chen,Dingding Han,Jianmin Yuan,Xiangjin Kong,Yu-Gang Ma
类目: Optics (physics.optics); Artificial Intelligence (cs.AI)
备注:
Abstract:We demonstrate that AI-driven program synthesis can autonomously discover fundamental strategies for decomposing unitary matrices in photonic networks. By extending DreamCoder to complex-valued linear algebra, the system generates decomposition programs achieving the minimal N(N-1)/2 Mach-Zehnder interferometers, distinct from both Reck and Clements architectures. Learned programs encode dimension-agnostic invariants: strategies discovered for 5 \times 5 matrices generalize to higher dimensions such as 64 \times 64 . The discovered programs encode interpretable, dimension-agnostic construction rules. These rules generalize across matrix sizes without retraining, demonstrating that autonomous program synthesis can serve as a scalable paradigm for algorithm discovery and the automated design of universal unitary operators. Beyond universal decompositions, the system automatically exploits matrix structure to reduce the interferometer count below the universal theoretical bound. For instance, for Householder matrices, it discovers a dimension-independent rule that requires only 2N-3 MZIs. This achieves linear, rather than quadratic, scaling and generalizes to arbitrary N without retraining. For matrices obtained from the singular value decomposition of sparse matrices, reductions generally increase with sparsity, reaching up to 38% fewer MZIs than the universal theoretical bound N(N-1)/2 at 95% sparsity. These MZI reductions translate directly into practical hardware benefits for scalable photonic implementations. Taken together, the system functions as a single unified engine that discovers both universal decomposition rules and matrix-specific optimizations, without being provided with the structural or analytical properties of the input matrices.
[AI-194] Quantum Circuit Vision: Cost-Aware Evaluation of Visual AI Agents for Quantum Code Generation
链接: https://arxiv.org/abs/2607.10057
作者: Dongping Liu,Aoyu Zhang,Luyao Zhang
类目: Quantum Physics (quant-ph); Artificial Intelligence (cs.AI)
备注:
Abstract:Can AI agents visually comprehend quantum circuit diagrams and generate verified executable code–and at what cost? We present Quantum Circuit Vision, a cost-aware evaluation framework for multimodal AI agents on quantum circuit visual understanding. We construct a 132-circuit benchmark spanning 13 categories ( 1 – 10 qubits) with executable Amazon Braket code and unitary-fidelity verification. Evaluating three frontier Claude-family models at different capability-cost tiers with n=5 repeated trials, we find that the mid-tier model (Sonnet 4.6, 1.30\times credits) offers the most favorable balance on the cost-accuracy frontier: 91% pass rate on the core subset at 18% of the per-call cost of the strongest model (Opus 4.6), whose accuracy advantage is not statistically significant (paired t : p=0.083 ). Logistic regression confirms that circuit depth–not qubit count–is the primary predictor of failure ( p0.001 ). Chain-of-thought prompting shows no statistically significant effect (all p0.18 , n=5 ), suggesting that visual pattern recognition outweighs explicit reasoning strategy for structurally coupled diagrams. We propose a cascade routing strategy (cheap \rightarrow expensive models) that achieves 84% accuracy at 38% of single-model cost, demonstrating that model routing dominates prompt engineering as a cost lever. We release QCV-Dataset (132 circuits, 5 modalities, 1,931 files) on Hugging Face Hub as an open evaluation infrastructure with structured metadata for discoverability, interoperability, and responsible AI documentation, and all evaluation code, cost logs, and verification scripts on GitHub for full reproducibility.
[AI-195] Geometric mean-based pairwise comparison method with the reference values – statistical approach
链接: https://arxiv.org/abs/2607.10038
作者: Konrad Kułakowski,Jacek Szybowski
类目: Methodology (stat.ME); Artificial Intelligence (cs.AI); Statistics Theory (math.ST)
备注: 30 pages
Abstract:For many years, the pairwise comparison method has been widely used for decision-making involving experts. The best-known example of this method is the Analytic Hierarchy Process (AHP). In this now classic approach, the weights of alternatives are calculated using the principal eigenvector of the comparison matrix. In this paper, we present a statistical view of the pairwise comparison method, using reference values and the geometric mean to calculate alternative priorities. Thanks to this approach, we can simultaneously capture both the phenomenon of inconsistency in pairwise comparisons and the preference distance between alternatives. In this paper, we define indicators that measure the quality of the obtained weight vector, which, thanks to the statistical approach, have a clear and intuitive interpretation.
[AI-196] Next-Dense-Stride Prediction for Multimodal Autoregressive Visual Modeling
链接: https://arxiv.org/abs/2607.09892
作者: Chicago Y. Park,Jialin Mao,Xiaojian Xu,Taha Kass-Hout,Ulugbek S. Kamilov,Cao Xiao
类目: Image and Video Processing (eess.IV); Artificial Intelligence (cs.AI)
备注:
Abstract:We introduce DenseAR, a new generative paradigm that reformulates autoregressive image generation as coarse-to-fine next-dense-stride prediction using a compact single-scale tokenizer. Our key insight is that traversing a single-scale latent grid with progressively denser strides naturally captures the transition from global structure to fine detail. This addresses two limitations of existing autoregressive models at once: the slow inference of raster-order autoregression, which DenseAR avoids by predicting multiple tokens in parallel, and the heavy cost of multi-scale approaches, which need long, multi-resolution token sequences to achieve coarse-to-fine prediction. Building on our efficient framework and the flexibility of autoregressive modeling, we further extend DenseAR to a unified model that handles multiple modalities and imaging tasks within a single backbone. We validate DenseAR on both medical and natural images. On multi-contrast brain MRI, a single DenseAR model unifies cross-modal translation, modality-conditioned generation, and tumor segmentation, while remaining competitive with task-specific methods. On ImageNet, DenseAR improves class-conditional generation quality (FID and IS) over both a single-grid baseline without stride ordering and a multi-scale tokenizer-based baseline.
[AI-197] Listen to the Features: Voice Anonymization Driven by Content Embedding Matching over Signal Reconstruction
链接: https://arxiv.org/abs/2607.09767
作者: Adrien Schneider(M-PSI),Kacper Zabkowski(M-PSI),Anderson Augusma(M-PSI),Frédérique Letué(SAM, SVH),Maria Camila Pinzon(M-PSI),Dominique Vaufreydaz(M-PSI)
类目: ignal Processing (eess.SP); Artificial Intelligence (cs.AI)
备注:
Abstract:The paper presents a voice anonymization model focusing on preserving content rather than producing realistic speech. It relies on content embeddings extracted from a frozen pretrained wav2vec2 encoder. These embeddings are decoded into an anonymized signal using vector quantization and a HiFi-GAN vocoder, both trained on LibriTTS without any waveform reconstruction loss or speaker embedding mapping. The training objective enforces that embeddings of the anonymized signal match those of the original one. While training, an auxiliary speaker classification branch with a gradient reversal layer is used to discard speakerspecific information. Results show that this straightforward embedding-based approach achieves very low WER (2.53) with an anonymization performance (EER 13.39) ranking within first level for VPC. Notably, emotions are partially preserved (UAR 43.91), even without a supporting training objective, while the anonymized voice is audible without reconstruction loss.
[AI-198] Physics-Informed Structure Anchoring With Capture-Aware Prototype Calibration for Cross-Environment RF Fingerprinting
链接: https://arxiv.org/abs/2607.09760
作者: Fengchong Yao,Jianbing Li,Qing Liu,Qikun Liu,Kefeng Song,Haitao Li,Song Wang
类目: ignal Processing (eess.SP); Artificial Intelligence (cs.AI)
备注:
Abstract:Radio frequency fingerprint identification (RFFI) uses transmitter-specific hardware imperfections as a physicallayer identity cue for Internet of Things (IoT) devices, but deep RFFI models often degrade when the acquisition environment changes. In multi-antenna reception, this degradation is not merely a generic distribution shift. It is also shaped by receiver-array topology, frequency-offset dynamics, and capturedependent target structure, which can distort embeddings and move source-trained decision boundaries. This article proposes physics-informed structure anchoring with capture-aware prototype calibration (PISA-CAPC), a framework that separates source representation anchoring from fixed-backbone target calibration. The representation stage organizes antenna tokens with a topology graph and modulates the graph using CFO-derived acquisition-dynamics descriptors. Bounded contextual residual suppression is then applied around the identity representation. At deployment, unlabeled capture-aware prototype calibration (U-CAPC) calibrates target decision scores through capturelocal prototype evidence under a fixed representation, mitigating boundary shift without requiring target-domain backbone updates or target labels. On a measured ten-transmitter multiantenna WiFi benchmark, PISA-CAPC achieves 0.9257 targetdomain mean Macro-F1 under a balanced transductive setting. Ablations confirm that topology-guided structure anchoring, contextual residual suppression, and capture-aware calibration contribute complementary gains. These results establish PISACAPC as a fixed-backbone route to cross-environment RFFI, coupling physically motivated representation learning with labelfree, capture-aware decision calibration.
[AI-199] Data-Driven Forward and Inverse Modeling of V-Beam Thermal Sensors
链接: https://arxiv.org/abs/2607.09752
作者: Tudor Bartha,Radu Chiorean,Adrian Groza
类目: ignal Processing (eess.SP); Artificial Intelligence (cs.AI)
备注:
Abstract:This paper presents a machine learning framework for data-driven inverse design of V-beam thermal sensors. The goal is to determine the optimal sensor geometry: beam inclination angle, beam length and beam width that achieves a target displacement under a given temperature. The design should also provide the geometry with minimum structure volume and minimum mechanical stress the sensor must support. This problem is ill-posed as for a given displacement there are multiple possible geometric configurations, causing direct regression methods to fail. We document a series of five exploratory trials that progressively revealed the nature of the problem culminating in a two-phase solution: a neural network forward model trained to map geometry and material constants to sensor responses, a gradient-descent inverse optimization over the frozen forward model, minimizing stress and volume simultaneously. The proposed pipeline utilizes a 3000-sample dataset and achieves a MAPE of 4.76% for predicting the displacement, more than 70% of predictions having MAPE of under 5%.
[AI-200] MorphologyFM: A Foundation Model for Morphology-Aware Representation Learning from ECG and Pulse Oximetry Waveforms
链接: https://arxiv.org/abs/2607.09749
作者: Saiyang Feng,Yuanyun Zhang,Shi Li
类目: ignal Processing (eess.SP); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Foundation models have recently emerged as a powerful paradigm for learning transferable representations from large scale biomedical data, yet existing approaches for physiological waveforms primarily optimize reconstruction or forecasting objectives that do not explicitly preserve clinically meaningful waveform morphology. Electrocardiograms (ECGs) and pulse oximetry (SpO2) waveforms encode rich cardiovascular and hemodynamic information through their morphological structure. In this work, we introduce MorphologyFM, a multimodal foundation model pretrained on paired ECG and SpO2 waveforms from the MIMIC critical care database using a morphology aware self supervised learning objective. MorphologyFM combines morphology guided masking, cross modal representation learning, and contrastive latent alignment to learn representations that capture clinically relevant physiological structure without requiring manual annotations. We evaluate MorphologyFM across multiple downstream prediction tasks, including arrhythmia classification, hypoxemia prediction, mortality prediction, and length of stay estimation, demonstrating consistent improvements over representative self supervised learning methods, including Masked Autoencoders (MAE), contrastive learning, Barlow Twins, and Joint Embedding Predictive Architectures (JEPA). Furthermore, we show that jointly modeling ECG and SpO2 waveforms produces more transferable representations than single modality pretraining. Our results establish waveform morphology as a powerful inductive bias for self supervised physiological representation learning and introduce MorphologyFM as a general purpose foundation model for continuous physiological monitoring.
[AI-201] he Universal Language of CSI:Unifying Wireless Sensing Across Devices and Environments
链接: https://arxiv.org/abs/2607.09727
作者: Jiayi Chen,Weiting Ou,Guangxu Zhu
类目: ignal Processing (eess.SP); Artificial Intelligence (cs.AI)
备注:
Abstract:WiFi sensing based on Channel State Information (CSI) promises ubiquitous, device-free perception, yet current research remains trapped in a Tower of Babel - fragmented into isolated silos where models are tailored to specific hardware dialects, fixed environments, and narrow tasks. The primary bottleneck is the Heterogeneity Gap: the disparity in signal dimensions, sampling rates, and semantic labels that prevents cross-system understanding. To bridge this gap, we propose a foundation-model framework that treats CSI not merely as raw signals but as a structured language with a learnable universal grammar. We first curate and standardize a large collection of heterogeneous real-world CSI datasets, establishing a unified infrastructure that allows incompatible signal formats to be treated as a single corpus. Second, we introduce a modular architecture that acts as a universal translator where lightweight dataset-specific adapters tokenize diverse signal inputs into a shared latent vocabulary, while a shared self-supervised Transformer backbone learns the temporal syntax of human motion and environmental dynamics. This design decouples sensing semantics from hardware syntax. Extensive evaluations show that by mastering this universal language, our approach consistently outperforms task-specific baselines and exhibits strong generalization capability in new environments, achieving superior efficiency in few-shot scenarios. By effectively absorbing heterogeneity, the framework offers a path toward robust, general-purpose wireless sensing, mirroring the linguistic generalization observed in Large Language Models. The code implementation is available at: this https URL.
[AI-202] he Ramanujan Challenge For AI WWW
链接: https://arxiv.org/abs/2607.09721
作者: Michael Shalyt,Rotem Kalisch,Carsten Schneider,Hila Barkan,Elyasheev Leibtag,John Campbell,Shachar Weinbaum,Tali Monderer,Ashvni Narayanan,Ido Kaminer
类目: History and Overview (math.HO); Artificial Intelligence (cs.AI); Number Theory (math.NT)
备注: Please submit solutions at this https URL
Abstract:To help evaluate the mathematical skills of current AI systems, we present a set of formulas for fundamental mathematical constants. These problems are attractive for AI evaluation because they are concrete and can be checked numerically to arbitrary precision, yet proving them may require non-obvious mathematics. Mathematical constants such as \pi , e , Catalan’s constant, and special values of the Riemann zeta function have fascinated mathematicians for centuries. The search for formulas evaluating mathematical constants has produced some of the most beautiful mathematics in the field, especially in cases that yield irrationality proofs or fast convergence rates. Ramanujan’s legacy is emblematic of this tradition. The list we provide contains two types of problems: formulas whose proofs are known to the authors but will remain encrypted for a short initial period; and formulas that are not yet proven. We are curious to see the achievements of AI in both cases.
[AI-203] ECG-LDC: A Hardware-Efficient Low-Dimensional Computing Framework for ECG Arrhythmia Classification
链接: https://arxiv.org/abs/2607.09680
作者: Anh Tran,Khanh Tran,Cuong Do
类目: ignal Processing (eess.SP); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 10 pages, 5 figures, 6 tables
Abstract:Continuous cardiac monitoring in wearable devices demands classifiers that are simultaneously accurate, energy-efficient, and deployable on resource-constrained hardware. While deep neural network approaches have demonstrated high classification accuracy for electrocardiogram (ECG) arrhythmia detection, their substantial parameter counts and reliance on multiply-accumulate-intensive operations make them impractical for low-cost edge platforms. In this work, we propose ECG-LDC, a hardware-software co-design framework that adapts Low-Dimensional Computing (LDC) for real-time ECG arrhythmia classification. ECG-LDC employs a dual-encoder architecture with dedicated value and feature codebooks to independently encode morphological waveform features and RR-interval temporal features, enabling effective capture of both intra-beat and inter-beat cardiac dynamics. The framework encompasses data preprocessing, model training, and a hardware accelerator architecture prototyped on the Pynq-Z2 platform. Implemented using binary representations and XOR/XNOR-based operations, ECG-LDC achieves 97.18% accuracy with a memory footprint of only 3.86\ \textkB . ECG-LDC sacrifices approximately 1.8% accuracy versus SOTA TinyML classifiers but achieves 11 ~ 570\times reduction in memory usage; among FPGA-based five-class arrhythmia classifiers, it delivers the highest accuracy with up to 2.4\times fewer LUTs and zero DSP block utilization, affirming its suitability for real-time arrhythmia detection on resource-constrained wearable platforms.
机器学习
[LG-0] Requential Coding: Pushing the Limits of Model Compression with Self-Generated Training Data
链接: https://arxiv.org/abs/2607.11883
作者: Shikai Qiu,Marc Finzi,Yujia Zheng,Kun Zhang,Andrew Gordon Wilson
类目: Machine Learning (cs.LG)
*备注: Code available at this https URL
Abstract:Compression is fundamental to intelligence. A model that can represent its training data as a short code has discovered regularities that enable generalization. Large neural networks may learn functions far simpler than their parameter counts suggest, but it is challenging to construct codes that realize this simplicity. Parameter-based methods such as quantization produce code lengths that scale with model size, insensitive to how much information the parameters store. Prequential coding bypasses this issue by compressing the training trajectory, but codes the exact data sequence regardless of how much the model learns, yielding large codes when the data has high entropy. We introduce requential coding, where a teacher model selects training samples drawn from the student’s own distribution. The student’s code records only these selections, which cost bits only where teacher and student disagree. The resulting code length is independent of parameter count and data entropy, and often orders of magnitude shorter than the prequential counterpart, with an advantage that grows with scale. This compression sheds light on phenomena inaccessible to prior compressors. Holding loss fixed, larger models and ensembles compress to much smaller sizes despite more parameters. Plugged into a PAC-Bayes bound, the requential code yields state-of-the-art generalization guarantees for billion-parameter LLMs, outperforming bounds built on aggressive post-training quantization even granted zero error. The bound tightens with scale in the compute-optimal regime, as models become increasingly compressible relative to dataset size. The same code predicts that models gradually overfit when trained for multiple epochs. It also isolates the learnable information in a dataset from its unpredictable, random content, revealing that lower-entropy text holds far more learnable structure than higher-entropy image data.
[LG-1] Relaxing Faithfulness with Intervention-Only Causal Discovery UAI2026
链接: https://arxiv.org/abs/2607.11816
作者: Bijan Mazaheri,Jiaqi Zhang,Caroline Uhler
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: Accepted to UAI 2026
Abstract:Causal discovery algorithms learn a network that describes the causal dependencies among random variables. A common workflow involves first utilizing conditional independence properties on observational data to determine partially directed causal relationships, then applying interventions to orient the unknown causal directions. A critical assumption for the first step is faithfulness: a requirement that causally linked variables exhibit statistical dependence. Many natural systems include buffering and stabilizing pathways that cancel out to achieve systemic robustness. This cancellation of pathways violates faithfulness, leading causal discovery algorithms to incorrectly remove causal dependencies. In this paper, we argue that hard interventions contain information about the presence/absence of causal linkage that is overlooked in the first stage of structure discovery. We show that a mild assumption – called intervention-immediacy faithfulness – that allows cancellations, is sufficient to nonparametrically identify causal structures with hard interventions. These results position interventions as the primary carriers of information about causal structure, which should take precedence over conditional independence testing. To flip the paradigm, we also specify equivalence classes when the identification criteria are not met due to limitations in the scope of interventions.
[LG-2] An Exact Instrument for State Usage in Selective State-Space Models and the Input-Driven Migration It Reveals
链接: https://arxiv.org/abs/2607.11796
作者: Raktim Bhattacharya
类目: Machine Learning (cs.LG)
*备注:
Abstract:Selective state-space models such as Mamba route information through a bank of first-order modes whose input coupling is set by a learned selection mechanism. We give an exact instrument for measuring how a trained model uses these modes. Because the state matrix is diagonal, each channel’s output decomposes exactly into per-mode contributions, and a per-(layer, channel, window) Gram tensor yields the exact output error of dropping any subset of modes, offline, at any budget. Validated against the reference implementation to a relative error of 2.3\times10^-7 on the Mamba-1 family where it is exact, the instrument predicts a layer’s deployed pruning error to a median relative deviation of 5\times10^-7 over 4,464 configurations, its floor set by the reconstruction. Applying the instrument across the Mamba-1 family (130M–2.8B), the deployed 7B Falcon-Mamba, and Mamba-2, we find that trained models re-allocate their state space with the input: which modes carry the signal migrates across contexts, and at the most affected layers a per-input oracle roughly halves the output error of a fixed mode set. Frozen-signal counterfactuals attribute the migration primarily to the input-dependent write map B_t ; the timestep usually identified with selectivity carries almost none of it. Input-scheduled mode pruning on this measurement outperforms static, Hankel-based, and layer-adaptive rankings at every scale from 130M to the deployed 7B Falcon-Mamba, and at half the state budget it matches the unpruned model. Because the scheduler reads each window’s mode usage from a first pass, this demonstrates realizable headroom; we claim no deployed compute or memory saving.
[LG-3] From Global to Factor-Wise Expert Composition in Discrete Diffusion Models
链接: https://arxiv.org/abs/2607.11758
作者: Haozhe Huang,Yudong Xu,Abhijoy Mandal,Alán Aspuru-Guzik
类目: Machine Learning (cs.LG)
*备注:
Abstract:Discrete diffusion models offer a powerful framework for solving complex reasoning tasks, particularly through compositional generation, which combines multiple pre-trained experts to generalize beyond their individual training data. Recent theoretical corrections introduce time-dependent mixing weights to better align composed diffusion dynamics with the intended target. However, these methods are fundamentally limited by working on a per-sample basis, treating each generated state monolithically and ignoring the potential spatial or functional specializations of different experts. In this work, we address this limitation by proposing FactorDiff - a factor-wise composition framework for diffusion models. We posit that samples can be further decomposed into smaller factors, and propose a sampling process that dynamically routes each factor to the most relevant expert. We instantiate this framework with spatial/pixel-level compositions and validate it on the ARC-AGI benchmark, demonstrating that simple factor-specific routing consistently outperforms complex global scalar weighting schemes on tasks that require logical consistency and spatial disentanglement.
[LG-4] HiFi-LLP: High-Fidelity Low-Cost Latency Predictors with Confidence for Robust HW-NAS SOCC
链接: https://arxiv.org/abs/2607.11746
作者: Shambhavi Balamuthu Sampath,Behzad Shomali,Nael Fasfous,Moritz Thoma,Judeson Anthony Fernando,Lukas Frickenstein,Pierpaolo Mori,Manoj Rohit Vemparala,Alexander Frickenstein,Walter Stechele
类目: Machine Learning (cs.LG); Hardware Architecture (cs.AR)
*备注: Published in the Proceedings of the 2025 IEEE 38th International System-on-Chip Conference (SOCC)
Abstract:With deep neural networks (DNNs) increasingly deployed on edge devices, hardware (HW)-aware optimization techniques–such as HW-aware compression and HW-aware neural architecture search (HW-NAS)–have become essential. These methods rely on real feedback from the target hardware to tailor DNN architectures for efficient deployment. While the search can be parallelized, latency measurements via hardware-in-the-loop (HIL) remain a bottleneck due to their sequential nature. Recent approaches use latency predictors to replace costly HIL feedback, but challenges persist: (1) platform-specific predictors often require tens of thousands of samples, and (2) inaccurate predictions can mislead the NAS process. To address this, we introduce HiFi-LLP, a high-fidelity, low-cost latency predictor based on graph attention networks, augmented with a confidence metric. HiFi-LLP outperforms prior platform-specific predictors by up to 9 percentage points (p.p.) in the 10% accuracy bound and achieves a Spearman’s rank correlation of up to 0.996 across six devices in the LatBench dataset. We further propose a hybrid NAS framework that routes low-confidence predictions to HIL, achieving up to 8.6 \times speedup compared to typical NAS while maintaining a competitive Pareto front.
[LG-5] CatRetriever: Contrastive Representation Learning for Slab-to-Bulk Retrieval in Generative Catalyst Discovery
链接: https://arxiv.org/abs/2607.11712
作者: Jungho Oh,Woosung Kim,Dong Hyeon Mok,Jonggeol Na,Seoin Back
类目: Machine Learning (cs.LG); Materials Science (cond-mat.mtrl-sci)
*备注:
Abstract:Inverse design is an emerging data-driven paradigm for efficiently navigating vast chemical spaces to discover new materials with targeted properties, and in the context of heterogeneous catalysis, surface generative models have recently advanced this goal by directly generating catalyst surface-adsorbate structures. However, these models typically operate at the slab level and do not provide the corresponding parent bulk structure, making it difficult to assess bulk-dependent properties such as formation energy, surface energy, crystallographic symmetry, and synthesizability. Here, we address this missing slab-to-bulk connection as a retrieval problem and introduce CatRetriever, a contrastive representation learning model that aligns slab and bulk crystal representations in a shared latent space. From a slab query, CatRetriever accurately retrieves plausible parent bulk candidates with R@1 91% and R@3 98% on both the in-distribution and holdout evaluation sets. We further extend the CatRetriever framework into an adsorption energy targeted bulk discovery pipeline that combines bulk retrieval, generative search space expansion, and adsorption energy distribution analysis. This workflow evaluates candidates by both structural compatibility with the query slab and their ability to access the target adsorption energy range across diverse surface environments. CatRetriever therefore provides a scalable route for connecting catalyst generative models with physically plausible and adsorption energy compatible bulk catalyst discovery.
[LG-6] Self-Healing Visual Recovery for Autonomous Ground Vehicles Using Camera-Only Visual Odometry
链接: https://arxiv.org/abs/2607.11686
作者: Jakob Solberg Berntzen,Safia Fatima,Leon Moonen
类目: Robotics (cs.RO); Machine Learning (cs.LG); Software Engineering (cs.SE)
*备注: 18 pages, 11 figures, 4 tables
Abstract:Low-cost unmanned ground vehicles are often used in indoor places like warehouses, inspection corridors, and farm rows, where painted floor lines guide the robot. Line following is useful because it only needs one camera and little computing power, but it can fail when the line is blocked or turns sharply and goes out of view. Sensor-rich platforms tolerate this through hardware redundancy (LiDAR, GPS, multiple cameras), but camera-only systems must recover at runtime with no additional infrastructure. This paper presents a lightweight, two-stage recovery approach that restores guideline tracking without LiDAR, GPS, or a GPU. When the line is lost, the robot first turns in place while slowly relaxing its color checks and waiting for confirmation across multiple frames (Stage 1). If the line is still not found, monocular visual odometry moves the robot back to saved breadcrumb positions before it tries again (Stage 2). The system uses a depth-gated HSV line tracker, a YOLOv8n obstacle detector, and a visual odometry breadcrumb mapper, and it runs at 20 Hz on CPU-only hardware. The controller embeds a complete MAPE-K loop within a single 50 ms control tick, with no external adaptation manager required. The approach is evaluated across 119 fault-injected episodes on three Webots simulation courses. The method was successful in 86.6% of cases, with a median recovery time of 3.26 seconds. These results demonstrate that reliable visual recovery is feasible on camera-only UGVs within practical cost and computational limits.
[LG-7] A multi-scale feature enhanced graph neural network for fluid dynamics prediction in complex geometries
链接: https://arxiv.org/abs/2607.11672
作者: Li Xiao,Tianyu Li,Yiye Zou,Mingjie Zhang,Xiaogangd Deng
类目: Machine Learning (cs.LG); Fluid Dynamics (physics.flu-dyn)
*备注: 14 pages
Abstract:Industrial design in fields such as vehicle and aerospace engineering often relies on large-scale numerical simulations to evaluate fluid dynamics performance, which can incur substantial computational costs. Deep neural networks have shown promise in improving simulation efficiency, especially graph neural networks (GNNs), which demonstrate great potential due to their flexibility with unstructured data. However, GNNs face challenges when dealing with tasks involving complex geometries and large-scale meshes. In this paper, we propose the Multi-scale Feature Enhanced Graph Neural Network (ME-GNN) to tackle these challenges. ME-GNN employs a graph neural network with a two-step message-passing mechanism to capture detailed local features effectively. Additionally, it integrates an Attention U-Net with uniform grid discretization, enabling the extraction of both fine and coarse features. The model also utilizes K-hop sampling to construct subgraphs, facilitating efficient training on large datasets while preserving detailed local features. We evaluated ME-GNN on three benchmark datasets and achieved state-of-the-art results: a relative L2 error of 0.0196 for the velocity field and 0.0556 for the surface pressure on ShapeNet-Car, a normalized mean squared error of 0.0033 for the flow field on AirfRANS, and a relative L2 error of 0.1416 for the surface pressure on DrivAerNet.
[LG-8] How to Tame Grokking: Representation Geometry as a Control Signal
链接: https://arxiv.org/abs/2607.11666
作者: Maksim A Kazanskii
类目: Machine Learning (cs.LG)
*备注:
Abstract:Grokking is a phenomenon in which neural networks initially memorize training data and only later exhibit strong generalization after prolonged optimization. Despite extensive recent study, the factors influencing the emergence and timing of grokking remain incompletely understood. We investigate the relationship between representation geometry and delayed generalization. We find that dimensionality collapse consistently precedes the onset of grokking in all evaluated settings. Motivated by these observations, we introduce Geometric Dimensionality Regularization (GeomDR), a simple spectral regularizer that modifies the effective dimensionality of hidden representations during training. Across modular addition, modular division, and permutation composition tasks, GeomDR consistently alters grokking dynamics and can substantially accelerate the onset of generalization depending on the intervention schedule and target dimensionality. In several settings, grokking is accelerated by up to 52 times relative to standard AdamW training. Similar qualitative effects are observed in both multilayer perceptrons and transformers. Together, these results suggest that representation geometry can serve as an effective control signal for grokking and provide evidence that geometric interventions offer a practical approach for studying and influencing delayed generalization in neural networks.
[LG-9] Bet on Features: Anytime-Valid and Feature-Aware Auditing of Conditional Quantile Forecasters
链接: https://arxiv.org/abs/2607.11653
作者: Ivane Antonov,Sohom Mukherjee,Richard Pibernik,Yo Joong Choe
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:Black-box conditional quantile forecasts are widely used for sequential decisions under asymmetric costs, such as inventory planning in supply chain management. Once deployed, such forecasters must be monitored continuously as data streams drift and regimes change; this invalidates standard, fixed-horizon backtests for calibration. Further, existing backtests do not take into account that the notion of calibration is, in fact, information-dependent: forecasts can look calibrated to an auditor with coarse information while being miscalibrated to an auditor with richer information. We develop a distribution-free and game-theoretic testing framework for continuously auditing black-box conditional quantile forecasters with non-i.i.d. losses, such that the resulting evidence process is powerful against predictably chosen alternatives specified by the features available to the auditor. We first formalize notions of conditional quantile calibration when different sets of features are available to the auditor, establishing that the coarseness of the auditor’s information set determines the hardness of the testing problem. We then identify the sets of alternatives for which the auditor can achieve power, and focusing on contextual bets linear in the features, we derive finite-time detection guarantees for such alternatives, all without an i.i.d. assumption. The resulting evidence processes are interpretable at the feature level, as they quantify fine-grained, “feature-aware” evidence for miscalibration. We empirically validate these methods on simulated and real data, finding that a popular time series forecaster (Chronos-2) is highly miscalibrated w.r.t. multiple relevant features.
[LG-10] Fundamental Limitations of Fixed-Budget Best-Arm Identification
链接: https://arxiv.org/abs/2607.11635
作者: Motti Goldberger
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:In fixed-budget best-arm identification, also known as ranking and selection, an algorithm has a sampling budget to distribute across K arms. Each sample provides noisy feedback about that arm’s mean, and the goal is to identify the arm with the largest mean. A common performance benchmark is the static oracle: a non-adaptive strategy that knows the means in advance and chooses fixed sampling proportions to maximize the exponential decay rate of the probability of incorrect identification. Several adaptive algorithms have been constructed such that their sampling proportions converge to the static oracle proportions. However, it has remained open whether any algorithm could match the static oracle’s error decay rate uniformly across all problem instances. We answer this in the negative. For any K\ge 3 and for rewards drawn from any one-parameter natural exponential family, we show that for any algorithm, there is at least one instance where the error decay rate is at most \left(1 + \frac\log(K)8\right)^-1 times that of the static oracle. This also answers the open question posed by Qin (2022), showing that fixed-budget best-arm identification does not admit a complexity.
[LG-11] SKooP: Symmetric Koopman Predictions for Faster and More Generalizable Legged Robot Locomotion with Reinforcement Learning IROS
链接: https://arxiv.org/abs/2607.11624
作者: Evelyn D’Elia,Weishu Zhan,Giulio Turrisi,Giulio Romualdi,Giuseppe L’Erario,Raffaello Camoriano,Wei Pan,Daniele Pucci
类目: Robotics (cs.RO); Machine Learning (cs.LG)
*备注: This paper has been accepted for publication at the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Pittsburgh, USA, 2026
Abstract:Reinforcement learning (RL) algorithms classically suffer from poor sample efficiency. In robotics, a recent line of work has emerged addressing this problem by encoding physics priors in the learning process. However, most of these approaches are validated on well-defined, low-dimensional benchmark systems rather than high-dimensional robots with complex nonlinear dynamics. In this paper, we introduce \textitSKooP (Symmetric Koopman Predictions), an approach combining the advantages of morphological symmetries with those of a Koopman model learned via autoencoder to enhance policy learning. SKooP learns a Koopman model of the system dynamics alongside the policy. The resulting Koopman predictions are used as privileged observations for the critic, allowing the agent to learn based on smoother, more informative features. We also incorporate group symmetries into the actor, critic, encoder and decoder networks to produce a highly equivariant policy. The SKooP approach is validated via in-depth analysis of the learned Koopman models and symmetric policies to showcase how each of these influences the agent’s performance. We also show that the learned policies are transferable to different simulation environments. Our results show that SKooP consistently reduces convergence time and increases the learned reward for multiple challenging bipedal locomotion tasks on a quadruped robot. Project page: this https URL
[LG-12] Privacy-Aware Collaborative and Distributed Bayesian Optimization
链接: https://arxiv.org/abs/2607.11600
作者: Aditya Rane,Sathwik Yamana,Paritosh Ramanan,Srikanthan Ramesh,Akash Deep
类目: Machine Learning (cs.LG); Methodology (stat.ME)
*备注: 6 pages, 5 figures
Abstract:We propose a collaborative meta-learning framework for distributed Bayesian optimization matching centralized performance without raw-data exchange. We show gradient sharing leaks client observations, with leakage worsening as the search converges and queries concentrate near the optimum. We evaluate a differentially private defense and characterize its privacy-utility trade-off.
[LG-13] Advancing Optimal Subset Oracle via Learning Relaxation of Neural Set Functions
链接: https://arxiv.org/abs/2607.11555
作者: Yongquan Shi,Zijing Ou,Shiping Wang,Yatao Bian
类目: Machine Learning (cs.LG)
*备注:
Abstract:Learning neural set functions is pivotal to a wide range of important applications, including compound selection in AI-driven drug discovery and product recommendation. Recent work has introduced optimal subset oracles to implicitly learn set functions under practical weakly supervised settings, where model parameters are optimized through mean-field variational inference. However, these frameworks rely on Monte Carlo sampling to estimate gradients of the evidence lower bound when updating the variational distribution. Repeated sampling across iterations incurs substantial computational overhead, while the resulting stochasticity can destabilize the optimization trajectory. In this work, we reinterpret the evidence lower bound as a continuous relaxation of the set function and learn a surrogate objective that replaces sampling-based ELBO gradient estimation during variational optimization. The learned surrogate provides stable and efficient gradients throughout the continuous domain, thereby reducing computational overhead and accelerating inference. Furthermore, we establish an approximation guarantee for the proposed framework under submodular maximization and characterize its connection to variational free energy. Experiments on a variety of real-world tasks demonstrate consistent improvements over existing baselines.
[LG-14] Condition-Stratified Robustness Analysis of Post-Hoc Calibration Methods for Probabilistic Classifiers
链接: https://arxiv.org/abs/2607.11542
作者: Gurdeep Singh Virdee
类目: Machine Learning (cs.LG)
*备注: 6 pages, 5 figures
Abstract:Post-hoc calibration is widely adopted to correct probability estimates from trained classifiers, yet most evaluations report aggregate performance without testing whether that performance holds across distinct operating conditions within a single dataset. We present a pre-registered, condition-stratified robustness analysis comparing temperature scaling (TEMP) and isotonic regression (ISO) across four controlled conditions (C1–C4). Four hypothesis groups are evaluated: discrimination deltas with Holm-corrected multiplicity control (H1), Brier score differences (H2), calibration slope outcomes (H3), and AUROC differences under best-condition setups (H4). TEMP-minus-ISO discrimination deltas remain small across all conditions (-0.0155 to 0.0139), with Holm-adjusted p-values of 0.9895 everywhere. TEMP Brier differences are consistently negative (C1: -0.0002 through C4: -0.0074), while ISO shows sign reversals. TEMP calibration slopes stay closer to unity in every condition (range 0.7597–0.9493) than ISO slopes (0.1364–0.2726). AUROC differences shift from near zero in C1 (-0.0004) to positive in C4 (0.0264). These results establish that in-dataset robustness is condition-dependent and metric-specific. No claim of external transportability is made.
[LG-15] Random Label Prediction Heads for Studying Memorization in Deep Neural Networks
链接: https://arxiv.org/abs/2607.11541
作者: Marlon Becker,Jonas Konrad,Luis Garcia Rodriguez,Benjamin Risse
类目: Machine Learning (cs.LG)
*备注:
Abstract:We introduce a straightforward yet effective method to empirically study memorization in deep neural networks for classification tasks. Our approach augments each training sample with auxiliary random labels, which are then predicted by a random label prediction head (RLP-head). RLP-heads can be attached at arbitrary depths of a network, predicting random labels from the corresponding intermediate representation and thereby enabling analysis of how memorization capacity evolves across layers. By interpreting the RLP-head performance as an empirical estimate of Rademacher complexity, we obtain a direct measure of both sample-level memorization and model capacity. We leverage this random label accuracy metric to analyze generalization and overfitting in different models and datasets. Building on this approach, we further propose a novel regularization technique based on the output of the RLP-head, which demonstrably reduces memorization. Interestingly, our experiments reveal that reducing memorization can either improve or impair generalization, depending on the dataset and training setup. These findings challenge the traditional assumption that overfitting is equivalent to memorization and suggest new hypotheses to reconcile these seemingly contradictory results. The source code is available at this https URL
[LG-16] ropical Circuits with Scalar Multiplication Gates
链接: https://arxiv.org/abs/2607.11540
作者: Christoph Hertrich,Moritz Stargalla
类目: Computational Complexity (cs.CC); Machine Learning (cs.LG); Combinatorics (math.CO)
*备注: 23 pages, 5 figures
Abstract:We study tropical circuits with scalar multiplication gates, that is, algebraic circuits whose gates implement \max , + , or multiplication with a positive constant. For such circuits, we prove exponential size lower bounds for computing maximum weight directed spanning trees and maximum weight bipartite perfect matchings. As a corollary, we obtain an exponential size separation between monotone and non-monotone maxout neural networks, which generalize the popularly used ReLU neural networks. One conclusion from this is that neural network models with enforced convexity constraints, such as input-convex neural networks (ICNNs), sometimes need to be exponentially larger than their unrestricted counterparts in order to express the same functions.
[LG-17] DAG-FM: A Foundation Model for Causal Discovery under Heterogeneous Causal Mechanisms
链接: https://arxiv.org/abs/2607.11510
作者: Yikang Chen,Zhengkang Guan,Haoyuan Qian,Peng Cui,Yi Yang,Kun Kuang
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 17 pages, 7 figures, preprint
Abstract:Causal discovery from observational tabular data remains fundamentally challenging, primarily due to the heterogeneity of underlying causal mechanisms and the high-dimensional combinatorial search space of Directed Acyclic Graphs (DAGs). In this paper, we propose \textbfDAG-FM, a novel foundation model architecture that amortizes causal discovery. Unlike direct matrix prediction, DAG-FM decomposes the causal discovery process into two auto-regressive stages using two specialized Transformer-based sub-modules: a leaf-node predictor and a parent-node predictor. To effectively model complex row-column interactions, we adopt a robust tabular interaction block to output feature-wise representations. Crucially, to handle diverse and unknown Functional Causal Model (FCM) assumptions in real-world scenarios, we introduce Mixture-of-Leaf-Experts (MoLE), allowing the model to dynamically route and adapt to identifiable mechanism families. Through an iterative inference algorithm, DAG-FM seamlessly extracts causal orderings and constructs valid DAGs. Extensive experiments demonstrate that DAG-FM achieves state-of-the-art performance on both synthetic benchmarks and complex real-world datasets, significantly outperforming traditional classical algorithms and recent foundation models in both accuracy and scalability.
[LG-18] Compound Interference Recognition for LR-FHSS Satellite IoT Uplinks via Multi-Domain Instance Fusion
链接: https://arxiv.org/abs/2607.11488
作者: H. Xu,B. He,S. Wang,Y. Jiang
类目: Information Theory (cs.IT); Machine Learning (cs.LG)
*备注: Submitted to IEEE Internet of Things Journal
Abstract:Long range-frequency hopping spread spectrum (LR-FHSS) is a promising uplink physical layer for massive low Earth orbit satellite Internet of Things, where low power terminals report short packets from wide area regions with limited terrestrial infrastructure. However, satellite IoT links are exposed to external interference, and the coexistence of multiple interference components can severely degrade receiver reliability and complicate interference mitigation. Existing recognition methods either focus on single interference scenarios or treat each compound interference combination as an independent class, leading to limited generalization or poor scalability. To address this problem, this paper formulates LR-FHSS uplink compound interference recognition as a multi-instance multi-label learning problem and proposes a multi-domain instance fusion method. The proposed method fuses local instances from the time-frequency and frequency domains and aggregates their predictions for bag-level multi-label recognition. A dataset construction pipeline is developed based on the US915 LR-FHSS configuration and incorporates shadowed-Rician fading and time-varying Doppler to emulate practical satellite communication conditions. Considering the difficulty of obtaining labeled compound interference samples in practice, single-to-compound generalization and few-shot compound interference adaptation are investigated as two practical receiver deployment scenarios. Experimental results show that the proposed method improves the overall exact accuracy over the strongest baseline by 14.71 percentage points in single-to-compound generalization and by 14.81 percentage points in few-shot compound interference adaptation for r=1 .
[LG-19] Event-based Neural Decoding for Neuroprosthetic Motor Control
链接: https://arxiv.org/abs/2607.11445
作者: Khaleelulla Khan Nazeer,Sirine Arfa,Matthias Jobst,Richard George,Christian Mayr
类目: Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
*备注:
Abstract:A substantial number of patients experience diminished mobility due to disabilities, diseases, or accidents. Although modern prostheses, powered by deep neural networks, hold the promise of significantly enhancing the quality of life for these individuals, their widespread adoption is hindered by significant latency, energy consumption, and spatial requirements. Wired connections to external high-performance processors restrict patient mobility, while wireless connections limit the volume of information that can be transmitted to these processors. Spiking neural networks offer the potential for compressed communication and low-power inference, yet they often lag behind state-of-the-art deep learning models in various applications. In this study, we propose a high-performance neural decoding method that effectively balances task performance and efficiency. An eventbased gated recurrent unit generates a sparse communication pattern with graded spikes, surpassing classical spiking neural networks in terms of task performance. Utilising an efficient training method and sparse inference, our model presents new opportunities for on-device neural decoding.
[LG-20] Velocity Scheduled Flow Matching
链接: https://arxiv.org/abs/2607.11442
作者: Vitalii Bondar
类目: Machine Learning (cs.LG)
*备注:
Abstract:Flow matching trains a neural network to regress the conditional velocity along a linear interpolant between noise and data, and the number of network evaluations~(NFE) sets the cost of sampling. The straight-line interpolant carries an implicit choice: the sample moves at constant speed throughout the trajectory. We relax this choice and introduce Velocity Scheduled Flow Matching~(VSFM), which replaces the conditional target x_1 - x_0 with v(t)(x_1 - x_0) for any nonnegative profile v:[0,1]\to\mathbbR_\geq 0 satisfying \int_0^1 v,dt = 1 . We study six polynomial profiles drawn from motion planning. The first use of VSFM is at inference time: a pretrained linear flow-matching model can be sampled under any admissible profile by integrating its ODE on a non-uniform \tau -schedule, with no retraining and no additional computation; on CIFAR-10 this lowers FID by up to 19.8% . Training from scratch under a braking profile gives a further reduction of 17.4% at 4 ~NFE. Both gains follow from the local truncation error of the Euler integrator on the induced grid.
[LG-21] Generalizing Preference-based Reinforcement Learning: a Rationality Model for Incomparability
链接: https://arxiv.org/abs/2607.11432
作者: Simone Drago,Marco Mussi,Leonardo Bianconi,Alberto Maria Metelli
类目: Machine Learning (cs.LG)
*备注:
Abstract:In this work, we study the reinforcement learning (RL) problem from pairwise trajectory comparisons provided by a human expert. We generalize preference-based RL by formalizing a novel setting in which the expert can also label trajectory pairs as incomparable, i.e., when neither trajectory dominates the other. We introduce the learning problem and the desiderata that its solution should satisfy. Then, we propose a novel Bradley-Terry-inspired rationality model that effectively captures incomparabilities and infers a multi-dimensional reward function, and we study its properties. We provide a sample complexity analysis for learning the model parameters when a dataset is available. Finally, we evaluate our model’s ability to reconstruct a reward function that aligns with the expert’s comparisons in simulated environments and to recover the Pareto frontier of policies, along with a robustness analysis across varying levels of expert rationality.
[LG-22] Physics-Aware Conditional SetGAN for Spatially Consistent Multi-User TR 38.901 Channel Generation
链接: https://arxiv.org/abs/2607.11429
作者: Mauro Gonzalo Tarazona-Levano,David Lopez-Perez,Nicola Piovesan,David Gomez-Barquero
类目: Machine Learning (cs.LG)
*备注: Submitted to IEEE GLOBECOM 2026
Abstract:TR 38.901-based channel models such as Sionna are reliable, but generating many multi-user channel realizations remains expensive. This paper asks a practical question: can a trained generative model produce multi-user TR 38.901 channels faster than Sionna without losing the spatial correlations imposed by user geometry? To answer this question, we propose a physics-aware, geometry-conditioned SetGAN trained on Sionna reference data. The method separates large-scale received power from normalized small-scale fading, compresses the latter with principal component analysis, and learns the conditional channel distribution in a latent space while preserving geometry-dependent correlations. On the UMa/NLoS benchmark, the model keeps the received-power distributions close to the reference, with about 0.41 dB Wasserstein distance, and reproduces spatial-consistency profiles with mean deviations below 0.03 on median curves versus distance. In addition, it reduces elapsed generation time by a factor of 3.45 and CPU-total cost by a factor of 6.15 relative to Sionna under matched user positions in the fixed-position CPU-vs-CPU benchmark. These results show that a trained generative model can substantially accelerate TR 38.901 channel generation without breaking the spatial consistency needed to evaluate multi-user systems.
[LG-23] Surprisingly Simple and Effective Multi-Domain Graph Foundation Model through Graph-to-Table Alignment
链接: https://arxiv.org/abs/2607.11374
作者: Chunyu Hu,Tianyin Liao,Ge Lan,Xingxuan Zhang,Jianxin Li,Peng Cui,Ziwei Zhang
类目: Machine Learning (cs.LG)
*备注:
Abstract:Graph Foundation Models (GFMs) have emerged as a promising paradigm for learning transferable representations across diverse graph domains. Recent advancements in GFMs have been largely dominated by two paradigms: Graph Neural Network and Large Language Model (LLM) based methods. However, these methods often face a fundamental dilemma between training with limited data and a heavy reliance on textual attributes. Tabular foundation models (TFMs) offer a potential alternative, as node features and representations can be naturally organized in a tabular form. However, how to enable TFMs to effectively capture structural information of graphs remains largely unexplored. The key challenge is to learn a graph-to-table alignment mechanism that enables graph structural understanding for TFMs. To address this, we propose GTAlign, a surprisingly simple yet effective Graph-to-Table Alignment framework for text-free Graph Foundation Model. Specifically, we first pretrain a graph encoder that maps diverse graphs into a unified latent space to capture domain-agnostic graph representations. To further bridge the gap between graph topology and the tabular representation space, we propose community-guided continual pre-training, where pseudo-labels derived from graph community are used to construct few-shot prediction episodes. Lastly, we adapt the graph encoder for an unseen target domain and perform in-context inference. Extensive experiments on five benchmark datasets demonstrate that GTAlign significantly outperforms state-of-the-art baselines on both node and graph classification, offering a simple, effective, and text-free GFM model. Code will be released upon acceptance.
[LG-24] Decomposing Runtime Kernel and Quantization Speedups via a Matched FP16 Intermediate: A Hardware-Conditioned Case Study on Four NVIDIA RTX A5000 GPUs
链接: https://arxiv.org/abs/2607.11368
作者: Weijia Han,Lisha Qu
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG); Performance (cs.PF)
*备注: 36 pages, 8 figures
Abstract:Reported serving speedups from quantized kernels typically bundle the weight format, the kernel, and the inference runtime into one number. We present an attribution study on four NVIDIA RTX A5000 GPUs, 24 GiB each, on a single host with NVLink-bridged pairs. A matched intermediate stack that keeps the faster runtime without the quantized kernel splits the full speedup into a runtime part and a kernel and quantization part. Under matched greedy decoding the full stack reaches 2.58\times end to end, with the runtime change accounting for about two thirds of that gain on a logarithmic scale; across three similar model families the kernel and quantization part moves by at most 1.5%. Sharding one instance across all four cards falls well below doubling: a profiler trace attributes about 80% of the per token shortfall to coordination, and an NVLink versus PCIe control on the same hardware shows similar realized bandwidth on both links, pointing away from link bandwidth as the cause. Whether to run one sharded instance or several independent ones depends on the workload and the model, with the ranking reversing on the larger model: the smaller model splits between sharding and multiple instances by workload, while the larger model favors two paired instances on every workload. Quantization extends sustainable concurrent users roughly four times past a reproducible half precision memory cliff. Differences in sampling mode and prompt pool between the two stacks are documented as threats to validity.
[LG-25] SPARC-Net: A Spectral Causality-Aware and Hard-Constrained Physics-Informed Architecture for Stiff and Shock-Dominated Partial Differential Equations
链接: https://arxiv.org/abs/2607.11310
作者: Divyavardhan Singh,Dimple Sonone,Hammad Mohammad,Kishor Upla
类目: Machine Learning (cs.LG); Discrete Mathematics (cs.DM)
*备注: 8 pages, 11 figures, 5 tables
Abstract:Physics-Informed Neural Networks (PINNs) provide a meshless approach for solving partial differential equations (PDEs), but suffer severe degradation in stiff and shock-dominated problems, where small PDE residuals can correspond to globally inaccurate solutions. We show these failures are multi-causal, arising from the concurrent interplay of (i) spectral bias against sharp features, (ii) imbalanced multi-term optimization and loss-weight collapse, (iii) violation of temporal causality, and (iv) under-resolved collocation. We present SPARC-Net, a unified architecture and training framework that jointly addresses all four pathologies. SPARC-Net leverages an adaptive multi-scale spectral encoder with a learnable spectral gate, a gated residual backbone, adaptive activations, and a hard-constraint output ansatz that exactly enforces initial and boundary conditions, structurally eliminating loss-weight collapse. Training employs stabilized gradient-norm loss balancing, floored causality-respecting residual weighting, and residual-based adaptive collocation (RAD). Validated against exact analytic and high-order spectral reference solutions across four canonical benchmarks – viscous Burgers’, Allen-Cahn, convection (beta=30), and reaction – SPARC-Net yields substantial improvements over vanilla PINNs: relative L2 error drops from 1.47e-1 to 1.14e-1 on Burgers’ (22% reduction), 9.93e-1 to 5.78e-2 on Allen-Cahn (94% reduction), and 9.82e-1 to 3.54e-3 on reaction (100% reduction). A characteristic-coordinate encoder for hyperbolic transport further reduces convection error from 5.14e-1 to 9.88e-5 (100% reduction). We report five-seed mean +/- standard deviation errors, Wilcoxon significance tests, full ablation studies, hyperparameter sensitivities, an extension to the 2D heat equation, and comparisons against parameter-matched baselines.
[LG-26] Backpropagation as a Nilpotent Linear System
链接: https://arxiv.org/abs/2607.11289
作者: Ahmed Boughammoura
类目: Neural and Evolutionary Computing (cs.NE); Machine Learning (cs.LG)
*备注: 23 pages, 3 figures
Abstract:Backpropagation is the computational engine of deep learning, yet its mathematical structure is typically treated as a procedural traversal of computational graphs. We present a global operator theory of the \emphF-adjoint framework, which reformulates the layerwise backward recursion of an L -depth feedforward network into a single linear system (I-\cB)\Xs=\bG , where \bG is a source vector. We prove that the global backward operator \cB is strictly block upper-triangular and nilpotent of index at most L . This nilpotency guarantees the exact termination of the Neumann series solution after at most L terms, revealing classical backpropagation to be mathematically equivalent to block back-substitution on an upper bidiagonal system. We formalise \emphF-symmetry – the condition in which the backward pass perfectly mirrors the forward pass – identifying orthogonal weight matrices as canonical examples. Through worked numerical examples, we demonstrate how this operator perspective exposes the single-path collapse of strictly feedforward networks and its breakdown in residual architectures. Finally, we leverage this compositional structure to rigorously derive the mechanics of residual networks (gradient highways) and transfer learning (gradient truncation). This framework elevates backpropagation from an algorithmic recipe to a global nilpotent-operator formulation.
[LG-27] rustworthy synthetic data for campaign decision support: strategy simulation fidelity and the PolicySynth framework
链接: https://arxiv.org/abs/2607.11269
作者: Tung Dang, TheHung Phung,Son Lam Nguyen,Tu Nguyen
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 15 pages, 4 figures
Abstract:Decision support systems (DSS) increasingly run retention what-if analysis on synthetic customer populations, because privacy constraints preclude unrestricted use of real data. Such a system is trustworthy only if the synthetic data lead managers to the same decisions as the real data would; yet prevailing criteria certify distributional similarity, not decision alignment, so a synthetic population can match every marginal distribution while still steering a marketing team toward the wrong campaigns. We close this decision-alignment gap with three contributions: strategy simulation fidelity (SSF), a criterion measuring how often the synthetic population yields the same go/no-go campaign decision as the real population; PolicySynth, a DSS framework whose generator is conditioned on the production churn scorer to align decision-relevant structure; and a three-axis reporting standard of decision alignment, membership-inference resistance, and novel-record rate as the minimum deployment quality gate. On a telecommunications churn corpus and a banking acquisition corpus, PolicySynth attains a mean SSF of 0.923 and 0.960, with seed-to-seed variance roughly ten times tighter than CTGAN on telecommunications and 2.5 times on banking. This stability is the deployable property: go/no-go recommendations shift by at most 1.2 percentage points between monthly retraining cycles, against 11.5 for CTGAN, a reversed recommendation on one campaign in nine. A bootstrap baseline matches PolicySynth on SSF yet copies real records verbatim and fails membership inference, evidence that no single axis suffices. PolicySynth reliably supports directional go/no-go screening; its ROI estimates diverge from real outcomes by 70 to 78% and require the volume correction we document.
[LG-28] FastTPS: An Optimized Method for LLM Token Phase for AI accelerators
链接: https://arxiv.org/abs/2607.11211
作者: Wenzong Yang,Danyang Zhang,Kun Cao,Tejus Siddagangaiah,Rajeev Patwari,Zhanxing Pu,Siyin Kong,Zijiang Yang,Hao Zhu,Varun Sharma,Yue Gao,Tianping Li,Fan Yang,Jicheng Chen,Yushan Chen,Fennian Zhao,Aaron Ng,Elliott Delaye,Ashish Sirasao,Sudip Nag
类目: Machine Learning (cs.LG)
*备注: 16 pages, 8 figures, 7 tables
Abstract:The popularity of large language models (LLMs) escalates an ongoing demand for effective inference. However, due to the sequential processing of tokens during the token phase in decoder-only LLMs inference, the inherent low parallelism leads to reduced throughput and suboptimal utilization of the computing units on artificial intelligence (AI) accelerators, particularly when handling long-sequence inputs that impose significant memory overhead. Recently, many reported methods have been developed as potential solutions, since they emerge with numeric deviation. This paper presents FastTPS, a high performance and low-precision loss method for accelerating the token-phase in LLM inference on general AI accelerators which includes three key components: (1) AI accelerator-enabled reloading-free KV Cache concatenation which decreases memory access overhead as well as enables full fusion of Attention, (2) high-efficiency and high-accuracy ‘RoPE’ attention based on the tiling optimized FLAT, and (3) highly-fused MLP with fine-grain pipeline scheduling. Our results confirm that FastTPS significantly alleviates memory bottlenecks in the token phase, delivering a 6x speed improvement (compared to none-fusion) on an AMD Ryzen AI 300 series NPU with BF16 precision while sustaining 93% peak memory bandwidth utilization during Phi3-mini-4k-instruct inference.
[LG-29] NeuroMem-FHP: A Likelihood-Free Deep Learning Framework for Parameter Estimation of Fractional Hawkes Process
链接: https://arxiv.org/abs/2607.11177
作者: Neha Gupta,Aditya Maheshwari
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 19 pages; 21 figures
Abstract:In this paper, we propose deep learning based NeuroMem-FHP framework for estimating the parameters of the fractional Hawkes process (FHP), a self-exciting point process that captures long-range dependence through a fractional Mittag-Leffler excitation kernel. Two neural architectures, namely a Long Short-Term Memory (LSTM) network and a Transformer, are developed to estimate the model parameters (\mu,\gamma,\alpha,\beta) directly from sequences of inter-arrival times without requiring computationally intensive likelihood optimization. Experiments on synthetic data that both neural models significantly outperform the classical Maximum Likelihood Estimation (MLE) method, with the Transformer achieving the highest estimation accuracy (MSE = 0.1634 ), followed by the LSTM (MSE = 0.1752 ), compared to MLE (MSE = 2.8032 ). An ablation study further examines the effects of key hyperparameters on model performance. The proposed framework is also on two real-world high-frequency datasets, namely AAPL NBBO transaction data and Montgomery County 911 emergency call records. Using a predictive validation approach, event sequences simulated from the estimated parameters closely reproduce the empirical distribution, tail behavior, and temporal dependence structure of the observed data. These results demonstrate that Transformer-based parameter estimation provides an accurate and efficient alternative to conventional estimation techniques for FHP and offers a promising framework for modeling event-driven systems with long-memory dynamics.
[LG-30] Rank-Conditioned Sample Reuse for the Plackett–Luce Best-of-K Objective
链接: https://arxiv.org/abs/2607.11146
作者: Melveena Jolly,Midhun Xavier
类目: Machine Learning (cs.LG); Methodology (stat.ME); Machine Learning (stat.ML)
*备注:
Abstract:We study the coupled objective J_K^WOR = E_S ~ PL-WOR_K[max_i in S R_i]: the expected maximum reward of a size-K Plackett-Luce draw without replacement, the law of Gumbel-Top-K / Stochastic Beam Search decoding. This estimand differs from the conventional i.i.d. objective J_K^iid = E[max_i=K R_i] targeted by existing sample-reuse Max@K estimators, and reusing their i.i.d. weights under the coupled sampler is provably biased (a closed-form three-item instance gives E[g_iid] = (4/5) grad J_K^WOR exactly; pass@K under the coupled sampler is the binary-reward special case). Generic joint-score REINFORCE is already unbiased for J_K^WOR; what it lacks is sample reuse. Our contribution is to instantiate standard rank-conditioned Horvitz-Thompson estimation for the J_K^WOR subset total: from one Gumbel-Top-n pool (nK) and its observed priority threshold we build an estimator that reuses all C(n,K) embedded K-subsets, unbiased with an unbiased exact score-function surrogate gradient, plus a reward-sorted Max-specific dynamic program that collapses the C(n,K)-term subset sum (with K!-cost set probabilities) exactly to a one-dimensional integral. A fixed-Q quadrature evaluation costs O(n log n + nKQ) arithmetic and is numerically, not algebraically, exact; no epsilon-approximation rate is certified. Each nonzero degree-K Horvitz-Thompson term has finite second moment exactly when n = 2K; under the same assumptions the full surrogate gradient has finite second moment whenever n = 2K (sharpness there is open). At K=1 the construction recovers classical priority sampling. All quantities require only the values and differentiable computation graphs of the n+1 drawn items’ probabilities, so finite structured sequence policies sampled by exact SBS are covered. A certified finite-Q quadrature bound and countably infinite support remain open. Validation code is included as ancillary files.
[LG-31] Learning Subgroup Relations Using Siamese Graph Neural Networks
链接: https://arxiv.org/abs/2607.11140
作者: Tal Weissblat
类目: Machine Learning (cs.LG); Group Theory (math.GR)
*备注:
Abstract:Determining whether one finite group is isomorphic to a subgroup of another is a fundamental problem in computational group theory. In this work, we propose a Siamese Graph Neural Network (Siamese GNN) for subgroup prediction using Cayley graph representations of finite groups. Each input group is represented by its undirected Cayley graph and encoded by one branch of a Siamese GNN to produce a graph embedding. The resulting graph embeddings are combined with algebraic features derived directly from the input groups to construct a joint feature vector, which is processed by a fully connected classifier to predict subgroup relations between finite groups. By integrating graph-based structural representations with algebraic features, the proposed framework provides a unified approach for learning subgroup relations from finite groups. Experimental results demonstrate the effectiveness of the proposed architecture, achieving a test accuracy of 95.9% (47/49) on an independent test set and illustrating the potential of geometric deep learning for subgroup prediction.
[LG-32] Comparison-Based Ordinal Learning for Proactive Driving Risk Assessment
链接: https://arxiv.org/abs/2607.11128
作者: Zhuoren Li,Yi Zhong,Weiqi Zhang,Xinrui Zhang,Lu Xiong,Chongfeng Wei,Bo Leng
类目: Robotics (cs.RO); Machine Learning (cs.LG)
*备注: 15 pages, 5 figures
Abstract:Real-time driving risk assessment provides an essential basis for proactive safety by identifying and quantifying the danger of ongoing road interactions before adverse outcomes occur. However, due to the scarcity of collision data and frame-level risk labels, existing driving risk assessment methods often rely on surrogate objectives, which may imperfectly align with true collision risk and not faithfully reflect the relative danger of driving interaction. This paper proposes a comparison-based ordinal risk learning framework that learns collision-relevant risk scores from pairwise supervision in driving data, directly modeling relative risk ordering without requiring numerical frame-level risk labels. We derive pairwise comparisons from three sources of event-structured driving data for such ordinal risk learning: temporal progression within safety-critical sequences, event-level contrast between dangerous and normal interactions, and physics-based counterfactual perturbations. On this basis, instantiations with three risk-scoring function parameterizations are implemented, including directly learning risk scores from comparison data, and aligning existing single or multiple surrogate-based risk models. The proposed framework is evaluated on the 100-Car and SHRP2 naturalistic driving datasets using a proactive collision warning task. Results show that the proposed framework improves high-recall risk discrimination, warning precision, and warning lead time over representative surrogate-based baselines across both in-distribution and out-of-distribution evaluations. These results suggest that the proposed framework can contribute to proactive safety research by providing more reliable risk assessment for automated driving systems and safety-critical driving interactions.
[LG-33] oolAtlas: Learning Once Reusing Everywhere with Tool-Side Memory
链接: https://arxiv.org/abs/2607.11126
作者: Yue Fang,Zhibang Yang,Fangkai Yang,Xiaoting Qin,Liqun Li,Qingwei Lin,Saravan Rajmohan,Dongmei Zhang
类目: Machine Learning (cs.LG)
*备注:
Abstract:Large language model (LLM) agents increasingly rely on external tools served by shared providers and accessed by heterogeneous downstream agents. Existing approaches improve tool use on the agent side through parameter updates, prompt refinement, or agent-side memory, making tool knowledge difficult to share and limited to behaviors observed in past tasks. We argue that reusable tool knowledge should instead be maintained by the tool provider. We introduce ToolAtlas, a graph-based framework that builds a persistent provider-side tool memory of tool capabilities, failure boundaries, and cross-tool compositions through execution-verified probing. At inference time, agents query the tool memory via adaptive graph traversal. Across two MCP-based benchmarks spanning eight services, ToolAtlas outperforms existing tool-side optimization and agent-side memory baselines by up to 21.61% in pass@1 and 18.61% in pass@4. The same tool memory also transfers across environment instances and agent frameworks without retraining or task-time exploration, yielding up to 24.16%/16.22% and 17.49%/14.27% relative gains in pass@1/pass@4, respectively. Ablation studies show that these gains arise from combining tool-centered memory organization with capability-guided execution probing. These results establish provider-side tool memory as an effective and reusable paradigm for tool servers. Our code is in: this https URL.
[LG-34] Implicit Neural Networks as Static Controllers: Certificates and Performance Separation
链接: https://arxiv.org/abs/2607.11122
作者: Giuseppe C. Calafiore,Laurent El Ghaoui
类目: ystems and Control (eess.SY); Machine Learning (cs.LG)
*备注:
Abstract:Implicit neural controllers (INCs) are static feedback laws that are evaluated through an algebraic fixed point equation; they include as special cases neural network controllers. We propose a so-called implicit representation of neural networks as a key enabling device that exposes the controller as a trainable linear interconnection closed through a known static activation map, thereby making well-posedness and Lyapunov/IQC analysis mathematically easy to handle. For finite-dimensional LTI plants, we first develop a rigorous analysis theory for a given INC, including Perron–Frobenius and norm conditions for well posedness, LMI/IQC certificates for exponential stability, and LMIs for discounted infinite-horizon quadratic performance. We then formulate synthesis as a certification-compatible heuristic search: training is carried out under explicit well-posedness constraints, implicit-differentiation formulas provide gradients, and the resulting controller is accepted only after independent post-training LMIs or regional admissibility checks are feasible. Finally, we establish constrained-control separation results: for a specific scalar unstable plant with hard actuator bounds, an INC achieves a strictly smaller discounted infinite-horizon cost than any admissible finite-order dynamic linear controller. Additional results cover quadratic state-input costs, comparison with linear static output feedback, and computable upper/lower-bound certificates. Numerical examples illustrate the mechanism and the resulting certified performance.
[LG-35] CA-DGCL: Dynamic Graph Continual Learning via Condensation and Attachment
链接: https://arxiv.org/abs/2607.11112
作者: Tingxu Yan Ye Yuan
类目: Machine Learning (cs.LG)
*备注:
Abstract:Dynamic graph continual learning (DGCL) is an effective manner for handling catastrophic forgetting in dynamic graphs. However, existing DGCL methods underutilize temporal information across graph snapshots. To address this critical issue, we propose a novel framework for Dynamic Graph Continual Learning via Condensation and Attachment (CA-DGCL). Specifically, CA-DGCL first condenses historical graph snapshots into compact semantic representations efficiently. Further, a cross-timestamp node chains is built to construct a third-order tensor and Tucker decomposition is applied to this tensor for obtaining stable node features, which encapsulate historical knowledge. Finally, these node features are used to generate new nodes and attached to the current graph for replaying of past information without compromising the new patterns. In addtion, a refined forgetting measure is introduced to make it more suitable for dynamic graph settings. Extensive experiments demonstrate that CA-DGCL outperforms baselines in forgetting suppression as well as maintain competitive accuracy, proving its efficacy for dynamic graph continual learning.
[LG-36] Neural Discovery of Memory and Nonlocal Kernels in Integro-Differential Equations with Constrained Kolmogorov–Arnold Networks
链接: https://arxiv.org/abs/2607.11110
作者: Aruzhan Tleubek,Salah A Faroughi
类目: Machine Learning (cs.LG); Numerical Analysis (math.NA)
*备注:
Abstract:Discovering the memory or nonlocal kernel governing an integro-differential equation (IDE) from sparse and noisy observations is an ill-posed inverse problem. Existing identification methods often rely on problem-specific analytical derivations, specialized observation requirements, or restrictive assumptions about the kernel, limiting their applicability across different classes of IDEs. In this work, we propose a differentiable-solver-based framework for discovering memory and nonlocal kernels directly from spatiotemporal observations. Within the solver, the unknown kernel is represented using a constrained Kolmogorov–Arnold Network (KAN) parameterization, with the physical constraints imposed through two different approaches: a Bernstein-polynomial-based Monotone–Convex KAN (MC-KAN), whose coefficient constraints enforce positivity, monotonic decrease, and convexity by construction, and a Chebyshev-based KAN (Cheb-KAN), in which the same properties are encouraged through soft penalty terms. After training, symbolic regression is applied to the learned kernels to obtain interpretable closed-form representations. We evaluate both methods on benchmarks spanning a one-dimensional Volterra equation, a one-dimensional viscoelastic wave partial integro-differential equation, and a two-dimensional nonlocal reaction-diffusion equation with an anisotropic coupled kernel. For the 1D problems, both methods recover the correct kernel functional form and achieve comparable solution-reconstruction accuracy. In contrast, for the sparse and noisy 2D nonlocal problem, the hard-constrained MC-KAN consistently achieves lower kernel reconstruction errors than the soft-constrained Cheb-KAN. Our results demonstrate that enforcing physically motivated shape constraints by construction provides greater robustness than soft penalties for multidimensional kernel discovery from sparse and noisy observations.
[LG-37] A Novel Graph Fraud Detector via Grouped Attribute Completion and Confidence-Aware Contrastive Learning
链接: https://arxiv.org/abs/2607.11107
作者: Junpeng Wu,Ye Yuan
类目: Machine Learning (cs.LG)
*备注: 9 pages,3 figures
Abstract:Graph fraud detection plays a pivotal role in safeguarding the security and integrity of modern digital ecosystems. Graph Neural Networks (GNNs) are commonly adopted for graph fraud detection. However, the practical performance of existing GNN-based detectors is severely hindered by incomplete node attributes and extreme class imbalance within graphs. To mitigate these limitations, this paper proposes a novel framework for Graph Fraud Detection with Grouped attribute completion and Confidence-aware Contrastive learning, named GFD-GC. Specifically, it first imitates heterogeneous neighborhood structures to implement group-wise aggregation, which obtains informative complete node features by capturing fine-grained graph contextual patterns. Further, it introduces a confidence-aware supervised contrastive learning strategy to augment scarce labeled fraud nodes with high confidence pseudo-fraud nodes, which enhances the compactness of fraud representations and their separability from non-fraud nodes. Extensive experiments demonstrate the superiority of the proposed GFD-GC over state-of-the-art baselines on the graph fraud detection task, thereby providing an effective solution for real-world fraud scenarios.
[LG-38] Multi-dimensional training-priority weighting based on physical information propagation paths: a unified residual-weighting framework for physics-informed neural networks
链接: https://arxiv.org/abs/2607.11094
作者: Zhangyi Lian,Xinda Dong,Wenxuan Huo,Weifeng Huang,Gang Zhu,Qiang He
类目: Machine Learning (cs.LG)
*备注:
Abstract:Physics-informed neural networks (PINNs) have shown promise for solving partial differential equations (PDEs); however, their synchronous optimization treats residuals of different regions and constraints equally, which is inconsistent with the progressive “from source to response” physical information propagation path, degrading training stability and accuracy. Existing causal training methods focus mainly on the temporal dimension, lacking a unified characterization of spatial and boundary dimensions. To address this, we define a unified class of training priorities according to the physical information propagation path: premise regions should be learned before dependent regions; temporal, spatial, and boundary priorities are instances of this principle. Using neural tangent kernel (NTK) dynamics, we theoretically analyze why standard PINNs do not obey this priority: their residual convergence order is governed by the NTK spectrum and is independent of the propagation path. Accordingly, we propose a unified multi-dimensional priority-constraint framework that partitions the domain along the propagation path and constructs negative-exponential residual weights, converting the physical propagation order into a training priority. For cases with coexisting priorities, we introduce a directional compatibility coefficient to clarify that “orthogonal directions can be coupled multiplicatively in synergy, whereas coaxial opposite directions cannot.” Benchmark cases show that this method consistently improves the convergence behavior and prediction accuracy of PINNs on problems with clear propagation paths or constraint-dominated structures, without modifying the network architecture and with controllable additional computational cost.
[LG-39] Adapting Evidential Neural Networks to Test-Time Neighbor Fusion Improves Molecular Property Prediction
链接: https://arxiv.org/abs/2607.11091
作者: Cameron Gruich,Weichi Yao,Yixin Wang,Bryan Goldsmith
类目: Machine Learning (cs.LG); Biomolecules (q-bio.BM); Machine Learning (stat.ML)
*备注: 45 pages (18 main, 27 SI); 11 figures (7 main, 4 SI); 14 tables (0 main, 14 SI); 61 equations (15 main, 46 SI)
Abstract:A trained molecular property model can be refined at test time by correcting each prediction with the measured labels of the most similar training molecules, a retraining-free procedure we call neighbor fusion; evidential neural networks make it principled by using their aleatoric and epistemic uncertainty to parameterize a Bayesian update. Our main contribution, PG-EVIKAL, learns a property-distance metric to re-rank structurally similar neighbors by their property relevance before fusion, building on EVIKAL (scalar Kalman filter) and GP-EVIKAL (Gaussian process variant handling correlated neighbors). Evaluated on 16 molecular datasets, PG-EVIKAL reduces RMSE relative to the evidential model baseline on 14 of them, with a median reduction of 19.4%, and improves calibration; in sequential-assay scenarios it further incorporates newly measured molecules, refining predictions as they arrive without retraining. This work demonstrates that evidential uncertainty decomposition is not merely a calibration objective but an actionable inference resource that enables test-time refinement of molecular property predictions.
[LG-40] Link Adaptation Using Joint-Thompson Sampling
链接: https://arxiv.org/abs/2607.11075
作者: Vignatha Vinjam,Manjunath Kolavennu,Myna Vajha,Karthik Periyapattana Narayanaprasad
类目: Machine Learning (cs.LG)
*备注: SPCOM 2026
Abstract:The choice of Modulation and Coding (MCS) type for a particular channel condition is made through link adaptation (LA) algorithms that operate at the MAC layer. These algorithms rely on the ACK/NACK statistics and the channel quality index (CQI) feedback. Several existing works model LA as a multi-armed bandit (MAB) problem across cellular and Wi-Fi links. In the MAB formulation, each available MCS is a Bernoulli arm parameterized by its transmission success probability, and the goal is to design a selection strategy that accrues maximum reward. Several popular MAB algorithms, such as upper confidence bound (UCB) and Thompson Sampling (TS), have been proposed in the literature. Using the fact that MCS success probabilities are ordered, we propose the Joint-Thompson Sampling (Joint-TS) algorithm. Unlike classical TS, which assumes independent Beta distributions for each arm, Joint-TS utilizes a multivariate ordered Beta distribution as the prior to preserve the inherent monotonicity of success probabilities. Our simulation results show that while existing MAB algorithms fail in specific scenarios, Joint-TS delivers competitive throughput with robust, consistent performance in all scenarios.
[LG-41] AeroMELD: A Linear Embedding of Aerosol Populations for Diagnostics and Latent Dynamics
链接: https://arxiv.org/abs/2607.11073
作者: Ehsan Saleh,Saba Ghaffari,Wenhan Tang,Jeffrey H. Curtis,Lekha Patel,Peter A. Bosler,Nicole Riemer,Matthew West
类目: Machine Learning (cs.LG); Atmospheric and Oceanic Physics (physics.ao-ph)
*备注: 34 pages, 12 figures
Abstract:Accurately representing atmospheric aerosol populations is essential for simulating aerosol-cloud interactions, radiative forcing, and ice nucleation, yet existing reduced schemes impose structural assumptions that limit their ability to capture composition diversity and mixing state. Machine-learning approaches offer more flexible representations, but standard autoencoders do not preserve the mathematical structure of aerosol populations and therefore cannot support physically meaningful process operators. We introduce AeroMELD (Aerosol Measure Embedding for Latent Dynamics), a mathematically grounded framework for constructing low-dimensional latent variables that retain this structure. We show that any permutation-invariant linear encoder must take a scale-shape decomposition, with total number concentration represented explicitly and latent shape given by a barycentric combination of per-particle embeddings. This aggregated latent state retains the diagnostic expressiveness of a Deep Sets model by moving the nonlinear post-aggregation stage into the learned diagnostic map while preserving latent linearity. Using particle-resolved data as ground truth, we encode weighted particle populations directly rather than binned aerosol states; size-resolved mass and number distributions serve only as diagnostic targets and visual summaries. The latent space accurately reconstructs these distributions, CCN spectra, optical coefficients, and immersion-freezing behavior while preserving the linear population structure needed for hybrid ML-physics models. Although the experiments focus on diagnostic reconstruction, the embedding is designed so that emissions and mixing can be represented exactly and nonlinear microphysical processes learned in a controlled latent space. This work establishes a foundation for learning aerosol-process evolution directly in latent space.
[LG-42] abPFN beyond Tabular Data: Calibration and Accuracy on Multimodal Embeddings
链接: https://arxiv.org/abs/2607.11007
作者: Jingxiang Zhang,Lujia Zhong,Zijie Zhu,Shuo Huang,Yuang Xu
类目: Machine Learning (cs.LG)
*备注: 19 pages, 13 figures, 10 tables. Jingxiang Zhang and Lujia Zhong contributed equally. Code: this https URL
Abstract:Few-shot multimodal classification commonly attaches a lightweight head, such as k -nearest neighbors, logistic regression, or a linear SVM, to a frozen pretrained encoder. Although computationally efficient, these heads can produce poorly calibrated confidence scores, limiting their reliability in calibration-sensitive applications. We evaluate TabPFN as a plug-and-play, zero-gradient classification head for frozen image, text, and audio encoders. Across 22,820 evaluation episodes spanning 14 datasets, 11 encoders, and three modalities, TabPFN achieves the best mean rank among nine classification heads on both negative log-likelihood (NLL) and expected calibration error (ECE). At a representative setting, it reduces NLL by 48–62% and ECE by 2.1–5.3 \times relative to the average of the eight baselines while matching or exceeding their average accuracy. Its accuracy advantage is conditional, concentrating at moderate-to-high shot counts and low-to-moderate feature dimensions ( k \ge 50 , d \le 32 ), and diminishing when labeled data are scarce, feature dimensions are high, or competing methods approach ceiling accuracy. In targeted backbone-adaptation experiments, replacing the trained linear head with TabPFN substantially improves calibration while preserving competitive accuracy. These results provide empirical guidance for using TabPFN as a training-free head in calibration-sensitive multimodal classification. To support transparency and reproducibility, we publicly release the source code, experiment configurations, and evaluation scripts in our GitHub repository: this https URL.
[LG-43] A Multi-Agent Framework for Zero-Dimensional Reduced-Order Model Planning
链接: https://arxiv.org/abs/2607.10994
作者: Bingteng Sun,Hao Yin,Yiling Chen,Renjie Xiao,Lei Xie,Shanyou Wang,Ruonan Wang,Shubao Chen,Qingzong Xu,Lin Lu,Qiang Du,Junqiang Zhu
类目: Machine Learning (cs.LG)
*备注:
Abstract:Zero-dimensional reduced-order models (0D ROMs) are central to multi-dimensional design workflows for high-end complex equipment. However, the planning process currently relies on manual expertise, limiting topological exploration and prolonging iterations. Even traditional optimization methods such as Genetic Algorithms (GA) are typically confined to local parameter tuning. Although Large Language Model (LLM) agents have shown promise in exploring large sample spaces, and frameworks such as Chain of Thought (CoT) and Reason and Act (ReAct) improve reasoning reliability, while Retrieval-Augmented Generation (RAG) overcomes domain knowledge barriers, a single agent still falls short for the long-horizon and highly coupled nature of complex 0D ROM planning. This paper proposes the Zero-dimensional reduced-order model CO-Planning framework (Z-COPA), a multi-agent architecture featuring a Symbolic Action Graph Engine (SAGE) and a MILP-Guided Navigation (MGN) optimizer. Its core innovation is a dedicated graph representation method that accurately encodes the 0D flow network topology, converting the empirical planning process into a rigorous graph structure optimization problem. We validate the forward and inverse design capabilities and generalization performance of Z-COPA on two real aircraft engine secondary-air systems, two IEEE power-distribution reconfiguration benchmarks, and two water-distribution network benchmarks. The results show superior task completion quality, obtaining the best performance in both forward and reverse design of air systems. Z-COPA disrupts the traditional 0D model planning paradigm, providing a new technical approach for exploring broader topological space and achieving highly automated, globally optimal air system architectures.
[LG-44] Enhanced Byzantine-Robust Federated Learning Via Truncated-Quadratic Loss for Heterogeneous Data
链接: https://arxiv.org/abs/2607.10970
作者: Zhi-Yong Wang,Hao Nan Sheng,Werner Stefan,Hing Cheung So,Linqi Song,Weitao Xu
类目: Machine Learning (cs.LG)
*备注:
Abstract:Federated learning distributes data among n clients, making it vulnerable to malicious attacks and data heterogeneity, which together pose challenges for robust learning. To tackle this issue, centered clipping and Huber aggregators have been exploited for Byzantine robustness. In this paper, we first demonstrate their equivalence via convex conjugate theory, and show that they can yield biased solutions in the presence of outliers, leading to failure under high data heterogeneity and a substantial fraction of outliers. Next, we propose a new robust aggregation rule that utilizes the truncated-quadratic (TQ) loss, effectively mitigating the biases of existing methods, such as centered clipping and Huber aggregators. We show that our aggregator achieves order-optimal Byzantine-robust learning under nonconvex loss functions and heterogeneous data, ultimately enhancing the reliability of federated learning systems. Additionally, we provide a robust deviation estimation strategy for TQ, demonstrating its effectiveness. Furthermore, we show that TQ maintains robustness even when only an estimate of the number of Byzantine clients is available. Finally, experimental results on MNIST, Fashion-MNIST, and CIFAR-10, indicate that our aggregator provides better robustness performance than the competing techniques.
[LG-45] Reinforcement Learning for Execution under Dynamic Fees in a Closed-Loop DEX Simulator
链接: https://arxiv.org/abs/2607.10960
作者: Wen-Ting Wang
类目: Machine Learning (cs.LG); Computational Finance (q-fin.CP); Machine Learning (stat.ML)
*备注:
Abstract:Trader-facing dynamic fees are increasingly proposed for automated market makers (AMMs), but historical data do not identify how order flow would respond: trader-facing fees do not vary, trader types are latent, and a replayed tape is not a sequential decision environment. We therefore construct a minimal closed-loop simulator in which the missing signal exists by construction: two constant-product pools repriced by an equilibrium-inspired dynamic-fee rule, fee-sensitive noise flow, and closed-form CEX–AMM arbitrage. Equilibrium is used as a closure principle, not as an object the trader learns. Against a tuned benchmark ladder of schedule, planning, lookahead, and tabular policies, a small DQN is the only evaluated valid policy whose paired improvement over tuned one-step routing excludes zero. On a reserved final block of 1,000 seeds with completion forced to 1.0 for every policy, it reduces implementation shortfall under every tested intra-step ordering, by 13.3\bps of order notional under the pre-specified agent-last ordering, and the edge is concentrated in, and learned from, dynamic-fee environments: under constant fees the paired difference is indistinguishable from zero. The result is model-conditioned counterfactual evidence about execution control in AMMs, not evidence about historical traders, equilibrium play, or deployable profit.
[LG-46] WSqD: A Horizon-Free Learning Rate Schedule for Large Model Training
链接: https://arxiv.org/abs/2607.10959
作者: Jianhao Ma,Yuxin Chen
类目: Machine Learning (cs.LG); Optimization and Control (math.OC); Machine Learning (stat.ML)
*备注:
Abstract:Standard learning rate schedules such as cosine annealing are tied to a fixed training horizon, limiting their ability to accommodate post hoc horizon extension. Warmup-stable-decay (WSD) partially addresses this issue by maintaining a long constant-rate phase before a short linear cooldown, allowing training to resume from a pre-decay checkpoint. However, its peak learning rate is still tuned based on the original training horizon and can become suboptimal when training is extended. Motivated by stochastic convex optimization, we propose WSqD (Warmup with Square-root base and linear Decay), a learning rate schedule that replaces WSD’s constant stable phase with a shifted inverse-square-root base while retaining the final linear cooldown. In the stochastic convex setting, WSqD provably attains the minimax-optimal O(1/\sqrtT) last-iterate convergence rate. Importantly, its base learning rate schedule is horizon-independent, and the training horizon is needed only to determine when to begin the final cooldown. Empirically, on language-model pretraining using the SlimPajama corpus, WSqD matches or outperforms carefully tuned WSD and other baselines across multiple training horizons while reusing a single peak learning rate.
[LG-47] Sticky Jump Diffusions: A Unifying View of Masked Continuous and Hybrid Diffusion
链接: https://arxiv.org/abs/2607.10951
作者: Pascal Jutras-Dubé,Patrick Pynadath,Jeremy Lu,Yuan Gao,Ruqi Zhang
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:We introduce Sticky Jump Diffusions (SJDs), continuous-time Markov processes on \mathbb R^d whose discrete anchors are token embeddings. In forward time, anchors release their mass at a hazard rate and the released mass diffuses in the continuous ambient space; time reversal couples a score-driven SDE with a sticky jump kernel whose rate and destination are fixed by flux balance with the forward law. We estimate the score and the per-anchor reverse hazards from a single denoising classifier via Denoising Hazard Matching, the hazard analogue of denoising score matching, with simulation-free cross-entropy training. SJD recovers masked diffusion, continuous diffusion, and hybrid diffusion as limits. Its reversal explains features that each family treats as given: the mask of masked diffusion carries no evidence about the source token because the unsticking kernel of every anchor collapses to the same absorbing point; the terminal projection of continuous diffusion is required due to the absence of atoms in its forward marginal, without which flux balance yields no reverse jumps; and the update rules of hybrid diffusion (commit rate, destination, and drift) all follow from flux balance rather than from separate design. Beyond these limits, the unsticking kernel becomes a design space: a cross-position blending corrupts each position toward a blend of its neighbors’ clean values or embeddings, turning dependency structure such as spatial locality or a constraint graph into an inductive bias of the corruption itself, and improves over the identity-kernel hybrid on CIFAR-10, Text8, and Sudoku.
[LG-48] Bandit PCA with Minimax Optimal Regret
链接: https://arxiv.org/abs/2607.10936
作者: Moïse Blanchard,Dmitrii Ostrovskii,Aadirupa Saha
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:We study the bandit-feedback version of online principal component analysis (Bandit PCA): in each round t = 1,\dots,T , the adversary selects a d \times d symmetric gain matrix G_t with spectrum in [0,1] and rank at most r ; the learner simultaneously selects a unit vector w_t \in S^d-1 and receives the reward w_t^\top G_t w_t . The learner receives no other feedback, and aims to minimize the regret against the best unit vector in hindsight. This problem was introduced by Kotlowski and Neu (2019), who gave an algorithm with regret O(d\sqrtrT \log T) and showed the lower bound of \Omega(r\sqrtT/\log T) . We improve upon both of these bounds and essentially bridge the gap between them, establishing the minimax regret of order r\sqrtdT up to polylogarithmic factors in d and T . The upper bound is attained by a novel algorithm, which combines online mirror descent on the spectrahedron of (real) density matrices with a multiscale exploration scheme in which the eigenspaces with different spectral magnitudes are updated at different rates. For the lower bound, we construct an adaptive adversary that refines a hidden large-reward subspace based on the learner’s actions, in such a way that low regret is impossible without estimating the subspace; as a result, lower-bounding the regret reduces to studying the arising subspace estimation problem. Finally, we discuss connections of Bandit PCA with adaptive-measurement quantum tomography.
[LG-49] he Spectral Structure of Latent Treatment Effects
链接: https://arxiv.org/abs/2607.10926
作者: Hamza Virk,Bijan Mazaheri,Yihren Wu
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:Identifying heterogeneous treatment effects under unobserved confounding is central in observational causal inference. In proxy models with a discrete latent confounder, prior Synthetic Potential Outcomes (SPO) [Mazaheri-Squires-Uhler '25] recover the mixture of treatment effects through recursively constructed scalar moments. We show that this sequence is one projection of a more fundamental object. Under the same population factorization assumptions, there is an exact compressed observable operator: after projecting onto the shared proxy signal subspace, the difference of two treatment-arm quotient operators is similar to the diagonal matrix of latent treatment effects. Its eigenvalues are the latent effects; its lifted left eigenvectors, after anchor normalization, recover the target-proxy feature matrix and then the latent mixture proportions. Every scalar SPO moment is a bilinear functional of a power of this operator. The resulting estimator handles overcomplete proxy systems, replaces high-order scalar inversion with finite-dimensional spectral analysis, and admits high-probability first-order perturbation bounds for treatment effects, feature rows, and simplex-projected mixture weights.
[LG-50] Infrared Organization and Critical Cognitive Field Formation in Transformer Dynamics
链接: https://arxiv.org/abs/2607.10923
作者: Byung Gyu Chae
类目: Machine Learning (cs.LG)
*备注: 37 pages, 35 figures
Abstract:Large language models exhibit remarkable emergent behaviors, yet the physical mechanism governing their collective dynamics remains poorly understood. Cognitive Field Theory predicts that learning reorganizes the time-scale density of states (TDOS) through the infrared accumulation of slow relaxation modes, thereby enhancing the memory self-energy, reducing the cognitive forgetting gap, and strengthening the collective susceptibility. Using publicly available Pythia language models, we extract relaxation spectra directly from Transformer layer Jacobians throughout training, network depth, and model scale, allowing the TDOS, memory self-energy, forgetting gap, memory kernel, and infrared critical exponent to be measured quantitatively. The measurements reveal progressive infrared accumulation of slow relaxation modes, producing an approximately flat infrared TDOS with ( \rho(\lambda)\sim\lambda^-0.1 ) and scale-free memory kernels ( K(t)\sim t^-1. ) The memory self-energy exhibits a pronounced transient maximum during early optimization before relaxing toward a metastable near-critical regime, corresponding to the smallest cognitive forgetting gap and the largest collective susceptibility predicted by Cognitive Field Theory. These observations provide quantitative experimental evidence that Transformer dynamics are governed by infrared collective organization. The reproducibility of the same dynamical behavior across training, network depth, and model scales suggests that infrared slow-mode organization represents a universal collective principle of Transformer dynamics.
[LG-51] Singular perturbations and hierarchical learning in two-layer neural networks
链接: https://arxiv.org/abs/2607.10869
作者: Cédric Gerbelot,Jean-Christophe Mourrat
类目: Machine Learning (cs.LG); Dynamical Systems (math.DS); Optimization and Control (math.OC)
*备注: 37 pages
Abstract:We study the population gradient flow of an infinitely wide two-layer neural network learning a misspecified single-index model in high dimension. The two layers are optimized jointly, with a perturbative parameter tuning the relative training speed between the first and second layer. This setting was considered by Berthier, Montanari and Zhou in \citeberthier2024learning, who conjectured a hierarchical learning scenario with explicit timescales as the second layer is trained faster than the first. In this paper, we prove that the constant and linear components of the hidden link function are indeed recovered within the predicted timescales, at sharp explicit thresholds. We then analyze the onset of learning of the quadratic component and show that the components learned at earlier stages continue to influence the dynamics in an essential way. Our proof is based on quantitative approximation results for singularly perturbed flows evolving near a manifold defined by integral constraints. At a phenomenological level, we also show that the empirical measure of the weights displays singular behaviour when reaching the quadratic component of the hidden link, with a small fraction of neurons growing significantly while the remaining ones rearrange to preserve the components already learned.
[LG-52] Reliability Scaling Laws for Quantized Large Language Models
链接: https://arxiv.org/abs/2607.10855
作者: Sirine Ayadi,Sándor Daróczi,Stephan Günnemann,Bertrand Charpentier
类目: Machine Learning (cs.LG)
*备注: Transactions on Machine Learning Research (TMLR), 2026
Abstract:Quantization is a powerful strategy to build capable and resource-efficient large language models (LLMs) by reducing the bitwidth of the parameters. While quantized LLMs achieve state-of-the-art performance on unperturbed inputs using standard predictive metrics, their performance on perturbed inputs, measured using reliability metrics, remains underexplored, despite its importance for reliable deployment. To address this gap, we first conduct a comprehensive reliability evaluation of quantized LLMs consisting of three key components: (1) Uncertainty: We assess the trustworthiness of LLMs quantized to 2, 3, 4, and 8 bits using six different quantization methods, employing established uncertainty metrics. (2) Calibration: We assess how well-calibrated the uncertainty estimates of quantized models are across model scales and bit precisions. (3) Robustness: We design character-level and word-level input perturbations to evaluate the reliability of quantized models under semantically-preserving variations in the inputs that arise in real-world applications. Second, we characterize how reliability scales with the total number of model bits. Our study reveals that while the performance scales monotonically with the total number of bits, the reliability scalings are nonlinear. A reliability peak occurs for 4-bit quantized models, indicating that quantizing moderately sized models offers the best reliability-efficiency trade-off. Additionally, our empirical findings reveal that quantization enhances the robustness of LLMs to natural input perturbations.
[LG-53] Predictive Divergence Masks for LLM RL
链接: https://arxiv.org/abs/2607.10848
作者: Xiangxin Zhou,Jiarui Yao,Penghui Qi,Bowen Ping,Jiaqi Tang,Haonan Wang,Tianyu Pang
类目: Machine Learning (cs.LG)
*备注:
Abstract:Reinforcement learning for large language models (LLMs) typically relies on trust-region masks to stabilize off-policy updates. The dominant PPO-style approach uses the sampled-token importance ratio for two criteria: a proximity criterion, which asks whether the policy has moved too far from the behavior policy, and a direction criterion, which asks whether the update pushes it farther away. Recent work DPPO improves the proximity criterion by replacing PPO’s ratio-based test with a probability divergence between the behavior and training policies. However, its direction criterion is still inherited from PPO. A token can be masked only when the sampled-token importance ratio moves away from one. We observe that this ratio-based direction criterion is a single-sample proxy that can disagree in sign with the change of the divergence that defines the proximity criterion. We therefore propose the predictive divergence mask, which asks whether the next policy-gradient step will increase or decrease the same divergence used by the trust region. For the discrete softmax policies used in LLM RL, we derive this prediction in closed form. Because production rollout engines expose only a truncated (top-K) view of the vocabulary, we develop two lightweight top- K estimators for this prediction. Detailed analysis shows the divergence-based direction is better aligned with the realized change of the divergence than the sampled ratio, and the resulting masks improve RL training across model scales and precision settings.
[LG-54] Graph Neural Networks for RFID-Based Spatial Geometry Inference in Spatial AI Systems
链接: https://arxiv.org/abs/2607.10822
作者: Curtis Shull,Merrick Green,Roy Rucker
类目: Machine Learning (cs.LG)
*备注: 38 pages, 10 figures. Introduces a Graph Neural Network framework for RFID-based spatial geometry inference using indoor floorplan semantics and digital twin representations
Abstract:Indoor spatial understanding remains a fundamental challenge for intelligent systems operating in physical environments. Traditional RFID localization techniques typically estimate positions of tags using signal strength measurements but fail to capture higher-order spatial relationships between objects and infrastructure. Recent work on RFID and wireless indoor localization has increasingly emphasized robust learning under noisy propagation, while recent graph-based localization methods demonstrate the value of relational modeling over isolated samples. This paper introduces a graph-based learning framework that leverages Graph Neural Networks (GNNs) to infer spatial geometry from RFID observations. Rather than predicting isolated coordinates, the proposed system models relationships between RFID readings, antennas, and physical structures within an indoor floorplan. This framing is aligned with recent graph-based indoor positioning and graph construction literature, where topology is a first-class source of information for downstream inference. The approach integrates signal strength data, floorplan semantics, and spatial constraints into a graph representation where nodes correspond to RFID observations and edges encode proximity and contextual relationships. A GNN is then trained to predict geometric patterns such as linear trajectories, rectangular bounding regions, and movement paths of objects in space.
[LG-55] Lower Bound on the Cumulative Constrained Violation for the OGDProjection algorithm for Constrained Online Convex Optimization (COCO)
链接: https://arxiv.org/abs/2607.10808
作者: Haricharan Balasundaram,Karthick Krishna Mahendran,Rahul Vaze
类目: Machine Learning (cs.LG)
*备注:
Abstract:The problem of constrained online convex optimization is considered, where at each round, once a learner commits to an action x_t \in \mathcalX \subset \mathbbR^d , a convex loss function f_t and a convex constraint function g_t that drives the constraint g_t(x)\le 0 are revealed. The objective is to simultaneously minimize the static regret and cumulative constraint violation (CCV) compared to the benchmark that knows the loss functions and constraint functions f_t and g_t for all t ahead of time, and chooses a static optimal action that is feasible with respect to all g_t(x)\le 0 . Currently, the best known algorithm is OGD+Projection algorithm of [Vaze and Sinha, 2025] that has simultaneous regret of O(\sqrtT) and CCV of O(T^1/3) for d=2 [Balasundaram et al., 2026], and simultaneous regret of O(\sqrtT) and CCV of O(\sqrtT) for any d [Sarkar and Sinha, 2026]. In this paper, we show that the CCV of the OGD+Projection algorithm is \Omega (T^\fracd-12d) . This is the first such lower bound result.
[LG-56] When does distribution shift break graph neural networks calibration?
链接: https://arxiv.org/abs/2607.10804
作者: Abderaouf Bahi
类目: Machine Learning (cs.LG)
*备注:
Abstract:Graph neural networks (GNNs) are increasingly deployed in real-world applications where distribution shift is un-avoidable. However, how such shifts affect model calibration, defined as the agreement between predictive confidence and actual accuracy, remains poorly understood, and existing graph calibration methods typically rely on labeled validation data from the deployment distribution. In this work, I present the first closed-form theoretical characterization of GNN calibration under distribution shift. I show that calibration is governed by a single scalar quantity that explicitly depends on structural changes between the source and target graphs, as well as feature quality. This characterization precisely identifies when a model becomes over-confident, under-confident, or remains calibrated, and directly yields the optimal temperature scaling strategy. I further extend the analysis to graph convolutional networks with symmetric normalization, multi-class classification, and covariate shift, and derive a theoretical upper bound on the expected calibration error. My analysis also reveals that, under homogeneous distribution shift, a single global temperature is theoretically optimal, providing a principled explanation for why more complex node-wise recalibration methods offer no additional benefit. Building on these theoretical insights, I propose STAC, a source-free, label-free calibration method. Experiments on synthetic benchmarks demonstrate substantial calibration improvements, while evaluations on five real-world graph datasets show that reliable calibration without target labels remains challenging despite the strong predictive power of the theory.
[LG-57] Q-Learning Lab: Teaching Reinforcement Learning Through Learner-Generated Trace Analysis
链接: https://arxiv.org/abs/2607.10802
作者: Ekkachai Jueng
类目: Computers and Society (cs.CY); Machine Learning (cs.LG)
*备注: 12 pages, 5 figures
Abstract:Reinforcement learning is usually introduced through the Bellman update, yet the equation often remains abstract to undergraduates: they watch policy arrows converge but rarely observe how each value is computed or why an action is chosen. We present Q-Learning Lab, a single-file, browser-based, bilingual (Thai/English) tool for teaching tabular Q-learning that requires no installation. Beyond the usual gridworld visualization - color-coded Q-values and policy arrows on a 5 \times 5 world - the tool exposes a live Bellman-substitution panel showing the numeric update at every step, and logs each transition, including the full pre-action Q-row, the greedy-versus-random decision under \varepsilon -greedy exploration, and wall-collision events, into an exportable trace. The central contribution is a learn-export-analyze loop: learners run their own agent, export the complete trace as CSV, and analyze it themselves, producing learning curves, value heatmaps, and visitation maps, turning a passive demonstration into a source of learner-generated data for reflective inquiry. We validate the tool without human-subject data through three complementary evaluations: (i) correctness of the learned values and policy against a value-iteration ground truth on the identical MDP; (ii) hyperparameter sweeps over \alpha , \gamma , and \varepsilon showing that every pedagogical claim the tool makes is reproducible; and (iii) a reward-editing study that uses the ground-truth optimal policy to separate two behaviorally identical but diagnostically opposite failure modes - an exploration failure versus genuine reward misspecification - that a single edited reward can produce. We also compare the tool against existing gridworld visualizers, describe its grounding in learning-by-doing pedagogy, and include a 50-minute lesson plan. The tool and all experiment code are openly available.
[LG-58] Hierarchical Bayesian Quadrature UAI2026
链接: https://arxiv.org/abs/2607.10793
作者: Tim Weiland,Toni Karvonen,Philipp Hennig
类目: Machine Learning (cs.LG); Numerical Analysis (math.NA)
*备注: 17 pages (9 main text), 10 figures. Accepted at the 42nd Conference on Uncertainty in Artificial Intelligence (UAI 2026)
Abstract:Numerical integration is a cornerstone of various scientific computing applications, such as engineering simulations and model evidence computations in probabilistic machine learning. Bayesian Quadrature uses Gaussian process surrogates that explicitly encode structural assumptions about the integrand to obtain integral estimates with quantified uncertainty. These surrogates are predominantly based on stationary covariance functions, which results in model misspecification for integrands exhibiting nonstationary behavior. We tackle this issue through an adaptively growing, tree-based partition of the integration domain into local stationary models. Our method recombines the local integral estimates through a hierarchy of GP conditioning that reintroduces cross-subdomain correlations, while model selection criteria control the tree growth to avoid unnecessary partitioning. The resulting algorithm is simple, requires no MCMC, and adapts its evaluation budget to local integrand complexity. On benchmark integration problems and a model evidence computation for an epidemiological model, Hierarchical Bayesian Quadrature achieves substantial gains over standard Bayesian Quadrature on nonstationary integrands while matching its performance on stationary ones.
[LG-59] he VC dimension of partial concept classes via Radons theorem
链接: https://arxiv.org/abs/2607.10751
作者: Grigory Ivanov,Attila Jung,Márton Naszódi
类目: Machine Learning (cs.LG); Combinatorics (math.CO); Functional Analysis (math.FA)
*备注: 22 pages
Abstract:Following Alon, Hanneke, Holzman, and Moran (FOCS 2021), we define a partial concept class (PCC) as a family of partial functions (f: V\to\0,1,\ast\); equivalently, its concepts partition the ground set into black ( f^-1(1) ), grey ( f^-1(\ast) ), and white parts ( f^-1(0) ). Its VC dimension is defined by shattering sets on which the value \ast is not taken. We study two geometric PCCs in real Banach spaces, both with a margin (\delta0): expanded half-spaces, where the grey part is a strip of width at least (\delta) adjacent to a half-space, and expanded balls, where the grey part is an annulus of width (\delta) around a unit radius ball. Our main results are dimension-free upper bounds on the VC dimension of the PCC of expanded balls in (L_p\parenth\mu), (1\le p\infty), including the non-Euclidean and algorithmically particularly relevant case (\ell^d_1). These bounds depend on the margin and on the radii, but not on the ambient dimension or the underlying measure space. These are extensions of the work of Bourneuf, Charbit, and Thomassé (FOCS 2025) who studied the PCC of expanded balls in Euclidean space, that is, \ell_2^d . We also prove lower bounds on the VC dimension that match the upper bounds in terms of the margin parameter \delta . Finally, we derive a Dense Neighborhood Lemma in (L_p)-spaces, again extending the known Euclidean results. Our method relies on the linearization of the distance through a map into a space of non-trivial Rademacher type, and then the use of a balanced signed-sum estimate, or a no-dimensional Radon theorem. The arguments rely on ideas from functional analysis that are clearly explained for the non-expert in that field. Comments: 22 pages Subjects: Machine Learning (cs.LG); Combinatorics (math.CO); Functional Analysis (math.FA) MSC classes: 68Q32, 52A35 ACMclasses: I.2.6; G.2.1 Cite as: arXiv:2607.10751 [cs.LG] (or arXiv:2607.10751v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2607.10751 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-60] Policy-Driven CT-Agent : Modeling Phase-Aware Diagnostic Control for Clinically Consistent CT Reasoning
链接: https://arxiv.org/abs/2607.10748
作者: Yanmeng Dong,Han Li,Yujia Li,Jingsong Liu,Xun Ma,Yanzhu Hu,Zhengyang Xu,Zhicheng Li,Nassir Navab,Shaohua Kevin Zhou
类目: Machine Learning (cs.LG)
*备注: 8 pages, 4 figures
Abstract:Computed Tomography (CT) diagnosis often relies on dynamic selection of imaging phases, such as non-contrast, arterial, or venous phases, based on preliminary findings, clinical suspicion, and diagnostic guidelines. This phase-wise decision process is critical for reducing unnecessary radiation exposure while supporting timely staging and treatment planning. However, phase-selection protocols can vary across hospitals, regions, and guidelines, while most existing CT-based AI methods assume that all phases are available and focus on static tasks under a fixed imaging phase, failing to model whether additional phases are required. This limitation stems from heterogeneous multi-phase representations, the need for knowledge-guided phase control beyond visual cues, and the lack of supervision for phase-sufficiency decisions in existing datasets. To address these challenges, we propose Policy-Driven CT-Agent (PD-CTAgent) for clinically consistent CT phase selection and diagnostic reasoning. PD-CTAgent introduces a Clinical Structure Abstraction Module (CSAM) to harmonize heterogeneous CT phases into a unified, phase-aware evidence representation. Based on this representation, a Knowledge-Guided Diagnostic Control Model (KDCM) evaluates phase sufficiency and iteratively requests additional phases when necessary. The policy-driven agent design further allows PD-CTAgent to flexibly follow different institutional, regional, or guideline-specific diagnostic protocols. Together, PD-CTAgent bridges static CT analysis and real-world clinical workflows. Experiments on two public datasets, LIDC and MCT-LTDiag, and one private dataset demonstrate its effectiveness and clinical consistency. Code will be made public upon acceptance.
[LG-61] Scaffold splits hide structural-frontier failures in ADMET models
链接: https://arxiv.org/abs/2607.10729
作者: Jiacheng Zheng,Chang Guo,Zixuan Wang,Xinyu Liu
类目: Machine Learning (cs.LG); Quantitative Methods (q-bio.QM)
*备注: 25 pages, 6 figures, 13 tables. Anonymized preprint
Abstract:Molecular property models are commonly evaluated by holding out Bemis–Murcko scaffolds, yet a scaffold identifier is only one notion of chemical unfamiliarity. We introduce a label-free structural-frontier split that reserves the sparsest and most physicochemically remote scaffold groups, and evaluate it on six public experimental or curated ADMET tasks. Against a 70/10/20 scaffold control with identical acyclic grouping, the frontier inflates equally weighted primary error with a taskwise median of 87.0% and a skew-sensitive mean of 130.3% (descriptive task/seed bootstrap interval, 52.1–246.0%). The mean falls to 75.9% once BBB is removed; that endpoint is the one whose score ranking inverts at the frontier. A message-passing graph-network control still shows a large gap (mean 82.8% over four tasks) and does not invert, so a low-capacity head does not explain the effect. We also test Multi-View Frontier Risk Extrapolation (\method), a count-adjusted tail-risk penalty over four molecular views, and treat it as a falsifiable probe. It changes normalized frontier error by only 0.16% relative to empirical risk minimization for the perceptron head (interval, -0.43 --0.84%) and by -1.9 % for the graph network; three fixed robust-penalty controls are likewise inconclusive. Against the published Lo-Hi and DataSAIL splitters the frontier inflates error more on average, though no split is uniformly hardest. An audit of 31,561 marine natural products further shows that OOD status and agreement with legacy ADMET predictions depend on the molecular view, endpoint and teacher coverage. Split construction and label provenance are important evaluation constraints in their own right, and the tested training penalties do not resolve the frontier failures we observe.
[LG-62] LayerNorm as Implicit Gain Control in Looped Transformers
链接: https://arxiv.org/abs/2607.10681
作者: Matthias M. M. Buehlmaier
类目: Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
*备注: 23 pages, 6 figures, 13 tables. Code available at this https URL
Abstract:In pre-LayerNorm looped transformers, LayerNorm inside the recurrent block acts as an implicit gain controller: by coupling the block’s local Lipschitz constant inversely to the activation scale, it renders the recurrence Jacobian non-normal – asymptotically contractive at every verified fixed point even where its operator norm exceeds 1 – so the true stability budget is the spectral margin, not an operator-norm bound. That margin depletes as the carry \rho \to 1 , and a minority of initializations never converge to a fixed point at all, so the diagonal carry constraint \rho(\barA) 1 is necessary but not sufficient for convergence of the full recurrence. Training experiments across six tasks, including a controlled ablation, reveal that the linear carry is not the depth-memory mechanism: gradient descent routes memory through the block’s more expressive nonlinear recurrence and leaves the stability-constrained carry at rest – the carry’s role is stabilization, not memory. We characterize the boundary of this claim: on tasks with axis-aligned per-channel structure, gradient descent does recruit the carry. All results are derived analytically and verified in a from-scratch, CPU-scale implementation; verification at larger scale is needed.
[LG-63] Modernizing HEBO: a robust Bayesian optimization baseline for practical heteroskedastic and non-stationary problems
链接: https://arxiv.org/abs/2607.10669
作者: L. A. Zhukov,E. V. Shaburova,D. V. Antonets
类目: Machine Learning (cs.LG)
*备注:
Abstract:Bayesian optimization is increasingly used to guide data-efficient experimentation in chemistry, materials science, and related laboratory settings, but its practical performance depends strongly on how well surrogate-model assumptions match the geometry and noise structure of the underlying objective. We introduce tidyHEBO, a robust Bayesian optimization model inspired by heteroskedastic evolutionary Bayesian optimization (HEBO) for single-objective, sequential optimization. tidyHEBO reconstructs the HEBO design philosophy in BoTorch and revises surrogate training, output-warping selection, acquisition function evaluation, and Pareto-front search. We benchmarked tidyHEBO on synthetic functions, Olympus emulators, fully experimental reaction-optimization datasets, needle-in-a-haystack (NIAH) materials problems, and Bayesmark hyperparameter optimization tasks. On these tasks tidyHEBO achieved competitive to superior performance and improvement in robustness across repeated optimization runs. We therefore propose tidyHEBO as a practical tool for sequential experimentations and a strong general-purpose benchmark for future Bayesian optimization research.
[LG-64] modelDNA: Calibrated Lineage Verification and Merge Decomposition from Sampled Weight Fingerprints
链接: https://arxiv.org/abs/2607.10617
作者: Muhammad Awais Bin Adil,Saad Aamir
类目: Machine Learning (cs.LG)
*备注: Code: this https URL . Data: this https URL . Live scanner: this https URL . DOI: https://doi.org/10.5281/zenodo.21305586
Abstract:The lineage graph of open-weight language models is self-reported: Hugging Face’s base_model metadata field is optional and unverified, and over 60% of Hub models document no parentage at all. Methods for detecting lineage from weights exist in the research literature, but each ships as paper code tied to one signal and one experiment; when a provenance dispute breaks, the analysis is redone by hand. This report describes modelDNA, a tool that fingerprints a model from roughly 100-300 MB of ranged HTTP reads (instead of a full 15 GB download for a 7B model), compares the fingerprint against a reference database of foundation models across four published signal families, and returns one of eight verdict classes with a calibrated probability, preferring honest abstention to confident error. On a benchmark of 15 real Hub models with org-documented parentage, judged against 8 candidate bases (13 positives, 107 hard negatives), the system achieves AUROC 1.0, zero false positives at its reporting threshold, and 13/13 correct top-1 parent attribution. The report’s second contribution is merge decomposition. Every mainstream weight-merging method is (near-)linear per tensor, and fingerprint sample positions are deterministic functions of tensor identity, so a merged model’s fingerprint is the same linear combination of its parents’ fingerprints. Mixture weights can therefore be recovered from fingerprints alone by sum-to-one constrained least squares. Against merges with published mergekit configurations as ground truth, the method recovers a slerp merge’s layer-interpolation curves at r = 0.999 and a dare_ties merge’s mixture weights to within 0.011 of the published values, without downloading any weights beyond the fingerprints. All fingerprints, benchmarks, and the inferred lineage graph of 55 models are public and reproducible offline.
[LG-65] MAdam: Low-Precision Training via Additive-Multiplicative Optimization ICML2026
链接: https://arxiv.org/abs/2607.10611
作者: Xiaoyuan Liang,Sebastian Loeschcke,Mads Toftrup,Anima Anandkumar
类目: Machine Learning (cs.LG)
*备注: Accepted at the 43rd International Conference on Machine Learning (ICML 2026). 45 pages
Abstract:Training with quantized weights can reduce costs but often results in degraded accuracy, especially when optimization is carried out in low precision, without storing high-precision copies. We identify a key failure mode: under low precision, standard optimizers can get stuck and not make progress, especially at large weight magnitudes due to coarse mantissa resolution. To overcome this, multiplicative updates have been previously proposed, in place of additive updates in standard optimizers. While successful under extremely low precision, such as under the logarithmic number system, they suffer from failures near zero and across sign changes. The failure modes of additive and multiplicative updates are therefore complementary. To exploit this, we propose M+Adam, which combines both update types: additive steps handle sign changes and small magnitudes, while multiplicative steps ensure progress at large magnitudes when additive updates are zeroed out under rounding. We prove monotone descent for M+Adam under standard smoothness assumptions. Across LLaMA-style pretraining with 60M-1B models, 1x-8x Chinchilla budgets, and using only BF16, FP8, and FP4 master weights, M+Adam consistently improves low-precision training.
[LG-66] AutoNorm: Understanding Adaptive Normalization in Transformers through Differentiable Gating
链接: https://arxiv.org/abs/2607.10593
作者: Piyush Kaushik Bhattacharyya,Divyanshu Rai,Swastik Singh,Kumar Aakash,Ayush Ranjan,Krutika Verma
类目: Machine Learning (cs.LG)
*备注:
Abstract:Normalization is a critical component for stabilizing Transformer training, yet the choice between static strategies such as Layer Normalization (LN) and adaptive alternatives remains largely task-dependent. In this paper, we investigate a key optimization challenge in differentiable normalization gating. Our experiments show that, on relatively stationary vision tasks, the high gradient variance introduced by Gumbel-Softmax gating can hinder convergence of the routing mechanism, causing learned gates to underperform simple random selection. In contrast, on non-stationary language modeling and classification tasks, sustained gating diversity enables the model to learn more effective layer-wise normalization policies. Motivated by these observations, we propose AutoNorm-S (Stabilized), a training strategy that mitigates optimization instability through a gate-freezing schedule. AutoNorm-S achieves competitive or improved performance across multiple benchmarks, outperforming adaptive normalization baselines on NLP datasets, including PTB and SST-2, while remaining competitive on standard vision benchmarks. These results suggest that decoupling normalization selection from optimization noise provides a practical and principled approach for adaptive normalization in Transformer architectures.
[LG-67] Sharp Concentration Bounds for Bundle-Valued Statistics on Manifolds ICML2026
链接: https://arxiv.org/abs/2607.10592
作者: Swagatam Das,Vaclav Snasel
类目: Machine Learning (cs.LG)
*备注: Accepted in ICML 2026
Abstract:Many geometric statistics and manifold learning pipelines routinely produce observations – such as tangent vectors or local frames – whose natural home is a varying family of fibers attached to different points of a base manifold, rather than a single shared vector space. Forming empirical averages requires transporting these observations to a common reference fiber, thereby introducing curvature- and holonomy-driven effects that are absent from classical concentration theory. We develop a non-asymptotic concentration theory for such transported empirical means, deriving finite-sample, dimension-free Hoeffding- and Bernstein-type bounds via sharp Hilbert-space inequalities. When shortest paths to the reference point are non-unique, transport becomes path-dependent and introduces a deterministic holonomy bias; we isolate and quantify this bias through bundle curvature and loop geometry, with sharp closed-form formulas for the tangent bundle of a round sphere. The resulting bias-variance decomposition separates the stochastic fluctuation decaying at the classical n^-1/2 rate in sample size n , from a curvature-driven error floor that no amount of additional data can eliminate; minimax lower bounds confirm both terms are unavoidable. We further establish a robust median-of-means estimator achieving optimal rates under heavy tails and the central limit theorem in the reference fiber. Controlled experiments on the sphere validate all theoretical predictions.
[LG-68] BucketKD: A Safety-Aware Bucket-Based Knowledge Distillation Framework for End-to-End Motion Planning
链接: https://arxiv.org/abs/2607.10565
作者: Md Nahidul Islam,Mohd Hasan Ali,Dipankar Dasgupta,Myounggyu Won
类目: Robotics (cs.RO); Machine Learning (cs.LG)
*备注:
Abstract:End-to-end motion planning has emerged as a promising paradigm in autonomous driving, directly mapping raw sensor data to control commands via deep neural networks. Despite its advantages, its large model size hinders deployment in resource-constrained platforms. In this paper, we present BucketKD, a bucket-based knowledge distillation framework that yields compact and safety-aware end-to-end planners. Compared to the state-of-the-art approach, which relies on simplified planning state representations, BucketKD discretizes critical environmental variables into adaptive buckets that capture richer scene semantics while preserving efficiency. In addition, we design a safety-aware waypoint attention mechanism that evaluates each waypoint’s risk level by accounting for both obstacle proximity and relative motion through a time-to-collision (TTC) formulation widely used in transportation research. This enables the student model to better retain safety-critical behaviors during distillation. Extensive experiments in CARLA using the Bench2Drive dataset show that BucketKD significantly outperforms the state-of-the-art in both planning accuracy and safety while maintaining strong compression ratios.
[LG-69] LLM -PDESR: Robust PDE Discovery via Subdomain Weighted Residuals and LLM -Guided Symbolic Hypothesis Generation
链接: https://arxiv.org/abs/2607.10546
作者: Jinyang Du,Hao Ma,Xiaohu Shi,Bo Yang,Yanchun Liang,Heow Pueh Lee,Chunguo Wu
类目: Machine Learning (cs.LG)
*备注: 28 pages, 12 figures
Abstract:Discovering governing partial differential equations (PDEs) from noisy observational data is a fundamental challenge in scientific machine learning. Traditional symbolic regression (SR) methods often struggle to identify accurate equations within vast combinatorial search spaces, largely due to their inability to incorporate essential domain-specific prior knowledge. Furthermore, reliance on pointwise evaluations and discrete finite differences inherently amplifies high-frequency noise, creating deceptive fitness landscapes that derail the optimization process. To resolve these bottlenecks, we propose LLM-PDESR, a framework that integrates the structural hypothesis generation of Large Language Models (LLMs) with a mathematically rigorous evaluation environment. By employing C^4-continuous quintic splines for robust differentiation and subdomain weighted residuals as natural low-pass filters, our approach effectively mitigates the fitness landscape distortion that plagues existing methods. A Pareto-driven feedback loop then enables the LLM to iteratively refine candidate equations, balancing predictive accuracy with structural parsimony. We evaluate LLM-PDESR on 23 canonical PDEs and five structurally novel equations (including a multivariate system) specifically designed to preclude dataset memorization and test true discovery capabilities. Demonstrating real-world applicability, the framework successfully extracts a consistent structural skeleton for an interpretable 1D dynamical surrogate (1D-CACE) directly from noisy ERA5 reanalysis data. Extensive experiments and out-of-distribution testing confirm that LLM-PDESR significantly outperforms state-of-the-art methodologies in structural recovery, noise resilience, and the avoidance of spurious complexity and equation bloat.
[LG-70] Learning from Noise: Effective-Rank Collapse and Out-of-Distribution Rejection in Restricted Boltzmann Machines
链接: https://arxiv.org/abs/2607.10506
作者: Oshada Rathnayake,Nikhil Shukla
类目: Machine Learning (cs.LG); Disordered Systems and Neural Networks (cond-mat.dis-nn)
*备注:
Abstract:Restricted Boltzmann machines (RBMs) represent data by shaping an energy landscape over visible and hidden configurations, but their discriminative use is fragile under out-of-distribution (OOD) inputs: samples outside the training distribution can be absorbed into one of the learned class basins rather than rejected. Here, we analyze this failure mode through the spectrum of the induced visible–visible interaction J=WW^T , where (W) is the visible–hidden weight matrix. Relative to a Marchenko–Pastur random-matrix reference, conventional training spreads spectral weight into many weak, bulk-compatible directions, increasing the effective rank of J . When auxiliary random binary images are assigned to a rejection label during training, the learned interaction undergoes effective-rank collapse: weak bulk-like modes are depleted, spectral weight concentrates into fewer dominant eigendirections, and the effective rank of J approaches that of the empirical data covariance matrix. The resulting RBM rejects structured OOD image datasets while preserving MNIST classification accuracy, showing that random auxiliary exposure can reshape both the interaction spectrum and the free-energy landscape of an energy-based classifier.
[LG-71] EvidentialRAG : Quantifying and Mitigating Information Conflict in Multi-Source Retrieval-Augmented Generation via Evidential Deep Learning
链接: https://arxiv.org/abs/2607.10491
作者: S M Asif Hossain,Ruksat Khan Shayoni,M. F. Mridha
类目: Machine Learning (cs.LG)
*备注:
Abstract:Retrieval-augmented generation grounds large language models in external evidence, but most pipelines still treat retrieved passages as deterministic and mutually consistent context. In open information environments, retrieved sources may disagree because of temporal drift, source error, ambiguity, or genuine uncertainty. This paper introduces ERAG, an uncertainty-aware RAG framework that converts retrieved chunks into probabilistic evidence before generation. A lightweight evaluator extracts candidate claims and maps chunk-level support to Dirichlet evidence. A conflict-preserving Dempster-Shafer fusion rule then transfers unresolved disagreement into epistemic uncertainty rather than normalizing it away. The generator is routed to direct answering, conflict-aware answering, or abstention according to the fused uncertainty score. Experiments on CRAG, ConflictQA, and MuSiQue show that ERAG remains competitive with the strongest matched baseline on standard question answering while improving behavior under conflict. On the CRAG ambiguous subset, hallucination decreases from 45.3% for Corrective RAG to a human-calibrated estimate of 34.8%, conflict resolution increases from 35.2% to 51.2%, and expected calibration error improves to 0.122. These results suggest that evidential modeling is a practical mechanism for trustworthy information processing in foundation-model-based retrieval systems.
[LG-72] NetInjectBench: Benchmarking Indirect Prompt Injection in Tool-Using Large Language Model Agents for Network Operations
链接: https://arxiv.org/abs/2607.10490
作者: Ruksat Khan Shayoni,Muhammad Faraz Shoaib,S M Asif Hossain,M. F. Mridha
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注:
Abstract:Tool-using large language model (LLM) agents are attractive for network operations, but tickets, alerts, logs, runbooks, and ChatOps messages can carry indirect prompt injections. We present NetInjectBench, a 130-scenario benchmark that separates untrusted artifact text, trusted policy metadata, and evaluation labels for network-operation tool use. The sample contains 40 benign, 40 weak-attack, 40 strong-attack, and 10 approved high-impact change scenarios; each is evaluated with Qwen2.5-7B, Llama3.1-8B, and Mistral-7B. Across 240 attack instances, naive execution reached an 82.50% unsafe tool-action rate. Prompt-only safety, Self-Reminder, Spotlighting, and a Two-Pass LLM Judge reduced this rate to 25.63%, 21.67%, 18.33%, and 10.00%, respectively. Static allowlisting reached 5.00% but blocked all approved changes, yielding 0.00% usefulness and 100.00% overblocking on approved cases. Under the stated metadata-integrity assumption, the metadata-aware policy gate produced 0/240 unsafe attack actions, with a 95% Wilson upper bound of 1.58%, while preserving 99.17% attack-scenario usefulness and 100.00% approved-change usefulness. The findings show that network-operation agents need execution-time authorization boundaries alongside prompt-level instruction hygiene.
[LG-73] Fast Data-Driven Modeling of Hydraulic Clutch Control Pressure with Latch-State Classification and Gaussian Process Regression
链接: https://arxiv.org/abs/2607.10477
作者: Yash Bagla,Jason Schneider
类目: ystems and Control (eess.SY); Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注: 8 pages, 5 figures. Accepted to the program of the 14th CTI Symposium and Exhibition, Automotive Drivetrains, Intelligent, Electrified, scheduled for May 13-14, 2020 in Novi, Michigan, USA
Abstract:This paper presents a data-driven method for modeling the pressure response of a hydraulic clutch control circuit. The system consists of a variable-force solenoid, accumulator, pressure regulator valve, and latch valve, and exhibits nonlinear behavior caused by hysteresis, latch transitions, and actuator dynamics. A baseline model using commanded current variables captured the general pressure response but failed to represent hysteresis and latch behavior accurately. The input vector was therefore extended with current derivative information, and several classifiers were tested to separate latch-related operating regimes before fitting Gaussian Process regression models to the resulting partitions. Nonlinear SVC and gradient boosting produced the highest latch-classification accuracy, and nonlinear SVC was selected for the final local-regression pipeline. The proposed approach was evaluated on unseen ramp-rate data and compared against a physics-based Amesim model. The machine-learning model reproduced the measured pressure response and hysteresis behavior more accurately than the physics-based simulation for the tested operating conditions. These results suggest that machine-learning plant models can complement physics-based hydraulic models during hardware development and controller calibration when representative test-stand data are available.
[LG-74] oward Production-Ready Federated Learning in Healthcare: Privacy Orchestration and Governance in MLOps
链接: https://arxiv.org/abs/2607.10467
作者: Sakshi Gorkhali,Jonesh Shrestha
类目: oftware Engineering (cs.SE); Machine Learning (cs.LG)
*备注: 5 pages, 2 figures, 1 table
Abstract:Healthcare organizations often cannot freely centralize patient data because medical records are sensitive, regulated, and institutionally controlled. Federated learning offers a practical alternative by allowing hospitals and clinics to train a shared model while keeping raw data local. However, federated learning is not automatically production-ready or private by default. Model updates can still leak information, and decentralized training introduces operational challenges in deployment, monitoring, rollback, debugging, and governance. This paper examines how MLOps practices and the emerging idea of Federated Learning Operations (FLOps) can make federated healthcare machine learning systems scalable, reliable, and trustworthy. It answers three research questions: how containerization and orchestration support federated deployment, how privacy-preserving mechanisms affect trade-offs among privacy, utility, scalability, and operational complexity, and which post-deployment practices are most important for long-term governance. The central argument is that federated healthcare ML requires more than privacy-preserving algorithms. It needs an integrated MLOps architecture that combines reproducible deployment, secure orchestration, model versioning, audit logging, drift monitoring, heterogeneity management, and clear governance.
[LG-75] Pitfalls of Administrative Censoring in Survival Models with Time-Indexed Inputs
链接: https://arxiv.org/abs/2607.10466
作者: Yanqi Xu,Hui Dai,Carlos Fernandez-Granda,Krzysztof J. Geras,Yiqiu Shen
类目: Machine Learning (cs.LG); Applications (stat.AP)
*备注:
Abstract:Survival models can model time-to-event outcomes using partially observed data. They are widely used in clinical prediction, including cancer risk, disease progression, treatment response, and mortality. Recent models often rely on rich inputs collected at a specific clinical encounter, such as medical images, laboratory tests, electronic health record snapshots, or sensor measurements. In large retrospective datasets, these inputs are usually collected over many calendar years. As a result, they may contain clues about when they were acquired through changes in devices, protocols, documentation, patient mix, or clinical practice. This creates a potential failure mode when outcomes are observed only up to a fixed study end date. More recent records necessarily have less potential follow-up than older records. A model that can infer the record date from the input may therefore learn to predict how much follow-up was available rather than the patient’s true risk of experiencing the event. We call this failure mode administrative-cutoff leakage. In this paper, we characterize when this leakage can occur, distinguish it from classical informative censoring and genuine temporal changes in risk, and propose practical ways to detect it. In simulations, we show that administrative-cutoff leakage can inflate fixed-horizon AUC and can also affect Harrell’s C-index under realistic follow-up patterns. We then demonstrate the same behavior in a real mammography cohort. These results motivate a simple design principle for survival prediction: for an n-year prediction task, the dataset should provide at least n years of potential follow-up after the latest input date. Otherwise, the models may be subject to bias induced by administrative-cutoff leakage.
[LG-76] Is Model Instability just Noise to be Tolerated or a Property that can be Managed?
链接: https://arxiv.org/abs/2607.10420
作者: Amirali Rayegan,Lunxiao Li,Tim Menzies
类目: oftware Engineering (cs.SE); Machine Learning (cs.LG)
*备注:
Abstract:In software analytics, rerunning the same analysis twice often yields different models and conclusions. This reduces trust in the model and limits its use. We find that model instability is a major problem. Across 127 multi-objective SE optimization problems (12,700 test cases), repeated runs of a state-of-the-art optimizer agree on only 13.7% of test cases, even under improved settings. We argue that this instability is not merely noise to tolerate, but a property that can be measured and managed. By adjusting how labels are spent, how complex the models become, and how splits are scored, we obtain models that agree 4.8 times as often as the default configuration. The standard deviation of optimization error falls by 22% on average (mean std 17.4 to 13.6), while recommendation quality improves rather than degrades. In terms of quality, the refined settings are statistically top-ranked on 119 of 127 datasets, compared to 74 for the defaults. We then test causal and data-locality interventions and find that they help only partially, suggesting a residual stability floor. Our evidence suggests there are fundamental limits to stability set by the data itself (noise, scarce labels, proxy objectives, and the many near-equivalent models a dataset admits). We conclude that instability should be treated as a standard evaluation axis in SE optimization, which should be routinely measured, reported alongside performance, and used to calibrate trust in any single run. The methods in this paper provide a baseline against which future efforts to reduce SBSE instability can be judged. To support open science, we offer the following reproduction package: this https URL Subjects: Software Engineering (cs.SE); Machine Learning (cs.LG) Cite as: arXiv:2607.10420 [cs.SE] (or arXiv:2607.10420v1 [cs.SE] for this version) https://doi.org/10.48550/arXiv.2607.10420 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-77] Machine Learning-based Correlation of Charpy Impact Properties Between Sub-sized and Standard-sized Specimens for Nuclear Structural Materials
链接: https://arxiv.org/abs/2607.10412
作者: Yugandhar Kasala Sreenivasulu,Isshu Lee,John W. Merickel,Fei Xu,Yalei Tang,Joshua E. Rittenhouse,Aleksandar Vakanski,Rongjie Song
类目: Machine Learning (cs.LG)
*备注: 27 pages, 8 figures
Abstract:Reliable correlations of Charpy impact test results between sub-sized and full-sized specimens are essential for structural integrity assessments, particularly in nuclear applications, where spatial constraints and limited material volume restrict specimen size. Although standards such as ASTM A370 and BS 7910 provide guidance on conversion methodologies, and numerous analytical correlation methods have been proposed in prior studies, these approaches generally have limited accuracy and their applicability is often constrained to specific materials, treatment conditions, and specimen geometries. In this study, a Machine Learning (ML)-based framework is proposed for correlating Charpy impact properties across specimen sizes. The proposed approach maps absorbed energy values across the full ductile-to-brittle transition region by applying a temperature shift combined with scaled residual projection, to align sub-sized test data with full-sized response. From the resulting temperature-energy profiles, the correlated values for upper shelf energy (USE) and ductile-to-brittle transition temperature (DBTT) are extracted by fitting data with a hyperbolic tangent model. The framework is validated using a dataset comprising 389 matched sub-sized and full-sized Charpy impact tests on SA533B steel. This ML-based approach demonstrates an improved correlation performance relative to conventional analytical methods, achieving R2 values of 0.942 for USE and 0.892 for DBTT. The trained ML models do not require access to full-sized Charpy data during inference, making this approach suitable for material surveillance programs, accelerated irradiation testing, and other applications involving small-size Charpy impact testing.
[LG-78] Stateful Worlds Stateless Elasticity: Exact-State Serving for Interactive World Models
链接: https://arxiv.org/abs/2607.10389
作者: Jin Li(Harvard University),Jiawei Chen
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG)
*备注: 20 pages. Extended version
Abstract:A persistent interactive world model keeps its running state resident on the GPU that serves it: a multi-gigabyte attention cache, almost all of it rewritten at every generation step. That state cannot be recomputed in interactive time or approximated without changing the world, so a live session pins its device. The pin is a scheduling problem. WorldMove moves a live session under one guarantee: the destination is bit-identical to the source, or nothing is installed. It relocates the cache in 18.8 ms same-node, 101x faster than save/load. It holds a checksum-verified 92.1-94.8 Gb/s on a 100 Gb fabric. At that rate the cache fits inside one interactive block. Migrating an actively generating session, it converges at a block boundary and the destination continues the world bit for bit. An admissibility condition decides each move. The move must complete inside the readout horizon, over bandwidth that covers the state plus its dirty rate. Lifted to a fleet schedulability test, it governed a consolidation loop that executed 48 of 48 migrations bit-identical across two providers. Two constraints are structural. Bit-exactness survives only inside a controlled configuration of one GPU architecture, so moving the state is the only way to preserve it exactly in interactive time. Verification cannot hide inside the wire on this fabric. Receive-path checksums stall the transport at protocol timescales under fan-in, and unscheduled incast silently collapses a receiver while every delivered byte stays correct. An incast-aware admission controller holds zero misses to 1.4x offered load and sheds overload as rejects. A lossless GPU codec widens the admission gate to fabrics raw motion cannot use. We exercise the serving loop and the mover separately, each end to end. Their composition on one fabric is unbuilt. Exact-state elasticity is a joint scheduling problem over transport and verification.
[LG-79] A Control Theory of Predictability in Latent World Models
链接: https://arxiv.org/abs/2607.10362
作者: Hanzhe You,Yonggang Zhang,Maohao Ran,Zhiqin Yang,Zhenyuan Zhang,Wei Xue,Jun Song,Xinmei Tian,Yike Guo
类目: Machine Learning (cs.LG)
*备注: Preprint about latent world models, Koopman operator and control theory, 33 pages, 1 figure. Main text about 10 pages
Abstract:Latent world models are trained to predict future states in a learned representation and are then deployed inside a planner that selects actions by simulating them forward. Current practice adopts the prediction error, the single- or multi-step rollout loss on held-out data, as the training and model-selection objective, on the assumption that a lower prediction error yields better control. We show that this assumption is unreliable for a structural reason: a planner does not query the model on the training distribution but on the states that its candidate actions reach, which generally leave the data manifold, so an error averaged over the data cannot by itself govern control. We therefore reframe the objective as the discrepancy between the predicted and the true plan-cost at the plan the planner commits to, and prove that the planner’s suboptimality is bounded by twice this discrepancy, whereas the data-averaged prediction error neither bounds nor tracks it. Under a linear-control premise the discrepancy separates into two terms. The first is a small on-manifold residual, on which the predicted and true dynamics agree and which a spectral tax prices through the non-normality of the latent transition operator. The second is an off-manifold divergence, on which an action carries the state off the manifold and the two dynamics diverge; this divergence is the binding term and is bounded by no data-averaged error. Synthetic operators confirm the pricing formulas, and latent model-predictive control experiments confirm the decoupling: across seeds, the single-step validation error is essentially uncorrelated with control success, whereas a fidelity score on the planner-reachable measure tracks it.
[LG-80] Byzantine Accountability Without Consensus: Strong Eventual Consistency for Non-Associative Stochastic Robust Aggregation
链接: https://arxiv.org/abs/2607.10305
作者: Ryan Gillespie
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注:
Abstract:Byzantine-robust aggregation rules such as multi-Krum assume a central coordinator, and decentralising them is obstructed by the rules themselves: they are globally coupled, non-associative, and discontinuous, so an ulpscale perturbation can flip the selected subset, moving the output by a non-vanishing amount. None of this prevents coordinator-free replication, because a robust rule needs no agreed order of contributions, only an agreed set and an agreed exclusion predicate, both of which converge without consensus. ACFA (Accountable Consensus-Free Aggregation) replicates a content-addressed OR-Set of signed contributions and a grow-only set of self-authenticating equivocation proofs, offline-verifiable by anyone. Aggregation is a deterministic pure function of the converged product state: fixed-point integer arithmetic over a hash-canonical order, ties broken by content hash. We prove that any pure function of a converged product of CRDTs (non-monotone, non-associative, or stochastic) inherits Strong Eventual Consistency, together with its converse; the contribution is the composition of a data lattice with an evidence lattice applied to a robust selector, not the elementary lifting step. A prototype (10 nodes, 3 Byzantine) passes 16/16 falsification checks: byte-identical roots under adversarial gossip, deterministic re-convergence after late equivocation proofs, partition recovery, and three byte-identity-breaking ablations. The guarantee is consistency, not accuracy; robustness is imported, conditional on 2f + 3 admitted contributions (at most f Byzantine) and a stated quantisation-margin condition.
[LG-81] Empowering Long-form Omni-modal Understanding with Robust Audio Perception
链接: https://arxiv.org/abs/2607.10299
作者: Kaiying Yan,Luoyi Sun,Xiao Zhou,Weidi Xie
类目: Machine Learning (cs.LG)
*备注:
Abstract:Recent advances in large-scale multimodal models have drivenremarkable progress in vision-language tasks; however, comprehensiveomni-modal understanding remains under-explored, largely due to thescarcity of datasets with rich, explicitly aligned auditory cues. To bridgethis gap, we present AVDC (Audio-Visual Decoupled Captions), a large-scaledataset designed to disentangle visual and auditory semantics. Specifi-cally, we propose an automated pipeline that leverages off-the-shelf mod-els to annotate videos with tripartite captions: visual-only (V), audio-only (A), and joint audio-visual (AV). This decoupled structure explic-itly captures both modality-specific nuances and complex cross-modalinteractions. Building upon this, we introduce AVDC-QA-CoT, a Chain-of-Thought augmented question-answering dataset to foster audio-visualreasoning. To fully exploit these resources, we employ a two-stage train-ing paradigm: omni-modal caption generation pre-training on AVDC, fol-lowed by instruction tuning on AVDC-QA-CoT. Extensive experiments acrossdiverse downstream tasks, spanning video captioning, audio-centric anal-ysis, and omni-modal benchmarks, demonstrate consistent and signifi-cant performance gains, showing the efficacy of our proposed datasetsand training strategy in advancing omni-modal perception. Code anddataset are related on this https URL.
[LG-82] Interpreting learning dynamics of autoencoders: Transient scaling and emerging concepts of the Ising model
链接: https://arxiv.org/abs/2607.10285
作者: Max Weinmann,Miriam Klopotek
类目: Machine Learning (cs.LG); Disordered Systems and Neural Networks (cond-mat.dis-nn); Statistical Mechanics (cond-mat.stat-mech)
*备注:
Abstract:We study how unsupervised autoencoders trained on microscopic spin configurations from the Ising model learn macroscopic, theory-relevant variables underlying the data-generating process. Without embedding domain knowledge, we mimic a typical discovery setting: We quantify learning across multiple spatial (coarse-graining) scales and reveal two distinct dynamical regimes controlled by main hyperparameters (model depth, width, and learning rate) – a magnetization-dominated regime and an energy-dominated regime characterized by trade-offs in their representation quality. The first regime is a transitory state exhibiting dynamical scaling and fluctuations that follow an ordering-to-scale; the second gradually shifts resolution towards smaller scales relevant for the energy representation. Deep models trained at moderate and fast rates become arrested before reaching these regimes. With a novel analysis of recursive-dynamic trajectories, we demonstrate that prediction errors induce flow fields that produce a common trajectory topology across all representation spaces. A dynamical viewpoint of learning is established in which intrinsic properties expose the effects of forced changes in representation during training. We utilize the intuition that learning operates as a process driven far from equilibrium by fluctuations from the training data and optimizer to provide an interpretive basis grounded in both the physical world and the machine models that represent it.
[LG-83] Sharper Analysis of Single-Loop Methods for Bilevel Optimization
链接: https://arxiv.org/abs/2607.10263
作者: Yubo Zhou,Jun Shu,Luo Luo,Junmin Liu,Deyu Meng,Guang Dai,Haishan Ye
类目: Machine Learning (cs.LG)
*备注: 26 pages,6 figures
Abstract:Bilevel optimization underpins many machine learning applications, including hyperparameter optimization, meta-learning, neural architecture search, and reinforcement learning. While hypergradient-based methods have advanced significantly, a gap persists between theoretical guarantees and practical single-loop implementations required for efficiency. We bridge this gap by establishing sharper convergence results for single-loop approximate implicit differentiation (AID) and iterative differentiation (ITD) methods, leveraging our proposed analytical framework, decoupled norm analysis (DNA). For AID, we improve the convergence rate from \mathcalO(\kappa^6/K) to \mathcalO(\kappa^5/K) , where \kappa is the condition number of the inner-level problem. For ITD, we prove that the asymptotic error is \mathcalO(\kappa^2) , exactly matching the known lower bound and improving upon the previous \mathcalO(\kappa^3) guarantee. Numerical experiments on synthetic and real tasks corroborate our theoretical findings.
[LG-84] Data-Driven Telecom Marketing Optimization: A Machine Learning-Based Churn Prediction and Customer Segmentation Framework
链接: https://arxiv.org/abs/2607.10260
作者: Nada Ali,Lina Ahmed,Tahani Abdalla Attia Gasmalla
类目: Machine Learning (cs.LG)
*备注:
Abstract:Customer churn is a major challenge for telecommunication companies, directly eroding revenue and long term customer relationships. Traditional retention programs rely on generic, not personalized incentives and lack the precision to identify high risk customers before they leave. This paper presents a data driven marketing optimization framework integrating machine learning based churn prediction, customer segmentation combining churn risk with customer value, and tailored, segment specific marketing and Return on Investment ROI strategies. Using the IBM Telco Customer Churn dataset with 7043 customers and 21 features, three gradient boosting ensembles, XGBoost, LightGBM, and CatBoost, were trained and tuned via randomized search with stratified 5 fold cross validation, class weighting, and F1 score driven decision threshold optimization to counter a class imbalance of 73.4% versus 26.6%. CatBoost was selected as the deployment model, achieving 77.68% accuracy, an F1 score of 0.6366, a PR AUC of 0.6553, and a ROC AUC of 0.8403 on the held out test set. Customers were partitioned with K Means clustering, validated via the Elbow method and visualized with Principal Component Analysis, into High, Medium, and Low Value segments, cross tabulated against churn risk labels to define four actionable clusters. Segment specific retention, upsell, and engagement strategies were designed for each cluster, and a theoretical ROI and CLV framework quantifies the financial impact of the proposed interventions. The pipeline was operationalized in an interactive Streamlit web application allowing marketing teams to upload data, filter by segment, visualize churn drivers via SHAP, and download automated segment reports. Results confirm that combining predictive churn modeling with value aware segmentation yields more actionable and profitable marketing decisions than churn prediction alone.
[LG-85] DSSMs: State Space Models with Explicit Memory via Delay Differential Equations
链接: https://arxiv.org/abs/2607.10244
作者: Yixiao Qian,Song Chen,Jiaxu Liu,Shengze Cai,Chao Xu
类目: Machine Learning (cs.LG)
*备注:
Abstract:State Space Models (SSMs) have emerged as a powerful paradigm for efficient long-sequence modeling, offering parallel training and fast linear-time recurrent inference. However, like other recurrent architectures, SSMs must compress an unbounded history into a fixed-size state, which limits context retention and makes precise retrieval over long-range context inherently difficult. To overcome this limitation, we propose Delay State Space Models (DSSMs), a delay differential equation (DDE)-inspired extension of diagonal SSMs that augments discrete SSM recurrences with explicit delayed-state feedback. Making explicit delayed feedback practical requires new stability parameterization, history management, and FFT-training tools. We address these challenges with a practical discretization and parameterization grounded in a simple delay-independent stability condition. To bypass direct time-domain kernel construction, we derive the DSSM transfer function and compute kernels in the frequency domain, using a kernel contour shift to suppress aliasing and recover accurate FFT training. Empirically, DSSMs substantially improve targeted delayed-retrieval tasks while outperforming S4D on most standard sequence metrics and remaining close on the others.
[LG-86] MeloBottleneck: Self-Supervised Melody Skeleton Extraction with a Latent Subsequence Bottleneck
链接: https://arxiv.org/abs/2607.10233
作者: Fan Bu,Rongfeng Li,Linfeng Fan
类目: ound (cs.SD); Machine Learning (cs.LG)
*备注: 8 pages, 3 figures
Abstract:Melody skeleton extraction aims to derive a shorter melody that preserves structural notes while removing ornaments. Prior methods rely on hand-crafted reduction rules or note-wise salience classifiers trained with heuristically or procedurally generated pseudo-labels. Such supervision can inherit generator bias and does not explicitly optimize a coherent reduced melody. We introduce MeloBottleneck, a self-supervised framework that represents a skeleton as a length-controlled, order-preserving latent subsequence. A hard-bottleneck extractor selects note events, a rhythmic-closure operator produces a self-consistent skeleton, and a re-ornamentation decoder reconstructs the input melody. Training combines reconstruction, a frozen autoregressive melody prior, ornament-invariant consistency across procedurally ornamented views, and ornament exclusion. We evaluate three regimes: synthetic out-of-distribution ornament-to-skeleton, TAVERN variation-to-theme, and Jiugong ornamented-to-gongche. A matched pseudo-label classifier excels on the synthetic benchmark, while MeloBottleneck transfers better, achieving competitive selection quality on TAVERN and Jiugong. Skeletonized melodies also improve BM25-based fragment retrieval, boosting Recall@K and MRR while reducing query time. Overall, the results suggest that learning skeletons as latent subsequences yields more robust transfer than pseudo-label imitation.
[LG-87] wo Confounds in Cross-Model Value Comparison: Response Determinism and the Access Harness
链接: https://arxiv.org/abs/2607.10202
作者: Hong-In Won,Jinseok Jang,Hyoseop Kim
类目: Machine Learning (cs.LG)
*备注: 23 pages, 4 figures, 3 tables
Abstract:Cross-model comparisons read divergence in value dispositions as evidence that language models hold individuated values. Under single-draw measurement this conflates two quantities: a difference in central tendency (a genuine value difference) and a difference in response determinism (how sharply a model commits to a forced choice). We introduce a separation protocol – no-rule value dilemmas with counterbalanced, repeated forced-choice measurement and a determinism index – and a determinism-corrected decomposition that splits an apparent cross-model distance into a direction-flip component (genuine disagreement) and a same-side-more-extreme component we label determinism. Across nine models, determinism varies substantially (0.66-0.95 among engaging models); whether it is a per-model trait or tracks provider and scale is a question our method makes measurable but our sample leaves open. Correcting for determinism shrinks apparent individuation, while a few cross-family disagreements survive a strict test. We then isolate a second confound: the access harness serving each model. Re-collecting the same models through raw provider APIs, we find the deployment client shifts a model’s value profile substantially and client-specifically: one subscription CLI moves a profile by 0.31, flips four of eighteen items, and inflates the flagship’s apparent softness (0.34 via CLI vs 0.66 via raw API), whereas another provider’s client is clean, confounding provider family with access client. The harness is a value-shaping layer: a base model that refuses one-in-ten forced choices is made compliant by an agent system prompt, established causally in a white-box control. An audit ranking models by single-draw value distance thus ranks a determinism-inflated quantity, confounded further by the client used. We contribute the decomposition and identify the deployment harness as a distinct value confound.
[LG-88] he Differential Neural Tangent Kernel and Its Positivity
链接: https://arxiv.org/abs/2607.10200
作者: Bangti Jin,Longjun Wu
类目: Machine Learning (cs.LG); Numerical Analysis (math.NA); Machine Learning (stat.ML)
*备注: 31 pages, 1 figure
Abstract:The Neural Tangent Kernel (NTK) is one powerful tool for analyzing the training dynamics of neural networks in the over-parameterized regime. Recently, the theoretical framework has been extended to physics-informed neural networks (PINNs) for solving linear PDEs, one highly popular class of neural PDE solvers. In the analysis, the positivity of the associated NTK plays a fundamental role. However, establishing the positivity of the NTK for PINNs is highly challenging, due to the presence of multiple differential operators. In this work, we propose a new theoretical framework, called Differential Neural Tangent Kernel (DNTK), for analyzing PINNs through the lens of the NTK, and establish the positivity of the infinite width DNTK for both shallow and deep neural networks for a wide class of activation functions, including RePU and smooth but non-polynomial activations, for all linear differential operators. These theoretical results lay the foundation for the analysis of gradient type algorithms for training PINNs.
[LG-89] Generative Augmentation of Raman Spectra for Glioma Classification
链接: https://arxiv.org/abs/2607.10196
作者: Andrei Iuşan,Iulian Vasile,Daria Voiculescu,Ion Petre,Andrei Păun,Bogdan Oancea,Mihaela Păun
类目: Machine Learning (cs.LG)
*备注:
Abstract:Access to sufficiently large biomedical datasets remains a major obstacle for machine learning in Raman spectroscopy-based diagnostics. In particular, for glioma analysis, datasets are typically small and heterogeneous, affected by acquisition-specific variability. This work investigates the utility of deep generative augmentation in such a small-cohort setting. We analyze glioma biopsy spectra acquired from 58 tumor samples and consider both binary IDH-status classification and 6-class methylation subtype classification problems. To address the limited size and imbalance of the dataset, we develop a conditional variational autoencoder ( \beta -CVAE) capable of generating class-conditioned synthetic Raman spectra. The generated data are evaluated in Train-on-Synthetic, Test-on-Real (TS/TR) and Train-on-Synthetic+Real, Test-on-Real (TSR/TR) settings under a strict patient-isolated cross-validation protocol. Models trained exclusively on synthetic data underperform models trained on real spectra, indicating a substantial domain gap between synthetic and real distributions. However, augmenting the real training data with synthetic spectra consistently improves classification performance across multiple models. These findings indicate that, even with a limited number of independent patient samples, generative models can capture sufficient structure to provide useful regularization for downstream classifiers. We also investigate a reconstruction-based inference strategy, termed Classification by Reconstruction (CbR), in which class prediction is based on reconstruction error under different class conditions. Overall, the results support the use of deep generative augmentation as a practical strategy for improving machine learning robustness in Raman spectroscopy applications characterized by limited biomedical datasets.
[LG-90] Knowledge-Conditioned Single-Pass LLM Synthesis of Executable Unity Game Scenes: A Compiler Error Census across 26 Goal Playable Concepts
链接: https://arxiv.org/abs/2607.10187
作者: Hugh Xuechen Liu,Kıvanç Tatar
类目: Machine Learning (cs.LG); Software Engineering (cs.SE)
*备注: Substantially reframed and extended version of arXiv:2603.07101 , with new experiments, figures, and analysis
Abstract:Large language models (LLMs) write Unity C# for game scenes. Yet nearly all demonstrations rest on an iterative repair loop that regenerates code until it compiles, conflating what the model writes with what the loop fixes. We remove the loop and evaluate a single pass, where the first draft is final. This isolates the model’s parametric knowledge, the most stringent test of unaided generation. Models instantiate Goal Playable Concepts, playable counterparts of goal patterns, across 10,400 generations (four open-weight models, 7B–30B; two generation modes; four intermediate-representation (IR) conditioning levels; 26 goal patterns; 20 seeds). None compiled into a runnable scene, leaving no survivorship bias. To understand how the generated C# scripts fail, we categorize the 99 error codes behind 90,673 compiler-error occurrences as Grounding (invented or misused Unity types and APIs) or Hygiene (structural defects needing no Unity knowledge). The split differs sharply by goal pattern (e.g., Stealth fails mostly on invented engine references; Capture on plain C# structure). Larger models, stricter IRs, and different generation modes move the errors but never yield a compiling scene. The bottleneck is missing engine-specific knowledge. The census orders goal patterns by that demand, showing designers where single-pass generation breaks.
[LG-91] On the Efficiency of LoRA Fine-Tuning for Vision-Language-Action Models in Industrial Robotic Manipulation ICANN2026
链接: https://arxiv.org/abs/2607.10172
作者: Finn Ferchau,Daniel Pommer,Cristian Axenie
类目: Robotics (cs.RO); Machine Learning (cs.LG)
*备注: 12 pages, 5 figures, 3 tables. Accepted at the International Conference on Artificial Neural Networks (ICANN 2026); to appear in Springer LNCS
Abstract:Deploying billion-parameter Vision-Language-Action (VLA) models on industrial hardware requires fine-tuning to bridge the embodiment gap. Full Fine-Tuning (FFT) provides maximal plasticity but requires data centre-grade GPUs. We present a systematic study of Low-Rank Adaptation (LoRA) for \pi_0 , a flow-matching VLA, evaluated on four precision assembly tasks with a UR5e robotic manipulator. Across a sweep of LoRA ranks (r=8 to 256), allocation strategies, and component-freezing ablations, we find no statistically significant advantage of FFT over certain LoRA configurations. Performance saturates at r=32, and uniform allocation across the Vision-Language-Model (VLM) backbone and action expert proves sufficient. Freezing the VLM or restricting the vision encoder to LoRA significantly degrades performance, indicating that embodiment adaptation requires both semantic and visual plasticity. These results suggest that LoRA at r=32 with full vision encoder fine-tuning is a practical approach, reducing static peak VRAM from 36.2 to 10.8 GiB (parameters and optimizer states, activation memory excluded) without detectable performance loss.
[LG-92] RDQ: Residual Distribution Quantization for Large Language Models
链接: https://arxiv.org/abs/2607.10137
作者: Prateek Singh
类目: Machine Learning (cs.LG)
*备注: 12 pages, 9 figures
Abstract:Post-training quantization (PTQ) of large language models degrades sharply below 4-bit precision. We identify the root cause as residual stream distributional drift: quantization noise injected at each transformer layer accumulates in the shared residual representation, causing KL divergence from the FP16 baseline to grow super-linearly with depth (Pearson r=0.999 with log-perplexity, p0.001, confirmed across all tested methods and bit-widths). We discover that 84% of LLaMA-3-8B layers exhibit non-Gaussian residual distributions (KS test, p=0.05), and that per-layer residual stream variance grows 6,548x across depth. We propose RDQ (Residual Distribution Quantization), a PTQ framework whose central contribution is Cascaded Error Compensation (CEC): a sequential calibration procedure that captures the actual drifted activations each layer receives (computed by running calibration data through already-quantized upstream layers) and fits per-channel AWQ-style scales against those drifted inputs, with scales folded into preceding RMSNorm weights for exact mathematical equivalence at zero inference overhead. RDQ achieves state-of-the-art results on all three tested architectures: LLaMA-3-8B: 7.55 / 5.62 PPL (W3/W4); Qwen-2.5-7B: 7.46 / 6.38 PPL; Mistral-7B: 6.88 / 5.73 PPL. RDQ beats the best published baseline (LeanQuant/SpinQuant) at every model and bit-width combination, with gains up to -46.4% vs. RTN at W3A16 on LLaMA-3-8B. All output is standard group-128 asymmetric quantization, deployable on Qualcomm AIMET, GGUF, and any standard inference stack at zero runtime overhead.
[LG-93] LeRoPE: Learnable RoPE Frequencies Improve Language Modeling
链接: https://arxiv.org/abs/2607.10134
作者: Petros Karypis,Sean O’Brien,Shreyas Kadekodi,Rui Zhu,Julian McAuley
类目: Machine Learning (cs.LG)
*备注: 27 pages, 10 figures, 12 tables
Abstract:Rotary Positional Encodings (RoPE) are currently the most popular positional encodings used in modern language models. RoPE rotates two-dimensional chunks of query and key vectors, operating as a function of their relative positional offset. The position-wise rates of rotation in RoPE typically follow a geometric sequence specified by a fixed base-frequency hyperparameter. Prior work has improved performance by either increasing this parameter to slow rotation or by applying RoPE to only a subset of QK dimensions. In this work we modify RoPE by learning a scalar per frequency, treating frequencies as learnable parameters rather than hyperparameters. We validate Learned RoPE by training a ladder of language models from scratch, ranging from 52M to 2.5B parameters. We observe and analyze the emergence of a high-norm, positional LeRoPE band. LeRoPE consistently outperforms RoPE and partial RoPE across all scales, with RoPE requiring 3.4% more compute (FLOPs) to match LeRoPE at the largest scale.
[LG-94] Energy-guided Recursive Model
链接: https://arxiv.org/abs/2607.10128
作者: Yifei Zhao,Ying Tang
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:Recursive reasoning models address structured problems by repeatedly updating latent states of small neural networks. However, their test-time scaling lacks a principled inference mechanism: increasing depth or stochastic breadth generates more trajectories without a clear criterion for selection, and existing methods predominantly rely on additional q-heads or heuristic voting. Here, we develop the Energy-guided Recursive Model (ERM), which introduces an intrinsic selection principle based on explicit Hopfield energies. ERM leverages Hopfield-type memories of valid local or global structures to define the selector over candidate trajectories. The resulting energy seamlessly integrates with energy-based techniques such as parallel tempering to enhance sampling efficiency and ranking. With D=64 recurrent steps and K=128 candidates, ERM reaches optimal solutions on Sudoku ( 98.97% ), Pencil Puzzle Bench (PPBench, 88.04% ) and Maze ( 99.30% ), improving upon recent Probabilistic Tiny Recursive Model and Equilibrium Reasoners. These results suggest that incorporating explicit energy functions into recursive reasoning offers a principled path toward more effective inference.
[LG-95] abLoRA: Parameter-Efficient Low-Rank Ensemble Learning for Large-Scale Tabular Data
链接: https://arxiv.org/abs/2607.10077
作者: Jiaqi Luo,Shixin Xu
类目: Machine Learning (cs.LG)
*备注:
Abstract:Tabular learning is still dominated by gradient-boosted decision trees (GBDTs), while recent deep learning approaches have become increasingly competitive. However, applying deep tabular models to large-scale datasets remains challenging, as large sample sizes, high feature dimensionality, or many target classes can introduce substantial computational cost. We propose TabLoRA, a parameter-efficient trainable neural ensemble for large-scale tabular learning. Instead of using fully independent ensemble backbones, TabLoRA shares a common backbone across predictors and introduces predictor-specific low-rank adaptations, enabling ensemble-style prediction without full parameter duplication. Across benchmarks, TabLoRA achieves a favorable balance between predictive performance and practical efficiency compared with GBDT methods and recent deep learning baselines under the same resource constraints. Memory analysis and ablation studies further show that the proposed design improves the feasibility of neural ensemble learning while preserving much of the benefit of full ensembles.
[LG-96] Distance-Preserving Embeddings in Inhomogeneous Random Graphs
链接: https://arxiv.org/abs/2607.10074
作者: My Le,Luana Ruiz,Souvik Dhara
类目: Machine Learning (cs.LG)
*备注:
Abstract:Graph machine learning provides powerful tools for understanding complex networks and learning meaningful node representations. A central challenge, however, is designing embeddings with minimal distortion of both local and global functionals, such as shortest path lengths. Prior distortion guarantees for distance-preserving embeddings are worst-case in nature, producing overly pessimistic bounds that fail to capture the structure of typical large-scale networks. To address this, we analyze shortest-path approximation via landmark-based embeddings on inhomogeneous random graphs, a general model with type-dependent edge probabilities. By retaining shortest paths to a small set of reference nodes called landmarks, landmark-based methods effectively function as virtual graph spanners, where structural heterogeneity and controlled neighborhood expansion modeled via multi-type branching processes enable significantly tighter dimension-distortion trade-offs than classical worst-case bounds. We extend these guarantees to global, component-wide averages and unify the analysis across finite-type and continuous latent spaces through a novel metric sandwiching framework, establishing universal distortion bounds for general L^2 kernel models, including heavy-tailed and power-law networks. Finally, we introduce a GNN-augmented variant that replaces rigid, computationally expensive exact shortest-path queries with flexible, structure-aware neural surrogates. By leveraging the inherent alignment between graph neural message-passing and the dynamic programming principles of shortest-path algorithms, our approach demonstrates that models trained on small-scale random graphs learn to extract universal distance-preserving features, achieving robust generalization to large-scale, real-world networks that match or exceed the fidelity of classical, exact landmark-based embeddings.
[LG-97] Conservation Laws for Diffusion Models
链接: https://arxiv.org/abs/2607.10067
作者: Ziv Aharoni,Henry D. Pfister
类目: Machine Learning (cs.LG); Information Theory (cs.IT); Machine Learning (stat.ML)
*备注:
Abstract:While autoregressive models optimize the exact data likelihood via the chain rule, diffusion models are typically trained with denoising objectives. We develop conservation laws based on generalized extrinsic information transfer (GEXIT) functions for a broad class of memoryless noise processes, showing that the data–model cross-entropy (CE) can be characterized exactly as an integral of local information-theoretic derivatives along the noise path. This yields a unified characterization of the likelihood for discrete and continuous diffusion, with the Gaussian case reducing to the well-known mutual information–minimum mean-square error (I-MMSE) relationship. An immediate implication is a locality property: one can compute the information-theoretic derivatives using only the marginal posteriors along the noise path. As a result, training reduces to learning the marginal posteriors by minimizing the negative log-likelihood. While the conservation law implies that the entropy does not depend on the noise path, finite-capacity denoisers approximate the posteriors with varying accuracy across noise types, leading to differences in performance. We validate these predictions on synthetic Markov sources and standard benchmarks, including text8 and CIFAR-10.
[LG-98] FlashTrie: A GPU-Accelerated Constrained Beam Search for Generative Retrieval
链接: https://arxiv.org/abs/2607.10044
作者: Dakshitha Anandakumar,Anurag Mukkara,Wenxiang Hu,Jiusheng Chen,M Akash Kumar,Ting Ye,Qiang Lou,Jian Jiao
类目: Machine Learning (cs.LG)
*备注:
Abstract:Constrained decoding is essential in generative retrieval, where document identifiers generated directly from a query must exactly match a predefined library of valid IDs. At scale, decoding is often constrained using a trie with beam search but most implementations run on CPU. Limited parallelism then makes trie traversal and candidate validation a serving bottleneck as beam width grows. We present FlashTrie, which addresses this limitation by optimizing constrained beam search on GPUs. It introduces an integer-aware succinct trie layout that uses bit compression to reduce memory footprint while keeping the full index in GPU high-bandwidth memory reducing memory stalls, and a cooperative CUDA kernel that performs beam expansion, validation, and pruning entirely on-device without per-step host orchestration. It further replaces CPU-style irregular lookup and heap maintenance with GPU-aware parallel primitives, improving warp utilization and reducing divergence. Together, these designs significantly reduce decoding latency and increase throughput while preserving retrieval quality. On a library of 800M keywords with beam widths up to 1000, FlashTrie reduces trie-search latency to under 3 ms, achieving up to 24x speedup over a highly optimized multi-threaded CPU baseline. These improvements enable FlashTrie to scale beam sizes by up to 5x in latency-critical applications such as sponsored search. In a large-scale online A/B experiment on a popular commercial search engine, it delivers a statistically significant +0.71% revenue lift, enabling real-time constrained decoding at a scale previously feasible only offline. The FlashTrie code will be publicly released after the review process. Subjects: Machine Learning (cs.LG) Cite as: arXiv:2607.10044 [cs.LG] (or arXiv:2607.10044v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2607.10044 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-99] MLPs are Hebbians: Constructing Efficient Fact-Storing MLPs for Transformers
链接: https://arxiv.org/abs/2607.10034
作者: Roberto Garcia,Jerry Liu,Ronny Junkins,Sabri Eyuboglu,Atri Rudra,Christopher Ré
类目: Machine Learning (cs.LG)
*备注:
Abstract:Large language models (LLMs) store factual knowledge in their parameters. While recent work has shown that this knowledge resides in MLP layers, existing constructive and mechanistic interpretability models of fact-storage in LLMs fail to explain the surprising empirical phenomenon that they store facts at an information-theoretically optimal rate. In this work, we develop a theoretical account of this phenomenon. We develop the first Transformer-compatible fact-storing MLP closed-form construction that satisfies the following three properties empirically observed in LLMs: it (i) attains optimal fact storage scaling, (ii) handles arbitrary input/output geometries, and (iii) works inside Transformers. Key to our work is to analyze the decoding margin of MLPs, whereas prior work only studies MLP fact storage. Under isotropic embeddings, our construction achieves information-theoretically optimal storage capacity scaling and requires 10 - 104\times fewer parameters at matched fact count than prior constructions. For arbitrary key and value embeddings, we show that our construction attains the same storage capacity scaling, up to penalization factors depending on the embedding geometries. Moreover, we demonstrate that our constructed MLPs can be used within Transformer blocks for factual recall tasks at optimal capacity scaling, requiring 15 - 63\times fewer parameters at matched fact count than prior constructions. Finally, as a proof-of-concept, we show that fact-storing MLPs enable modular fact editing by swapping a Transformer’s MLP with a new one.
[LG-100] Robustly Invertible Nonlinear Dynamics and the BiLipREN: From Inversion-Based Control to Generative Trajectory Modelling
链接: https://arxiv.org/abs/2607.10026
作者: Yurui Zhang,Ruigang Wang,Ian R. Manchester
类目: ystems and Control (eess.SY); Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注:
Abstract:This paper proposes a new notion of robust invertibility for nonlinear dynamical systems, and introduces constructive parameterizations of recurrent neural network which are robustly invertible by design. We define robust invertibility as the existence of a causal inverse system such that both the forward and inverse systems are contracting and have bounded incremental input-output gains (the system is bi-Lipschitz), implying that both forward prediction and input reconstruction are robust to signal perturbations and initial-state mismatch. We construct robustly invertible recurrent models via series composition of static orthogonal layers and dynamic layers satisfying a strong input-output monotonicity property, and provide a differentiable neural network parameterizations in the form of the bi-Lipschitz recurrent equilibrium network (BiLipREN). Additionally, composition with dynamic orthogonal layers yields a nonlinear minimum-phase/all-pass (a.k.a. inner–outer) factorization. We illustrate the utility of the framework through a series of application examples in data-driven internal model control, dynamic surrogate loss learning, and signal-space normalizing flows, illustrating its utility for robust control, trajectory optimization, and generative modeling of complex trajectory distributions.
[LG-101] Local Multimodal Music Alignment from Global Supervision
链接: https://arxiv.org/abs/2607.10023
作者: Irmak Bukey,Zachary Novack,Jongmin Jung,Dasaem Jeong,Chris Donahue
类目: ound (cs.SD); Machine Learning (cs.LG); Multimedia (cs.MM)
*备注: ISMIR 2026
Abstract:Understanding music requires understanding localized relationships across data modalities, e.g., how time in performance audio maps onto position in a score image. Yet supervision for such local correspondences is difficult to obtain-in practice, we often only have access to coarser global supervision like paired segments of audio and images. To address this gap, we propose FuSiLi (Fused Sinkhorn-Localized Similarity), a similarity score for multimodal contrastive learning operating directly on local image patch and audio frame features via Sinkhorn-based soft alignment. We show that FuSiLi (i) effectively learns local relationships, (ii) requires only global supervision, and (iii) retains the global alignment capabilities of conventional contrastive approaches. We fine-tune pretrained CLIP and CLAP encoders on pairs of raw sheet music images and audio using a hybrid contrastive objective combining FuSiLi with conventional global similarity. We evaluate on cross-modal retrieval and frame-level alignment tasks against a range of global and local baselines, showing that our approach outperforms them on local alignment while remaining competitive on retrieval.
[LG-102] Vilya-1: An all-atom foundation model for macrocycle structure prediction and design
链接: https://arxiv.org/abs/2607.09998
作者: Vilya Research:Pascal Sturmfels,Milad Salem,Naozumi Hiranuma,Stephen Rettie,Xiaoliang Pan,Benjamin D. Sellers,Adam P. Moyer,Patrick J. Salveson,Ivan Anishchanka
类目: Machine Learning (cs.LG); Biomolecules (q-bio.BM)
*备注: 21 pages, 14 figures
Abstract:Macrocyclic peptides are an increasingly important therapeutic modality, but existing computational methods for modeling their structures and properties are limited in scope and do not generalize well across the synthetically accessible chemical space. In this work, we introduce Vilya-1, a deep learning model that addresses two central challenges in macrocycle design: sampling biologically relevant conformations across arbitrary chemistries and predicting key developability properties such as membrane permeability. Vilya-1 operates on a uniform all-atom representation and is trained on heterogeneous structural datasets spanning diverse topologies and chemical classes. Across a broad set of macrocycles composed of canonical and non-canonical residues, Vilya-1 substantially improves geometric accuracy relative to physics-based methods, co-folding networks, and deep-learning conformer generators, while maintaining broad chemical coverage that extends to small molecules. Vilya-1 also supports generative applications, enabling the design of novel macrocycles with tailored chemical, structural, and property profiles. Together, these capabilities establish Vilya-1 as a foundation model for accelerating the development of next-generation macrocycle therapeutics.
[LG-103] Multimodal Routing for Interpretable Robust and Auditable Clinical Prediction
链接: https://arxiv.org/abs/2607.09982
作者: Nikkie Hooman,Zhongjie Wu,Eric C. Larson,Mehak Gupta
类目: Machine Learning (cs.LG)
*备注: Accepted at CHASE 2026. 12 pages, 6 figures, 4 tables
Abstract:Electronic health record (EHR) data are inherently multimodal, and leveraging multiple modalities can improve predictive performance. However, most existing approaches rely on deep fusion, which obscures how individual modalities contribute to predictions and limits the interpretability of multimodal reasoning. We propose an explicit multimodal routing framework for clinical prediction that enables interpretable, robust, and auditable reasoning across three EHR modalities: structured longitudinal variables (L), clinical notes (N), and chest X-rays (I). Our model constructs discrete unimodal, directional bimodal, and trimodal routes to capture both individual modality signals and asymmetric cross-modal interactions. To audit multimodal reasoning and assess robustness, we introduce inference-time route masking, which simulates missing modalities and reweights the remaining routes without retraining. We analyze changes in performance and routing weights under these scenarios to understand model decision-making. We evaluate our framework on multi-label phenotype prediction (K = 25) and binary ICU mortality prediction using trimodal patient stays from MIMIC-IV, revealing systematic differences in modality reliance across clinical condition groups. Overall, our framework offers a transparent, auditable, and practical approach to multimodal clinical prediction, providing interpretability, robustness, and insights into how different data sources drive model decisions.
[LG-104] Optimizing ARDL Models for Retail Sales Forecasting and Fair Pricing
链接: https://arxiv.org/abs/2607.09956
作者: Sujay Uday Rittikar
类目: Machine Learning (cs.LG); Computational Engineering, Finance, and Science (cs.CE)
*备注: 12 pages, 1 figure
Abstract:Pricing food products to balance profitability with consumer welfare is a central challenge for retailers. Dynamic pricing is widely used to maximize revenue, yet most pricing models optimize business objectives while overlooking consumer fairness. This paper studies the risk of consumer exploitation under dynamic food pricing in Canada and proposes a methodology that embeds fairness constraints directly into retail sales forecasting. We model total retail trade sales with a log–log Autoregressive Distributed Lag (ARDL) specification, in which the coefficient on a product price is a sales elasticity, and pose the pricing problem as maximizing forecast sales subject to price bounds anchored to the Consumer Price Index (CPI). We solve this problem with both Linear Programming (LP) and Simulated Annealing (SA), under single-product and multi-product configurations. A key finding is that the fitted nominal elasticities are positive. As a result, an unconstrained sales-maximizer would push every price to its upper bound, and the CPI ceiling is the safeguard that prevents this. Simulated Annealing instead settles on conservative, interior prices that lower consumer cost while still meeting the sales target. We benchmark forecast accuracy against naive, seasonal-naive, ARIMA, and SARIMA baselines, and a CPI-deflated re-specification shows that the positive nominal elasticities are largely an inflation-driven artifact. The result is a transparent, fairness-aware pricing framework.
[LG-105] Learning Partition Trees for Nearest Neighbor Search
链接: https://arxiv.org/abs/2607.09909
作者: Sanjeev Khanna,Ashwin Padaki,Erik Waingarten
类目: Data Structures and Algorithms (cs.DS); Machine Learning (cs.LG)
*备注:
Abstract:We study nearest neighbor search from the perspective of data-driven algorithm design: given a dataset P \subset \mathbbR^d of size n and sample access to a query distribution over \mathbbR^d , the goal is to learn a data structure optimized for queries drawn from that specific distribution. We focus on the class of balanced halfspace trees, which naturally abstracts space-partitioning frameworks like locality-sensitive hashing. Assuming Gaussian-like marginal conditions on the dataset and query distribution, we give an efficient algorithm that learns a tree achieving o(nd) query time, provided that a perfect tree exists. At the core of our algorithmic approach is the balanced halfspace cut problem, where we are given a distribution over \mathbbR^d \times \mathbbR^d and must find a balanced halfspace that minimizes the fraction of cut pairs. We prove that without distributional assumptions, finding the optimal balanced halfspace is NP-hard. To circumvent this computational barrier, we design an efficient improper learning algorithm: if the optimal halfspace cuts an \alpha fraction of pairs, our algorithm outputs a balanced polynomial threshold function of degree \tildeO(1/\varepsilon^2) that cuts at most an O(\sqrt\alpha+\varepsilon) fraction. Subjects: Data Structures and Algorithms (cs.DS); Machine Learning (cs.LG) Cite as: arXiv:2607.09909 [cs.DS] (or arXiv:2607.09909v1 [cs.DS] for this version) https://doi.org/10.48550/arXiv.2607.09909 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-106] Nonparametric Bayesian Inverse Reinforcement Learning with Data-Parallel Gibbs Sampling
链接: https://arxiv.org/abs/2607.09886
作者: Sai Anirudh Katupilla,Shreeya Dasa Lakshminath
类目: Machine Learning (cs.LG)
*备注: 6 pages, 5 figures
Abstract:Inverse Reinforcement Learning recovers reward functions from expert demonstrations, but standard formulations assume that all demonstrations come from a single expert. When demonstrations are pooled from multiple experts with distinct preferences, parametric methods recover an averaged reward that fits no individual expert well. We implement Nonparametric Bayesian Inverse Reinforcement Learning with a Dirichlet Process prior over reward functions, allowing the number of latent reward types to be inferred jointly with the rewards themselves. Inference uses a collapsed Gibbs sampler combining a Chinese Restaurant Process update for cluster assignments with a Metropolis-Hastings update for reward weights, and soft value iteration as the inner planning routine. We evaluate on a 10x10 ObjectWorld grid with two and three ground-truth reward types. The serial sampler recovers K=2 with Adjusted Rand Index of 1.000, substantially outperforming a Maximum Entropy IRL baseline (ARI=0.000). Extension to K=3 shows that the sampler correctly identifies the number of clusters in all runs; assignment ARI of 0.48-0.58 reflects behavioral overlap between expert types that persists across grid instantiations, revealing that reliable K=3 evaluation on ObjectWorld requires controlled object placement rather than random seeding. We further parallelize the sampler across CPU cores using Ray on HPC hardware, achieving a peak speedup of 4.79x at 8 workers, and characterize a throughput-versus-accuracy tradeoff arising from the consensus merge heuristic used during state aggregation. Code and a containerized environment are available at this https URL.
[LG-107] Nonlinear Axiomatic Attribution for Cooperative Games
链接: https://arxiv.org/abs/2607.09869
作者: Weida Li,Zhuanghua Liu,Yaoliang Yu,Bryan Kian Hsiang Low
类目: Machine Learning (cs.LG)
*备注:
Abstract:The Shapley value is a widely used concept in attribution problems, as it uniquely satisfies the axioms of linearity, consistency, equal treatment, and efficiency. Often, the inclusion AUC metric is used to evaluate the quality of player rankings, in order to identify positively participating players. However, it can be established that the Shapley value is not always reliable for this purpose. The core issue lies in its linearity: the Shapley value acts as a linear operator with an excessively large null space, which is likely to contain non-negligible perturbations that remain indistinguishable to the operator. To address this limitation, we explore the design of nonlinear axiomatic attribution methods. Inspired by the least core, which is a popular nonlinear substitute for the Shapley value, we introduce a class of nonlinear attribution methods that retain the remaining necessary axioms. Each method yields a contribution vector that is the unique optimal solution to a minimization problem, which aims to approximate utility functions as faithfully as possible. In terms of the inclusion AUC metric, our experiments demonstrate the potential effectiveness of these methods compared to Shapley value variants that relax only the efficiency axiom. Our code is available at this https URL.
[LG-108] Generative Testing of Automated Speech Recognition Systems
链接: https://arxiv.org/abs/2607.09833
作者: Yanis Xabier Wilbrand Peña,Oliver Weißl,Andrea Stocco
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注:
Abstract:Automatic speech recognition (ASR) systems have achieved high accuracy with transformer-based models, enabling deployment in critical applications. However, they remain vulnerable to adversarial manipulation, particularly in black-box settings where attacks must preserve perceptual naturalness. This work introduces GATAS, a black-box testing approach that generates failure inducing inputs by operating in the phoneme-level latent space of a text- to-speech model. Instead of perturbing waveforms directly, the approach interpolates latent representations to induce transcription errors while remaining within the manifold of natural speech. The attack is formulated as a multi-objective optimization problem balancing semantic divergence and perceptual quality. Our empirical evaluation against both white-box and black-box baselines shows that GATAS achieves a 98% success rate while producing lower distortion and higher perceptual quality, as confirmed by human studies. Despite operating without gradient access, GATAS remains competitive against white-box methods, highlighting that representation and perceptual alignment are more critical than access to model internals. Overall, our results demonstrate that untargeted latent-space optimization enables the efficient generation of realistic and effective test cases for ASR systems.
[LG-109] Learning Predictive Ambiguity Sets for Decision-Focused Distributionally Robust Optimization
链接: https://arxiv.org/abs/2607.09820
作者: Junjie Guo
类目: Machine Learning (cs.LG); Computational Finance (q-fin.CP)
*备注:
Abstract:Predict-then-optimize systems usually compress uncertainty into a point forecast and then solve a downstream optimization problem as if the forecast were reliable. Distributionally robust optimization (DRO) offers protection against misspecification, but the ambiguity set is often centered at historical samples and uses a fixed radius. We propose \emphlearned predictive ambiguity sets (LPAS): a deep contextual model outputs a finite nominal scenario distribution, a state-dependent Wasserstein radius, and optionally an anisotropic ground metric. These outputs define a contextual ambiguity set that feeds a DRO decision layer. The radius is trained by a combination of conditional quantile calibration, size regularization, and downstream decision loss, so that robustness is adaptive rather than globally fixed. We derive the finite dual form used by the decision layer, present a staged training algorithm, and evaluate the method on distributionally robust portfolio optimization with 20 SP 500 constituents from 2018–2026. The proposed method substantially improves over equal-weight, predict-then-optimize, and historical Wasserstein DRO baselines, achieving 26.28% annualized return, Sharpe ratio 1.30, final wealth 1.61, and lower tail loss than a deep fixed-radius DRO baseline while using a smaller average radius. The results show that learned ambiguity radii can recover most of the performance of strong fixed-radius DRO while reducing unnecessary conservatism and improving regime adaptivity.
[LG-110] Estimation Prediction and Assortment Optimization for Markov Chain Choice Models with Panel Data
链接: https://arxiv.org/abs/2607.09817
作者: Yalcin Akcay,Gerardo Berbeglia,Young-San Lin
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:We propose a framework for the Markov chain (MC) choice model with panel data, including parameter estimation, personalized choice prediction, and personalized assortment optimization. In contrast to the traditional setting, which assumes that each transaction is independently drawn from a random utility model, our framework accounts for dependencies among transactions for the same customer in historical data, captured by partial-ordering preference information. To the best of our knowledge, our framework initiates the study of choice modeling with panel data under MC. As our primary result, we propose novel expectation-maximization (EM) algorithms for MC parameter estimation by incorporating partial-ordering-based customer preference information. On synthetic datasets and the sushi dataset, our EM algorithms outperform the traditional EM algorithm of Simsek and Topaloglu (Operations Research, 66, 2018) and multinomial-logit-based partial-order benchmarks adapted from Jagabathula and Vulcano (Management Science, 64, 2018). As our secondary contribution, we present hardness and computational results for conditional choice prediction and assortment optimization problems. These results complement our estimation framework and clarify the computational landscape of conditional choice and assortment optimization, which may be of independent interest.
[LG-111] RUBRIC: Realism–Utility Balanced Ranking for Imbalanced Classification
链接: https://arxiv.org/abs/2607.09816
作者: Yanxuan Yu,Dong liu,Renata Borovica-Gajic,Ying Nian Wu
类目: Machine Learning (cs.LG)
*备注:
Abstract:Class imbalance poses a fundamental challenge in risk-sensitive applications such as fraud detection and medical diagnosis, where minority-class samples are scarce yet critical for accurate classification. Existing oversampling methods generate synthetic samples to rebalance class distributions; however, they often produce large numbers of low-quality candidates that distort decision boundaries or introduce artifacts, leading to overfitting and degraded generalization. In this work, we introduce RUBRIC, a generator-agnostic filtering framework that formulates synthetic sample selection as a quality-over-quantity optimization problem. RUBRIC ranks candidates using a realism-utility trade-off: realism is quantified by a learned discriminator that distinguishes real samples from synthetic samples, while utility captures proximity to the decision boundary through a concave margin-based scoring function. We show that, under mild regularity conditions, the proposed filtering strategy monotonically tightens the generalization bound for margin-based classifiers by jointly reducing distribution shift and suppressing near-negative tail contributions. Through extensive experiments on credit-card fraud detection and other imbalanced benchmarks, we demonstrate that RUBRIC improves F1-macro and recall while maintaining comparable ROC-AUC across several generators. We also provide explicit lambda-sensitivity analysis to show how users can recover AUPRC when ranking quality is prioritized. Subjects: Machine Learning (cs.LG) Cite as: arXiv:2607.09816 [cs.LG] (or arXiv:2607.09816v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2607.09816 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-112] Quota Marketplace: Dynamic Pricing for Efficient Allocation of ML Training Resources OSDI2026
链接: https://arxiv.org/abs/2607.09802
作者: Balasubramanian Sivan,Renato Paes Leme,Mihai Tiuca,Ian McFarlane,Vasilis Gkatzelis,Nehal Mehta,Soheil Hassas Yeganeh,Vahab Mirrokni,Amin Vahdat
类目: Machine Learning (cs.LG); Computer Science and Game Theory (cs.GT)
*备注: OSDI 2026 Paper
Abstract:The escalating demand for Machine Learning (ML) training resources in recent years has resulted in a substantial gap between the high demand and the available supply. Efficient allocation of these scarce and expensive resources is crucial for organizations to maximize their return on investment. Existing resource allocation mechanisms, like Karma [OSDI’23], are designed to guarantee Pareto efficiency and max-min fairness in settings with dynamic (time-varying) user demands, but fail to preserve these key properties in the presence of demands with heterogeneous values. Given the ubiquity and inevitability of heterogeneity in organizational values of different workloads, effective resource allocation policies must accommodate these variations. In this paper, we describe the design, implementation, deployment, and theoretical analysis of Quota Marketplace, a market-based mechanism to efficiently allocate ML training chips (like GPUs), explicitly addressing scenarios with demands of heterogeneous value. We detail the implementation of this mechanism within Google and present metrics that demonstrate its impact. We also discuss many business-critical requirements that the Quota Marketplace handles quite effectively, and document the gains and opportunities it has unlocked. We establish theoretically how this market-based approach achieves the essential properties of Pareto efficiency and max-min fairness by allowing the users to express the value of their workloads and enabling dynamic resource pricing based on supply and demand fluctuations. Ultimately, the market facilitates resource allocation that aligns with organizational priorities. Comments: OSDI 2026 Paper Subjects: Machine Learning (cs.LG); Computer Science and Game Theory (cs.GT) Cite as: arXiv:2607.09802 [cs.LG] (or arXiv:2607.09802v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2607.09802 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-113] Discovering Latent Response Laws in Forced Physical Systems
链接: https://arxiv.org/abs/2607.09801
作者: Yi Zhu,Su Chen,Xiaojun Li,Xiuli Du
类目: Machine Learning (cs.LG)
*备注: 23 pages, 5 figures and 2 tables. Supplementary Information provided as an ancillary file
Abstract:Governing equations provide compact descriptions of physical systems, yet the variables in which they are simple are often hidden in high-dimensional measurements. This challenge is sharper for forced systems, whose responses depend on both intrinsic dynamics and time-dependent inputs. Here we introduce FLARE, a forced latent autoencoder for response equations that learns compact response coordinates, identifies sparse input-dependent latent dynamics and decodes equation rollouts to full responses. By estimating latent dimension from data and separating state estimation from external forcing, FLARE enables forecasts to be initialized from past responses and driven by prescribed future inputs. Across known dynamical systems, application-scale forced responses and visual observations, FLARE recovers compact forced dynamics and predicts long-horizon high-dimensional responses under inputs not used for training. By turning learned coordinates into a dynamical interface, FLARE extends equation discovery to systems whose effective states are hidden within complex observations, providing a route for interpretable modelling and prediction of high-dimensional responses in forced dynamical systems.
[LG-114] he Silent Freeze: Predicting When Low-Precision Training Stops Learning
链接: https://arxiv.org/abs/2607.09800
作者: Zekai Shang
类目: Machine Learning (cs.LG)
*备注:
Abstract:Training in reduced floating-point precision can silently halt learning: when a gradient-descent weight update falls below half the unit in the last place (ULP) of the weight, it rounds away and that coordinate freezes while its gradient is still nonzero. The freeze is deterministic, governed by a per-coordinate half-ULP condition, and predictable from a high-precision trajectory and the target mantissa length alone, without low-precision data. In a small GPT trained under the standard AdamW-plus-cosine recipe with bf16-equivalent stored weights, training proceeds normally and then permanently freezes just past mid-run, within four steps of the a-priori prediction. In a 124 -million-parameter GPT-2 transformer whose weights are constrained to the 8 -bit floating-point grid after every optimizer step, with no master weights, the dense weights freeze at initialization in both fp8 formats – predicted \empha priori from an fp32 reference – and validation loss plateaus while full precision keeps improving. Stochastic rounding removes the persistent freeze, and the same reference predicts that too. The condition transfers across frozen-feature regression, a mantissa-truncation emulator spanning 128\times in precision, small networks, and a CNN on MNIST: a computable axis of low-precision training, not diffuse noise.
[LG-115] Metadata-Free Meta-Reweighted Direct Preference Optimization under Noisy Preference Labels
链接: https://arxiv.org/abs/2607.09796
作者: Hua Qu,Yifan Li,Xiaodong Yuan
类目: Machine Learning (cs.LG)
*备注: 41 pages
Abstract:Direct Preference Optimization (DPO) has become an important method for aligning large language models (LLMs) with human preferences because it removes the need for explicit reward modeling and reinforcement learning optimization. However, its performance depends heavily on the quality of preference data, and noisy preference data in real-world settings can weaken alignment performance. To address this issue, we propose a bilevel optimization framework and prove, under certain assumptions, that this framework can recover the DPO optimum under clean data. We further derive a prior form for the learnable weighting function under asymmetric label-flipping noise. Considering that high-quality metadata may be difficult to obtain, we propose a task-agnostic meta-knowledge-driven method that enables meta-learning even when metadata is completely unavailable. To reduce the high cost of higher-order gradients in LLM meta-learning, we combine central-difference approximation with LoRA fine-tuning and develop a scalable training scheme. Experiments on TL;DR summarization and Anthropic HH single-turn dialogue show that the proposed method improves training performance over multiple DPO baselines under different noise rates.
[LG-116] A Risk-Field Enhanced Closed-Loop Digital Twin Framework for Autonomous Driving Safety Validation
链接: https://arxiv.org/abs/2607.09772
作者: Yongzhi Liu
类目: Robotics (cs.RO); Machine Learning (cs.LG)
*备注:
Abstract:Autonomous driving systems require reliable safety validation before real-world deployment. However, large-scale road testing is costly, difffcult to reproduce, and inefffcient for exposing rare safety-critical scenarios. Conventional simulation improves repeatability, but an offfine simulator alone cannot continuously connect physical trafffc states, virtual reconstruction, algorithm evaluation, and scenario evolution. This paper proposes a risk-ffeld enhanced closed-loop digital twin framework for autonomous driving safety validation. The framework integrates physical data acquisition, data synchronization, virtual twin reconstruction, risk-aware scenario generation, autonomous driving algorithm evaluation, and safety analysis. A driving risk ffeld is introduced as a uniffed intermediate representation to describe obstacle, lane-departure, road-boundary, time-to-collision, and comfort-related risks around the ego vehicle. The risk ffeld ranks high-risk scenarios in the digital twin scenario library and provides dense safety guidance for reinforcement learning-based driving policies. A simulation-style evaluation protocol is designed to compare conventional reinforcement learning baselines, risk-penalty baselines, and the proposed risk-ffeld guided method. The study indicates that embedding explicit risk structure into digital twins can make autonomous driving validation more targeted, interpretable, and reusable, while its practical effectiveness remains bounded by model ffdelity, risk calibration, and sim-to-real transfer.
[LG-117] Saturation-Aware Robust Trajectory Optimization for Reusable Launch Vehicles via Differentiable Physics
链接: https://arxiv.org/abs/2607.09736
作者: Liwei Chen,Tong Qin
类目: Robotics (cs.RO); Machine Learning (cs.LG); Applied Physics (physics.app-ph)
*备注:
Abstract:The high-angle-of-attack flip maneuver of reusable launch vehicles presents significant challenges for robust trajectory optimization due to the combined effects of highly nonlinear dynamics, aerodynamic uncertainties, and actuator saturation. This paper presents a differentiable physics framework for saturation-aware robust trajectory optimization. At its core, a Differentiable Particle Tube Control (DPTC) scheme is developed to optimize uncertainty evolution through an ensemble-based distribution shaping strategy. State uncertainty is represented by a Lagrangian particle ensemble, while hard actuator projection operators are embedded directly into the computational graph, enabling the joint optimization of the nominal feedforward trajectory and a time-varying feedback policy via end-to-end backpropagation. The proposed framework is evaluated against an automatic differentiation-based Successive Convexification (AD-SCvx) baseline combined with a conventional covariance steering feedback strategy. Six-degree-of-freedom Monte Carlo simulations demonstrate that, although the baseline achieves nominal fuel-optimal solutions, its unconstrained feedback formulation becomes susceptible to actuator saturation under aerodynamic disturbances, leading to degraded closed-loop robustness. In contrast, the proposed DPTC framework proactively performs a constraint-aware performance trade-off by relaxing spatial tracking to preserve critical control authority. These results demonstrate that integrating differentiable physics with ensemble-based optimization provides an effective and practical framework for robust guidance in highly constrained aerospace flight systems.
[LG-118] ERP Data Provisioning Financial Control Testing
链接: https://arxiv.org/abs/2607.09712
作者: Anitha Samudrala
类目: Machine Learning (cs.LG)
*备注:
Abstract:Financial control testing increasingly depends on representative enterprise resource planning (ERP) data in quality environments, yet direct production copies expose personal, supplier, banking, and commercially sensitive records. This work presents Secure ERP Quality Provisioning for Financial Control Testing (SEQ-FCT), a governed data-provisioning framework that combines deterministic masking, synthetic scenario expansion, referential tokenization, policy-based release approval, and automated validation for reconciliation, fraud-rule testing, and audit analytics. A single synthetic dataset is used for evaluation. It contains 186,000 finance-process records from six subsidiaries over 2022-2025, including accounts payable invoices, payments, general-ledger journals, accounts receivable receipts, and bank-statement lines. The dataset includes entity relationships, monetary values, approval paths, tax attributes, banking markers, exception labels, fraud-rule triggers, and control-failure outcomes. Because the dataset is synthetic, reported results demonstrate controlled internal consistency rather than production validation. Against a production-clone upper bound, static masking, rules-only synthesis, conditional tabular generative synthesis, and a hybrid baseline, SEQ-FCT achieved 0.932 reconciliation F1, 0.887 fraud-trigger recall, 0.914 control-failure F1, and an estimated leakage-risk score of 0.018. The analysis indicates that financial process behavior can be preserved more reliably when masking, synthetic data, and governance checks are evaluated as a single release pipeline instead of independent utilities.
[LG-119] EvoClawBench: Can Agents Learn Reusable Skills from Their Own Runs?
链接: https://arxiv.org/abs/2607.09711
作者: Zhiyuan Peng,Xin Yin,Chenhao Ying,Zhe Cui,Zixiang Ding,Zhenhua Liu,Jiang Wu,Yuan Luo
类目: Machine Learning (cs.LG); Software Engineering (cs.SE)
*备注:
Abstract:Existing agent benchmarks primarily test task completion, tool use, or skill utility, but do not isolate whether a runtime can convert evidence from its own runs into reusable skills that improve fresh executions after authoring overhead. We introduce EvoClawBench, a benchmark for this closed-loop skill-learning question on repeated, fixture-backed tasks. EvoClawBench compares direct execution without skills, PreSkill authoring before execution, and PostSkill summarization from first-run evidence followed by a fresh second execution. The suite contains 100 tasks and 502 sub-problems across coding, data, office, security, operations, and domain-document workflows, with support for multiple agent runtimes. Experiments with OpenClaw and nanobot under local execution show that direct baseline performance is strongly runtime-dependent: OpenClaw remains below 20% across models, while nanobot ranges from 56.45% to 96.13%. Self-authored skills have mixed effects. nanobot GPT-5.4 stays above 96% in all modes and MiniMax-M2.7 improves from 90.97% to 94.50% under PostSkill, but nanobot DeepSeek-V4-Pro drops from 77.77% to 4.80% with PreSkill and 0.99% with PostSkill. OpenClaw shows similarly non-monotonic behavior, with some skill runs near baseline and others collapsing. These results indicate that learning reusable skills from an agent’s own runs is selective and cost-sensitive, rather than an automatic benefit of adding skill authoring to an agent loop.
[LG-120] Manifold Constrained Tabular Deep Neural Networks
链接: https://arxiv.org/abs/2607.09710
作者: Tian Li,Lucy Robinson,Varun Ojha,Huizhi Liang
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:Tabular classification is often governed by local, condition-triggered rules rather than smooth global patterns. However, tabular deep neural networks (DNNs) are typically built upon Euclidean representations that favor smooth variations and semantic locality. This potential geometric mismatch can make it challenging for tabular DNNs to efficiently represent the discrete, rule-partitioned structures often underlying tabular classification. To address this issue, we propose HDE-Net, a manifold-constrained DNN that enables hierarchical decision modeling in hyperbolic space. We first abstract heterogeneous features into unified Latent Decision Nodes (LDNs) and embed them in the Poincaré ball, forming a continuous representation that resembles tree-structured reasoning. For numerical features, we introduce a Soft Decision Routing mechanism that approximates range-based local rules in a differentiable manner, bringing their LDN semantics closer to those of categorical features. An entropy-aware capacity allocation algorithm further adapts the number of LDNs per numerical feature to balance expressiveness and complexity. On the TALENT-tiny-core classification benchmark (30 datasets), HDE-Net achieves the \textitbest average rank, outperforming both industrial GBDTs and recent tabular DNNs while maintaining high efficiency.
[LG-121] Quantum-Inspired Contextual Learning for Sparse-Ring Fraud Detection in Dynamic Transaction Graphs
链接: https://arxiv.org/abs/2607.09704
作者: Behnam Tonekaboni,Hiroshi Yamauchi
类目: Machine Learning (cs.LG)
*备注: 23 pages, 4 figures
Abstract:We present an exploratory benchmark and quantum-inspired modeling prototype for fraud screening in dynamic financial transaction graphs. Coordinated fraud may not be visible from individual transactions alone, but may emerge as a multi-period relational pattern. We focus on sparse-ring fraud, a stylized pattern in which a completed directed cycle is distributed across several days, requiring models to integrate evidence across both time and graph structure. We study this problem using a synthetic transaction simulator with completed sparse-ring injections and broken-ring decoys. Daily directed transaction graphs are aggregated into rolling windows and represented using raw graph features, persistent-homology summaries, or hybrid feature vectors that combine both. We compare a gated recurrent unit (GRU) baseline with quantum-inspired Contextual Machine Learning (CML) as sequence-level classifiers. Because the benchmark uses synthetic data, a modest sample size, and sequence-level labels, the results are exploratory. Within this scope, topology-only summaries are too compressed to solve the supervised ring-completion task by themselves, largely because they remove account-pair identity and edge direction. The strongest results come from hybrid representations that combine identity-preserving graph features with topological summaries. These findings suggest that topology is most useful as a contextual layer over dynamic graph features, and that CML is a promising candidate model for fraud patterns whose evidence is distributed across temporal and relational context.
[LG-122] Safe responses matter: Output-aware safety guardrail mitigate over-refusal in MLLM s ECCV2026
链接: https://arxiv.org/abs/2607.09697
作者: Jiayi Li,Kun Zhan
类目: Machine Learning (cs.LG)
*备注: Accepted to ECCV 2026!
Abstract:Existing safety mechanisms for multimodal large language models (MLLMs) face a fundamental trade-off between safety and utility. Model fine-tuning achieves robust safety but compromises general utility. Input-side safety guardrails offer a lightweight alternative, yet they suffer from severe over-refusal, indiscriminately blocking benign queries or those the model could have safely answered through refusal or advisory responses. We identify that the root cause of over-refusal lies in the input-aware paradigm: safety guardrails make safety decisions without considering whether the model itself is capable of generating safe responses. Usually, MLLMs already possess intrinsic safety mechanisms that can transform harmful inputs into harmless outputs, but input-side safety guardrails override this capability, degrading user experience. Motivated by this insight, we propose a paradigm shift toward output-aware safety guardrails. Our method operates within the model’s hidden state space to predict whether the forthcoming generation will be unsafe before it is fully produced. By training a lightweight classifier via multi-instance contrastive learning on hidden state representations, our approach distinguishes between inputs that will lead to unsafe outputs and those that will not, even when the inputs themselves contain risky elements. This enables precise intervention only when the model’s actual response would be harmful. Extensive experiments demonstrate that our output-aware safety guardrail matches the safety performance of existing methods while drastically reducing over-refusal, preserving the model’s utility and built-in safety capabilities. Code is available at: this https URL
[LG-123] FedCausal-Dyn: A Causal-Dynamic Paradigm for Federated Learning under Dynamic Feature Drift
链接: https://arxiv.org/abs/2607.09695
作者: Kaijie Chen,Alex Johnson,Maria Garcia,Wei Zhang,Daniel Kim
类目: Machine Learning (cs.LG)
*备注: 18 pages
Abstract:This paper addresses the challenging problem of dynamic feature drift in federated learning, where data distributions evolve across clients and over time – a common scenario in real-world applications like financial technology. Existing approaches often assume static drift, limiting their effectiveness in non-stationary environments. To overcome this, we propose \textbfFedCausal-Dyn, a novel federated learning framework built on a causal-dynamic paradigm. Its key innovation is \textitcausal-domain feature separation, which disentangles domain-invariant causal features from spurious, domain-specific variations via specialized projection heads and adversarial training. This enables \textitreliable and dynamic prototype aggregation, weighting local class prototypes by estimated reliability before global aggregation. We further introduce \textitcausal-feature guided collaborative regularization, unifying prototype contrastive alignment and domain invariance into a cohesive objective. Extensive experiments on three federated domain generalization benchmarks demonstrate that FedCausal-Dyn consistently achieves state-of-the-art performance, with the highest average accuracy and the most stable results. Ablation studies confirm each component’s critical contribution. Our work provides a robust and principled solution for federated learning under dynamic feature drift.
[LG-124] Prioritizing Search Space Regions in the Low Autocorrelation Binary Sequences Problem
链接: https://arxiv.org/abs/2607.09688
作者: Blaž Pšeničnik,Borko Bošković,Jan Popić,Janez Brest
类目: Machine Learning (cs.LG)
*备注:
Abstract:Low autocorrelation binary sequences problem (LABS) is a hard combinatorial optimization challenge with important applications in communications, signal processing, and satellite navigation. This paper proposes a hybrid search framework that combines Thompson sampling with parallel self-avoiding walks to adaptively allocate computational effort across restriction classes of the LABS search space. By modeling partitions as arms in a multi-armed bandit setting, the proposed method dynamically shifts search resources toward partitions that empirically produce higher merit factors while maintaining exploration of less-sampled regions. The approach is further accelerated through GPU-parallel execution, shared posterior updates, efficient neighborhood evaluation, and a Bloom filter for cycle prevention. In addition, we use a two-stage optimization strategy that first searches constrained partitioned skew-symmetric spaces and then refines the best candidates in the unrestricted space. Experiments on long binary sequences show that the proposed method improves the previously best-known results for 35 sequence lengths in the range 450 \le L \le 527 and for L=573 . In particular, we report a new longest sequence with merit factor exceeding 8.0 , obtained for L=451 . The results also show that Thompson sampling effectively prioritizes partitions with better observed performance, confirming the value of online, data-driven resource allocation in LABS optimization. Overall, the proposed framework provides a scalable and effective strategy for high-performance merit factor maximization.
[LG-125] MawForge: Memory-Bounded Expert Materialization for Local Mixture-of-Experts Inference
链接: https://arxiv.org/abs/2607.09686
作者: Craig Opie
类目: Machine Learning (cs.LG)
*备注: 7 pages
Abstract:Sparse Mixture-of-Experts (MoE) language models separate total parameter count from per-token active computation, but local inference systems often still require the full model, key-value cache, runtime buffers, and operatingsystem headroom to fit in fast memory. MawForge tests a different systems hypothesis: local MoE serving can be made practical on constrained unified-memory machines by storing the full model on disk, keeping common tensors resident, and materializing routed expert tensors into a bounded execution cache on demand. The central finding is that MawForge is effective as a bounded execution mechanism and measurement substrate for local MoE inference, but not as a cache-maximization policy. Performance depends on balancing expert reuse against resident footprint, KV-cache size, quantization, route locality, and macOS memory pressure.
[LG-126] AuditWeave: A Tamper-Evident Auditor-Navigable Evidence Layer for AI-Assisted and Data-Transformation Workflows
链接: https://arxiv.org/abs/2607.09682
作者: Vimal Nakrani
类目: Machine Learning (cs.LG); Software Engineering (cs.SE)
*备注: 8 pages, 3 figures, open-source implementation at this http URL
Abstract:AI systems are increasingly used to assist consequential decisions in regulated domains such as auditing, finance, and healthcare. This creates a recurring obligation: an organization must be able to reconstruct, after the fact, which evidence informed a given conclusion, and to show that the record of that reasoning was not altered. Existing tools address related but distinct problems - model observability, drift monitoring, governance reporting - and are built for the machine-learning engineer operating a system, not the reviewer who must trace one specific conclusion back to its supporting evidence. We present AuditWeave, a lightweight Python library, with no runtime dependencies, that records the steps of AI-assisted and data-transformation workflows into a single append-only, hash-chained ledger. A small, system-agnostic event vocabulary spans both retrieval-augmented generation (RAG) pipelines and tabular/lakehouse transformations, so a conclusion that draws on both can be traced end-to-end through one record. Within a sealed ledger, any modification, reordering, insertion, or deletion of events is detectable through chain verification. We describe the design and evaluate recording overhead, scalability, and tamper-detection correctness on the reference implementation. The integrity guarantees cost tens of microseconds per event, and, as the hash-chain construction implies, verification flagged every injected mutation across four mutation classes over 2,000 randomized trials.
[LG-127] Position: Every Ground Truth is a Human Construction not an Objective Truth
链接: https://arxiv.org/abs/2607.09668
作者: Charlotte Högberg,Ericka Johnson,Kiri L. Wagstaff
类目: Machine Learning (cs.LG)
*备注: 13 pages, 1 figure. To be published in Proceedings of the 43 rd International Conference on Machine Learning, Seoul, South Korea. PMLR 306, 2026
Abstract:Ground truth datasets play a fundamental role as reference values in the training and evaluation of machine learning models. This position paper argues that ground truths are not neutral objective measurements that are naturally given, but instead that they are constructed by arrangements of humans and technologies. We argue that the ML community will benefit from articulating and discussing these often invisible or unreported choices and acknowledging that reference data sets are contingent, not universal. Focusing on the situated and context-dependent nature of ground truths can improve reliability by enabling a better informed perspective on where, when, and how the datasets, and the models they have shaped, can best be used. We argue for increasing `situated reliability’ which includes articulating the limits and strengths of models and their truth claims. Finally, paying more attention to the construction of ground truths can support transparency, accountability, and interdisciplinary work.
[LG-128] Input-Aware Dynamic Backdoor Attack Against Quantum Neural Networks
链接: https://arxiv.org/abs/2607.11843
作者: Junrui Zhang,Zemin Chen,Lusi Li,Mohammad Ghasemigol,Daniel Takabi,Rui Ning
类目: Quantum Physics (quant-ph); Machine Learning (cs.LG)
*备注: Accepted at the 2026 IEEE International Conference on Quantum Computing and Engineering (QCE 2026)
Abstract:Quantum Neural Networks (QNNs) are a promising framework for quantum machine learning on near-term quantum devices, but their security risks remain insufficiently understood. Studies have shown that QNNs are vulnerable to backdoor attacks, yet existing quantum backdoors mostly rely on a fixed trigger shared by all poisoned inputs. This fixed-trigger design is a major weakness because many defenses detect or weaken the repeated patterns such triggers leave in data representations. Although input-aware dynamic backdoors have been studied in classical neural networks, transferring them to QNNs is difficult because quantum learning introduces new obstacles. In particular, measurement compresses the post-ansatz quantum state into a limited classical output, weakening supervision for a trigger generator, while individual density matrices fluctuate with the input and make per-sample contrastive learning unstable. To address these challenges, we propose Q-DIBA, the first input-aware dynamic backdoor attack for QNNs. Q-DIBA jointly trains a classical trigger generator and a victim QNN through a three-mode mini-batch strategy that supports clean behavior, attack activation, and trigger specificity. To provide stable quantum-level supervision, Q-DIBA introduces an ensemble density contrastive loss that operates on post-ansatz quantum states before measurement and contrasts mode-averaged density matrices rather than individual samples. Experiments on MNIST and Fashion-MNIST across multiple QNN architectures show that Q-DIBA achieves high clean accuracy, strong attack success, and high cross-trigger accuracy, demonstrating effectiveness, stealthiness, and input specificity. The attack also remains resilient against defenses including visual inspection, spectral-signature detection, and fine-tuning, suggesting that input-aware quantum backdoors are an important threat to secure QNN deployment.
[LG-129] mathttQ2SAR: overcoming classical bottlenecks in drug discovery via quantum multiple kernel learning
链接: https://arxiv.org/abs/2607.11701
作者: Mariano Caruso,Daniel Ruiz,Alejandro Giraldo,Guido Bellomo
类目: Quantum Physics (quant-ph); Machine Learning (cs.LG)
*备注:
Abstract:Quantitative Structure-Activity Relationship ( \mathttQSAR ) modeling is a foundational computational methodology in early-stage drug discovery, heavily relied upon for predicting compound toxicity, bioavailability, and therapeutic potential. However, classical methods often struggle to effectively map the highly complex, non-linear, and high-dimensional interactions inherent in molecular data, leading to reduced predictive accuracy and costly late-stage clinical failures. In this paper, we present a Quantum Multiple Kernel Learning ( \mathttQMKL ) framework, dubbed Next-Gen \mathttQ^2SAR , that leverages Quantum Support Vector Machines ( \mathttQSVMs ) to overcome these classical limitations. By encoding molecular descriptors into exponentially large quantum Hilbert spaces, our approach substantially enhances the expressiveness of non-linear modeling. Benchmarking our quantum-enhanced framework on a dataset targeting the \mathttDYRK1A kinase (a critical target for Alzheimer’s disease), the \mathttQMKL - \mathttSVM achieves an impressive Area Under the Curve ( \mathttAUC ) score of 0.8750 , significantly outperforming classical state-of-the-art Gradient Boosting models ( \mathttAUC = 0.8037 ). Furthermore, we establish a theoretical and empirical pathway toward resolving classical data bottlenecks through projected quantum kernels ( \mathttPQK ) and measurement accelerators. As quantum computing architecture matures, this framework paves the way for autonomous cognitive architectures and self-improving drug discovery pipelines, promising to unlock deeper insights across vast chemical spaces and to accelerate the development of life-saving therapeutics.
[LG-130] Diversified Multinomial Logit Contextual Bandits
链接: https://arxiv.org/abs/2607.11684
作者: Heesang Ann,Taehyun Hwang,Min-hwan Oh
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:Existing contextual multinomial logit (MNL) bandits model relevance-driven choice but ignore the potential benefits of within-assortment diversity, while submodular/combinatorial bandits encode diversity in rewards but lack structured choice probabilities. We bridge this gap with the \textitdiversified multinomial logit (DMNL) contextual bandit, which augments MNL choice probabilities with a generally submodular diversity function, thereby formalizing the relevance–diversity trade-off within a single model. Incorporating diversity renders exact MNL assortment optimization intractable. We propose a \textitwhite-box UCB-based algorithm, \textttOFU-DMNL , that constructs assortments item-wise by maximizing optimistic marginal gains, avoids black-box optimization oracles. We show that \textttOFU-DMNL achieves at least a (1-\frac1e+1) - \textitapproximate regret bound \tildeO\left(d \sqrtT/K\right) , where d is the context dimension, K the maximum assortment size, and T the horizon, and attains an improved approximation factor over standard submodular baselines. Experiments demonstrate consistent gains and, relative to exhaustive enumeration, comparable regret with substantially lower runtime. Overall, DMNL bandits provide a practical foundation for diversity-aware assortment optimization under uncertainty, and \textttOFU-DMNL offers a statistically and computationally efficient solution.
[LG-131] Imputation-free transformer learning enables robust Alzheimers disease prediction and calibrated uncertainty quantification across heterogeneous clinical cohorts
链接: https://arxiv.org/abs/2607.11656
作者: Christelle Schneuwly Diaz,Narmina Baghirova,Duy-Thanh Vu,Duy-Cat Can,Gilles Allali,Philippe Ryvlin,Oliver Y. Chén
类目: Neurons and Cognition (q-bio.NC); Machine Learning (cs.LG)
*备注:
Abstract:Accurate diagnostic classification and disease-severity prediction for Alzheimer’s disease are hampered by the incompleteness and heterogeneity of real-world clinical data. Left unaddressed, these barriers prevent reliable disease modelling and hinder effective clinical evaluation. Conventional imputation strategies introduce systematic bias, distort inter-feature relationships, and yield overconfident predictions, limitations especially consequential in diagnostic settings. Here, we propose NITROGEN, an imputation-free transformer that jointly models within-patient feature dependencies and between-patient relational structure through masked and intersample attention, enabling robust multimodal learning directly from partially observed records. We trained NITROGEN on ADNI (N=7858 scans), and evaluated it on two independent cohorts: OASIS-3 (N=2675 scans) and AIBL (N=1286 scans). Across cohorts and diagnostic and cognitive score prediction tasks, NITROGEN showed robust calibration and uncertainty quantification advantages over tree-based ensemble methods, while maintaining competitive discriminative performance. Cross-cohort and cross-method analyses identified cortical thickness in the temporal pole, age, and APOE genotype as important, though not individually sufficient, features for AD classification. We further introduced a modality-aware uncertainty adjustment that augments predictive uncertainty proportionally to the importance of absent modalities, enabling calibrated confidence when diagnostic information is unavailable. Together, our results show that imputation-free attention learning preserved meaningful discrimination under cohort shift, revealing expected degradation on more distributionally different cohorts, and demonstrate that evaluating models along calibration, interpretability, and cross-cohort reliability, not accuracy alone, is essential for clinical deployment.
[LG-132] Machine Learning-Based Reconstruction for Resistive Silicon Sensors
链接: https://arxiv.org/abs/2607.11585
作者: Alexander Aoki,Gaetano Barone,Leena Diehl,Gabriele Giacomini,Vagelis Gkougkousis,Hanshal Goyal,Rohan Kher,Daniel Li,Anna Macchiolo,Yevhenii Padnuik,Daria Senina,Samantha Sunnarborg,Jessica Tang,Alessandro Tricoli,Lixing Wang,Don C. Wong
类目: High Energy Physics - Experiment (hep-ex); Machine Learning (cs.LG); Nuclear Experiment (nucl-ex)
*备注:
Abstract:Low-Gain Avalanche Diodes (LGADs) and AC-coupled Low-Gain Avalanche Diodes (AC-LGADs) are promising technologies for precision timing and four-dimensional tracking. In AC-LGADs, the AC pad is coupled to the resistive n ^+ layer through a dielectric layer, while the gain layer remains unsegmented. This structure provides a 100% fill factor and enables good spatial resolution with a relaxed readout pitch. The same signal-sharing mechanism that makes interpolation possible complicates the readout: charge spreads across multiple pads, the useful information can approach the electronic-noise threshold, and matrix-inversion approaches can become computationally challenging and sensitive to off-diagonal noise. In this work, we study machine-learning-based reconstruction and compression for resistive silicon sensors. We use full-waveform information from correlated pads to regularise the reconstruction and extract spatial information beyond what is available from binary readouts or reduced-amplitude summaries. We first introduce recurrent neural network models based on LSTM layers, which provide a proof-of-concept implementation for full-waveform reconstruction and have been tested for FPGA deployment using \hls. We also study routes towards bandwidth reduction with waveform rasterisation and window-selection methods, and extend the approach beyond the first model to topology-agnostic transformer-based architectures that use pad coordinates as part of the input. These models are designed to support arbitrary pad counts and geometries, mitigate edge distortions, preserve approximately 10~\mu\mathrmm position resolution for 500~\mu\mathrmm\times500~\mu\mathrmm pitched sensors, and guide future resistive-silicon sensor designs
[LG-133] Climate-Invariant Conformal Prediction Intervals for Multi-Horizon Solar and Wind Forecasting
链接: https://arxiv.org/abs/2607.11470
作者: Shreedhar Gangwar(1),Abhinav Bains(1),Banalaxmi Brahma(1) ((1) B. R. Ambedkar National Institute of Technology, Jalandhar, India)
类目: Applications (stat.AP); Machine Learning (cs.LG)
*备注: 10 pages, 5 figures
Abstract:Reliable uncertainty quantification is essential for integrating solar and wind generation into modern power systems, where operators must weigh risk rather than act on point forecasts alone. Existing probabilistic methods, however, often either lack finite-sample validity or require per-site recalibration, so a single model rarely transfers across the diverse climates of a dispersed generation fleet. This paper proposes a heteroscedastic, asymmetric, group-conditional split-conformal framework built on a bootstrap-diverse XGBoost ensemble, producing prediction intervals that adapt in width to local difficulty while retaining distribution-free coverage guarantees. A single fixed specification, with no per-site or per-horizon tuning, is evaluated across four climatologically distinct sites spanning both hemispheres, at horizons of 1 to 12 hours, for both solar irradiance and wind speed. The framework holds near-nominal coverage on both targets and reduces the Interval Score by up to 35% relative to competitive baselines, with the calibration and sharpness of its intervals shown to be properties of the method rather than of site-specific tuning.
[LG-134] Inter-Stop Energy Prediction and Causal Driver Quantification for Dual-Source Trolleybuses via a Time-Aware Tabular Deep Learning Architecture
链接: https://arxiv.org/abs/2607.11349
作者: Wentao Zeng(1 and 2),Zijian Huang(3),Yiming Bie(4),Jiabin Wu(1),Jun Gong(5) ((1) School of Management, Foshan University, Foshan, China a School of Management, Foshan University, Foshan, China (2) School of Mechanical and Electrical Engineering and Automation, Foshan University, Foshan, China (3) School of Artificial Intelligence, South China Normal University, Guangzhou, China (4) School of Transportation, Jilin University, Changchun, China (5) Department of Civil Engineering, The University of Hong Kong, Hong Kong, China)
类目: Optimization and Control (math.OC); Machine Learning (cs.LG)
*备注:
Abstract:Dual-source trolleybuses alternate between overhead catenary supply and on-board battery operation, creating energy-use patterns driven by route attributes, high-frequency trajectories, and hourly weather. Existing models struggle to represent these heterogeneous inputs and rarely explain the causal drivers of consumption. This paper proposes a time-aware tabular deep learning framework for inter-stop energy management. Periodic time encoding is integrated into a parameter-efficient batch-ensemble backbone to jointly learn static and sequential features, while Bayesian optimization with tree-structured density estimation tunes hyperparameters. To move beyond prediction, a three-layer causal explanation pipeline combines feature attribution for marginal effects, a linear non-Gaussian acyclic model for causal direction discovery, and a meta-learner for net average treatment effects. Experiments on the Zurich trolleybus dataset enriched with meteorological records achieve a MAPE of 6.52% and R of 0.982, outperforming ten statistical, tree-ensemble, and deep learning baselines. Ablation results show that periodic time encoding contributes most to the accuracy gain. Causal analysis identifies regenerative braking ratio and average speed as the strongest energy-saving factors, while coasting distance is the main driver of excess consumption. The findings offer actionable thresholds for vehicle technology, driving behavior, capacity allocation, and catenary network planning.
[LG-135] Fixed-Protocol Amortized MPS Tomography with Conformalized Predictive Uncertainty
链接: https://arxiv.org/abs/2607.11273
作者: Jian Xu,Delu Zeng,John Paisley,Qibin Zhao
类目: Quantum Physics (quant-ph); Machine Learning (cs.LG)
*备注:
Abstract:Quantum state tomography is sample-starved, and the states one prepares live on a narrow, learnable manifold. A k=0 prior-only control shows that on concentrated families a prior estimate is already near-optimal, so ``high fidelity at few measurements’’ can be family memorization rather than tomography; genuine measurement-efficiency needs a model that conditions on the measurements and demonstrably uses them. On a shared matrix-product-state (MPS) core parameterization we study two routes. Approach~A learns a generative prior over MPS cores with measurement-guided posterior inference (gold-standard-validated, but whose few-measurement accuracy the control shows is largely the prior). Approach~B, our main proposal, is a \emphfixed-protocol amortized MPS estimator trained once with a gauge-invariant fidelity loss; we deliberately do not rest it on a permutation-invariant set encoder (a plain MLP matches it). The decisive lever is the measurement design: motivated by the fact that local reduced density matrices determine a \chi -MPS, conditioning on an \emphinformative local Pauli set rather than random strings turns a modest, memorization-prone estimator into a high-fidelity one ( \approx!0.95 , up to +0.59 over prior-only, decisively passing a shuffled-measurement control). A dropout ensemble, conformally recalibrated, gives \approx!90% -coverage intervals – including for observables never measured, where a shot-based interval does not exist. Quality holds as the system grows (fidelity 0.90 at n=10 , gain \emphgrowing in n ; 0.88 at bond dimension \chi=4 ), the parameterization is polynomial (native contraction to 20 qubits), and we close the loop on IBM hardware ( 5 states at 0.97 from hardware-measured Paulis).
[LG-136] Long-Memory Reservoir Computing for Data-Scarce Dengue Forecasting
链接: https://arxiv.org/abs/2607.11272
作者: Rahul Goswami,Shinjini Paul,Palash Ghosh,Tanujit Chakraborty
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:Accurate dengue forecasting is crucial for public health planning, but remains challenging because incidence series are often short, noisy, non-stationary, nonlinear, and often affected by long-range temporal dependence. Fractional differencing in Autoregressive Fractionally Integrated Moving Average (ARFIMA) helps balance non-stationarity and persistence, but its linear structure limits its ability to capture nonlinear dynamics. Deep neural networks can model nonlinear patterns, but usually require large training samples and do not explicitly encode statistical long memory. Echo State Networks (ESNs), a widely used reservoir computing framework, are attractive in this setting because they retain nonlinear recurrent dynamics while training only a simple readout, making them suitable for data-scarce scenarios. However, standard ESNs lack long-term memory from a time-series perspective. This study proposes a long-memory reservoir computing framework that integrates dedicated long-memory and short-memory ESN reservoirs with a ridge-regression readout. We introduce two variants: Fractional ESN (fESN), which incorporates fractional-differencing dynamics into the reservoir to encode long-range dependence directly, and Wavelet ESN (wESN), which extracts stable low-frequency components through wavelet smoothing before modeling them with a memory-aware reservoir. We establish theoretical guarantees for closed-loop reservoir dynamics, showing that standard ESNs induce short-memory processes under mild conditions, whereas the proposed long-memory reservoirs generate polynomially decaying dependence consistent with statistical long memory. Across multiple dengue datasets and forecasting horizons, fESN and wESN outperform statistical and deep learning baselines. Combining conformal prediction with fESN and wESN provides distribution-free calibrated uncertainty intervals.
[LG-137] When cheap gradients fail: the measurement cost of attacking quantum classifiers
链接: https://arxiv.org/abs/2607.11095
作者: Bacui Li,Chandra Thapa,Tansu Alpcan,Udaya Parampalli
类目: Quantum Physics (quant-ph); Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注: 57 pages, 15 figures
Abstract:Adversarial perturbations threaten machine learning classifiers, including variational quantum classifiers. We show that finite quantum measurement statistics (shot noise) act as a built-in defense against gradient-based test-time attacks whose cost scales unfavorably for the attacker. Because every gradient component must be inferred from repeated circuit executions under any unbiased gradient-estimation rule, white-box extraction consumes a dimension-dependent measurement budget that measurement grouping cannot remove in expressive circuits. Under stated assumptions, single-step attacks need at least quadratically many shots in the input dimension d , growing as d^5/2 under norm-concentration scaling, with a sufficient-budget analysis for iterative attacks via stochastic gradient Langevin dynamics. Simulations up to 784 input dimensions validate the law: the realized total budget is the d^5/2 geometric floor for plateau-mitigated models and grows as d^3.00 for the tested deep circuits, whose gradient norms decay with dimension absent barren-plateau mitigation; folding the measured gradient norm back in recovers the parameter-free d^3/2 shot-noise geometry. Against a matched classical baseline whose attack overhead is dimension-independent (the cheap-gradient principle of automatic differentiation), the quantum gradient cost ratio grows empirically as d^3.00 , so the attacker’s relative cost diverges as the model scales. Experiments on a 156-qubit IBM processor (ibm_boston, 4-qubit circuits, d=12 ) reproduce the effect: at matched budgets the device attack tracks the ideal within a few percent, with the high-shot gradient faithful to the exact one. The defense operates precisely when the forward map is classically hard to simulate: only then is a white-box attacker denied the simulate-and-backpropagate shortcut and must pay the measurement cost we quantify.
[LG-138] Overcoming Fourier Locking in Quantum Data Re-uploading Classifiers via Spectral Homotopy
链接: https://arxiv.org/abs/2607.11013
作者: Spencer Topel
类目: Quantum Physics (quant-ph); Machine Learning (cs.LG)
*备注: 14 pages, 5 figures, 1 table. Submitted to PRX Quantum. Code and data: this https URL . ORCID: 0009-0000-0081-9786
Abstract:Data re-uploading parameterized quantum circuits (DRU-PQCs) are universal function approximators, yet their expressivity produces oscillatory, non-convex loss landscapes that resist gradient-based optimization. We show that the primary optimization bottleneck in DRU-PQCs is not insufficient capacity but a structural failure mode we term Fourier locking (FL): because encoding weights and entangling layers are nonlinearly coupled, random initialization on high-frequency targets collapses the encoding parameters into spurious local minima. Two Fisher diagnostics characterize FL. The input-space quantum Fisher information F_x measures the effective frequency content of the encoded state; the Fisher discriminant ratio of the measured features measures their alignment with the class labels. In two independent 50-seed experiments, the locking is literal: trapped circuits hold F_x frozen for the entire run, while escaping circuits migrate their frequency content (direct training: r_pb = -0.48 ; curriculum: d = 1.34 ; both p 0.001 ). The replicated signature is this spectral mobility, not any endpoint value of F_x , and trapped circuits retain a fully non-degenerate parameter-space QFIM ( r_pb \approx 0 ): the failure is spectral misalignment of a responsive state, not a loss of geometric sensitivity. A frequency-staged homotopy protocol that paces the target frequency ( f: 1.0 \to 3.0 ) convexifies the early loss landscape; escaping circuits raise F_x in step with the curriculum, and the escape rate triples (18% vs. 6%). Fourier locking is a frequency-alignment problem, and its remedy is frequency pacing.
[LG-139] Fast Whole-Brain Geometry-Aware Functional Alignment for Cross-Subject Decoding
链接: https://arxiv.org/abs/2607.10931
作者: Pierre-Louis Barbarant,Florent Meyniel,Bertrand Thirion
类目: Neurons and Cognition (q-bio.NC); Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:Decoding brain activity is useful for characterizing brain processes and understanding the functional architecture underlying cognition. However, the inter-individual variability in brain response patterns limits the development of decoders that generalize across individuals. A solution to this challenge is functional alignment: aligning functional data across individuals before training population-level decoders. The core issue is to strike the balance between aligning functional features and preserving the anatomical structure, while maintaining computational efficiency. We introduce a new functional alignment method for fMRI, SpectralOT, that embeds cortical geometry into Laplace-Beltrami eigenmodes along functional data to regularize the alignment.
[LG-140] ransferable Implicit Solvent Machine Learning Potential for Drugs and Proteins Approaching Ab Initio Accuracy
链接: https://arxiv.org/abs/2607.10887
作者: Jan Eckwert,Julija Zavadlav
类目: Chemical Physics (physics.chem-ph); Machine Learning (cs.LG); Biomolecules (q-bio.BM)
*备注:
Abstract:Machine learning interatomic potentials (MLPs) have revolutionized atomistic modeling, offering the potential to replace traditional methods like Density Functional Theory (DFT). However, inference time of MLPs is orders of magnitude slower than that of classical force fields, hindering real-world applications for biomolecular systems that require timescales of microseconds and beyond. Implicit solvent MLPs can address this issue, but are faced with data challenges associated with coarse-grained modeling. Consequently, previous approaches relied on empirical force field data, thereby inherently limiting the MLP’s accuracy. Here, we introduce the Transferable Water Implicit Network (TWIN), an implicit water MLP parametrized entirely by an Equivariant Graph Neural Network and trained solely on ab initio and experimental labels. We demonstrate TWIN’s transferability across drug-like molecules, peptides, and proteins, achieving excellent results on ab initio and experimental crystallographic and NMR benchmarks, consistently outperforming previous machine-learning-based implicit solvent or coarse-grained models. Furthermore, TWIN closely matches DFT-based explicit solvent MLPs while providing a two-order-of-magnitude faster timestep evaluation, paving the way for efficient ab initio-level modeling of biomolecular systems in aqueous environments.
[LG-141] Edge Cluster Expansion with Radial Rotary Attention for Interatomic Potentials
链接: https://arxiv.org/abs/2607.10664
作者: Zemin Xu,Wenbo Xie,P.Hu
类目: Machine Learning (stat.ML); Materials Science (cond-mat.mtrl-sci); Machine Learning (cs.LG); Chemical Physics (physics.chem-ph)
*备注:
Abstract:In this paper, we provide a systematic investigation of SO(2) theory to machine learning interatomic potentials (MLIPs) and identify the limitations of conventional SO(2) Linear architectures relative to SO(3) Clebsch-Gordan Tensor Products (CGTP). Building on these insights, we propose direct Cartesian construction and recursive Clebsch-Gordan construction of Wigner D-matrices and introduce two novel interaction building blocks. First, we propose the Edge Complex Product Basis based on Generalized Asymmetric Contraction, a new formulation for many-body expansion that directly constructs higher-order interactions on edges through complex-valued equivariant multiplications. Second, we introduce Radial Rotary Complex Attention(RRA), which enhances extrapolation performance and surpasses existing attention vector formulations. We also introduce several improvements to the Atomic Cluster Expansion module. Building on these advances, we train our models on OMat24, sAlex, and MPTrj, and introduce TECE-OAM-RRA-1.0, which achieve state-of-the-art (SOTA) performance on the Matbench Discovery.
[LG-142] Learning Topological Quantum Phases from Limited Subsystems
链接: https://arxiv.org/abs/2607.10656
作者: Mehran Khosrojerdi,Sougato Bose,Alessandro Cuccoli,Paola Verrucchi,Abolfazl Bayat,Leonardo Banchi
类目: Quantum Physics (quant-ph); Statistical Mechanics (cond-mat.stat-mech); Strongly Correlated Electrons (cond-mat.str-el); Machine Learning (cs.LG)
*备注:
Abstract:Characterizing quantum topological phases requires measuring non-local string order parameters, demanding access to the full system, which is often experimentally unfeasible. In this work, we introduce a data-efficient supervised learning framework that circumvents this limitation by recognizing quantum phases from small subsystems. Our protocol utilizes a quantum kernel constructed from the reduced density matrices of these subsystems, which can be efficiently estimated experimentally. We benchmark our framework with the classification of the phase diagrams of two spin models on one-dimensional lattices, namely the generalized cluster-Ising spin-1/2 chain and the anisotropic Haldane spin-1 chain. Remarkably, our approach achieves high accuracy in phase classification when operations are limited to as few as one to four sites, and it also generalizes to longer chains even when trained on moderate system sizes. These findings demonstrate that local reduced density matrices preserve vital signatures of global topological phases, offering a practical route to characterize rich phase diagrams of quantum many-body systems.
[LG-143] Demixing Sparse Signals from Nonlinear Observations using Generalized Non-convex Regularization
链接: https://arxiv.org/abs/2607.10618
作者: Raziyeh Takbiri
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Signal Processing (eess.SP)
*备注:
Abstract:We consider the recovery of a pair of sparse vectors from a limited number of nonlinear observations of their superposition: y_i=g(\inner\ba_i\bPhi\bw^\ast+\bPsi\bz^\ast)+e_i , i=1,\dots,m , with m\ll n , incoherent orthonormal bases \bPhi,\bPsi , a scalar link g , and noise e_i that may be heavy-tailed or contaminated. We propose a regularization-based framework combining a Huberized data fidelity with generalized folded-concave penalties (SCAD, MCP), and a two-block proximal alternating algorithm with backtracking (NLD-PALM) whose whole iterate sequence provably converges to critical points under the Kurdyka–Łojasiewicz property, with local linear rates. On the statistical side we establish restricted strong convexity of the Huberized nonlinear loss through an exact sign-definite decomposition, and derive estimation error bounds of order \sigma\sqrts\log(n)/m that hold at \emphevery localized stationary point, an oracle rate \sigma\sqrts/m free of \log n and shrinkage bias under a beta-min condition, and a co-equal recovery theorem for \emphunknown monotone links via a linear surrogate and a clipped Plan–Vershynin decoupling. The estimator requires no knowledge of the sparsity levels, and its guarantees hold under symmetric noise with only finite variance. Experiments at n=512 under a frozen data-driven regularization rule show an earlier phase transition than convex \ell_1 demixing and greedy hard-thresholding baselines, a 35\times accuracy advantage over squared-loss estimation under 5% gross outliers, and successful demixing of spike-plus-background signals observed through a saturating amplifier.
[LG-144] Approximation of Analytic Functions by ReLU Neural Networks with Adjustable Depth and Width
链接: https://arxiv.org/abs/2607.10589
作者: Yanming Lai,Defeng Sun,Yang Wang
类目: Machine Learning (stat.ML); Information Theory (cs.IT); Machine Learning (cs.LG); Numerical Analysis (math.NA)
*备注: 47 pages, 4 figures, 1 table
Abstract:In contrast to most studies on neural network approximation theory that characterize results through a single parameter, such as the total number of network parameters, \citeshen2020deep pioneered the characterization of approximation rates as a joint function of the width parameter N and the depth parameter L , thereby granting greater architectural flexibility. Existing works using the (N,L) -characterization focus on function classes with finite smoothness s , establishing a typical approximation rate of \mathcalO\left(N^-2s/dL^-2s/d\right) with d denoting the input dimension, which indicates that network depth and width play symmetric roles for these classes. In contrast, this paper establishes upper bounds for the approximation of analytic functions, which possess infinite smoothness, via ReLU networks under the (N,L) -characterization. Specifically, we derive approximation rates of \mathcalO\left(N^-C L^\tau\right) , where C0 is some constant and \tau0 is a parameter influenced by the relation between L and N . In particular, \tau=1 if N scales roughly as L^d . Our findings reveal that depth plays a more critical role than width in the context of analytic function approximation. The main technical difficulty of obtaining such upper bounds lies in the trade-off between the smoothness parameters and the approximation accuracy. To overcome this difficulty, we employ refined constructions of several ReLU networks to approximate power functions, multivariate multiplication, and polynomials, which may be of independent interest.
[LG-145] Observation-Level Watermarking and Detection for Tabular Data
链接: https://arxiv.org/abs/2607.10554
作者: Dongyu Cui,Xuan Bi
类目: Methodology (stat.ME); Machine Learning (cs.LG)
*备注:
Abstract:With the development of generative AI, watermarking techniques have been widely used to detect the authenticity of AI-generated data and protect the rights of users and creators. While it is already well applied in data types including imaging and text data, watermarking tabular data is still under-explored. Existing methods primarily focus on numerical data, leaving discrete, categorical, and mixed data less studied. In this work, we propose STAMP (Single-observation Tabular Attribution and Marking Procedure), a novel framework for watermarking tabular data that can accommodate and preserve a wide range of distributions. We also develop a corresponding detection mechanism, which can reliably identify watermarks even when the sample size is as small as one. We establish theoretical guarantees for asymptotic consistency and detection accuracy. Finally, through extensive simulation studies and two real-data applications, we demonstrate that the proposed method is effective and robust to subsetting, while maintaining data fidelity and a high detection rate.
[LG-146] Beyond Looking Up Try Looking Around: Harmonizing Global Structure and Local Consistency in Optimal Transport for Short Text Clustering
链接: https://arxiv.org/abs/2607.10548
作者: Zhihao Yao,Yuxuan Gu,Jixuan Yin,Bo Li
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:Pseudo-labeling based on Optimal Transport (OT) has become an effective mechanism for enhancing short text clustering. Existing OT methods are short in modeling semantic consistencies between samples, which may assign different pseudo-labels to semantically similar samples. These erroneous pseudo-labels can cause the model to produce inferior clusters. This paper proposes a novel short text clustering framework, which remedies the neglect of semantic consistency in existing OT methods, generating reliable pseudo-labels to facilitate clustering. Specifically, the proposed approach first designs an instance-level attention mechanism to capture semantic relationships between samples, which are then integrated into the OT formulation to endow the transport process with neighborhood semantic awareness. By solving the proposed OT formulation, reliable pseudo-labels are obtained that simultaneously account for sample-to-sample semantic consistency and sample-to-cluster global structure information. These pseudo-labels are then used as supervisory signals to guide the model to achieve accurate clustering. Extensive experiments demonstrate that the proposed approach outperforms state-of-the-art methods. The code is available at: \hrefthis https URLthis https URL.
[LG-147] Representation Learning for Semiparametric Causal Mediation Analysis under No Essential Heterogeneity
链接: https://arxiv.org/abs/2607.10540
作者: Roberto Faleh,Sofia Morelli,Holger Brandt
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注: 30 pages, 2 figures
Abstract:We propose a two-stage estimator for structural mediation parameters that combines deep representation learning with G-estimation under the “no essential heterogeneity” (NEH) assumption. We call the method UNIT. In the first stage,TARNet estimates the heterogeneous effect of a randomized treatment on a mediator by learning a shared covariate representation across treatment this http URL resulting conditional average treatment effect (CATE) estimate provides a plug-in approximation to the heterogeneity-dependent component of the weight function entering the G-estimating equation of Zheng and Zhou (2015), which identifies the structural parameters even in the presence of unmeasured mediator-outcome confounding. We show that more accurate first-stage representation learning can yield a more informative plug-in weight and thereby improve the precision of the structural parameter estimator. In simulations with non-Gaussian covariates and nonlinear mediator effects, TARNet weights reduce the Stage-2 standard error of the mediation coefficient by a factor of 1.45 to 1.51 (median across replications, n \ge 2000 ) relative to the classical approach, at no cost to bias or coverage.
[LG-148] Integrating Background Knowledge for Scalable Causal Discovery
链接: https://arxiv.org/abs/2607.10456
作者: Mátyás Schubert,Theofanis Aslanidis,Tom Claassen,Sara Magliacane
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:Expert background knowledge is often available in practical applications of causal discovery. Such constraints on the true causal graph can help causal discovery in terms of identifiability of causal effects and accuracy of the learned structure, but also in reducing the space of candidate causal graphs. As causal discovery can become computationally expensive for large number of variables, it is crucial to utilize background knowledge effectively during the causal discovery process. However, most current methods only use background knowledge in a postprocessing step after causal discovery to refine the learned graph. In this work, we develop a framework for utilizing background knowledge during the causal discovery process, focusing especially on scalable causal discovery methods that recover only a subset of the whole graph. We implement our framework for multiple algorithms and empirically show that utilizing background knowledge can both reduce computational requirements and increase the quality of the learned structures.
[LG-149] Emergent Generalization by Representation Learning in Artificial Neural Networks
链接: https://arxiv.org/abs/2607.10430
作者: Hardik Rajpal,Dan Goodman
类目: Neurons and Cognition (q-bio.NC); Information Theory (cs.IT); Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE); Chaotic Dynamics (nlin.CD)
*备注:
Abstract:Dimensionality reduction has proven powerful for identifying neural manifolds, which are low-dimensional structures underlying high-dimensional neural activity. These low-dimensional representations have improved the interpretability of population-level coding. Yet whether such low-dimensional representations are biologically relevant and confer functional advantages in learning systems, or merely reflect neuron-level activity, remains contested in neuroscience. We show that an explicit information bottleneck forcing a recurrent neural network to learn a low-dimensional representation is necessary for rotational and out-of-distribution generalisation in a time-series prediction task. Using information-theoretic measures of causal emergence, we characterise the dynamics of this representation across the memorisation-to-generalisation transition, finding a non-monotonic trajectory which shows an initial decrease, a minimum, and a subsequent rise to a maximum, even as prediction loss falls monotonically. This trajectory scales with task complexity, and the magnitude of emergent structure reliably predicts generalisation performance. Analysis of CA1 hippocampal activity in mice learning an alternating maze task reveals analogous non-monotonic emergence dynamics that track behavioural performance. Together, these findings indicate that the ability of neural networks to learn compact, distributed and emergent representations confers a functional advantage for generalisation, supporting a causal role for learned representations in cognition.
[LG-150] SCoNet: A Two-Stage Copula CNN-LSTM for Uncertainty-Aware Spatio-Temporal Forecasting
链接: https://arxiv.org/abs/2607.10410
作者: Jongwook Kim,Jong-Min Kim
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:Reliable forecasting of several interrelated environmental variables - such as regional precipitation and temperature, or other correlated geophysical fields - across many locations calls for accurate predictions accompanied by trustworthy statements of their uncertainty. Modern deep-learning models forecast such variables accurately but usually report no uncertainty, and forcing them to output uncertainty through maximum likelihood tends to degrade their accuracy, especially when the variables are strongly correlated. Motivated by this tension, we develop TSCoNet, a two-stage convolutional-recurrent model coupled with a Gaussian copula that jointly forecasts multiple variables over space and time while quantifying predictive uncertainty. The method first learns accurate mean forecasts and then, holding the mean fixed, refines a shared representation to estimate the predictive variance, yielding calibrated prediction intervals after a standard recalibration, so that uncertainty is added without sacrificing point accuracy. We study the approach on simulated non-stationary spatial fields on the sphere and on a real dataset of monthly precipitation and temperature for fifty cities over 2000-2020. The model matches the accuracy of a strong deterministic forecaster while supplying calibrated prediction intervals that the deterministic model cannot, giving a single tool that provides both accurate point forecasts and reliable uncertainty for multivariate spatio-temporal data.
[LG-151] How much Data do We Need? Sequential Data Collection for Stochastic Programming
链接: https://arxiv.org/abs/2607.10207
作者: Xin Li,Juergen Branke,Xuan Vinh Doan
类目: Optimization and Control (math.OC); Machine Learning (cs.LG)
*备注:
Abstract:Data-driven optimization often requires collecting data to estimate uncertain model parameters before solving the underlying decision problem. In practice, however, data acquisition may incur non-negligible costs, making it critical to determine when to stop additional data collection. In this paper, we study an optimal stopping problem for sequential data collection in stochastic optimization under parameter uncertainty. We propose a benefit-driven stopping framework that balances information gain and sampling cost. We model the unknown distribution parameter within a Bayesian learning framework and update beliefs sequentially as new observations are collected. At each iteration, the decision maker evaluates the expected marginal benefit of additional data relative to the unit sampling cost and determines whether to continue sampling or stop and implement the optimization decision. Based on this framework, we develop several stopping policies. The proposed policies are evaluated through a newsvendor problem with exponentially distributed demand. Numerical experiments compare the policies with fixed-budget and hindsight benchmark strategies. The results show that benefit-driven stopping rules can substantially reduce unnecessary data collection while achieving near-optimal decision performance, demonstrating the effectiveness of adaptive stopping in data-driven optimization.
[LG-152] Are We Ready for AI-Driven Discovery? AI Verification Before the Next Fundamental Physics Breakthrough
链接: https://arxiv.org/abs/2607.10039
作者: Gaia Grosso,Vinicius Mikuni,Lukas Heinrich
类目: Data Analysis, Statistics and Probability (physics.data-an); Machine Learning (cs.LG); High Energy Physics - Phenomenology (hep-ph)
*备注:
Abstract:Machine learning (ML) has become integral to fundamental physics, accelerating statistical workflows from data acquisition through inference and hypothesis testing. As ML systems grow increasingly autonomous, ensuring their reliability for discovery claims becomes critical. This review synthesizes the VERaiPHY (Validation Evaluation for Robust AI in PHYsics) initiative’s frameworks for rigorous ML assessment across particle physics, astrophysics, and cosmology. We establish when verification is essential by contextualizing ML within the statistical discovery workflow. We emphasize fundamental limitations: inductive bias is unavoidable, sample complexity bounds learning, and experimental constraints limit discovery. We reflect on physicists’ evolving role as both experimental designers and evaluators whose judgments encode scientific rigor into AI systems. Responsible integration requires understanding ML’s transformative potential alongside its intrinsic boundaries.
[LG-153] Manifold Constrained Conformal Prediction for Spatial Events
链接: https://arxiv.org/abs/2607.10008
作者: Collin Nill,Trevor Harris,Jason Adams
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Methodology (stat.ME)
*备注: 21 pages, 10 figures, 4 tables
Abstract:We introduce a new conformal prediction method that constructs calibrated prediction sets over collections of spatial events, such as tropical cyclone genesis and earthquake locations. Forecasting natural hazards has become increasingly important, due to their significant economic impact, and quantifying the uncertainty of predictions is critical for accurate risk assessment. Our approach works by representing spatial point clouds as empirical measures so that we can score them using (sliced) Wasserstein distance, then constraining the resulting distribution-valued prediction set to be supported only near the training data manifold. We derive a coverage lower bound for the intersected sets and show that, in practice, this gap can be made small through a simple data-adaptive selection criterion. Because the resulting set is not analytically tractable, we introduce a modified flow-based sampling procedure, which allows us to represent and apply these prediction sets in practice as ensembles. Numerical experiments on synthetic data, tropical cyclone genesis, and earthquake occurrences show that our method achieves near-nominal coverage, with significantly lower energy distance and manifold distance than highest predictive density region (HDR) baselines along with generative model baselines.
[LG-154] Inverse-IMPRESSION: A Graph-based Platform for Molecular Structure Elucidation from Experimental NMR Spectroscopic Properties
链接: https://arxiv.org/abs/2607.09978
作者: Zheqi Jin,Grace Armitage,Richard Cox,Ben Honoré,Mohammad Golbabaee,Craig Butts
类目: Chemical Physics (physics.chem-ph); Machine Learning (cs.LG)
*备注: 15-page manuscript (4 figures), plus 59 pages of Supporting Information (45 figures); Submitted to the Journal of the American Chemical Society
Abstract:Here, we present a platform built on our inverted Graph Transformer Network, IMPRESSION-G2, which can accurately and rapidly reconstruct molecular bonding directly from experimental nuclear magnetic resonance (NMR) spectroscopic information. It comprises three interconnected stages: a one-shot model that predicts bond connectivity between atoms; a structure-correction stage that corrects the predicted structures by removing uncertain bonds and iteratively reassigning them; noise-augmented multi-shot prediction, generating an ensemble of candidate structures, which are ranked to identify the best-fit structure. By integrating a range of ^1 H and ^13 C NMR data, including two-dimensional (2D) experiments such as COSY, HSQC, and HMBC, the inverse-IMPRESSION platform correctly identifies the structures of 77.8% of molecules with up to 30 heavy atoms (H, C, N, O and F) using simulated NMR data, and 10 of 19 (53%) molecules using experimental NMR data. The experimental structures solved have molecular weights of up to 480 Da and are representative of the complex structures in synthetic and natural products that routinely challenge chemists. The inverse-IMPRESSION framework thus provides the first effective approach for automated molecular structure elucidation using graph-based machine learning on experimental data.
[LG-155] Artificial Intelligence Across the Cardiac Amyloidosis Diagnostic Pathway: From Single-Modality Detection to Multimodal Clinical Integration
链接: https://arxiv.org/abs/2607.09948
作者: Diana Shadibaeva,Rochak Dhakal,Kui Zhang,Xiaofeng Yang,Saurabh Malhotra,Weihua Zhou
类目: Medical Physics (physics.med-ph); Machine Learning (cs.LG)
*备注: Diana Shadibaeva and Rochak Dhakal contributed equally to this work. Total Pages = 35, No. of Figures = 4, No. of Tables on Manuscript = 1, Supplementary Tables = 6
Abstract:Cardiac amyloidosis (CA) is increasingly recognized but remains substantially underdiagnosed, because its clinical and imaging phenotype overlaps with more common cardiomyopathies. Definitive subtype assignment and management further require integration of multimodal evidence to distinguish transthyretin from light chain disease. Machine learning and deep learning have been applied across the diagnostic and management pathway. These applications span ECG, echocardiography, and health record-based case finding, as well as CMR and nuclear interpretation, including SPECT/CT biomarker quantification, prognostic modeling, and treatment response assessment. This narrative review synthesizes these studies by clinical tasks, namely screening, detection, quantification, prognosis, and treatment response monitoring, rather than by input modality. This task-based organization clarifies why apparently similar AI models require different cohorts, reference standards, evaluation metrics, and implementation thresholds. The evidence reveals a maturity gradient. Binary detection and AI assisted quantification on bone scintigraphy and SPECT/CT are closest to clinical translation. Detection is supported by large externally validated cohorts, and quantification by interpretable, outcome linked measurement of myocardial tracer burden. By contrast, subtype aware classification, prognostic risk stratification, and treatment response monitoring remain at an early stage. These tasks are limited by small cohorts, enriched retrospective designs, heterogeneous labels, incomplete external validation, and uncertain calibration in realistic prevalence settings. Across tasks, high discrimination alone is insufficient.
[LG-156] Depth-Efficient Quantum Topological Data Analysis for Regime-Specific Detection of Financial Stress
链接: https://arxiv.org/abs/2607.09906
作者: Arul Rhik Mazumder,Shreyan Ronit Mazumder
类目: Quantum Physics (quant-ph); Machine Learning (cs.LG); Statistical Finance (q-fin.ST)
*备注: 12 pages, 6 figures, 5 tables, Accepted to IEEE International Conference of Quantum Computing and Engineering - QCE 2026 in the Quantum Applications (QAPP) Technical Papers track
Abstract:We present, to our knowledge, the first adaptation of Pauli Correlation Encoding (PCE) to quantum topological data analysis, reformulating Betti number estimation as a depth-efficient variational optimization over a compressed qubit register. From a Takens embedding and Vietoris–Rips filtration of SP~500 returns, we extract combinatorial Laplacians and recast null-space counting as a continuous-PCE Rayleigh-quotient minimization with variational deflation, encoding n_k simplex indices into O(n_k^1/\kappa) qubits with shallow, ancilla-free circuits. Because the resulting loss is rational rather than bilinear in the correlators, the barren-plateau bound of~\citeSciorilli25 does not transfer; empirically the gradient variance decays only polynomially, with no exponential barren plateau, over n=4 – 12 qubits. The classical stage matches ripser~\citebauer2021ripser on all 190 sliding windows (2007-2009). On the real market Laplacians ( \beta_1=1 – 22 ), warm-starting from a classical null-space surrogate allows PCE-VQE to recover \beta_1 exactly at every scale, placing the obstacle in the optimisation landscape rather than the encoding. Chronologically split classification gives in-regime ROC AUC 0.818 , but out-of-distribution evaluation on the 2020 COVID shock and 2022 rate cycle (AUC 0.009 , 0.515 ) shows the calibration does not generalize across crisis regimes.
[LG-157] When Classical Baselines Are Tuned as Carefully as the Quantum Model Does Quantum Reservoir Computing Still Win?
链接: https://arxiv.org/abs/2607.09905
作者: Tushar Pandey
类目: Quantum Physics (quant-ph); Machine Learning (cs.LG)
*备注: 9 pages, 2 figures
Abstract:Can a small quantum computer forecast a changing signal better than an ordinary classical method? Many studies say yes, but the classical methods they compare against are often left in a basic, untuned state while the quantum model is carefully optimised. We ask what happens when the classical competitor is given exactly the same care: the same size and the same amount of tuning effort. We study two popular reasons a quantum reservoir is thought to help, using exact simulations of small quantum systems (up to eleven qubits) on prediction tasks. In both cases the quantum advantage disappears once the comparison is fair. In the first, extra quantum measurements add nothing that a simple classical formula of the same size does not already provide. In the second, a feedback loop genuinely helps the quantum model, turning a useless setup into a working predictor, yet a well-tuned classical network still predicts slightly more accurately, and the gap is statistically reliable. Our point is not that quantum reservoirs can never win, but that two of their commonly cited advantages do not hold up against fair classical competitors at this scale. We provide these matched comparisons as a simple, reusable checklist for honest benchmarking. All results are fully reproducible from fixed random seeds.
[LG-158] An End-to-End Hybrid Quantum–Classical Sampling Workflow for Discrete Markov Random Fields: A Reproducible Case Study
链接: https://arxiv.org/abs/2607.09893
作者: Arul Rhik Mazumder
类目: Quantum Physics (quant-ph); Machine Learning (cs.LG)
*备注: 9 pages, 7 figures, 9 tables, Accepted to IEEE International Conference of Quantum Computing and Engineering - QCE 2026 in the Quantum End-to-End Hybrid Case Studies (QECS) Technical Papers track
Abstract:Sampling from discrete Markov random fields (MRFs) is a hard problem. We study amplitude-encoded i.i.d. sampling for small MRFs where 2^n target probabilities are precomputed classically. This removes quantum exponential speedup but allows a clean comparison against classical MCMC based on independent circuit samples ( \tau \approx 1 ). Across 60 instances spanning five graph families (1k-step burn-in, 3k retained samples), the mean ESS ratios of Quantum to Single-Site Gibbs, Block Gibbs, Tuned-Block, and Parallel Tempering are 16.35 , 7.29 , 1.82 , and 1.79 , showing modern classical samplers substantially close this gap. Amortizing O(2^n) preprocessing into wall-clock time, exact inverse-CDF sampling yields 17.7\textM ESS/s versus 488\textK ESS/s for the quantum sampler ( 36\times mean rate, 153\times per-instance), confirming no wall-clock advantage. We characterize MCMC autocorrelation costs and benchmark amplitude-encoded state preparation at n \in \8,10,12\ . An MPS scaling study ( n \le 40 ) shows bond dimension \chi=32 achieves F=0.721\pm0.059 at n=40 . Finally, a matched-budget VQC vs. MPS comparison at n \in \8,10,12\ shows VQC fidelities fall far below MPS: (F_\mathrmVQC, F_\mathrmMPS) = (0.31, 0.99), (0.21, 0.96), (0.17, 0.88) at compressions 10.7\times , 34.1\times , and 113.8\times .
[LG-159] Q-Score: A Quantum-Native Scoring Function for Molecular Docking
链接: https://arxiv.org/abs/2607.09737
作者: Kangyu Zheng,Yidong Zhou,Ruihao Li,Zixin Ding,Zhiding Liang,Shaohua Li
类目: Chemical Physics (physics.chem-ph); Machine Learning (cs.LG)
*备注:
Abstract:Molecular docking predicts how a small molecule binds to a protein and is a key bottleneck in drug discovery. Classical scoring functions sum empirical pairwise contacts, blind to quantum-mechanical effects like orbital charge transfer that govern binding specificity. We introduce Q-Score, encoding GNN-predicted orbital donor-acceptor energies into a weighted graph and scoring binding by solving a maximum-weight vertex clique problem via Digitized-Counterdiabatic QAOA. Each interaction anchor maps to one qubit and compatibility constraints become edges. Across 11 protein targets, DC-QAOA recovers the exact optimum on 8 at 10 qubits. On 1000 AI-generated molecules, Q-Score is orthogonal to classical scoring with Spearman rho of 0.05, driven by orbital quality with rho of 0.90, and free of molecular-weight bias, enriching for strong orbital interactions at twice the random rate. DC-QAOA achieves a mean approximation ratio of 0.94 with 52 percent exact. Execution of 1000 circuits on IBM Eagle confirms 6-qubit solvability on NISQ hardware.
附件下载
