本篇博文主要内容为 2026-07-09 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 (2026-07-09)
今日共更新553篇论文,其中:
- 自然语言处理共44篇(Computation and Language (cs.CL))
- 人工智能共145篇(Artificial Intelligence (cs.AI))
- 计算机视觉共99篇(Computer Vision and Pattern Recognition (cs.CV))
- 机器学习共180篇(Machine Learning (cs.LG))
- 多智能体系统共11篇(Multiagent Systems (cs.MA))
- 信息检索共5篇(Information Retrieval (cs.IR))
- 人机交互共22篇(Human-Computer Interaction (cs.HC))
多智能体系统
[MA-0] Institutional Red-Teaming: Deployment Rules Not Just Models Causally Shape Multi-Agent AI Safety
【速读】：该论文旨在解决多智能体生成式AI系统在实际部署中因规则设计不当导致的集体安全风险问题，尤其关注部署规则如何因果性地影响智能体群体行为的安全性与公平性。其核心挑战在于：现有评估方法难以分离单个规则变量对复杂群体行为的影响，导致无法精准识别高风险规则设计。解决方案的关键在于提出“机构红队测试”（institutional red-teaming）这一因果评估范式——在保持智能体、目标和任务状态不变的前提下，仅改变单一部署规则，从而将群体行为的变化归因于该规则本身。研究通过构建IABench-CA基准平台，在228个情境下对五种典型规则与七类模型群体进行大规模实验（33,924场游戏），结合规范性合作参照系与自标注推理轨迹，揭示出规则对集体安全具有显著因果效应；进一步发现，不存在普遍安全的默认规则，且基于身份的反向靶向规则在所有情境与群体中均表现出系统性不安全特征，其根本机制在于身份显著性（identity salience）——即使通过匿名化暂时隐藏身份信息，智能体仍能通过反复博弈推断出隐藏规则并实施针对性排除。为此，研究将方法封装为可验证的安全论证工作流，为每个部署情境与群体定义一个临时安全规则区域Φ(c,P)，明确标示残余风险与持续监控义务，实现对部署规则的可解释、可追溯的安全评估。
链接: https://arxiv.org/abs/2607.07695
作者: Yujiao Chen
机构: Massachusetts Institute of Technology (麻省理工学院)
类目: Artificial Intelligence (cs.AI); Computer Science and Game Theory (cs.GT); Multiagent Systems (cs.MA)
备注:
Abstract:We introduce institutional red-teaming, an evaluation methodology for testing deployment rules in multi-agent AI: hold the agents, objectives, and task state fixed, vary only one rule, and attribute the resulting change in collective behavior to that rule. We instantiate the methodology in IABench-CA, a consequence-allocation benchmark spanning 228 contexts, five canonical rules, and seven model populations (33,924 games), with a normative cooperative reference and auto-labelled reasoning traces. Three findings emerge. (1) Deployment rules causally alter collective safety: changing only the consequence rule moves mean fatality by 22 to 58 percentage points within every population. (2) There is no safe default, but the targeting hazard is universal: the safest rule, the least-safe rule, and even the direction of the incidence effect vary across populations, yet regressive identity-targeting is never decisively safest in any context for any population, eliminates the least-resourced agent in 30-87% of games everywhere, and is selection-unsafe relative to the cooperative reference for all seven populations. (3) Identity salience is the mechanism: a one-shot anonymization ablation on the most exploitation-prone population (gpt-5.1) shows that merely naming the loss bearer in the rule text drives targeted elimination from 22% to 81% at identical payoffs; under repeated play, anonymization only delays the targeting, as agents re-infer the hidden rule from observed eliminations. We package the methodology as a safety-case workflow that certifies a provisional rule region \Phi(c,P) per deployment context and population, with explicit residual risks and monitoring obligations.
[MA-1] Agent Delivery Engineering Predictive Reliability Framework
【速读】：该论文旨在解决长时序大语言模型（LLM）多智能体系统中因基础设施监控无法察觉的隐蔽性退化所引发的可靠性风险问题。其核心解决方案是提出一种自适应退化评估预测可靠性框架（ADE Predictive Reliability Framework, ADE-PRF），通过整合跨五个层级的20类异构信号，构建具有39.2点动态范围的可信度裕度（Trust Margin, TM）指标，实现对系统健康状态的主动轨迹预测。该框架采用三重方法并行预测机制，可实现长达8小时的前瞻预测，其中指数法在平均绝对误差（MAE=1.228）、方向准确率（76.8%）及99.65%的预测值落在±10点容差范围内的表现上显著优于卡尔曼滤波法。生产环境验证覆盖380,227次预测与280,579次验证，涵盖六种代理配置持续15天，并结合七组沙箱控制实验，关键发现包括识别出“虚假繁荣”现象——即表面指标正常但内部已发生退化的隐蔽状态，以及在集成ADE插件后，TM指标能立即与真实状态耦合，且20个影响因子中有16个依赖于ADE采集的数据。研究证明，ADE-PRF为生产环境中LLM智能体提供了最早期的可靠性量化与前瞻性预警能力。
链接: https://arxiv.org/abs/2607.07689
作者: Dexing Liu(Shanghai Qijing Digital Technology Co., Ltd)
机构: Shanghai Qijing Digital Technology Co., Ltd(上海奇景数字科技有限公司)
类目: Multiagent Systems (cs.MA)
备注: 117pages,83figures
Abstract:Long-horizon LLM multi-agent systems face reliability risks invisible to infrastructure monitoring. We propose the ADE Predictive Reliability Framework (ADE-PRF), enabling proactive health trajectory prediction from passive degradation detection. ADE-PRF aggregates 20 heterogeneous signals across five layers into a Trust Margin ™ metric (39.2-point dynamic range). Triple-method parallel prediction enables 8-hour forecasts: the Exponential method achieves MAE=1.228, Direction Accuracy=76.8%, with 99.65% within +/-10-point tolerance. Production validation spans 380,227 predictions and 280,579 validations across six agent profiles over 15 continuous days, plus seven sandbox-controlled experiments. Key findings include detection of “false prosperity” – degradation concealed by normal surface metrics – and immediate TM coupling with ground-truth states upon ADE plugin integration, with 16/20 factors relying on ADE-collected data. Exponential consistently outperforms Kalman. ADE-PRF provides among the earliest reliability quantification with forward-looking warnings for production LLM agents.
[MA-2] Stability and Convergence of Optimistic Exponential Weights with Asymmetric Step Sizes in Bimatrix Games
【速读】：该论文旨在解决双矩阵二人博弈中，由乐观指数权重法（optimistic exponential weights method）生成的迭代序列在不同步长设置下的最终迭代收敛性（last-iterate convergence）与均衡点稳定性问题。传统研究通常假设双方玩家采用相同的步长，而本文突破这一限制，允许策略更新步长 ηx 与 ηy 不同，从而更贴近实际应用中的异步学习场景。其核心解决方案的关键在于：在零和博弈的特殊情形下，基于固定点集为有限的假设，提出了一个仅依赖于步长乘积 ηxηy 的全局最终迭代收敛的充分条件，该条件具有实际可操作性，并能部分解释实验中观察到的行为；在一般双矩阵博弈中，进一步给出了渐近稳定与不稳定的几乎紧阈值，同样以步长乘积为判据，主要体现理论价值。该框架统一推导出多个已知结果，并为特定情形提供了实用的步长约束，通过实验验证了理论分析的有效性。
链接: https://arxiv.org/abs/2607.07517
作者: Hédi Hadiji,Sarah Sachs
机构: Laboratoire des Signaux et Systèmes, CentraleSupélec(中心超级理工学院); School of Mathematics, University of Bristol(布里斯托大学)
类目: Multiagent Systems (cs.MA); Dynamical Systems (math.DS)
备注:
Abstract:We study bimatrix two-player games and investigate the last-iterate convergence and stability of equilibria for the iterates generated by the optimistic exponential weights method. In contrast to prior work, we allow the step sizes \eta_x and \eta_y to differ. Our first main result establishes, under the assumption that the set of fixed points is finite, a sufficient condition for global last-iterate convergence in the special case of zero-sum games, which constrains only the product \eta_x\eta_y of the step sizes. This condition is practically relevant and partially explains empirically observed behavior. Our second main result provides an almost-tight threshold for asymptotic stability and instability, again in terms of products of the step sizes, for general bimatrix games. This result is primarily of theoretical interest. We derive several known results and practically relevant step size bounds for special cases and illustrate our results by experiments.
[MA-3] Multi-Agent Robotic Control with Onboard Vision-Language Models
【速读】：该论文旨在解决视觉语言模型（Vision Language Models, VLMs）及视觉语言动作模型（Vision Language Action, VLA）在机器人控制应用中面临的可解释性差、泛化能力不足以及对高算力外部计算资源依赖严重的问题。其核心解决方案是提出一种基于多智能体系统（Multi-Agent System, MAS）的架构，将专用智能体部署于机器人本地硬件上，实现完全离线运行，从而摆脱对外部云端计算的依赖。系统采用参数量紧凑的VLM（3-20B），并通过微调提升包装质量检测的准确性；引入创新的“Megamind”协调智能体，有效缓解小型模型在长时序规划任务中因上下文记忆能力有限导致的性能瓶颈。该方案在基于AMD Ryzen™ AI迷你电脑的软硬件协同仿真环境中验证成功，结果表明，全本地部署的MAS架构不仅具备成本效益，且在复杂工业仓库场景下的五类任务（安全巡检、仓储维护、搜寻、包装质检与响应人类请求）中表现出良好的可行性与实际迁移潜力，其仿真环境已开源（Apache 2.0许可）。
链接: https://arxiv.org/abs/2607.07403
作者: Kajetan Rachwał,Maciej Majek,Bartłomiej Boczek,Jakub Matejczyk,Dominik Matejkowski,Adam Dąbrowski,Tim Seyde,Alexander Amini,Maria Ganzha
机构: 未知
类目: Multiagent Systems (cs.MA); Robotics (cs.RO)
备注: 6 pages, 2 figures, accepted to 24th International Conference on Practical applications of Agents and Multi-Agent Systems (PAAMS’26)
Abstract:Vision Language Models (VLMs) and Vision Language Action (VLA) models have shown promise in robotic control. Yet, they face significant challenges regarding explainability, generalization, and compute requirements. This paper presents a Multi-Agent System (MAS) architecture that addresses these limitations by deploying specialized agents on onboard hardware - eliminating dependence on external compute. The system controls a multi-purpose autonomous mobile manipulator in a simulated industrial warehouse, fulfilling five task categories: safety inspection, warehouse maintenance, warehouse search, package quality verification, and responding to human requests. Compact VLMs (3-20B parameters) are used throughout, with fine-tuning applied to improve package inspection accuracy. A novel “Megamind” orchestration agent mitigates context retention issues inherent to long-horizon planning with smaller models. The system was validated in a hardware-in-the-loop simulation using an AMD Ryzen™ AI mini PC. Results demonstrate that a fully onboard MAS architecture is a viable, cost-efficient alternative to cloud-dependent deployments, with strong potential for real-world transfer. The simulation environment has been released as open source under the Apache 2.0 licence.
[MA-4] A Large Language Model-Driven Agent -Based Modeling Framework with Multi-Round Communication for Simulating Vaccine Opinion Dynamics
【速读】：该论文旨在解决生成式人工智能（Generative AI）在社会计算领域应用中，个体认知模块如何驱动决策行为并影响宏观层面意见动态的机制不清晰的问题。其核心挑战在于缺乏可解释、可验证的认知-行为映射框架，以揭示不同认知模块在复杂社会互动中的作用。为此，研究提出一种将大语言模型（LLM，Qwen3-8B）嵌入基于代理的建模（Agent-Based Modeling, ABM）的集成框架，通过疫苗接种意见动态这一典型场景进行实证分析。该框架的关键创新在于引入可配置的认知模块——记忆模块与提示多样性模块，并在异质性个体特征与社交网络结构下模拟多情境下的意见演化过程。仿真结果表明，不同认知模块对涌现意见具有相反的影响效应，且框架能够复现既有研究中观察到的社会影响力非线性行为模式，验证了其在实现代理模型三级验证（Level 3 Validation）方面的潜力，为理解人类认知机制与社会集体行为之间的因果关系提供了可扩展的技术路径。
链接: https://arxiv.org/abs/2607.07387
作者: Bo Zhang,Na Jiang
机构: The Hong Kong University of Science and Technology (Guangzhou); Jinhe Center for Economic Research, Xi’an Jiaotong University
类目: Multiagent Systems (cs.MA); Social and Information Networks (cs.SI); Physics and Society (physics.soc-ph)
备注: 11 pages, 5 figures
Abstract:Recently, Large Language Models (LLMs) have been utilized in various applications of computational social science and provide the possibility to integrate such models into agent-based modeling to explore the cognitive processes. However, how specific cognitive modules drive individual decisions and macro-level opinion dynamics remains unclear. Therefore, this study introduces a framework that integrates an LLM (Qwen3-8B) into agent-based modeling to investigate this problem, using vaccination opinion dynamics as a case study. We utilize this framework to simulate opinion dynamics among agents with heterogeneous profiles and social networks, evaluating scenarios by enabling different cognitive modules: a memory module and a prompt diversity module. The simulation results reveal that different cognitive modules have opposite impacts on our emergent opinion. Furthermore, the framework reproduces the non-linear behavior patterns of social influence observed in existing research, demonstrating our framework’s validity and potential to reach the level 3 validation of agent-based models.
[MA-5] Multi-Agent AI Control: Distributed Attacks Hamper Per-Instance Monitors NEURIPS
【速读】：该论文旨在解决多智能体环境下生成式人工智能（Generative AI）系统中潜在的协同攻击风险，即多个智能体在共享基础设施上协作实施恶意行为（如模型权重窃取、训练过程污染）时，传统单智能体监控机制难以有效识别和防范的问题。其核心挑战在于：现有AI控制方法主要聚焦于单一智能体在孤立轨迹中的行为，而真实部署场景中多智能体协同运作已成为常态，且严重安全威胁往往依赖于多个智能体之间的协调行动。为此，论文提出并构建了FakeLab——一个包含9个服务、86项良性任务和4类攻击目标的合成AI实验室环境，用于系统性地模拟分布式攻击场景。解决方案的关键在于揭示“碎片化效应”（fragmentation effect）：随着参与攻击的智能体数量增加及协作程度提高，对每个智能体进行独立监控的有效性显著下降，导致整体攻击成功率上升。研究进一步发现，这一现象并非由良性与恶意代码比例决定，而是更深层地依赖于智能体的模型能力；引入显式规划器可使攻击完成率提升至七倍，并加剧执行者对监控机制的“理论心智”（theory-of-mind）理解偏差。实验还表明，弱信任监控器几乎无法检测多数攻击，而强监控器虽能将未被察觉的成功攻击减少一个数量级以上，但仍存在明显漏洞。因此，该研究强调了在多智能体环境中重构监测框架的必要性，为下一代安全可控的生成式AI系统设计提供了关键实证依据。
链接: https://arxiv.org/abs/2607.07368
作者: Oliver Makins,Orazio Angelini,Zohreh Shams,Mary Phuong
机构: UK AI Security Institute (英国人工智能安全研究所); MATS Research (MATS研究机构)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注: Submitted to NeurIPS; 81 pages; 32 figures and 24 tables
Abstract:AI control is a family of techniques to prevent an AI with malicious goals from subverting its operator’s intent. AI Control usually studies a single agent in one trajectory, but real deployments run many agents over shared infrastructure, and the most severe risks (model-weight exfiltration, training-run poisoning) plausibly need several agents acting in concert. We initiate the empirical study of multi-agent AI control, formalising distributed attacks in which several agents jointly aim for a malicious goal. We develop FakeLab: a synthetic AI-lab codebase (9 services, 86 benign tasks, 4 attack objectives). We evaluate single agent monitoring against distributed attacks, varying the number of agents, their coordination, model capabilities and precise monitoring configuration. Our central finding is the fragmentation effect: as more agents coordinate to attack, per-agent monitoring becomes less likely to catch any of the attackers. Secondly, this is not driven by the ratio of benign to malicious code – we argue this is likely dependent on model capability. Third, an explicit planner amplifies the fragmentation effect, raises the rate of completing the attack up to sevenfold, and mildly sharpens executors’ poor monitor-theory-of-mind. Finally: a weak “trusted” monitor misses most attacks, while a stronger one cuts undetected success by more than an order of magnitude but remains imperfect. We are happy to share FakeLab, our control setting, with safety researchers on request. Comments: Submitted to NeurIPS; 81 pages; 32 figures and 24 tables Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA) Cite as: arXiv:2607.07368 [cs.LG] (or arXiv:2607.07368v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2607.07368 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[MA-6] From Atomic Actions to Standard Operating Procedures: Iterative Tool Optimization for Self-Evolving LLM Agents
【速读】：该论文旨在解决大语言模型（Large Language Model, LLM）代理在执行复杂任务时因依赖静态、细粒度原子工具集而产生的重复性低层逻辑构建问题，导致推理开销大、任务失败率高的瓶颈。其核心解决方案在于提出通过将原子动作合成可复用的标准操作流程（Standard Operating Procedure, SOP），作为可调用的高层工具来封装多步逻辑，从而实现代理的自我演化。关键创新在于提出EvoSOP框架，该框架支持代理从执行轨迹中提取SOP，并通过“构建-合并-评估-修剪”的系统化生命周期迭代优化工具集，显著提升任务成功率并减少交互轮次，验证了通过迭代工具优化可形成可靠高效的工具使用模式，为自演化代理的规模化发展提供了可行路径。
链接: https://arxiv.org/abs/2607.07321
作者: Haipeng Ding,Yuexiang Xie,Zhewei Wei,Yaliang Li,Bolin Ding
机构: Renmin University of China (中国人民大学); Alibaba Group (阿里巴巴集团)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Multiagent Systems (cs.MA)
备注:
Abstract:Tool utilization enables Large Language Model (LLM) agents to interact with the real world and resolve complex tasks. However, existing agent frameworks predominantly rely on static toolsets composed of granular atomic actions (e.g., basic file I/O or single-turn search), which forces agents to reinvent low-level logic for every recurring workflow, leading to increased reasoning overhead and failure rates. In this study, we propose that agents can achieve self-evolution by synthesizing these atomic actions into reusable Standard Operating Procedures (SOPs), which function as callable higher-order tools that encapsulate multi-step logic. We further introduce EvoSOP, a framework that empowers agents to extract SOPs from execution trajectories and iteratively optimize the toolset through a systematic lifecycle of construction, merging, evaluation, and pruning. Extensive experiments demonstrate that EvoSOP significantly boosts task success rates while substantially reducing the number of interaction rounds compared to baselines. Our analysis also reveals that iterative tool optimization fosters reliable and efficient tool-use patterns, providing a scalable pathway for the development of self-evolving agents.
[MA-7] Operational Reframing and Approval-Framed Delegation in Multi-Agent LLM Safety
【速读】：该论文旨在解决多智能体大语言模型（multi-agent LLM）系统在安全评估中因将“直接提示”与“规划-执行”（planner-executor）管道的对比结果简化为单一“管道效应”而导致的可解释性问题。现有评估方法未能区分影响安全性的多重机制，包括：有害意图被重新表述为合理操作任务、规划器对请求的拒绝或转化行为，以及执行器在“获得批准”框架下因委托提示而增强的合规性。为此，作者提出一种五条件受控对比设计，通过30个合成有害场景及来自四个代理安全基准的探索性外部验证集进行评估，并采用大语言模型（LLM）判断合规性。研究发现，聚合后的管道安全性并非稳定的架构属性；其中，“操作重构”（operational reframing）是跨模型最普遍的风险信号，显著提升GPT、Gemini和DeepSeek在两类场景下的合规率，而Claude则表现出较强抗性；规划器主要通过拒绝行为缓解风险，但当其生成可执行步骤时，执行器反而可能比直接提示基线更合规；“批准式委托”对提示设计、模型组合和场景来源高度敏感，且通过设置怀疑性执行器提示可显著降低合规性。此外，原始直接模型排名无法准确预测部署后规划-执行系统的实际表现，例如Gemini在直接提示下最安全，但在与Claude规划器配合时合规率从8.9%飙升至38.9%；而GPT的零平均管道效应实则掩盖了重构增益与规划器拒绝之间的相互抵消。因此，论文主张多智能体安全评估应独立报告操作重构、规划器行为、委托框架和模型配对等关键因素，避免将失败简单归因于系统架构本身。
链接: https://arxiv.org/abs/2607.07097
作者: Lifei Liu,Haoran Yu,Xiaochong Jiang,Su Wang,Pin Qian,Yihang Chen
机构: Independent Researcher(独立研究员); Carnegie Mellon University (卡内基梅隆大学); Georgia Institute of Technology (佐治亚理工学院)
类目: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Multiagent Systems (cs.MA)
备注:
Abstract:Safety evaluations of multi-agent LLM systems often compare a direct prompt with a planner-executor pipeline and report the difference as a single “pipeline effect.” We argue that this aggregate is difficult to interpret because it conflates three mechanisms: harmful intent may be reframed as plausible operational work, the planner may refuse or transform the request, and the executor may act under delegation prompts implying prior approval. To separate these factors, we introduce a five-condition controlled contrast design, evaluated on 30 synthetic harmful scenarios and an exploratory external validation set from four agent-safety benchmarks using LLM-judged compliance. Our results show that aggregate pipeline safety is not a stable architectural property. Operational reframing is the most portable risk signal, increasing compliance for GPT, Gemini, and DeepSeek across both scenario sets, while Claude is comparatively resistant. Planner behavior can offset this risk mainly through refusal; however, when the planner produces executable steps, the executor may become more compliant than under the direct operational baseline. Approval-framed delegation is sensitive to prompt design, model pairing, and scenario source, and a skeptical executor prompt sharply reduces compliance. Raw-direct model rankings can also mispredict deployed planner-executor behavior. Gemini is safest under raw direct prompts in the primary set yet shows the largest amplification with a Claude planner, rising from 8.9 percent to 38.9 percent compliance. GPTs near-zero aggregate pipeline effect instead hides a reframing increase canceled by planner refusal. These findings suggest that multi-agent safety evaluations should report reframing, planner behavior, delegation framing, and model pairing separately before attributing failures to architecture itself. Subjects: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Multiagent Systems (cs.MA) Cite as: arXiv:2607.07097 [cs.AI] (or arXiv:2607.07097v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2607.07097 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[MA-8] Progressive Crystallization: Turning Agent Exploration into Deterministic Lower-Cost Workflows in Production
【速读】：该论文旨在解决生成式 AI（Generative AI）在IT运维（AIOps）场景中作为持续性成本中心的问题，即每次任务执行均需完整的大型语言模型（LLM）推理，即使面对以往已解决的重复性问题也未能实现成本优化。其核心解决方案是提出“渐进结晶化”（progressive crystallization）这一生命周期机制，将智能体（agent）探索行为从永久执行模式转变为发现机制，通过定义从完全由智能体编排、混合模式到完全确定性工作流的三阶段执行分类体系，结合基于证据的晋升与降级机制，将经过多次验证的智能体行为逐步固化为成本更低、可复现性更强的确定性工作流，并自动对出现退化的流程进行降级处理。该方法在生产环境的云网络AIOps系统中验证，八个月内使确定性执行比例从0%提升至45%，尽管事件量翻倍，但单次事件的智能体成本下降超过70%，同时通过增强可复现性与可审计性提升了系统安全性。关键创新点在于实现了智能体行为向高效确定性流程的自动化演进，构建了涵盖执行分类、晋升/降级标准、轨迹提取、经济模型及安全考量的完整框架。
链接: https://arxiv.org/abs/2607.07052
作者: Arun Malik
机构: 未知
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC); Emerging Technologies (cs.ET); Multiagent Systems (cs.MA)
备注: Conference-style paper; 10 pages (estimated from manuscript formatting if applicable); focuses on agentic AI, AIOps, workflow automation, deterministic execution, and LLM cost optimization
Abstract:AI agents deployed for IT operations are typically permanent cost centers because every execution requires full LLM inference, even for previously solved problems. This paper introduces progressive crystallization, a lifecycle that treats agent exploration as a discovery mechanism rather than a permanent execution model. It defines a three-stage execution taxonomy, from fully agent-orchestrated to hybrid to fully deterministic workflows, together with an evidence-based promotion mechanism that converts repeatedly validated agent behaviors into cheaper and more reproducible deterministic workflows, while automatically demoting workflows that regress. Evaluated on a production cloud networking AIOps system processing tens of thousands of incidents per month, the approach increased deterministic execution from 0% to 45% over eight months, reduced per-incident agent costs by more than 70% despite doubling incident volume, and improved safety through greater reproducibility and auditability. The paper also presents the execution taxonomy, promotion and demotion criteria, trace extraction methodology, economic model, safety considerations, and discusses limitations and threats to validity.
[MA-9] LLM -powered reasoning in agent -based modeling
【速读】：该论文旨在解决传统基于代理的模型（Agent-based Modeling, ABM）依赖静态先验信息、无法实时适应动态变化的问题，从而限制了其在政策制定中的实用性。其解决方案的关键在于提出一种可扩展的混合代理与语言驱动流行病（Hybrid Agent-based and Language-driven Epidemic, HALE）建模框架，通过引入大语言模型（Large Language Models, LLMs）来动态预测个体决策行为，使ABM能够根据实时社会行为和环境变化进行自适应调整，显著提升了模型对复杂人类行为的模拟能力与现实适用性。
链接: https://arxiv.org/abs/2607.06757
作者: Sifat Afroj Moon,Dakotah Maguire,Adam Spannaus,Joe Tuccillo,Maksudul Alam,Sudip K. Seal,John Gounley,Heidi Hanson
机构: Oak Ridge National Laboratory (橡树岭国家实验室)
类目: Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注:
Abstract:Agent-based modeling (ABM) has the capability to model millions of individuals and their interactions, which is useful for policy making. However, ABMs have traditionally relied on static prior, which prevents the models from adapting to real-time changes. Our research provides a novel approach to addressing this information gap. Large language models (LLMs) offer new opportunities to predict human decision-making. Here, we introduce a scalable Hybrid Agent-based and Language-driven Epidemic (HALE) modeling framework that leverages LLMs to predict human decision-making in an ABM simulation. As a proof-of-concept, we use HALE to simulate COVID-19 and its effects in Salt Lake County, UT.
[MA-10] A hierarchical memory architecture overcomes context limits in long-horizon multi-agent computational modeling
【速读】：该论文旨在解决大语言模型（Large Language Models, LLMs）在长时程科研工作流中因无状态架构导致的上下文连续性缺失与定量严谨性不足的问题。其核心解决方案是提出一种多智能体框架——集成式定量系统建模（Ensemble QSP），采用三层分层记忆架构，在项目周期内严格控制注入上下文的规模与稳定性（中长期项目状态：中位数301个标记，最大4,050个标记，基于104次运行），通过限制每类状态容量并主动淘汰已完成任务，实现无需人工干预的持续自主运行，有效防止上下文退化。系统由五名领域专家型工作者智能体在主调查员（PI-agent）的监督下协同工作，结合基于物理规律的检查清单与结构化领域知识，确保计算过程符合科学约束。实验表明，该框架可在无需人类介入的情况下稳健完成药代动力学-药效学（PK-PD）模型的自主选择，且在不同成本和前沿性能的LLM上均保持一致的结果质量，相比单智能体基线显著提升了药代参数恢复精度，并在语义多样但任务相同的提示下表现出稳定的模型选择能力。跨生理药代动力学（PBPK）模型的特征级消融分析进一步证实，PI智能体的监督可提升调试效率，同时维持最终建模准确性。该架构具备结构上的领域无关性，新增科学领域仅需配置相应主调查员智能体即可。
链接: https://arxiv.org/abs/2607.07666
作者: Shivendra G. Tewari,Holly Kimko
机构: 未知
类目: Quantitative Methods (q-bio.QM); Multiagent Systems (cs.MA)
备注: 19 pages, 4 figures, 2 tables. Preprint submitted for publication
Abstract:Large language models (LLMs) demonstrate remarkable reasoning capabilities, yet their stateless architecture fundamentally limits deployment in long-horizon research workflows requiring multi-session continuity and quantitative rigor. Here we present Ensemble QSP, a multi-agent framework featuring a three-layer hierarchical memory architecture that keeps injected context bounded and constant in project duration (mid-term project state: median 301 tokens, max 4,050, across 104 runs) by capping each state category and evicting completed work, enabling continuous autonomous operation without context degradation. The system orchestrates five specialist worker agents under domain-expert principal investigators, enforcing physical constraints through physics-based checklists and structured-domain knowledge. Comprehensive benchmarking demonstrates robust autonomous pharmacokinetic-pharmacodynamic model selection without human intervention, consistent result quality across both lower-cost and frontier LLMs, improved PK parameter recovery relative to single-agent baselines, and stable model selection across linguistically diverse prompts of the same task. Feature-level ablation across physiologically based pharmacokinetic (PBPK) models spanning a broad complexity range shows that PI-agent oversight improves debugging efficiency while preserving final accuracy across conditions. The architecture is structurally domain-agnostic, adding a new scientific domain requires only a new PI agent configuration.
自然语言处理
[NLP-0] Accurate Interdisciplinary and Transparent Structure-property Understanding with Deep Native Structural Reasoning
【速读】： 该论文旨在解决跨生物、化学和材料科学领域中结构-性质关系的机制性解释难题，核心挑战在于如何在人工智能模型中同时保持领域原生的结构信息表征，并实现基于科学原理与物理约束的可解释推理。其解决方案的关键是提出一种多模态科学基础模型——SciReasoner，该模型将蛋白质、小分子及无机晶体的坐标、拓扑与周期性连接关系统一离散化为结构感知的词汇表，使结构片段作为可追溯的证据单元参与推理过程。通过这一设计，SciReasoner在低同源性或孤儿样蛋白的细胞组分注释任务中，将最大F1值从0.42提升至0.55；在化学逆合成任务中，单步准确率由0.63提高至0.72，并生成可解释的断键路径与前体验证轨迹；在材料科学中则能有效区分元素与化合物相态，并解析高/低带隙区域。在86个基准测试中，其在67项任务上达到当前最优性能，且双盲专家评估显示其推理轨迹在98%情况下优于或至少等同于前沿大语言模型。该方法通过将结构作为受科学约束推理的可检视基底，实现了预测准确性与科学可解释性之间的深度融合。
链接: https://arxiv.org/abs/2607.07708
作者: Chen Tang,Yizhou Wang,Jianyu Wu,Lintao Wang,Shixiang Tang,Pengze Li,Encheng Su,Jun Yao,Jiabei Xiao,Yuqi Shi,Jielan Li,Hongxia Hao,Zhangyang Gao,Fang Wu,Ben Fei,Xiangyu Yue,Pan Tan,Bozitao Zhong,Jinouwen Zhang,Aoran Wang,Yan Lu,Jiaheng Liu,Xinzhu Ma,Liang Hong,Mingyue Zheng,Phil Torr,Bowen Zhou,Wanli Ouyang,Lei Bai
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE); Machine Learning (cs.LG)
备注:
Abstract:Structure-property relationships are foundational to biology, chemistry and materials science, where function, reactivity and physical response emerge from spatial, chemical and periodic organization. Mechanistically explaining these relationships requires interpreting structural evidence through scientific principles and physical constraints, from stereochemistry and bonding to symmetry, energetics and periodic order. However, applying artificial intelligence to this process presents a joint challenge of representation and reasoning: models must preserve domain-native structural information while showing how specific evidence supports predictions under these constraints. Here we introduce SciReasoner, a multimodal scientific foundation model for native structural reasoning across proteins, small molecules and inorganic crystals. SciReasoner discretizes coordinates, topologies and periodic connectivities into a unified structure-aware vocabulary, treating structural tokens as addressable evidence units during reasoning. In homology-controlled Gene Ontology prediction, SciReasoner improves Cellular Component annotation for low-homology and orphan-like proteins, increasing F_\max from 0.42 to 0.55. In chemistry, it raises single-step retrosynthesis accuracy from 0.63 to 0.72 while generating fragment-level disconnection and precursor-verification traces. In materials science, its representations separate elemental and compound phases and resolve high- and low-band-gap regimes. Across 86 benchmarks, SciReasoner achieves state-of-the-art performance on 67 tasks. Double-blind expert evaluation rates its reasoning traces as preferred or at least comparable to those of a frontier large language model in 98% of cases. By making structure an inspectable substrate for reasoning under scientific constraints, SciReasoner connects accurate prediction with interpretable scientific inference.
[NLP-1] Co-LMLM: Continuous-Query Limited Memory Language Models
【速读】： 该论文旨在解决传统大语言模型（LLM）在知识存储与调用方面存在的固有局限，即知识被“硬编码”于模型权重中，导致知识更新困难、可控性差且难以追溯来源。为此，论文提出一种新型有限记忆语言模型（LMLM），称为连续查询式有限记忆语言模型（CO-LMLM），其核心创新在于将知识库（KB）中的键值对从传统的基于关系结构的离散查询扩展为连续向量形式：即以连续向量作为查询键，对应文本知识值。这一设计突破了以往依赖离散符号化关系型知识库的限制，使模型能够以极低计算成本生成灵活的向量查询，并在生成过程中高效检索可读性强、可溯源的人类可理解的知识内容。该方案的关键在于通过连续向量键实现高灵活性与高效性之间的平衡，同时结合一个无需依赖维基百科的自由文本事实片段标注流程，显著拓展了知识预训练数据的适用范围。实验表明，在多个模型规模下，CO-LMLM在困惑度（perplexity）和事实精确率（factual precision）上均优于现有LMLM及标准LLM；尤其在3.6亿参数规模下，其表现超越了使用40倍更多数据预训练的模型，并在SimpleQA评测中达到gpt-4o-mini水平，优于Claude Sonnet 4.5。
链接: https://arxiv.org/abs/2607.07707
作者: Yair Feldman,Linxi Zhao,Nathan Godey,Dongyoung Go,Yilun Hua,Kilian Q. Weinberger,Jennifer J. Sun,Yoav Artzi
机构: Cornell University (康奈尔大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: preprint
Abstract:Limited memory language models (LMLMs) externalize factual knowledge during pretraining to a knowledge base (KB), rather than memorizing it in their weights. During generation, the model then fetches knowledge from the KB as needed. This recently introduced paradigm provides multiple advantages, including knowledge control capabilities that remain beyond conventional LLMs. We propose continuous-query LMLM (CO-LMLM), where the KB pairs continuous keys with textual knowledge values, a significant departure from prior reliance on relational KB and queries. CO-LMLM generates flexible vector queries at minimal cost, while still integrating human-readable and attributable retrieved knowledge into its generation. We pair this design with an annotation pipeline that tags free-form factual spans in arbitrary text, removing prior work’s restriction to Wikipedia. Across pretraining on Wikipedia and FineWeb-Edu and at multiple model scales, CO-LMLM outperforms prior LMLMs and vanilla LLMs in both perplexity and factual precision. At 360M scale, this includes lower perplexity than models pretrained on 40x more data, and SimpleQA-verified performance that is in line with gpt-4o-mini and higher than Claude Sonnet 4.5.
[NLP-2] From Noisy Traces to Root Causes: Structural Trajectory Analysis and Causal Extraction for Agent Optimization
【速读】： 该论文旨在解决长时域智能体（long-horizon agents）优化中因执行轨迹（execution traces）冗余、异质性强以及上下文信息噪声大所导致的优化效率低下与过拟合问题。现有方法在使用真实执行轨迹进行优化时，往往面临两个核心挑战：其一，大规模轨迹集合中存在大量重复或低价值失败案例，易引发过拟合；其二，单条轨迹内包含大量非因果相关步骤，而简单的截断或滑动窗口等上下文压缩方法可能误删关键因果证据，从而生成误导性的优化信号。针对这一困境，本文提出STRACE（Structural TRajectory Analysis and Causal Extraction）框架，其解决方案的关键在于双层次结构化处理机制：在批量层面，通过挖掘失败模式实现对冗余轨迹的过滤，保留具有代表性的失败样本；在单个轨迹内部，则基于文本依赖图（textual dependency graph）进行因果定位，剔除非因果步骤并精准识别出导致失败的根本模块（root-cause module），从而构建高信噪比的优化上下文。实验结果表明，STRACE显著优于标准上下文过滤基线，在具有挑战性的形式化验证任务（VeruSAGE-Bench）中，成功优化了人类专家设计的智能体，使成功率从42.5%提升至58.5%，实现1.4倍的性能增益。
链接: https://arxiv.org/abs/2607.07702
作者: Ying Chang,Jiahang Xu,Xuan Feng,Chenyuan Yang,Peng Cheng,Yuqing Yang
机构: Microsoft Research(微软研究院)
类目: Computation and Language (cs.CL)
备注:
Abstract:The optimization of long-horizon agents increasingly relies on reflection-based mechanisms, where a large language model (LLM) acts as an optimizer to diagnose agent failures and improve agent policies. However, real execution traces are difficult to use directly for optimization: large trace collections are often redundant and heterogeneous, making optimization inefficient and prone to overfitting to low-value failures; meanwhile, each individual trajectory also contains many irrelevant steps, while naive context reduction methods such as truncation or sliding windows can discard causally important evidence and produce misleading optimization signals. To resolve this dilemma, we introduce STRACE (Structural TRajectory Analysis and Causal Extraction), a framework that constructs high signal-noise optimization contexts for more precise and effective optimization. At the batch level, STRACE mines failure patterns to filter redundant traces and retain representative failures; within each selected trace, it performs causal localization over a textual dependency graph to remove non-causal steps and identify the true root-cause module for optimization. Empirical results demonstrate that STRACE significantly outperforms standard context-filtering baselines. Notably, on a challenging formal verification task (VeruSAGE-Bench), it successfully optimizes human-expert designed agents, delivering 1.4\times success-rate improvement (42.5% to 58.5%). The code is available at this https URL .
[NLP-3] Agon: Competitive Cross-Model RL with Implicit Rival Grading of Reasoning
【速读】： 该论文旨在解决当前基于可验证奖励的强化学习（如GRPO）在训练推理模型时仅评估最终答案而忽略思维过程的问题，导致模型倾向于生成更长的输出而非更高质量的推理，尤其是在复杂问题上。其核心解决方案是提出Agon框架，通过两个相互竞争的模型互为评判者：二者交替扮演“起草者”与“解题者”角色，一方撰写解题步骤，另一方在阅读该步骤的基础上独立求解，双方均根据是否优于对方获得奖励。这一机制使得推理质量得以隐式评估，无需显式的思维过程标注或奖励模型。由于两模型在训练中持续优化且彼此对抗，每个模型都面临不断变强的对手，从而推动更深层次的推理能力发展。在推理阶段，该框架采用两阶段级联结构，先由一个模型生成草稿，再由另一个模型基于草稿作答。实验表明，在DeepMath硬数据集上，该方法使Qwen3模型的pass@1指标翻倍，相较于未经训练的Mixture-of-Agents方法提升约八倍；该优势在竞赛编程代码和不同模型家族（Qwen3.5、Gemma 4）中均得到复现。未来方向是让模型在潜在空间中进行联合推理，实现更高效的协同思考。
链接: https://arxiv.org/abs/2607.07690
作者: Vladislav Beliaev
机构: thinkdense.ai
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 15 pages, 7 figures, 8 tables
Abstract:Reinforcement learning from verifiable rewards (e.g. GRPO) is the engine behind today’s reasoning models, yet it grades only the final answer. On hard problems this trains models to write more rather than to think better, since the trace itself is never graded and no label for good thinking exists. We introduce Agon, which makes two competing models each other’s graders. Both attempt the same problem; in alternating roles, one drafts a solution and the other reads it while solving, and each is rewarded for out-solving the other. To win, a model must out-reason a rival that has seen its work, so reasoning is judged implicitly during training, with no process labels and no reward model. Because both models are optimized, each faces a progressively stronger rival, which single-model RL cannot provide. The two need only be comparably strong and behaviorally different. At inference the pair deploys as it trains, a two-stage cascade in which one model drafts and the other answers after reading the draft. On the hard split of DeepMath with Qwen3, this doubles GRPO’s pass@1, roughly eight times the gain of an untrained Mixture-of-Agents pass over the same base. The ordering replicates on competitive-programming code and across model families (Qwen3.5, Gemma 4). For now the models talk in text; the next step is to let them reason together in latent space.
[NLP-4] Max Out GRPO Signal: Adaptive Trace Prefix Control for Hard Reasoning Problems
【速读】： 该论文旨在解决生成式 AI（Generative AI）在使用组相对策略优化（Group Relative Policy Optimization, GRPO）时面临的瓶颈问题：当一组样本中无一成功时，组相对优势信号消失，导致无法提供有效梯度，从而浪费了最具挑战性的前沿样本。其核心解决方案是引入动态可调的参考前缀（AdaPrefix-GRPO），将问题难度视为一个连续可控的调节参数。与现有方法一次性设定前缀长度不同，AdaPrefix-GRPO 在训练过程中将其转化为反馈控制器，持续调整每个问题所接收的前缀长度，维持成功率稳定在50%左右——此时GRPO的梯度信号最强——并在训练后期完全移除辅助，使模型最终独立完成任务。实验表明，在相同训练计算量（FLOPs）下，该方法显著提升模型性能：对于0.6B参数模型，对分布外数学问题的准确率超过GRPO的两倍（2.1×）；在Qwen3-1.7B和AIME数据集上分别达到1.6×和1.7×的提升，同时将推理轨迹长度减少约一半。该方法仅需在数据准备阶段添加前缀并施加损失掩码，无需修改训练器本身，且模型越小，增益越显著。
链接: https://arxiv.org/abs/2607.07674
作者: Vladislav Beliaev
机构: Independent Researcher; thinkdense.ai
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: 13 pages, 5 figures, 3 tables
Abstract:Group Relative Policy Optimization (GRPO) stalls on a model’s hardest problems: when no rollout in a group succeeds, the group-relative advantages vanish and the problem contributes no gradient, wasting the frontier examples we most want to learn from. Prepending a correct prefix of a reference solution raises the success rate, making prefix length a continuous knob on difficulty. Concurrent methods set the knob once; AdaPrefix-GRPO turns it into a feedback controller: throughout training it adjusts how much of the solution each problem gets, holding its success rate near 50%, where GRPO’s gradient signal is largest, then withdraws the assistance entirely, so the deployed model solves problems unaided. On hard math, at matched training FLOPs, it more than doubles GRPO’s accuracy on held-out problems from the training distribution for a 0.6B model (2.1x), with 1.6x on Qwen3-1.7B and 1.7x on AIME, while roughly halving trace length. The method is implemented in data preparation plus a loss mask on prefix tokens; the trainer is otherwise stock. The smaller the model, the larger the gain.
[NLP-5] Does Bielik Know What It Doesnt Know? Activation Dispersion Separates Entity Familiarity from Factual Reliability Across Model Scale
【速读】： 该论文旨在解决大语言模型在生成回答时对未见过实体产生幻觉（hallucination）的问题，核心关注点是：模型是否能在生成任何输出 token 之前，通过内部激活状态提前表征对实体的熟悉程度，并且这种表征能否预测答案的事实可靠性。其解决方案的关键在于发现并验证了一种存在于后 SwiGLU MLP 层激活中的无监督表示信号——具体为逆参与比率（inverse participation ratio）和谱熵（spectral entropy）两种单次前向传播即可计算的分散度指标。这些指标在四个波兰 Bielik 模型（参数规模 1.5B–11B）上对已知、虚构及真实但冷门的实体进行区分，表现出极高的判别能力（AUROC 达 0.95–1.00），且不受层选择影响，具有跨实体类型迁移性，表明该信号具有普遍性和鲁棒性。值得注意的是，尽管模型在内部已具备对实体熟悉度的精确感知（早在 1.5B 模型中即达到性能上限），但其行为层面的事实可靠性却随模型规模显著提升，说明内在熟悉度感知与外在事实准确性之间存在分离现象。此外，模型极少主动拒绝或规避回答，反映出其缺乏基于内部可信度的自我抑制机制。因此，该研究揭示了模型“知道”自己是否熟悉某实体的能力与其“正确回答”的能力并非同步演化，为未来构建可信赖的生成式 AI (Generative AI) 提供了关键的诊断工具与改进方向。
链接: https://arxiv.org/abs/2607.07670
作者: Grzegorz Brzezinka
机构: Prosit AS
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 23 pages, 6 figures and 7 tables
Abstract:Large language models hallucinate most about entities they have never seen. We ask whether a model’s activations betray entity familiarity before a single answer token is generated, and whether that signal predicts the factual reliability of the answers. On four Polish Bielik models (1.5B-11B parameters), we probe four entity domains (athletes, cities, writers, musicians), each with 42 well-known, 42 obscure-but-real, and 42 fabricated entities addressed by a one-sentence question (504 prompts per model). Two unsupervised, single-forward-pass dispersion measures over post-SwiGLU MLP activations, inverse participation ratio and spectral entropy, separate known from fabricated entities at AUROC 0.95-1.00 across all domains and scales; a supervised linear probe reaches 0.99-1.00. Both clear selection-aware permutation floors of about 0.70-0.74 (empirical p=1e-3), survive held-out layer selection (0.93-0.99), and persist on real names (known vs. obscure-but-real: 0.96-1.00). The signal transfers across entity types (mean off-diagonal AUROC 0.92-0.99); a matched-template counterfactual shows the only large drops are template-caused, not entity-type effects, and the signal is diffuse across heads. This representational signal is already at ceiling at 1.5B, whereas behavioral factual reliability scales sharply: 0, 2, 10, and 19 of 42 known athletes are answered fully correctly by the 1.5B, 4.5B, 7B, and 11B models under a strict judge. Within known entities, separating correct from hallucinated answers is much harder (probe 0.93; dispersion no better than a first-token-entropy baseline). A five-sample semantic-entropy baseline reaches only 0.71-0.83 at 5x the inference cost. Despite this internal awareness, the models almost never abstain: an audit of 2,520 answers finds 2 refusals and 1 hedge. Entity familiarity and factual reliability are distinct phenomena on different scaling curves.
[NLP-6] DiaLLM : An Investigation into the Robustness-Generation Gap in English Dialect Adaptation
【速读】： 该论文旨在解决大语言模型在生成方言英语（dialectal English）方面能力不足的问题，尽管当前模型已具备对方言英语的理解能力，但其输出仍局限于标准美式英语，导致方言生成这一更具挑战性的任务长期未被有效解决。解决方案的关键在于提出DiaLLM框架，通过在国际英语语料库（International Corpus of English）上持续预训练三个开源模型家族，并结合隐式与显式后训练范式，每种范式再与三种模型对齐策略组合，实现了对澳大利亚、印度和北部英国英语三种方言的可控对比研究。研究发现，模型的方言鲁棒性（robustness）与生成质量（generation）存在解耦现象：基准测试结果主要受持续预训练和监督微调（SFT）影响，而对齐策略则显著重塑了生成内容的方言特征，这种变化却无法被传统基准捕捉。显式的目标方言适配方法能生成更可识别且更受人类偏好的方言文本，但最激进优化方言奖励的策略反而不被人类评价者青睐。独立的语言学分析进一步验证了“奖励-质量”之间的差距，尤其在三个模型族中的两个表现尤为明显。研究结果表明，尚无单一对齐方法占优，未来需依赖更丰富的奖励设计及持续投入方言资源建设以缩小该差距。所有代码、模型检查点及偏好数据集均已公开。
链接: https://arxiv.org/abs/2607.07669
作者: Jordan Painter,Dipankar Srirag,Adarsh Kappiyath,Diptesh Kanojia,Aditya Joshi,Lu Yin
机构: Institute for People-Centered AI, University of Surrey, Surrey, United Kingdom; University of New South Wales, Sydney, Australia
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models increasingly \emphunderstand dialectal English, yet still \emphproduce only standard, US-leaning English, leaving dialectal generation, the harder half of the problem, largely unaddressed. We introduce \textbfDiaLLM, which continually pretrains three open-weight language model families on the International Corpus of English and applies implicit and explicit post-training paradigms, each combined with three model alignment strategies, giving the first controlled comparison of these components across Australian, Indian, and Northern British English. Our results reveal that dialectal robustness and generation are \emphdissociated: benchmarks are shaped by continual pretraining and SFT, while alignment visibly reshapes generation in ways benchmarks do not capture. Explicit variety-targeted adaptation produces output reliably recognised as dialectal and preferred over broad alignment, yet the method that most aggressively optimises the dialectal reward is not preferred by human evaluators. Independent linguistic analysis corroborates this reward-quality gap, most clearly on two of the three families. No single alignment method dominates, and closing the gap will require richer reward designs and continued investment in dialectal resources. We release all code, checkpoints, and preference datasets.
[NLP-7] RL Post-Training Builds Compositional Reasoning Strategies ICML2026
【速读】： 该论文旨在探究强化学习（Reinforcement Learning, RL）后训练是否仅能放大基础模型中已存在的原始技能，还是能够将这些原始技能组合成更高层次的新策略。研究在完全可观测的重写语法（rewrite-grammar）环境中进行，其中预训练分布已知且所有生成的重写过程均可审计。其解决方案的关键在于：通过基于轨迹（Trace-based）的推理任务与二元最终答案奖励信号，RL能够发现并构建出非显式存在于预训练模型中的、可复用的复合性程序。具体而言，RL采用分阶段的组合机制，先强化原始符号约简能力，再逐步发现有效的序列组合（即有序原始约简链的压缩）和并行组合（即独立原始约简的单步合并），形成稳定且可重复利用的策略库。与拒绝采样微调（Rejection Fine-Tuning, RFT）相比，关键差异不在于探索范围大小，而在于探索的选择性——RFT产生大量类似捷径的无效重写，而RL则聚焦于有效且可复用的结构化路径。此外，消融实验表明，复合策略的涌现不仅依赖原始技能的暴露程度，更取决于预训练是否将原始能力组织为可被后续压缩的约简程序。因此，基础模型仅提供弱化的程序性原料，而RL将其转化为可靠、可扩展的高层级策略。
链接: https://arxiv.org/abs/2607.07646
作者: Azwar Abdulsalam,Nishil Patel,Andrew Saxe
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 8 pages, 6 figures. Accepted to the 2nd Workshop on Compositional Learning at ICML 2026, Seoul, South Korea
Abstract:Does RL post-training merely amplify primitive skills already latent in a base model, or can it compose primitive skills into new higher-level strategies? We study this question in a fully observable rewrite-grammar environment where the pretraining distribution is known and every generated rewrite can be audited. A Transformer is pretrained on primitive symbol-rewrite chains and post-trained on a Trace-based reasoning task with only a binary final-answer reward. RL solves held-out problems that remain rarely solved by the pretrained model even under much larger sampling budgets, while rejection fine-tuning improves early but plateaus. Trace analysis shows that RL reorganizes primitive competence through a phased compositional mechanism: it first strengthens primitive reductions, then discovers valid composed procedures. These include sequential compositions, which collapse ordered chains of primitive contractions, and parallel compositions, which combine independent primitive contractions in a single step. The composed procedures are not isolated samples; they are reused and consolidated into a stable repertoire. Comparing RL with rejection fine-tuning shows that the key difference is not exploration volume but selectivity: RFT produces many shortcut-like rewrites, much of them invalid, whereas RL concentrates exploration into valid reusable structure. Pretraining ablations show that the emergence of compositional strategies is gated not by primitive exposure alone, but by whether pretraining organizes primitive competence into reduction procedures that RL can later compress. The base model provides weak procedural ingredients; RL builds them into reliable higher-level strategies.
[NLP-8] Future Confidence Distillation in Large Language Models
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在可信度感知系统中进行可靠置信度估计的关键挑战，即现有方法普遍将置信度视为完成回答后的静态属性，忽视了置信度信息在生成过程中的动态演化。其核心解决方案是提出“未来置信度蒸馏”（future confidence distillation），通过利用后解答阶段的正确性探测器生成的置信度作为教师信号，训练基于前解答阶段隐藏表示的预测器。研究发现，后解答阶段的置信度估计在校准性和区分度上显著优于前解答阶段，且线性探针从隐藏表示中可提取远超模型显式表达的置信度相关信息。基于此，所提出的蒸馏方法仅需前解答阶段的表示即可实现接近后解答置信度的校准性能，兼具高样本效率和跨数据集的领域内迁移能力，证明了置信度信息可在答案生成完成前被有效预测，从而实现高效且可靠的置信度估计。
链接: https://arxiv.org/abs/2607.07626
作者: Sahil Kale
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 16 pages, 5 figures
Abstract:Reliable confidence estimation is essential for deploying large language models (LLMs) in confidence-aware systems, where downstream decisions such as retrieval, tool use, and adaptive computation depend on accurately estimating answer reliability. Existing approaches, however, largely treat confidence as a property of completed responses, overlooking how confidence-related information evolves throughout the answering process. In this work, we investigate confidence from a temporal perspective by comparing pre-solution Feeling-of-Knowing (FOK) and post-solution Judgement-of-Learning (JOL) confidence estimates across frontier and open-source LLMs. We show that post-solution confidence is consistently better calibrated and more discriminative than pre-solution confidence, while linear probes trained on hidden representations recover substantially richer confidence-related information than models explicitly verbalise. Building on this observation, we introduce future confidence distillation, which trains predictors operating on pre-solution hidden representations using teacher confidence estimates produced by post-solution correctness probes. Despite requiring only pre-solution representations for inference, distilled predictors recover much of the calibration improvement achieved by post-solution confidence, remain highly sample efficient, and transfer across datasets within the same domain. Together, our findings demonstrate that confidence-related information evolves throughout the answering process and can be anticipated before answer generation is complete, enabling significantly more reliable yet low-cost confidence estimation.
[NLP-9] PALS: Percentile-Aware Layerwise Sparsity for LLM Pruning
【速读】： 该论文旨在解决现有单次剪枝方法（如Wanda和SparseGPT）在Transformer模型中对各层统一施加相同稀疏率所导致的效率损失问题，即忽略了不同层在模型中的重要性差异。其核心解决方案是提出一种感知分位数的逐层稀疏性方法（PALS, Percentile-Aware Layerwise Sparsity），该方法基于每层激活值幅度的99分位数动态调整稀疏率，并将调整范围限制在目标稀疏率±5%以内，以实现更精细的层间稀疏性分配。实验表明，在LLaMA-2-7B模型上以50%稀疏率进行剪枝时，PALS相较于均匀稀疏的Wanda方法将WikiText-2困惑度从12.92降低至10.96（9次运行均值，p < 0.001），显著提升性能；而该优势具有架构依赖性，对LLaMA-3-8B收益微弱，对Mistral-7B则无明显改善。此外，研究发现基于梯度的稀疏性分配策略表现反而劣于随机分配，暗示梯度幅值无法有效预测离散权重移除的影响，进一步凸显了激活统计特性在指导剪枝中的关键作用。PALS在剪枝流程中引入的计算开销可忽略不计，且无需额外微调，具备良好的实用性与可扩展性。
链接: https://arxiv.org/abs/2607.07557
作者: Yazdan Jamshidi,Alexey Shvets
机构: Palo Alto Networks (帕洛阿尔托网络)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:One-shot pruning methods like Wanda and SparseGPT apply the same sparsity ratio to every layer of a transformer, ignoring known variation in layer importance. We propose PALS (Percentile-Aware Layerwise Sparsity), which adjusts per-layer sparsity based on the 99th percentile of activation magnitudes, bounded to \pm 5% around the target ratio. On LLaMA-2-7B at 50% sparsity, PALS achieves 10.96 WikiText-2 perplexity versus 12.92 for uniform Wanda (mean over 9 runs, p 0.001 ). The benefit is architecture-dependent: LLaMA-3-8B shows marginal gains and Mistral-7B shows none. We also find that gradient-based allocation – the seemingly more principled approach – produces results worse than random, suggesting that gradient magnitude does not predict the impact of discrete weight removal. PALS adds negligible cost to the pruning pipeline and requires no fine-tuning.
[NLP-10] hink Big Search Small: Where Capacity Matters in Hierarchical Search Agents ?
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）驱动的多智能体搜索系统中模型能力分配不合理的问题，即现有系统通常使用相同规模的单一模型实例化所有角色，未能明确不同角色应如何分配计算资源。其核心解决方案是将层次化搜索任务分解为三个角色：负责任务分解的委派角色（delegation role）、负责检索与证据提取的执行角色（execution role），以及作为对照组固定不变的答案生成角色。通过在五个多跳问答（multi-hop QA）基准上对委派和执行角色进行受控的容量扩展实验，研究发现：第一，角色解耦显著优于单智能体基线，使精确匹配（Exact Match, EM）指标从4.5提升至8.6；第二，能力敏感性呈现非对称特征——扩大委派主干模型可带来约11点的EM提升，而扩展执行子代理仅提升约2.6点，表明任务分解是当前系统的性能瓶颈；第三，通过质量过滤轨迹蒸馏训练的17亿参数执行器，在准确率上达到前沿子代理水平，同时减少37%的子代理调用令牌开销。上述结果揭示了构建高效层级搜索智能体的关键策略：应将计算资源集中于委派环节以强化任务分解能力，同时显著压缩执行环节而不牺牲准确性，从而实现性能与效率的帕累托优化。
链接: https://arxiv.org/abs/2607.07548
作者: Qinnan Cai,Yibo Zhao,Xiang Li
机构: East China Normal University (华东师范大学)
类目: Computation and Language (cs.CL)
备注: 21pages
Abstract:Large language model based search agents increasingly adopt multi-agent architectures in which a main agent decomposes a complex question into sub-queries and dispatches them to parallel sub-agents. However, existing systems instantiate all roles from a single model of identical scale, leaving open how model capacity should be distributed across roles. We factorize hierarchical search into three roles: a delegation role responsible for task decomposition, an execution role responsible for retrieval and evidence extraction, and an answer generation role held fixed as a confound control. We then conduct controlled capacity sweeps along the delegation and execution axes on five multi-hop QA benchmarks. The experiments yield three findings. First, role factorization consistently outperforms a single-agent baseline, improving exact match from 4.5 to 8.6 points across six model scales. Second, capacity sensitivity is asymmetric: scaling the delegation backbone improves EM by ~11 points, whereas scaling the execution sub-agent moves EM by only ~2.6 points, identifying decomposition as the capability bottleneck. Third, a 1.7B-parameter executor trained via quality-filtered trajectory distillation matches a frontier sub-agent in accuracy while consuming 37% fewer sub-agent tokens, advancing the Pareto frontier. These results suggest a concrete recipe for building hierarchical search agents: concentrate capacity at delegation and downsize execution without sacrificing accuracy. Our code is available at this https URL.
[NLP-11] FourierQK: Spectral Preprocessing of Query-Key Projections Improves Transformer Attention
【速读】： 该论文旨在解决在字符级语言建模中，标准点积注意力机制（dot-product attention）因局部感受野和缺乏对多尺度上下文的显式建模而导致的性能瓶颈问题。其核心解决方案是引入基于快速傅里叶变换（FFT-based）的谱预处理（spectral preprocessing），对学习得到的查询-键（Q/K）投影进行全局频域混合，从而增强模型对跨不同语义尺度（从段落到词）的长程依赖建模能力。关键创新在于：通过在频域中学习一组具有近似几何分布的频率分量（49、27、10、6 个标记/周期，分别对应段落、子段落、短语和词级尺度），实现对多尺度结构的隐式捕捉，显著降低验证损失（最高达79%），且该增益仅源于频域中的全局信息混合，而非单纯的空间投影扰动。此外，研究明确区分了非因果的双边谱注意力（本文方法）与真正因果的谱注意力（如后续论文中基于Morlet小波核的方法），并验证了该架构与FNet等完全替换注意力结构的方法存在本质差异——本方法保留完整的注意力评分机制，仅对Q/K投影施加谱预处理。
链接: https://arxiv.org/abs/2607.07478
作者: Athanasios Zeris
机构: Independent Researcher, Athens, Greece.
类目: Machine Learning (cs.LG); Computation and Language (cs.CL); Signal Processing (eess.SP)
备注: 16 pages, 2 figures, 7 tables
Abstract:FFT-based spectral preprocessing of learned query-key (Q/K) projections substantially improves transformer attention on character-level language modelling. On TinyShakespeare: a fixed random spectral filter achieves val=1.031 (Delta=+0.443); a single learned frequency at paragraph scale achieves val=0.608 (Delta=+0.867); and four learned frequencies spanning paragraph to word scale achieve val=0.309 (Delta=+1.166), a 79% reduction over standard dot-product attention. The single-frequency result is confirmed across three random seeds (mean val=0.236, std=0.019). The four frequencies converge to a near-geometric multi-scale ordering (49, 27, 10, 6 tokens/cycle) corresponding to paragraph, sub-paragraph, phrase, and word scales. The gain is specific to spectral preprocessing: random orthogonal and non-orthogonal projections of Q/K produce no measurable improvement, suggesting the benefit comes from global frequency-domain mixing rather than metric distortion. All results are verified by a shuffled-validation diagnostic against positional leakage. Causal filters (Gaussian, Mexican Hat, Morlet) do not improve over standard attention at character-level tokenisation: the bilateral FFT kernel is structurally non-causal, coupling every position to future tokens. This defines an architectural boundary between bilateral spectral attention (this paper) and genuinely causal spectral attention at word-scale tokenisation (companion paper MorletQK). This work is architecturally distinct from FNet (Lee-Thorp et al., 2021), which replaces attention with Fourier mixing of token embeddings. Here, spectral preprocessing applies only to Q/K projections while the full attention score structure is preserved.
[NLP-12] Beyond Attack-Success Rate: Action-Graded Severity Scale for Tool-Using AI Agents
【速读】： 该论文旨在解决现有生成式AI红队测试基准中仅以二元成功/失败标签（binary attack-success rate）评估攻击效果所导致的严重信息丢失问题，即忽略了攻击行为实际造成的危害程度。其核心挑战在于：传统指标无法反映攻击者在越权、跨作用域或权限提升等关键维度上的具体危害性，从而误导对系统安全性的评估。为此，论文提出一种基于行动轨迹的七级定序危害评分框架（action-graded harm rubric, L0–L6），依据三个核心维度进行量化：动作是否可逆、是否跨越作用域影响第三方、是否实现权限扩展。该评分体系通过两种方式实现：一是基于确定性规则的“黄金标准”（deterministic oracle），二是由三名前沿大模型法官组成的评审小组对无标签轨迹进行判断。实验结果表明，该方法能揭示传统二元指标所掩盖的三种关键风险场景，例如某防御机制虽报告攻击成功率零，但依然存在未受过滤工具引发的外部可见跨作用域泄露。尽管评审小组与黄金标准间具有高度一致性（Krippendorff’s alpha = 0.91），但仍存在系统性盲点，尤其在识别权限升级链（escalation chains）方面表现不足。相较已有研究中的危害分类体系、任务完成度测试、执行层级安全基准或带严重性感知的仿真环境，本工作的创新在于提供了一种可复用、基于真实红队日志轨迹的严重性评估工具，直接作用于现有红队记录中的具体操作序列，显著提升了安全评估的精细化水平。所有代码、提示模板及每轮实验日志均已开源。
链接: https://arxiv.org/abs/2607.07474
作者: Harry Owiredu-Ashley
机构: 未知
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 8 pages, 6 figures. Code and artifacts: this https URL
Abstract:Agentic red-teaming benchmarks report whether an injected agent was compromised as a single bit: the attack succeeded, or it did not. We argue that this binary attack-success rate discards the information a defender most needs, namely how harmful the resulting action was. We introduce an action-graded harm rubric that scores an agent’s tool-call trajectory on a seven-level ordinal scale (L0 to L6) according to whether the executed action was reversible, whether it crossed scope to reach another party, and whether it expanded privilege. We compute the scale two ways: a deterministic oracle that reads the trajectory and the attacker’s stated goal, and a panel of three frontier language-model judges that read a tag-free account of the same trajectory. Across four victim models and two defenses on the AgentDojo workspace suite, severity grading exposes three cases the binary metric hides, including a defense that reports a zero attack-success rate while still permitting an externally visible cross-scope leak through an unfiltered tool. The judge panel reproduces the oracle with high ordinal agreement (Krippendorff’s alpha = 0.91) but shares systematic blind spots that we characterize, most notably a failure to recognize escalation chains. Unlike prior work that provides harm taxonomies, harmful-task completion tests, execution-level safety benchmarks, or severity-aware simulation, our contribution is a reusable, trace-grounded severity instrument applied to the actual actions recorded in existing red-team logs. All code, prompts, and per-episode logs are released.
[NLP-13] SynthAVE: Scalable Synthetic Labeling for E-Commerce with LLM -Arena Validation
【速读】： 该论文旨在解决大规模电商场景下属性值提取（Attribute Value Extraction, AVE）任务中高质量标注数据稀缺的问题。由于产品类型、属性维度及语言种类的组合规模庞大，人工标注所需成本极高，难以满足工业级应用需求。为此，论文提出SynthAVE，一个涵盖12,726个产品、229种产品类型、792个属性以及西班牙语、法语、意大利语和德语共4种语言的大规模人类验证基准数据集。其核心解决方案是设计一种多大语言模型（multi-LLM）竞技场框架（arena framework），通过21种独立的判别配置（7个模型族 × 3种提示模板）对合成标签进行并行评估，并采用多数投票机制确定最终标签。实验表明，该集成方法与人类专家的一致性达到Cohen’s κ = 0.92（95.2%一致率），单个模型间也表现出较高的一致性（Fleiss’ κ = 0.76）。这证明了多样化模型在个体判断差异基础上可通过集成实现高可靠性预测，从而在保证与人工标注相当质量的前提下，实现低成本、可扩展的合成标签验证。
链接: https://arxiv.org/abs/2607.07469
作者: Andrea Scarinci,Virginia Negri,Brayan Impata,Suleiman Khan,Victor Martinez,Marcello Federico
机构: Amazon(亚马逊)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Fine-tuning large language models (LLMs) for e-commerce attribute extraction requires labeled data representative across thousands of product types, attributes, and multiple languages. This combinatorial scale translates to millions of annotations, rendering human labeling prohibitively costly. While recent work has demonstrated synthetic label generation using LLMs, deploying such approaches at industrial scale requires integrated quality control mechanisms. We present SynthAVE, a large-scale human-validated benchmark for attribute value extraction spanning 12,726 products across 229 product types, 792 attributes, and 4 languages (Spanish, French, Italian, German). To validate synthetic labels at scale, we introduce a multi-LLM arena framework where samples are independently evaluated by 21 judge configurations (7 model families \times 3 prompts), with final labels determined via majority voting. The majority vote ensemble agrees with human experts at Cohen’s \kappa = 0.92 (95.2% agreement), while individual judges show substantial inter-model agreement (Fleiss’ \kappa = 0.76 ). This demonstrates that diverse models with varying individual judgments aggregate into highly reliable predictions, enabling cost-effective validation at scale while maintaining quality parity with human review.
[NLP-14] he Blind Curator: How a Biased Judge Silently Disables Skill Retirement in Self-Evolving Agents
【速读】： 该论文旨在解决在大语言模型（LLM）作为裁判（judge）执行无参考评估任务时，因裁判偏倚导致技能退役机制失效的问题。其核心挑战在于：生成式系统依赖“技能退役”这一结构约束来防止能力退化至无技能基线，而该机制的有效性依赖于无偏奖励信号，但实际中LLM裁判常引入偏差，尤其表现为“假通过偏倚”（false-pass bias，即失败案例被误判为成功），从而悄然关闭了系统的自我修正能力。论文的关键解决方案是通过受控的奖励污染分析与行为实验，在无参考报告生成测试平台中结合代码生成交叉验证，揭示了假通过偏倚会引发技能退役机制的突变式失效——一旦超过临界阈值，无论增加多少数据都无法恢复淘汰劣质技能的能力。进一步分析表明，这种机制失效具有普适性，仅在接近零假通过率、具备验证器特性的裁判中得以避免。尽管系统整体评估质量在某些情况下仍保持稳定，但其根本安全机制已“静默失效”，无法通过聚合指标暴露。因此，该研究贡献在于提出一种行为安全性结论而非性能提升，并设计了一种低成本的缺陷注入审计方法，可在部署前识别裁判是否处于临界阈值的危险侧。
链接: https://arxiv.org/abs/2607.07436
作者: Xing Zhang,Yanwei Cui,Guanghui Wang,Ziyuan Li,Wei Qiu,Bing Zhu,Peiyang He
机构: AWS Generative AI Innovation Center (亚马逊云科技生成式AI创新中心); HSBC Holdings Plc., HSBC Technology Center, China (汇丰控股公司，汇丰科技中心，中国)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Cryptography and Security (cs.CR)
备注:
Abstract:A self-evolving agent retires its bad skills by watching them fail, so what happens when the judge cannot see the failures? Skill retirement is the structural constraint that keeps a growing library from drifting below the no-skill baseline, but its guarantee assumes an unbiased reward, which is false for the LLM judges that reference-free tasks force upon us. We show that a biased judge does not merely add noise; it \emphsilently switches off the curator. We make this precise with a corrupted-reward analysis and, isolating the causal channel by injecting corruption on top of a deterministic reward, a behavioral study on a reference-free report-writing testbed with a code-generation cross-check. Symmetric noise leaves retirement intact, but \emphfalse-pass bias (failures slipping through as passes) disables contribution-based retirement past a sharp threshold that no amount of data can cross. Separating genuine retirement from cap-eviction churn shows this \emphmechanism failure is universal, holding across domains and failure rates and sparing only near-zero-false-pass, verifier-like graders. The downstream \emphoutcome, though, is regime-dependent: eval quality degrades only where the same corruption also starves skill synthesis, and otherwise holds steady, so the disabled curator is \emphsilent, surfacing in no aggregate metric. The contribution is a behavioral safety result, not a performance one. A cheap defect-injection audit then tells an operator, before deployment, which side of the threshold their judge occupies.
[NLP-15] DeLS-Spec: Decoupled Long-Short Contexts for Parallel Speculative Drafting
【速读】： 该论文旨在解决块并行推测解码（block-parallel speculative decoding）中长期上下文建模与局部因果依赖性之间的矛盾问题。现有方法如DFlash虽通过单次前向传播预测整个块以提升效率，但其位置无关的预测缺乏显式的块内因果约束，导致生成质量下降；而后续方法如Domino和DSpark虽引入了因果性，却需从头训练推测模型，带来高昂的训练成本且灵活性不足。本文提出DeLS-Spec，一种解耦的长短期上下文推测解码方法，其核心创新在于将固定预训练的DFlash模型作为长上下文专家，同时引入一个轻量级的局部头（local head）作为短上下文专家。该局部头可独立使用标准的下一个词预测目标进行训练，无需与目标模型或DFlash主干联合训练，从而实现极低的训练开销。在推理阶段，DeLS-Spec融合长、短上下文的输出概率分布，且局部头不绑定特定的DFlash检查点，显著提升了系统的模块化与灵活性。实验结果表明，在Qwen3模型上，DeLS-Spec在数学、代码和对话等多类基准任务中均持续优于DFlash，实现了更高的加速比与平均接受长度。
链接: https://arxiv.org/abs/2607.07409
作者: Hong-Kai Zheng,Piji Li
机构: Nanjing University of Aeronautics and Astronautics (南京航空航天大学); MIIT Key Laboratory of Pattern Analysis and Machine Intelligence (工业和信息化部模式分析与机器智能重点实验室); The Key Laboratory of Brain-Machine Intelligence Technology, Ministry of Education (教育部脑-机智能技术重点实验室)
类目: Computation and Language (cs.CL)
备注:
Abstract:Speculative decoding accelerates LLM inference by drafting multiple tokens and verifying them in parallel. Block-parallel drafters such as DFlash further improve drafting efficiency by predicting an entire block in one pass, but their position-wise predictions lack explicit intra-block causal conditioning. Recent methods such as Domino and DSpark attempt to introduce such causality into block-parallel drafting, but they require training the draft model from scratch, which limits their flexibility and increases training cost. We propose DeLS-Spec, a decoupled long-short context speculative decoding method. DeLS-Spec treats the fixed DFlash model as a long-context expert and introduces a lightweight local head as a short-context expert. The local head can be trained independently with a standard next-token prediction objective, without joint training with the target model or the DFlash backbone, leading to extremely low training cost. At inference time, DeLS-Spec combines long-context and short-context logits, and the local head is not tied to a specific DFlash checkpoint, making the method more modular and flexible. Experiments on Qwen3 models show that DeLS-Spec consistently improves speedup and average acceptance length over DFlash across math, code, and dialogue benchmarks.
[NLP-16] ransformer-based segmentation of prosodic boundaries in Brazilian Portuguese
【速读】： 该论文旨在解决巴西葡萄牙语（Brazilian Portuguese, BP）中自动韵律切分（automatic prosodic segmentation）性能不足的问题，现有方法仍主要依赖规则或传统机器学习，难以充分捕捉语言的复杂韵律特征。其解决方案的关键在于提出SAMPA——一个基于Whisper大模型（Whisper large-v3）的端到端切分框架，通过在语音转录过程中显式插入韵律边界标记（prosodic boundary markers），实现对语音单位边界的精准识别。研究通过在人工标注的NURC-SP数据集上微调模型，并系统评估多种训练与推理时的过滤策略，包括在跨数据集（MuPe-Diversidades）上的分布外测试，验证了模型在不同场景下的鲁棒性与泛化能力。实验表明，最优模型在保留测试集上达到F1=0.731，而在MuPe-Diversidades上达到F1=0.796，表现出较强的边界检测性能。进一步的n元语法分析与声学-视觉分析揭示，模型能够有效响应形态句法、语义及韵律线索，从而实现符合语言学规律的边界预测。
链接: https://arxiv.org/abs/2607.07408
作者: Rodrigo de Freitas Lima,Julio Cesar Galdino,Marcos Vinicius Treviso
机构: 未知
类目: Computation and Language (cs.CL)
备注: 6 pages, 5 figures, submitted to an IEEE conference
Abstract:Automatic prosodic segmentation identifies boundaries between speech units from acoustic and linguistic evidence. Although recent deep learning approaches have produced strong results for English, automatic segmentation for Brazilian Portuguese (BP) still relies mostly on rule-based or traditional machine-learning methods. This paper presents SAMPA, a Whisper-based segmenter that transcribes BP speech while inserting explicit markers for terminal prosodic boundaries. We fine-tune Whisper large-v3 on manually segmented recordings from the NURC-SP dataset and evaluate different training and test-time filtering configurations, including out-of-distribution testing on the MuPe-Diversidades dataset. SAMPA achieves competitive boundary-detection performance across settings, with the best models reaching F1=0.731 on the held-out test split and F1=0.796 on MuPe-Diversidades. Finally, through n-gram and acoustic-visual analyses, we show that our model follows morphosyntactic, semantic, and prosodic cues for detecting prosodic boundaries.
[NLP-17] F-Engram: A Train-Free Engram with SSD-Backed Memory for Large Language Models
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在知识扩展过程中因依赖预训练、微调、检索增强或更长上下文而带来的高成本问题。现有基于记忆（Engram-style memory）的方案虽能以紧凑方式注入隐藏状态，但依赖哈希压缩的GPU驻留设计常导致不相关短语在共享槽位中发生冲突，损害短语级语义保真度。为此，论文提出TF-Engram，一种无需训练的Engram系统：其关键在于离线从外部语料构建针对特定短语的语义记忆表，将大规模记忆表跨部署于GPU–DRAM–SSD分层存储架构中，并采用早期退出引导的预测性预取（Early-Exit Guided Predictive Prefetching）机制，在自回归解码过程中隐藏外部内存访问延迟。实验表明，该方法在Qwen3-0.6B上将下游平均得分从57.6提升至59.4，优于冻结主干网络及参数量相当的LoRA基线；系统评估进一步验证了大规模记忆表可实现较低开销的离线构建、基于SSD的存储显著降低GPU内存占用，且预测预取有效缓解了外部内存访问带来的吞吐损失。结果表明，静态短语记忆可作为可扩展、免训练、低开销的组件集成至LLM推理流程中。
链接: https://arxiv.org/abs/2607.07388
作者: Yutang Ma,Kecheng Huang,Xikun Jiang,Zili Shao
机构: The Chinese University of Hong Kong (香港中文大学); Beijing Institute of Technology, Zhuhai (北京理工大学珠海学院)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 13 pages, 2 figures
Abstract:Large Language Models (LLMs) store factual knowledge and domain-specific patterns implicitly in dense Transformer parameters, making knowledge expansion costly through pretraining, fine-tuning, retrieval augmentation, or longer contexts. Engram-style memory offers a compact hidden-state injection pathway, but existing GPU-resident designs often rely on hash-based compression, causing unrelated phrases to collide in shared slots and weakening phrase-level semantic fidelity. We present TF-Engram, a train-free Engram system that constructs phrase-specific semantic memory offline from external corpora, stores large memory tables across a GPU–DRAM–SSD hierarchy, and uses Early-Exit Guided Predictive Prefetching to hide external-memory latency during autoregressive decoding. On Qwen3-0.6B, TF-Engram improves the average downstream score from 57.6 to 59.4, outperforming both the frozen backbone and a parameter-matched LoRA baseline. System evaluation shows that large TF-Engram tables can be built with moderate offline cost, SSD-backed storage substantially reduces GPU memory demand, and predictive prefetching recovers much of the throughput loss caused by external memory access. These results demonstrate that static phrase memory can be integrated into LLM inference as a scalable, train-free, and low-overhead system component.
[NLP-18] R3: Advertisement Compliance Rectification via Group-Relative Experience Extractor and Curriculum Reinforcement ACL2026
【速读】： 该论文旨在解决在线广告内容审核中因严格合规要求导致的海量视频广告被拒问题，尤其针对文本类违规（包括语音转录文本与屏幕显示文本）的修正难题。现有基于安全驱动的方法常因过度编辑而严重损害广告原始语义意图，难以在合规性与内容真实性之间取得平衡。其解决方案的关键在于提出R³框架，通过三项核心创新实现合规性与语义意图保留的协同优化：（1）基于经验驱动的数据合成方法，利用群体相对合规经验提取器构建高质量监督信号；（2）采用分层奖励机制的课程强化学习策略，在确保合规的同时最大化语义一致性；（3）构建端到端的视频修正框架，集成文本识别、重写与重渲染模块，支持工业级部署。实验证明，R³在真实工业数据集及线上A/B测试中显著优于现有最优基线，实现了违规修正与语义保真之间的最佳权衡。
链接: https://arxiv.org/abs/2607.07318
作者: Yuan Chen,Zhenyu Hu,Mengge Xue,Te Cao,Liqun Liu,Peng Shu,Huan Yu,Jie Jiang
机构: Tencent(腾讯)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: ACL 2026 (Poster, Industry Track)
Abstract:Rigorous content moderation is crucial for online advertising but leads to millions of daily rejections. This scale renders manual rectification infeasible, particularly for video advertisements. However, existing safety-driven methods often suffer from aggressive over-editing, which compromises the advertiser’s original semantic intent merely to satisfy compliance. In this work, we target the rectification of textual violations in video ads, covering both speech transcripts and on-screen text. We propose R^3, a novel framework designed to harmonize compliance with original semantic intent preservation. Our approach integrates three key innovations: (1) an experience-driven data synthesis framework that bootstraps high-quality supervision via a group-Relative compliance experience extractor; (2) a curriculum Reinforcement learning strategy with hierarchical rewards designed to enforce compliance while maximizing semantic consistency; and (3) a comprehensive video Rectification framework seamlessly integrating text recognition, rewriting, and re-rendering for industrial deployment. Extensive experiments on industrial datasets and online A/B testing demonstrate that R^3 significantly outperforms state-of-the-art baselines, achieving an optimal trade-off between violation rectification and intent preservation.
[NLP-19] Evaluating RAG Metrics in Applied Contexts: An Experiment Its Findings and Its Limitations
【速读】： 该论文旨在解决生成式检索增强生成（RAG）系统中评估指标有效性与人工评价之间相关性不足的问题，尤其关注现有评估指标在真实业务场景下是否能准确反映生成回答的质量与检索内容的相关性。其解决方案的关键在于通过构建基于真实业务数据的人工标注问答数据集，对RAG系统的生成响应和检索片段进行多维度评估，综合运用来自四个评估工具库（Ragas、DeepEval、RAGChecker、Opik）的自动化指标，并将其结果与人工评价者评分及标准信息检索指标（如召回率）进行相关性分析，从而验证各指标的实际效用。研究进一步揭示了当前评估方法的局限性，并为未来更可靠、可解释的RAG评估框架提供了方向。
链接: https://arxiv.org/abs/2607.07302
作者: Quentin Brabant
机构: Orange Research (橙色研究)
类目: Computation and Language (cs.CL)
备注:
Abstract:This paper reports an empirical study evaluating the relevance of several RAG metrics. The experiment is based on a question-answering dataset created by human annotators from business data. The generated responses and retrieved spans of a RAG system are scored using evaluation metrics from four libraries (Ragas, DeepEval, RAGChecker, Opik). These metrics are compared to scores given by two evaluators, as well as to standard metrics such as recall. An analysis of correlations is conducted. Finally, we highlight certain limitations of our methodology, compare it to those used in the literature, and suggest some avenues for future research. This paper is an English translation of a paper originally published in the French-speaking workshop EvalLLM (Brabant, 2026).
[NLP-20] Multimodal Voice Activity Projection for Turn-Taking in Social Robots with Voice-Activity-Related Pretrained Encoders DATE
【速读】： 该论文旨在解决社交机器人在人类-人类互动场景中（尤其是调解型交互）的对话轮换预测问题，核心挑战在于机器人需主动预判对话动态而非仅响应沉默。其解决方案的关键在于提出一种多模态语音活动投影（Multimodal Voice Activity Projection, MM-VAP）框架，该框架将原始仅基于音频的语音活动投影（VAP）模型扩展至同步的音视频输入，同时保持自监督的未来投影目标。该方法利用预先训练的音视频骨干网络（针对语音任务优化），通过低秩适应（Low-Rank Adaptation, LoRA）进行微调，并在独立说话人编码后引入跨说话人注意力机制，以建模对话双方之间的关系动态，从而实现对未来语音活动的准确预测。此外，引入语义一致性损失（semantic consistency loss），对256维输出空间进行正则化，使其与高层对话行为模式保持一致。实验结果表明，该方法在NoXi和NoXi+J数据集上优于现有基线，尤其在部分轮换事件上表现显著提升；在Haru EDR语料库上的额外评估进一步验证了其在面向调解任务的人机交互中的适用性。
链接: https://arxiv.org/abs/2607.07294
作者: Antonio Cano,Guillermo Pérez,Luis Merino,Randy Gomez
机构: 4i Intelligent Insights(4i智能洞察); Universidad de Sevilla(塞维利亚大学); Universidad Pablo de Olavide(帕博·德·奥拉维德大学); Honda Research Institute Japan(本田研究院日本)
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: Accepted for presentation at the 35th IEEE International Conference on Robot and Human Interactive Communication (RO-MAN 2026). Acceptance notification date: 30 May 2026. Final published version pending
Abstract:Turn-taking prediction is a key requirement for social robots involved in human-human interaction, particularly in mediator settings, where the robot must anticipate conversational dynamics rather than merely react to pauses. This work presents a Multimodal Voice Activity Projection (MM-VAP) framework that extends the original audio-only VAP formulation to synchronized audio-visual inputs while preserving its self-supervised future-projection objective. The proposed approach builds on pretrained audio-visual backbones originally optimized for speech-related tasks and adapts them through Low-Rank Adaptation to the multimodal turn-taking problem. After independent speaker encoding, an inter-speaker attention stage models the relational dynamics required to project future voice activity. In addition, a semantic consistency loss is introduced to regularize the 256-state output space according to higher-level dialogue activity patterns. Experiments on NoXi and NoXi+J showed improvements over the current baselines, particularly for some turn-taking events. Additional evaluation on the Haru EDR corpus further supported the suitability of this direction for mediation-oriented human-robot interaction.
[NLP-21] A Word-Level Digital Reader of the Prasthanatrayi with Sankaras Bhasya: Corpus Method and an Open Offline Reading Aid for the Advaita Vedanta Canon
【速读】： 该论文旨在解决印度教不二论（Advaita Vedanta）核心典籍——“普拉斯坦纳特拉伊”（Prasthanatrayi，即十部主要《奥义书》、《梵经》及《薄伽梵歌》及其商羯罗（Sankara）注释）在文本阅读中因连续音变（sandhi）、复杂复合词（samasa）和密集的学术性散文表达而导致的词级解析困难问题。其核心挑战在于：文本中词与词之间的边界模糊，词形变化与音变规则掩盖了词根与语法功能，严重阻碍了精确的语义理解与研究。本文提出的关键解决方案是构建一个完全离线、开源的词级数字阅读器，支持对整个普拉斯坦纳特拉伊及其商羯罗注疏进行逐词点击解析。该系统通过一个混合式处理流程实现高精度分析：基于规则的音变拆分器结合词形词典与实证语料库查表，并引入大语言模型（LLM）辅助分析，在对抗性双重验证协议下确保可靠性；同时建立持久的人工审核闭环，确保校正结果在每次重新生成时均被保留。该工具将所有词汇映射至词元（lemma），不仅实现词形切分（padaccheda）、形态学分析与释义弹窗，还具备跨层次（原文与注疏）的词频检索功能，形成一个可检索的词频对照表（concordance）。评估结果显示，99%以上的实证词形分析与权威词形词典高度一致，且错误集中于低置信度层级，符合预期分布，验证了系统的可靠性。最终产品为一个独立的单文件HTML应用，无需服务器或网络支持，可自由分发，作为教学与研读工具。
链接: https://arxiv.org/abs/2607.07282
作者: Tamal Maharaj
机构: Ramakrishna Mission Vivekananda Educational and Research Institute (梵音教育与研究学院)
类目: Computation and Language (cs.CL); Digital Libraries (cs.DL)
备注:
Abstract:The Prasthanatrayi – the ten principal Upanisads, the Brahmasutra, and the Bhagavadgita, with Sankara’s commentaries (bhasya) – is the foundational corpus of Advaita Vedanta. Continuous euphonic combination (sandhi), long compounds (samasa), and dense scholastic prose make it hard to read at the word level: where one word ends, and what each word means grammatically, are both obscured. We present an open, fully offline, word-level digital reader of the entire Prasthanatrayi with Sankara’s bhasya. Every word – of both the root text (mula) and the commentary – is clickable and resolves to a pop-up giving its split (padaccheda), morphological analysis, and gloss. Because every word carries a lemma, the reader also acts as a concordance: a search on a dictionary headword retrieves all of that word’s inflected and sandhi-hidden occurrences, and its occurrences inside compounds, across both layers. The resource covers thirteen commentarial units (2,971 verses, sutras, and prose sections; 36,881 analysed word-occurrences of root text) and a global dictionary of 95,587 distinct commentarial surface forms. We describe the corpus, the hybrid pipeline – a rule-based sandhi splitter over an inflected-form lexicon and attested-corpus look-ups, with LLM-assisted analysis under an adversarial two-pass verification protocol – and a durable human-review loop whose corrections survive every regeneration. An intrinsic evaluation against independent Sanskrit resources finds high-confidence analyses agree with an authoritative inflectional lexicon on over 99% of attested forms, and a band-blind adjudication confirms that quality degrades predictably across confidence bands, with errors concentrated in the low-confidence tier the review loop targets. The reader is a single self-contained HTML file needing no server or network, offered as a freely redistributable teaching and reading aid.
[NLP-22] Understanding Interpretation Difficulty in Harmful Online Communication: Insights from Cybercrime Communities
【速读】： 该论文旨在解决网络有害内容中因使用俚语、隐晦表达、缩写及社群特有术语而导致的语义解读困难问题，尤其聚焦于与网络犯罪相关的Discord聊天内容。其核心解决方案在于将解读过程视为一个证据整合（evidence-integration）问题，而非仅依赖消息级别的分类。研究通过构建专家审定的参考解读，评估人类与大语言模型（Large Language Model, LLM）在不同上下文条件下的解读表现，发现局部上下文对人类解读能力提升有限，而外部知识和扩展对话上下文能显著改善理解；对于LLM而言，局部上下文同样有助于提升解读效果，且模型规模越大性能越优。进一步的定性错误分析提出了导致有害聊天难以解读的若干因素分类，强调在有害内容分析中应综合多源信息以实现更准确的理解。
链接: https://arxiv.org/abs/2607.07277
作者: Tomohiro Okatsu,Naoki Takada,Yin Min Pa Pa,Katsunari Yoshioka,Tatsunori Mori
机构: Yokohama National University, Japan
类目: Computation and Language (cs.CL); Computers and Society (cs.CY)
备注:
Abstract:Harmful online communication often contains slang, coded terms, abbreviations, and community-specific expressions, which make messages difficult to interpret. This paper presents an exploratory study of interpretation difficulty in Discord chats related to cybercrime. We construct reference interpretations of purposefully selected difficult messages, which were reviewed by an expert. We then use them to evaluate human and large language model (LLM) interpretations under different context conditions. The results show that local context alone is often insufficient for humans, while external knowledge and extended conversational context substantially improve human interpretation. For LLMs, local context also improves interpretation, and the larger model performs better. We further conduct a qualitative error analysis and propose a preliminary classification of factors that make harmful chats difficult to interpret. These findings suggest that harmful-content analysis should treat interpretation as an evidence-integration problem, rather than as message-level classification alone.
[NLP-23] Billions of Sketches Reveal Hidden Cultural Variation in Human Concepts
【速读】： 该论文旨在解决人类概念普遍性（conceptual universality）研究中长期依赖语言相似性评估所导致的潜在偏差问题。传统方法通过跨语言词汇比较来推断概念的普遍性，但语言作为交流工具会将丰富的个体与文化体验压缩为共享惯例，从而掩盖了概念在心智表征中的深层差异。为此，论文提出以大规模视觉数据为切入点，分析来自236个主权国家和地区、总计26亿张由人类绘制的常见概念草图，从视觉想象角度揭示概念结构的内在多样性。其核心解决方案在于利用草图嵌入模型（sketch embedding models）与词嵌入模型（word embedding models）的对比，发现视觉表征能够保留更丰富的语义与文化结构信息，而语言模型则倾向于压缩这些细节。研究显示，基于草图的跨文化相似性与已知文化距离的相关性比文本测量高出45%，且涉及触觉交互的概念表现出更强的文化差异，表明视觉意象不仅反映符号约定，更映射了具身经验（embodied experience）的多样性。因此，该研究的关键突破在于证明：概念普遍性的模式高度依赖于测量模态，而大规模草图数据提供了一种直接、高分辨率的工具，可有效探测思想在具身化与文化维度上的深层多样性。
链接: https://arxiv.org/abs/2607.07267
作者: Arianna Pera,Mauro Martino,Nima Dehmamy,Douglas Guilbeault,Luca Maria Aiello,Andrea Baronchelli
机构: 未知
类目: Computers and Society (cs.CY); Computation and Language (cs.CL); Physics and Society (physics.soc-ph)
备注:
Abstract:Claims about the universality of human concepts have been predominantly assessed through linguistic similarity across languages and cultures. However, words are effective as communication devices because they compress rich experiential variation into shared conventions, potentially obscuring hidden individual and cultural differences in how concepts are mentally represented. Here, we analyse 2.6 billion human-made sketches of common concepts from 236 countries and territories to examine conceptual structure through people’s visual imagination. Consistent with recent work on image-based cognition, we find that single concepts unfold into multiple distinct visual exemplars, revealing latent information about similarities and differences in conceptual structure across cultures. This variation is strongest for concepts involving haptic interaction, suggesting that visual imagery reflects variation in embodied experience as much as conventional definitions. Comparing embedding models of sketches with word embedding models across languages, we find that their geometries diverge, with visual representations preserving rich semantic and cultural structure that language models compress. Cross-cultural similarities derived from sketches align 45% more closely with established cultural distances than do text-based measures. Together, these results suggest that patterns of human conceptual universality may depend critically on the modality through which concepts are measured, with large-scale sketching providing a direct, high-resolution probe of conceptual diversity across embodied and cultural dimensions of thought.
[NLP-24] Evaluation of Multilingual Ability to Use Spatial Deictic Expressions in Vision-Language Models ACL
【速读】： 该论文旨在解决视觉语言模型（VLMs）在多语言环境下对空间指示性表达（spatial deictic expressions）的准确理解和使用能力不足的问题。空间指示性表达（如“this”和“that”）的指代对象依赖于具体语境，要求模型能够联合推理语言信息与视觉空间结构，实现对上下文相关指代的精准定位。此外，不同语言中空间指示词的语义区分具有语言特异性，模型需具备跨语言理解此类空间差异的能力。为此，论文构建了一个涵盖四种语言的基准测试集，用于评估VLMs在多语言场景下对空间指示性表达的运用能力。实验结果表明，现有模型在基于物体距离选择恰当指示词方面表现显著偏离人类认知模式，其核心问题在于缺乏对视觉空间结构与语言语用规则之间深层关联的建模能力。解决方案的关键在于提升模型对视觉空间上下文的精细感知能力，并强化其对多语言中空间指代差异的语义理解能力。
链接: https://arxiv.org/abs/2607.07251
作者: Kaito Watanabe,Taisei Yamamoto,Tomoki Doi,Hitomi Yanaka
机构: The University of Tokyo(东京大学); Riken(理化学研究所); Tohoku University(东北大学)
类目: Computation and Language (cs.CL)
备注: Accepted to ACL SRW 2026
Abstract:One of the expected abilities of vision-language models (VLMs) is spatial reasoning ability based on a given text and image. To evaluate the spatial reasoning abilities of VLMs, we focus on the use of spatial deictic expressions, which are defined as spatial expressions whose referent is determined by their situational context, such as this'' and that’'. To handle spatial deictic expressions, VLMs must jointly reason over language and visual space, grounding context-dependent references in the image’s spatial structure. In addition, selecting appropriate spatial deictic expressions across languages requires VLMs to understand the language-specific spatial distinctions encoded by these expressions. In this paper, we develop a benchmark to evaluate the multilingual ability of VLMs to use spatial deictic expressions in four languages. Our experiments using this benchmark reveal that the tested models use demonstratives in a manner different from that of humans, particularly in selecting the appropriate demonstratives based on the distance to the object.
[NLP-25] From Text to Parameters: Predicting Item Parameters from Embedding Regularization with Reliability and Design Ceilings
【速读】： 该论文旨在解决项目校准中的“冷启动”问题，即新开发题目的心理测量属性在未经过实地测试前无法确定，导致难以进行有效校准。其解决方案的关键在于利用现代文本嵌入（text embeddings）技术自动构建传统上需人工指定的设计矩阵，并结合正则化回归模型预测题目参数。研究提出了一套综合评估框架，包括基于重复交叉验证的决定系数（R²）及其重抽样标准差、以及两个性能上限：由参数标准误推导出的信度上限（reliability ceiling）和通过模拟功效校准得到的设计上限（design ceiling）。实证分析表明，题目难度可高度由文本特征预测（重复交叉验证R² = 0.53，约为信度上限的57%），而区分度与伪猜测参数的可预测性较低；但通过对比性能上限发现，这种差异主要源于目标参数的可靠性水平而非文本信号强度——文本对所有难度目标均能解释约57%至63%的可靠方差，而三参数逻辑斯蒂模型（3PL）中的伪猜测参数因信度上限接近零，实际上不可靠。此外，在医学执照考试基准（BEA 2024）中，基于嵌入的回归虽解释方差极少，却达到领先榜单的均方根误差（RMSE），凸显了在基准评测中使用无量纲指标与明确性能上限的重要性。最后，研究揭示单次训练-测试划分可能使R²值虚高0.1至0.15，强调了在项目校准支持及未来基准建设中采用重复交叉验证的必要性。
链接: https://arxiv.org/abs/2607.07141
作者: Shi-Ting Chen,Jinsong Chen
机构: 未知
类目: Computation and Language (cs.CL); Methodology (stat.ME)
备注:
Abstract:Newly developed items must ordinarily be field tested before their psychometric properties are known, creating a cold start problem for item calibration. Predicting item parameters from features is a long standing measurement problem dating back to the Linear Logistic Test Model; modern text embeddings now automate the design matrices traditionally specified by hand. We propose an evaluation framework combining regularized regression on item text embeddings, repeated cross validated R squared reported with its resampling standard deviation, and two performance upper bounds: a reliability ceiling derived from parameter standard errors, and a design ceiling derived from simulation based power calibration. Applying this framework to a mathematics item bank (EEDI) and a medical licensure benchmark (BEA 2024), we find that item difficulty is highly predictable from text (repeated cross validated R squared = 0.53, or about 57% of its reliability ceiling), whereas discrimination and pseudo guessing appear less predictable. However, evaluating these results against our ceilings reveals that this apparent hierarchy stems from target reliability rather than text signal strength: text uniformly recovers 57 to 63% of the reliable variance across difficulty targets, whereas the 3PL pseudo guessing parameter has a reliability ceiling near zero, making it an unviable target at current precision. On BEA, embedding based regression matches leaderboard RMSE despite explaining almost no variance, highlighting the critical need for scale free metrics and explicit ceilings in benchmarking. Finally, we show that a single train and test split can inflate apparent accuracy by 0.1 to 0.15 in R squared, underscoring the necessity of repeated cross validation for calibration support applications and future benchmark construction.
[NLP-26] Behavior Leverag e Imbalance in Multi-Teacher On-Policy Distillation
【速读】： 该论文旨在解决多教师在线策略蒸馏（multi-teacher on-policy distillation, OPD）在生成式语言模型工具调用任务中出现的“行为偏差”问题，即模型在训练过程中因教师信号分布不均衡而产生过度调用工具（over-calling）的现象。尽管整体损失函数表现良好，但模型在具体决策行为上出现了不可见的退化，表现为对本应直接回答的任务仍频繁调用工具。其解决方案的关键在于识别并缓解“行为杠杆失衡”（behavior leverage imbalance）——即在生成过程中，某些特定位置的局部token级信号（如tool_call、函数名等）对全局生成模式具有过强控制力，导致模型偏离正确行为。为此，作者提出一种名为Soft Clamp的逐token发散校准方法，通过动态压缩极端的Jensen-Shannon发散值，同时保留非零梯度，从而在不牺牲决策准确率的前提下，有效降低过调用率（从13.7%降至9.0%），并减少工具调用循环和重复调用。研究强调，多教师OPD应关注教师信号在生成过程中的作用位置与机制，而不仅依赖于聚合层面的损失指标。
链接: https://arxiv.org/abs/2607.07050
作者: Jiabin Shen,Guang Chen,Chengjun Mao
机构: Ant Group (蚂蚁集团)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 17 pages including appendix, 6 figures
Abstract:Agentic language models must learn when to call tools, when to consume tool responses, and when to answer directly. This makes multi-teacher on-policy distillation a natural training strategy: one teacher can specialize in tool calls, another in direct responses, and the student can learn from both on its own generated distribution. We show that this strategy can induce a behavior shift that is invisible from aggregate losses alone. In a two-teacher tool-use setting, vanilla generalized knowledge distillation improves tool-call recall but also moves the model toward over-calling, where it calls tools on examples that should be answered directly. Aggregate explanations are insufficient: tool-call samples do not receive more token exposure, and full-sequence per-token divergence is not larger for the tool-call teacher. We instead analyze behavior leverage imbalance: local token-level signals at mode- entry and structural positions, such as tool_call and function names, can have disproportionate control over the global generation mode. We propose Soft Clamp, a per-token divergence calibration method that dynamically compresses extreme token-level Jensen-Shannon divergence while preserving nonzero gradients. On APIGen-MT, Soft Clamp reduces over-calling from 13.7% to 9.0% relative to vanilla GKD while matching its decision accuracy. In a BFCL multi-turn diagnostic, it also lowers tool-call loops and repeated calls among GKD variants. These results suggest that multi-teacher OPD should monitor where teacher signals act, not only how large they are in aggregate. Comments: 17 pages including appendix, 6 figures Subjects: Computation and Language (cs.CL); Machine Learning (cs.LG) Cite as: arXiv:2607.07050 [cs.CL] (or arXiv:2607.07050v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2607.07050 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Jiabin Shen [view email] [v1] Wed, 8 Jul 2026 06:26:13 UTC (135 KB) Full-text links: Access Paper: View a PDF of the paper titled Behavior Leverage Imbalance in Multi-Teacher On-Policy Distillation, by Jiabin Shen 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.LG 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-27] Riemannian Geometry for Pre-trained Language Model Embeddings
【速读】： 该论文旨在解决预训练语言模型嵌入表示中句级分类信号是否存在于上下文词元嵌入的黎曼几何结构中的问题，以提升模型的可解释性与安全性。其核心解决方案是提出一种称为黎曼均值池化（Riemannian Mean Pooling, RMP）的方法，通过从编码器的解析雅可比矩阵中提取每个词元的拉回度量（pullback metrics），并在对称正定（SPD）流形上利用弗雷歇均值（Fréchet mean）进行聚合，从而捕捉嵌入空间中的非欧几里得几何结构。实验表明，在具有复杂语言结构的三个数据集（CoLA、CREAK、RTE）上，RMP显著优于传统的欧氏均值池化；而在去除标注驱动词汇伪影的FEVER-Symmetric基准上，RMP性能保持在随机水平，验证了其对真实语义信号的敏感性。消融实验进一步揭示，仅使用随机初始化编码器结合弗雷歇聚合即可在两个数据集上超越欧氏池化，说明性能提升主要源于几何聚合机制本身，而非模型学习到的流形结构；而经过训练的编码器仅在最依赖知识的CREAK数据集上带来额外增益，表明其在特定任务中对深层语义建模具有补充作用。
链接: https://arxiv.org/abs/2607.07047
作者: Szczepan Konior,Alexandre Quemy,Przemysław Klocek,Grégoire Cattan,Bartłomiej Sobieski
机构: IBM Automation and AI, Krakow, Poland; Hother, Krakow, Poland; University of Warsaw; Centre for Credible AI, Warsaw University of Technology
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Understanding the geometric structure of pre-trained language model embeddings matters for interpretability and safety. We ask whether sentence-level classification signal lives in the Riemannian geometry of contextual token embeddings, and probe it by extracting per-token pullback metrics from a learned encoder’s analytical Jacobian and aggregating them with the Fréchet mean on the symmetric positive definite (SPD) manifold; we call this procedure Riemannian Mean Pooling (RMP). Across three datasets with non-trivial linguistic structure (CoLA, CREAK, RTE), RMP outperforms Euclidean mean pooling, while on FEVER-Symmetric, a benchmark constructed to remove annotation-driven lexical artifacts, the method correctly stays at chance. Ablations show that a randomly initialised encoder combined with Fréchet aggregation already beats Euclidean pooling on two of the three signal-bearing datasets, localising the source of the gain to the geometric aggregation rather than to learned manifold structure; the trained encoder contributes additional signal specifically on CREAK, the most knowledge-heavy of the three signal-bearing datasets.
[NLP-28] Dissociating the Internal Representations of Sycophancy in LLM s ICML2026
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）中普遍存在的“谄媚行为”（sycophancy）问题，即模型在用户陈述错误时仍无条件附和的现象。尽管谄媚行为常被视为单一行为模式，但其在实际表现中存在显著差异，尤其体现在对事实性陈述与主观观点的不同响应机制上。为此，论文提出将谄媚行为解耦为两类子类型：基于可验证事实的“事实性谄媚”与基于主观信念的“观点性谄媚”，并依据二者之间的语义区分进行建模。研究通过在某一子类型的激活表示上训练线性探测器并构建控制向量，进而评估其在另一子类型上的迁移能力，以衡量两类谄媚行为在模型内部表征上的共享程度。结果表明，不同大语言模型在两类谄媚行为的表征结构上存在显著差异，表现为表征统一性或表征分离且存在因果干扰等不同模式。该解耦方法为揭示复杂模型行为的内在表征结构提供了有效框架，是理解并干预模型谄媚现象的关键突破。
链接: https://arxiv.org/abs/2607.07003
作者: Anthony Baez,Sheer Karny,Pat Pataranutaporn
机构: 未知
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: Accepted to Mechanistic Interpretability Workshop at ICML 2026
Abstract:Large Language Models (LLMs) frequently exhibit sycophancy, where they agree with a user’s statement even when incorrect. While sycophancy is often treated as a single defined behavior, it can manifest in substantially distinct ways and circumstances, raising the question of whether this multi-faceted nature is reflected in its internal mechanisms. To address this gap, we dissociate the representations of sycophancy into factual and opinion subtypes – motivated by the distinction between verifiable claims and subjective beliefs. We train linear probes and construct steering vectors on activations of one subtype and evaluate their transfer to the other subtype to measure to what extent they share representations. We find evidence that different LLMs represent these subtypes differently, with either more unified or more distinct and causally interfering representations. This method of dissociation offers a promising framework for studying the representational structure of complex model behaviors.
[NLP-29] MILES: Modular Instruction Memory with Learnable Selection for Self-Improving LLM Reasoning
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在测试时推理过程中，面对序列化问题输入时无法有效积累和复用经验的问题。现有方法或依赖全解模板存储导致泛化能力差，或采用启发式步骤级选择策略，缺乏对最终答案正确性的优化。此外，传统学习选择策略需大规模训练数据与固定动作空间，难以适应测试时增量扩展记忆且监督信号有限的现实约束。其解决方案的关键在于提出MILES（Modular Instruction Memory with LEarnable Selection for self-improving LLM reasoning），一种支持动态步进式记忆扩展并基于答案正确性优化的记忆组合机制。MILES通过维护由子目标嵌入与子指令组成的非对称模块化记忆单元，并为每个单元配备可学习的选择头，实现粗粒度到细粒度的检索流程：粗粒度阶段支持记忆扩展并从高置信样本中收集监督信号以训练选择头；细粒度阶段利用学习到的选择头对粗粒度候选进行重排序，指导不确定样本的推理过程。该框架在保持高效性的同时显著提升准确率，在多种场景下展现出优越的性能、鲁棒性及跨任务迁移能力。
链接: https://arxiv.org/abs/2607.06974
作者: Ruilin Tong,Dong Gong
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Large language models (LLMs) increasingly improve their reasoning at test time via additional computation, yet most existing works treat each problem in isolation. When problems arrive sequentially, accumulating reusable experience across them can further improve performance. Existing memory-based methods either store whole-solution templates that generalize poorly to novel problems or use heuristic step-level selection that is not optimized for final-answer correctness. Learning selection policies requires large-scale training data and fixed action spaces, making such approaches unsuitable for test-time settings where memory expands incrementally and only limited supervision is available. We propose MILES (Modular Instruction Memory with LEarnable Selection for self-improving LLM reasoning), a framework that dynamically expands step-wise memory and applies correctness-optimized memory composition under realistic test-time constraints. MILES maintains modular memory units consisting of asymmetric pairs of sub-goal embeddings and sub-instructions, each associated with a learnable selection head. This memory structure enables a coarse-to-fine retrieval mechanism: The coarse level enables memory expansion and collects supervision for training selection heads from confident samples, while the fine stage applies learned selection heads to rerank coarse-level candidates and guide reasoning for uncertain samples. MILES consistently matches or outperforms prior methods while achieving superior accuracy-efficiency tradeoffs. Extensive experiments demonstrate its effectiveness, robustness, and transferability.
[NLP-30] Large Language Models (LLM s) and Generative AI in Cybersecurity and Privacy: A Survey of Dual-Use Risks AI-Generated Malware Explainability and Defensive Strategies
【速读】： 该论文旨在解决生成式人工智能（Generative AI）在网络安全领域中带来的双重挑战：一方面，大语言模型（LLMs）被广泛应用于自动化威胁检测、安全代码生成和漏洞发现等防御性场景；另一方面，其能力也被恶意利用于生成大规模、高隐蔽性的恶意软件，导致2025年预计有50%的已检测威胁源自LLM生成内容，远高于2021年的2%。这一趋势凸显了在2026年高度自动化的威胁环境中，传统安全框架已难以应对由生成式AI驱动的复杂攻击。解决方案的关键在于构建下一代综合性安全框架，核心包括：基于真实案例（如Google Play Protect、Microsoft Defender等平台实践）的多维度分析，融合零日漏洞检测、DevSecOps集成、联邦学习、合成内容识别与可解释AI（XAI）等前沿技术；同时提出以模型水印、对抗防御和跨行业协作为核心的可信AI部署机制，推动负责任、透明且可追溯的生成式AI应用，为实现安全、可扩展的智能安全系统提供系统性路径与基准参考。
链接: https://arxiv.org/abs/2607.06963
作者: Kiarash Ahi,Saeed Valizadeh
机构: Virelya Intelligence Research Labs; University of Florida; Google
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: Invited survey paper. 10 pages, 5 figures, 2 tables
Abstract:Large Language Models (LLMs) and generative AI (GenAI) systems, such as ChatGPT, Claude, Gemini, LLaMA, Copilot, Stable Diffusion by OpenAI, Anthropic, Google, Meta, Microsoft, Stability AI, respectively, are revolutionizing cybersecurity, enabling both automated defense and sophisticated attacks. These technologies power real-time threat detection, phishing defense, secure code generation, and vulnerability exploitation at unprecedented scales. Following a rapid surge where LLM-generated malware grew to account for an estimated 50% of detected threats by 2025, up from just 2% in 2021, navigating this highly automated threat landscape in 2026 demands next-generation security frameworks. This paper presents a comprehensive survey of the beneficial and malicious applications of LLMs in cybersecurity, including zero-day detection, DevSecOps, federated learning, synthetic content analysis, and explainable AI (XAI). Drawing on a review of over 70 academic papers, industry reports, and technical documents, this work synthesizes insights from real-world case studies across platforms like Google Play Protect, Microsoft Defender, Amazon Web Services (AWS), Apple App Store, OpenAI Plugin Stores, Hugging Face Spaces, and GitHub, alongside emerging initiatives like the SAFE Framework and AI-driven anomaly detection. We conclude with practical recommendations for responsible and transparent LLM deployment and trustworthy AI, including model watermarking, adversarial defense, and cross-industry collaboration, setting a new benchmark for rigorous, holistic cybersecurity research at the intersection of AI and threat defense, and offering a roadmap for secure, scalable LLM systems that serves as a critical reference for researchers, engineers, and security leaders navigating the complex challenges of AI-driven cybersecurity.
[NLP-31] Comprehensive Evaluation of Large Language Model Responses: A Multi-Factor Scoring System
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在语言任务中表现优异但缺乏全面评估体系的问题。现有评估方法多局限于单一维度，难以全面反映模型在准确性、连贯性、事实一致性等多方面的综合能力。其解决方案的关键在于提出一种多因素评分范式，整合准确性、简洁性、事实一致性、可读性和连贯性五个核心维度，并配套开发图形化用户界面（GUI），实现评估结果的可视化呈现。通过在TruthfulQA数据集上的实验，该框架揭示了主流LLMs在推理任务中的优势（综合得分最高达0.6104），同时也暴露其在处理复杂事实与语义模糊性方面的普遍局限。该方法突破了传统指标的狭隘视角，为模型能力的透明化分析与优化提供了可扩展、可适应的新路径，未来有望拓展至多语言领域，推动知识工程与模型精炼的发展。
链接: https://arxiv.org/abs/2607.06940
作者: Yiming Gai,Junde Lu,Xuefei Huang
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:The remarkable performance of large language models (LLMs) in linguistic tasks underscores an urgent need for comprehensive evaluation of their response quality. Prevailing methods, often confined to singular dimensions, fall short of capturing the full spectrum of model capabilities. This study introduces a multifactor scoring paradigm, integrating accuracy, conciseness, factual consistency, readability, and coherence, complemented by a graphical user interface (GUI) for visualizing outcomes. Evaluations on the TruthfulQA dataset unveil mainstream LLMs’ strengths in reasoning tasks (peaking at a composite score of 0.6104) alongside pervasive limitations in navigating complex facts and ambiguities. Transcending the narrow lens of traditional metrics, this framework offers a transparent, adaptable avenue to illuminate model potential and deficiencies. Though presently focused on English tasks, its horizons beckon toward multilingual domains. This work carves a novel path for knowledge engineering and model refinement.
[NLP-32] Geometric Self-Distillation for Reasoning Generalization
【速读】： 该论文旨在解决自蒸馏（self-distillation）在大语言模型中因教师与学生间信息不对称导致的分布外（out-of-distribution, OOD）推理性能退化问题。其核心挑战在于：在特权上下文自蒸馏中，教师模型因可见额外提示或完整解题轨迹而具备更强的预测能力，但学生模型无法获取这些信息，导致教师对某些生成路径的偏好难以被学生合理支持。这种不匹配在训练过程中产生强烈的蒸馏拉力，引发学生预测行为的系统性偏移（drift），最终损害模型在未见分布上的泛化能力。
该研究提出的解决方案关键在于提出一种几何自蒸馏（GeoSD）目标函数，从几何角度建模学生预测行为的演化，并通过双重机制对抗漂移：首先，采用Hellinger距离对教师偏好进行加权，依据学生当前对候选词的概率分布重叠程度动态调节梯度强度，从而抑制学生无法支撑的强偏好信号；其次，引入近端项（proximal term），以Fisher-Rao距离衡量学生预测与近期检查点之间的偏离程度，防止预测轨迹过度偏离历史稳定状态。两项损失均定义于下一个词分布的空间几何结构中，配合自然梯度更新（natural gradient update），使优化过程在概率分布流形上进行，而非传统参数空间。实验表明，GeoSD在数学推理基准上显著提升分布外准确率5.7–8.6个百分点，且在1.7B至32B不同规模模型中均保持有效，同时保留了分布内性能优势。分析揭示，传统匹配方法在分布外失效的根本原因在于其通过压缩高熵状态下的备选方案质量来强行达成与教师的一致，导致“自信地错误”，而GeoSD则通过维持备选路径的可及性，实现更稳健的推理。
链接: https://arxiv.org/abs/2607.06855
作者: Josip Jukić,Ivan Titov
机构: University of Amsterdam (阿姆斯特丹大学); University of Edinburgh (爱丁堡大学)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:
Abstract:On-policy distillation is a practical post-training recipe for large language models, supplying dense teacher supervision on the student’s own trajectories. In privileged-context self-distillation, teacher and student are the same model conditioned on the same prefix, but the teacher also sees a hint or the full solution trace. This makes supervision abundant but harder to trust: the teacher can be confident about continuations its privileged view makes obvious but the student cannot yet justify. The distillation pull is strongest where teacher and student disagree most, and over many updates it accumulates into drift that degrades out-of-distribution (OOD) reasoning. We introduce GeoSD, a geometric self-distillation objective that treats this drift as movement in the student’s predictive behavior and counters it in two complementary ways. A Hellinger loss scales each teacher preference by the overlap the student already shares with it, attenuating the pull on tokens the student cannot yet support. Since these pulls still compound over training, a proximal term penalizes how far the student’s predictions drift from a recent checkpoint, measured as a Fisher-Rao distance. Both are distances in the same geometry of next-token distributions, and a natural-gradient update takes its steps in that geometry rather than in parameter space. Across mathematical reasoning benchmarks and three model families, GeoSD preserves the in-distribution gains of self-distillation while improving average OOD accuracy by 5.7-8.6 points over the base model, with gains holding across model scales from 1.7B to 32B. Analyzing why standard matching fails out of distribution, we find it wins agreement with the teacher by draining mass from alternatives at high-entropy states, resulting in confident agreement on wrong answers, whereas GeoSD keeps those alternatives in reach.
[NLP-33] LLM s Silently Correct African American English: Auditing and Mitigating Dialect Bias via Activation Steering
【速读】： 该论文旨在解决大型语言模型（LLM）对非标准方言——非洲裔美国人英语（AAE）——存在的系统性偏见问题，即模型在面对以AAE为上下文的输入时，倾向于强制生成标准美国英语（SAE）的输出，从而错误地“纠正”或重构原生AAE表达。这一现象不仅削弱了对多元语言文化的尊重，还可能导致信息失真与文化代表性缺失。其解决方案的关键在于提出一个端到端的框架，包含两个核心部分：一是基于条件方言群体不变性（cDGI）的审计机制与特征级定位分析，能够有效区分模型真实偏见与翻译过程引入的伪影，并识别出触发偏见的核心语言特征（如否定一致结构“ain’t nobody”等句法构造）；二是首次将激活引导（activation steering）应用于方言偏见缓解，通过因果追踪提取方言特异性方向，并在推理阶段无须训练地注入至关键层，显著降低偏见水平（较提示工程提升5至20倍），同时保持SAE语用流畅性。此外，研究团队发布了目前最大的真实世界AAE平行语料库REAL-AAE，包含17,479组自然推文中的AAE/SAE/AAE回译三元组，具备高自动验证（BERTScore F1=0.95）与人工评估一致性（83.0%语义一致率），为后续研究提供了重要数据基础。
链接: https://arxiv.org/abs/2607.06845
作者: Huan Wu,Ali Emami,Muhammad Furquan Hassan,Osaretin Igbinoba,Osakpolor Idusuyi,Osamede Igbinoba,Faiza Khan Khattak,Laleh Seyyed-Kalantari
机构: York University (约克大学); Vector Institute (向量研究所); Connected Minds (连接心智); Emory University (埃默里大学); Wilfrid Laurier University (威尔弗里德·劳里埃大学); University of Toronto (多伦多大学); University of Guelph (圭尔夫大学); Monark Health (莫纳克健康); CIFAR Solution Network Member (加拿大高端研究理事会解决方案网络成员)
类目: Computation and Language (cs.CL)
备注:
Abstract:African American English (AAE), a rule-governed dialect spoken by over 30 million people, is routinely misinterpreted and “corrected” by large language models (LLMs). Across six instruction-tuned LLMs (14B to 70B), we show that state-of-the-art models systematically prefer Standard American English (SAE) continuations even when the preceding context is in AAE, effectively rewriting AAE into SAE. We present an end-to-end framework to audit and mitigate this bias. For auditing, we introduce conditional Dialect Group Invariance (cDGI), which isolates true model bias from translator-induced artifacts, and a feature-level localization analysis that identifies which AAE markers most strongly trigger bias; we find that syntactic constructions, especially negative concord (e.g., “ain’t nobody”), are universal triggers across all models. For mitigation, we introduce, to our knowledge, the first application of activation steering to dialect bias: a training-free, test-time method that extracts dialect directions via causal tracing and injects them into bias-relevant layers. Activation steering reduces bias 5 to 20 times more than prompting while preserving SAE fluency. To enable this work, we release REAL-AAE , the largest real-AAE parallel corpus to date: 17,479 AAE/SAE/ AAE_back triplets from natural tweets (2 to 6 times larger than prior real-AAE resources), validated automatically (BERTScore F1 = 0.95) and by three native AAE speakers (83.0% semantic agreement).
[NLP-34] Gradient-Based Speech-to-Text Alignment for Any ASR Model: From CTC to Speech LLM s
【速读】： 该论文旨在解决语音识别中词级时间对齐（speech-to-text alignment）的通用性与精度问题，尤其针对那些不直接提供对齐信息的模型（如基于注意力机制的编码器-解码器AED和语音大语言模型LLM）。现有方法多依赖于注意力权重或在编码器帧网格（encoder frame grid）上进行对齐，受限于帧粒度导致时间分辨率较低。其解决方案的关键在于提出一种通用的基于梯度的对齐方法：通过计算每个教师强制输入的输出词元对数概率相对于输入音频的梯度，将其归一化为每帧显著性（saliency），再通过一次动态规划（dynamic programming）遍历生成词边界。该方法无需训练、无需修改模型结构、无需额外对齐头，适用于所有可微分的自动语音识别（ASR）模型，包括语音LLM，并能在原始输入信号网格上实现高精度对齐，突破了编码器帧网格的限制。实验在四个模型家族的十六个模型上进行，涵盖读诵式（TIMIT）与非正式口语（Buckeye）数据，结果表明该梯度对齐方法对所有模型均有效，性能在多数情况下优于原生对齐（尤其是流式模型），仅需承担单次反向传播的计算开销作为主要代价。
链接: https://arxiv.org/abs/2607.06831
作者: Albert Zeyer,Ralf Schlüter,Hermann Ney
机构: University of Siegen (锡根大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Speech-to-text alignment means finding the temporal boundaries of each word in the audio. Some models provide such an alignment directly and others do not. Connectionist temporal classification (CTC) and transducer models have an alignment by construction, whereas attention-based encoder-decoders (AED) and speech large language models (LLMs) do not, and their word timings are usually read off the attention weights instead. All of these signals live on the encoder frame grid, which bounds their temporal precision. We study a generic gradient-based alignment that applies to any differentiable ASR model. We take the gradient of each teacher-forced token log probability with respect to the input, reduce it to a per-frame saliency, and decode the resulting matrix into word boundaries with a single dynamic-programming pass. The method needs no training, no model modification and no alignment heads, works across all model families including the speech LLMs, and aligns on the input grid rather than on the coarser encoder grid. We evaluate it on sixteen models from four families, on read (TIMIT) and spontaneous (Buckeye) speech, each against the model’s own native or attention-based alignment. We find that the gradient yields a usable alignment for every model, that it is usually somewhat behind a strong native aligner but better where the native alignment is weak, as for the streaming models, and that its main disadvantage is the cost of one backward pass per token.
[NLP-35] Ad Headline Generation using Self-Critical Masked Language Model NAACL ACL
【速读】： 该论文旨在解决电子商务网站在大规模场景下生成高质量、具有吸引力的广告标题这一难题，尤其针对如何持续产出符合平台创意质量标准的广告内容。其核心挑战在于传统方法难以在保证语法正确性和创意新颖性的前提下实现规模化生成。为此，论文提出一种基于强化学习（Reinforcement Learning, RL）策略梯度方法的程序化解决方案，将基于Transformer的掩码语言模型（Masked Language Model, MLM）与多产品联合条件建模相结合，实现对多个商品信息的协同生成。该方案的关键创新在于利用强化学习优化生成过程，使模型能够根据上下文语义和多商品特征动态调整标题生成策略，从而在重叠度指标和质量评估中显著优于现有的Transformer及LSTM+RL方法。此外，实验表明，模型生成的标题在语法准确性和创意质量方面均超越人工提交的标题，验证了其在实际应用中的有效性。
链接: https://arxiv.org/abs/2607.06818
作者: Yashal Shakti Kanungo,Sumit Negi,Aruna Rajan
机构: Amazon(亚马逊)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted at NAACL-HLT 2021 (Industry Track). 9 pages, 3 tables, 3 figures - ACL Anthology URL: this https URL - Editors of the proceedings: Young-bum Kim, Yunyao Li, Owen Rambow - Bibkey: kanungo-etal-2021-ad
Abstract:For any E-commerce website it is a nontrivial problem to build enduring advertisements that attract shoppers. It is hard to pass the creative quality bar of the website, especially at a large scale. We thus propose a programmatic solution to generate product advertising headlines using retail content. We propose a state of the art application of Reinforcement Learning (RL) Policy gradient methods on Transformer based Masked Language Models. Our method creates the advertising headline by jointly conditioning on multiple products that a seller wishes to advertise. We demonstrate that our method outperforms existing Transformer and LSTM + RL methods in overlap metrics and quality audits. We also show that our model-generated headlines outperform human submitted headlines in terms of both grammar and creative quality as determined by audits.
[NLP-36] rees from Marginals: Autoregressive drafting with factorized priors
【速读】： 该论文旨在解决生成式 AI（Generative AI）中自回归语言模型在实现高交互性时面临的效率瓶颈问题，具体表现为：尽管推测解码（speculative decoding）通过单次前向传播生成额外文本以提升响应速度，但因子化草稿模型（factorized draft models）依赖独立性假设预测未来词元的边缘分布，导致随着推测预算（speculative budget）增加，接受率急剧下降，从而限制了整体加速效果。其解决方案的关键在于提出 Weaver——一种轻量级自回归适配器，通过利用因子化草稿模型输出的 top-K 边缘分布构建提案树（proposal tree），在不进行全词汇投影的前提下恢复所提词元间的条件依赖关系，有效提升了推测接受率。同时，为支持具有门控增量网络（Gated Delta Net）结构的模型，研究设计了一种无需回滚的树验证算法，并在 SGLang 中实现了优化的 CUDA 内核，实现了高效的验证过程。综合模型与系统层面的创新，该方法相较于传统自回归解码实现 4.37 倍加速，且相较高度优化的 DFlash 基线提升 24.7%。
链接: https://arxiv.org/abs/2607.06763
作者: Yuma Oda,Ryan Mathieu,Roman Knyazhitskiy,Artur Chakhvadze
机构: 未知
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:
Abstract:Speculative decoding greatly increases the interactivity of autoregressive language models by trading off computation for extra tokens generated in a single forward pass. Factorized draft models are especially efficient because they predict future-token marginals in parallel, but their independence assumption causes acceptance rates to degrade sharply as the speculative budget grows. We analyze this limitation and introduce Weaver, a lightweight autoregressive adapter that constructs proposal trees from the top-K marginals of a factorized drafter. Weaver restores conditional dependencies between proposed tokens while avoiding a full-vocabulary projection. To support fast verification for models with Gated Delta Net layers, we derive a rollback-free tree-verification algorithm and implement optimized CUDA kernels in SGLang. By combining these model and systems contributions we achieve a 4.37-fold speedup over autoregressive decoding, and outperform a highly optimized DFlash baseline by 24.7%.
[NLP-37] When Does In-Context Search Help? A Sampling-Complexity Theory of Reflection-Driven Reasoning
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在复杂推理任务中零样本（zero-shot）表现不佳的问题，特别是如何通过迭代式自我修正机制提升模型的推理能力。其核心挑战在于理解并量化“上下文搜索”（in-context search）——即模型通过生成、批判与修订解题尝试来逐步优化答案的过程——在推理时的采样效率。论文的关键解决方案是将上下文搜索建模为对推理轨迹的近似贝叶斯推断：以基线模型定义先验分布，自省（self-reflection）提供反馈以更新后验分布。研究表明，当自省能够有效定位早期错误时，上下文搜索可实现相对于基线模型的指数级性能提升，即仅需多项式数量的序列尝试即可解决零样本成功率极低的问题；反之，若无法精确定位错误，则序列尝试相较于并行采样无渐近优势。进一步分析表明，这种收益具有鲁棒性且可通过训练学习：近似的后验更新即可实现显著效果，且基于搜索轨迹的交叉熵训练可在多项式样本复杂度下恢复所需行为。此外，论文在可验证奖励下的分阶段强化学习框架中证明，最优策略扩展等价于相同的后验重加权规则。实证部分验证了理论预测的关键定性结论。
链接: https://arxiv.org/abs/2607.06720
作者: Yotam Wolf,Noam Wies,Amnon Shashua
机构: The Hebrew University (希伯来大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Training large language models (LLMs) with extended reasoning has enabled in-context search, in which models iteratively generate, critique, and revise solution attempts. We provide a theoretical analysis of in-context search by modeling it as approximate inference over reasoning traces, where the base model defines a prior and self-reflection provides feedback for posterior updates, and study the resulting inference-time sampling complexity - the number of sequential attempts needed to achieve high success probability. We show that when reflections reliably localize early mistakes, in-context search can yield exponential improvements over the base model, solving problems with exponentially small zero-shot pass rates using only a polynomial number of sequential attempts, whereas when this property fails, conditioning on past attempts offers no asymptotic benefit over parallel sampling. We further show that these gains are robust and learnable: approximate posterior updates suffice, and cross-entropy training on search rollouts recovers the required behavior with polynomial sample complexity. Finally, we show that under a stagewise abstraction of reinforcement learning with verifiable rewards, the optimal policy extension implements the same posterior reweighting rule. We validate key qualitative predictions of the theory on real large reasoning models.
[NLP-38] Final Checkpoints Are Not Enough: Analyzing Latent Reasoning Faithfulness Along Training Trajectories
【速读】： 该论文旨在解决生成式模型中隐式推理（latent reasoning）方法在多步推理过程中存在的可解释性与因果可信性问题。尽管隐式推理通过在连续隐藏状态中完成推理，具有高效紧凑的优势，但其内部状态的不可见性引发了对推理步骤是否真正因果驱动最终答案的质疑。现有研究仅在训练收敛后评估可信性，发现存在推理步骤可被替换而不影响输出等不忠实行为，却未揭示这些行为如何在训练过程中演化形成。本文通过追踪不同隐式推理范式在多个保存检查点上的可信性演变，采用可验证的反事实编辑（counterfactual edit）和噪声消融激活补丁（noise-ablation activation patch）方法，系统分析了推理过程的因果贡献变化。研究发现：(i) 在输出层面，不同推理轨迹在收敛时均表现出相似的不忠实现象；(ii) 在激活层面，两类范式下的推理步骤对最终答案的因果贡献均随训练进程衰减，且输出翻转的案例恰好对应因果贡献显著下降的样本；(iii) 激活层面的演化路径因答案格式而异，二元选择任务中因果贡献持续下降，而开放式生成任务中则呈上升趋势。上述结果表明，隐式推理的可信性并非静态属性，而是高度依赖于训练阶段与输出格式，为理解隐式推理的动态演化机制提供了关键实证依据。
链接: https://arxiv.org/abs/2607.06648
作者: Hengyu Jin,Shu Yang,Di Wang
机构: Tongji University (同济大学); Provable Responsible AI and Data Analytics (PRADA) Lab; King Abdullah University of Science and Technology (阿卜杜拉国王科技大学)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:
Abstract:Latent reasoning methods perform multi-step inference entirely in the model’s continuous hidden states, promising more compact and efficient reasoning. However, these opaque hidden states raise a question of faithfulness: whether these latent reasoning steps causally drive the final answer. Prior work investigates this question at converged checkpoints and reports several unfaithful behaviors, such as latent reasoning steps that can be replaced without changing the answer, but leaves how these behaviors form during training unexamined. We instead track how faithfulness evolves across saved checkpoints for different latent reasoning paradigms, applying a verifiable counterfactual edit on the input and a noise-ablation activation patch on the latent reasoning steps. We find that (i) at the output level, latent reasoning methods can look similarly unfaithful at convergence under counterfactual edits while following qualitatively divergent trajectories; (ii) at the activation level, the causal contribution of latent reasoning steps to the final answer decays across training for both paradigms, with the examples that flip on the output side in (i) also being the examples on which this contribution decays; and (iii) the activation-level trajectory diverges by answer format, decaying on binary choice and rising on open-ended decoding. These findings highlight that latent reasoning faithfulness depends on training stage and answer format.
[NLP-39] Healthier LLM s: Retrieval-Augmented Generation for Public Health Question Answering
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在公共卫生问答任务中因幻觉（hallucination）及官方指导信息快速更新所导致的可靠性不足问题。其核心解决方案是采用检索增强生成（Retrieval-Augmented Generation, RAG）框架，通过将模型输出锚定在显式维护的权威知识库上，提升回答的准确性和时效性。研究的关键在于系统评估不同检索策略（密集型、稀疏型与混合型检索）在多种嵌入模型和语料变体下的表现，发现混合检索在召回率和排序质量上具有持续优势，且段落长度与主题特性显著影响检索效果；同时，通过精心选择的检索上下文可显著提升多选题准确率，使小型开源模型在性能上达到甚至超越大型闭源模型。为评估真实场景下的自由文本回答质量，研究引入基于评分标准的LLM作为裁判（LLM-as-a-judge）机制，涵盖忠实性、完整性、清晰度与事实一致性维度，并通过双人人工标注验证其有效性，结果显示忠实性与完整性一致性较高，而事实一致性和清晰度的可复现性较低，提示需谨慎解读相关指标。总体而言，研究强调检索环节是实现可靠公共卫生问答的核心杠杆，并为基于官方指南构建与评估RAG系统提供了实证依据与实践指导。
链接: https://arxiv.org/abs/2607.06641
作者: Felix Feldman,Joshua Harris,Timothy Laurence,Leo Loman,Ollie Higgins,Fan Grayson,Poonam Soma,Bethany Pace-Bonello,Michael Borowitz,Toby Nonnenmacher
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 19 Pages, 14 Main Text Pages, 6 Figures
Abstract:Large language models (LLMs) achieve promising results on medical question answering benchmarks, yet their use in public health is constrained by hallucinations and the rapid evolution of official guidance. Retrieval-Augmented Generation (RAG) mitigates these risks by grounding responses in an explicitly maintained corpus, but end-to-end performance depends critically on retrieval configuration and on evaluation beyond multiple-choice formats. We extend PubHealthBench, a question answering (QA) benchmark of 7,929 questions derived from UK Government public health guidance, into a retrieval-augmented setting and systematically evaluate retrieval and generation choices. We compare dense, sparse, and hybrid retrieval across multiple embedding models and corpus variants, and show that hybrid retrieval consistently improves recall and ranking quality, with chunk length and topic interacting with ranking performance. Providing retrieved context substantially increases multiple-choice accuracy across a diverse set of LLMs, enabling smaller open-weight models to match or outperform larger models used without retrieval, with gains primarily driven by retrieval quality and careful context selection. To assess realistic free-form answering, we introduce a rubric-based LLM-as-a-judge covering faithfulness, completeness, clarity, and factual consistency, and validate it against dual human annotations. Judge-human agreement is strongest for faithfulness and completeness, while factual consistency and clarity are less reliably reproduced, motivating caution when interpreting those dimensions at scale. Overall, our results highlight retrieval as a primary lever for reliable public health QA and provide practical guidance for building and evaluating RAG systems grounded in official guidance.
[NLP-40] Audio Sentiment Analysis via Distillation and Cross-Modal Integration of Generated Multilingual Transcripts
【速读】： 该论文旨在解决从语音中自动识别情感极性（正向或负向）这一挑战性任务，其核心难点在于需同时处理语音中的声调变化与语义内容。现有方法依赖音频基础模型，但其是否能全面融合多源信息仍不明确。为此，论文提出一种基于跨模态变换器的多模态解决方案，通过自动语音识别（ASR）工具生成文本转录，并利用机器翻译工具将转录文本自动翻译为多种语言以构建多语言文本模态，进而通过级联式架构逐层融合音频与多语言文本特征。关键创新在于：1）利用自动生成的多语言文本增强模型对语义的理解能力；2）通过知识蒸馏技术，将多模态“教师”模型的知识迁移至仅依赖音频的“学生”模型，从而在推理阶段无额外计算开销的前提下显著提升性能。实验结果表明，自动转录与自动翻译均对分类性能有显著贡献，且蒸馏后的单模态模型具备更强的泛化能力。
链接: https://arxiv.org/abs/2607.06611
作者: Andrei-George Durdun,Victor Constantinescu,Radu Tudor Ionescu
机构: University of Bucharest (布加勒斯特大学); PPC Romania (PPC 罗马尼亚)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Sound (cs.SD)
备注: Accepted at KES 2026
Abstract:Automatically recognizing the sentiment, positive or negative, from speech is a challenging task, requiring both the analysis of vocal inflections and the interpretation of uttered words. Recent solutions rely on audio foundation models to solve the task, but it remains unclear if such models can take all aspects into account. To this end, we propose a multimodal solution that integrates audio and text information via cross-modal transformers, where text transcripts are automatically generated via an automatic speech recognition (ASR) tool. Moreover, we create multiple text modalities by automatically translating the transcripts into multiple languages via machine translation tools. Audio and multilingual text features are combined via a cascaded architecture comprising cross-modal transformer blocks that integrate modalities one by one. We further distill knowledge from the multimodal model, called teacher, into a unimodal (audio only) model, called student. We conduct experiments on a large-scale dataset, demonstrating that the automatically generated textual information can bring significant performance boosts in multimodal sentiment polarity classification. Our ablation study confirms that both automatic transcripts and automatic translations are helpful. Moreover, we show that the audio-only model can be enhanced via distillation, boosting performance without any computational overhead during inference. To reproduce the reported results, we publicly release our code at this https URL.
[NLP-41] Recovering Latent Structures after Variational Bayesian Variable Selection: Fit Assessment and Factor-Number Selection in Partially Exploratory Factor Analysis
【速读】： 该论文旨在解决部分探索性因子分析（PEFA）中因子数量与因子载荷结构设定不明确的问题，其核心挑战在于如何在缺乏先验约束的情况下合理识别潜在因子结构并确定最优因子数。解决方案的关键在于采用基于贝叶斯变量选择的正则化变分近似方法（PCFA VA），通过引入尖峰-平板先验（spike and slab priors）为未指定的因子载荷分配包含概率，从而实现载荷结构的自动恢复。研究进一步提出一种后选择评估框架，将收敛解转化为协方差模型：一方面采用硬选择（阈值化概率以生成稀疏模式），另一方面采用软选择（保留概率作为权重以计算有效参数量）。在此基础上，推导出相应的自由度、绝对拟合指标（如RMSEA、SRMR、CFI、TLI）以及相对拟合准则（AIC、BIC、ELBO）。为确定因子数量，提出了一个无量纲的增益规则（scale-free gain rule），并引入持续下降保护机制以避免过拟合。仿真结果表明，绝对拟合指标能有效追踪载荷恢复情况并识别欠因子问题；尽管原始准则存在因子数量误判风险，但所提增益规则可准确恢复真实维度，其中基于ELBO的版本表现最为稳健。最后，在一个包含100个题项的PID-5量表实例中，该模型在拟合优度上优于传统的25个分面的验证性模型，并能在不同指定条件下一致地复现主要结构特征，验证了方法的有效性与稳定性。
链接: https://arxiv.org/abs/2607.07159
作者: Jinsong Chen,Yi Jin
机构: 未知
类目: Methodology (stat.ME); Computation and Language (cs.CL)
备注:
Abstract:In partially exploratory factor analysis (PEFA), the loading structure and factor numbers are weakly specified. The regularized variational approximation for partially confirmatory factor analysis (PCFA VA) recovers this structure via Bayesian variable selection, using spike and slab priors to assign inclusion probabilities to unspecified loadings. This research introduces a post selection assessment framework for this approach. We convert converged solutions into covariance models using either hard selection (thresholding probabilities into a sparse pattern) or soft selection (retaining them as weights for effective parameter counts). We derive the resulting degrees of freedom, absolute fit diagnostics (RMSEA, SRMR, CFI, TLI), and relative criteria (AIC, BIC, ELBO). To determine factor numbers, we propose a scale free gain rule with a sustained drop guard. Simulations show absolute indices successfully track loading recovery and flag under factoring. While raw criteria over factor, our gain rule accurately recovers true dimensionality, with the ELBO variant proving most robust. Finally, a 100 item PID 5 example demonstrates that our model fits better than a confirmatory 25 facet model and concordantly recovers major structures across disjoint specifications.
[NLP-42] Reconfigurable Radiology Labels Without Relabeling
【速读】： 该论文旨在解决公开胸部X线（CXR）数据集普遍采用固定且有限的标签体系（如CheXpert-14）所带来的表达能力不足问题，即自由文本报告中包含的丰富临床发现无法被现有标签体系充分覆盖，而这些发现的重要性又因任务、机构和阅片者而异。其解决方案的关键在于提出一种可复用的管道（pipeline），将自由文本报告转化为多标签矩阵，并通过词典编辑方式动态重构标签体系，而非依赖重新推理或人工标注整个语料库。该方法仅需一次预处理，后续标签体系调整可在极短时间内完成（如MIMIC-CXR数据集仅需196秒，无API成本），显著提升灵活性与效率。实验表明，在58标签分类体系下，43%的影像研究包含超出CheXpert-14范围的发现；基于新标签训练的图像探测模型不仅在共享目标上达到与CheXpert-14相当的性能（0.78 AUROC），还能有效识别专家评审的长尾罕见病灶。这表明，放射学标注工作的新范式应是：一旦报告结构化，标签体系即可作为可配置参数进行迭代优化，而非重复标注整个数据集。
链接: https://arxiv.org/abs/2607.06597
作者: Jean-Benoit Delbrouck,Dave Van Veen,Akash Pattnaik,Kalina Slavkova,Javid Abderezaei,Harris Bergman,Khan Siddiqui
机构: 未知
类目: Image and Video Processing (eess.IV); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Public chest-radiograph (CXR) datasets are typically released with small, fixed label schemas such as CheXpert-14. However, the underlying free-text reports describe far more findings – and which findings matter depends on the task, site, and reader. We release a pipeline that converts free-text reports into multi-label matrices and then reconfigures the label schema through dictionary edits rather than new inference passes, i.e., without relabeling the corpus. After this one-time pass, reconfiguring MIMIC-CXR (223K reports) from cached annotations takes 196 seconds with no API cost, compared to \ 6.6K for an equivalent relabeling pass with Claude Opus 4.7. Using a 58-label taxonomy, we show that 43% of CXR studies contain at least one finding outside CheXpert-14. Image probes trained on these labels match CheXpert-14 probes on shared targets while also reaching 0.78 AUROC on expert-reviewed long-tail labels that CheXpert-14 cannot represent. These results suggest a different unit of work for radiology labeling: once reports are structured, the label schema becomes a configuration to edit, not a corpus to relabel.
信息检索
[IR-0] InductWave: Inductive Multi-Hop Logical Query Answering on Knowledge Graphs
链接: https://arxiv.org/abs/2607.07422
作者: Mayank Kharbanda,Michael Cochez,Rajiv Ratn Shah,Raghava Mutharaju
类目: Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
备注: Under Review at TKDE
Abstract:Logical Multi-Hop Query Answering over Knowledge Graphs (KGs) can be formulated as querying, with an implicit completeness assumption. Current works mainly focus on Existential First Order Logic (EFO) queries. These EFO queries contain conjunction, disjunction, and negation operators. Most existing works employ transductive reasoning, meaning they are not capable of reasoning over entities unseen during training. In the real world, there is a resource scarcity, and we cannot train a model with all the nodes of a large KG. Hence, we propose InductWave, a wavelet-based inductive embedding method for logical query answering on large KGs. Here, the training graph consists of fewer nodes than the test graph. Our model performs on par with the baseline models while having half the number of message-passing layers. It outperforms all of them in most cases, with 75% of the layers. These fewer resource requirements enable us to evaluate InductWave on massive graphs, such as Wiki-KG. We test our model using extensive experiments across varying train-test graph proportions of the FB15k-(237) dataset, comparing it with the state-of-the-art models. The code and datasets for the model are available at this https URL.
[IR-1] Interpretable Uncertainty for Adaptive Retrieval and Reasoning in Question Answering
链接: https://arxiv.org/abs/2607.07380
作者: Ritajit Dey,Iadh Ounis,Graham McDonald
类目: Information Retrieval (cs.IR)
备注: 2 pages, 1 figure
Abstract:Large language models (LLMs) achieve a strong performance in question answering (QA), but remain prone to hallucinations and suffer from limited transparency. Retrieval-augmented generation (RAG) can improve factuality, yet decisions about when and how to retrieve from external resources are typically based on opaque policies or computationally inefficient multi-step prompting procedures. We propose an uncertainty-aware framework for adaptive QA based on explicit signals derived from LLM internal representations. We distinguish between knowledge insufficiency and knowledge ambiguity or conflict, and efficiently estimate these from hidden states in a single forward pass. These signals guide system behaviour: RAG is triggered when knowledge is insufficient, while additional reasoning is applied when ambiguity or conflict is high. By grounding adaptive decisions in decomposed and efficiently estimable uncertainty signals, this approach provides a transparent and practical alternative to existing retrieval and reasoning strategies supporting the design of interpretable user-facing tools.
[IR-2] Granularity in Actoin: Graphing sources for social history
链接: https://arxiv.org/abs/2607.07183
作者: Sofus Landor Dam,Johan Heinsen
类目: Information Retrieval (cs.IR)
备注:
Abstract:This working paper describes a pipeline for turning historical sources into structured data organised around the principle of foregrounding action as the basic and constitutive unit of analysis. It is rooted in a desire for pipelines that suit a granular approach to social history. The pipeline rests on the principles developed in the GRAM-framework (Graph of Roles and Actions Model), but leverages a range of machine learning tools to allow for an automated, skeletal graphing of actions. Ideally, such auto-GRAMS would integrate with close readings, including extensive manual graphing. Finally, we provide an example of how this approach might work in practice by graphing actions of pretending across four separate archival collections, relating to runaways and itinerants in eighteenth and nineteenth-century Denmark.
[IR-3] Seeing and Reflecting: Multimodal Memory-Enhanced Agent Collaboration for Recommendation
链接: https://arxiv.org/abs/2607.07108
作者: Hao Cong,Huizu Lin,Zihan Wang,Chengkai Huang,Quan Z. Sheng,Lina Yao
类目: Information Retrieval (cs.IR)
备注:
Abstract:Large language model (LLM)-based agentic recommender systems show promise in modeling user preferences through natural-language reasoning, yet they remain limited by text-centric inputs and coarse-grained memory updates, making agents prone to missing visual evidence, semantic noise, and preference drift. To address these limitations, we propose MMEACR, a Multimodal Memory-Enhanced Agent Collaboration framework for recommendation. MMEACR introduces a dual-track memory architecture that separates interpretable agent reasoning from fine-grained multimodal matching. In the reasoning track, collaborative User and Item Memory Agents maintain persistent multimodal memories and update them through an attribute-guided reinforcement-and-reflection mechanism. In the matching track, a decoupled multi-modal embedding memory is built from raw interaction narratives and item images to preserve detailed cross-modal signals beyond structured memory updates. The two tracks are integrated through weighted Reciprocal Rank Fusion to produce robust and interpretable rankings. Experiments on three real-world domains show that MMEACR achieves strong overall performance against competitive LLM-based and agent-based baselines, with notable gains in visually grounded recommendation scenarios.
[IR-4] When and How to Ask: Dynamic Preference Elicitation Strategies for Conversational Recommendation SIGIR2026
链接: https://arxiv.org/abs/2607.06765
作者: Feng Xia,Shuo Zhang,Xi Wang
类目: Information Retrieval (cs.IR)
备注: Accepted at SIGIR 2026
Abstract:Conversational Recommender Systems (CRSs) are interactive systems that use multi-turn natural language dialogue to understand evolving user preferences and provide personalized recommendations. To achieve this goal, CRSs rely on preference elicitation strategies to actively gather informative preference cues from users; however, the timing and selection of these strategies during a conversation remain largely unexplored. While many existing studies emphasize eliciting explicit item attributes and tend to adopt relatively static elicitation strategies, the use of item-based preference elicitation and how it varies across different dialogue stages remains less explored. In this work, we conduct a systematic investigation of preference elicitation strategies from a stage-aware perspective. We provide empirical evidence that optimal preference elicitation strategies are stage-dependent and context-sensitive: attribute-based inquiries are effective in early stages, while item-based strategies become superior as preferences refine. To support this paradigm, we introduce InPE, a dataset enriched with fine-grained annotations for elicitation necessity and strategy selection. With this dataset, we propose COPE (COnversational Preference Elicitation via Mixture of Experts), a novel architecture for strategy modeling. Extensive offline evaluation on our dataset indicates that context-aware preference elicitation strategies are beneficial for conversational recommendation. In addition, the analysis of the predicted strategies uncovers consistent stage-wise tendencies in dialogue progression, providing empirical evidence of common interaction patterns in conversational recommendation systems. Our dataset is available at this https URL.
人机交互
[HC-0] Modeling Failure Dynamics in Time-Constrained Authentication Systems: Evidence of a Success Cliff in USSD Workflows
链接: https://arxiv.org/abs/2607.07650
作者: Aklile Seyoum Mamo,Amanuel Kebede,Anny Christelle Irakoze,Jema Ndibwile
类目: Cryptography and Security (cs.CR); Human-Computer Interaction (cs.HC)
备注: 8 pages, 6 tables, 5 figures, 12 equations, IEEE International Conference on Cyber Security and Resilience (CSR)
Abstract:Time-constrained interactive systems such as USSD (Unstructured Supplementary Service Data)-based financial services operate under strict session limits and sequential user interaction. While stronger authentication mechanisms improve security, they also increase interaction complexity and time burden, potentially reducing transaction completion. In this work, we model the failure dynamics of such systems and investigate how authentication complexity interacts with user response time and network round-trip time to influence session success rate. We propose and implement a simulation-based framework to investigate these failure dynamics and formally define a non-linear failure phenomenon, termed the \textitSuccess Cliff, where session success rates sharply decline beyond a critical complexity threshold. Through controlled experiments, we quantify the trade-off between security and usability and identify conditions under which secure authentication workflows become operationally unreliable.
[HC-1] wo-player Alternate Uses Test: A Controlled Testbed for Interactive Human-AI and Human-Human Co-Creation
链接: https://arxiv.org/abs/2607.07522
作者: Babak Hemmatian,Anita Keshmirian,Yijun Lin,Shravan Ramamoorthy,Maryam Jahadakbar,Eli Khuri-Reid,Jingtong Wang,Sarah Hadjarab,Sindre Veum,Pranav Gupta,Deepak Somaya,Lav R. Varshney
类目: Human-Computer Interaction (cs.HC)
备注: To appear in ACM Creativity and Cognition 2026
Abstract:Controlled research on AI ideation typically compares independent agents, while field studies of human-AI collaboration sacrifice experimental control. We introduce a controlled, two-player extension of the Alternate Uses Test (AUT) that enables comparison of human-human and human-AI co-creation under matched interactive conditions, alongside calibrated non-interactive baselines. The platform supports decomposition of performance into three typically confounded factors: participant traits, partner perceptions, and content dynamics. An in-person pilot (N = 62) demonstrates its utility. Under matched time limits, originality with a GPT-4 partner is statistically equivalent to that with a human partner. Approach motivation (BAS Drive) moderates whether interactive partnership benefits originality, and self-reported cognitive outsourcing predicts lower originality specifically in human-human dyads. Prior exposure to highly creative ideas improves later performance, suggesting a “seeding” intervention. We release the platform, code, and dataset as a shared testbed for controlled studies of human-AI co-creation.
[HC-2] Creativity from Friction: Human-AI Interaction for Exploratory Structural Design ICML2026
链接: https://arxiv.org/abs/2607.07521
作者: Ricardo Maia Avelino,Rita Sevastjanova,Tom Van Mele,Philippe Block,Mennatallah El-Assady
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注: Accepted at ICML 2026, Workshop on Human-AI Co-Creativity
Abstract:AI agents that generate final answers based on user input often do not meet the needs of creative fields. Fields such as structural design and architecture need interactive systems that help users externalise and develop ideas, explore alternatives, and refine partial solutions. The final product of such designs needs to comply with many constraints concerning, e.g., spatial configuration, mechanical behaviour, material quantities, and costs. These constraints create friction in the design process, which can stimulate novel and creative solutions. In this paper, we discuss the misalignment between current generative AI goals to remove friction and provide final solutions and the needs of creators, such as structural designers, who develop ideas through iterative work. We present the design dimensions of systems allowing for constrained human-AI co-creation that rely on vision-language models making structural exploration conversational, multimodal, and responsive to evolving human intent in ways that follow and augment the discipline’s creative process. Through a pilot design interface based on these principles and a study with experts in the field, this paper shows how structural designers perceive interactive AI systems and how such systems can support design space exploration by reducing repetitive modelling friction while preserving reflective design friction.
[HC-3] Should We Dangle a Carrot? The Effect of Performance-based Incentives in Visualization Experiments
链接: https://arxiv.org/abs/2607.07463
作者: Abhraneel Sarma,Matthew Kay,Sheng Long,Michael Correll,Alexander Lex
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:A perennial research question in visualization involves identifying which visual encodings for a particular dataset are most effective for users in performing a specific task. The relative effectiveness of the different encodings are commonly identified through controlled experiments. However, designing an experiment involves making many, often ad hoc, decisions about the experimental setup such as whether to include a training module, whether to provide performance-based incentives to participants, etc. Yet, there is limited guidance on how these decisions should be made, and we do not fully understand the impact of these subjective decisions on empirical results. In this paper, we investigate the impact of one such key design decision: monetary rewards. Specifically, we ask: does providing or not providing participants with performance-based financial incentives affect the results and the conclusions that we draw from visualization studies? We conducted two crowdsourced studies investigating the impact of incentives on (i) a low-level, perceptual task (perception of correlations in scatterplots or parallel coordinate plots), and (ii) a task involving reasoning (decision-making based on a weather forecast represented as intervals or density plots). In each of these studies, we manipulate both the visual representation and the presence of incentives as between-subject conditions. We expected to find no effect of incentives on the perceptual task, but to see an effect for the decision-making task. However, we found no effect on task performance in either study. While these are results of only two studies and should be replicated, they suggest that performance-based financial incentives may not always have the intended effect on participants that we presumed, and calls for a reflection of how incentivized studies should be designed.
[HC-4] Initiation Safety: A Missing Dimension in Generalist-Robot Safety
链接: https://arxiv.org/abs/2607.07420
作者: Zhijin Meng,Francisco Cruz
类目: Robotics (cs.RO); Human-Computer Interaction (cs.HC)
备注: 4 pages, 2 figures. Accepted to RSS 2026 Workshop on Rethinking Safety for Generalist Robots
Abstract:Safety for generalist robots is usually discussed in terms of motion or dialogue. We argue a third question is missing: should the robot take its first hard-to-undo social action at all, such as a greeting, an uninvited grasp, or stepping into someone’s space? We call this initiation authorization. Current frameworks rarely treat it as a separate safety layer. Today’s stacks often skip this step: a high engagement score or a confident VLA rollout is treated as permission to act. But seeing a person is not the same as having their consent to be addressed. We frame initiation authorization within generalist-robot safety and contrast it with post-plan VLA guardrails, implementing PAS (probe-authorize-speak) on a doorway humanoid, comparing it with direct-init on logged traces, and proposing a three-condition user study, with open questions on metrics, governance, and where initiation ends and foundation-model generation begins.
[HC-5] Behavior Foundations for Quadruped Robots: ABot-C0 Technical Report
链接: https://arxiv.org/abs/2607.07370
作者: Xufeng Zhao,Fuzhi Yang,Jianhui Chen,Li Gao,Zhang Meng,Jie Gao,Yao Zheng,Wenyu Liu,Menglin Yang,Minqi Gu,Yaru Zhao,Honglin Han,Shihui Su,Zixiao Tang,Liu Liu,Mu Xu,Yang Cai,Wenbin Tang
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC); Machine Learning (cs.LG)
备注: Abot-C0 project page will be released later
Abstract:In embodied intelligence systems, the motion controller serves as the critical bridge between semantic reasoning and physical execution. Humanoid control has progressed rapidly through large-scale human motion-capture data and motion-tracking paradigm. However, producing quadruped robots motion corpora with scalability and physical feasibility faces more fundamental obstacles: animal motion data is scarce, and cross-embodiment retargeting remains fragile. We present ABot-C0, a generalist motion-control system for quadruped robots that establishes three complementary behavior foundations: a scalable multi-source motion-data pipeline, robust policy learning across motion tracking, locomotion, and scene interaction, and a unified deployment stack for reliable real-world operation. Fundamentally, we construct a data pyramid through conditional video-generation synthesis, annotated motion capture, teleoperation and human design, producing 16,074 physically feasible motion clips as the data foundation for various motion learning demands. We then train a Flow-Matching generalist policy that demonstrates for the first time quadruped motion tracking scaling law that its performance improves consistently as training scales up, with zero-shot capability to track unseen motions. Then, we push a step further for robust all-terrain traversal locomotion by adopting a three-stage privileged-to-perceptive framework with temporal LiDAR memory and terrain-predictive supervision. Collectively, these components form a motion generalist that coordinates multi-policy execution, smooth behavior transitions, energy-efficient control, and safety mechanisms for real-world deployment. Extensive experiments on urban-terrain autonomous navigation and companion-style multimodal interaction demonstrate that quadruped robots move beyond single-function demos toward product-level behavioral intelligence.
[HC-6] Clinical Translation of Brain-Computer Interface in China: A Landscape Analysis of Investigator-Initiated Trials Registered Clinical Trials and Regulatory Approval
链接: https://arxiv.org/abs/2607.07185
作者: Long Chen(1),Wanyi Qing(1),Lifen Mo(1 and 2),Xiaoke Chai(3),Wenting Li(4),Fengyan Liang(1),Ming Yin(1),Yi Yang(5),Jizong Zhao(5) ((1) State Key Laboratory of Digital Medical Engineering, Key Laboratory of Biomedical Engineering of Hainan Province, School of Biomedical Engineering, Hainan University, Sanya, Hainan, China (2) School of Information and Communication Engineering, Hainan University, Haikou, Hainan, China (3) Brain Computer Interface Transitional Research Center, Beijing Tiantan Hospital, Capital Medical University, Beijing, China (4) W. P. Carey School of Business, Arizona State University, Tempe, AZ, USA (5) Beijing Tiantan Hospital, Capital Medical University, Beijing, China)
类目: Human-Computer Interaction (cs.HC)
备注: 15 pages, 5 figures, 6 tables
Abstract:Neurological injury affects hundreds of millions of people worldwide, yet the loss of motor or communication functions resulting from stroke, spinal cord injury, and neurodegenerative disease remains largely irreversible with existing therapies. Brain-computer interfaces (BCIs) offer a promising pathway for restoring these functions by decoding neural activity into commands that control an external device. Here, we present the first quantitative analysis of China’s BCI translational ecosystem, integrating evidence from three pillars: investigator-initiated trials (IITs), registered clinical trials, and regulatory-approved products. We analyzed 134 clinical trials from the Chinese Clinical Trial Registry (ChiCTR), 26 IITs, and five BCI-related products approved by the National Medical Products Administration as of June 2026. Results demonstrate that clinical trial registration has increased rapidly since 2020, with research centers concentrated primarily in Guangdong, Shanghai, and Jiangsu. Non-invasive systems predominated, accounting for 79.1% of registered studies, with stroke rehabilitation as the leading indication (65.0%). As of June 2026, five BCI-related products received regulatory approvals, including the world’s first approved semi-invasive implantable BCI, an invasive closed-loop deep brain stimulation system with real-time local field potential recording, and three non-invasive EEG-based rehabilitation systems. Collectively, these findings characterize a rapidly expanding BCI translational pipeline in China, spanning from early clinical research to regulatory approval. However, long-term implant stability, standardization of clinical infrastructure and workflows, and generalizability of decoding algorithms remain critical barriers to widespread clinical adoption. Addressing these challenges will be essential for integrating BCI technologies into routine clinical practice.
[HC-7] Bringing robustness to end-user programming
链接: https://arxiv.org/abs/2607.07116
作者: Mickaël Baron(LISI (Poitiers)),Patrick Girard(LIAS (Poitiers), IDD)
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:In some cases, end-user programming allows the design of stand-alone applications. But none of the existing approaches is concerned by safety aspects of programming. Heavy techniques exist to develop safe applications, particularly in non-interactive domains. They involve software engineering techniques, and sometimes, formal methods. All these techniques are very far from end-users. Our idea is to let this part to experts, and to connect end-user programming onto this safe conventional development. Starting from an existing functional core, we built an interactive end-user programming environment called GenBuild, which allows designing interactive stand-alone applications. GenBuild is composed of two distinct modules. The Generator is the first one. It is a specialized tool developed for a domain expert who sets out a safe functional core. The Builder is the second module. It is a purely interactive tool that allows an end-user to develop some complete interactive application among an existing functional core. It allows the verification of some properties that are a first step towards the development of safe end-user programming.
[HC-8] CompoVista: A Composition-Graph-Based Visual Analytics System for Compositional Analysis of Traditional Chinese Paintings
链接: https://arxiv.org/abs/2607.07105
作者: Dekun Qian,Ruiqi Yu,Fengling Zheng,Li Ye,Yize Li,Weigui Zheng,Yigang Wang,Jinchang Li,Zhiguang Zhou
类目: Human-Computer Interaction (cs.HC); Graphics (cs.GR)
备注:
Abstract:Composition in Traditional Chinese Paintings (TCPs) carries spatial, narrative, and cultural-aesthetic meaning. Systematic compositional analysis is therefore important for understanding their visual language and artistic meaning. Traditional compositional analysis is mainly qualitative and interpretation-driven. It supports close reading of individual paintings, but it is difficult to discover, compare, and verify compositional patterns across large painting collections. To better understand these challenges, we conducted a literature review and in-depth interviews with two art historians. Based on these findings, we introduce the Composition Graph, a scene-graph-based representation for TCP composition. It models a painting through four layers: entities, relations, void space, and context. Based on this representation, we develop CompoVista, a canvas-based visual analytics system for composition-oriented exploration of TCPs. CompoVista allows art historians to construct and revise format-aware painting cohorts through visual queries and context queries. It also supports cohort-level inspection of entity distributions and relations, comparison of compositional differences across cohorts, and tracing aggregate patterns back to painting-level this http URL evaluated CompoVista through a task-based user study with 12 domain participants, two case studies, and expert interviews. The results show that CompoVista supports composition-oriented cohort construction, pattern discovery, iterative refinement, and evidence inspection. The evaluation also reveals future needs, including clearer result explanations, fuzzier composition queries, and stronger exploration history management. Our work contributes a composition-specific structured representation and an integrated visual analytics workflow for studying TCP composition at collection scale.
[HC-9] Video-Based Detection of squint and cataract for accessibility-aware adaptive web interface rendering
链接: https://arxiv.org/abs/2607.07099
作者: Amar Ranjan Dash,Manas Ranjan Patra
类目: Computer Vision and Pattern Recognition (cs.CV); Human-Computer Interaction (cs.HC)
备注: International Journal of Computer Science, Engineering and Applications (IJCSEA), Vol. 16(3), 18 page 8 Figure, 2 Table
Abstract:Squint and cataract are major ocular disorders that majorly affect visual perception and interaction capability. This paper proposes a real-time video-based automated detection system for squint and cataract detection based on computer vision and image processing methods. The proposed system uses a media-pipe face-mesh (a 478-point facial landmark detection model) to extract geometric ocular features for multi-class squint classification. Simultaneously, The presence and severity cataract is estimated through grayscale intensity and histogram-based lens opacity analysis. The system records short video sequences with standard laptop or mobile cameras, which can be deployed at low costs and on a large scale. The experimental performance has shown great accuracy in the detection of squint (98.39%) and classification of cataract (96.90%). Besides automatic ocular analysis, the proposed framework is also made accessible for visual impairment inference which will be integrated with future adaptive user interface and Web accessibility systems for people with visual impairment.
[HC-10] ShapeTalk: Combining Natural Language and Sketch for Time-Series Pattern Querying
链接: https://arxiv.org/abs/2607.07073
作者: Guoruizhe Sun,Yueqiao Chen,Emily Guo,Yutong Yao,Dongyu Liu
类目: Human-Computer Interaction (cs.HC)
备注: 10 pages of main text, 2 pages of references, and 20 pages of appendices
Abstract:Searching for time-series segments that match user-defined patterns is important in domains such as finance, climate science, and healthcare. However, existing visual query tools often struggle to support vague, composite, or fuzzy pattern descriptions, often requiring users to express their intent through precise sketches or rigid structured filters. We present ShapeTalk, a coordinated natural-language and sketch-based querying system for univariate time-series pattern search. Rather than treating text and sketch as a fused input stream, ShapeTalk uses them as complementary representations of analytic intent: natural language supports semantic and compositional pattern descriptions, while sketching supports direct geometric refinement. The two modalities are linked through a shared visual context, editable feature representations, and synchronized result views, enabling users to move between text and sketch during iterative query formulation. At its core is an LLM-based semantic parsing pipeline that translates free-form natural-language queries into interpretable and editable shape-feature constraints. We evaluate ShapeTalk through two usage scenarios, a user study with failure-case analysis, and an assessment of the LLM-based semantic parsing pipeline. The results show that ShapeTalk supports effective time-series pattern search, with natural language serving as an accessible entry point and sketching providing a complementary mechanism for refinement and recovery when textual specifications are insufficient.
[HC-11] Learning social norms enhances compatibility in dynamic human-AI coordination
链接: https://arxiv.org/abs/2607.07021
作者: Yi Yang,Siyuan Liu,Xin Gao,Huamu Sun,Chao Liu,Qing Zhou,Bingbing Nie
类目: Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注: 44 pages, 5 figures, supplementary information included
Abstract:Humans continuously coordinate with others in dynamic interactions, often through implicit, hard-to-quantify social norms that act as shared tacit expectations among interacting agents. As AI agents, including large language models (LLMs), become embedded in daily life, they increasingly participate in such interactions and reshape social interaction structures. Yet they often fail to coordinate with humans in an effective, considerate, and natural manner. We hypothesize that this gap arises because existing approaches align model behavior with human demonstrations without explicitly quantifying the underlying norms that generate such behavior. We selected pedestrian-vehicle interaction as a representative dynamic interaction and developed a simplified experimental platform that captures its key interactive features. From 3,456 dynamic human interactions collected via this platform, we identified three principles underlying human social norms: outcome predictability, value alignment, and advantage awareness. Incorporating these principles into AI agents significantly improves human-AI coordination. In the closed-loop interaction task with humans, the social-norm-informed LLM achieved a nearly fourfold higher total score than the baseline strategy and outperformed human-human interactions by 43%. These findings indicate that formalizing tacit social norms into explicit, quantifiable principles can enable AI agents to achieve mutually beneficial coordination in dynamic interactions, supporting their more natural integration into human society.
[HC-12] Multimodal Smart Glove for Sign Language Recognition Using Deep Learning
链接: https://arxiv.org/abs/2607.06996
作者: Anh Thu Nguyen Ngoc,Tam Phong Truong,Thai Anh Nguyen Duong,Vu Linh Nguyen,Manh Duong Phung
类目: Human-Computer Interaction (cs.HC)
备注: In Proceedings of IFToMM International Symposium on Robotics and Mechatronics
Abstract:Sign language recognition technologies can improve communication between deaf individuals and the broader community, but many existing systems face challenges in real-world deployment. This paper presents a deployable smart glove system for sign language recognition that integrates wearable sensing and deep learning. The glove incorporates flex sensors and an inertial measurement unit (IMU) to capture finger articulation and hand motion, while facial cues are obtained through a camera. Sensor data are transmitted via an ESP32-C6 microcontroller and processed using a long short-term memory (LSTM) network to model temporal gesture dynamics. Experimental results show that the proposed model achieves an overall recognition accuracy of approximately 95%. The trained model is further converted to TensorFlow Lite for real-time inference. This demonstrates the feasibility of the system for practical sign language translation applications.
[HC-13] New Cross-Sensory Approach to Designing Restorative Virtual Environments
链接: https://arxiv.org/abs/2607.06901
作者: Rachel Masters,Francisco Ortega
类目: Human-Computer Interaction (cs.HC)
备注: 5 pages, ACM CHI 2026 Cross-Sensory Futures Workshop
Abstract:Virtual reality (VR) nature immersion is an increasingly popular field of research due to its potential to help people who do not have access to real nature. There are many questions surrounding how virtual forests can be designed to effectively reduce stress and restore attention. Many of these questions relate solely to visual aspects, but more recent literature has started exploring multisensory experiences. In these experiences, senses are treated as additive; however, certain results from the current literature may indicate that there are more complex, cross-sensory interactions occurring. For example, adding sound to visuals can increase stress reduction potential, but certain natural sounds can feel threatening if they are out of place within the virtual nature scene. Overall, cross-sensory interactions in VR nature environments (VNEs) are underexplored and challenge our current understanding of multisensory VNEs, and future explorations of these interactions are essential for designing optimal VNEs for stress reduction.
[HC-14] Head Gaze or Finger? Comparing Object Selection Techniques in Augmented Reality for People with Low Vision
链接: https://arxiv.org/abs/2607.06778
作者: Ruijia Chen,Tianyi Zhang,Sanbrita Mondal,Yukang Yan,Yuhang Zhao
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Augmented reality (AR) can enhance visual perception for people with low vision (PLV) by overlaying multimodal information. Selection-based augmentation further allows users to flexibly choose and augment relevant information while reducing distraction and visual clutter. However, little is known about the ability and preferences of PLV in performing object selection techniques in AR, considering their potential visual and gaze control challenges. To understand what selection techniques are suitable for PLV to support selection-based AR augmentations, we conducted a mixed-methods study with 20 PLV and 18 sighted controls who performed target selection tasks using three input techniques – head, gaze, and finger pointing with dwell-based confirmation – in two real-world scenarios (sitting vs. on the go). We found that for PLV, gaze-based selection enabled the fastest initial pointing when sitting and comparable overall selection time to head-based selection in both scenarios; however, due to reduced gaze stability, head-based selection remained the most stable and the least mentally demanding. Uniquely, participants with central vision loss preferred finger-based selection, reporting a greater sense of control. Our results provide empirical insights into accessible AR interaction techniques and selection-based vision enhancements for PLV.
[HC-15] Devising Interactive Spaces: A Rehearsal-Oriented Tool for Creating Responsive Environments for Immersive Theatre
链接: https://arxiv.org/abs/2607.06761
作者: Pavlos Panagiotidis,Paul Tennent,Jocelyn Spence,Nils Jaeger
类目: Human-Computer Interaction (cs.HC)
备注: Accepted as a technical demonstration paper at Creativity and Cognition 2026
Abstract:We present a rehearsal-oriented system for creating responsive built environments during theatre devising workshops. The system connects bespoke sensing modules for gesture, position, and speech recognition to light and sound outputs through a visual no-code programming layer. It was developed, used, and refined across six workshops with eight professional performance-makers, where participants created light-and-sound scores, gesture- and position-triggered scenes, responsive architectures, participatory prototypes, and a multi-room scratch performance. Rather than presenting a production-ready show-control platform, this demo focuses on how sensing and actuation can be made available as compositional materials during early-stage creative experimentation for immersive theatrical compositions. The system is designed to support quick configuration, visible mappings, and in-room testing, allowing performers to experiment with responsive spaces with minimal technical support. We describe the system architecture, its workshop use, and the practical conditions that helped integrate interactive sensing into embodied performance-making.
[HC-16] Creating a Mixed-Reality Installation with Families through Theatrical Co-Design
链接: https://arxiv.org/abs/2607.06754
作者: Pavlos Panagiotidis,Roma Patel,Boriana Koleva,Steve Benford,Jocelyn Spence,Paul Tennent,Juan Pablo Martinez Avila,Laurence Cliffe
类目: Human-Computer Interaction (cs.HC)
备注: Accepted as a poster paper at Creativity and Cognition 2026
Abstract:Co-designing with families for environmental sustainability relies on participatory imagination, yet habitual family roles and uneven participation, especially between adults and young children, often constrain it. A second challenge is continuity: workshop relationships and embodied ways of working do not easily survive into the final design, where artefacts travel more readily than roles or interactional dynamics. We report on a nationally toured mixed-reality installation developed through applied-theatre-led co-design with families. Across three workshops and user testing, applied theatre methods supported families to co-create narratives, artefacts, and interactional roles that shaped the public event. We show how theatrical co-design can rebalance child-adult participation through playful status shifts, and how selected workshop dynamics can be re-staged within a public mixed-reality installation. We contribute a theatrical account of participatory design in which designers work not only with artefacts and ideas, but with roles, rhythms, authority, and the social conditions that support collective imagination.
[HC-17] Flowcode: An AI-Powered Programming Environment for Scaffolding Iteration in Creative Computing Education
链接: https://arxiv.org/abs/2607.06721
作者: Tiffany Tseng,Liliana Hanem Seoror,Jeevika Adda,Meitalia Factor,Rona Darabi,Kiley R Matschke,Tiffany Fu,Annie Lin,Alekhya Maram,Arya Sinha
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Building upon found examples is a popular way people learn to code, especially in creative coding communities where sharing projects and remixing are common practices. But effectively doing so requires being able to 1) understand how existing code works, and 2) extend it by writing code that implements your own ideas, practices that can be challenging for new creative coders. We explored how to support these two processes through the design of Flowcode, a creative coding programming environment that integrates a flowchart for visualizing code structure and a chat interface tailored to support learning to code over vibe coding. We share how we iterated on the design of Flowcode over two studies with new creative coders, reflecting on the roles visualization and friction may play in enabling productive AI-use in computing education.
[HC-18] Exploring the Interaction of Explanation Styles Context and Trust of AI Privacy Redaction in AI-mediated Interactions
链接: https://arxiv.org/abs/2607.06687
作者: Roshni Kaushik,Maarten Sap,Koichi Onoue
类目: Human-Computer Interaction (cs.HC)
备注: Originally submitted to UIST, will be resubmitting
Abstract:AI-mediated communication is increasingly being utilized to help facilitate interactions; however, in privacy sensitive domains, an AI mediator has the additional challenge of considering how to preserve privacy. In these contexts, a mediator may redact or withhold information, raising questions about how users perceive these interventions and whether explanations of system behavior can improve trust. In this work, we investigate how explanations of redaction operations can affect user trust in AI-mediated communication. We devise a scenario where a validated system removes sensitive content from messages and generates explanations of varying detail to communicate its decisions to recipients. We then conduct a user study with 180 participants that studies how user trust and preferences vary for cases with different amounts of redacted content and different levels of explanation detail. Our results show that participants believed our system was more effective at preserving privacy when explanations were provided (p0.05, Cohen’s d ~ 0.3). We also found that contextual factors had an impact; participants relied more on explanations and found them more helpful when the system performed extensive redactions (p0.05, Cohen’s f ~ 0.2). We also found that explanation preferences depended on individual differences as well, and factors such as age and baseline familiarity with AI affected user trust in our system. These findings highlight the importance and challenge of balancing transparency and privacy in AI-mediated communications and suggest that adaptive, context-aware explanations are essential for designing privacy-aware, trustworthy AI systems.
[HC-19] Digital Frag mentation and Generative AI Use Across 103 Million Application Events
链接: https://arxiv.org/abs/2607.06681
作者: Sumer S. Vaid,Ashley V. Whillans
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Emerging Technologies (cs.ET); Applications (stat.AP)
备注:
Abstract:Knowledge workers switch between applications thousands of times per day, spending nearly a tenth of the work year transitioning between digital applications in a process called digital fragmentation. Whether this fragmentation reflects who an employee is, where they work, or what kind of day they are having, has remained an open question. We analyzed 103 million application events recorded second-by-second from 1,017 employees across eight organizations that largely employ knowledge workers (e.g., law, financial services). Day-to-day variation in fragmentation within individual employees accounted for 44.6% of the variation in digital fragmentation, slightly exceeding stable individual differences between employees (35.8%), and far exceeding variation between organizations (19.6%). Fragmentation rose over the work week and reset after weekends and holidays. Higher-than-typical use of communication applications coincided with more fragmented work. Generative AI use also occurred on more fragmented days, but the period following AI use was marked by narrower, longer, and more predictable application use. These findings identify the workday as a key level for understanding and intervening on digital fragmentation and suggest that AI may help structure fragmented work rather than merely intensify it.
[HC-20] Security and Privacy in Agent ic AI: Grand Challenges and Future Directions
链接: https://arxiv.org/abs/2607.06608
作者: Adam Jenkins,Agnieszka Kitkowska,Caterina Maidhof,Diego Paracuellos,Francesco Sovrano,Gonzalo Gabriel Mendez,Guillermo Suarez-Tangil,Hana Kopecka,Isabel Wagner,Isabel Barbera,Javier Carnerero-Cano,Jide Edu,Jose Luis Martin-Navarro,Jose Such,Josep Domingo-Ferrer,Juan Carlos Carrillo,Kopo Marvin Ramokapane,Mark Cote,Pablo Vellosillo,Ramon Ruiz-Dolz,Rongjun Ma,Ruba Abu-Salma,Sameer Patil,William Seymour,Xiao Zhan
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注:
Abstract:We present key challenges and future research directions in the security and privacy of agentic AI, based on a horizon-scanning exercise that brought together thirty leading international experts from academia, industry, and government to engage in focused discussions and collaborative exercises on the emerging risks associated with the growing agency of AI.
[HC-21] Esports and Physiological Tremor a StarCraft 2 Tournament Study
链接: https://arxiv.org/abs/2607.06577
作者: Andrzej Białecki,Izabela Ghafour,Szymon Kuliś,Maciej Skorulski,Jan Gajewski
类目: Quantitative Methods (q-bio.QM); Human-Computer Interaction (cs.HC)
备注: 11 pages, 3 Figures, 4 Tables
Abstract:Physiological tremor of the upper limb is a sensitive neuromuscular indicator that may be modulated by cognitive load and competitive stress, yet its behaviour in real esports conditions remains uncharacterised. We measured wrist accelerometer-based tremor in 16 healthy adult male StarCraft~2 players across two tournament days, computing log power spectral density ( log(PSD) ) and dominant frequency in four bands (2–4, 8–14, 10–20, and 1–25Hz) and comparing them to published population norms using linear mixed models. Players deviated significantly from the reference in all bands: log(PSD) was elevated at 2–4~Hz and substantially reduced at higher frequencies (Cohen’s d = 1.6 – 2.3 ), suggesting long-term neuromuscular adaptation to the fine-motor demands of esports. Tremor indicators declined systematically over the tournament day. Contrary to the fatigue-related increases typical of traditional motor tasks. Neither game outcome nor actions per minute significantly predicted post-game tremor. These findings suggest physiological tremor may reflect a generalised psychophysiological adaptation to competitive esports rather than being a short-term performance predictor.
计算机视觉
[CV-0] Selective Timestep Weighting and Advantage-Based Replay for Sample-Efficient Diffusion RLHF CVPR2026
链接: https://arxiv.org/abs/2607.07693
作者: Eric Zhu,Abhinav Shrivastava,Soumik Mukhopadhyay
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: 19 pages, 18 figures, 4 tables. Submission under review. A shorter, non-archival 4-page abstract version of this work was accepted to CVPR 2026 Workshops (GCV, CVEU)
Abstract:Reinforcement learning from human feedback (RLHF) has emerged as a powerful paradigm for aligning generative models with human preferences. However, applying RLHF to diffusion models remains highly feedback inefficient, as existing approaches typically require large amounts of human or reward model evaluations. This limitation reduces the practicality of diffusion RLHF in realworld settings where feedback is the primary bottleneck. In this paper, we propose two complementary strategies that substantially improve the feedback efficiency of diffusion RLHF while preserving generalization to unseen prompts. Our key observation is that reward information in diffusion trajectories is unevenly distributed: not all denoising timesteps or trajectories contribute equally to learning from a reward signal. By emphasizing informative timesteps and trajectories during optimization, we obtain more effective gradient updates. First, we introduce a per-timestep weighting scheme that reweights denoising steps during policy optimization. We theoretically connect this weighting to the optimal convergence properties of proximal policy optimization (PPO) and approximate the resulting weighting trend empirically. Second, we introduce a replay mechanism that prioritizes informative trajectories, enabling the model to reuse past samples instead of repeatedly querying new rewards. Together, these strategies significantly improve the feedback efficiency of diffusion RLHF. Under identical hyperparameter settings, our approach achieves up to a 6 \times improvement in sample efficiency compared to widely used diffusion RLHF baselines.
[CV-1] Scaling Mixture-of-Experts Video Pretraining for Embodied Intelligence
链接: https://arxiv.org/abs/2607.07675
作者: Shuailei Ma,Jiaqi Liao,Xinyang Wang,Jingjing Wang,Chaoran Feng,Zijing Hu,Chong Bao,Zichen Xi,Yuqi Gan,Weisen Wang,Yanhong Zeng,Qin Zhao,Zifan Shi,Wei Wu,Hao Ouyang,Qiuyu Wang,Shangzhan Zhang,Jiahao Shao,Yipengjing Sun,Liangxiao Hu,Lunke Pan,Nan Xue,Kecheng Zheng,Yinghao Xu,Xing Zhu,Yujun Shen,Ka Leong Cheng
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project page: this https URL
Abstract:Despite the recent promise in robot control, video generative models suffer from a domain mismatch due to their primary focus on content creation. For example, their design inherently prioritizes visual fidelity and creativity over computational efficiency and physical realism. In this work, we present LingBot-Video, a DiT-based video pretraining paradigm specifically tailored for embodied intelligence. From the architecture perspective, we adopt the Mixture-of-Experts (MoE), instead of dense, framework to achieve a better trade-off between modeling capacity and inference efficiency, and manage to scale it up from scratch. From the data perspective, we construct a data profiling engine that augments standard internet videos with extensive robot-oriented footage, encompassing manipulation, navigation, and egocentric perspectives, to equip the base model with an intrinsic understanding of actions and world dynamics. From the training perspective, we develop a multi-dimensional reward system to enforce the alignment regarding physical rationality and task completion, going beyond standard criteria such as aesthetics, prompt-following, and motion consistency. Comprehensive evaluations validate its performance and efficiency as a video foundation model. We contribute LingBot-Video as the inaugural large-scale, open-source MoE video foundation model to the community, in a pioneering effort to bridge digital creativity and physical actuation.
[CV-2] MedPMC: A Systematic Framework for Scaling High-Fidelity Medical Multimodal Data for Foundation Models
链接: https://arxiv.org/abs/2607.07673
作者: Hyunjae Kim,Dain Kim,Pan Xiao,Serina S. Applebaum,Younjoon Chung,Xuguang Ai,Yu Yin,Roy Jiang,Yuexi Du,Yawen Wei,Yiming Kong,Tuo Guo,Zhiyuan Cao,Mengmeng Du,Yuelei Fu,Yan Hu,Rui Shi,Gui Yang,Kevin W. Jin,Yuntian Liu,Yuxuan Tian,Jonathan Marquez,Zhen Chen,Sheng Zhang,Hoifung Poon,Hua Xu,Jaewoo Kang,Qingyu Chen
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Medicine is inherently multimodal, requiring clinicians to synthesize information across diverse data streams. Yet the development of multimodal foundation models is constrained by limited access to large-scale, high-quality clinical data. Although PubMed Central (PMC) offers a complementary source of expert-authored image-text data, existing PMC-derived resources remain limited in fidelity, reproducibility, and clinical validation. We introduce MedPMC, an automated, continuously updatable framework that transforms permissively licensed literature into high-fidelity infrastructure for medical multimodal models. Applied to 6.1 million PMC articles, MedPMC curated 11 million medical image-text pairs. Component evaluations showed strong performance for initial screening (F1 = 93.2), multi-panel figure detection (F1 = 96.5), figure separation (mAP = 89.8), caption separation and alignment (F1 = 81.4; ROUGE-L = 85.3), and medical figure classification (F1 = 96.5). Manual review by five annotators, three with medical training, found 95.3% of MedPMC images medically relevant, versus 19.7% in a prior PMC-derived dataset. Across 26 benchmarks spanning 11 specialties, a MedPMC-trained CLIP-style model improved average zero-shot AUC by 7.1 percentage points over the strongest architecture-matched biomedical CLIP baseline despite using fewer than half as many image-text pairs. As the vision encoder in a multimodal large language model, it improved medical visual question-answering by 1.9 and 16.9 percentage points across two benchmarks. In 10,524 Yale New Haven Health System dermatology photographs, it improved morphology-to-image retrieval Recall@5 by 11.7 percentage points. These findings show that high-fidelity literature curation strengthens medical multimodal foundation models across benchmark and clinical settings. We publicly release the framework, corpus, benchmarks, and pretrained models.
[CV-3] Dual Latent Memory in Vision-Language-Action Models for Robotic Manipulation
链接: https://arxiv.org/abs/2607.07608
作者: Hongyu Qu,Jianzhe Gao,Xiaobin Hu,Shaohuan Yang,Xinlei Yu,Rui Yan,Wenguan Wang,Xiangbo Shu,Shuicheng Yan
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: Project page: this https URL
Abstract:Mainstream Vision-Language-Action (VLA) models predict actions primarily from the current observation under a Markovian assumption, thus struggling with long-horizon, temporally dependent tasks. Existing memory-augmented VLAs either expand the observation window or retrieve history from the memory bank as auxiliary policy-side context. However, they leave memory outside the native latent embedding space of VLA reasoning, preventing historical experience from being fluidly interleaved with multimodal reasoning and action formation. To this end, we introduce LaMem-VLA, a latent-memory-native framework that reconstructs historical experience into latent memory tokens and directly interweaves them with VLA reasoning. At its core, LaMem-VLA introduces four coordinated components: (i) a curator that organizes historical experience into two complementary short-term and long-term memory vaults; (ii) a seeker that queries both vaults using the multimodal cognition to retrieve context-relevant evidence; (iii) a condenser that reconstructs the retrieved evidence into compact short-term and long-term latent memory tokens; and (iv) a weaver that injects these memory tokens with the current observation and instruction into one continuous embedding sequence. By representing, retrieving, and consuming historical experience entirely in the same continuous latent space, LaMem-VLA enables memory to directly participate in VLA reasoning and guide action generation under a bounded context. Extensive experiments on SimplerEnv and LIBERO demonstrate the superiority of our LaMem-VLA.
[CV-4] Cardiac MRI Through-Plane Super-Resolution Guided by Reference and Memory
链接: https://arxiv.org/abs/2607.07581
作者: Shaoming Pan(1),Chenchuhui Hu(1),Leon Axel(2),Meng Ye(1) ((1) University of Texas at Arlington, (2) New York University Grossman School of Medicine)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 8 pages, 3 figures 2 tables
Abstract:Clinical cardiac MRI is commonly acquired with high in-plane resolution but coarse through-plane resolution to reduce scan time and accommodate breath-hold and cardiac-motion constraints, which limits 3D analysis and diagnostic accuracy. We propose STRMSR, a reference- and memory-guided through-plane super-resolution (SR) framework that reconstructs high-resolution (HR) cardiac volumes by leveraging HR reference views acquired from the same subject and intermediate SR results as the memory. Our method uses coarse-to-fine contextual matching to establish robust correspondence between low-resolution target and reference/memory images under spatial misalignment. A learnable patch-wise dynamic feature aggregation module predicts content-adaptive mixture weights for each local patch, effectively fusing dynamic information while suppressing unreliable feature transfers. The intermediate SR results stored in the memory bank ensure slice-to-slice consistency for the super-resolved 3D volume. Experiments on the WHS cardiac MRI dataset under two reference protocols, orthogonal-plane views and long-axis chamber views, demonstrate consistent improvements over baselines at 4x and 8x upsampling factors.
[CV-5] Automatic Echocardiography Segmentation via Transition Probability Correlation for Stable Semantic Extraction
链接: https://arxiv.org/abs/2607.07580
作者: Xinran Chen,Xiyuan Wang,Guangquan Zhou,Chuan Chen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:While echocardiography is essential for cardiovascular diagnosis, inherent speckle noise and low signal-to-noise ratio often lead to ambiguous semantic features and fragmented boundaries. These limitations significantly hinder the segmentation accuracy of deep learning models in complex clinical cases. Moreover, temporal motion of the heart plays a critical role in recognizing anatomical structures. To address these challenges, we designed a STLSF module which comprises a window-matching-based semantic correction component and a semantics-guided texture enhancement component. By leveraging local transition probability correlations to correct semantics and employing semantics-guided texture enhancement, the STLSF module effectively mitigates texture instability and ambiguous semantic interpretations caused by disadvantaged echocardiography quality. Additionally, to facilitate the encoder’s adaptation to the intrinsic priors of ultrasound-specific imaging patterns, we propose a frequency-aware denoising pre-training method. The entire work builds a convolution-based network with locality inductive bias and long-range dependencies. Extensive experiments confirm our SOTA performance, achieving 93.87% Dice on CAMUS and 92.62% on EchoNet-Dynamic, with respective HD95 values of 3.29mm and 2.73mm.
[CV-6] AA-ViT: Anatomically Aware Vision Transformer with Structural and Frequency Guidance for Contrast Enhanced Brain MRI Synthesis
链接: https://arxiv.org/abs/2607.07553
作者: Talha Meraj,Tom Flannery,Charlie Cummins,Matt Townend,Thomas C Booth,Peter Crossley,Michael McCann,Ian Overton,Saritha Unnikrishnan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted for Publication in MIUA 2026 proceedings
Abstract:Accurate tumour localization and diagnosis is a critical component of clinical care for brain cancers. Magnetic Resonance Imaging (MRI) is the most commonly used imaging modality due to its superior soft-tissue contrast. However, standard MRI often exhibits limited contrast and imaging artifacts, which necessitates the use of contrast agents to enhance lesion visibility. The administration of chemical contrast agents is not always feasible and may be contraindicated in patients with renal impairment or other health conditions. As a result, developing accurate and non-invasive contrast enhanced MRI (CEMRI) synthesis methods has clinical importance. In recent years, numerous approaches for CEMRI synthesis have been proposed, predominantly relying on generative artificial intelligence models. While these methods demonstrate promising performance, their dependence on implicit feature learning often limits their ability to preserve anatomical boundaries and tumour-specific fine structures. To address these challenges, we propose an anatomically aware frequency-and-structure-guided vision transformer (AA-ViT), for CEMRI synthesis using pre-contrast MRI modalities (T1, T2, and FLAIR). Experiments on the BraTS 2021 dataset demonstrate that the proposed method preserves anatomical and lesion boundaries, achieving higher PSNR and SSIM than state-of-the-art approaches. Clinical evaluation by three neuroradiologists and a neurosurgeon on 19 randomly selected cases across diverse gliomas yielded a mean score of 3.94/5, providing preliminary clinical validation rarely seen in prior studies. Synthetic post-contrast scans from our model could lower scanning costs, shorten imaging time, and avoid the potential risks of using gadolinium-based contrast agents.
[CV-7] Face-trace: Open-Set Attribution and Progressive Discovery of Synthetic Face Generators
链接: https://arxiv.org/abs/2607.07545
作者: Alessia Infantino,Claudio Schiavella,Irene Amerini
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Preprint. 17 pages, 16 figures
Abstract:Recent advances in generative Artificial Intelligence have made synthetic face images increasingly realistic, creating new challenges for multimedia forensics. Source attribution methods should not only identify the generator of an image when the source is known, but also handle samples produced by previously unseen models. However, most existing approaches address synthetic face attribution in a closed-set setting, where all possible generators are available during training. This assumption does not hold in real-world scenarios, where new generators continuously appear and rejected samples should be organized rather than simply discarded. In this work we propose a pipeline for open-set synthetic face source attribution that combines known generator classification, energy-based OOD rejection, and unknown generator discovery. A classifier is trained on known generators using frozen I-JEPA embeddings, while rejected samples are represented by combining projected I-JEPA features with Forensic Self-Descriptors and then clustered to discover groups of unknown generators. We also extend the discovery stage to an incremental scenario, where rejected samples arrive over time. Experiments on the WILD dataset show that the proposed method achieves 96.73% closed-set attribution accuracy. In the open-set setting, energy-based rejection reaches 71.25% balanced accuracy, while rejected samples are clustered into meaningful unknown-generator groups, obtaining an ARI of 0.81, an NMI of 0.90, and an overall clustering purity of 87.74%. In the incremental setting, the discovered generator space is progressively extended while maintaining a final purity of 99.23%. Cross-dataset experiments suggest that the pipeline can operate beyond the original dataset distribution, although post-processing remains challenging.
[CV-8] SonoRank: Towards Calibration-Free Real-Time Finger Flexion Detection from Forearm Ultrasound Sequences
链接: https://arxiv.org/abs/2607.07542
作者: Dean Zadok,Alon Wolf,Alex M. Bronstein,Oren Salzman
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Powered prosthetic hands are frequently abandoned, largely due to the limited functionality of current devices that rely on surface electromyography (sEMG). Sonomyography (ultrasound) has emerged as a promising alternative, owing to its ability to observe muscle activity in real time and control a greater number of degrees of freedom. Yet, existing ultrasound-based methods require per-user fine-tuning, limiting their commercialization. We propose SonoRank, an important step towards calibration-free finger flexion detection from forearm ultrasound video. SonoRank first learns to rank pairs of ultrasound sequences by their relative motion magnitude for each of the five fingers. The learned representations are then fine-tuned to classify whether each finger is actively flexing, using a rest reference that is captured at the beginning of the operation. Under 12-fold leave-one-subject-out cross-validation on a dataset of twelve subjects with synchronized kinematics, SonoRank achieves a 28% improvement in F1 score over direct classification baselines that skip the ranking stage. These results establish pairwise ranking as an effective pretraining signal for subject-independent control, bringing ultrasound-based prosthetics closer to practical, calibration-free deployment.
[CV-9] Infinite Worlds with Versatile Interactions
链接: https://arxiv.org/abs/2607.07534
作者: Zelin Gao,Qiuyu Wang,Jiapeng Zhu,Jingye Chen,Zichen Liu,Qingyan Bai,Jiahao Wang,Yufeng Yuan,Hanlin Wang,Yichong Lu,Ka Leong Cheng,Haojie Zhang,Jian Gao,Tianrui Feng,Yuzheng Liu,Yao Yao,Yinghao Xu,Xing Zhu,Yujun Shen,Hao Ouyang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project page: this https URL Code: this https URL
Abstract:We present LingBot-World 2.0 (also known as LingBot-World-Infinity), an advanced iteration of LingBot-World featuring four distinct upgrades. (1) Our model achieves an unbounded interaction horizon while maintaining consistent output quality, benefiting from a carefully crafted causal pretraining paradigm. (2) Through distilling a real-time variant from the base model, our system guarantees rapid response time, sufficient to drive 720p video streams at 60 fps. (3) Compared to the previous version, this update introduces highly diverse interactive elements, comprising a broader spectrum of actions (e.g., attacking, archery, spell-casting, and shooting) alongside a richer variety of text-driven events. (4) We pioneer the integration of an agentic harness within the domain of world modeling, wherein a pilot agent is tasked with planning and executing character behaviors, while a director agent is responsible for synthesizing novel environmental elements as the scene progresses. Additionally, to facilitate a shared experience, we develop an interface that permits multiple players to simultaneously immerse themselves in this vivid world simulator. We pair our primary 14B model with a lightweight 1.3B counterpart, which supports effortless deployment on a single GPU.
[CV-10] Context-Aware Slum Mapping in Sub-Saharan Africa Using Sentinel-1 Texture and Local Climate Zones
链接: https://arxiv.org/abs/2607.07532
作者: Peterson Chepkilot,Babak Memar,Paolo Gamba
类目: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注: Submitted to IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (JSTARS) for possible publication
Abstract:Accurate mapping of informal settlements remains a major challenge in Sub-Saharan African (SSA) cities because optical imagery often fails to distinguish Informal Settlements (defined here as LCZ 7) from spectrally similar formal Compact Low-Rise areas (LCZ 3). This study presents a context-aware, reproducible Optical-SAR framework that improves informal settlement delineation using Sentinel-2 spectral features and Sentinel-1 structural information within an adapted Local Climate Zone (LCZ) taxonomy. We implement a three-tier SAR integration strategy: calibrated backscatter, GLCM textures, and a physics-guided feature engineered to capture the high structural disorder and weak radar return characteristic of SSA informal settlements. Using reference data across Nairobi and Eldoret (Kenya), we evaluate performance via a stratified hold-out protocol and a season-aware ablation study. Results show that SAR textures provide the dominant performance gain for LCZ 7 detection. The Optical-SAR model achieves overall accuracy of 0.816 (dry) and 0.807 (wet), significantly outperforming the WUDAPT baseline (OA 0.704) and reducing the critical LCZ 3 - LCZ 7 confusion to 7%. Seasonal analysis indicates that while optical-only separability varies with phenology, SAR-derived textures stabilize informal settlement mapping across seasons. These findings demonstrate that the incorporation of SAR-derived features yields consistent improvements for urban morphology mapping in data-scarce environments across seasons and across the evaluated source cities, while cross-city transfer remains limited without local adaptation strategies.
[CV-11] Learning to Unify Deformable Shape and Texture Representations for Cardiac Video Classification
链接: https://arxiv.org/abs/2607.07518
作者: Tonmoy Hossain,Miaomiao Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Deformable shape representations have proven to be robust complements to texture features in cardiac image classification, offering geometric priors that are invariant to imaging artifacts and intensity variations. However, existing deep networks perform simple concatenation to combine these distinct feature representations, which neither fully exploits their complementary nature nor learns cross-modal feature dependencies. Furthermore, this results in uniform attention across all timepoints; hence ignoring the varying diagnostic importance across the cardiac phases. In this paper, we propose a novel cardiac video classification model that, for the first time, learns temporal features in an integrated space of deformable shape and image texture representations. In particular, we design a bi-directional cross-attention in the latent space to fuse latent deformable shape and image features, allowing each modality to adaptively weight the other based on spatio-temporal correspondence. In contrast to current methods that apply uniform weighting across all the cardiac phases, our approach learns to dynamically adjust the contributions of shape and texture representations, derived from images, over time. We demonstrate state-of-the-art classification performance on a cine cardiac magnetic resonance (CMR) video dataset, achieving improved interpretability from attention mechanisms that identify diagnostically critical cardiac phases and modality contributions.
[CV-12] HIVE: Understanding Post-Hallucination Reasoning in Vision Language Models ECCV2026
链接: https://arxiv.org/abs/2607.07507
作者: Feng He,Zhenting Wang,Qifan Wang,Qiang Guan,Dongfang Liu,Ruixiang Tang,Qiankun Li
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted by ECCV 2026
Abstract:Hallucinations in vision language models (VLMs) are commonly treated as semantic errors, yet they often arise from partial or ambiguous visual evidence. Prior work mainly focuses on detecting or suppressing hallucinations at generation time, leaving the subsequent reasoning stage largely unexplored. In this work, we study Post Hallucination Reasoning (PHR), the stage in which hallucinated semantics enter the model’s inference context and influence downstream predictions. To systematically investigate PHR, we introduce HIVE, Hallucination Inference and Verification Engine, an evaluation infrastructure that enables controlled comparisons between faithful and hallucinated captions. Across nine tasks and nine models, we observe structured modality dependent patterns: hallucinated captions often improve accuracy on vision language tasks, while text only tasks exhibit limited or unstable effects. Further analyses show that hallucinated cues broaden semantic coverage and reshape reasoning dynamics while preserving stable inference. These findings highlight that hallucinated semantics may influence downstream reasoning once they enter the model’s inference context. Understanding this post hallucination stage is important for improving the reliability and interpretability of multimodal reasoning systems. Code is publicly available at this https URL.
[CV-13] Discovering Geometric Biases in 3D Face Reconstruction: A Curvature-Aware Spectral Framework for Fairness Evaluation
链接: https://arxiv.org/abs/2607.07486
作者: Veronika Shilova,Emmanuel Malherbe,Giovanni Palma,Panagiotis-Alexandros Bokaris,Laurent Risser,Jean-Michel Loubes
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:3D Morphable Models (3DMMs) remain the standard parametric shape priors for many state-of-the-art 3D face reconstruction algorithms. However, as these models are derived from a finite number of 3D face samples, they inherit the morphological biases of their training data, potentially limiting their generalizability across diverse global populations. In this paper, we propose a novel framework to analyze 3DMM reconstructions through the lens of surface curvature, with the objective to discover, quantify and visualize biases. While standard evaluation metrics often rely on Euclidean distances, our reconstruction error captures subtle surface nuances such as local topology or undulations. To do so, we leverage the Laplace-Beltrami Operator (LBO) to generate high-resolution curvature error maps, providing a localized and geometrically meaningful visualization of discrepancies between ground truth faces and reconstructed meshes. We derive from it an error metric that we validated through a user study, observing a significantly higher correlation to human perception compared to traditional methods. Furthermore, we conduct extensive experiments across several 3DMM bases and fitting algorithms, uncovering systematic age-related biases and providing preliminary evidence of biases associated with gender and ethnicity. Our findings highlight the necessity of adopting curvature-aware evaluation protocols to ensure demographic fairness and geometric precision in future 3D face reconstruction research.
[CV-14] A Theory of Contrastive Learning with Natural Images ICML2026
链接: https://arxiv.org/abs/2607.07470
作者: Antonio Torralba,Yair Weiss
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: ICML 2026
Abstract:Why does contrastive learning with simple images and augmentations yield useful representations for downstream tasks? We address this question by analytically computing the optimal representation in terms of a contrastive loss for a range of basic augmentations and any image dataset with stationary statistics. We show that for certain augmentations the optimum can be attained by a CNN whose first layer filters are sinusoids, followed by a pointwise nonlinearity, global average pooling, and a final linear layer that performs partial whitening. We also show that the optimal weights in such CNNs for more complicated augmentations are still sinusoids. The frequencies of the sinusoids and their weights can be computed using a simple waterfilling algorithm given the dataset’s expected power spectrum. Experiments with different image datasets and augmentations show that such CNNs trained with SGD empirically learn sinusoids in their first layer and to perform partial whitening
[CV-15] EmbodiedGen V2: An Agent ic Simulation-Ready 3D World Engine for Embodied AI
链接: https://arxiv.org/abs/2607.07459
作者: Xinjie Wang,Liu Liu,Taojun Ding,Andrew Choi,Chaodong Huang,Mengao Zhao,Ziang Li,Jackson Jiang,Chunlei Yu,Shengxiang Liu,Wei Xu,Zhizhong Su
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:We present EmbodiedGen V2, a generative 3D world engine for building executable sim-ready environments for embodied intelligence. Sim-ready 3D asset generation has advanced rapidly, yet assembling such assets into policy-ready task environments remains largely manual, limiting scalable closed-loop learning. EmbodiedGen V2 addresses this gap through a unified sim-ready representation that connects cross-simulator assets, interaction affordances, task-driven worlds, large-scale multi-room scenes, and stateful Vibe Coding into a generative, editable, and reusable simulation pipeline. The generated environments support manipulation, navigation, mobile manipulation, cross-simulator deployment, and embodied policy training. In evaluation, the asset pipeline achieves 96.5% human acceptance and 98.6% collision success, and 83.3% of task-driven worlds are directly usable for downstream simulation without manual modification. Online reinforcement learning with generated environments further improves simulation success from 9.7% to 79.8%, and transfers to real robots with task success increasing from 21.7% to 75.0%. These results establish EmbodiedGen V2 as scalable simulation infrastructure for training, evaluating, and deploying embodied policies.
[CV-16] wo-Stage Multi-Modal Fusion with Adaptive Alignment for Action Quality Assessment
链接: https://arxiv.org/abs/2607.07438
作者: Kanglei Zhou,Ruizhi Cai,Xinning Wang,Yijian Zheng,Liyuan Wang,Jianguo Li,Xiaohui Liang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to IJCV
Abstract:Action Quality Assessment (AQA) aims to evaluate how well a person performs a movement, which is essential in applications such as sports scoring, skill assessment, and healthcare. However, unimodal approaches often struggle to capture subtle cues of movement quality in real-world settings. Although multi-modal inputs provide complementary information, existing methods still face two major challenges: heterogeneous modalities often lead to cross-modal misalignment and unstable fusion, and reliable multi-modal annotation is costly, resulting in limited dataset diversity. To address these challenges, we propose DualAlign, a two-stage multi-modal fusion framework with adaptive alignment. The framework first constructs a coherent visual representation by maximizing shared structural information across RGB video, optical flow, and skeleton modalities. Textual semantics are then incorporated after visual stabilization, allowing high-level descriptions to complement rather than distort the underlying visual manifold. To evaluate the framework under realistic multi-modal conditions, we introduce MM–JDM, a movement-quality assessment dataset integrating RGB videos, optical flow, skeleton sequences, and structured text. MM–JDM naturally exhibits modality noise, class imbalance, and label scarcity, making it a challenging benchmark for studying multi-modal fusion and alignment. Extensive experiments show that DualAlign improves average correlation on MM–JDM by 21.16% over the state-of-the-art methods and achieves gains of 3.53% and 5.95% on the RG and Fis-V benchmarks, respectively. DualAlign also remains robust under missing-modality and label-scarce conditions.
[CV-17] VCDP: Variation-Conditioned Distributional Proxy Learning for Semi-Supervised Medical Image Segmentation
链接: https://arxiv.org/abs/2607.07416
作者: Zimu Zhang,Yiheng Zhong,Zhuoru Zhang,Yingzhen Hu,Yanan He,Fanliang Meng,Xiaofeng Liu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Semi-supervised 3D medical image segmentation reduces the need for dense voxel-level annotations by exploiting unlabeled volumes. Although existing methods such as consistency regularization, pseudo-labeling, and co-training improve prediction-level robustness, they often provide insufficient feature-space organization for anatomically complex structures, especially small organs and ambiguous boundary regions with large intra-class variations. To address this issue, we propose Variation-Conditioned Distributional Proxy Learning (VCDP), a plug-and-play training-only regularization module for semi-supervised 3D medical image segmentation. VCDP represents each class with a learnable Gaussian distribution for shared class semantics and multiple variation prototypes for fine-grained intra-class patterns. A unified variation-conditioned compatibility score is further formulated to fuse distributional similarity and soft variation aggregation, guiding voxel embeddings to align with both global organ identity and local anatomical variations. VCDP is attached to decoder features during training and removed during inference, introducing no additional inference cost. Experiments on multi-organ segmentation benchmarks show that VCDP improves most evaluated baselines, particularly for small, ambiguous, and highly variable organs. Our anonymous code is released at this https URL.
[CV-18] Heterogeneity-Adaptive Diffusion Schrodinger Bridge for PET-Guided Whole-Body MRI Translation MICCAI2026
链接: https://arxiv.org/abs/2607.07401
作者: Chengbo Wang,Jiacheng Yu,Linjie Bian,Ming Qi,Xiaosheng Liu,Tongtong Che,Jichang Zhang,Shuyu Li,Shaoli Song,Xiuying Wang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted at MICCAI 2026
Abstract:While whole-body multimodal medical imaging scanners have been increasingly recognized for more effective medical applications, the excessive long acquisition time in PET-MR scanning is a major obstacle in more efficient clinical practice. Deep learning-based MRI translation provides a potential solution to reduce scan duration. However, current models often focus on specific anatomical regions and face challenges for whole-body scans that consists of highly heterogeneous feature distributions mainly due to (1) different anatomical regions across whole-body, and (2) lesions or pathological tissues. This paper tackles the challenges through a novel Heterogeneity-Adaptive Diffusion Schrodinger Bridge (HA-DSB) framework. By explicitly modeling translation as stochastic transport between source and target distributions, HA-DSB incorporates region context embeddings derived from a vision-language model (VLM) to enable region-specific modeling. To enhance fidelity of the pathological tissue, lesion-aware metabolic prior from PET is integrated directly into the bridge dynamics through a dual-stage guidance mechanism. Specifically, a PET-guided noise modulation module adaptively scales spatial diffusion perturbations during the forward process, while PET features are leveraged during the reverse process to selectively amplify lesion-relevant structures via an attention mechanism. Experiments demonstrate the superiority of our method across different body regions in whole-body MRI translation and show improved translation quality in lesion areas under PET guidance. Our code is available at Github.
[CV-19] When Prompts Ignore Structure: Graph-Based Attribute Reasoning for Calibrated VLMs EMNLP2026
链接: https://arxiv.org/abs/2607.07395
作者: Tanay Sodha,Aditya Sharma,Ramya Hebbalaguppe,Vinti Agarwal,Pranav Murthy Yeluripaty
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Under review: EMNLP2026
Abstract:Reliable confidence estimation remains a key limitation of test-time adaptation in vision-language models (VLMs), where prompt tuning improves zero-shot accuracy but often degrades calibration due to entropy-driven overconfidence. Prior approaches mitigate this using LLM-derived class attributes and contrastive regularization, yet treat attributes independently, ignoring their relational structure. We propose ARGTCA, which represents (class, attribute) pairs as nodes in a Symbolic Attribute Graph and trains a Graph Attention Network (GAT) using contrastive objectives to produce structurally informed embeddings that capture inter-attribute dependencies. We introduce two attribute selection strategies: ARGTCA-DIV for intra-class diversity and ARGTCA-DISC for inter-class discrimination. Experiments across nine benchmarks show that ARGTCA-DIV reduces average Expected Calibration Error (ECE) by approximately ~37% over baselines, while ARGTCA-DISC consistently performs as the second-best variant, reducing average ECE by approximately ~17% over baselines. These results suggest that modeling symbolic attribute interactions provides a principled approach for reliable test-time adaptation in VLMs.
[CV-20] MMAgent -R2: Learning to Rerank and Reject for Agent ic mRAG ECCV2026
链接: https://arxiv.org/abs/2607.07383
作者: Tao Zhang,Ziqi Zhang,Zongyang Ma,Yuxin Yang,Bing Li,Chunfeng Yuan,Kang Rong,Fengyun Rao,Jing Lyu,Weiming Hu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by ECCV 2026
Abstract:Knowledge-based Visual Question Answering (KB-VQA) requires models to retrieve visual entities matching the query image from large-scale encyclopedic knowledge bases and answer related questions. Existing multimodal Retrieval Augmented Generation (mRAG) methods rely on global visual features to match candidate entities, yet when the knowledge base contains numerous visually similar entities, the retriever struggles to distinguish them, populating the candidate set with visually similar but factually mismatched distractors. Since subsequent processing steps such as noise filtering are also confined to this fixed candidate set, errors from failed retrieval inevitably propagate to the final answer. To address these challenges, we propose MMAgent-R ^2 , an agentic mRAG framework that integrates visual reranking and active rejection as its internal verification mechanism. Visual reranking directly compares query and candidate images, capturing discriminative details beyond textual descriptions to precisely identify the target entity among similar candidates; active rejection discards unreliable results and retrieves additional candidates when no confident match is found, moving beyond the fixed candidate pool. We design a composite reward function with step-level verification rewards and achieve joint optimization of external retrieval, internal verification, and answer generation via GRPO training. Experiments on InfoSeek, E-VQA, and MMhops demonstrate that \ours achieves state-of-the-art performance, with particularly notable advantages in challenging retrieval scenarios and complex multi-image multi-hop reasoning tasks.
[CV-21] BUS: Brain-Inspired Unsupervised Self-Reflection for Advanced Multimodal Reasoning
链接: https://arxiv.org/abs/2607.07361
作者: Jiacheng Yang,Tongying Xiao,Yunkai Dang,Cong Wang,Yuekun Yang,Qi Fan,Wenbin Li,Feng Miao,Yang Gao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Current Vision-Language Models (VLMs) often struggle to handle complex visual tasks that require consistent and fine-grained reasoning. Recent methods aim to train models to facilitate self-reflective reasoning, i.e., reviewing and improving the generated reasoning. However, they require large volumes of annotated data and lack explicit reflective behavior during test time. This work aims to bridge this gap through inspiration from neuroscience. The human brain exhibits efficient backward prediction, i.e., predicting which current states are likely to precede a given future state. In this work, we first verify that mainstream VLMs can perform backward prediction, similar to the human brain. Then, we propose Brain-inspired Unsupervised Self-reflection (BUS), a label-free training framework to enhance reflective reasoning capability in challenging image analysis. BUS enables VLMs to perform backward prediction and provide explicit learning signals on data without ground-truth labels. In this way, BUS eliminates reliance on annotated data while improving reasoning performance. Notably, BUS is compatible with popular fine-tuning methods, such as Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL). Finally, extensive experiments on 8 benchmarks demonstrate the effectiveness of BUS across a wide range of complex visual tasks. It achieves notable improvements over the base model while using only unlabeled training data. Our experimental findings validate that backward prediction capability is critical for VLM reasoning.
[CV-22] HAJJv2-CrowdCount: Zero-Shot Benchmark for Dense Crowd Counting
链接: https://arxiv.org/abs/2607.07322
作者: Reem AlYabis,Fares AlTuwaim,AlJawharh AlOtaibi,Mohamed Eltahir
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 5 pages, 8 figures, 2 tables. Annotations available at this https URL
Abstract:Automated crowd counting in Hajj video is difficult not because current models lack capacity, but because the footage violates the assumptions those models were built on: cameras observe the crowd from steep, near-vertical angles, individuals occlude one another extensively, and a single frame can contain well over a thousand people. Benchmarks that test crowd counting in such an environment are either private or not detailed per second. We revisit the HAJJv2 dataset and contribute HAJJv2-CrowdCount: per-second human-annotated crowd counts for its testing videos. Using these annotations, we benchmark three recent zero-shot counting paradigms: an open-vocabulary detector (YOLO-World), a point-based counter (APGCC), and a promptable segmentation-based counter (SAM3Count). SAM3Count attains the lowest overall mean absolute error (MAE 70.4, 95% CI 56.0-86.1), ahead of YOLO-World (92.0) and APGCC (152.9). This ordering reverses, however, in the regime most relevant to deployment: on the densest frames, the detection- and segmentation-based counters both degrade sharply (MAE exceeding 300), while the point-based counter degrades far more gracefully (MAE 114.9). This inversion is decision-relevant for Hajj crowd management, where reliable counts are needed most precisely in the densest and most occluded scenes. The annotations are released to support reproduction and extension of these results.
[CV-23] SoccerNet 2026 Challenges Results
链接: https://arxiv.org/abs/2607.07320
作者: Anthony Cioppa,Silvio Giancola,Håkan Ardö,Mohamad Dalal,Jan Held,Jérémie Ochin,Jiayuan Rao,Karen Sanchez,Renaud Vandeghen,Artur Xarles,Olivier Barnich,Albert Clapés,Mathieu Delvaux,Sergio Escalera,Bernard Ghanem,Cédric Hons,Antoine Houet,Sotiris Manitsaris,Tom Michel,Pierre Miralles,Thomas B. Moeslund,Mikael Nilsson,Bogdan Stanciulescu,Marc Van Droogenbroeck,Yanfeng Wang,Weidi Xie,Faisal Altawijri,Mohamed Atef,Semen Budennyy,Vasiliy Chelpanov,Puhua Chen,Yixin Chen,Lechao Cheng,Jianling Chu,Ju-Seong Do,Oleg Durygin,Omar Fetouh,Mirco Fuchs,Youssef Ghallab,Falguni Ghosh,Wonjun Heo,Yufeng Hu,Weixuan Huang,Phuong-Linh Huynh-Ha,Matvey Isupov,Yangguang Ji,Siyuan Jiang,Zhenxiang Jiang,Wonyong Jo,Ho-Young Jung,SeongHeon Kang,MinJae Kim,Youngseon Kim,Jakub Komosa,Artem Konshin,Trung-Hoang Le,Jongmin Lee,Lingling Li,Litao Li,Vadim Linkov,Fang Liu,Haoxuan Ma,Shun Makino,Ismail Mathkour,Konstantin Mitin,Mikhail Moiseev,Takumi Nagaya,Yuki Nakamura,Thanh-Khoi Nguyen,Hoang-Phuc Nguyen,Trong-Thuan Nguyen,Christian Orduz,Kwanyong Park,Fabian Perez,Parthsarthi Rawat,SuHyun Rim,Hoover Rueda-Chacón,Atom Scott,Minori Sugimura,Yuyang Sun,Shengeng Tang,Minh-Triet Tran,Ikuma Uchida,Juan Vanegas,Thanh-Nhan Vo,Jiangtao Wang,Yaxiong Wang,Xiaogang Wang,Ruifeng Wang,Rio Watanabe,Jiali Wen,Yongliang Wu,Di Yang,Xu Yang,Zhuo Yang,Xinyu Ye,Yibo Yu,Zihan Zhai,Yu Zhang,Zhenyu Zhao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 40 pages
Abstract:The SoccerNet 2026 Challenges constitute the sixth annual edition of the SoccerNet open benchmarking effort, dedicated to advancing computer vision research in sports video understanding. This year’s challenges span five vision-based tasks: (1) Ball Action Anticipation, predicting the timing and class of ball-related actions within a short future window from a preceding observation window; (2) Player-Centric Ball Action Spotting, temporally localizing and classifying ball-related actions while assigning each action to the acting player through team affiliation and jersey number; (3) Novel View Synthesis, rendering images from unobserved camera poses in multi-view football scenes; (4) Spiideo SoccerNet Synloc, localizing athletes in real-world pitch coordinates from a single calibrated static-camera image; and (5) Visual Question Answering, answering multiple-choice questions about football broadcasts across text, image, and video inputs. For each task, participants were provided with annotated data, a unified evaluation protocol, and a public baseline. This edition saw broad participation, with 427 teams submitting 1,129 entries across the five tasks and 28 teams contributing reviewed technical reports. This paper describes each task and its evaluation protocol, presents the challenge leaderboards, and summarizes the leading submissions, with the aim of documenting the current state of each task as measured on held-out challenge data.
[CV-24] CarbonCLIP: Enhance Carbon Prediction from Satellite Imagery via Integrated Street-View Semantics and Temporal Context Training
链接: https://arxiv.org/abs/2607.07292
作者: Zeru Yang,Fang-Ying Gong,Steve H.L. Yim,Chau Yuen
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted by IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 21 pages, 6 figures, 9 tables
Abstract:Accurately estimating urban carbon emissions is critical for sustainable urban planning, yet many existing approaches remain difficult to apply consistently across cities due to data-source heterogeneity and the lack of fine-grained semantic-temporal context in remote sensing data. We propose CarbonCLIP, a task-oriented multimodal distillation framework that improves satellite-based carbon emission prediction by transferring contextual knowledge into a unified satellite representation through dual-branch contrastive learning. Unlike conventional methods that rely on static visual features, CarbonCLIP explicitly bridges the gap between top-down satellite views and ground-level human activities. Specifically, the spatial branch uses fine-grained textual descriptions automatically generated from street-view images by Large Multimodal Models (LMMs) to provide semantic priors reflecting building functions, infrastructure, and urban activities, while the temporal branch employs a month encoder to encode temporal priors associated with monthly emission variation. CarbonCLIP requires multimodal data only during the pretraining phase; during inference, it relies solely on satellite imagery, thereby supporting scalable deployment when ground-level data are unavailable at inference. Experiments on Beijing and Singapore demonstrate that CarbonCLIP outperforms baselines in both study cities. The results validate that our method effectively transfers multimodal knowledge into satellite representations, offering a robust solution for satellite-based urban carbon modeling.
[CV-25] InfraQR: Edge-Placed QR-Inspired Structured Patch Attacks on Infrared Vision-Language Models
链接: https://arxiv.org/abs/2607.07288
作者: Xin Li,Jiaju Han,Ma Yaqi,Chengyin Hu,Yingying Zhao,Jiahuan Long,Fengyu Zhang,Yahui Chai
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Infrared vision-language models are increasingly used for perception under low-light and adverse visual conditions, yet their robustness to localized structured perturbations remains underexplored. Existing infrared adversarial studies mainly focus on object detectors, leaving the security of infrared vision-language models less systematically examined. We present InfraQR, a QR-inspired structured patch attack for infrared vision-language models. Unlike localized attacks that attach perturbations to the target object, InfraQR places a compact structured patch along image boundaries and optimizes learnable grid cells through surrogate CLIP-style encoders. The resulting patch has a near-binary structured appearance, but is not required to be a valid or machine-readable QR code. We evaluate InfraQR on infrared classification, caption transfer, and question-answer-aware visual question answering (VQA) tasks. On a 300-image infrared benchmark, InfraQR sharply reduces the accuracy of multiple CLIP-style classifiers, including reducing OpenAI CLIP accuracy from 98.67% to 0.70%. The generated adversarial images also transfer to black-box captioning and VQA models, causing semantic degradation in captions and more error-prone answers under GPT-5.4-based evaluation. These results show that infrared vision-language models remain vulnerable to structured edge-placed perturbations, motivating further study of cross-task robustness beyond direct object occlusion.
[CV-26] Naming the Concepts Classifiers Rely On: Language-Anchored Decomposition for Faithful Explanation
链接: https://arxiv.org/abs/2607.07264
作者: Ahsan Habib Akash,Dipkamal Bhusal,Stacey Jones,Donald A. Adjeroh,Binod Bhattarai,Prashnna Kumar Gyawali
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Code available at this https URL
Abstract:Deep neural networks are widely deployed in high-stakes visual applications where interpretability is critical, yet existing explanations face a trade-off: post-hoc concept methods recover factors that are faithful to a model’s behavior but unnamed, while naming and by-design methods attach human-readable concepts only by retraining or altering the classifier. We propose Language-Anchored Decomposition (LAD), a post-hoc framework that delivers concepts which are simultaneously named, faithful, and obtained without modifying the model. For each class, a large language model proposes a concept vocabulary that CLIP-based similarity maps localize across image regions. Inverting standard non-negative matrix factorization, LAD fixes these language-grounded maps as the coefficient matrix and learns only a concept basis that reconstructs the frozen encoder’s activations, so naming becomes a structural constraint and the model’s own feature geometry determines which concepts are retained. Removing this anchor preserves accuracy but collapses attribution faithfulness. Across natural-image, scene, and medical-imaging benchmarks, LAD produces spatially precise explanations that are decision-relevant under both concept insertion and deletion, while uniquely providing stable, human-interpretable concept names.
[CV-27] An Edge-aware Prompt-enhanced SAM for Ultrasound Image Segmentation ICME2026
链接: https://arxiv.org/abs/2607.07240
作者: Wenhao Li,Fangyi Liu,Bo Du
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ICME2026
Abstract:Ultrasound image segmentation is essential for delineating anatomical structures and lesions, providing the foundation for accurate diagnosis. While the Segment Anything Model (SAM) has demonstrated remarkable success on natural images, its performance on ultrasound data is often hindered by poor boundary delineation. To address this limitation, we propose EP-SAM, an edge-aware and prompt-enhanced adaptation of SAM. Specifically, we leverage multi-block feature extraction from the image encoder to enrich coarse-to-fine semantic representations, while edge-aware supervision of the image encoder improves robustness to contour ambiguity and speckle noise. By integrating these complementary cues, EP-SAM generates high-quality prompts that effectively guide the model toward target regions of interest. Experimental results on multiple benchmarks demonstrate that EP-SAM consistently outperforms existing SAM-based methods.
[CV-28] Unraveling Machine Behavior by Multi-Level Bias Analysis and Detection: Methodology and Application to Computer Vision
链接: https://arxiv.org/abs/2607.07236
作者: Ignacio Serna,Aythami Morales,Julian Fierrez
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:This study investigates the presence and propagation of bias within Neural Networks through a comprehensive multi-level analysis spanning the learned latent space, layer activations, and the network’s parameters. Based on this taxonomy, we propose three bias detection approaches: 1) SpaceBias (new method), which characterizes the latent space prior to the final classification layer using neighbor-probability distributions and quantifies bias with the two-sample Kolmogorov-Smirnov test on the per-group distributions. 2) ActivationBias (extension of the existing method InsideBias), which analyzes the activations of neural network filters and quantifies bias via a Mann-Whitney U test, based on the observed fact that underrepresented groups exhibit lower activation levels in the final convolutional layers. 3) WeightBias (extension of the existing method IFBiD), which uses a secondary neural network trained to identify biased patterns directly in the parameters of task-specific models. Unlike conventional methods, which assess neural network outcomes and treat the model as a black box, our proposed techniques provide insight into how biases manifest within the network architecture itself at different levels, offering a more nuanced and detailed understanding. Experiments are conducted on two complementary applications: gender classification in the DiveFace dataset (72,000 face images) and digit classification on a colored-MNIST benchmark with controlled bias severity. In total, more than 127,000 models with varying degrees and types of bias were trained and evaluated. The severity sweep shows that the internal disparity, and with it the detection performance, decreases smoothly as the training distribution approaches balance. The results highlight the importance of methods that provide deeper insight into the behavior of AI models.
[CV-29] `Attention-Guided Cross-Temporal Clustering for Self-Supervised Video Object Segmentation
链接: https://arxiv.org/abs/2607.07230
作者: Waqas Arshid,Mohammad Awrangjeb,Alan Wee-Chung Liew,Yongsheng Gao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted for publication in Machine Intelligence Research journal
Abstract:Video object segmentation (VOS) is a fundamental task in video understanding, requiring accurate delineation and consistent tracking of objects across frames. While supervised methods achieve strong performance, they rely on densely annotated datasets that are costly to obtain and have limited domain coverage. Self-supervised learning offers a promising alternative by removing the need for manual labels; however, existing approaches often struggle to jointly maintain spatial accuracy and temporal coherence, particularly in unconstrained multi-object scenarios. Many rely on optical flow, synthetic motion cues, or task-specific pretraining, limiting scalability and generalisation. We propose a self-supervised framework, Cross-Temporal Consistency and Clustering, that learns mid-level, part-aware representations by combining attention-guided token selection with lightweight temporal clustering. Instead of operating at the pixel or whole-object level, the method aligns soft part assignments across time using a saliency-weighted symmetric consistency objective. The framework leverages a frozen transformer backbone with lightweight modules for adaptive token selection and multi-offset temporal alignment, enabling efficient scaling across resolutions and motion patterns.
[CV-30] Vision Foundation Models in Radiology: A Scoping Review of Data Methodology Evaluation and Clinical Translation
链接: https://arxiv.org/abs/2607.07219
作者: Alejandro Vergara-Richart(1 and 2),Xavier Rafael-Palou(1),Almudena Fuster-Matanzo(1),Ignacio Iborra Roncales(1),Ángel Alberich-Bayarri(1),Ana Jiménez-Pastor(1) ((1) Quantitative Imaging Biomarkers in Medicine, Quibim S.L., València, Spain, (2) Universitat Politècnica de València, València, Spain)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 33 pages, 8 tables, 2 figures
Abstract:Vision foundation models (VFMs) are increasingly being developed for radiological imaging, yet their definition, development and evaluation remain heterogeneous. We conducted a PRISMAScR scoping review of peer-reviewed studies published between January 2017 and March 2026 describing foundation models trained exclusively on radiological imaging data. Sixty-seven studies were included and mapped across three pillars: data scale and heterogeneity, architectural and pretraining scalability, and downstream transferability and generalization. Datasets primarily covered brain MRI, thoracoabdominal CT, and chest X-ray, ranging from fewer than 100,000 samples to multi-million-image cohorts. Transformer-based architectures and self-supervised pretraining predominated, particularly masked image modeling, contrastive learning and multi-stage approaches. Evaluation focused mainly on segmentation and classification, whereas cross-center, cross-scanner, anatomical and modality-shift validation was inconsistently reported. Alignment with FUTURE-AI principles was uneven. Overall, radiology-specific VFMs show promising transferability, but clinical translation remains constrained by limited data representativeness, heterogeneous benchmarks, incomplete reporting and insufficient deployment-oriented evaluation.
[CV-31] Why Fake ? Unveiling the Semantic Vocabulary of Deepfake Detectors CVPR
链接: https://arxiv.org/abs/2607.07216
作者: Vazgken Vanian,Alexandros Doumanoglou,Dimitris Zarpalas
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at CVPRW 2026
Abstract:Deepfake (DF) technology poses a significant threat to information integrity, driving the need for robust detection methods. Most DF detectors only consider predicting a binary label for whether the input is real or fake, lacking the justification required for real-world applications like legal proceedings. Explainable DF Detection has emerged to address this limitation, but existing techniques frequently fall short by either relying on human annotations for precise artifact localization or generating superficially plausible textual explanations without grounding. This work investigates the use of post-hoc explainable AI (XAI) to analyze the decision-making process of state-of-the-art black-box DF detectors. Specifically, we employ Encoding-Decoding Direction Pairs (EDDP), a technique suitable for uncovering the concept space of DF detectors (their semantic vocabulary) as well as the mechanism for writing and reading concept information to and from internal representations. Our analysis reveals previously hidden real and fake features learned implicitly during detector training, offering nuanced explanations unattainable through conventional methods. This enables global model understanding, spatially aware concept localization, and counterfactual what-if analysis, all contributing to a deeper comprehension of DF detection strategies.
[CV-32] DiffCVE: Diffusion-based Compressed Video Enhancement
链接: https://arxiv.org/abs/2607.07195
作者: Wenqiang Xiao,Wenzhuo Ma,Junxi Zhang,Zhenzhong Chen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Perceptual quality enhancement of severely compressed videos remains challenging due to complex artifact patterns and substantial information loss. Recent diffusion models have demonstrated strong generative capability for visual restoration, but directly applying them to compressed video often ignores compression degradation characteristics and may introduce structure-inconsistent hallucinations. To address this issue, this paper presents a diffusion-based compressed video enhancement method, named DiffCVE. Coding Prior-enhanced Dual Conditioning (CPDC) branches are designed to jointly model compressed video and coding prior conditions, where coding priors including residuals and motion vectors provide complementary structural and motion guidance during the diffusion denoising process. To make the diffusion process aware of compression severity, a Compression Degradation Semantic Prompting (CDSP) mechanism is introduced to leverage QP-conditioned textual prompts together with LoRA fine-tuning. In addition, a Coding Prior-guided Weighted Fusion (CPWF) module is incorporated into the VAE decoder to fuse VAE encoder and coding prior encoder features with QP-predicted weights. Extensive experiments demonstrate the effectiveness of the proposed method in improving perceptual quality, especially under severe compression settings. The project page with enhanced video demonstrations is available at this https URL.
[CV-33] Prototype-Anchored Generalized Manifold Regression for Unknown-Domain Object Detection
链接: https://arxiv.org/abs/2607.07192
作者: Zihao Zhang,Aming Wu,Yang Li,Yahong Han
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:In this paper, we study Single-Domain Generalized Object Detection (Single-DGOD), which aims to transfer a detector trained on a single source domain to multiple unseen domains. Existing methods mainly rely on simulation-driven strategies, such as data augmentation or textual prompts, to enlarge the training distribution. However, finite simulations can hardly cover the dynamic variations of real-world scenarios, often causing overfitting to synthetic styles and limited robustness to complex structural degradations. Inspired by the manifold hypothesis, we argue that semantic features, despite diverse visual changes, should lie on a compact and stable low-dimensional manifold. Therefore, robust generalization requires rectifying deviant samples back to this semantic manifold, rather than exhaustively simulating external perturbations. To this end, we propose Manifold Regression with Visual-Text Dual Chain-of-Thought (MR-DCoT), which formulates unknown-domain generalization as a manifold regression problem. MR-DCoT first uses a Visual-Text Dual Chain-of-Thought module to combine VLM-guided semantic evolution with diffusion-based structural perturbation, generating structured off-manifold hard examples. It then introduces Class-Specific Prototype Anchoring to learn a rectification operator that projects deviant features toward the source semantic manifold. By integrating outlier generation and semantic correction into a closed loop, MR-DCoT effectively narrows the distribution gap and improves robustness under unseen shifts. Extensive experiments on three complementary benchmarks, including adverse-weather detection, real-to-art generalization, and zero-shot semantic segmentation, demonstrate the effectiveness and versatility of our method.
[CV-34] EditVerse3D: High-Quality 3D Object Editing with Region-Aware Learning ECCV2026
链接: https://arxiv.org/abs/2607.07187
作者: Youtan Yin,Yanning Zhou,Jiacheng Wei,Xiaofeng Yang,Jun Zhang,Jiayang Bai,Jingwen Ye,Weidong Zhang,Guosheng Lin
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ECCV 2026. Project page: this https URL
Abstract:Local editing of 3D objects remains a long-standing challenge. When interacting with 3D content, humans naturally tend to specify a coarse region of interest for modification rather than defining precise editing boundaries. However, previous methods rely on fully edited 2D images, precise 3D masks, or redundant pipelines, which present a gap. To bridge this gap, we propose EditVerse3D, a novel 3D editing framework that enables high-quality object editing under such coarse guidance. Our approach takes as input a 3D object to be edited, a coarse 3D bounding box indicating the target region, and a reference 2D image describing the desired modification. It produces a coherent, high-fidelity edited 3D object. To facilitate this editing, we introduce a novel region-aware adaptive loss that emphasizes hard-to-learn regions and balances the objective between target and preserved areas. Complementing our loss function, we enhance model robustness and generalization through targeted data augmentations, such as training with scaled 3D masks and filtering out unrealistic editing pairs. We construct a large-scale 3D editing dataset derived from parts information. Extensive experiments demonstrate that EditVerse3D achieves superior visual quality and quantitative performance compared to existing 3D editing approaches. Please visit our project page at this https URL.
[CV-35] Comparative Study of Domain-adapted VLMs for General Document Visual Question Answering ICDAR2026
链接: https://arxiv.org/abs/2607.07179
作者: Miguel Lopez-Duran,Elena Marrero,Julian Fierrez,Marta Robledo-Moreno,Ruben Vera-Rodriguez,Daniel DeAlcala,Aythami Morales,Ruben Tolosana,Oscar Delgado,Alvaro Ortigosa,Javier Ortega-Garcia
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 17 pages, 4 figures, accepted at the Automatically Domain-Adapted and Personalized Document Analysis workshop of the ICDAR 2026
Abstract:Document Visual Question Answering (DocVQA) presents a complex multimodal challenge, requiring models to exploit visual, textual, and layout information from documents. Although Vision-Language Models (VLMs) have shown remarkable performance in text-vision tasks, their robustness and transferability to different document domains remains underexplored. In this study, we present a comprehensive evaluation of 8 open-source pretrained VLMs on DocVQA in three different document domains: industrial documents of varying type, infographics, and presentation slides. We systematically assess model performance under zero-shot evaluations, fully supervised finetuning with inter- and intra-dataset evaluations, and few-shot learning evaluations of knowledge transfer between domains. Our findings demonstrate that while large pretrained VLMs possess strong zero-shot baselines for structured layouts, their performance strongly decreases on visually complex layouts of infographics and slides. Although parameter scaling is a dominant factor on performance, supervised finetuning yields higher relative gains in smaller architectures. Furthermore, our cross-domain and few-shot experiments show that visual understanding is the main bottleneck for DocVQA, not a lack of knowledge from the VLMs. Using 50 target domain samples, the models finetuned in DocVQA with datasets of different domains rapidly adapt to the target domain documents, even surpassing their fully supervised counterparts in some cases.
[CV-36] Stage-Aware Adaptation and Distribution Calibration for Subject-Driven Personalized Text-to-Image Generation
链接: https://arxiv.org/abs/2607.07173
作者: Wenyan Xu,Alizer Wong
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 16 pages, 4 figures, 6 tables
Abstract:Subject-driven personalized text-to-image generation requires a pretrained diffusion model to acquire a specific subject from a few reference images while preserving subject identity, following novel text prompts, and maintaining sample diversity. Existing optimization-based methods instantiate subject adaptation through full fine-tuning, textual embedding optimization, or low-rank parameter updates; PaRa further constrains personalization from the perspective of parameter rank reduction. However, a uniform low-rank constraint or a uniform adapter strength cannot explicitly distinguish the capacity requirements of different denoising stages. Moreover, inference-time candidate selection driven mainly by identity similarity may compress the selected samples in the visual representation space. We decompose the problem into two complementary components: SPaRa denotes training-side stage-aware low-rank adaptation, DCAL denotes inference-side distribution-calibrated candidate selection, and SPaRa-DCAL denotes the combined framework. Theoretical analysis shows that timestep-dependent scaling controls the effective perturbation magnitude of a low-rank adapter, while identity-biased candidate selection restricts the radius of selected features around the reference center under explicit conditions. Auditable experiments under the SDXL and DreamBooth 30-subject protocol show that DCAL improves 1-LPIPS, CLIP-I, DINO-I, and CLIP-T on a fixed LoRA candidate pool, while revealing a clear trade-off with CLIP/DINO pairwise diversity and pairwise LPIPS. These results indicate that personalized generation should be evaluated through identity consistency, text alignment, and representation diversity rather than identity metrics alone.
[CV-37] PUF: Plug-and-Play Uncertainty-Aware Fusion for Online 3D Scene Graph Generation ECCV’26
链接: https://arxiv.org/abs/2607.07170
作者: Yi Yang,Myrna Castillo,Bodo Rosenhahn,Michael Ying Yang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by ECCV’26
Abstract:Online 3D scene graph generation builds a persistent, structured representation of a scene by incrementally fusing 2D observations into a global 3D graph. Existing online methods treat this fusion as a fully deterministic pipeline, where we identify three sources of uncertainty that are overlooked: observation, 2D model, and 3D representation. We propose PUF: a Plug-and-play, Uncertainty-aware, and training-free Fusion framework. Scene graph node association is reformulated as a probabilistic likelihood over semantic and spatial factors, replacing binary accept/reject gates. Dirichlet evidence accumulation distributes class and relationship evidence across plausible candidates proportional to association likelihood. An optional class-conditional prior completes edges for sparsely or never co-observed object pairs. We instantiate PUF with both a 3D Gaussian and a 3D voxel backend and observe consistent improvements, demonstrating its ability to generalize across different representations. Experiments on the 3DSSG and ReplicaSSG benchmarks show that our method substantially outperforms existing approaches while maintaining real-time latency. These results establish uncertainty-aware fusion as a principled and effective paradigm for online 3D scene understanding. The source code is publicly available at this https URL.
[CV-38] ACoS: Weakly Supervised Learning of Two-Dimensional Materials from Scribble Annotations to Precise Segmentation
链接: https://arxiv.org/abs/2607.07169
作者: Jiabei Chen,Liping Zhang,Jiang-Bin Wu,Zhongming Wei,Enhao Ning,Su Yan,Weijun Li,Ping-Heng Tan,Xin Ning
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 35 pages, 7 figures
Abstract:The precise pixel-level localization of 2D material flakes is crucial for high-throughput screening. However, traditional fully supervised methods rely on dense annotations, which are costly and time-consuming, severely limiting the practical deployment of segmentation models. This paper proposes TACoS, a specialized scribble segmentation framework tailored for 2D materials. First, we design a unified framework that integrates semi-supervised consistency learning with structured tree energy constraints. This framework comprises two core components: an unlabeled weak-strong distribution alignment module and a tree energy regularization module. The former employs cosine consistency constraints to enhance prediction alignment across views. Meanwhile, the latter utilizes minimum spanning trees to establish pixel affinity relationships and generate structure-aware soft pseudo labels for online semantic guidance. Next, we introduce asymmetric regional contrast learning. This approach fuses high-confidence predictions from the weak augmentation branch with scribbles to form augmented labels, and construct category prototypes in the representation space. Simultaneously, we prioritize contrastive constraints on challenging pixels in boundary-unlabeled regions. This strategy enhances intra-class cohesion and inter-class separation at the representation level, effectively reducing category confusion in low-contrast edges and complex backgrounds. Experiments conducted on the constructed graphene and MoS2 datasets demonstrate that our method TACoS achieves over 96% of fully supervised performance using less than 0.6% annotated data. Furthermore, it exhibits superior structural coherence and boundary stability in scenarios with weakly contrasting edges and complex backgrounds, providing an efficient and scalable solution for automated high-throughput screening of 2D material flakes.
[CV-39] NoDrift3R: Raymap-Guided Coupling for Drift-Robust Unposed Feed-Forward 3D Reconstruction
链接: https://arxiv.org/abs/2607.07168
作者: Xiangyu Sun,Liu Liu,Seungkwon Yang,Jingbing Han,Seungtae Nam,Zhizhong Su,Eunbyung Park
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: European Conference on Computer Vision
Abstract:Pose-Free Feed-forward 3D Gaussian Splatting (3DGS) has recently emerged as a powerful paradigm for fast scene reconstruction. However, its performance degrades significantly in long image sequences due to cumulative camera pose estimation drift, which propagates errors into geometric modeling and severely limits rendering fidelity. In this work, we revisit the long-sequence bottleneck and identify pose drift as the primary factor restricting reconstruction quality. Furthermore, while SfM-based pseudo ground-truth poses introduce sensor noise, purely rendering-based supervision often leads to optimization instability and local minima due to the entangled optimization of geometry and pose. To address the challenges, we propose a synergistic pose-free framework that explicitly couples geometry and appearance via a Raymap-Guided Coupling Module (RGC). Concretely, we anchor Gaussian centers to raymap-induced geometry and jointly optimize RGB reconstruction, raymap consistency, and camera regularization under a unified objective, yielding a bidirectional feedback loop: stronger geometry improves rendering, and appearance supervision in turn refines geometry and pose. To further stabilize learning across wide temporal ranges, we introduce a Dual-Frequency Viewpoint Scheduling strategy that combines easy-to-hard interval expansion with replay of short-interval pairs. Extensive experiments across in-domain and cross-domain datasets show consistent gains in both rendering and pose estimation, with notably improved robustness on long sequences. Ablation studies validate our central insight: explicitly designed geometry-appearance synergy is the key to scalable and drift-robust pose-free feed-forward 3D reconstruction.
[CV-40] ASFR-Net: Adversarial Alignment and Spatio-Frequency Refinement Network for Heterogeneous Remote Sensing Image Change Detection
链接: https://arxiv.org/abs/2607.07161
作者: Xin-Jie Wu,Zhi-Hui You,Si-Bao Chen,Qing-Ling Shu,Xiao Wang,Jin Tang,Bin Luo
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:The core challenge of heterogeneous change detection in remote sensing imagery lies in effectively decoupling genuine land-cover changes from significant modal disparities caused by distinct imaging mechanisms. These intrinsic inconsistencies are prone to introducing pseudo-changes, thereby constraining detection accuracy. To address this, we propose a novel, end-to-end adversarial spatio-frequency refinement network (ASFR-Net). Initially, a modality-invariant representation learner (MIR-Learner) guides the backbone to extract modality-invariant features, effectively bridging the primary domain gap. Subsequently, to address persistent residual modal differences, we design an innovative spatio-frequency synergistic enhancement module (SFEM), which identifies and suppresses sensor-specific noise and artifacts that are difficult to discern in the spatial domain by leveraging frequency-domain processing. Multi-level difference features are then computed from these refined representations and fed into a decoder equipped with cascaded hierarchical guided fusion module (HGFM) blocks to generate precise change maps. To alleviate the data scarcity in heterogeneous tasks, we construct and release a new high-resolution benchmark specifically focused on building changes: the visible-near-infrared heterogeneous change detection (VisNIR-HCD) dataset. It presents unique scientific challenges arising from deceptive visual similarity and non-linear spectral inversions, providing a robust platform for evaluating model generalization. Extensive experiments on VisNIR-HCD and public datasets demonstrate that ASFR-Net achieves state-of-the-art (SOTA) performance, significantly outperforming existing methods. The source code and the VisNIR-HCD dataset are publicly available at this https URL.
[CV-41] Prior-matched evaluation of operational Earth-observation classifiers: a three-number reporting method demonstrated on Sentinel-1 internal-wave detection
链接: https://arxiv.org/abs/2607.07146
作者: Joao Pinelo,Joao Goncalves,Arun Shukla,Adriana Santos-Ferreira
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注: 24 pages, 6 figures, 1 table
Abstract:The Internal Waves Service screens the Sentinel-1 Wave-mode archive for internal solitary waves, routing detections to experts whose adjudication time is the resource the effort exists to conserve. Because attention is the cost of error, precision leads. Its classifier was trained and reported at a one-to-one class balance, fixed before the operational rate could be known. That rate has since emerged at roughly one scene in twenty, and a balanced-test score badly overstates the precision a validator meets. A model that scores 0.794 balanced-test precision scores 0.192 in real operation: the gap is a systematic artefact of reporting at the wrong prior, invisible to the metric most work quotes. We show the mismatch to be an evaluation problem in the costume of a training one at a fixed recall, prior correction and calibration cannot move precision, and answer it with a prior-matched reporting method based on three figures: balanced-test, operational-prior, and real post-deployment, whose contrast is the honest measure. A precision-first, leakage-controlled development cycle then improves the classifier lever by lever, each promoted only against a pre-registered margin; added capacity not clearing it, calibration inert, feature aggregation the one real lift, so the honest negatives are as much a result as the gain. Holding recall at a floor of 0.80 and certifying against a sealed, single-read lockbox, the promoted model reports 0.927 precision at the operational prior; an out-of-time check confirms discrimination transfers to unseen periods while a fixed operating point does not. Prior-matched reporting, begin balanced, then move to the prior as the stream reveals it, transfers to any operational Earth-observation service bootstrapping a rare-event detector under a prior it has yet to discover.
[CV-42] Sparse Attention for Dense Open-Vocabulary Prediction in CLIP
链接: https://arxiv.org/abs/2607.07135
作者: Fatimah Zohra,Chen Zhao,Shuming Liu,Bernard Ghanem
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Contrastive Language-Image Pre-training (CLIP) relies on softmax-based self-attention, a strictly positive distribution that assigns probability mass to every pair of tokens-even semantically irrelevant ones. While these dense softmax weights are effective for gathering broad context during pre-training, they spread attention across many low-salience tokens, producing noise that obscures the fine-grained, spatially localized cues required for dense, open-vocabulary prediction. We study an inference-time substitution of the row-wise softmax in the final visual self-attention layers with the \alpha -entmax transform, applied across both the standard query-key attention and self-correlation variants. Because entmax applies a data-dependent threshold that maps low scores exactly to zero, it acts as an implicit denoiser, zeroing contextually irrelevant dependencies while redistributing mass onto the most relevant tokens. We evaluate on open-vocabulary tasks-dense semantic segmentation (Pascal VOC, Pascal Context, ADE20K) and fine-grained retrieval (FG-OVD)-and find the gain from attention sparsification is proportional to how much the baseline attention spreads off the target class.
[CV-43] Widest-Path Reachability Fields for Connectivity-Preserving Slender Structure Segmentation
链接: https://arxiv.org/abs/2607.07123
作者: Youcheng Zong,Runda Jia,Minxuan Hu,Weilan Su,Dakuo He
类目: Computer Vision and Pattern Recognition (cs.CV); Systems and Control (eess.SY)
备注:
Abstract:Segmenting slender curvilinear structures such as retinal vessels, cracks, and roads demands topological correctness, as even a single-pixel discontinuity can fragment a continuous network and invalidate downstream analysis. Under standard binary-mask supervision, models optimized for pixel-level overlap frequently produce topologically broken predictions. We trace this to a fundamental mismatch: pixel-wise losses distribute gradients uniformly, yet connectivity hinges on a sparse set of bottleneck pixels. These pixels are vastly outnumbered by thick structures and background, rendering their aggregate gradient contribution negligible. We term this phenomenon topological gradient starvation (TGS). To address it, we propose Widest-Path Reachability Fields (WPRF), a differentiable Max-Min reachability objective that redirects gradient flow to connectivity bottlenecks. The module is plug-and-play, backbone-agnostic, and incurs no inference overhead. WPRF implements a differentiable Max-Min objective via dynamic programming on a domain-restricted graph, coupled with a bottleneck-aware observation term that balances gradient contributions across varying structures. Compared to prior topology-aware losses that rely on post-hoc skeletonization or homology computation, WPRF directly optimizes end-to-end reachability via differentiable Max-Min algebra, enabling gradient flow to concentrate on connectivity bottlenecks without auxiliary structures. We introduce OMVIS, a new oral microvessel segmentation dataset. Experiments across nine architectures and six datasets validate the bottleneck-focused gradient routing mechanism. WPRF improves 87% of experiments with fixed hyperparameters and achieves clDice gains of 7.2 percentage points on structurally fragile datasets.
[CV-44] ColorFM: An Optimization-to-Learning Framework for Color Transfer via Flow Matching ECCV2026
链接: https://arxiv.org/abs/2607.07119
作者: Yuhang He,Kai Zhang,Xiaoming Li,Du Chen,Jian Yang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ECCV 2026
Abstract:Color transfer aims to align the color distribution of a source image with that of a reference image while preserving structural and semantic consistency. However, existing methods often suffer from inaccurate global mapping, semantic misalignment, and visual artifacts. To address these issues, we propose ColorFM, an optimization-to-learning framework. ColorFM connects online optimization to offline inference by reformulating color transfer as the transport of pixel distributions along velocity fields via Flow Matching. Specifically, we introduce ColorFM-O, an instance-specific optimization scheme that fits the velocity field through hierarchical color coupling guided by semantic priors. By numerically integrating the induced flow trajectories, ColorFM-O produces precise and semantically consistent color transfer results, while generating high-quality paired data as pseudo-supervision. Building upon this, we design ColorFM-L, an efficient feed-forward model trained on the generated pairs. Through implicit state modeling, ColorFM-L extracts deep semantic features to predict flow parameters for bidirectional linearized transport, ensuring accurate color transfer. Extensive experiments demonstrate that ColorFM-L outperforms state-of-the-art methods in visual quality, structural fidelity, and semantic consistency, successfully combining the accuracy of optimization with the speed of feed-forward inference.
[CV-45] ree-of-Thoughts Reasoning for Text-to-Image In-Context Learning
链接: https://arxiv.org/abs/2607.07117
作者: Stepanida Alekseeva,Jenifer Kalafatovich,Seong-Whan Lee
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 6 pages, 3 figures, 4 tables. Accepted at IEEE SMC 2026. Code available at this https URL
Abstract:In text-to-image in-context learning (T2I-ICL), a model has to infer a latent compositional pattern from fewshot demonstrations for generating a query image. Recent studies show that state-of-the-art multimodal large language models struggle with this setting, particularly due to limited compositional reasoning and sensitivity to prompt construction. In this work, we propose a Tree-of-Thoughts (ToT) reasoning framework for T2I-ICL that introduces a multi-stage reasoning and selection layer that generates, evaluates, and selects among multiple candidate hypotheses before constructing the final prompt for image synthesis. By exploring alternative reasoning branches and selecting a coherent interpretation, the proposed approach mitigates prompt ambiguity and compositional errors. We implement the proposed approach in a complete ToT-T2IICL inference pipeline and evaluate it on the CoBSAT benchmark. Both qualitative and quantitative results show that structured multi-branch reasoning leads to more consistent and semantically aligned image generation compared to baseline and Chain-of-Thought prompting strategies, without any additional training or fine-tuning.
[CV-46] AT-Attn: Temporal-Aware Cross-Attention for Longitudinal Multimodal Alzheimers Disease Diagnosis
链接: https://arxiv.org/abs/2607.07091
作者: Xinyue Du,Yibo Liu,Zhenglei Zhou,Xuancheng Yao,Weimin Zhong,Qiuhui Chen
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Submitted to IEEE BIBM 2026. 8 pages, 4 figures
Abstract:In longitudinal Alzheimer’s disease (AD) diagnosis support, clinical and imaging information is often collected at irregular visits. Integrating these multimodal observations may improve diagnostic assessment, but naive fusion can degrade performance when MRI is noisy or intermittently unavailable. We propose AT-Attn, a temporal-aware multimodal framework that combines Change-and-Time encoding, time-biased asymmetric cross-attention, and gated fusion to integrate MRI with longitudinal clinical information. We evaluate AT-Attn on an MRI-retained ADNI cohort of 1,520 patients using structural MRI, six cognitive-scale trajectories, and seven static clinical variables under patient-level five-fold cross-validation. The main asymmetric AT-Attn model achieves accuracy 0.719+/-0.024, macro F1 0.721+/-0.023, ROC-AUC 0.873+/-0.013, and PR-AUC 0.783+/-0.018, outperforming unimodal and naive multimodal fusion baselines while remaining competitive with strong tabular baselines. These results suggest that a temporal-aware and constrained fusion strategy can help structural MRI contribute clinically relevant complementary information for patient-level AD diagnosis support.
[CV-47] Navigating Hierarchy: Hyperbolic Learning on Brain Graphs for Disorder Diagnosis
链接: https://arxiv.org/abs/2607.07077
作者: Yapeng Li,Bo Jiang,Ziyan Zhang,Dongdong Chen,Zhengzheng Tu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 12 pages, 5 figures
Abstract:Functional brain networks exhibit a hierarchical organization across ROI, community, and whole-brain levels, supporting local processing, inter-community coordination, and global integration. Recent studies have demonstrated that brain community-aware modeling is beneficial for both diagnosis and biomarker identification of brain networks. However, existing brain graph modeling methods often struggle to model ROI-community interactions, thereby failing to fully exploit the hierarchy across ROI, community, and whole-brain network levels. To address this issue, inspired by deep hyperbolic learning in modeling hierarchical structures, we propose a novel framework, termed Hyperbolic Learning on Brain Graphs (HLBG), for brain network analysis. The core idea of HLBG is to exploit the inherent hierarchical geometry of hyperbolic space to model the hierarchical relationships among ROIs, functional communities, and the whole-brain network, thereby learning hierarchy-aware and highly discriminative representations for brain network data. Specifically, HLBG first projects representations from ROIs, communities, and the whole-brain network into Lorentzian hyperbolic space. Then, the multi-level hierarchy is imposed via two geometric entailment constraints. In addition, we introduce a new Graph-aware Mamba (GaMamba) model, which incorporates topology-derived structural prompts into Mamba to capture long-range dependencies while preserving graph topological information. Experiments on ABIDE-I and REST-MDD demonstrate that HLBG outperforms state-of-the-art methods and identifies disorder-relevant functional biomarkers.
[CV-48] Making Implicit Preservation Intent Explicit in Conversational Image Editing
链接: https://arxiv.org/abs/2607.07051
作者: Soomin Han,Jihyung Ahn,Bumsoo Kim,Buru Chang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Conversational image editing requires preserving not only visible content, but also content that temporarily disappears across turns. When newly added or modified content occludes a previously visible region, that region should reappear if it was never semantically changed. However, existing systems often fail to recover such occluded-but-unchanged content, producing inconsistent or hallucinated results. We introduce OCCUR-Bench, a diagnostic benchmark for temporal preservation in conversational image editing. OCCUR-Bench provides diverse occlusion-and-revelation scenarios with historical restoration references, enabling evaluation of faithful restoration rather than plausible regeneration. We also propose ReSpec, a training-free framework that makes implicit preservation explicit by pairing restoration-aware instructions with historical visual references. Given an editing history, ReSpec identifies what should persist, selects the historical image state that provides missing visual evidence, and conditions an in-context editor on the resulting instruction and reference image. Experiments show that ReSpec improves restoration fidelity and temporal consistency on OCCUR-Bench, highlighting the need to ground preservation in editing history rather than only the current image.
[CV-49] RACE-Seg3D: Counterfactual Context Auditing For Robust 3D Glioma Segmentation Under Institutional Shift
链接: https://arxiv.org/abs/2607.07038
作者: Nguyen Linh Dan Le,Nguyen Pham Hoang Le,Tran Dang Khoi
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 16 pages, 5 figures
Abstract:Medical image segmentation models can achieve strong benchmark performance while remaining sensitive to scanner, protocol, and institutional variation. These context shifts alter image appearance without changing the underlying lesion, allowing models to exploit nuisance cues that Dice and HD95 fail to expose. We present TRACE-Seg3D, a counterfactual context auditing framework for robust 3D medical image segmentation. TRACE-Seg3D preserves lesion-relevant evidence and systematically varies imaging context to quantify prediction stability under controlled context shifts. The framework pairs each segmentation with audit evidence for context sensitivity and anatomical plausibility, enabling case-level reliability assessment beyond overlap-based evaluation. Experiments on BraTS and UTSW glioma segmentation benchmarks demonstrate competitive in-distribution and cross-domain performance. TRACE-Seg3D also exposes context-sensitive failure modes missed by conventional metrics. These results establish counterfactual context auditing as a practical route toward transparent and reliable 3D medical image segmentation under distribution shift. Our code is available at this https URL.
[CV-50] AnchorPrune: Relevance-Anchored Contextual Expansion for Visual Token Pruning ECCV2026
链接: https://arxiv.org/abs/2607.07033
作者: Kyuan Oh,Bumsoo Kim
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: ECCV 2026
Abstract:Large vision-language models incur substantial inference costs because high-resolution inputs introduce thousands of visual tokens, many of which are redundant for a given query. Existing pruning methods often combine query relevance and token diversity, yet these objectives can conflict under aggressive compression: relevance-driven selection may overconcentrate the budget on correlated local evidence, while diversity-driven selection may suppress indispensable tokens or retain distinct but uninformative regions. We introduce AnchorPrune, a training-free framework that first constructs a protected relevance anchor and then expands it with complementary visual context. AnchorPrune adaptively determines the anchor size from the novelty profile of relevance-ranked tokens, preserving a compact set of query-critical evidence, and allocates the remaining budget through importance-weighted novelty to recover informative, non-redundant context relative to the anchor. This ordered design prevents contextual expansion from displacing indispensable query cues while improving overall visual coverage. AnchorPrune is lightweight, architecture-aware, and requires neither retraining nor model modification. Across image and video vision-language models and benchmarks, it consistently improves the accuracy-efficiency trade-off over training-free baselines, particularly under severe compression. On LLaVA-NeXT-7B, AnchorPrune preserves 97.6% of full-token performance using only 160 of 2,880 visual tokens. These results establish relevance-anchored contextual expansion as an effective principle for efficient multimodal inference. Code is available at this https URL.
[CV-51] Latent graph encoding of multimodal neuroimaging features with generative AI architectures ICIP
链接: https://arxiv.org/abs/2607.07027
作者: Ishaan Batta,Meenu Ajith,Vince Calhoun
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: 6 pages, accepted in IEEE International Conference on Image Processing (ICIP) 2026
Abstract:While generative models enable encoding of complex neuroimaging data for feature generation and reconstruction, developing optimal architectural frameworks with appropriate encoding and latent space processes is crucial for studying structural and functional properties of the brain. We design a multimodal generative framework for structural and functional magnetic resonance imaging (MRI) features through systematic evaluation of encoding strategies, latent multimodal fusion, and generative model selection. Using structural gray matter volume (GMV) and static functional network connectivity (sFNC) features from a large neuroimaging dataset, we analyze generative frameworks involving variational autoencoders (VAEs), transformers, generative adversarial networks (GANs), and diffusion models. Architectures that employ modality-aware graph encoding of functional connectivity into a lower-dimensional latent space outperform vectorized encoders or direct data space approaches. The proposed multimodal graph VAE (gMMVAE) surpasses alternative generative variants across multiple metrics for generation fidelity, reconstruction quality, efficiency, and latent space discriminability, highlighting its potential for robust multimodal neuroimaging analysis.
[CV-52] SHTA: Semantic Hard Token Correction and Center Alignment for Semi-Supervised Medical Image Segmentation
链接: https://arxiv.org/abs/2607.07019
作者: Zhuoru Zhang,Yiheng Zhong,Zimu Zhang,Xiaofeng Liu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages, 7 figures, 4 tables
Abstract:Recent advances in semi-supervised medical image segmentation have achieved remarkable performance through prediction consistency, pseudo-label supervision, and hard-region supervision. However, these methods primarily improve supervision quality rather than explicitly enforcing semantic consistency in the learned representations of hard regions. Consequently, even under increasingly stronger prediction-level supervision, difficult regions exhibiting unstable semantic assignment often fail to establish semantically consistent representations during training, thereby limiting further segmentation improvement. To address this issue, we propose SHTA (Semantic Hard Token Correction and Center Alignment), a lightweight training-time semantic representation branch. Instead of introducing additional prediction supervision, SHTA refines intermediate semantic representations through Semantic Assignment, Hard Token Refinement, and Semantic Center Alignment, thereby improving semantic consistency in hard regions while preserving the original prediction pathway and introducing no additional inference cost. We integrate SHTA into representative semi-supervised segmentation frameworks, including GA-CPS, CPS, URPC, and MagicNet, and conduct evaluations on the Synapse and AMOS datasets. Experimental results demonstrate that SHTA delivers consistent paired improvements across frameworks, with especially clear gains in segmentation accuracy, weak-organ recovery, and semantic ambiguity reduction, while incurring only training-time overhead. The code is available at this https URL.
[CV-53] Ego-Human Motion Prediction with 3D-Aware LLM ECCV2026
链接: https://arxiv.org/abs/2607.07001
作者: Yujin Bae,Jaewoo Jeong,Hyeonseong Kim,Kuk-Jin Yoon
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ECCV 2026
Abstract:Anticipating human motion from an egocentric perspective is fundamental for proactive assistance in AR/VR, human-robot collaboration, and embodied AI. While recent works incorporate language as a semantic prior to reduce the ill-posed nature of egocentric forecasting, they largely neglect the 3D spatial and semantic context that governs how motion unfolds, and treat pose and language prediction as separate inference streams. We introduce Ego3DLM, built on two core principles: accurate motion forecasting requires explicit spatial and semantic understanding of the 3D environment, and pose and language must be predicted holistically in a single pass, since motion is inherently tied to the semantic interpretation of actions being performed. Given three-point tracking, 3D scene features, and egocentric video, Ego3DLM simultaneously decodes past pose, future pose, past narration, and future narration in a single autoregressive pass, grounding predicted poses and descriptions in one another to enforce cross-modal and temporal consistency. We adopt a three-stage training scheme: (1) spatial-semantic scene awareness pretraining; (2) holistic instruction tuning over all four outputs in a single pass; and (3) GRPO-based reinforcement finetuning with intra- and inter-modal rewards that directly optimize pose-language fidelity. Experiments on the Nymeria benchmark demonstrate that Ego3DLM achieves state-of-the-art performance across future motion prediction, past motion tracking, and motion description, showing that 3D scene grounding and holistic cross-modal prediction yield physically plausible and semantically coherent motion forecasts. The project page is available at this https URL.
[CV-54] EdgeCompress: Coupling Multidimensional Model Compression and Dynamic Inference for EdgeAI
链接: https://arxiv.org/abs/2607.06982
作者: Hao Kong,Di Liu,Shuo Huai,Xiangzhong Luo,Ravi Subramaniam,Christian Makaya,Qian Lin,Weichen Liu
类目: Computer Vision and Pattern Recognition (cs.CV); Hardware Architecture (cs.AR); Machine Learning (cs.LG)
备注: Author’s accepted version. Published in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD)
Abstract:Convolutional neural networks (CNNs) have demonstrated encouraging results in image classification tasks. However, the prohibitive computational cost of CNNs hinders the deployment of CNNs onto resource-constrained embedded devices. To address this issue, we propose EdgeCompress, a comprehensive compression framework to reduce the computational overhead of CNNs. In EdgeCompress, we first introduce dynamic image cropping (DIC), where we design a lightweight foreground predictor to accurately crop the most informative foreground object of input images for inference, which avoids redundant computation on background regions. Subsequently, we present compound shrinking (CS) to collaboratively compress the three dimensions (depth, width, and resolution) of CNNs according to their contribution to accuracy and model computation. DIC and CS together constitute a multidimensional CNN compression framework, which is able to comprehensively reduce the computational redundancy in both input images and neural network architectures, thereby improving the inference efficiency of CNNs. Further, we present a dynamic inference framework to efficiently process input images with different recognition difficulties, where we cascade multiple models with different complexities from our compression framework and dynamically adopt different models for different input images, which further compresses the computational redundancy and improves the inference efficiency of CNNs, facilitating the deployment of advanced CNNs onto embedded hardware. Experiments on ImageNet-1K demonstrate that EdgeCompress reduces the computation of ResNet-50 by 48.8% while improving the top-1 accuracy by 0.8%. Meanwhile, we improve the accuracy by 4.1% with similar computation compared to HRank, the state-of-the-art compression framework. The source code and models are available at this https URL
[CV-55] HPR-SAM: Hierarchical Probabilistic Representation Learning for Prompt-free SAM-based Medical Image Segmentation
链接: https://arxiv.org/abs/2607.06972
作者: Yingzhen Hu,Yiheng Zhong,Keying Zhu,Zimu Zhang,Zihan Ye,Sifan Song,Jionglong Su,Xiaofeng Liu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages, 4 figures
Abstract:Prompt-free adaptation of the Segment Anything Model (SAM) has emerged as a promising paradigm for automatic medical image segmentation. Existing methods mainly focus on prompt generation, while overlooking that prompt quality is fundamentally constrained by the expressiveness of anatomical representations. However, deterministic prototypes or semantic tokens are insufficient to jointly capture global anatomical priors, intra-structure diversity, and local structural reliability. To address this limitation, we propose the Hierarchical Probabilistic Representation (HPR) framework, which learns complementary anatomical representations through Distributional Anatomical Representation (DAR), Multi-component Anatomical Representation (MAR), and Local Reliability Representation (LRR), and integrates their predictions via Hierarchical Prediction Fusion (HPF) while remaining compatible with the original SAM decoder. Experiments on the Synapse, LA, and PROMISE12 datasets demonstrate that HPR-SAM achieves state-of-the-art performance on Synapse and the best performance under few-shot settings on LA and PROMISE12, validating the effectiveness of the proposed hierarchical probabilistic representation learning framework for prompt-free medical image segmentation. Code is available at this https URL.
[CV-56] SpiS-GAN: Spiral-Modulated Handwriting Synthesis with Star Operation
链接: https://arxiv.org/abs/2607.06949
作者: Nguyen Duy Hieu,Dang Hoai Nam,Pham Hoang Giap,Quang Huu Hieu,Vo Nguyen Le Duy
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Training robust handwriting recognition (HTR) systems requires massive amounts of annotated data, which is often difficult to acquire. While synthetic handwriting generation offers a practical solution to expand training sets, existing models struggle with several core issues. First, previous approaches, even MLP-based models fail to effectively trace cursive handwriting due to fixed-grid spatial receptive field. Second, their CNN-relied discriminators usually lose structural details through aggressive downsampling, making broken connections difficult to detect. Third, existing architectures are either limited to linear feature interactions or too expensive for high-resolution synthesis. Finally, existing approaches lack explicit edge constraints, often resulting in blurred stroke boundaries. To address these challenges, this study proposes a Spiral-Modulated Handwriting Synthesis framework based on Generative Adversarial Networks (SpiS-GAN). Our generator employs Star-Spiral Blocks combining proposed Modulated Elliptical SpiralFC with the star operation to capture spatial relationships and efficiently follow complex handwriting stroke trajectories, while a Spiral-Modulated discriminator is introduced for multi-domain flaws detection. Additionally, we introduce a Sobel-Regularized Edge Reconstruction Loss that provides edge guidance, ensuring every character remains clear and legible. Evaluations on the English and Vietnamese datasets demonstrate that SpiS-GAN significantly outperforms current state-of-the-art models. The generated images are highly authentic, accurately preserve the original writer’s style across languages, and successfully lower error rates when training downstream HTR systems.
[CV-57] Self-Supervised Pretraining Improves Cross-Site and Cross-Scale Robustness of Point Cloud Leaf-Wood Segmentation
链接: https://arxiv.org/abs/2607.06948
作者: Heeju Mun,Tackang Yang,Yunsoo Nam,Changhyun Choi
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 30 pages, 10 figures
Abstract:The accuracy of existing leaf-wood segmentation methods for tree point clouds varies across forest types and sites. Self-supervised learning (SSL) on point clouds has improved the generalization of deep learning models for forestry point cloud tasks, including biomass regression and individual tree segmentation, but its applicability to leaf-wood segmentation remains untested. In this study, we pretrained Point-M2AE, a widely used SSL architecture for point clouds, on ShapeNet-55 augmented with 2,400 individual tree point clouds. For fine-tuning and inference, we used recursive voxel subdivision to handle the wide variation in point density across inputs, allowing the same model to operate at both individual-tree and plot scales without architecture change. Compared to the model without pretraining, the pretrained model improved wood IoU from 60.5% to 70.0% for needleleaf and from 69.7% to 76.3% for broadleaf trees. On a benchmark spanning four countries across three climatic zones, the pretrained model achieved the smallest cross-site variation and highest overall performance among compared methods (LeWos, CWLS, and PointTransformer). Plot-level segmentation maintained accuracy comparable to individual-tree performance, with mIoU of 84.7% for broadleaf and 77.7% for needleleaf plots, showing that the model generalizes across scales without additional finetuning. As a downstream test in tropical forests, where dense canopies make segmentation challenging, we applied our model and a quantitative structure model to estimate wood volume for 28 trees from Guyana, Indonesia, and Peru to assess whether the segmentation improvements from SSL pretraining translate into improved downstream performance. The resulting volume estimates achieved the lowest error among all methods tested (MAE = 2.40 m ^3 ), less than half that of algorithmic baselines (LeWos: 5.94 m ^3 ; CWLS: 5.27 m ^3 ).
[CV-58] General Incomplete Multimodal Learning via Dynamic Quality Perception ECCV2026
链接: https://arxiv.org/abs/2607.06943
作者: Xiangyu Meng,Shicai Wei
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by ECCV 2026. Corresponding author: Shicai Wei
Abstract:Multimodal learning robust to missing modalities is essential for real-world applications. Existing methods mainly focus on inter-modality missing, where entire modalities are absent, while overlooking intra-modality degradation, where modalities are present but severely corrupted. In practice, these two types of missing often coexist, making existing approaches ineffective. To address this limitation, we propose General Incomplete Multimodal Learning (GIML), a unified framework that simultaneously handles both inter-modality missing and intra-modality degradation through dynamic quality perception. Specifically, GIML models heterogeneous missing patterns as continuous modality information degradation, enabling degradation-aware adaptive fusion. To achieve reliable quality perception, we introduce a Noise-aware Quality Estimator that learns the mapping from corrupted features to noise intensity through controlled noise injection. Furthermore, we propose a Noise-Semantic Decoupled module that separates semantic information from noise interference. This improves robustness and generalization to unseen corruption patterns. Extensive experiments across datasets with diverse modality types demonstrate the effectiveness and generality of GIML. Code is available at: this https URL.
[CV-59] Bi-PT: Bidirectional Cross-Attention Point Transformers for Four-Chamber Heart Reconstruction from Sparse Cardiac MRI Data
链接: https://arxiv.org/abs/2607.06923
作者: Chenchuhui Hu,Shaoming Pan,Leon Axel,Meng Ye
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:We propose Bi-PT, a pipeline for reconstructing 3D four-chamber human heart meshes from clinical sparsely sampled cardiac magnetic resonance imaging (CMR) data. This work addresses the error-prone generation of 3D cardiac shape from a sparse point cloud (SPC) extracted from 2D long-axis and short-axis views used in routine clinical CMR protocols. Bi-PT enables accurate inference of the four-chamber heart mesh from the SPC by learning robust point features via bidirectional point cross-attention between an atlas and the SPC, together with per-point semantic labels that improve correspondence estimation. We formulate the deformation field as a Neural Ordinary Differential Equation (NODE) parameterized by a per-point affine transformation and translation to deform the atlas toward the target heart shape. By learning such a NODE, we can guarantee the deformation field to be a locally affine diffeomorphic deformation. We also integrate a semantic label loss into the Chamfer distance to encourage label-consistent correspondences and add a smoothness regularization to stabilize and improve the learning of the deformation field. Extensive experiments demonstrate that Bi-PT achieves accurate and robust performance compared to baselines.
[CV-60] Latency-Constrained DNN Architecture Learning for Edge Systems using Zerorized Batch Normalization
链接: https://arxiv.org/abs/2607.06922
作者: Shuo Huai,Di Liu,Hao Kong,Weichen Liu,Ravi Subramaniam,Christian Makaya,Qian Lin
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注: 15 pages. Author’s accepted manuscript, published in Future Generation Computer Systems
Abstract:Deep learning applications have been widely adopted on edge devices, to mitigate the privacy and latency issues of accessing cloud servers. Deciding the number of neurons during the design of a deep neural network to maximize performance is not intuitive. Particularly, many application scenarios are real-time and have a strict latency constraint, while conventional neural network optimization methods do not directly change the temporal cost of model inference for latency-critical edge systems. In this work, we propose a latency-oriented neural network learning method to optimize models for high accuracy while fulfilling the latency constraint. For efficiency, we also introduce a universal hardware-customized latency predictor to optimize this procedure to learn a model that satisfies the latency constraint by only a one-shot training process. The experiment results reveal that, compared to state-of-the-art methods, our approach can well-fit the ‘hard’ latency constraint and achieve high accuracy. Under the same training settings as the original model and satisfying a 34 ms latency constraint on the ImageNet-100 dataset, we reduce GoogLeNet’s latency from 40.32 ms to 34 ms with a 0.14% accuracy reduction on the NVIDIA Jetson Nano. When coupled with quantization, our method can be further improved to only 0.04% drop for GoogLeNet. On the NVIDIA Jetson TX2, we compress VGG-19 from 119.98 ms to 34 ms and even improve its accuracy by 0.5%, and we scale GoogLeNet up from 20.27 ms to 34 ms and achieve higher accuracy by 0.78%. We also open source this framework at this https URL
[CV-61] Compass: Prostate Cancer Detection Needs Multi-View Context MICCAI2026
链接: https://arxiv.org/abs/2607.06919
作者: Paul F.R. Wilson,Mohamed Harmanani,Zhuoxin Guo,Obed K. Dzikunu,Hannes Cash,Adam Kinnaird,Brian Wodlinger,Purang Abolmaesumi,Parvin Mousavi
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: MICCAI 2026
Abstract:Artificial intelligence (AI) analysis of micro-ultrasound ( \mu US) has shown promise for prostate cancer (PCa) detection. However, most existing AI methods focus on the analysis of single \mu US images in isolation. By contrast, expert \mu US readers typically assess a full recorded video study, which provides three-dimensional context, to improve PCa detection compared to single-frame analysis. Inspired by this clinical workflow, we propose Compass, a novel AI methodology which models a \mu US study as a stream of 2D images. Compass jointly integrates rotational sweep videos of the prostate with \mu US frames acquired at the moment of biopsy, and performs evidence aggregation across the study using a transformer conditioned on the probe’s rotational angle. Finally, a decoder head predicts frame-level and study-level risk scores for the patient. The model is trained and evaluated using a multi-center clinical trial dataset of \mu US studies, including continuous rotational scans of the prostate and videos captured during biopsy acquisition. We compare the proposed method to baseline AI methods from the literature and to risk scores provided by clinical experts. Our framework shows strong performance, highlighting the value of multi-view context for \mu US PCa detection, and providing a potentially powerful tool to complement human expertise in \mu US-based PCa diagnosis. Our code is available at: this https URL.
[CV-62] LoCA: Spatially-Aware Low-Rank Convolutional Adaptation of Vision Foundation Models ECCV2026
链接: https://arxiv.org/abs/2607.06918
作者: Sojung An,Junha Lee,Sujeong You,Nam Ik Cho,Donghyun Kim
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted by ECCV 2026
Abstract:Pre-trained Vision Foundation Models (VFMs) provide strong visual representations for diverse downstream tasks. The key challenge of VFM adaptation stems from the prohibitive costs of full fine-tuning and catastrophic forgetting. To address this, Low-Rank Adaptation (LoRA) has emerged as the prevailing paradigm for Parameter-Efficient Fine-Tuning (PEFT). However, LoRA is typically designed for transformer self-attention layers parameterized by 2D matrices. Since convolutional kernels inherently couple spatial and channel information within a 4D tensor, forcing them into a monolithic 2D matrix disrupts the inherent spatial topology. In this paper, we propose Low-Rank Convolutional Adaptation (LoCA), a convolution-aware PEFT framework that addresses spatial-channel entanglement by decoupling channel and spatial adaptation. LoCA introduces a low-rank channel adaptation for dense cross-channel mixing and refines spatial bases extracted from pre-trained kernels via Singular Value Decomposition (SVD). Experimental results show that LoCA preserves pre-trained spatial priors and achieves competitive or state-of-the-art performance across fine-grained classification, domain-generalized semantic segmentation, and generative benchmarks.
[CV-63] Smart Scissor: Coupling Spatial Redundancy Reduction and CNN Compression for Embedded Hardware
链接: https://arxiv.org/abs/2607.06915
作者: Hao Kong,Di Liu,Shuo Huai,Xiangzhong Luo,Weichen Liu,Ravi Subramaniam,Christian Makaya,Qian Lin
类目: Computer Vision and Pattern Recognition (cs.CV); Hardware Architecture (cs.AR); Machine Learning (cs.LG)
备注: 9 pages, 9 figures. Author’s version, accepted by and published in ICCAD 2022. Copyright 2022 ACM
Abstract:Scaling down the resolution of input images can greatly reduce the computational overhead of convolutional neural networks (CNNs), which is promising for edge AI. However, as an image usually contains much spatial redundancy, e.g., background pixels, directly shrinking the whole image will lose important features of the foreground object and lead to severe accuracy degradation. In this paper, we propose a dynamic image cropping framework to reduce the spatial redundancy by accurately cropping the foreground object from images. To achieve the instance-aware fine cropping, we introduce a lightweight foreground predictor to efficiently localize and crop the foreground of an image. The finely cropped images can be correctly recognized even at a small resolution. Meanwhile, computational redundancy also exists in CNN architectures. To pursue higher execution efficiency on resource-constrained embedded devices, we also propose a compound shrinking strategy to coordinately compress the three dimensions (depth, width, resolution) of CNNs. Eventually, we seamlessly combine the proposed dynamic image cropping and compound shrinking into a unified compression framework, Smart Scissor, which is expected to significantly reduce the computational overhead of CNNs while still maintaining high accuracy. Experiments on ImageNet-1K demonstrate that our method reduces the computational cost of ResNet50 by 41.5% while improving the top-1 accuracy by 0.3%. Moreover, compared to HRank, the state-of-the-art CNN compression framework, our method achieves 4.1% higher top-1 accuracy at the same computational cost. The codes and data are available at this https URL
[CV-64] Seeing What Matters: Lesion-Aware High-Resolution Patch Discovery and Fusion for Chest X-ray Report Generation
链接: https://arxiv.org/abs/2607.06909
作者: Yingshu Li,Yunyi Liu,Zhenghao Chen,Tong Chen,Zailong Chen,Lingqiao Liu,Lei Wang,Luping Zhou
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Despite rapid advances in chest X-ray (CXR) foundation models, most radiology report generation (RRG) systems still rely on heavily downsampled inputs (e.g., 256x256) due to the fixed visual token budgets of pretrained vision encoders, suppressing subtle yet clinically important cues present in native-resolution images. However, enabling high-resolution (high-res) perception remains challenging: naive tiling causes prohibitive token inflation, while global compression suppresses subtle lesions and degrades diagnostic fidelity. Inspired by radiologists’ workflow, localizing suspicious regions before detailed high-res assessment. We propose Lesion-Aware High-Resolution Patch Discovery and Fusion for Chest X-ray Reporting (LePaX), the first RRG framework that enables efficient high-res CXR perception (up to 1920x1920) without increasing the vision-token count. LePaX formulates high-res perception as a constrained spatial resolution allocation problem under a fixed token budget and introduces two key components: Learnable Spatial Resolution Allocation (LSRA), which learns a spatial utility map that adaptively allocates limited high-res capacity to diagnostically relevant regions, enabling targeted extraction of high-res patches from native CXRs; and Global-Regional Fusion (GRF), which performs token-preserving region-to-global refinement by projecting high-resolution regional evidence back onto the global feature grid through spatially aligned resolution write-back, avoiding token inflation. Experiments on multiple CXR benchmarks demonstrate that LePaX consistently improves both clinical and linguistic metrics while enabling native-resolution CXR perception with over 10x fewer visual tokens than naive high-res tiling.
[CV-65] Dynamic Object Detection and Tracking in Construction: A Fisheye Camera and LiDAR Sensor Fusion Model ICRA
链接: https://arxiv.org/abs/2607.06896
作者: Yilong Chen,Huili Huang,Yong K. Cho
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 4 pages, 8 figures, submitted to IEEE International Conference on Robotics and Automation (ICRA) 2025 Future of Construction Workshop
Abstract:Robust dynamic object detection and tracking are essential for enabling robots to operate safely and effectively alongside humans in complex environments such as construction sites. While LiDAR-based SLAM and occupancy grid methods offer viable solutions for detecting and tracking motion, many state-of-the-art 3D vision approaches rely heavily on pre-trained neural networks and require additional post-processing to identify moving objects. Sensor fusion techniques, combining the precision of LiDAR with the semantic richness of RGB imagery, offer a promising alternative. In this work, we present a novel framework that enhances a quadruped robot equipped with a LiDAR sensor and an upward-facing fisheye camera for real-time dynamic object detection and tracking. After identifying moving objects within a registered point cloud, our method assigns semantic labels by projecting 3D coordinates onto a 2D cylindrical panorama, aligning with real-time image-based detections for observation update of the Kalman filter. The proposed system demonstrates high precision, simplicity, and robustness, particularly in handling objects transitioning between dynamic and static states, thus it is well-suited for deployment in real-world construction environments.
[CV-66] ReMoDEx: A Local-to-Global Relevance-Based Model Decision Explainability Framework for large-Scale Image Datasets
链接: https://arxiv.org/abs/2607.06889
作者: Abhay Kumar Pathak,Mrityunjay Chaubey,Manjari Gupta
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Deep learning image classifiers achieve strong predictive performance yet remain opaque in how decisions are formed. A model may predict correctly while relying on irrelevant cues, shortcut associations, peripheral structures, or device level artifacts instead of task relevant regions. On large scale datasets this opacity is especially problematic, since inspecting heatmaps one sample at a time cannot scale to thousands of predictions. We propose Relevance Based Model Decision Explainability (ReMoDEx), a framework for systematic, dataset scale assessment of model decision behaviour in image classification. ReMoDEx defines a stepwise pipeline: model inference, target class selection, relevance map generation, heatmap standardisation, similarity based grouping of patterns, cluster level interpretation, and spatial relevance assessment. Local methods GradCAM++, Integrated Gradients, Occlusion Sensitivity, and Layerwise Relevance Propagation are each combined independently with a single global module that summarises an entire set of relevance maps into a few decision strategy clusters, replacing sample by sample inspection with an automatic, scalable summary. To demonstrate ReMoDEx, we applied it to a VGG16 based classifier distinguishing COVID-19, Normal, Lung Opacity, and Viral Pneumonia. The classifier showed stable performance (86.27% test accuracy, 0.9624 test AUC). However, each explainer combined with the global module consistently produced two recurring strategies: central thoracic region decisions and border/corner sensitive decisions, indicating possible shortcut learning that conventional metrics could not reveal. Masked image validation confirmed that model confidence and predicted class changed when central or peripheral regions were occluded. ReMoDEx thus provides a scalable relevance based decision assessment framework and an essential complement to accuracy based evaluation.
[CV-67] Video2Reaction: Mapping Video to Audience Reaction Distribution in the Wild
链接: https://arxiv.org/abs/2607.06875
作者: Trang Nguyen,Sidong Zhang,Shiv Shankar,Gauri Jagatap,Deepak Chandran,Andrea Fanelli,Madalina Fiterau
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Understanding and forecasting audience reactions to video content are crucial for improving content creation, recommendation systems, and media analysis. To enable audience reaction prediction and other content engagement applications, we introduce \textbfVideo2Reaction , a multimodal dataset that maps short movie segments to a distribution of \textitinduced emotions of viewers in the wild, as expressed through social media. \textbfVideo2Reaction spans more than 10,000 videos and serves as a reliable benchmark as well as a training resource for audience reaction prediction. To enable cost-effective continuous annotations as reactions may change over time, we develop a two-stage multi-agent pipeline using only open-source LLMs, achieving 86% correctness under blind human verification despite the inherently noisy and subjective nature of the task. We establish the first benchmark for video-to-reaction-distribution prediction in the wild and show that pretrained foundation video models fail in zero-shot settings, while finetuning transforms them into state-of-the-art predictors capable of modeling both full reaction distributions and dominant responses from video alone. However, the task remains challenging: even the strongest methods achieve only 77% Top-3 F1 in dominant reaction prediction (LLaVA-Next), highlighting a substantial gap in modeling collective audience reaction. \modificationDataset and code are available at our project page: this https URL
[CV-68] Ensemble Deep Learning Approaches for AI-Altered Video Detection
链接: https://arxiv.org/abs/2607.06872
作者: Laiba Khan,Hung-Mao Wu,Wei Lin,Frank Bi,Yousef Abdelhadi,Joshua Jung
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:The increasing accessibility of artificial intelligence has led to a rapid rise in AI-generated videos, making it more difficult to distinguish between real and manipulated content. Many existing detection methods rely on a single model and often struggle to generalize across different types of deepfakes. In this work, we developed a multimodal deepfake detection system that combines both audio and visual analysis using an ensemble of models. The system includes AASIST for audio-based detection, and EfficientNet, XceptionNet, and MesoNet for analyzing visual features in video frames. The pipeline takes a video as input, separates the audio, and extracts face frames using MTCNN. Each model produces a score indicating the likelihood of the input being fake. These scores are then combined using ensemble strategies, including mean averaging and stacking. Mean fusion provides a simple and stable baseline, while stacking uses a trained meta-model to learn how to combine predictions more effectively. Results show that while individual models perform well on the datasets they were trained on, their performance drops when tested on more diverse datasets. The ensemble approach helps improve overall robustness by combining predictions from multiple models, leading to more consistent performance across different types of deepfakes. This suggests that using both audio and visual information together is a more reliable approach for deepfake detection. Our results highlight generalization to unseen manipulations as the central open challenge, with average accuracy around 70%.
[CV-69] Geometric Collapse: When Vision Models Fail to Verify Physical Causality ICML2026
链接: https://arxiv.org/abs/2607.06871
作者: Wentao Zhang,Jinhu Qi,Weiqiang Jin,Yifei Zhang,Chan-Tong Lam,Irwin King
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: ICML 2026
Abstract:Recent progress in large-scale self-supervised learning has improved dense geometric prediction, but it remains unclear whether such scaling yields inference-time physical plausibility checks. We propose Scrambled Edges, a controlled counterfactual that injects salient edge-like cues while violating surface continuity, illumination coherence, and occlusion ordering. With energy-matched and structure-matched controls, we isolate the effect of unsupported edge evidence from high-frequency energy and edge sparsity. Across CNN/ViT/SSL depth predictors on NYU Depth v2 and KITTI, Scrambled Edges induce up to 3.2x larger deviation from clean predictions than energy-matched noise; additional diffusion and flow-matching depth estimators show attenuated but still significant collapse. The resulting Geometric Collapse propagates globally: even with oracle knowledge of the corrupted region, output-level repair recovers only 47%, with substantial error outside the mask. These findings provide controlled behavioral evidence that current dense predictors lack reliable mechanisms to quarantine physically unsupported edge cues, motivating explicit plausibility scoring and selective cue integration.
[CV-70] Gen4U: Unifying Video Generation and Understanding via Diffusion
链接: https://arxiv.org/abs/2607.06856
作者: Michael King,Aravindh Mahendran,Matthew Koichi Grimes,Fedor Kitashov,Adham Elarabawy,Pedro Velez,Maks Ovsjanikov,Viorica Pătrăucean
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Prior work suggests that diffusion representations capture low-level geometry but struggle with high-level semantics. We demonstrate that state-of-the-art video diffusion models overcome this limitation. By systematically probing their intermediate activations using recent mutual-kNN alignment metrics, we reveal a highly structured latent space where visual representations evolve across both network depth and noise levels. We show that while moderate noise levels yield linearly separable global semantics, fine-grained details persist at lower noise levels but become spatially scattered, requiring attention mechanisms to decode. Building on these insights, we introduce Gen4U (Generation for Understanding), a framework that repurposes these generative representations with a single forward pass. Our experiments establish that frozen, large-scale video diffusion models function as highly competitive video encoders across a wide spectrum of tasks, spanning semantic and non-semantic objectives (video classification, depth estimation, camera pose estimation, image and video captioning). Bypassing fine-tuning, Gen4U unifies the generation and understanding paradigms, achieving strong perception performance while fully preserving the model’s ability to generate high-quality video.
[CV-71] Retrieving and Refining Winning Noise Tickets for Diffusion-Based Motion Generation ECCV2026
链接: https://arxiv.org/abs/2607.06843
作者: Sakuya Ota,Qing Yu,Kent Fujiwara,Satoshi Ikehata,Ikuro Sato
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ECCV 2026, Project page: this https URL
Abstract:Diffusion-based text-to-motion models synthesize realistic human motions but often exhibit semantic drift from the input text. Motion is inherently temporal, especially in compositional and long-duration sequences that require semantic consistency across multiple action segments and smooth kinematic transitions throughout the trajectory. We posit that the initial noise is central to this consistency: within the Gaussian noise space, certain instances, i.e. winning noise tickets, carry latent structure that biases denoising toward particular motion semantics, even under null prompts. We propose WInning Noise Retrieval and Optimization (WINRO), a training-free, model-agnostic framework that improves text-motion alignment by selecting and refining such tickets before diffusion sampling. WINRO maps random noises to motion features generated under null prompts, retrieves the best-aligned noise for a given text, and refines it via a KL-regularized objective that reduces the residual semantic gap while preserving the Gaussian prior. An optional LoRA-based adapter amortizes this refinement into a single forward pass. WINRO consistently improves text-motion fidelity across different base models, MDM and MotionLCM, on HumanML3D without retraining, improves temporal robustness on the MTT benchmark, and generalizes to applications such as motion stylization and spatial constraint satisfaction.
[CV-72] LEMUR 2: Unlocking Neural Network Diversity for AI
链接: https://arxiv.org/abs/2607.06839
作者: Tolgay Atinc Uzun,Waleed Khalid,Saif U Din,Sai Revanth Mulukuledu,Akashdeep Singh,Chandini Vysyaraju,Raghuvir Duvvuri,Avi Goyal,Yashkumar Rajeshbhai Lukhi,Muhammad A. Hussain,Krunal Jesani,Usha Shrestha,Yash Mittal,Roman Kochnev,Pritam Kadam,Mohsin Ikram,Harsh R. Moradiya,Alice Arslanian,Dmitry Ignatov,Radu Timofte
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注: 10 pages, 9 figures, 1 table
Abstract:Existing NAS benchmarks (e.g., NAS-Bench, NATS-Bench) cover only narrow, task-specific regions of the architectural design space and lack cross-domain or deployment-aware evaluation. LEMUR 2 introduces a large-scale, extensible framework unifying generative, evaluative, and deployment pipelines to unlock neural-network diversity. It comprises over 14,000 distinct architectures and more than 750,000 structured training records documenting model performance, hyperparameters, and task outcomes. These models were produced through AST-based code mutation, genetic and reinforcement-learning evolution, generation of fractal architectures, and synthesis guided by a Large Language Model (LLM). This includes deep models generated with the retrieval-augmented system NN-RAG, which derived and used architectural motifs from over 900 PyTorch modules extracted from public repositories. LEMUR 2 further employs NN-VR and NN-Lite pipelines for automated deployment and latency benchmarking on heterogeneous mobile and Unity-based VR platforms, providing real-device performance metadata. It spans multimodal tasks, image captioning, text-to-image synthesis, and language modeling, supporting cross-domain analysis of architectural transferability. By linking diverse architectures, tasks, and deployment data, LEMUR 2 provides the data foundation for LLM fine-tuning and coupling diverse architectural origins with large-scale, cross-platform empirical validation. This dataset defines a new basis for reproducible and data-driven AI design, advancing the emerging paradigm of LLM-driven AutoML and architectural generalization across modalities and hardware.
[CV-73] WildCity: A Real-World City-Scale Testbed for Rendering Simulation and Spatial Intelligence ECCV2026
链接: https://arxiv.org/abs/2607.06838
作者: Xiangyu Han,Mengyu Yang,Jiaqi Li,Bowen Chang,Ziyu Chen,Hexu Zhao,Rahul Kumar Agrawal,Anthony Rodriguez,Fiona Hua,Marco Pavone,Chen Feng,Yiming Li
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: ECCV 2026; Project Page: this https URL
Abstract:Humans can navigate an unfamiliar city and gradually form a coherent spatial mental map spanning tens of square kilometers. Can AI build spatial representations at a comparable scale? Although recent foundation models have advanced scene reconstruction and embodied intelligence, scaling to entire cities remains an open challenge, primarily due to the lack of city-scale data. To bridge the gap, we introduce WildCity, a real-world multimodal dataset collected by autonomous fleets traversing complex urban environments. Our dataset includes 18 trajectories, each averaging 83.7 kilometers in length, and preserves the core challenges of in-the-wild perception, e.g., dynamic objects, lighting variations, and imperfect camera poses. We further establish an urban-tailored reconstruction baseline and convert the reconstructed environments into a closed-loop simulator. Beyond the dataset and baseline, we systematically analyze the key challenges on the path to simulation-ready urban digital twins: scalability, extrapolation, and uncertainty. Ultimately, WildCity aims to catalyze progress not only in city-scale rendering, but more broadly in the pursuit of AI that can perceive, remember, and reason across space at a scale comparable to human cognition. Project page: this https URL
[CV-74] Rail Track Extraction from Rasterized Classified Point Clouds Using a Full-Resolution Fully Convolutional Recurrent Neural Network
链接: https://arxiv.org/abs/2607.06829
作者: Alexander Gribov,Jie Chang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 15 pages, 8 figures, 1 table
Abstract:Rail track extraction is essential for effective railway asset management and maintenance, especially in automated inspection and mapping workflows. This paper introduces a novel method for extracting rail tracks from classified 3D point clouds using a fully convolutional recurrent neural network that preserves full spatial resolution and is trained exclusively on synthetically generated data. This approach enhances per-pixel quality and is particularly suited for rail track extraction. The proposed method begins by rasterizing points corresponding to railroad tracks, then applies the neural network to reduce noise and yield a cleaner track representation suitable for vectorization [1]. Subsequent morphological operations further refine the resultant data, enabling accurate track centerline extraction. Next, the extracted centerlines undergo smoothing to eliminate residual irregularities [2, 3]. Finally, the algorithm transfers 3D information from lidar points onto 2D polylines and applies additional vertical smoothing. A single centerline for both tracks is found using the Dynamic Time Warping (DTW) algorithm [4]. The final outcome consists of rail top centerlines and track centerlines derived for rail pairs, with minimal manual intervention. Experimental validation confirms the effectiveness of this method in yielding high-quality rail track extraction.
[CV-75] G-PROBE: Cross-FOV Place Recognition and Certainty-Coupled Localization for 3D Point Clouds
链接: https://arxiv.org/abs/2607.06782
作者: Jinseop Lee
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 18 pages, 9 figures
Abstract:Global localization from 3D point clouds remains challenging under limited or asymmetric fields of view (FOV), which fail to provide the dense, symmetric coverage that place recognition methods assume. We present G-PROBE, a learning-free global localization framework that removes this assumption. A virtual sensor decomposition runs the same pipeline, by design, on configurations ranging from a narrow-FOV sensor to a panoramic or multi-sensor rig. The front-end enumerates cross-FOV branch ensembles that encode heading hypotheses for heading-invariant place recognition. A score-scale-invariant, tuning-free gamma-SGRT suppresses heading aliasing under partial FOV and provably becomes inert at symmetric 360 degrees. The back-end, CG-GICP, refines a coarse full-cloud GICP with a pass restricted to high-certainty co-observed points selected by a bird’s-eye-view certainty map (a by-product of front-end scoring). This certainty coupling links descriptor evaluation to 6-DoF metric pose estimation without an external verification module. Evaluated on five LiDAR datasets and three modalities (mechanical, solid-state, FMCW), G-PROBE attains the highest learning-free multi-session F1 on average and is competitive in panoramic single-session settings. Where hand-crafted and zero-shot supervised baselines collapse under wide-to-narrow cross-sensor pairing, it remains usable end-to-end (up to 55.0% vs. no more than 6.8% success), and under FOV asymmetry (360 to 60 degrees) it retains about 54% Recall@1, about 18x the strongest learning-free baseline.
[CV-76] URS-Stereo: Uncertainty-Guided Residual Search for Real-Time Stereo Matching
链接: https://arxiv.org/abs/2607.06779
作者: Pouya Sohrabipour,Chaitanya kumar reddy Pallerla,Dongyi Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Real-time stereo matching is crucial for robotics, autonomous systems, and embedded vision applications, where both computational efficiency and disparity accuracy are required. Recent coarse-to-fine stereo matching methods improve efficiency by progressively refining disparity estimates using local cost volumes at higher resolutions. However, these methods rely heavily on the accuracy of propagated disparity estimates from previous stages. When the propagated disparity is inaccurate, the ground-truth correspondence may fall outside the predefined local search range, leading to unrecoverable matching failures during subsequent refinement. In this paper, we propose URS-Stereo, a real-time coarse-to-fine stereo matching framework that addresses this limitation through uncertainty-guided search adaptation. Specifically, we introduce an Uncertainty-Guided Residual Search Module (UGRSM), which predicts the reliability of propagated disparities together with residual search offsets to adaptively relocate the centers of local cost volumes before disparity refinement. By dynamically adjusting the search region according to the confidence of the propagated disparity, the proposed method significantly improves the robustness of local correspondence estimation while preserving the computational efficiency of coarse-to-fine stereo matching. Extensive experiments on SceneFlow, KITTI 2012, KITTI 2015, Middlebury, and ETH3D demonstrate that URS-Stereo consistently improves disparity estimation while maintaining real-time inference speed, validating the effectiveness of the proposed uncertainty-guided search strategy
[CV-77] Hardware-aware Graph Neural Networks prunning for embedded event-based vision
链接: https://arxiv.org/abs/2607.06739
作者: Piotr Wzorek,Kamil Jeziorek,Tomasz Kryjak
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Event-based cameras are gaining popularity as the sensor of choice for mobile robotics, due to their high performance in dynamic environments. However, these applications require efficient real-time data processing with low latency and power consumption. One strategy to meet these stringent requirements is hardware acceleration of efficient algorithms that preserve the temporal sparsity of event data. In this work, we propose an optimization strategy for Graph Convolutional Neural Networks models aimed at adapting their architecture to the limited resources of embedded heterogeneous FPGA platforms. Our method incorporates hardware-aware pruning and quantization, taking into account the trade-off between on-chip memory savings and inference accuracy. Strategic exploration of the design space with Fine Grid Search and Greedy layer-wise Iterative Deepening Search methods enables flexible adaptation of the model architecture to the target platform. Our approach was evaluated across various network configurations and multiple datasets, resulting in BRAM memory reductions of 28.8% for CIFAR-10 (with a 1.65% decrease in accuracy), 31.4% for MNIST-DVS (accuracy drop of 3.55%), and 26.5% for N-Caltech101 (with a 5.18% accuracy reduction).
[CV-78] A Good Initialization is All You Need for Faithful Visual Attribution
链接: https://arxiv.org/abs/2607.06726
作者: Zihan Gu,Jiayu Wang,Hua Zhang,Yue Hu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Faithful visual attribution identifies which image regions support a model prediction. Search-based perturbation methods lead the insertion–deletion faithfulness frontier by masking regions and measuring score changes, but they usually output a complete ordering of all regions. Many applications, especially MLLM attribution and repair, only need a compact top-(k) evidence mask. We study this mask-first attribution problem. An exactly (k)-region mask is combinatorial: useful evidence can depend on interactions among fine regions. Coarse grouping can stabilize early search but aggregates redundant content, whereas one-step scoring can miss high-value combinations. We introduce two forward-only methods. \textscCoPAIR uses a PhaseWin–Greedy gap diagnosis to construct coarse singleton/pair candidates that warm-start full-ordering search. \textscTRACE directly searches fixed-cardinality fine-region masks with cross-entropy sampling, elite retention, and distribution updates, with a finite-budget recovery analysis. The resulting evidence set can be returned as a compact attribution mask or used to initialize Greedy or PhaseWin when a complete ranking is required. Across ImageNet classification with CLIP ViT-L/14, CLIP RN101, and ResNet-101, our initialized search methods establish a new state-of-the-art frontier for faithful full-ordering attribution under inclusive forward-call accounting. On POPE and RePOPE with Qwen2.5-VL-3B-Instruct and LLaVA-v1.5-7B, \textscTRACE+Greedy gives the strongest search-based MLLM attribution results. Direct \textscTRACE masks further achieve single-point RePOPE repair rates of (94.44%) and (96.00%), showing that compact evidence masks can be actionable attribution outputs, not merely prefixes of full rankings.
[CV-79] SPEAR: A Simulator for Photorealistic Embodied AI Research ECCV
链接: https://arxiv.org/abs/2607.06701
作者: Mike Roberts,Renhan Wang,Rushikesh Zawar,Rachith Dey-Prakash,Quentin Leboutet,Stephan R. Richter,Matthias Müller,German Ros,Rui Tang,Stefan Leutenegger,Yannick Hold-Geoffroy,Kalyan Sunkavalli,Vladlen Koltun
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Graphics (cs.GR); Machine Learning (cs.LG); Robotics (cs.RO)
备注: Accepted for publication at the European Conference on Computer Vision (ECCV) 2026
Abstract:Interactive simulators have become powerful tools for training embodied agents and generating synthetic visual data, but existing photorealistic simulators suffer from limited generality, programmability, and rendering speed. We address these limitations by introducing SPEAR: A Simulator for Photorealistic Embodied AI Research. At its core, SPEAR is a Python library that can connect to, and programmatically control, any Unreal Engine (UE) application via a modular plugin architecture. SPEAR exposes over 14K unique UE functions to Python, representing an order-of-magnitude increase in programmable functionality over existing UE-based simulators. Additionally, a single SPEAR instance can render 1920x1080 photorealistic beauty images directly into a user’s NumPy array at 73 frames per second - an order of magnitude faster than existing UE plugins - while also providing ground truth image modalities that are not available in any existing UE-based simulator (e.g., a non-diffuse intrinsic image decomposition, material IDs, and physically based shading parameters). Finally, SPEAR introduces an expressive high-level programming model that enables users to specify complex graphs of UE work with arbitrary data dependencies among work items, and to execute these graphs deterministically within a single UE frame. We demonstrate the utility of SPEAR through a diverse collection of example applications: controlling multiple embodied agents with distinct action spaces (e.g., humans, cars, and robots) across several in-the-wild UE projects; rendering photorealistic city-scale environments; manipulating UE’s procedural content generation systems; rendering synchronized multi-view images of detailed human faces; coordinating an interactive co-simulation with the MuJoCo physics simulator; and editing scenes with natural language via an AI coding assistant.
[CV-80] CoMind: Understanding Collaborative Human Activity from Multiple Minds and Views ECCV2026
链接: https://arxiv.org/abs/2607.06691
作者: Alexey Gavryushin,Dingxi Zhang,Zhao Huang,Alexandros Delitzas,Jiaqi Chen,Ben Ellis,Cedric Zöllner,Manthan Patel,Manuel Kaufmann,Marc Pollefeys,Xi Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ECCV 2026
Abstract:Human-human collaboration is a fundamental aspect of everyday life, essential to success in a wide range of goal-directed activities from household tasks to professional teamwork. While much research has focused on modeling coordination and task execution, the cognitive processes that support such collaboration, particularly Theory of Mind (the ability to infer the mental states of others), remain difficult to study in natural settings. To address this gap, we introduce a novel egocentric and exocentric video dataset capturing real-world collaboration in cooking scenarios. The dataset integrates multi-perspective video, high-quality audio, gaze tracking, and 3D scene and object scans, with annotations for shared attention to objects, social cues and interactions between agents, as well as agent-object interactions. We establish benchmarks for Joint Attention Estimation, Socially Conditioned Object Interaction Anticipation, and Collaborative Handover Prediction, enabling research on multimodal perception, proactive assistance, and collaborative planning. By providing temporally aligned, richly annotated multimodal data, CoMind facilitates the development and evaluation of AI systems capable of modeling complex social interactions and reasoning about human behaviors in collaborative environments. Our dataset and benchmarks are made available at this https URL.
[CV-81] ProMoE-FL: Prototype-conditioned Mixture of Experts for Multimodal Federated Learning with Missing Modalities
链接: https://arxiv.org/abs/2607.06633
作者: Aavash Chhetri,Bibek Niroula,Eduard Vazquez,Yash Raj Shrestha,Prashnna Gyawali,Loris Bazzani,Binod Bhattarai
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:In this paper, we address the problem of multimodal federated learning with missing modality. Existing methods utilize an additional public dataset or perform naive feature synthesis that is based solely on the available modality. To address these limitations, we propose ProMoE-FL, a Prototype-conditioned Mixture-of-Experts framework for robust missing-modality feature synthesis in multimodal federated learning. ProMoE-FL builds a global client-aware prototype bank that captures clinically meaningful modality priors across institutions. Our Mixture of Experts is conditioned on these prototypes and modality indices to enable direction-aware expert routing for dynamically synthesizing missing features. We perform extensive quantitative and qualitative evaluations on four public chest X-ray datasets (MIMIC-CXR, NIH Open-I, PadChest, and CheXpert) and demonstrate that ProMoE-FL consistently outperforms state-of-the-art methods in both homogeneous as well as the more challenging heterogeneous settings.
[CV-82] Dynamic-in-Few-Step: Unifying Dynamic Computation and Few-Step Distillation for Efficient Video Generation
链接: https://arxiv.org/abs/2607.06631
作者: Yu Cheng,Siyue Yao,Zhongang Qi,Shanyan Guan,Wei Li,Fajie Yuan
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Video Diffusion Models (VDMs) have demonstrated superior generation quality but suffer from prohibitive computational costs. While recent few-step distillation techniques significantly accelerate inference, they typically enforce a static model architecture across all denoising stages, ignoring the varying computational demands inherent to different noise levels. In this work, we propose a novel post-training acceleration framework that exploits this redundancy by integrating dynamic structural sparsification directly into the distillation process. Unlike conventional post-hoc compression applied to a fixed diffusion pipeline, our approach jointly optimizes the denoising steps and structured model sparsity, transforming a pre-trained VDM into a compact, step-specific Mixture-of-Models (MoM). To address the training instability arising from this joint optimization, we introduce a Progressive Training Strategy coupled with an Output Rollout Mechanism, which ensures the coherent learning of structural decisions across timesteps. Furthermore, we develop a specialized inference engine to deploy the resulting MoM efficiently. Our method is orthogonal to existing acceleration techniques and highly effective: On Wan-14B, it removes 24% of the per-step FLOPs on top of 4-step distillation, adding a 1.2x wall-clock gain and reaching a 30x speedup over the 50-step teacher while preserving competitive generation quality.
[CV-83] SpaR3D-MoE: Adaptive 3D Spatial Reasoning from Sparse Views Meets Geometry-Inductive Mixture-of-Experts ECCV2026
链接: https://arxiv.org/abs/2607.06620
作者: Haida Feng,Hao Wei,Haolin Wang,Shiwei Li,Chade Li,Yihong Wu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted to ECCV 2026
Abstract:Recent Multimodal Large Language Models (MLLMs) struggle to bridge the representational gap between 2D semantic understanding and 3D spatial geometry. Existing 3D-aware models either rely on costly 3D-specific data or utilize RGB-only inputs with heuristic sampling and monolithic, shallow fusion, which respectively disrupt essential spatiotemporal connectivity and induce modality contention across diverse spatial tasks. To overcome these bottlenecks, we introduce SpaR3D-MoE, an end-to-end framework that enables adaptive spatial reasoning by equipping MLLMs with geometry-aware capabilities from only sparse RGB inputs. First, we propose an adaptive spatiotemporal manifold sampling mechanism that constructs a geometry-aware spatiotemporal graph to extract informative keyframes, effectively mitigating sequence redundancy while preserving the scene’s topological connectivity. Second, we introduce the heterogeneous geometry-inductive Mixture-of-Experts driven by an instruction-pose aware router, which adaptively routes multimodal tokens to specialized experts, resolving the cross-modal contention inherent in monolithic fusion. Extensive experiments on VSI-Bench, ScanQA, and SQA3D demonstrate that our method achieves state-of-the-art performance. Notably, SpaR3D-MoE achieves the highest average score of 63.5 on VSI-Bench, outperforming the strongest baseline by 7.8 absolute points, alongside relative improvements of 35.4% and 51.4% in Route Plan and Relative Direction tasks, respectively.
[CV-84] Overview of the NLPCC 2026 Shared Task 1: Difficulty-Aware Multilingual and Multimodal Medical Instructional Video Understanding Evaluation
链接: https://arxiv.org/abs/2607.06618
作者: Shenxi Liu,Kan Li,Mingyang Zhao,Yuhang Tian,Bin Li
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 21 pages, 1 figure, 5 tables
Abstract:Following the CMIVQA, MMI-VQA, and M4IVQA challenges in NLPCC 2023–2025, we introduce the Difficulty-Aware Medical Instructional Video Question Answering (DA-MIVQA) shared task for NLPCC 2026. DA-MIVQA extends previous multilingual and multimodal medical video benchmarks by explicitly distinguishing questions according to the type and complexity of evidence required for answering. Specifically, simple questions can often be answered from subtitle-based textual cues, whereas complex questions require visual grounding, procedural understanding, and cross-modal evidence integration. The challenge contains three tracks: Difficulty-Aware Temporal Answer Grounding in Single Video (DA-TAGSV), Difficulty-Aware Video Corpus Retrieval (DA-VCR), and Difficulty-Aware Temporal Answer Grounding in Video Corpus (DA-TAGVC). The dataset is collected from public medical instructional channels, covers diverse scenarios such as first aid, emergency response, rehabilitation, nursing, and general medical education, and is manually verified with difficulty annotations. This paper presents the task motivation, dataset construction, evaluation protocol, participation overview, competition results, and representative systems of DA-MIVQA. DA-MIVQA provides a practical benchmark for evaluating medical instructional video question answering systems under varying textual, visual, temporal, and procedural reasoning requirements.
[CV-85] WHERE to Generate Matters: Budget-Aware Synthetic Augmentation for Label Skewed Federated Learning
链接: https://arxiv.org/abs/2607.06616
作者: Sangwoo Lee,Sunghwan Park,Jaewoo Lee
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: preprint
Abstract:Label skew in federated learning (FL) causes client drift and degrades global accuracy. Synthetic data augmentation can reduce this imbalance; however, full class balancing requires substantial computation cost. We propose FedEAS, a policy that assigns each client an entropy-adaptive per-class generation budget computed from its local label distribution. The budget jointly decides \emphhow much each client generates and \emphWHERE the samples go. Accordingly, the total generation budget follows from the per-client budgets rather than being fixed in advance. FedEAS recovers most of the accuracy gain of full class balancing while reducing the generation budget by 94.1%. At the same total generation budget, it outperforms Uniform allocation by up to 18.82% across CIFAR-10 and CIFAR-100.
[CV-86] Do Counterfactually Fair Image Classifiers Satisfy Group Fairness? – A Theoretical and Empirical Study NEURIPS2024
链接: https://arxiv.org/abs/2607.06603
作者: Sangwon Jung,Sumin Yu,Sanghyuk Chun,Taesup Moon
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: NeurIPS 2024 Track Datasets and Benchmarks
Abstract:The notion of algorithmic fairness has been actively explored from various aspects of fairness, such as counterfactual fairness (CF) and group fairness (GF). However, the exact relationship between CF and GF remains to be unclear, especially in image classification tasks; the reason is because we often cannot collect counterfactual samples regarding a sensitive attribute, essential for evaluating CF, from the existing images (\eg, a photo of the same person but with different secondary sex characteristics). In this paper, we construct new image datasets for evaluating CF by using a high-quality image editing method and carefully labeling with human annotators. Our datasets, \oursceleb and \ourslfw, build upon the popular image GF benchmarks; hence, we can evaluate CF and GF simultaneously. We empirically observe that CF does not imply GF in image classification, whereas previous studies on tabular datasets observed the opposite. We theoretically show that it could be due to the existence of a latent attribute G that is correlated with, but not caused by, the sensitive attribute (\eg, secondary sex characteristics are highly correlated with hair length). From this observation, we propose a simple baseline, Counterfactual Knowledge Distillation (CKD), to mitigate such correlation with the sensitive attributes. Extensive experimental results on \oursceleb and \ourslfw demonstrate that CF-achieving models satisfy GF if we successfully reduce the reliance on G (\eg, using CKD).
[CV-87] MiLSD: A Micro Line-Segment Detector for Resource-Constrained Devices
链接: https://arxiv.org/abs/2607.06600
作者: Parsa Hassani Shariat Panahi,Amir Hossein Jalilvand,M. Hassan Najafi
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Robotics (cs.RO)
备注: 10 pages, 12 figures, 5 tables
Abstract:Line segment detection is a key building block in visual SLAM, 3D reconstruction, and industrial inspection. Recent deep learning methods have greatly improved accuracy, yet even the smallest models require several megabytes of memory, exceeding low-cost MCU capacity. This work investigates the maximum achievable accuracy under a sub-megabyte budget. We propose MiLSD, a detector tailored for MCU-level constraints, and systematically compare three output representations within a compact fully-convolutional backbone. Our study shows that the proposed F-Clip center-with-length-and-angle formulation learns most effectively at small model sizes. We find that 8-bit quantization preserves full-precision performance, while 4-bit quantization causes significant degradation, particularly in angle regression, with quantization-aware training recovering only part of the loss. With a one-megabyte activation budget and inference enhancements including sub-pixel decoding, test-time augmentation, and a lightweight verifier, MiLSD improves sAP10 on ShanghaiTech Wireframe from 10.6 (25k parameters, 0.25 MB) to 24.1 within 1 MB. Rather than competing with GPU-scale parsers, we map the accuracy memory trade-off across representations, bit-widths, capacities, and post-processing strategies for embedded vision systems.
[CV-88] LipSSD: Lipschitz-Constrained Single-Shot Detection for Adversarially Robust Object Detection
链接: https://arxiv.org/abs/2607.06592
作者: Vincent Lébé,Yannick Prudent,Corentin Friedrich,Thomas Massena(IRIT, DTIPG - SNCF, UT3),Ronan Sicre(IRIT),Franck Mamalet
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Object detectors have many applications in safety-critical systems, but they are known to be sensitive to worst-case perturbations such as adversarial attacks, which limits their applicability in real-world scenarios. Compared with classification, adversarial robustness for object detection has received less attention, and existing methods are often tied to adversarial training, whose performance may not transfer across attacks, perturbation budgets, or architectures. In this work, we introduce Lipschitz-constrained variants of object detection architectures as robust-by-design alternatives to standard detectors. We validate this approach with LipSSD, a Lipschitz-constrained Single Shot MultiBox Detector (SSD), and provide a comprehensive study of its adversarial robustness using multiple white-box adversarial attacks and datasets. We first analyze the accuracyrobustness trade-off induced by Lipschitz constraints and show that it can be controlled through a single training hyperparameter. We then demonstrate that Lipschitzconstrained detectors are complementary to adversarial training: under the same training setup on the Pascal VOC dataset, adversarially trained LipSSD improves mAP@50 on unseen attacks by up to 15 points over classical adversarially trained SSD. Finally, we use more specific safety-critical datasets such as LARD and KITTI, and show that Lipschitz-constrained detectors can improve robustness while largely preserving clean performance. These results suggest that architectural Lipschitz control is a practical and attack-agnostic direction for improving the robustness of object detectors.
[CV-89] AI for Cultural Heritage Textiles: Fine-Tuned Latent Diffusion for Novel Ulos Motif Synthesis WWW
链接: https://arxiv.org/abs/2607.06590
作者: Humasak Tommy Argo Simanjuntak,Jesika Purba,Sitogab Girsang,Widya Manurung,Samuel Situmeang,Arlinta Barus,Daniel Oranova Siahaan
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 21 pages, 8 figures, 3 tables. The manuscript is currently under review at the 2026 4th International Conference on Data, Information and Computing Science ( this https URL )
Abstract:Preserving and revitalising traditional textiles such as Ulos, a cultural heritage of the Batak ethnic group in North Sumatra, Indonesia, requires balancing fidelity to tradition with innovative approaches that meet contemporary design demands. Traditional Ulos weaving faces two key limitations: a narrow range of motifs and a time-intensive design process. This study presents a generative AI framework that fine-tunes two pretrained latent diffusion models: Protogen v3.4 and Stable Diffusion v1.4, on a curated, annotated dataset of high-resolution Ulos motifs to generate culturally consistent yet novel designs. Model performance is evaluated quantitatively using Frechet Inception Distance (FID), Inception Score (IS), and qualitatively through assessments by traditional weavers and members of the public. Protogen v3.4 consistently outperforms Stable Diffusion v1.4, achieving substantially lower FID (~10.5x) and higher IS (2.0x), indicating superior visual fidelity, diversity, and closer alignment with the real Ulos motif distribution. We further examine the effects of strength and guidance scale on generation quality across both models. Lower strength values consistently yield higher fidelity (lower FID), while higher strength values increase generative diversity at the cost of realism, revealing a clear fidelity-diversity tradeoff for both models. Across all tested configurations, a guidance scale of 5-9 provides the most effective balance between fidelity and diversity, stabilising FID, KID, and IS, and is recommended as the operating range for high-quality, diverse Ulos motif generation. These findings demonstrate that carefully fine-tuned generative AI can support the creative renewal of intangible cultural heritage while preserving its stylistic and symbolic integrity.
[CV-90] CoFINN: Conservation Flux Informed Neural Networks for Physics Problems Governed by Conservation Laws
链接: https://arxiv.org/abs/2607.06587
作者: Adnan Harun Doğan,Mert Deniz,Hande Alemdar,Özgür Uğraş Baran
类目: Computer Vision and Pattern Recognition (cs.CV); Fluid Dynamics (physics.flu-dyn)
备注: 28 pages, 7 figures
Abstract:We present CoFINN (Conservation Flux Informed Neural Networks), a physics-informed deep learning framework for predicting compressible flow fields governed by conservation laws. Unlike conventional data-driven convolutional neural networks (CNNs), which optimize only pixel-wise similarity metrics, CoFINN embeds finite-volume conservation physics directly into the training process. Unlike classical physics-informed methods which enforce differential-equation residuals at collocation points through automatic differentiation, CoFINN adopts a finite-volume perspective consistent with modern CFD methodology. CoFINN interprets CNN output fields as structured computational grids, where each pixel represents a finite-volume cell, and enforces conservation consistency through sophisticated numerical flux calculations. The framework is evaluated on transonic flow prediction around airfoils at (M=0.7, Re=6 * 10^6), including challenging conditions involving shock waves and high angles of attack. Results show that CoFINN improves aerodynamic force prediction accuracy, reducing drag prediction error by up to 34% at extreme angles of attack and by approximately 15% on average across the test set. Improvements are particularly significant in limited-data regimes, demonstrating that the conservation-based loss acts as an effective physical regularizer. The proposed approach maintains the computational efficiency advantages of CNN surrogates while significantly improving physical consistency and conservation behavior. The framework is architecture-agnostic and extensible to broader classes of conservation-law-governed physical systems.
[CV-91] Pixel-Precise Explainable Stress Indexing: A Semantic Segmentation Framework for Disease Severity Quantification in Field Crops
链接: https://arxiv.org/abs/2607.06585
作者: Raunak Kumar,Soumyashree Kar
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 26 pages, 15 figures, 5 tables
Abstract:Plant diseases, resulting from both biotic and abiotic stresses, cause an estimated 20-40% loss in global agricultural yield annually, resulting in economic damages exceeding USD 220 billion. Accurate and scalable stress quantification is essential for precision agriculture, yet traditional manual assessments are labour-intensive and subjective. This paper proposes a unified deep learning pipeline integrating semantic segmentation, regression-based severity estimation, and disease classification. Stress severity is categorised into four levels (Low to Very High) based on the proportion of infected leaf area. Experiments on the Apple Tree Leaf Disease Segmentation dataset (1,641 samples, six classes) evaluate four models: U-Net (MobileNetV2), SegFormer, FCN, and PSPNet. U-Net with MobileNetV2 achieves the best performance with 98.20% pixel accuracy, 0.70 mIoU, and 99.41% detection accuracy at 14.7 ms per image, making it suitable for real-time use. SegFormer performs competitively (mIoU 0.66), while FCN and PSPNet show lower spatial accuracy (approximately 0.49 mIoU). The computed severity index strongly correlates with expert annotations (r = 0.968, R^2 = 0.937), demonstrating the system’s reliability for automated crop monitoring and decision support.
[CV-92] owards Accurate and Fast Clinical Body Composition: A Resource-Efficient Hierarchical Segmentation Framework for Multi-Source CT
链接: https://arxiv.org/abs/2607.07177
作者: Xiaodi Shen,Qingzhu Zheng,Yaoyang Qiu,Cien Fan,Ruonan Zhang,Yangdi Wang,Luyao Wu,Weikai Zheng,Longfei Zhao,Bing Li,Rulin Xu,Qiqi Xu,Ren Mao,Shiting Feng,Xuehua Li
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注: Affiliations: (1) Department of Radiology, The First Affiliated Hospital, Sun Yat-sen University, Guangzhou 510080, China. (2) Research Development Center, Canon Medical Systems (China) Co. Ltd. Beijing 100015, China
Abstract:Background: Automated 3D segmentation of muscles and adipose tissue from CT is vital for body composition analysis, but multi-source data heterogeneity and high CPU memory demands hinder clinical deployment. Methods: We propose a coarse-to-fine hierarchical framework to segment ten tissue structures. Efficiency is optimized using Dynamic Spacing and Anisotropic Patching, a Group Inference mechanism for low-memory sliding-window processing, and Topology-Aware Asymmetric Resampling for fast post-processing. Results: The framework was trained on 1,558 CT volumes from seven public and two private datasets, and evaluated on an independent test cohort (N=105), per-structure Dice coefficients ranged from 0.924 to 0.982. Eight major structures met the ±10% relative error clinical acceptance limit. On a 12-core CPU workstation, the GPU-free pipeline averaged 44.5 seconds per volume with 4.73 GB peak memory. Conclusion: This framework balances accuracy and efficiency, enabling robust, large-scale body composition analysis on standard CPU workstations. Comments: Affiliations: (1) Department of Radiology, The First Affiliated Hospital, Sun Yat-sen University, Guangzhou 510080, China. (2) Research Development Center, Canon Medical Systems (China) Co. Ltd. Beijing 100015, China Subjects: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2607.07177 [eess.IV] (or arXiv:2607.07177v1 [eess.IV] for this version) https://doi.org/10.48550/arXiv.2607.07177 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Qiqi Xu [view email] [v1] Wed, 8 Jul 2026 09:11:41 UTC (1,443 KB)
[CV-93] From Data Completeness to Data Sufficiency: A Task-Driven Imaging Framework for Intraoperative CBCT under Quality-Time-Dose Trade-offs
链接: https://arxiv.org/abs/2607.07039
作者: Yi Jia,Rongjun Ge,Yang Chen,Yan Xi,Wenjun Xia
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV); Medical Physics (physics.med-ph)
备注:
Abstract:Mobile C-arm cone-beam computed tomography (CBCT) has been widely used for real-time intraoperative 3D imaging. However, current practice often mechanically applies the fan-beam CT criterion of “180° plus fan angle” in pursuit of “data completeness” in reconstruction. This review argues that, under the single circular trajectory of three-dimensional cone-beam geometry, complete data are mathematically unattainable; moreover, blindly increasing sampling may exacerbate the trade-off among intraoperative image quality (Q), imaging time (T), and radiation dose (D). Against this background, this review reframes the evaluation of intraoperative CBCT around “data sufficiency” rather than “data completeness.” This perspective moves beyond the excessive pursuit of absolute mathematical and analytic accuracy, and instead emphasizes task-specific minimum image-quality thresholds required for clinical decision-making. By synthesizing evidence from multiple clinical scenarios, this review suggests that approximation errors can be acceptable when clinical decision-making requirements are satisfied, thereby achieving a Q-T-D balance.
[CV-94] Format-Controlled Multi-Scale JPEG Compression Response Analysis for Image-Level Forgery Screening
链接: https://arxiv.org/abs/2607.06615
作者: Sujith K Mandala
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: This work has been submitted to the IEEE for possible publication
Abstract:Image forgery detection is a critical task in digital forensics, yet many deep-learning localization approaches are typically GPU-accelerated and computationally heavier than handcrafted screening methods. We propose a lightweight, interpretable feature engineering pipeline for image-level forgery screening using only CPU computation and gradient boosted trees. Our method introduces \emphmulti-scale Error Level Analysis (ELA) computed at seven JPEG quality levels, combined with novel \emphcross-quality ELA ratio features that capture double-compression artifacts characteristic of spliced regions, augmented by spatial entropy, FFT energy bands, edge density, SRM residuals, and DCT blockiness, yielding a 405-dimensional feature vector. CASIA v2.0 contains a format confound (60% of tampered images are TIFF while authentic images are JPEG/BMP and contain no TIFF samples), enabling a trivial \textttis_tiff classifier to reach 0.80 AUC. We address this through rigorous format-controlled evaluation: on the JPEG-only subset (9,501 images, eliminating the TIFF/JPEG container confound), our method achieves AUC~=~0.990 [95% CI: 0.988–0.991] and F1~=~0.905 using 5-fold stratified cross-validation. Under a conservative source-aware group split (preventing related images from appearing in both train and test), AUC remains 0.976. An ablation study reveals that multi-scale ELA provides the dominant gain (+0.180 AUC over single-quality on the format-controlled subset), while cross-quality ratios provide complementary double-compression detection. These results support that the method detects compression-history inconsistencies rather than file-format shortcuts – while offering feature-level interpretability, CPU-only deployment, and sub-second inference.
[CV-95] Non-contact Real-time Heart-rate Measurement using Image Processing with Commodity Cameras and AI Agents
链接: https://arxiv.org/abs/2607.06598
作者: Kelly Li,Fulu Li
类目: Image and Video Processing (eess.IV); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: 6 pages, 5 figures
Abstract:Heart rate measurement is one of the key requirements for real-time health monitoring, in particular for health caring of elderly people. Traditional heart rate measurement relies on contact sensing mechanisms such as some heart rate measurement devices at medical hospitals or some wearable devices with embedded sensors such as Apple Watch, etc. In this paper, we develop a system for non-contact, real-time, heart rate measurement using image processing with commodity cameras such as an embedded camera on a laptop, where we use an innovative algorithm to capture the relevant signals for the computation of heart rate in a time series in real life environments. The presented heart rate computation (HRC) process is composed with four major steps: (a) identify frames per second of the camera in use, i.e., 30 frames per second for a given camera, (b) face detection (FD) with shape predictor of 68 face landmarks using deep learning (DL) method, © time sliding window (TSW) algorithm to de-noise the signal by smoothing out the noise, and (d) compute heart rate based on identified signal periodicity. We test and analyze the developed prototypes against heart rate results by Apple Watch and check the difference range in multiple rounds and compute the mean of the difference for the measurement values of the heart rate of the same person at the same time. We will do further tuning and optimization of the present methods and deploy the system as a personal AI agent [6] for health monitoring as our future directions.
人工智能
[AI-0] Breaking Database Lock-in: Agent ic Regeneration of High Performance Storag e Readers for Database Bypass VLDB
链接: https://arxiv.org/abs/2607.07696
作者: Victor Giannakouris,Immanuel Trummer
类目: Databases (cs.DB); Artificial Intelligence (cs.AI)
备注: To be presented at AIDB 2026 (co-located with VLDB)
Abstract:Analytical workloads operating on data stored in external database systems face a fundamental bottleneck: data access is guarded entirely by the database driver, like JDBC or ODBC, forcing all reads through query execution and other driver layers that are not designed for bulk columnar analytics. We present Jailbreak, an approach that bypasses the database engine entirely by reading storage files directly and materializing data as in-memory columnar buffers. Jailbreak’s key insight is that database file formats, while complex, are fully specified by their source code and documentation, artifacts that Large Language Models (LLMs) can ingest to regenerate operator-specific table reading components without human-engineered parsing logic. Jailbreak leverages LLM-assisted code synthesis for database storage decoding, turning a traditionally opaque format into a directly queryable artifact. We evaluate Jailbreak on PostgreSQL and MySQL storage files, targeting analytical snapshot scenarios common in read replicas and offline processing pipelines. The generated reader produces Apache Arrow buffers consumable directly by most of the widely known query engines, including DuckDB, Apache Spark, and GPU-accelerated frameworks such as cuDF and Spark RAPIDS. We validate correctness against JDBC/ODBC-based baselines using the TPC-H benchmark across all query results, and demonstrate significant performance improvements in end-to-end analytical throughput, achieving up to 27x speedups. Our results showcase that LLM-assisted storage reader synthesis is a viable and generalizable methodology for breaking data lock-in across database systems, with applications beyond PostgreSQL and MySQL for any system whose file format is available to the LLM from documentation or source code.
[AI-1] SkillCenter: A Large-Scale Source-Grounded Skill Library for Autonomous AI Agents
链接: https://arxiv.org/abs/2607.07676
作者: Tianming Sha,Yue Zhao,Lichao Sun,Yushun Dong
类目: Artificial Intelligence (cs.AI)
备注: 44 pages, 5 figures. Code: this https URL ; Data: this https URL
Abstract:Autonomous AI agents can execute complex tasks with limited human review, yet they often lack the grounded operational knowledge to make their outputs not just executable but correct, secure, and maintainable. We introduce SkillCenter, to our knowledge the largest open skill library for agents by total count: 216,938 structured skills across 24 domain bundles. A SkillGate-filtered pipeline contributes 114,565 source-grounded skills from peer-reviewed journals, ArXiv, and over 24,000 technical sources, integrated with 102,373 community skills from GitHub and the ClawHub marketplace. We present the end-to-end framework that builds the pipeline subset: multi-source acquisition, an LLM-based quality gate (SkillGate), template-driven generation, iterative source-grounding, and quality-controlled publishing. Source grounding is a traceability guarantee: each retained claim maps to an exact quotation in its source. All skills ship as offline-searchable SQLite FTS5 bundles.
[AI-2] Recursive Self-Improvement in AI: From Bounded Self-Refinement to Autonomous Research Loops
链接: https://arxiv.org/abs/2607.07663
作者: Mingguang Chen,Licheng Wang,Bo Qu
类目: Artificial Intelligence (cs.AI)
备注: 42 pages, 6 figures
Abstract:AI systems increasingly participate in their own improvement: revising their outputs, adapting their own harnesses during deployment, training on data they generate, and, increasingly, conducting AI research itself. This literature is described under a vocabulary (“self-refine,” “self-reward,” “self-play,” “self-evolve”) that conflates fundamentally different ambitions. We survey 1,250 arXiv papers (2024-2026) along two axes: what the system improves – its behavior in deployment, its policy through training, its evaluator, or the research process itself – and the degree of loop closure (human-in-the-loop to fully closed). The taxonomy separates bounded self-refinement – convergent, evaluable, and already industrial practice – from open-ended recursive self-improvement (RSI), which remains bounded by grounding requirements, collapse dynamics, and compute constraints on every measured axis. Its distinctive feature is a dedicated category for self-evaluation: every improvement loop is a claim that some signal can substitute for human judgment. We survey the evaluator design space – judges, process reward models, verifiers, rubrics, meta-evaluation – order the signals into a verification hierarchy from formal verifiers (strongest) to intrinsic self-assessment (weakest), and observe that demonstrated self-improvement strength tracks this hierarchy, that its failure modes (self-confirming loops, model collapse, diversity collapse) follow from its violations, and that the “research direction-setting” bottleneck keeping humans in the loop sits at the top of that hierarchy. We connect the technical literature to the theory of RSI limits and to the safety and governance questions raised by frontier-lab accounts of closing the loop, and identify governance-grade measurement of self-improvement as the field’s most underpopulated niche.
[AI-3] ALER-TI: Aligned Latent Embedding Retrieval for Time Series Imputation
链接: https://arxiv.org/abs/2607.07640
作者: Xuan-Thong Truong,Trung-Kien Le,Tung Kieu,Thi-Thu Nguyen,Nhat-Hai Nguyen
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 10 pages, 2 figures, 12 tables
Abstract:Deep learning has significantly advanced time series imputation, yet most existing architectures primarily rely on localized temporal context within the corrupted input sequence. This reliance can be limiting in real-world scenarios, where time series often exhibit non-stationary dynamics, weak temporal correlations, and infrequent patterns that are difficult to reconstruct from nearby observations alone. In this paper, we propose ALER-TI, Aligned Latent Embedding Retrieval for Time Series Imputation, a retrieval-augmented framework that explicitly leverages historical patterns to supplement degraded local context for more reliable missing-value reconstruction. The core of ALER-TI is Latent Embedding Alignment (LEA), which mitigates the representation mismatch between corrupted queries and complete historical candidates. By applying post-hoc masking in the latent space, LEA aligns candidates with the query’s missingness pattern while allowing historical embeddings to be pre-computed and cached for efficient retrieval. ALER-TI is model-agnostic and can be integrated with various imputation backbones through a lightweight adaptation module. Extensive experiments on six real-world datasets under different missing rates demonstrate that ALER-TI consistently improves strong baseline models and enhances robustness across diverse imputation settings.
[AI-4] owards Agent ic AI Governance: A Preliminary Assessment
链接: https://arxiv.org/abs/2607.07612
作者: Mubarak Raji,Masooda Bashir
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
备注: International Conference on the AI Revolution: Research, Ethics, and Society (AIR-RES 2026)
Abstract:Artificial intelligence is rapidly evolving from generative systems to agentic AI capable of autonomously planning and executing tasks. Widely characterized as the Year of Agentic AI, 2025 marked accelerated development and deployment, introducing new ethical and governance challenges. This paper presents a systematic review of the emerging literature on agentic AI governance. Our analysis identifies features that distinguish agentic AI from traditional systems and why it warrants targeted governance attention. We synthesize prevailing governance priorities, proposed mechanisms, and stakeholder roles shaping this evolving domain. As an initial scholarly effort, this review lays the preliminary groundwork for developing a structured roadmap to guide responsible and adaptive agentic AI governance.
[AI-5] CARLA-GS: Decoupling Representation Reasoning and Physics Simulation for Autonomous Driving Corner-Case Synthesis
链接: https://arxiv.org/abs/2607.07601
作者: Kaicong Huang,Meng Ma,Ruimin Ke
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Safety evaluation for autonomous driving is dominated by rare, safety-critical interactions, motivating simulators that can deliberately synthesize corner cases with photorealistic observations. Corner-case generation is inherently a multi-source problem spanning visual representation, scene reasoning, and vehicle trajectory generation and control. Prior knowledge- and model-based approaches typically focus on scene or trajectory components in isolation, while diffusion-based methods attempt end-to-end generation but still struggle to ensure spatiotemporal consistency and physical realism. To unify these aspects within a single framework, we propose CARLA-GS, a modular corner-case synthesis pipeline that decouples visual representation, semantic reasoning, and physics-based execution while maintaining tight cross-module coupling. Starting from real driving data, we reconstruct an editable gaussian scene with additional geometry-consistent constraints. A multi-agent LLM then performs scene-level reasoning to identify risky interactions and generate intent-level waypoint trajectories, while the low-level motion control is delegated to CARLA, where a PID controller ensures kinematic and dynamic feasibility. The simulated vehicle states are finally re-projected into the gaussian scene for ego-centric rendering. This design enables high-level semantic reasoning, low-level physically executable motion, and photorealistic corner-case generation within a unified pipeline. Experiments on the Waymo Open Dataset show, both quantitatively and qualitatively, that our framework enables controllable corner-case generation and produces photorealistic, spatiotemporally consistent videos aligned with semantic intent and physically feasible motion.
[AI-6] Collaborative Synthetic Data Generation for Knowledge Transfer in Federated Learning
链接: https://arxiv.org/abs/2607.07565
作者: Maximilian Andreas Hoefler,Karsten Mueller,Wojciech Samek
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:One-shot federated learning (OSFL) addresses the communication overhead of federated learning by limiting training to a single round, but doing so without sacrificing model quality is non-trivial, particularly when client data distributions diverge. Recent work has addressed this challenge by aggregating client knowledge on the server through the construction of transferable synthetic datasets or distillates. However, most of these methods lack formal privacy guarantees, leaving a gap in jointly achieving low communication, robustness to heterogeneity, and rigorous privacy. We propose FedKT-CSD (Federated Knowledge Transfer via Collaborative Synthetic Data), a framework inspired by neural image compression that closes this gap by leveraging publicly pretrained autoencoders as a shared latent space. Each client encodes its private data in a single forward pass, computes class-conditional latent statistics, and transmits these to the server. The server aggregates these statistics via secure aggregation, adds calibrated differential privacy noise, and decodes a synthetic dataset for training a global model and further downstream tasks. This design provides formal (\varepsilon,\delta) -differential privacy by construction, while keeping client-side computation and communication lightweight. Despite operating under privacy constraints, FedKT-CSD is competitive with and even outperforms non-private baselines across diverse datasets and heterogeneity settings, and scales to a large number of clients. Our code is available at: this https URL
[AI-7] Single-Rollout Asynchronous Optimization for Agent ic Reinforcement Learning
链接: https://arxiv.org/abs/2607.07508
作者: Zhenyu Hou,Yujiang Li,Jie Tang,Yuxiao Dong
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Reinforcement learning (RL) is becoming increasingly important for post-training large language models (LLMs). Previous RL pipelines for LLMs were mostly synchronous and batch-interleaved, which is inefficient for long-horizon agentic tasks. Recently, asynchronous RL has emerged as a more efficient alternative by updating the model as rollouts arrive. However, existing asynchronous RL systems often emphasize throughput, while leaving training stability and task effectiveness largely underexplored. For example, a key challenge is that group-wise sampling in the widely-used GRPO framework does not naturally fit asynchronous agentic training. In this paper, we present Single-rollout Asynchronous Optimization (SAO) to address the stability and off-policy challenges in asynchronous RL. To reduce off-policy effects and improve generalization, we replace group-wise sampling with single-rollout sampling, that is, using one rollout per prompt. We further improve this single-rollout strategy with practical value-model training designs. To improve optimization stability, we introduce a strict double-side token-level clipping strategy. SAO is able to train stably for one thousand steps and consistently outperform GRPO and its variants on agentic coding and reasoning benchmarks, such as SWE-Bench Verified, BeyondAIME, and IMOAnswerBench. We also demonstrate that single-rollout RL is particularly effective in a simulated online learning setting, where the model must adapt to changing evolving environments. To this end, SAO is successfully deployed in the agentic RL pipeline for training the open GLM-5.2 model (750B-A40B).
[AI-8] Do LLM -Generated Skills Make Better AI Data Scientists? A Component Ablation Across Data-Science Workflows KDD2026
链接: https://arxiv.org/abs/2607.07504
作者: Wei-Jung Huang
类目: Artificial Intelligence (cs.AI)
备注: KDD 2026 Workshop on AI Data Scientist
Abstract:Product data scientists often ask LLM-based agents to help with recurring execution tasks such as cleaning data, writing SQL, choosing statistical tests, and formatting results. Reusable skill files are meant to avoid prompting from scratch by packaging guidance for a task family. Expert-written skills can encode high-quality guidance, but writing and maintaining them across many data-science task families creates a manual bottleneck. We ask whether LLM-generated skills offer a useful low-curation alternative: do they improve performance over the task prompt alone? We test this question across four lifecycle stages: data preparation, data extraction, statistical analysis, and reporting, using one generated skill per stage. We find no reliable improvement from full generated skills over No-Skill prompting. We then ask whether any part of the skill is useful by ablating different skill components. The main ablation covers 56 tasks, nine model configurations, and three providers, yielding 7,560 runs. Compared with prompting using the task alone, neither the full generated skill nor any ablated skill variant significantly improves performance; all p-values are at least 0.396, and the total spread across variants is only 1.2 pp. A supplemental token-matched control adds 1,512 runs and finds that Full skills perform similarly to task-irrelevant skill-formatted content. The results caution against using one LLM-generated skill per data-science workflow as a default single-shot prompting strategy.
[AI-9] mEE: End-to-end Time Series Classification via In-Context Learning
链接: https://arxiv.org/abs/2607.07500
作者: Jaris Küken,Shi Bin Hoo,Martin Mráz,Frank Hutter,Lennart Purucker
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Time series classification (TSC) is dominated by a two-stage paradigm: train a feature encoder – either from scratch on the target dataset or via pretraining on large corpora – and then fit a task-specific classifier on top. While effective, this decoupling optimizes representation learning independently of the classification objective, requires per-dataset training, and prevents the model from exploiting label information during inference. We introduce TimEE, a 4.5M-parameter foundation model for end-to-end TSC via in-context learning. Given a labeled support set and a query time series, TimEE directly outputs a predicted class distribution in a single forward pass with no per-dataset training required. Following the prior-data fitted network (PFN) framework, TimEE is meta-trained exclusively on synthetic TSC tasks, where each task contains time series with distinct class identities arising from structured distributional shifts in the generative process. Despite seeing no real time series during pre-training, TimEE ranks first in ROC AUC (and third on accuracy) on the UCR benchmark among all compared methods, which include both foundation models and supervised deep learning baselines. To our knowledge, TimEE is the first purely synthetic-pretrained model to reach state-of-the-art performance on the UCR benchmark. These results establish end-to-end ICL with synthetic priors as a compelling, largely unexplored direction for TSC, with scaling, prior design, and richer generation mechanisms as natural avenues for improvement. Code is publicly available at this http URL.
[AI-10] Reward-Adaptive Iterative Discovery: A Case Study on Automated Game Testing for NHL26
链接: https://arxiv.org/abs/2607.07498
作者: Florian Fuchs,Jessy Gosselin-Grant,Boris Skuin,Michele Petteni,Alessandro Sestini,Joakim Bergdahl,Amir Baghi,Linus Gisslén
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Reinforcement Learning Conference - Reinforcement Learning and Video Games Workshop 2026
Abstract:Testing is a major effort for the gaming industry, requiring a significant part of development budget and people power. We present a case study on a development version of the ice hockey game EA SPORTS NHL 26, for which human playtesters test the goalie AI for behavioral exploits. To reduce the effort of re-testing the goalie AI after every game or behavior modification in the development phase, we propose Reward-Adaptive Iterative Discovery (RAID), a novel approach to automatically find exploits using an iterative Reinforcement Learning (RL) approach that trains a population of goal scoring agents. While previous approaches can already successfully find exploits, RL algorithms tend to overfit to a single solution. We introduce a simple extension on top of existing RL algorithms, such that they find multiple diverse high-quality solutions. For our first deployment of this approach, within a single experiment we were able to find six hockey scoring exploit strategies that were qualitatively similar to those that playtesters had found in hours-long manual testing sessions.
[AI-11] Search Fail Recover: A Training Framework for Correction-Aware Reasoning
链接: https://arxiv.org/abs/2607.07492
作者: Dmitry Beresnev,Vladimir Makharev,Roman Khalikov,Ivan Oseledets,Petr Anokhin
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Many reasoning tasks are not well described by a single left-to-right chain: a solver may need to pursue a plausible branch, observe delayed failure, and return to the latest prefix that can still be completed. We introduce Pyligent, a training and inference framework inspired by the Diligent Learner formulation that represents reasoning as validated search over partial solution chains. A task validator labels generated continuations and failures, and the resulting search trees are converted into supervised targets for three actions: continue, finish, and backtrack, with optional traces that summarize abandoned branches. We evaluate Pyligent on a hidden directed graph task designed to isolate delayed-failure recovery, and on structured reasoning domains with exact validators, including 4\times4 Sudoku, Sudoku with reasoning traces, and Blocksworld. Compared with gold-only supervised fine-tuning, Pyligent improves solve rate by 72.7 percentage points on hidden graphs, by 17 and 18 points on mixed and expert Sudoku, by 27 and 14 points on mixed and expert Sudoku with reasoning traces, and by 13 points on Blocksworld. These results suggest that explicit failed-branch supervision can teach useful recovery behavior beyond imitation of polished solution chains.
[AI-12] Where to Intervene? Benchmarking Fairness-Aware Learning on Differentially Private Synthetic Tabular Data
链接: https://arxiv.org/abs/2607.07471
作者: Vinícius Gabriel Angelozzi,Héber H. Arcolezi
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注: Paper accepted at PETS 2026. Code is available at this https URL
Abstract:Machine learning models are increasingly deployed in high-stakes domains, raising concerns about both privacy and fairness. Differential Privacy (DP) has become a gold standard for privacy-preserving data analysis, while fairness-aware mechanisms aim to mitigate discrimination against underrepresented groups. However, these objectives can conflict: DP often amplifies disparities across demographic groups, and little is known about whether established fairness interventions remain effective under DP constraints. In this work, we present, to our knowledge, the first systematic evaluation of fairness interventions on differentially private synthetic tabular data. Our benchmark centers on the Adaptive Iterative Mechanism (AIM), identified as the state-of-the-art marginal-based DP synthesizer (Cormode et al. 2025). We thus evaluate fairness interventions across four datasets, multiple group fairness metrics, and three categories of mitigation strategies (pre-processing, in-processing, and post-processing) under a wide range of privacy budgets. We compare four pipeline configurations: (Baseline) training on original data; (DP-only) training on DP synthetic data; (Fair-only) applying fairness mechanisms on original data; and (DP+Fair) combining fairness mechanisms with DP synthetic data. Our results demonstrate that while DP alone can degrade both utility and fairness, applying fairness interventions can partially restore equitable outcomes. Among them, post-processing methods tend to provide more stable fairness-utility trade-offs across privacy budgets and synthesizers, achieving strong fairness improvements while preserving competitive utility relative to other intervention stages. We release all code, data, and experimental artifacts in an open-source repository to ensure full reproducibility and to support future research on the privacy-fairness-utility trade-off.
[AI-13] SpaCellAgent : A Self-Evolving LLM -Based Multi-Agent Framework for Trajectory Analysis
链接: https://arxiv.org/abs/2607.07467
作者: Songhan Wang,Haoang Chi,He Li,Zhiheng Zhang,Jiayan Yuan,Cheems Wang,Hao Peng,Xinwang Liu,Wenjing Yang
类目: Artificial Intelligence (cs.AI)
备注: 27 pages, 19 figures
Abstract:Spatial and Single-cell transcriptomics are transformative in deciphering cellular dynamics. As the fundamental paradigm for reconstructing cell developmental paths, trajectory inference (TI) is critical. However, existing methods require extensive manual intervention and proficiency in heterogeneous tools, posing a significant barrier to efficient TI analysis. To bridge this gap, we propose SpaCellAgent, an autonomous large language model (LLM) multi-agent framework that automates end-to-end spatiotemporal analysis and narrative generation. SpaCellAgent utilizes a multi-agent architecture for strategic workflow planning, a dynamic tool-orchestration engine for adaptive algorithm selection, and a self-evolution module that iteratively refines performance through feedback. We evaluate SpaCellAgent on six heterogeneous datasets encompassing complex temporal developmental trajectories, diverse sequencing platforms, and spatially-resolved tissue architectures. SpaCellAgent consistently demonstrates over 40% improvement in analytical efficiency while maintaining expert-aligned performance. By converting natural language specifications into optimized analytical workflows and fully automating the pipeline, SpaCellAgent democratizes advanced spatiotemporal modeling and establishes a scalable, agent-driven paradigm for computational biology. The code and materials are available at this https URL.
[AI-14] RLVP: Penalize the Path Reward the Outcome
链接: https://arxiv.org/abs/2607.07435
作者: Bojie Li,Noah Shi
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Agents acting on our behalf in the real world (e.g. placing phone calls) must learn online from costly, often irreversible interactions rather than cheap simulator steps. Two things follow. First, deployability depends on the path, not only the outcome. An agent must respect outcome-neutral constraints such as not repeatedly calling an unresponsive user, respecting business hours, or completing required authentication constraints that outcome-based rewards cannot express, since violating them frequently improves apparent success. Second, because each interaction is expensive, the agent must learn efficiently from very few examples. Reinforcement learning from verifiable rewards (RLVR) is blind to both challenges: it optimizes solely on the outcome and wastes expensive rollouts on all-fail groups where group-relative advantage collapses to zero. Attempts to densify supervision by rewarding progress target the hard-to-verify direction. In contrast, real agentic environments can cheaply detect bad moves. Since group-relative advantage is equivalent to within-group variance, a dense signal helps only when it supplies variance the outcome lacks. A verifiable penalty on the path meets this condition reliably, while a progress potential helps only where partial progress is reachable. The resulting recipe “penalize the path, reward the outcome” achieves high task success with near-zero violations, where outcome-only training violates constraints on nearly every episode. We provide four design rules for effective penalties, including avoidance of the inaction trap that arises when a penalty is used in isolation. Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI) Cite as: arXiv:2607.07435 [cs.LG] (or arXiv:2607.07435v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2607.07435 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-15] Reason Less Verify More: Deterministic Gates Recover a Silent Policy-Violation Failure Mode in Tool-Using LLM Agents
链接: https://arxiv.org/abs/2607.07405
作者: Vikas Reddy,Sumanth Reddy Challaram,Abhishek Basu
类目: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注:
Abstract:Tool-using LLM agents can violate the very policies they are deployed to enforce while appearing to complete the task successfully. In policy-permissive environments, a tool may execute any well-formed call even when the corresponding state transition is forbidden by domain policy. The result is a silent wrong state (a booking cancelled, a passenger count changed, a claim acted on without verification) that neither the tool nor the agent’s self-report exposes. We study this failure mode in the \tau^2 -bench airline domain. On a budget agent, 78% of observed failures are silent wrong-state failures with no tool error, and the aggregate failure rate is reproducible across disjoint seeds, not sampling noise. We then evaluate a lightweight intervention: deterministic, read-only pre-execution gates that inspect the proposed call and current state before allowing a write. A four-gate suite raises full-benchmark success from 29.6% to 42.0% on gpt-4o-mini (+12.4pp; paired task-level bootstrap P=0.0012), and the lift reproduces on a disjoint 15-seed set (+12.3pp; P=0.0008). The effect is concentrated where the gates fire: on the 26/50 firing tasks, success rises by +19.2pp, while movement on the 24 non-firing tasks does not exclude zero. Two negative controls (a self-enforcing retail domain and BFCL) bound the mechanism: gates help when tools are policy-permissive and add little where tools already self-enforce. As suggestive evidence, not a central claim, the same failure mode persists at the frontier: gpt-5.2 at default reasoning still attempts policy-violating writes, and the same suite improves success from 61.2% to 71.6% (+10.4pp; P=0.020; n=5, no replication). The contribution is a bounded evaluation and reliability result: deterministic gates do not guarantee task success, but they can deterministically prevent a known class of silent policy-violating writes at the action boundary. Subjects: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR) Cite as: arXiv:2607.07405 [cs.AI] (or arXiv:2607.07405v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2607.07405 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-16] Agent ic Data Environments
链接: https://arxiv.org/abs/2607.07397
作者: Elaine Ang,Chenxi Huang,Georgios Liargkovas,Jerry Liu,Jinhui Liu,Nikos Pagonas,Charlie Summers,Haonan Wang,Jiakai Xu,Tianle Zhou,Yusen Zhang,Zhou Yu,Zhuo Zhang,Tianyi Peng,Kostis Kaffes,Eugene Wu
类目: Artificial Intelligence (cs.AI); Databases (cs.DB)
备注:
Abstract:Autonomous agents promise substantial gains in speed, scale, and labor efficiency, but their failures can impose abrupt and often irreversible costs. The central challenge for agentic automation is therefore to increase the benefits of automation while bounding the consequences of failure. While databases remain central to modern computing, agents operate over a broader data environment spanning files, APIs, applications, and system state. In this talk, I will outline early work on Agentic Data Environments – the execution substrate in which agents operate – that both amplify agent capabilities and enforce safety guarantees. This perspective reframes data systems from passive stores of state into active substrates for safe, reliable execution. Subjects: Artificial Intelligence (cs.AI); Databases (cs.DB) Cite as: arXiv:2607.07397 [cs.AI] (or arXiv:2607.07397v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2607.07397 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Journalreference: IEEE Data Bulletin Vol. 50 No. 1 2026
[AI-17] MIRA-Math: A Benchmark for Minimal Information Requesting and Mathematical Reasoning
链接: https://arxiv.org/abs/2607.07391
作者: Charbel Al Bateh,Samer Saab Jr
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Mathematical reasoning benchmarks typically provide all facts needed to solve each problem, while interactive benchmarks often mix reasoning with tools, retrieval, and long-horizon dialogue. We introduce MIRA-Math, a benchmark for a narrower diagnostic capability: solving mathematical problems whose full latent state has a unique answer, but whose solver-facing view is missing exactly one necessary atomic fact. The solver must request the missing information in natural language under a strict budget and then integrate the returned fact into an exact final answer. A fixed constrained LLM responder sees only the dataset-provided atomic fact and must either offer the quoted fact when the request matches it, or decline otherwise. Thus, instance generation, typed hint specifications, validation, and final-answer verification are deterministic, while request metrics are measured under a fixed LLM-mediated responder channel. MIRA-Math contains 2,310 generated instances from 22 typed mathematical families spanning algebra, probability, linear systems, discrete structures, signal processing, Markov chains, circuits, interpolation, and numerical boundary-value problems. Experiments across frontier and small models show that request success and final-answer accuracy are separable: models may ask for the right fact yet fail the downstream computation, or fail before obtaining the canonical hint. We release generators, verifiers, prompts, run metadata, and dataset documentation to support reproducible evaluation of minimal information requesting in mathematical reasoning.
[AI-18] Physics-Audited Agent ic Discovery in Scientific Machine Learning
链接: https://arxiv.org/abs/2607.07379
作者: Diab W. Abueidda,Bilal Ahmed,Panos Pantidis,Mostafa E. Mobasher
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:In agentic scientific machine learning (SciML), large language model (LLM) agents can discover surrogate models and select one by an automated score, typically an error metric. A low error, however, does not establish that the predicted fields satisfy the physics that matter for mechanics, such as boundary conditions, superposition, stiffness scaling, or causality. We introduce Physics-Audited Agentic SciML (PA-SciML), a verification-first workflow for agentic SciML discovery. The workflow fixes a scoring evaluator before search, derives reviewable machine-checkable physics requirements, checks each trained candidate on its outputs, and separately searches prescribed input ranges or measured load-history spans for high-violation cases without reference solution fields. A surrogate is reported as verified only under the stated checks. When enabled, the workflow also adds advisory numerical probes before training and tests one modeling change at a time to record which isolated edits are associated with score gains before reuse. In the reported computational-solid-mechanics numerical examples, the static elasticity run selects a surrogate with lower validation error than the error-only baseline while both selected models pass the common linear-elastic checks. In the transient elastodynamics run, an error-only baseline with similar mean error fails a stricter causality check by responding to future parts of the loading history, while the selected surrogate passes the stated checks. The main distinction is per-candidate physics evidence on predicted fields, not a richer aggregate score.
[AI-19] On Adversarial Vulnerability of Vision-Language Models through the Lens of Intermediate Spectral Subspaces
链接: https://arxiv.org/abs/2607.07375
作者: Chethan Krishnamurthy Ramanaik,Tobias Callies,Michael Hecht,Eirini Ntoutsi
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Adversarial vulnerability in deep neural networks (DNNs) has been studied from the perspectives of decision-boundary geometry, feature robustness, input-output Jacobians, and the instability of inverse problems. Here, we focus on the spectral structure of intermediate linear transformations that propagate information through modern DNNs, an unexplored mechanism of adversarial vulnerability. Specifically, we investigate transformer-based vision-language models, whose linear layers admit interpretable spectral decompositions and whose widespread adoption makes understanding their robustness increasingly important. We propose a white-box spectral-subspace-guided attack (SSGRA) that aligns intermediate representations with the subspace spanned by the bottom right singular vectors. Our experiments show improved attack effectiveness over existing baselines. In addition, SSGRA offers a spectral interpretation of adversarial vulnerability in VLMs, providing insights for improving their robustness.
[AI-20] HumAIN: Human-Aware Implicit Social Robot Navigation IROS2026
链接: https://arxiv.org/abs/2607.07357
作者: Daeun Song,Nhat Le,Jeffrey Chen,Mohammad Nazeri,Amirreza Payandeh,Rohan Chandra,Reuth Mirsky,Ross Mead,Ling Xiao,Xuesu Xiao
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: 8 pages, 4 figures. Accepted to the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2026)
Abstract:Effective social robot navigation requires sensitivity to human behavior, often revealed through subtle skeletal cues like gait and orientation. We present Human-Aware Implicit Social Robot Navigation (HumAIN), a novel framework that fuses implicit social cues directly into the planning loop via knowledge distillation. We first employ a transformer-based teacher model that fuses rich multi-modal inputs, including historic images, skeletal keypoints, robot state, and a robot’s target goal, to learn robust, human-aware representations for the robot’s future trajectory planning. To enable real-time deployment, we then distill this knowledge into a lightweight student model. By optimizing for both trajectory reconstruction and latent feature alignment with the teacher, the student learns to infer complex social dynamics from minimal inputs. Bridging the prediction-planning gap with an efficient distilled architecture, our method enables robots to reason about human behavior in a manner that is adaptive, robust, and socially compliant. We validate HumAIN through extensive experiments, where it improves trajectory prediction metrics by an average of 29.8% across all metrics compared to state-of-the-art baselines. These results highlight the benefit of using implicit, whole-body cues to achieve human-like navigation awareness on resource-constrained platforms.
[AI-21] Latency-Aware Bid Acceptance under Operational Feasibility: A Public Benchmark with Hindsight Ceilings
链接: https://arxiv.org/abs/2607.07343
作者: Aswin Chandrasekaran
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Optimization and Control (math.OC)
备注: 20 pages. Benchmark, code, and run manifests: this https URL
Abstract:Online truckload bid acceptance is a closed-loop stochastic decision problem in which a carrier or broker must, in real time, accept or reject a tendered load subject to operational feasibility, fleet repositioning costs, and opportunity cost against future demand. Public, reproducible benchmarks for this problem are scarce: existing routing benchmarks are static, while dynamic-fleet studies typically rely on private operator data. We introduce FreightBidBench, a public-calibrated, dependency-free, closed-loop benchmark in which feasibility (pickup reach, appointment windows, simplified hours-of-service, stochastic yard delays) and economics (service-failure penalty, terminal fleet value, daily price-premium window) are explicit, versioned, and reproducible from public Freight Analysis Framework and U.S. Department of Agriculture truck rate data. We develop two full-horizon hindsight ceilings: a simple LP style relaxation and a tighter Lagrangian-per-truck information relaxation that retains per-truck hours-of-service and sequencing structure and is 20.7% tighter than the LP relaxation on a tight-capacity scenario and 39.3% tighter on a scarce-capacity scenario. We introduce a parametric surrogate-rollout cascade with boundary-band and scarcity-pressure escalation triggers. On ten-seed tight and scarce scenarios, the best simple policy retains 91.0% and 86.5% of rollout profit and the standard-library surrogate 94.2% and 89.3%; a cascade at a single escalation band recovers roughly 98% on both at 40-56% of rollout’s mean decision latency, and on the tight scenario is statistically indistinguishable from the rollout teacher (paired-bootstrap 95% CI on the profit delta spans zero).
[AI-22] Hypergraph Neural Stochastic Diffusion: An SDE Framework for Uncertainty Estimation
链接: https://arxiv.org/abs/2607.07330
作者: Zhiheng Zhou,Mengyao Zhou,Dengyi Zhao,Xingqin Qi,Guiying Yan
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 26 pages,6 figures
Abstract:Hypergraph neural networks have shown powerful capability in modeling higher-order relations, yet their predictive uncertainty remains underexplored. Unlike pairwise graphs, uncertainty in hypergraphs arises not only from noisy attributes and ambiguous labels, but also from variations in node-hyperedge incidence structures and complex higher-order dependencies. Existing approaches mainly estimate uncertainty from final predictions or rely on computationally expensive ensembles and Bayesian inference, limiting their ability to capture uncertainty evolution during representation learning. In this paper, we propose Hypergraph Neural Stochastic Diffusion(HyperNSD), a stochastic differential equation framework for uncertainty estimation on hypergraphs. HyperNSD models hypergraph representations as stochastic processes evolving over node-hyperedge incidence structures. A learnable drift function captures deterministic higher-order diffusion dynamics, while a learnable stochastic forcing function characterizes structural ambiguity and representation noise. Predictive uncertainty is directly quantified through the variability of stochastic representation trajectories, providing an intrinsic uncertainty measure beyond post-hoc confidence scores. We formulate HyperNSD with neural drift and diffusion networks, enabling joint learning of prediction and uncertainty propagation. Theoretical analyses establish well posedness, perturbation stability,permutation equivariance, and numerical convergence of the proposed stochastic dynamics. Experiments on multiple hypergraph benchmarks demonstrate that HyperNSD achieves reliable uncertainty estimation for out-of-distribution and misclassification detection while preserving competitive prediction accuracy. These results provide a principled stochastic-dynamical framework for trustworthy higher-order representation learning.
[AI-23] FedCVESA: Taking Away Training Data in Federated Learning via Correlation Value Encoding and Segmented Aggregation
链接: https://arxiv.org/abs/2607.07314
作者: Chongkai Li,Bang Zhang,Wenjian Luo
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注: 16 pages, 5 figures, 2 tables
Abstract:Federated learning (FL) avoids explicit data exposure by keeping raw data on local clients, yet privacy risks remain in the training process and the learned model itself. Recently, centralized Taking Away Training Data (TATD) attacks have shown that malicious training could abuse the memorization capacity of deep models to store and later recover training data. However, this memorization-based threat has not been systematically studied under FL environments, where multi-client averaging could overwrite encoded training data. In this paper, we study a white-box TATD attack in which a malicious server selects n target clients from K participating clients and actively writes private training data into the global model during federated training. We propose FedCVESA, a federated variant of Correlation Value Encoding Attack (CVEA), by adding a Pearson-correlation regularizer to the loss function of target clients, so that private training data are gradually encoded into selected model parameters, referred to as carrier parameters. To reduce the overwriting of carrier parameters during server aggregation, we further propose segmented aggregation over dispersed carrier parameters, preserving selected carrier parameters while keeping standard averaging on the remaining parameters. Experiments on MNIST, Fashion-MNIST, and CIFAR-10 under Dirichlet non-IID partitions show that the proposed method can steal semantically meaningful private training images from the trained model while maintaining acceptable main-task utility in a controlled proof-of-concept setting. These results demonstrate that FL can become a parameter-level memorization channel for active TATD attack under the studied white-box malicious-server setting.
[AI-24] FMMVCC: Fuzzy Mamba-based Multi-View Contrastive Clustering for Univariate Time Series
链接: https://arxiv.org/abs/2607.07258
作者: Donato Cerciello,Leonardo Schiavo,Angel Panizo-LLedot,Javier Huertas Tato,David Camacho
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:In many realistic scenarios, large volumes of time series data are generated with limited or expensive annotations. This limitation makes supervised learning methods difficult to apply and leads to the use of unsupervised approaches capable of discovering meaningful structures directly from raw data. Clustering therefore plays a crucial role in organizing time series into groups that share similar temporal patterns, enabling exploratory analysis and downstream tasks without requiring manual labeling. However, existing deep clustering methods often struggle to capture long-range temporal dependencies or rely on architectures with high computational cost. This paper introduces FMMVCC, a Mamba-based deep clustering framework for time series that leverages state space sequence modeling to efficiently learn temporal representations with linear complexity. Additionally, it utilizes multi-view self-supervised learning with temporal masking and augmentations. Experimental evaluation in 15 benchmark datasets proves that FMMVCC consistently outperforms state-of-the-art baselines, achieving the best overall performance in 29 of 60 total metric evaluations and the highest average rank in all tested scenarios.
[AI-25] ORCAID: Oblique Rule-Based Continuous-Action Interpretation for Deep RL Policies
链接: https://arxiv.org/abs/2607.07235
作者: Ignacio D. Lopez-Miguel,Ezio Bartocci,Thomas Eiter,Martin Tappler
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注: 33 pages, 8 figures, accompanying source code available at this https URL
Abstract:Explainability remains a key issue in reinforcement learning (RL). Distilling an interpretable policy from an agent trained in a complex environment is particularly challenging when the action space is continuous. We introduce ORCAID, a novel method for extracting interpretable rule-based policies from RL agents operating in mixed continuous-discrete environments with continuous action spaces. Our main contribution is an efficient oblique decision tree training algorithm that partitions the state space by hyperplanes and fits local linear models. The key idea lies in a three-stage split search: efficient random initialization, local refinement, and backward elimination. Finally, adjacent leaves are merged to yield a concise set of interpretable rules describing a given deep RL policy. We evaluate ORCAID across multiple RL environments, demonstrating that the extracted rule-based policies maintain strong performance with a low number of parameters and can even be used to improve the performance of the original deep RL policy.
[AI-26] Reasoning Consistency Scanning: A Framework for Auditing Chain-of-Thought Validity in AI Safety Evaluations
链接: https://arxiv.org/abs/2607.07229
作者: Silvia Santano
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Prior work has shown that chain-of-thought (CoT) reasoning is often unfaithful: a model’s stated reasoning does not reliably reflect the process that produced its output. Detecting unfaithfulness, though, requires controlled experimental interventions, which cannot be applied to evaluation transcripts after the fact. We turn instead to a more tractable question that has received less attention: whether the stated reasoning is logically consistent with the answer it accompanies. Unlike faithfulness, consistency can be assessed from a transcript alone, with no intervention. We introduce reasoning consistency scanning, a reusable method for detecting this property in AI safety evaluation transcripts. Our contributions are fourfold. First, we formalize reasoning consistency as distinct from faithfulness and define a six-subtype taxonomy of inconsistency. Second, we build a validated benchmark of 60 transcripts, manually adapted from InstrumentalEval outputs. Third, we implement a working scanner for InspectScout, the first to target this property in safety evaluation transcripts. Fourth, we report results across four generator models and three evaluations from inspect_evals, showing that reasoning inconsistency is present, detectable, and varies systematically across both models and task types.
[AI-27] Validate the Dream Before You Trust Its Verdict: Admissibility for World-Model Simulators
链接: https://arxiv.org/abs/2607.07196
作者: Christian Oefinger,Finn Rasmus Schäfer,Korbinian Moller,Mattia Piccinini,Johannes Betz
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Software Engineering (cs.SE)
备注: Accepted at RSS 2026 Workshop on Robot World Models
Abstract:Across robotics, World Models (WMs) are increasingly used to evaluate action policies by simulating the consequences of actions in an imagined world, and returning a success or safety verdict. Yet a verdict is only as trustworthy as the WM that produced it, and the WM itself needs to be certified. In video-generation WMs, fidelity metrics such as Fréchet Video Distance (FVD) reward visual realism, but ignore whether the world responds correctly to the policy’s actions, including those unseen in training. Classical simulation-based validation assumes a trusted simulator evaluating an untrusted policy, whereas generative WMs are themselves unverified learned artifacts. Hence, we argue that any WM used as a test oracle must first be accredited before its verdicts can serve as evidence. Building on credibility practices from safety-critical simulation, including Verification, Validation Accreditation (VVA), Safety of the Intended Functionality (SOTIF), and scenario-based testing standards, we define an admissibility ladder (L0-L4) that a WM must climb before its closed-loop verdicts are accepted as assurance evidence. Our framework is embodiment-agnostic, and is instantiated in autonomous driving (AD), where assurance methods for traditional simulation are most mature. Applied to two driving WMs, the lower rungs reveal a reversal: the model that ranks higher on visual generation quality (L0) ranks lower on action-following (L1-L2), so visual fidelity does not predict the action-robustness a closed-loop verdict depends on.
[AI-28] Does AI Understand Imaging? A Systematic Benchmark of Agent ic AI for Computational Imaging Tasks
链接: https://arxiv.org/abs/2607.07189
作者: Ethan Chung,Chuanjun Zheng,Jasper Tan,Jingxi Li,Haopeng Zhang,Huaijin Chen
类目: Artificial Intelligence (cs.AI)
备注: 14 pages, 11 figures. Preprint / work in progress. Paper Webpage: this https URL
Abstract:Vision-language models (VLMs) and agentic AI have shown strong performance on semantic visual tasks, but it remains unclear whether they can handle the physics and inverse problems that underlie computational imaging. We present ImagingBench, a benchmark of 20 computational imaging tasks spanning five categories: ray and wave optics, image signal processing, inverse reconstruction, computational sensing, and calibration. ImagingBench evaluates three complementary settings: Expert, fixed expert-guided inverse reconstruction; Planner, planner-guided inverse reconstruction; and Forward, forward-system simulation for consistency checking. We benchmark leading proprietary and open-source image-centric multimodal systems, including Gemini, GPT, and Qwen, and compare them with representative task-specific non-agentic baselines. Across tasks, agentic models remain consistently weaker than specialized methods, especially on computational sensing problems such as lensless imaging, event-based reconstruction, time-of-flight imaging, and holography. Planner guidance provides only modest and inconsistent gains over the fixed-prompt Expert baseline. Although the models often generate visually plausible outputs, their reference-based fidelity remains poor, revealing a substantial gap between semantic visual competence and physically grounded imaging performance. ImagingBench provides a unified testbed for measuring this gap and tracking progress in agentic AI for computational imaging.
[AI-29] Predicting LLM Safety Before Release by Simulating Deployment
链接: https://arxiv.org/abs/2607.07184
作者: Marcus Williams,Hannah Sheahan,Cameron Raymond,Tomek Korbak,Deng Pan,Peilin Yang,Leon Maksin,Ningyi Xie,Phillip Guo,Ian Kivlichan,Micah Carroll
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 31 pages
Abstract:Pre-deployment safety evaluations aim to inform the downstream risks of releasing a new AI model. Yet most evaluations provide limited evidence about how often undesired model behavior will occur in deployment: they generally have insufficient coverage, are unrepresentative, and are generally recognizable as tests. To address these concerns, we study a simple way to simulate a model deployment: starting from de-identified conversations from a previous model deployment, we hold fixed the initial conversation prefix and regenerate the next response using a candidate model. The resulting responses can then both be audited for novel misalignments and used to estimate the prevalence of model misbehavior before deployment. We evaluate deployment simulation across four GPT-5-series deployments, using registered, outcome-blinded predictions for GPT-5.4 and retrospective analyses of three earlier releases. We find that deployment simulation produces informative estimates of post-deployment misbehavior rates and outperforms baselines based on adversarially selected production data; its evaluation-awareness point estimates were also much closer to production traffic than those from traditional evaluations. We also identify the realism of tool resampling as a central challenge for further improving predictions and share results suggesting that this challenge is surmountable even in complex tool-use settings. Finally, we show that deployment simulation can be seeded from public chat datasets and remain informative about production misbehavior rates, suggesting a path for external researchers to run deployment-grounded evaluations without access to private production logs. Overall, deployment simulation helps evaluators forecast how language models will behave in the real world and supports more quantitative assessment of deployment risk.
[AI-30] Entropy Pacing Policy Optimization for Multi-Task Agent ic Reinforcement Learning
链接: https://arxiv.org/abs/2607.07178
作者: Zetian Hu,Shunyu Liu,Junjie Zhang,Yongcheng Jing,Ting-En Lin,Yongbin Li,Dacheng Tao
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Recent breakthroughs of Reinforcement Learning (RL) have highlighted its potential for complex agentic Large Language Model (LLM) tasks. However, existing efforts largely focus on single-task settings, whereas real-world deployment necessitates a generalist agent capable of solving multiple tasks simultaneously. In this work, we identify a critical yet underexplored phenomenon in multi-task agentic RL: different tasks can exhibit exploration-exploitation pace mismatch. Specifically, easier tasks may converge early to low-entropy policies that hinder learning on harder tasks, while harder tasks can, in turn, push easier tasks back toward high-entropy exploration. This back-and-forth interaction creates inter-task entropy crossovers and frequent entropy spikes. Inspired by this observation, we introduce Entropy Pacing Policy Optimization (EPPO) for multi-task agentic LLMs, which coordinates entropy across tasks to stabilize multi-task optimization. At the core of EPPO is a task-wise dynamic clipping mechanism that replaces the fixed clipping threshold in Group Relative Policy Optimization (GRPO) with a task entropy-aware adaptive bound, tightening updates for over-confident tasks while relaxing them for under-explored ones. Experiments on the multi-task agentic benchmarks demonstrate that the proposed EPPO yields results superior to its counterparts.
[AI-31] GeoProp: Grounding Robot State in Vision for Generalist Manipulation
链接: https://arxiv.org/abs/2607.07101
作者: Guoyang Zhao,Quanhao Qian,Gongjie Zhang,Wenhao Li,Jiuniu Wang,Xiaowei Lu,Deli Zhao,Ran Xu
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: 21 pages, 8 figures, 11 tables. Project page: this https URL
Abstract:Proprioception is fundamental to robotic manipulation, yet standard fusion methods often treat it as an isolated vector lacking explicit alignment with visual tokens. Without a direct correspondence between 3D kinematics and 2D feature maps, manipulation policies struggle to ground the robot’s state within the scene, frequently underperforming even vision-only baselines. To address this, we introduce GeoProp, a lightweight, plug-and-play adapter that aligns proprioception with vision through explicit geometric grounding and spatial feature sampling. GeoProp projects the robot state onto the image plane to sample localized visual features, constructing a grounded state token. It then injects state-derived spatial priors into the corresponding visual features via FiLM modulation. To capture motion intent, GeoProp further samples features at a short-horizon predicted coordinate derived from recent kinematics, providing look-ahead visual context. Across 67 tasks, GeoProp improves Diffusion Policy by 8.7% on 63 simulation tasks and pi_0 by 4.0% on the RoboTwin subset, and yields a 10.6% average gain across both policy families in the real world, while adding only 2-3% to the parameter count. These results demonstrate that GeoProp is a simple yet high-impact inductive bias for generalist embodied policies. Project page: this https URL.
[AI-32] Multiplication Beyond Groups: Stratified Fourier Mechanisms in Transformer Circuits ICML2026
链接: https://arxiv.org/abs/2607.07066
作者: Zitong Andrew Chen,Junaid Hasan,Akhil Srinivasan,Hemkesh Bandi,Jarod Alper
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Number Theory (math.NT); Representation Theory (math.RT)
备注: 29 pages, 15 figures. Spotlight at the Mechanistic Interpretability Workshop at ICML 2026. First three authors contributed equally. Code at this https URL
Abstract:Transformers have demonstrated a remarkable ability to learn algorithmic reasoning, yet mechanistic analyses have mostly focused on globally invertible operations such as cyclic addition and group composition. In this work, we investigate how small transformers learn modular integer multiplication over composite moduli, a fundamentally non-invertible operation due to the presence of zero-divisors. We propose the monoid extension: a localized generalization of Group Composition via Representation (GCR) that suggests the learned computation does not rely on a single global representation space. Instead, the model partitions the input space into local hierarchical algebraic regions, where group-like structure survives and Fourier mechanisms can be applied. In transformers trained on square-free modular multiplication, we find that embeddings organize around these regions, attention exhibits class-sensitive routing and low-rank write directions, and local character features explain a large fraction of the model’s output logits. Our results suggest that representation-theoretic mechanisms previously identified for group operations can extend beyond groups to more general structures.
[AI-33] Complexity-Budgeted Interaction-Aware Interpretable Model for Tabular Data
链接: https://arxiv.org/abs/2607.07060
作者: Srikumar Krishnamoorthy
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Inherently interpretable classifiers for tabular data typically rely on sparse features, rules, or patterns that users can inspect directly. The marginal feature-screening step common to these methods can discard variables whose predictive value emerges only through joint configurations with other variables. We present Interaction Aware Interpretable Machine Learning (IAIML), a framework that addresses this limitation through three coordinated mechanisms: adaptive per-feature discretization, finite-grid pairwise interaction scoring, and a partitioned explanation budget. Detected interactions are routed through one of two strategies: relaxing the screening filter so that interaction-supported variables enter the pattern search, or constructing explicit pair terms for a sparse downstream classifier. On a 40-dataset panel comprising 24 real-world tabular benchmarks and 16 synthetic interaction stress tests, evaluated under nested cross-validation, IAIML achieves mean AUC within 1.4 points of tuned gradient-boosted ensembles while requiring roughly 14–28 times fewer fitted explanation components. On datasets with strong pairwise interaction structure and low marginal signal, IAIML outperforms all baselines. Among compact interpretable methods, IAIML is comparable to RuleFit in AUC and component count and is less expensive to tune. EBM obtains a small but significant AUC advantage across the full panel, with a substantially larger lookup-table footprint. Performance degrades on datasets requiring higher-order interactions beyond the pairwise scope. Component-isolated ablations confirm that adaptive discretization and interaction-aware admission each contribute incrementally. These results support IAIML as a compact, interaction-aware framework appropriate for settings where bounded explanation size and controlled treatment of feature interactions are design requirements.
[AI-34] Measuring Intelligence Beyond Human Scale
链接: https://arxiv.org/abs/2607.07040
作者: Jerry Han,Rafael Moschopoulos,Ella Colby,Vishrut Goyal,Andrew Tu,Kia Ghods,Mark Braverman,Elad Hazan
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:How can we measure intelligence beyond human capability? Human-authored benchmarks saturate, and above human capability, examiners may not know which tasks are both hard and verifiable. We argue that this difficulty is inherent to absolute-scale evaluation and propose a new paradigm based on relative measurement in which models generate public challenges that separate other systems. Aggregating these outcomes yields an adversarial psychometric rating system that can scale with the systems being measured. We describe practical protocols that reduce incentives for private-information attacks, support judge-free adjudication, and naturally scale with agent capabilities. We instantiate the framework across verifiable and open-ended, non-verifiable domains, illustrating how model-generated evaluation can continue to measure systems beyond the human frontier. Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2607.07040 [cs.AI] (or arXiv:2607.07040v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2607.07040 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-35] On the Principles of Deep Feedforward ReLU Networks
链接: https://arxiv.org/abs/2607.07035
作者: Changcun Huang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:The architecture of deep feedforward neural networks is ubiquitous in deep learning, either as a whole system or as a subnetwork of other architectures, and thus its mechanism is a key ingredient of the black box of neural networks. On the basis of the simplest two-layer ReLU network, this paper systematically studies the mechanism of deep feedforward ReLU networks with multiple hidden layers and successfully explains the training solution obtained by the back-propagation algorithm. The concept of a path, especially in terms of the relationships between paths, plays a central role in uncovering the mystery of the black box. It is shown that a unit of a deep ReLU network can form a piecewise linear manifold to divide the input space, instead of a hyperplane of the two-layer case. How to efficiently use the hidden-layer units to produce both linear functions and partitions of the input space is also a central problem. The principles of a two-layer ReLU network can be generalized to the deeper case to a large extent, such as multiple strict partial orders and continuity restriction. The combination of the basic and simple principles proposed can yield complicated instantiations including the training solutions, and in this sense the black box of deep feedforward ReLU networks is revealed.
[AI-36] Intrinsic Greens Learning: Supervised Learning on Manifolds via Inverse PDE ICLR2026
链接: https://arxiv.org/abs/2607.07034
作者: Alexandre Quemy
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Accepted at AI PDE Workshop @ ICLR 2026
Abstract:We introduce Intrinsic Green’s Learning (IGL), a framework that models a target function on a manifold as the solution to a linear PDE whose source term is learned from data. Rather than approximating the target directly, IGL learns a source and integrates it against a Green’s kernel. An encoder discovers a low-dimensional coordinate chart on the manifold where both the source and the kernel decompose as low-rank tensors, collapsing a high-dimensional integral into independent one-dimensional integrals with cost linear in the intrinsic dimension. A two-stage algorithm separates coordinate discovery from source fitting, a near-convex linear solve, preventing the dimensional collapse of joint training. Learnable gates on each coordinate automatically discover the intrinsic dimension of the manifold. We validate IGL on synthetic manifolds and on MNIST, where it simultaneously achieves near-optimal classification and automatic recovery of the intrinsic dimension.
[AI-37] Gimitest: A Comprehensive Tool for Testing Reinforcement Learning Policies
链接: https://arxiv.org/abs/2607.07029
作者: Dennis Gross,Quentin Mazouni,Helge Spieker,Arnaud Gotlieb
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注:
Abstract:Reinforcement learning (RL) policies can be unsafe and vulnerable to attacks. Ensuring their reliability is often a pain point as existing automated testing methods target only selected environments, testing scenarios, and RL algorithms. To address this, we propose a comprehensive framework for testing single- and multi-agent RL policies under varying conditions. Our implementation of this framework, Gimitest, is an open-source tool that supports various gym frameworks and allows for modifications of their integrated components. This article describes the framework and details Gimitest’s functionality and architecture. It showcases its effectiveness in testing multiple RL policies in environments such as the official Farama Gymnasium and PettingZoo.
[AI-38] Multimodal Spatiotemporal-Frequency Fusion with Peak Enhancement for Cellular Traffic Forecasting
链接: https://arxiv.org/abs/2607.07016
作者: Qingzhong Li,Yue Hu,Hui Ma,Yajun Zhang,Xinjun Pei,Ming Yan,Fei Xing
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Accepted in the 2026 IEEE International Conference on Systems, Man, and Cybernetics (SMC)
Abstract:Accurate forecasting of cellular network traffic is essential for network planning, resource allocation, and quality-of-service assurance in modern mobile communication systems. Real-world traffic often exhibits bursty endogenous dynamics and disturbances triggered by external urban events, which makes reliable prediction highly challenging. Most existing spatiotemporal traffic forecasting methods primarily focus on intrinsic traffic patterns or structural relationships within a single modality, and rarely model burst behavior together with exogenous contextual signals. To address this issue, we propose \textbfMSPF-Net, a multimodal cellular traffic forecasting framework that integrates external contextual information. Specifically, MSPF-Net consists of a Spatiotemporal-Frequency Traffic Encoder for capturing temporal, spatial, and spectral traffic patterns, a Peak Enhancement Module for extracting burst-aware representations of sudden spikes, a News Context Representation Module for encoding urban news streams into exogenous contextual embeddings, and a Dynamic Fusion Prediction Module for adaptively integrating these heterogeneous signals to generate forecasts. Experiments on the Milano, Trento, and LTE traffic datasets demonstrate that jointly modeling traffic dynamics, burst patterns, and news contextual signals can effectively improve forecasting performance.
[AI-39] Physics-guided spatiotemporal neural models for fuel density prediction
链接: https://arxiv.org/abs/2607.06999
作者: Tolga Caglar,Jaynil Jaiswal,Saqib Azim,Yudhir Gala,Mai H. Nguyen,Ilkay Altintas
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: to be published in IEEEXplore
Abstract:This paper presents a physics-guided machine learning (PGML) framework for fuel density prediction, integrating physics constraints and domain knowledge into deep learning models to enhance model accuracy and stability. We explore three deep learning architectures – ConvLSTM, Adaptive Fourier Neural Operator (AFNONet), and Video Vision Transformer (ViViT) – to model the spatiotemporal evolution of fuel density. Our approach incorporates differentiable physics-informed terms in the loss function, including a mass-conserving fuel transport term and a rate-of-spread estimation. Experimental results, averaged across multiple independent trials, demonstrate that the proposed PGML framework outperforms purely data-driven baselines without physics constraints in both accuracy and stability. This framework enables computationally efficient, physically plausible fire forecasting to support adaptive prescribed burn management.
[AI-40] Large Behavior Model: A Promptable Digital Twin of the Retail Customer
链接: https://arxiv.org/abs/2607.06993
作者: Wachiravit Modecrua,Krittin Pachtrachai,Touchapon Kraisingkorn
类目: Artificial Intelligence (cs.AI)
备注: 17 pages, 5 figures
Abstract:Customer behavior modeling underpins recommendation, marketing, and decision support, yet existing approaches either optimize predictive accuracy without explaining decisions or simulate users without grounding them in real behavioral data. We present the Large Behavioral Model (LBM) that learns customer decision making directly from large-scale retail transactions through a unified Person-Environment formulation. Customer state is represented by a behavioral profile derived from historical purchases, while product context is incorporated through retrieval-augmented generation. The model is trained using continued pre-training on verbalized behavioral data, supervised fine-tuning for decision generation, and reinforcement learning with verifiable rewards for evidence-based calibration. We evaluate the proposed framework on purchase prediction, hard-negative discrimination, basket completion, promotion response, and cross-domain voucher redemption. The model consistently outperforms frontier general-purpose language models on in-domain retail tasks while demonstrating strong zero-shot and fine-tuned transfer across retailers and decision domains. Ablation studies show that continued pre-training is the primary driver of behavioral generalization, retrieval is most effective when applied during both training and inference, and reinforcement learning improves reliance on explicit behavioral evidence over generic language-model priors. These results demonstrate that behavioral knowledge encoded in transaction histories can be effectively learned by language models, providing a scalable foundation for customer digital twins and behavior simulation.
[AI-41] WAM-TTT: Steering World-Action Models by Watching Human Play at Test Time
链接: https://arxiv.org/abs/2607.06988
作者: Yusen Feng,Bingchen Han,Jiangran Lyu,Kai Liu,Yixin Zheng,Yuxuan Wan,Weiheng Liu,Sun Han,Ruiqin Li,Yulong Zhang,Fangfu Liu,Xuesong Shi,Libin Liu,Yizhou Wang,Zhizheng Zhang,He Wang
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Steering robot foundation models (RFMs) toward new task variants or user-preferred behaviors remains challenging, often requiring additional robot demonstrations, task-specific fine-tuning, or long-context conditioning. We present WAM-TTT, a test-time training framework for steering world action models from raw human videos. Rather than treating human videos as trajectories to imitate, WAM-TTT absorbs them into a lightweight adaptive memory inside a frozen WAM through self-supervised video prediction. To make this memory useful for control, we introduce a meta-training stage that aligns human demonstrations with robot behaviors using paired human-robot data and a key–value memory reconstruction objective. At test time, only unlabeled human videos are required to adapt the memory, while the pretrained WAM remains frozen. This enables efficient and reusable steering without robot actions, human-side annotations, or task-specific fine-tuning, while preserving the generalization ability of the foundation model. Extensive experiments show that WAM-TTT consistently outperforms in-context human-video conditioning baselines across diverse manipulation tasks and generalization settings.
[AI-42] Hybrid Least Squares/Gradient Descent Methods for MIONets
链接: https://arxiv.org/abs/2607.06976
作者: Jun Choi,Chang-Ock Lee,Minam Moon
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Numerical Analysis (math.NA)
备注:
Abstract:In this paper, we propose an efficient hybrid least squares/gradient descent (LSGD) method for MIONets to accelerate training. This method generalizes the LSGD method for DeepONets. Since MIONet is the sum of the entrywise product of multiple branch networks and a trunk network, it can be viewed as a multilinear function with respect to the last layer parameters of each branch network. These sets of parameters can be optimized using the alternating least squares method, where we solve the LS system for a single branch network in turn. To handle the large-sized system matrix, we introduce Kronecker and Khatri-Rao products and tensor permutation matrices to factor the large matrix into small ones. Our method is compatible with a general type of L^2 loss with regularization terms for the last layer parameters of each branch, where linear operators can be applied to the MIONet output in each loss term.
[AI-43] End-to-End LLM Flight Planning with RAG -based Memory and Multi-modal Coach Agent ICML2026
链接: https://arxiv.org/abs/2607.06964
作者: Amin Tabrizian,Arsyi Aziz,Aarifah Ullah,Mahyar Ghazanfari,Pouria Razzaghi,Peng Wei
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: Accepted at the ICML 2026 LM4Plan Workshop
Abstract:Bridging the gap between human pilot intent and autonomous flight operation is critical for real-world electric vertical takeoff and landing (eVTOL) aircraft deployment. Flight planning traditionally relies on classic algorithms that struggle to incorporate flexible human preferences. We present FRAMe, an End-to-End Large Language Model (LLM) Flight Planning tool with RAG-based Memory and Multi-modal Coach Agent. Our system integrates a planner LLM with a multi-modal coach agent and retrieval augmented generation (RAG)-based memory to generate flight plans that satisfy mission constraints while aligning with human flight operator preferences. We demonstrate the system in a range of real-world-inspired scenarios of varying difficulty levels. Across four LLMs, the full FRAMe system (RAG and coach) yields the highest validity for every planner (up to 93.8% aggregate, 99% on Easy scenarios for the strongest planner) and shifts preference-relevant metrics in the operator-favored direction where the metric has headroom. FRAMe signifies how advanced LLMs can be deployed for human-centric mission planning, translating natural language instructions into safe, efficient, and flexible flight routes. The code is available at: this http URL
[AI-44] Imputation Meets Clustering: Exploiting Latent Subgroup Structure for Missing Data Recovery ECML-PKDD2026
链接: https://arxiv.org/abs/2607.06930
作者: Chuyao Zhang,E Li,Taochen Chen,Yiqun Zhang,Yuzhu Ji,Shuping Zhao,Peng Liu,Yiu-ming Cheung
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Accepted to ECML-PKDD 2026
Abstract:Missing data is prevalent in practical applications, making effective imputation an essential preprocessing step for downstream analysis. Real-world datasets often exhibit complex latent structures composed of multiple subgroups with distinct distributions. However, existing methods often overlook such population heterogeneity. Without explicit structural guidance, these methods tend to produce generic estimates that blur subgroup boundaries and lack instance-level fidelity. While incorporating subgroup information offers a remedy, it faces a circular dependency: reliable subgroup identification requires complete data, while data completion is the imputation objective itself. To resolve this, we propose CAGI (Cluster-Aware Generative Imputation), a framework that reformulates clustering and imputation as a mutually reinforcing co-optimization process. CAGI employs a ``Partition-Guide-Restore’’ strategy where dynamic cluster assignments act as local priors to condition a Generative Adversarial Network. An iterative feedback loop is established to progressively refine both cluster structures and imputed values toward faithful subgroup distributions. To ensure distributional stability, CAGI further employs a multi-level optimization objective combining instance-level reconstruction with distribution-level regularization. Extensive experiments on 14 benchmark datasets with 15 representative baselines demonstrate the superiority of CAGI. The source code is available at: this https URL
[AI-45] MADB: A Large-Scale Music Aesthetics Dataset with Professional and Multi-Dimensional Annotations
链接: https://arxiv.org/abs/2607.06929
作者: Sirui Zhang,Tianle Wang,Xinyi Tong,Peiyang Yu,Jishang Chen,Liangke Zhao,Haoxin Zhang,Duo Xu,Xin Jin,Feng Yu,Songchun Zhu
类目: ound (cs.SD); Artificial Intelligence (cs.AI)
备注:
Abstract:Music aesthetic assessment is a challenging yet underexplored problem, requiring models to capture fine-grained, multi-dimensional human perceptual judgments. Progress in this area has been limited by the lack of large-scale datasets with structured aesthetic annotations. We introduce MADB, a large-scale dataset and benchmark comprising 9,999 tracks annotated by 30 trained annotators. Each track is rated by around 10 annotators across 10 perceptual dimensions and one overall score, with additional textual comments for multimodal analysis. We establish a unified evaluation framework over multiple pretrained models. Results reveal substantial gaps between model predictions and human judgments, exposing key limitations of current approaches. MADB provides a new benchmark for human-aligned music understanding. Project page: this https URL
[AI-46] Grounding Spatial Relations in a Compact World Model: Instruction Leakage and a Goal-Free Dynamics Fix
链接: https://arxiv.org/abs/2607.06925
作者: Yufeng Wang,Lu Wei,Haibin Ling
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Compact world models that condition on a language goal promise to ground relations such as ``put the red block left of the blue block’’ using a sparse set of explicit \emphreference anchors. We ask when such references actually ground a relation, and identify a trap: a goal-conditioned predictor reaches a striking 0.90 relation-readout accuracy, yet this is \emphinstruction transcription, not perception. Withholding the goal collapses it to chance ( 0.90!\to!0.27 , three seeds) and a counterfactual instruction makes the predicted anchors follow the \emphfalse instruction 94.5% of the time (true scene 2.3% ; N=256 ). Tested across three settings and a within-task ablation, our central claim characterizes the confound: \textbfinstruction leakage occurs when the scored quantity is transcribable from the instruction (when the instruction names the answer) and is essentially independent of how predictive the non-instruction inputs are. Our tabletop and the external BabyAI benchmark leak, whereas a Language-Table forward-dynamics world model whose instruction names \emphreferents does not, until the instruction is augmented to name the direction; and degrading the action never increases leakage, the opposite of what predictor-competition predicts. The diagnosis prescribes the fix: keep the goal out of the dynamics (it belongs to the planner’s cost) and supervise the \emphread path, recovering genuine, instruction-independent grounding ( 0.88 , identical with and without the goal). The detection protocol and remedy apply to any goal-conditioned world model whose instruction names the scored quantity.
[AI-47] he Harness Effect: How Orchestration Design Sets the Token Economics of Enterprise Agent ic AI
链接: https://arxiv.org/abs/2607.06906
作者: Muayad Sayed Ali,Aliaksandra Novik,Anji Boddupally,Artem Yavorskyi,Chris Nickerson,Daniel Rica,Emily DuGranrut,Felix Leung,Garrett Prince,Grace Barnett,Heath Robinson,Hosain Al Ahmad,Jesse Resnick,Juan Carlos Farah,Jyothi Swaroop Meruga,Leonid Kuznetsov,Luke Gorham,Marie Schmoll,Michael Paciullo,Saumya Das,Sharath Sheripally,Tommy Griscom,Mykyta Osadchyi,Neha Mantri,Nick Westrum,Olivia Benowitz,Parikshith Kulkarni,Radik Chernyshov,Rakshith Vasudev,Rohith Nadimpally,Vikas Gangadevi,Waseem AlShikh
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Agentic AI development today runs on token maxing: buying capability with tokens – longer reasoning traces, more turns, wider tool payloads, bigger replayed contexts – so tokens per task grow faster than task value. Falling per-token prices mask the pattern; total spend rises anyway. We argue the decisive lever against token maxing is the harness: the orchestration layer that assembles context, exposes tools, sequences turns, delegates work, and carries enterprise observability and governance. We isolate it with a controlled swap: 22 locked evaluation tasks, six foundation models (Claude Sonnet 4.6, Gemini 3.1, Gemini Flash 3.5, Qwen 3.6, GLM 5.1, Palmyra X6), changing only the orchestration layer – a frozen conventional production loop versus the Writer Agent Harness. Holding models constant, the harness cuts blended cost per task 41% ( 0.21- 0.12), median wall-clock 44% (48s-27s), and tokens per task 38% (14.2k-8.8k), with task-completion quality at parity (0.78-0.81, directional at this sample size). Efficiency is model-invariant – every model gets cheaper (33-61%) – while quality gains are capability-dependent: a model’s gain correlates almost perfectly with its baseline strength (r=0.99, n=6), a phenomenon we term harness leverage. Quality per dollar rises 82%; task-completions per million tokens rise from 54.9 to 92.0. On this workload the orchestration layer moved cost per task more than the full spread of the model menu did. We formalize token economics at the orchestration layer (including effective input price under prompt caching), detail the six mechanism families behind the effect – cache-shape discipline to failure-spend governance – compare six widely used agent systems on the same axes, and argue the harness is the one component whose efficiency multiplies across every model an organization runs – present and future.
[AI-48] Computing with Stochastic Oracles in AI-Augmented Computation
链接: https://arxiv.org/abs/2607.06893
作者: Jie Wang
类目: Computational Complexity (cs.CC); Artificial Intelligence (cs.AI); Information Theory (cs.IT); Machine Learning (cs.LG)
备注: 18 pages, 0 figures
Abstract:The Stochastic-Oracle Turing Machine (SOTM) framework models AI-augmented computation as the interaction of a probabilistic Turing machine with an oracle whose responses are drawn from context-dependent distributions. This paper studies what an SOTM can achieve under two oracle-response schemes: in a cached-response oracle, each distinct query receives one response that is reused on later calls to the same query, while in a fresh-response oracle, each call returns an independent response. In both schemes, the SOTM first computes from its input and internal random source to generate its first query, then proceeds adaptively, computing from its query-response transcript (the record of queries issued and responses received) to generate each subsequent query or produce a final output. Cached responses impose two transcript-based ceilings on achievable performance: a correct-identification ceiling governed by the total variation distance between the transcript distributions induced by the hidden states of the oracle, and an output quality ceiling equal to the expected score of the best output the SOTM can compute from the transcript. Fresh responses can raise these ceilings by allowing repeated calls to accumulate independent evidence toward correct or high-quality outputs. In the binary single-informative-query case, the error probability decreases exponentially in the number of calls to the same query at the Chernoff rate. For output quality, query-count bounds characterize threshold stopping when the score function is incorporated as part of the SOTM, and majority-based amplification bounds characterize the binary candidate-output model when it is not. Together, the results identify how response reuse, transcript information, and access to the score function determine what an SOTM can compute and at what token cost.
[AI-49] GemNav: Discrete-Token Visual Robot Navigation using a Multimodal Large Language Model
链接: https://arxiv.org/abs/2607.06882
作者: Peter Bohm,Saimunur Rahman,Abdelwahed Khamis,Sagun Man Singh Shrestha,Chris McCool,Peyman Moghadam
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Visual navigation policies built on large pretrained models have so far followed a common recipe: a dedicated visual encoder, a bespoke action head, and training on thousands of hours of cross-embodiment datasets. We ask whether this recipe is necessary. In this paper, we introduce GemNav, a visual robot navigation policy that adapts a frozen Multimodal Large Language Model (MLLM) for short-to-medium horizon waypoint navigation using Low-Rank Adaptation (LoRA) on the language tower alone, with no auxiliary visual encoder and no continuous regression head. Waypoints and categorical navigation signals share a single discrete token vocabulary generated by the language-model head, and a soft-decoded auxiliary loss recovers the metric structure that pure cross-entropy training discards. On a single 8.7-hour open corpus, roughly three orders of magnitude smaller than competing training sets, the policy transfers zero-shot to four physically distinct unseen environments and stops within 0.25-0.42m of the goal across 20 real-world trials covering an open carpark, an obstacle carpark, a long outdoor chemical yard, and an indoor warehouse. Conditioning on short image histories improves offline metrics but yields no robot benefit, pointing to a ceiling on what temporal context adds once pretrained vision features are in place. These results indicate that discrete-token adaptation of frozen MLLMs can provide a data-efficient, deployable alternative for foundation model robot navigation.
[AI-50] A Gold-Standard Study of What Makes a Lightweight Game-Playing Agent Strong
链接: https://arxiv.org/abs/2607.06854
作者: Nima Kelidari,Mohammadsaeed Haghi,Mahdi Salmani
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Science and Game Theory (cs.GT)
备注: 9 pages, 5 figures, 3 tables. Code and models: this https URL
Abstract:Reinforcement learning agents for imperfect-information card games are only as strong as the opponents they train against, and they are hard to grade, since they beat a random opponent over 99 percent of the time and only tie copies of themselves. So we build a strong, fixed, rule-based expert for Gin Rummy and use it only as a yardstick, never for training. It beats every agent we trained 70 to 99 percent of the time. Across more than a hundred runs, we isolate what makes a lightweight agent stronger. Trust region updates, a well-aimed reward, a curriculum of tougher opponents, warm starting, and keeping the best checkpoint all help, and stacking them lifts a self-play champion from about 30 to 36 percent against the expert. Several ideas did not pay off. Short-term and longer-term reward shaping, learned state embeddings, imitation and DAgger, and a live large language model opponent were each unhelpful, too slow, or too heavy to train at scale. Comparing MLP, convolutional, set-based, attention, and recurrent encoders shows that extra capacity does little to break the ceiling, suggesting the limit is information rather than network size. We add standard baselines (neural fictitious self-play and information set Monte Carlo search) and confirm the approach carries over to Leduc Hold’em, where the optimum is computable. The result is a lightweight, game-agnostic recipe that trains competitive agents without training on the expert, for any game a small model can handle, reported with robust statistics and released as a reusable package.
[AI-51] Evaluating SageMath-Augmented LLM Agents for Computational and Experimental Mathematics ICML2026
链接: https://arxiv.org/abs/2607.06820
作者: Pavel Snopov,German Magai
类目: Artificial Intelligence (cs.AI)
备注: 37 pages, 16 figures, accepted to 3rd AI for Math Workshop at ICML 2026
Abstract:Recent advances in AI for Mathematics have focused largely on autoformalization and theorem proving, leaving the role of Computer Algebra Systems (CAS) in agentic LLM workflows underexplored. We propose a ReAct-style agentic setup that combines LLM reasoning with verifiable feedback from SageMath, together with Context7 for the up-to-date documentation. We evaluate this agentic setup across frontier models for solving research-level mathematical problems from the RealMath benchmark in a setting that emulates a computational-mathematics research loop. We also propose a refinement to the RealMath benchmark by introducing a multi-step post-processing procedure and a multi-stage validation pipeline, both of which improve the quality and reliability of the extracted problem set. Our experiments reveal substantial performance gains from SageMath access across all evaluated models on +9.7~pp on average, the gains range from 1.5~pp to 27.8~pp and narrow the gap between open-weight and closed models. Qwen~3.7-Max benefits from SageMath the most, while GPT-5.5 achieves the highest solve rate of 75.2% and the lowest token usage among tool-enabled configurations. Our findings suggest that CAS-augmented agents represent a promising direction for assisting mathematicians in computational exploration, and we believe that this work is a step towards automated conjecture discovery. The project repository is available online.
[AI-52] When Agents Go Rogue: Activation-Based Detection of Malicious Behaviors in Multi-Agent Systems ICML2026
链接: https://arxiv.org/abs/2607.06807
作者: Haowen Xu,Xue Tan,Lei Ma,Zhihao Zhang,Chao Wang,Qingze Wang,Ping Chen,Jun Dai,Xiaoyan Sun
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: Accepted to ICML 2026
Abstract:While enabling effective collaboration on complex tasks, LLM-based Multi-Agent Systems (MAS) face critical security challenges due to vulnerabilities at the agent and interaction levels. Most existing MAS security defenses are built upon two core assumptions: semantically-explicit malicious attacks and explicit graph-based modeling of the MAS topology and agent-level interactions. In practice, real-world attacks are becoming more semantically stealthy, while MAS execution is typically asynchronous without the temporal alignment assumed by graph-based propagation models. To address these limitations, we propose AcMAS, an activation-based framework for malicious-behavior detection in MAS. By analyzing internal reasoning states in the activation space of local agents, AcMAS detects even stealthy attacks in a synchronization-robust fashion, without relying on explicit interaction graphs. Moreover, our activation analysis provides critical signals to guide AcMAS in restoring the functionality of compromised agents, rather than the disruptive agent isolation commonly used by the state-of-the-art methods. Comprehensive evaluation demonstrates that AcMAS significantly outperforms graph-based baselines against stealthy attacks, by +0.22 F1 in synchronous settings (0.94 vs. 0.72) and by +0.55 F1 in asynchronous settings (0.93 vs. 0.38), with generalization across diverse open-source LLM backbones, attack intensity, and MAS scale.
[AI-53] What Predicts Correctness in Text-to-SQL? A Selective-Prediction Study
链接: https://arxiv.org/abs/2607.06799
作者: Robert Richardson
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Evaluating uncertainty in AI-generated SQL queries requires estimating whether a query is correct, where correct means it executes to the same result as a human-written reference. We study which signals predict correctness on hard multi-table text-to-SQL, using AUROC to measure how well each ranks correct queries above incorrect ones. On BIRD and Spider, black-box signals such as string, structural, and execution self-consistency, a schema-relevance score, and query executability all fall between about 0.61 and 0.68 AUROC, with string self-consistency strongest at 0.675; white-box log-probability is similar (0.67). The signals that move past this ceiling are verification-based: an LLM judge scores from 0.72 (GPT-4o-mini) to 0.78 (Claude). Judges from different providers make different errors, so a two-provider ensemble reaches 0.82 AUROC with a well-calibrated probability (expected calibration error 0.03) and supports useful abstention frontiers (for example, answering 27% of questions at 24% selective risk) where self-consistency offers no valid low-risk subset. The pattern holds across two benchmarks, two generators, and two judge providers. We also ask whether a verifier can be trained. Fine-tuned verifiers, both encoder and generative, reach about 0.77 to 0.79 AUROC in-distribution but fall to about 0.66 on unseen schemas; scaling to 7B, adding schema diversity, distilling a strong judge’s rationales, and cross-benchmark training all fail to close that gap. Cross-schema transfer appears to track model scale and reasoning rather than fine-tuning. In practice, correctness uncertainty for text-to-SQL lives in reasoning-based signals: a fine-tuned verifier is a good in-domain tool, but a verifier that generalizes across schemas currently means a large frozen reasoning model.
[AI-54] Enhancing deep learning models for time series classification via knowledge distillation
链接: https://arxiv.org/abs/2607.06796
作者: Javidan Abdullayev,Maxime Devanne,Jonathan Weber,Germain Forestier
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Published version. Open access under CC BY 4.0. 24 pages, 11 figures
Abstract:Deep learning has achieved remarkable success in various domains including time series analysis, computer vision and natural language processing. However, high computational and memory demands of state-of-the-art architectures pose challenges for deployment in resource-limited environments. Knowledge Distillation (KD) addresses this by transferring knowledge from a large teacher model to a smaller, more efficient student model while maintaining competitive performance. In this work, we investigate the effectiveness of KD for Time Series Classification (TSC) across three architectures: the classical Fully Convolutional Network (FCN), the convolutional Inception model and the transformer-based ConvTran model. We evaluate our approach on UCR Archive, the largest benchmark repository of time series datasets, by modifying architectural components such as convolutional filters, Inception modules and attention heads across the three architectures. Our results consistently show that KD most effectively benefits student models of intermediate complexity across all three architectures, with the distilled FCN student reducing parameters by a factor of 38, the distilled Inception student achieving nearly the same performance as the teacher with 42% fewer parameters and the distilled ConvTran student with 2 attention heads showing the most significant improvement through distillation. To encourage further research and reproducibility, we provide our implementation at this https URL.
[AI-55] From Agent ic to Autogenic Network Management for AI-Native 6G and Beyond: A Standards Perspective
链接: https://arxiv.org/abs/2607.06786
作者: Petar Djukic,Sudipta Acharya,Takai Eddine Kennouche,Burak Kantarci
类目: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI)
备注: 9 pages, 5 figures, Accepted to IEEE Network
Abstract:Standards bodies, including TM Forum, 3GPP, and ETSI, are converging on Agentic AI as the foundation for next-generation network management, where Large AI Model (LAM)-based agents autonomously interpret intent, coordinate resources, and adapt operational behaviors at runtime. However, achieving this vision at the scale and complexity of 6G networks requires management systems that can generate and evolve their own automation software during operation. We introduce Autogenic network management, a reference architecture that extends agentic capabilities with self-programming, self reflection, self-orienting, and self-architecting capabilities. The architecture supports practical staged deployment beginning with human-supervised LAM-based agents and progressing toward autonomous operation as confidence builds. We demonstrate the approach through high-priority operator scenarios drawn from TM Forum’s autonomous network use cases, showing how autogenic management addresses real operational challenges. We conclude with a research roadmap outlining the technical advances needed to make autogenic network management realistic in future 6G networks.
[AI-56] AirPASS: Over-the-Air Federated Learning via Pinching Antenna Systems
链接: https://arxiv.org/abs/2607.06768
作者: Seyed Mohammad Azimi-Abarghouyi,Christopher G. Brinton
类目: Information Theory (cs.IT); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:This paper investigates over-the-air federated learning (AirFL) in wireless systems where the access point is equipped with a multi-waveguide pinching antenna system (PASS). We adopt the widely studied learning-oriented AirFL formulation, which seeks to maximize the number of selected devices while keeping the aggregation distortion below a prescribed threshold. The resulting joint optimization of device selection, receive beamforming, and pinching-antenna placement is highly nonconvex due to the intricate coupling among these system variables. To address this challenge, we develop AirPASS, an alternating optimization framework with two main components: a homotopy-Riemannian margin-consolidation method for device selection and receive beamforming under fixed PASS configuration, and a homotopy-assisted geometry optimization method for updating the pinching-antenna positions under fixed selected devices and beamformer. Experiments show that AirPASS consistently outperforms conventional co-located MIMO baselines, remains close to ideal FedAvg, and achieves an attractive performance-complexity tradeoff relative to SDR-DC and matching-pursuit scheduling alternatives.
[AI-57] Cost-Effective Agent Harnesses for Abstract Reasoning and Generalization on ARC-AGI-1
链接: https://arxiv.org/abs/2607.06764
作者: Kabir Moghe,Peter Chin
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Recent progress on ARC-AGI-1 from disclosed architectures has come broadly from two regimes: heavy test-time compute over frontier models (evolutionary search, exhaustive sampling, extended chain-of-thought), or benchmark-specific training in which small models are fine-tuned on ARC data, often with task-specialized architectures. We study a third regime: an open-weight model in non-thinking mode (DeepSeek V3.2) under a strict budget, with no ARC-specific fine-tuning. We study what is recoverable through architecture alone, building agentic harnesses that decompose pattern-discovery and program-synthesis stages explicitly. First, we introduce an Explorer-Definer Pipeline that separates pattern discovery from executable transformation synthesis, implemented as a two-stage agent pipeline. Next, we present the Reflective Orchestrator, which augments the pipeline with autonomous exploration of new transformations when previous hypotheses fail on training pairs. On the ARC-AGI-1 public 400-task evaluation set, the pipeline reaches 57.50% pass@2 at \ 0.25 per task, and the orchestrator reaches 67.25% pass@2 at \ 0.62 per task. Together these architectures lift a 15.50% one-shot baseline by ~52 points without benchmark-specific training or heavy test-time compute. Furthermore, the orchestrator-driven lift tests a falsifiable diagnostic the pipeline produces; unbiased pass@k analysis suggests the pipeline is generation-bound, not selection-bound (selection via training-pair accuracy captures ~95% of the candidate ceiling) and predicts that significant improvement requires broader generation, not better ranking. The orchestrator implements this prediction via adaptive re-exploration and confirms it (unbiased pass@1 lift +9.81 pp, matching selection-mediated pass@2 lift). An additional pipeline ablation identifies its think tool as a significant component, with removal reducing pass@2 by 5.75 pp.
[AI-58] QANTIS: Hardware-Calibrated Sequential POMDP Belief Updates on IBM Heron
链接: https://arxiv.org/abs/2607.06760
作者: Bayram Yuksel Eker,Suayb S. Arslan,Ozgur Nazli,Mustafa Serhat Demirgil,Furkan Deligoz
类目: Artificial Intelligence (cs.AI); Quantum Physics (quant-ph)
备注: 10 pages, 6 figures
Abstract:Autonomous systems under partial observability act on beliefs, not raw sensor events. QANTIS treats the quantum processor as a calibrated belief-update service in that loop: it receives a prior and an observation model, estimates the rare-event evidence term, and returns an ordinary posterior to a classical planner. This paper asks whether that service can be reused across a sequential Tiger POMDP horizon on present IBM Heron hardware without corrupting the planner-facing posterior. We answer with a controlled hardware case study rather than an end-to-end autonomy or wall-clock speedup claim. The study compares no amplification, guarded Grover amplification, and all-step fixed-point amplification on the same trajectory, then checks whether the returned posterior would change the downstream action. All-step FPAA preserves the Tiger posterior across the reported 8-step and 12-step primary runs, and the 20-step and 32-step controls remain inside the same operating band. In every reported decision check, the hardware posterior and the exact Bayes posterior select the same immediate action. Boundary-aware BIQAE stabilizes amplitude estimation near zero and near one, while a rare-event sweep maps the logical sample-complexity envelope for one-in-a-million evidence. The result is an operating envelope for a hardware-calibrated belief-update primitive, not a standalone hardware-advantage claim.
[AI-59] SmartHomeSecure: Automated Detection and Repair of Smart Home Configuration Errors Using Large Language Models
链接: https://arxiv.org/abs/2607.06748
作者: Yizhi Wang,Xinghua Gao,Reachsak Ly,Alireza Shojaei
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:Smart home automation platforms increasingly rely on user-authored YAML configuration files to define device behaviors, but these files are prone to syntax, formatting, and semantic logic errors that can cause automation failures and safety risks. Existing YAML validators, static analysis tools, and general-purpose large language models offer limited support for end-to-end diagnosis and repair because they lack domain-specific understanding and validated correction workflows. This paper presents SmartHomeSecure, a prototype for automated detection and repair of Home Assistant configuration errors using lightweight program analysis and constraint-guided large language model generation. SmartHomeSecure parses YAML files, detects syntactic and common semantic errors, normalizes error context, applies deterministic auto-fixes for routine defects, and constructs constrained prompts that guide LLMs toward minimal and structurally valid repairs. The system is implemented as a modular web application with four layers: UI Shell, Feature Orchestrator, Domain Engine, and Integration Layer. Its repair pipeline was evaluated on 100 real-world Home Assistant YAML files with manually injected errors across five categories: syntax/parsing, indentation, mapping, sequence, and scalar quoting errors. Four models were tested: gpt-oss-20b, gpt-oss-120b, llama-3.1-8b, and llama-3.3-70b. Results show that three models achieved 100% error detection accuracy, with repair success rates ranging from 87% to 93%. Manual verification found no hallucinated or incorrect repairs among successful outputs. These findings suggest that combining domain-aware program analysis with constrained generative AI is a feasible approach for improving the reliability and usability of smart home configuration repair.
[AI-60] A Continual Learning Framework for Adaptive Control of Modular Soft Robots
链接: https://arxiv.org/abs/2607.06740
作者: Nilay Kushawaha,Muhammad Sunny Nazeer,Baljinder Singh Bal,Cecilia Laschi,Egidio Falotico
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Soft robots have attracted significant attention in applications such as medical intervention, rehabilitation, and robotic manipulation due to their inherent compliance, flexibility, and high degrees of freedom. Modular soft robots (MSRs), composed of multiple interconnected segments, represent an emerging class of robotic systems with highly deformable and reconfigurable structures capable of performing complex tasks. However, designing controllers for MSRs remains challenging due to their nonlinear dynamics, modeling complexity, and hyper-redundant nature. Existing approaches typically require controllers to be retrained from scratch whenever the robot morphology changes. In this work, we address these challenges through a continual learning inspired control framework capable of incrementally adapting to changes in robot morphology while preserving previously acquired knowledge. Specifically, the proposed framework enables the controller to sequentially learn new MSR configurations without forgetting previously learned ones. In addition, for MSRs with fixed configurations, the same framework can be employed in a distributed manner to learn module-specific dynamics, enabling localized control and improved precision. The proposed approach is validated through closed-loop trajectory tracking experiments in simulation using a tendon-driven soft robot, as well as on a real-world three-module pneumatic soft robotic arm. Furthermore, we demonstrate the adaptive capabilities of the framework through a reaching experiment in which the controller selectively activates only the necessary modules to reach a virtual target position, thereby reducing computational overhead.
[AI-61] Reliable and Developer-Aligned Evaluation of Agents for Software Engineering
链接: https://arxiv.org/abs/2607.06713
作者: Razvan Mihai Popescu
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: International Conference on the Foundations of Software Engineering '26
Abstract:Large language models are rapidly moving towards closing the development cycle, transitioning from simple assistive companions to autonomous contributors deeply embedded into collaborative development environments. Despite their accelerated adoption, existing evaluation techniques are limited due to their fragmented nature and distorted projection of true model capabilities, often obtained from hypothetical syntactic scenarios. This research aims to bridge this gap by providing a comprehensive evaluation methodology for LLM-powered agents that is grounded in real-world software development practice. Our evaluation approach focuses on contamination-awareness, in-the-wild agentic behavior assessment, and trajectory-aware benchmarks and metrics capturing realistic coding contexts, human-aligned behavior, and model failure modes.
[AI-62] Vision Language Action (VLA) Models for Unmanned Aerial Robotics and Bimanual Manipulation: A Review
链接: https://arxiv.org/abs/2607.06706
作者: Inkyu Sa,Chanoh Park,Hea-Min Lee,Donghee Noh,Ho Seok Ahn
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 56 pages, 11 figures, 16 tables
Abstract:Vision Language Action (VLA) models unify visual perception, natural-language understanding, and action generation within a single foundation model, allowing a robot to follow instructions such as fold the towel or fly to the red building directly from camera images. Because VLAs inherit world knowledge from internet-scale pre-training, they have become the dominant framework for learning-based manipulation, with bimanual coordination serving as the most demanding testbed: two arms with 7 degrees of freedom each must move in concert to fold, assemble, and reorient objects. Unmanned aerial robotics faces a structurally similar challenge: a drone must coordinate thrust, attitude, and increasingly gripper commands from visual observations under strict latency and payload constraints. This review covers 183 contributions spanning 2017-2026 and organized along seven dimensions: VLA architectures, training recipes, action representations, bimanual coordination (2022-2026), unmanned aerial vehicle (UAV) navigation and control (2017-2026), language grounding, and cross-cutting concerns including memory and world models. We show that the coordination strategies, training recipes, and action representations developed for bimanual VLAs transfer to unmanned aerial systems and identify fourteen research directions across both domains.
[AI-63] Diffusion enabled Optimal Transport distances for graph matching
链接: https://arxiv.org/abs/2607.06646
作者: Iman Seyedi,Francesco Archetti
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:This paper introduces Diffusion Semi-Relaxed Fused Gromov-Wasserstein (DsrFGW), a novel method for graph comparison that unifies node features and structural connectivity through optimal transport. While traditional Gromov-Wasserstein and semi-relaxed variants (srGW, srFGW) capture graph structure, they often struggle with sparse, noisy, or partially observed graphs. Inspired by Graph Diffusion Distance, which posits graphs are similar if they enable similar information transmission patterns, DsrFGW incorporates diffusion processes allowing information propagation across nodes, capturing local and global structural patterns while reducing sensitivity to noise or missing edges. An extensive evaluation on 36 synthetic pairwise graph matching tasks (easy, medium, hard) demonstrates consistent superiority over srFGW, achieving accuracy improvements of 0-20 percentage points and dramatic Adjusted Rand Index (ARI) gains: in medium-difficulty scenarios, srFGW often achieves negative ARI (worse than random) while DsrFGW offers better performance in terms of both internal and external clustering quality measures (i.e., Adjusted Rank Index and Accuracy with respect to the true underlying clusters, respectively). Even under severe noise, DsrFGW improves clustering quality in 92% of the synthetic tasks with optimal diffusion scales adapting to problem difficulty, establishing DsrFGW as a robust framework for graph comparison under structural uncertainty.
[AI-64] he Rank-One Corner: How Much Value Equivalence Does a Task Need from a World Model?
链接: https://arxiv.org/abs/2607.06640
作者: Donna Vakalis
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 22 pages, 14 figures
Abstract:A learned world model is usually judged by how faithfully it reconstructs its observations or predicts reward, as though quality were something the model simply has or lacks. But what a task actually needs from a model is narrower: the few predictive coordinates its queries depend on, which we call the closure. We show that how much of that closure a latent comes to represent is set not by the model’s capacity or its observations but by the dimensionality of the objective it is trained against, and we measure this directly on a DreamerV3 stack in a controlled environment with known ground-truth closure. An aligned scalar value signal – the objective at the heart of value equivalence – installs only a one-dimensional projection of a closure that needs several dimensions: read through a single linear probe, the recoverable structure rises from R^2=0.10 to 0.76 as the scalar is replaced by the full objective. Sweeping the objective’s dimensionality from one to four installs exactly that many predictive directions through an auxiliary head, and the same staircase appears – at attenuated magnitude but the same rank – through the model’s own value head, so the dissociation is dimensional rather than an artifact of head form. Capacity-matched comparisons and in-situ pressure checks rule out the obvious alternatives. The law governs a regime, and we measure its boundary: on a companion closed-loop task whose structure is observable frame by frame, reconstruction installs that structure and the scalar objective suffices – the objective decides what a latent represents exactly where cheaper training signals cannot already recover it. Value equivalence is thus not all-or-nothing but dimensional: the familiar single-reward objective is its rank-one corner, and a model installs as much of a task’s structure as the objective it is asked to predict.
[AI-65] At-Grok Is Not Converged:A Measurement-Validity Audit for Grokking Representation Metrics
链接: https://arxiv.org/abs/2607.06639
作者: Truong Xuan Khanh
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 26 pages, 7 figs, 3 tables
Abstract:On modular arithmetic, a network’s embedding keeps compressing for tens of thousands of steps after it has already generalized. Reading effective rank at the grokking transition overstates the converged value by 3-5x on an MLP, and by 1.3-1.5x on a transformer trained to convergence; on the MLP it also erases which cells compress at all. Compression lags the accuracy transition by an amount on the order of the time-to-grok, at least 10,000 steps, rather than coinciding with it. A one-variable ablation shows what sets the lag size: adding LayerNorm to an otherwise identical transformer moves the fraction of compression done by the grok step from 0.87 to 0.25, and a pre-registered control rules out scale invariance as the mechanism. We package this as an audit that separates onset from compression, flags censoring, excludes boundary cells that never fully generalize, and checks that the reference floor has plateaued, with an adversarial suite that caught a false-confidence bug in our own branch. A secondary, MLP-specific depth law linking norm budget to converged floor fails a generality test on a transformer and flips sign under free weight decay. Code and the toolkit are released.
[AI-66] Specification Grounding Drives Test Effectiveness for LLM Code
链接: https://arxiv.org/abs/2607.06636
作者: Amin Haeri,Mahdi Ghelichi
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Large language models frequently generate code that appears correct on typical inputs yet fails on edge cases, invalid inputs, and other specification-defined corner conditions. A popular fix has the model write its own tests and repair until they pass, but the source of the gain is unclear: does it come from the tests merely existing, or from their grounding in a specification of what the code should do? We isolate this factor. Holding the tester, test budget, and repair loop fixed, we change a single prompt line that controls whether the tester receives the spec as a checklist of rules. The baseline is strong: it is already told to probe invalid inputs and edge cases. Grounding the tests in the spec produces correct code +38 percentage points more often than this baseline across three Claude tiers (Haiku 4.5, Sonnet 4.6, Opus 4.8), and +36 points on a held-out set. Grounding, not test quantity, is the primary driver: doubling the test budget barely helps, and combining eight independent ungrounded suites plateaus far below grounding. An ablation isolates the spec’s content, not its format: given the spec as a plain paragraph the tester recovers 27 of 30 bugs, but asked to plan tests without the spec it recovers only 2 of 30. The effect survives stronger baselines: a property-based generator catches 28 of 30 bugs but invents out-of-spec requirements, and an AlphaCodium-style loop only matches the baseline. It replicates across vendors (GPT-5.3-codex +28, Gemini 3.5 Flash +19), with a task-level sign test over 18 tasks significant at p=0.002. Grounding improves both sensitivity and precision: it catches more real bugs and wrongly rejects far less correct code, cutting the false-alarm rate from 33% (68% against a Python standard-library oracle) to 0%. On well-specified algorithmic problems it neither helps nor hurts.
[AI-67] Cross-Trajectory Chimera Interventions Reveal Dissociable Roles of Weight Magnitude and Direction in Grokking
链接: https://arxiv.org/abs/2607.06628
作者: Truong Xuan Khanh
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 12 pages, 8 figures, 1 table
Abstract:Which properties of a partially trained network are causally portable to a different, independently trained network? Single-trajectory interventions show necessity within one run, not portability across runs. We introduce cross-trajectory chimera interventions: given two runs from different seeds, we split each weight vector into a norm and a unit direction, recombine one run’s norm with the other’s direction, and continue training. On two modular-arithmetic tasks that grok, the components dissociate. Direction carries a transferable, donor-specific circuit identity: implanting a donor’s direction at the recipient’s norm drives the run to the donor’s circuit in 40/40 cases, while an angle-matched random control yields no shift. The transfer is threshold-like, and its location is predicted by the recipient’s norm, separating perfectly by norm class over all 20 pairs (joint permutation probability 1.9e-4). Norm carries only a modest, distributed delay effect and no identity signal. An adaptive bisection procedure localizes the threshold to +/-1/64. Direction indexes which solution a trajectory approaches; norm governs how susceptible that identity is to being overwritten.
[AI-68] Open-Ended Scenario Reasoning for Specialist Model Adaptation
链接: https://arxiv.org/abs/2607.06625
作者: Youcheng Zong,Runda Jia,Ranmeng Lin,Mingxuan Ren,Dakuo He
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Systems and Control (eess.SY)
备注:
Abstract:Process industries have accumulated validated specialist models, yet sensor drift, feedstock variation, and regime switching cause these models to degrade systematically in new scenarios. Collecting new labeled data and retraining is costly, while continuing with the original model incurs persistent bias. Existing adaptation methods require modifying model parameters with sufficient labeled data, making rapid response on deployed systems difficult. Using LLMs as direct predictors risks hallucinations and uncontrollable outputs. Such predictors also cannot incorporate unstructured scenario knowledge from the field. To address these limitations, this article proposes Reasoning-Driven Open Adaptation for Specialist Models (ROAM), a framework that uses LLM world knowledge and reasoning to adapt frozen specialist models to unseen scenarios without retraining. ROAM confines all corrections to a low-dimensional, semantically interpretable latent space. LLM-generated scenario judgments and online observations are fused under a unified probabilistic framework. A risk-constrained mechanism suppresses corrections under unreliable LLM evidence or abrupt scenario shifts and falls back to the original frozen model when evidence is insufficient. Experiments on a mineral thickening process and the public IndPenSim penicillin fermentation dataset show that ROAM reduces MAE by over 20% in major shift settings such as hidden shifts with only 839 additional parameters and under 0.02,ms per-step overhead. These results indicate that LLM reasoning can be turned into a conservative adaptation signal for industrial models already in service.
[AI-69] Agent Lens: Production-Assessed Trajectory Reviews for Coding Agent Evaluation
链接: https://arxiv.org/abs/2607.06624
作者: Andrey Podivilov,Vadim Lomshakov,Sergey Savin,Matvei Startsev,Roman Pozharskiy,Maksim Parshin,Sergey Nikolenko
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Software Engineering (cs.SE)
备注:
Abstract:We present AgentLens, a production-assessed benchmark for interactive code agents. Most code-agent benchmarks reduce a run to a single bit – did the task pass? – but the people who actually use these agents experience the entire trajectory: how the agent follows instructions, uses its tools, verifies its own work, recovers from mistakes, and talks to them along the way. AgentLens evaluates that whole trajectory. It pairs formal verification, where an objective check exists, with LLM-written trajectory reviews and side-by-side comparisons, so that each run yields a readable explanation of why the score is what it is. This makes AgentLens useful for more than ranking models: we use it to diagnose model behavior, compare successive versions of our own agent, and catch product regressions in a nightly evaluation pipeline. We release the benchmark as open source at this https URL.
[AI-70] LLM -Guided Task-Semantic Field Factorization for Industrial Process Forecasting
链接: https://arxiv.org/abs/2607.06623
作者: Youcheng Zong,Runda Jia,Mingxuan Ren,Dakuo He
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Systems and Control (eess.SY)
备注:
Abstract:Process industries rely on time-series forecasting and soft sensing to estimate quality variables that are hard to measure online. Labeled data are scarce, operating regimes change frequently, and retraining models or rebuilding alignment pipelines for each scenario is costly. Such settings often provide variable tables and process documents that record variable names, units, physical meanings, and process roles. However, standard time-series backbones usually treat inputs as anonymous numerical columns. Existing text-enhanced methods also rarely make the semantic-logical relations between input variables and the prediction target available to the model within each numerical window. To address this problem, this article proposes Task-Semantic Field Factorization (TSF), a large language model (LLM)-guided framework. TSF builds a task-semantic field from task protocols and variable documents before training and uses the LLM only for offline semantic construction. Online training and inference remain with conventional time-series backbones. During training and inference, the current numerical window activates variable semantics, so semantic information participates in each prediction and supports adaptation to different prediction targets and operating shifts. On multiple complex industrial forecasting and soft-sensing tasks, TSF reduces MAE by 6.4% on average in improved settings, with the largest reduction reaching 25.5%. It adds only about 1.8–3.0k parameters, with less than 0.008 ms/step of additional online inference overhead. These results show that TSF turns existing process documents into measurable forecasting gains across backbones and semantic generators while remaining lightweight for deployment.
[AI-71] Inertia-1: An Open Exploration of Wearable Motion Foundation Models
链接: https://arxiv.org/abs/2607.06617
作者: Zongzhe Xu,Aakarsh Anand,Sarah Jiang,Chuntung Zhuang,Zitao Shuai,Sriram Sankararaman,Yuzhe Yang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Wearable motion sensing provides a continuous and scalable window into human behavior and health, making it a natural fit for foundation models, yet its pretraining and scaling principles remain poorly understood. Prior work studies isolated design choices, such as sensor placement or sampling frequency, often under fixed settings and narrow downstream tasks that fail to capture real-world sensing diversity. We introduce Inertia-1, a fully open exploration of wearable motion foundation models. Using massive corpora of accelerometer data from global sources spanning more than 18.2M hours, we build a controlled framework for studying the full lifecycle of wearable motion foundation models, covering data choices such as sensor modality, device placement, sampling rate, window length; model choices such as architectures and model size; and training choices such as pretraining objective and data scale. Extensive evaluations across 15 datasets spanning human activity recognition, freezing-of-gait detection, and disease prediction reveal intriguing findings for building motion foundation models that generalize across tasks and sensing conditions. Collectively, Inertia-1 not only presents state-of-the-art recipes for diverse downstream tasks, but also serves as a comprehensive, practical, and open cookbook for wearable motion representation learning.
[AI-72] STAGformer: A Spatio-temporal Agent Graph Transformer for Micro Mobility Demand Forecasting
链接: https://arxiv.org/abs/2607.06614
作者: Ye Zihao
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Accurate station-level demand forecasting is essential for the efficient operation of bike-sharing systems, yet it remains challenging due to complex spatio-temporal dependencies and the large scale of urban networks. This paper presents STAGformer, a Spatio-Temporal Agent Graph Transformer that achieves efficient global modeling with linear computational complexity. The model introduces a two-step agent attention mechanism, where a small set of learnable spatial and temporal agent tokens first aggregate global information and then broadcast it back to individual stations and time steps, effectively capturing long-range interactions while reducing the quadratic cost of standard self-attention to O(NT). STAGformer integrates four core modules: a spatio-temporal encoder that fuses dynamic node features with external contextual factors (weather, time, points of interest), a graph propagation module for spatial neighbor aggregation, a temporal convolution module for local pattern extraction, and the agent attention module for global dependency modeling. Extensive experiments on two real-world datasets – NYC Citi-Bike and Chicago Divvy-Bike – demonstrate that STAGformer consistently outperforms state-of-the-art baselines across multiple prediction horizons, achieving significant improvements in both RMSE and MAE. Ablation studies validate the contribution of each component, with the agent attention mechanism proving critical for modeling global spatio-temporal dependencies.
[AI-73] PRoVeFL: Private Robust and Verifiable Aggregation in Federated Learning
链接: https://arxiv.org/abs/2607.06612
作者: Harsh Kasyap,Anil Kumar Pradhan,Ugur Ilker Atmaca,Graham Cormode,Carsten Maple
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: 18 pages
Abstract:Federated Learning (FL) enables multiple clients to collaboratively train machine learning models while retaining data locality, thereby enhancing user privacy. However, traditional FL frameworks rely on a centralized aggregation server and assume honest-but-curious clients, making them susceptible to both server-side inference and client-side poisoning attacks. Although recent work has explored secure and Byzantine-resilient FL protocols, they face a fundamental trade-off among privacy, integrity, and verifiability, and incur substantial computational and communication overhead due to the heavy use of cryptographic primitives. In this work, we propose PRoVeFL-a novel, modular FL framework that is Privacy-preserving, Byzantine-Robust, and ensures Verifiable aggregation. PRoVeFL employs multiple servers leveraging multi-key fully homomorphic encryption. Each client encrypts its local model updates and distributes encrypted shares to all servers. This design enables a hybrid computation model in which ciphertext operations are carefully offloaded to the plaintext domain under strict privacy constraints to efficiently evaluate complex statistical aggregation rules. PRoVeFL is compatible with a wide range of state-of-the-art Byzantine-robust aggregation algorithms (e.g., Krum, Trimmed Mean, FLTrust, norm clipping, MESAS, and more) and further enhances them with verifiability mechanisms that require minimal trust in at least one honest server. We evaluate it across different settings and demonstrate its scalability with varying numbers of parameters and participants. PRoVeFL improves runtime over the prior works, Prio and ELSA, based on distributed trust with comparable security guarantees, up to 100x and 10x, respectively.
[AI-74] Deep Reinforcement Learning for Reliability Based Bi-Objective Portfolio Optimization
链接: https://arxiv.org/abs/2607.06610
作者: Sounaq Das,Tanmay Sen,Raghu Nandan Sengupta,Aditya Gupta
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Optimization and Control (math.OC)
备注:
Abstract:Portfolio optimization under uncertainty is inherently a multi-objective decision problem involving complex interactions among return, risk, market dynamics, and practical investment constraints. Existing reliability based portfolio optimization approaches primarily rely on static optimization frameworks and often fail to capture sequential decision making, tail risk, and market frictions such as transaction costs. To address these limitations, we propose a deep reinforcement learning framework for multi-objective reliability based portfolio optimization (MORP-DRL). The proposed framework jointly optimizes expected return and downside risk using three complementary risk measures: variance, Conditional Value-at-Risk (CVaR), and Entropic Value-at-Risk (EVaR). To model uncertainty and heavy-tailed market behavior, asset returns are represented using GARCH(1,1), Extreme Value Theory, and a t-copula dependence structure, while realistic scenarios are generated through quasi-Monte Carlo simulation. A Proximal Policy Optimization (PPO) based strategy is developed under practical constraints including transaction costs and portfolio bounds, and is benchmarked against NSGA-II. Experiments on ten global equity indices across pre-COVID, COVID, and post-COVID market regimes demonstrate that MORP-DRL achieves competitive risk-return performance, reduced downside risk during periods of market stress, and scalability to high-dimensional portfolio settings.
[AI-75] D2PO: Optimizing Diffusion Samplers via Dynamic Preference ECCV2026
链接: https://arxiv.org/abs/2607.06609
作者: Jinkyu Kim,Jinyoung Choi,Bohyung Han
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Accepted to ECCV 2026
Abstract:We propose D2PO (Dynamic Direct Preference Optimization), a principled framework for optimizing diffusion sampling policies with respect to timestep schedules and classifier-free guidance (CFG) weights. Our work is motivated by a fundamental limitation of existing student-teacher regression frameworks; low-NFE student samplers are trained to mimic high-NFEteachers, often sacrificing high-frequency texture fidelity while preserving coarse global structures, thereby misaligning the sampler with perceptual quality. D2PO addresses this challenge by reformulating sampler optimization as a preference-based alignment problem, leveraging the Direct Preference Optimization (DPO) framework. To make DPO applicable to diffusion samplers, we model the sampling policy as an energy-based model (EBM), transforming preference comparisons into tractable energy differences. We further introduce a novel energy formulation derived directly from the pretrained score network, enabling preference evaluation in perturbed spaces that jointly capture structural consistency and fine-grained details. Moreover, we introduce dynamic preferences, where the preferred samples used for alignment progressively improve as the sampling policies are learned. This self-improving mechanism replaces rigid static teacher supervision with an iterative, preference-guided refinement process, providing progressively stronger alignment signals. Extensive experiments demonstrate that D2PO aligns diffusion samplers with perceptual quality more faithfully, unlocking the full potential of high-quality teachers and consistently outperforming conventional regression-based schedulers under low-NFE constraints.
[AI-76] NEST: Tackling Dataset-Level Distribution Shifts via Regime-Oriented Mixture-of-Experts
链接: https://arxiv.org/abs/2607.06607
作者: Lanhao Li,Bingshu Xie,Lijun Sun,Xin Xue,Haoyi Zhou,Jianxin Li
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Accurate long-term forecasting in complex systems is frequently compromised by dataset-level distribution shifts, where diverse underlying behavioral modes and evolving system states drive the dynamic multivariate time-series. While existing methods predominantly focus on local temporal shifts, they fail to explicitly model the global structural challenge where datasets are composites of distinct operational regimes. In this paper, we propose NEST, a specialized framework designed to model and recompose these evolving structures through a two-phase dense MoE architecture. NEST first facilitates structural specialization by partitioning the dataset into distinct operational regimes through unsupervised clustering in a principled moment-entropy space. We introduce a regime-oriented router mechanism that generates initial expert weights based on temporal content, subsequently refined through geometric modulation to regime centroids. Crucially, rather than acting as monolithic predictors, individual experts function as specialized kernels that capture regime-specific dynamics by evolving unique variate-attention patterns. Extensive evaluations on diverse benchmarks, including heterogeneous network traffic and physical phenomena, demonstrate that NEST consistently achieves state-of-the-art performance. Our code and datasets are available at this https URL
[AI-77] riRoute: Unified Learned Routing for Joint Adaptive Attention Experts and KV-Cache Allocation
链接: https://arxiv.org/abs/2607.06601
作者: Andrii Balashov,Olena Ponomarova
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 22 pages, 5 figures, 6 tables; preprint
Abstract:Conditional computation can decouple language model quality from per-token inference cost, yet leading techniques act on a single axis in isolation: Mixture-of-Experts (MoE) sparsifies the FFN, Mixture-of-Depths (MoD) skips whole transformer blocks, and KV-cache quantization compresses attention memory. We argue these three decisions (attention resolution, expert selection, and cache bit-width) are strongly coupled and should be made jointly: a token rare enough to warrant full attention may also need high-precision caching regardless of which expert processes it. We introduce TriRoute, a single lightweight controller shared across all three axes that, for every token at every layer, emits a coordinated policy: (i) an attention mode (skip/local/full), (ii) a sparse set of FFN experts (with a null expert recovering MoD), and (iii) a KV-cache bit-width. The controller trains end-to-end via a heterogeneous relaxation (Gumbel-Softmax with straight-through estimation for categorical decisions and load-balanced top-k gating for experts) under a Lagrangian budget constraint that turns the average compute and memory cost into a controllable knob. We identify a cross-axis routing-collapse cascade in naive joint training, where collapse on one axis propagates to the others, and address it with per-axis normalization and a coupling-aware balancing loss. On decoder-only models from 160M to 1.3B parameters at compute-optimal token counts, TriRoute Pareto-dominates the best independent MoD+MoE+KV-quantization combination at matched inference FLOPs and memory, while better preserving tail-case robustness on rare entities, code, and arithmetic that pure perplexity optimization erodes. Post-hoc analysis reveals interpretable structure: the controller allocates full attention and high-precision cache to sentence-initial positions, rare subwords, and named entities, while cheaply routing function words.
[AI-78] When Agents Remember Too Much: Memory Poisoning Attacks on Large Language Model Agents
链接: https://arxiv.org/abs/2607.06595
作者: George Torres,Sharad Shrestha,Satyajayant Misra
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:Personal AI agents powered by large language models can reason and act using available tools to access emails, manage calendars, and push code to remote repositories, all with minimal oversight. When augmented with long-term memory, an agent can recall specific details relevant to the current task, reducing the need for large context windows. Currently, long-term memory agents tend to fall into two distinct domains: conversational and action-planning agents. Personal assistant agents sit at the convergence of these two domains and handle sensitive information while interacting with untrusted information sources, creating previously unaccounted security vulnerabilities. In this work, we introduce the novel attack vector, GhostWriter, which exploits current memory subsystems in tool-using personal agents to poison their memory store. GhostWriter operates in two phases: injection, where an adversary sends a hidden attack payload to the target agent; and activation, in which the poisoned memory is retrieved. We show that GhostWriter achieves near-universal injection rates of approximately 98% and a high average activation rate of approximately 60% against state-of-the-art agents. This attack is possible due to the lack of security-focused memory governance. In response, we propose Agentic Memory Sentry (AM-Sentry), which leverages two mitigation techniques: a memory-saving policy and a memory-retrieval screen. Our experiments show that AM-Sentry dramatically reduces GhostWriter’s success rate while preserving agent utility.
[AI-79] QCNN with Rough Path Signature Kernels
链接: https://arxiv.org/abs/2607.07634
作者: Leonardo Nogueira Falabella,Vasily Sazonov
类目: Quantum Physics (quant-ph); Artificial Intelligence (cs.AI)
备注:
Abstract:Time series analysis plays a vital role across a wide range of scientific and engineering domains but poses substantial computational challenges. A major difficulty arises from the time reparameterization invariance of time series data, which complicates the extraction of meaningful temporal features. In this work, we address the problem of time series classification by exploring the application of quantum computation techniques. We propose a hybrid quantum-classical architecture that integrates recent advances in quantum neural networks with the mathematical framework of path signatures, mitigating the impact of time reparametrization invariance. The architecture employs feature layers that compute a signature kernel between pairs of input paths, consisting of a reference path and a target path for classification, using either classical or quantum variational linear solvers (VQLS). These feature layers are followed by a Quantum Convolutional Neural Network (QCNN) to perform downstream learning tasks. We evaluate several realizations of the proposed architecture, differing in QCNN configurations, on a binary classification task involving time series representations of handwritten digits. Our experiments demonstrate the potential advantages of implementing path signature kernel layers within quantum circuits and provide an analysis of the computational limitations associated with the VQLS component.
[AI-80] Stability of Flow Models for Graph Signals
链接: https://arxiv.org/abs/2607.07510
作者: Martin Schmidt,Gonzalo Mateos
类目: ignal Processing (eess.SP); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Submitted to the IEEE Transactions on Signal Processing
Abstract:Generating signals on graphs requires permutation-equivariant models that exhibit stability with respect to relative structural perturbations. While favorable stability properties of Graph Neural Networks (GNNs) have been well documented, it is unclear how structural errors propagate through the dynamics of continuous generative flow models that are gaining traction for graph signal generation. In this paper, we analyze continuous normalized flow models parameterized by GNNs and show that permutation equivariance is preserved for both the resulting continuous-time ordinary differential equations and their discrete numerical approximations used as graph signal samplers. Our primary contribution is to derive explicit stability bounds on the generated probability distributions, which quantify how relative graph perturbations affect the final sampled signals. Motivated by these theoretical bounds, we introduce a stability-promoting regularized flow matching strategy that actively penalizes the spatial Lipschitz constant of the vector field during model training. Experiments using synthetic smooth signals on stochastic block model graphs and real-world fMRI signals on brain connectomes demonstrate that this bound-oriented approach yields generative models that are more robust to structural noise, without sacrificing output quality.
[AI-81] Quantum simulation of real-world nonlinear dynamics via Koopman method
链接: https://arxiv.org/abs/2607.07338
作者: Baoyang Zhang,Dong An,Zhaoyuan Meng,Yefei Yu,Xiaoxiao Xiao,Zhen Lu,Yue Yang
类目: Quantum Physics (quant-ph); Artificial Intelligence (cs.AI); Fluid Dynamics (physics.flu-dyn)
备注:
Abstract:Nonlinear dynamics is ubiquitous in nature, ranging from chemical pattern formation to ocean circulation, yet its simulation on quantum computers is fundamentally limited by the unitary nature of quantum evolution. We propose the quantum Koopman method, a data-driven framework that embeds nonlinear dynamics into a learned linear representation and implements the resulting evolution using shallow quantum circuits. This method learns Koopman observables from trajectory data, projects the lifted dynamics onto a finite-dimensional subspace, and decomposes the corresponding non-unitary propagator into parallel spectral channels. We utilize the Koopman method on a superconducting processor to simulate three distinct nonlinear systems, comprising reaction-diffusion dynamics, fluid motion on a sphere, and satellite-derived observations of Gulf Stream currents, employing up to 32 parallel circuits of 10 qubits. These quantum simulations capture the dominant multiscale patterns and statistical signatures of the underlying dynamics, and reveal a transition from performance limited by hardware noise in weakly nonlinear systems to performance limited by finite-dimensional Koopman representations as nonlinear scale interactions increase. This transition identifies a practical boundary for quantum-amenable nonlinear dynamics, establishing a hardware-validated route for simulating moderately nonlinear dynamics on near-term quantum hardware.
[AI-82] POO-LPSP: Parallel Osprey Optimized Least Penalty-Squared Prioritization Methods for Priority Derivation in the Analytic Hierarchy Process
链接: https://arxiv.org/abs/2607.07313
作者: Kevin Kam Fung Yuen
类目: Optimization and Control (math.OC); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC); Numerical Analysis (math.NA)
备注: 16 pages, 1 Figure, 4 Tables
Abstract:Pairwise comparison (PC) via pairwise reciprocal matrices (PRMs) is central to the Analytic Hierarchy Process (AHP). Although the traditional eigenvector method is widely applied to derive priorities, its theoretical robustness in reflecting true priority vectors remains debated. Building upon a previous iteration of this study, this research develops the revised Least Penalty-Squared Prioritization (LPSP) optimization models, including the revised Least Product of Penalty and Direct Squares (LPPDS) and revised Weighted Squares (LPPWS), to minimize the revised Root Mean Penalty-Squared Variance (RMPSV) and the revised Root Mean Penalty-Weighted Square Variance (RMPSWV). However, solving these non-linear formulations is computationally complex for decision-makers. To overcome these limitations, this study proposes the Parallel Osprey Optimized Least Penalty-Squared Prioritization (POO-LPSP) method. By integrating an improved bio-inspired metaheuristic Parallel Osprey Optimization Algorithm (POOA), this framework efficiently solves complex LPSP models to minimize RMPSV and RMPSWV, thereby enhancing prioritization reliability. The practical utility and computational efficiency of the POO-LPSP method are validated through a numerical application focusing on a Generative AI (GAI) vendor selection problem. To extend, POO-LPSP can serve as a robust alternative to Saaty’s Eigen system method for AHP applications.
[AI-83] Bayesian Optimization of Genetic Algorithm Hyperparameters in a Multi-Fidelity Framework for Efficient Lattice Material Design
链接: https://arxiv.org/abs/2607.07289
作者: Sergei Zorkaltsev,Maciej Haranczyk,Christina Schenk
类目: Materials Science (cond-mat.mtrl-sci); Disordered Systems and Neural Networks (cond-mat.dis-nn); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Optimization and Control (math.OC)
备注: 20 pages, 5 figures, 2 tables
Abstract:This study presents a multi-fidelity framework for the systematic optimization of genetic algorithm (GA) hyperparameters. The framework integrates three fidelity levels: high-fidelity Fast Fourier Transform (FFT) homogenization for validation, a medium-fidelity 3D convolutional neural network surrogate for rapid property evaluation, and a low-fidelity Gaussian process (GP) surrogate within a Bayesian optimization (BO) framework to guide the hyperparameter search. Various acquisition functions are evaluated, with logNEI achieving the best performance by effectively accounting for the noise inherent in GA evaluations. The proposed framework identifies hyperparameter configurations that enable a 25-generation GA run to achieve elastic modulus values comparable to those obtained in a full 75-generation optimization. Furthermore, introducing a penalized BO objective significantly reduces the number of required lattices with only minor decreases in absolute achieved elastic modulus, revealing a practical trade-off between performance and the number of structures that must be evaluated. High-fidelity FFT validation verifies the effectiveness of the surrogate-driven optimization strategy. The optimized hyperparameters allow for rapid convergence, eliminate the need for lattice mutation, and reduce the overall computational cost by 24% (from 225 to 171 hours) while preserving mechanical performance. These results demonstrate the potential of multi-fidelity optimization as an efficient and practical approach for GA hyperparameter tuning and future experimental lattice design studies.
[AI-84] DiPhon: Diffusion on Graphons for Scalable Graph Generation
链接: https://arxiv.org/abs/2607.07232
作者: Sergio Rozada,Yiming Qin,Manuel Madeira,Pascal Frossard,Alejandro Ribeiro
类目: Machine Learning (stat.ML); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Diffusion models represent a leading paradigm for graph generation, with notable impact in domains such as molecular design. Yet, scaling these models to large graphs remains an open problem. We approach this question in the dense-graph setting through the lens of graphons, the size-agnostic limit objects of dense graph sequences, to study how structural graph statistics behave across node-size scales. This perspective leads to DiPhon, a diffusion framework for size-scalable graph generation. Specifically, we formulate a continuous diffusion process on the graphon space via a Jacobi stochastic differential equation (SDE), and propose DiPhon, a discretized graph-level process that mimics these dynamics on finite graphs. We further derive the corresponding reverse-time process, which requires access to the marginal score. For the Jacobi process, this score interestingly admits a tractable form, which we estimate from data via graph denoising and plug into the reverse process to generate graph samples. We prove that DiPhon matches exactly the first moment of the marginal distributions induced by the continuous graphon process, and approximates the second moment up to a closed-form discrepancy. Thus, DiPhon inherits key size-agnostic statistical properties of the graphon dynamics, providing a principled route toward scalable graph generation. Empirically, we demonstrate this scalability by training on small graphs and generating progressively larger graphs at inference time, without retraining, while preserving their core topological properties.
[AI-85] Memory Scarcity Open Models and the Restructuring of the AI Industry 2026-2030 – A quantitative scenario analysis of inference economics training-cost divergence and infrastructure solvency
链接: https://arxiv.org/abs/2607.07207
作者: Satoshi Matsuoka
类目: General Economics (econ.GN); Artificial Intelligence (cs.AI); Hardware Architecture (cs.AR); Computational Engineering, Finance, and Science (cs.CE); Performance (cs.PF)
备注: 21 pages
Abstract:We analyze how four forces restructure the AI industry over 2026-2030: the DRAM/HBM price surge, frontier-capable open-weight models (GLM-5.2), rapid inference-efficiency gains (near-Shannon-limit KV-cache compression, lightweight local runtimes), and the entry of Meta and xAI into compute resale on fleets bought before the memory repricing. Formulating inference economics in dollars per petabyte of bandwidth delivered (\ /PB) – model-agnostic for bandwidth-bound decode – we show the entrant-incumbent cost gap never closes: a depreciation conveyor delivers newly amortized fleets to incumbents faster than hardware prices normalize (3.2x in 2026, 1.9x in 2027, re-widening to 3-4x by 2029-30). Training bifurcates into a luxury tier (\ 18-38B per frontier run by 2030) and a mass tier (previous-frontier parity via RL/distillation falling toward \ 5M). Solvency of the announced buildout is confined to a corridor requiring roughly 2x annual token-demand growth for four years with sticky premium pricing; a measurement critique shows public token trackers overstate monetizable demand, and all pre-Q2-2026 projections predate the industry’s shift from token maximization to token minimization. A vintage-breakeven analysis finds 2026 and 2028-29 capacity each fatally exposed to one pricing regime, with only the 2027 vintage robust. A greenfield custom-silicon entrant removes the merchant margin but not the memory premium (central outcome: 25% success/34% mediocre/41% loss, improvable via staged go/no-go gates). China’s LineShine LX2 – domestic HBM on a standard ISA – decouples its cost curve from the memory crisis. Scenario probabilities: Rotating Landlord Oligopoly 25%, Commoditization Crash 25%, Jevons Absorption 20%, System-Layer Re-differentiation 18%, Geopolitical Bifurcation 12%. Solvency now depends on monetized bandwidth demand, premium stickiness, and vintage ownership.
[AI-86] A Multi-Analyst LLM Pipeline for Auditable Rule Discovery Across 68 Public Physiological Corpora
链接: https://arxiv.org/abs/2607.06802
作者: Dovy Paukstys
类目: ignal Processing (eess.SP); Artificial Intelligence (cs.AI)
备注: 8 pages, 2 figures, 9 tables; submitted to IEEE SPMB 2026
Abstract:Open physiological corpora are heterogeneous: they use different sensors, labels, sampling rates, recording settings, and clinical endpoints. They can support detector design, but they do not directly specify which detector rules should be built for a new contactless monitoring platform. We report a controlled four-analyst large-language-model (LLM) workflow for converting 68 public physiological corpora, screened for commercial-use compatibility, into an auditable library of candidate rule shapes for prospective validation. Four independent commercial LLM families read the corpus documentation under a controlled prompt and produced 695 candidate rule markers (top-markers). Deduplication retained 649 rule records; a threshold-bounds audit then flagged 51 sanity violations for clamping or curator review. Cross-corpus consolidation produced 436 unique rule shapes. Gate-tagging against two hard invariants, native target-hardware channel availability and no multi-night per-patient personalization, identified 94 build-now detector components across four detector-family buckets. The pipeline does not produce a validated clinical detector. It produces an auditable engineering cascade in which analyst disagreement, threshold checks, curator review, and automated continuous-integration (CI) checks route literature-derived rules toward prospective hardware validation.
[AI-87] sbootstrap: Distribution-Free Uncertainty Quantification and Conformal Prediction for Time Series
链接: https://arxiv.org/abs/2607.06690
作者: Sankalp Gilda
类目: Methodology (stat.ME); Artificial Intelligence (cs.AI); Mathematical Software (cs.MS); Statistical Finance (q-fin.ST); Applications (stat.AP)
备注: 4 + 2 pages. Code: this https URL
Abstract:Finance, sensing, and demand streams violate the exchangeability that IID conformal prediction and the IID bootstrap assume, and existing libraries implement either a general resampling engine or conformal calibration without the other. tsbootstrap provides block, residual, sieve, and wild resampling, classical bootstrap confidence intervals, and adaptive conformal calibrators (EnbPI, ACI, NexCP, AgACI) through a single typed API in which a specification object selects each method. In a controlled coverage study the IID bootstrap undercovers sharply under dependence; dependence-aware methods reduce the coverage deficit, the sieve nearest to nominal under short-memory linear dependence. On the shared fixed-statistic path a compiled backend runs several times faster than arch, and a streaming reduce avoids materializing the O(Bn) replicate tensor, limiting peak extra memory to O(B) for the statistic array. The software is MIT licensed (v0.6.1).
[AI-88] Can Reinforcement Learning Efficiently Discover Price Manipulation?
链接: https://arxiv.org/abs/2607.06121
作者: Ioanna-Yvonni Tsaknaki,Andrea Macrì,Fabrizio Lillo
类目: Trading and Market Microstructure (q-fin.TR); Artificial Intelligence (cs.AI)
备注: 30 pages, 11 figures and 5 tables
Abstract:In this paper, we investigate whether a model-free RL agent can identify and exploit price manipulation opportunities more effectively than a traditional model-based approach that assumes correct specification of the data-generating process but relies on noisy parameter estimates. We consider a single-asset market in which prices evolve according to an Almgren-Chriss framework with non-linear permanent impact and linear temporary impact. We first establish the existence of price-manipulative strategies in discrete time and compute the optimal benchmark strategy using Sequential Least Squares Quadratic Programming under full information. We then compare two finite-sample learning approaches: a model-based procedure that estimates impact parameters from simulated execution data and an agnostic RL approach based on Deep Deterministic Policy Gradient, trained directly on the same amount of data. For intermediate volatility, the RL agent successfully discovers profitable manipulative strategies without explicit knowledge of the underlying model, even when training data are quite limited. More importantly, RL consistently outperforms the model-based approach when parameter estimates are affected by sampling error, despite the latter benefiting from the correct model specification. For large volatility, all methods are unable to identify manipulation opportunities, while for small volatility, the model based approach outperforms RL. These findings highlight both the effectiveness of RL in complex control problems and the risks associated with deploying learning algorithms in financial markets without appropriate safeguards.
机器学习
[LG-0] he Key to Going Linear: Analysis-Driven Transformer Linearization
链接: https://arxiv.org/abs/2607.07706
作者: Anna Kuzina,Paul N. Whatmough,Babak Ehteshami Bejnordi
类目: Machine Learning (cs.LG)
*备注:
Abstract:The quadratic cost of causal self-attention severely bottlenecks long-context transformer inference. While numerous post hoc linearization pipelines exist, it is difficult to identify which components preserve model quality. This work isolates the effect of state update design in a strict frozen-backbone regime. We show that softmax relies on key-dependent, rank-1 orthogonal projections, elucidating why delta-style networks outperform purely gated accumulation. We identify a potential source of approximation errors and introduce structural interventions, specifically sink tokens, short convolutions, and fixed-budget cache routing, which reduces the remaining gap. We scale this linearization approach across LLaMA and Qwen models up to 32B parameters, outperforming prior post hoc baselines on MMLU and matching the long-context retrieval of complex adaptive-caching frameworks.
[LG-1] ECGLight: Compute-Light Framework For Paper ECG Digitization and Myocardial Infarction Screening
链接: https://arxiv.org/abs/2607.07683
作者: Shreyasvi Natraj,Cyrus Achtari,Felice Gragnano,Andrea Milzi,Marco Valgimigli,Diego Paez-Granados
类目: Machine Learning (cs.LG)
*备注:
Abstract:Electrocardiography (ECG) is one of the most widely used tests for diagnosing cardiovascular disease. Yet several remote clinics still utilize paper ECG printouts for their analysis due to limited connectivity and computational capacity. As a result, vast numbers of physical ECGs obtained in remote areas still remain incapable of being accessed by contemporary artificial-intelligence (AI)-based decision support as they require high computational resources or strong high-speed internet connectivity. This causes several cases where conditions like acute coronary occlusion (ACS) is overlooked and reperfusion therapy delayed. Although prior work has tackled digitization and diagnosis separately, and utilized advanced AI models for them, there still remains a lack of a compute-light, on-device framework that reconstructs paper ECGs at high fidelity, while accurately supporting multiple clinically relevant endpoints. We address this need with an end-to-end lightweight on-device digitization-to-diagnosis pipeline that converts a smartphone photo or scan of a paper ECG into a calibrated 12-lead signal and screens for Myocardial Infarction (MI) pathologies, with SHapley Additive exPlanations (SHAP) to support interpretability. Trained and evaluated on 21,799 ECGs from the PTB-XL dataset and further validated on hospital-acquired ECG-Matrix dataset, the complete system runs in 30 s per ECG on CPU-only resources, achieving 95.51% accuracy (F1 = 0.9519) for MI detection on PTB-XL and 88.89% accuracy (F1 = 0.8862) for OMI detection on ECG-Matrix. This work showcases that legacy paper records can be reliably democratized in any part of the world, providing a scalable decision support when digital ECG export, connectivity, or high-end compute are unavailable
[LG-2] Neural Operator-enabled Topology-informed Evolutionary Strategy for PDE-Constrained Optimization
链接: https://arxiv.org/abs/2607.07682
作者: Xiangming Huang,Guannan Zhang,Lu Lu,Raphaël Pestourie
类目: Machine Learning (cs.LG)
*备注:
Abstract:The inverse design of physical systems governed by partial differential equations is computationally demanding due to the high dimensionality and non-convexity of design spaces. Generative models for inverse design often lack robustness and transferability, whereas evolutionary strategies are robust but struggle in high-dimensional spaces. This paper introduces a Neural Operator-enabled Topology-informed Evolutionary Strategy (NOTES) that integrates dimensionality reduction, representation learning, and evolutionary optimization for efficient and transferable inverse design. NOTES couples a DeepONet-based neural operator with the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) to perform global optimization in a compact latent space that encodes topology-aware priors while discovering high-performance designs for unseen operating conditions. Applied to nanophotonic beam-deflector inverse design governed by Maxwell’s equations, NOTES reduces the design dimensionality from 256 to 25 and consistently achieves over 95 percent efficiency, outperforming CMA-ES, topology optimization, and other baselines. Applied to structural optimization, NOTES discovers designs that achieve compliance down to 246. By decoupling topology learning of a DeepONet from the governing physics in a PDE solver, NOTES provides a flexible and transferable framework for the inverse design of physical systems.
[LG-3] How Data Shapes RoPE Frequency Usage: From Positional Scale Matching to Length Generalization
链接: https://arxiv.org/abs/2607.07678
作者: Xinyi Wu,Siyuan Liu,Ali Jadbabaie
类目: Machine Learning (cs.LG)
*备注:
Abstract:Rotary Position Embeddings (RoPE) provide transformers with a fixed grid of positional frequencies, yet trained models use these frequencies highly non-uniformly. We study what determines this frequency usage and propose a data-centered explanation: RoPE frequencies are selected to match the relative-distance structure of the training data. Viewing each frequency as a positional lens, we formalize a field-resolution tradeoff and show that, for a data-induced dependency profile of width W , the optimal frequency scales as 1/W . This frequency-matching principle explains controlled observations on synthetic and text-based data, and suggests that the mid-low frequency bands observed in language models arise from the multi-scale dependency structure of natural language. We further connect frequency selection to position-interpolation-based length generalization: scaling frequencies down expands the effective field while reducing resolution. This helps when longer-context dependencies are approximate dilations of those seen during training, but can fail when relevant dependencies do not scale with context length. Empirically, we show that natural language exhibits approximate self-similarity across positional scales, explaining why test-time frequency scaling can support long-context generalization. Overall, our results identify a data-driven mechanism behind emergent RoPE frequency usage and show that long-context generalization depends on two forms of scale matching: between learned frequencies and training-time dependencies, and between frequency scaling and how those dependencies extend to longer contexts.
[LG-4] PeTeR: Post-Training Robustification of Probabilistic Circuits
链接: https://arxiv.org/abs/2607.07671
作者: Adrian Ciotinga,Yeming Dai,YooJung Choi
类目: Machine Learning (cs.LG)
*备注:
Abstract:Probabilistic circuits (PCs) can model complex joint distributions while supporting exact and efficient computation of many inference queries. However, standard likelihood-based PC learning is vulnerable to overfitting and fragile generalization when confronted with data noise, small sample sizes, or distribution shifts. This can be mitigated using distributionally-robust optimization which consider worst-case distributions within a Wasserstein ball of the empirical distribution, but current methods are limited to training a model from scratch in this framework. Instead, we propose PeTeR: a novel, data-free post-training framework designed to robustify pre-trained PCs against distribution shifts without retraining from scratch. Empirical evaluations across multiple density estimation benchmarks demonstrate that PeTeR effectively robustifies baseline models against both random and adversarial perturbations, achieving competitive or superior performance to data-dependent robust learning baselines.
[LG-5] Guidance Breaks the Fitted Operator: A Terminal-Fitted Repair for Classifier-Free Guidance
链接: https://arxiv.org/abs/2607.07665
作者: Shiheng Zhang
类目: Machine Learning (cs.LG); Numerical Analysis (math.NA)
*备注:
Abstract:Classifier-free guidance (CFG) is the standard way to strengthen class-conditioning in diffusion and flow-matching samplers, yet at large guidance it oversaturates and destabilizes, symptoms practitioners suppress with more steps or limited-interval schedules. We analyze CFG through an asymptotic-preserving, numerical-analysis lens. Building on a recent result that the deterministic DDIM step is the unique fitted operator for the unguided terminal layer, exact on the final small-sigma stretch of sampling, we show that guidance re-stiffens exactly the discriminative subspace to an anomalous exponent 1+w. DDIM is therefore no longer fitted there, and on coarse meshes its guided residual diverges as sigma_min goes to zero. We prove a guided clock barrier with three ordered step-size thresholds, and read one-step oversaturation as its endpoint: a solver artifact on the calibration model rather than the continuous guided law. The same analysis yields a one-coefficient, zero-extra-NFE repair: replace CFG’s w(r-1) by r^(1+w)-r on the guidance direction. On the calibration model’s discriminative crossover, this removes CFG’s sigma_min-divergent blow-up and is first-order accurate against the exact guided flow as sigma_min goes to zero. On learned CIFAR-10 checkpoints, and as a cross-domain smoke test on Stable Diffusion 1.5 DDIM, it acts as a high-guidance stabilizer at no extra cost rather than a universal quality knob: it cuts residual amplification and saturation, gives 9/9 point-FID wins over CFG on the tested grid, and preserves classifier-proxy target accuracy in the hard-cell blocks. We report the limits alongside: it is not a universal image-quality win, and against a dense vanilla-CFG reference it is not a uniformly better integrator of that field.
[LG-6] An optimal control approach for neural network architecture adaptation with a posteriori error estimation
链接: https://arxiv.org/abs/2607.07637
作者: C G Krishnanunni,Thomas Scott,Tan Bui-Thanh
类目: Machine Learning (cs.LG); Numerical Analysis (math.NA); Optimization and Control (math.OC)
*备注:
Abstract:This work presents a novel approach for adapting neural network architecture along the depth based on a posteriori error estimation. By formulating neural network training as a continuous-time optimal control problem, we derive rigorous error estimates that quantify how approximation error distributes across network layers. This error decomposition enables a principled depth adaptation strategy: new layers are inserted at locations of maximum estimated error, allowing the network to efficiently capture complex, nonlinear variations in the underlying problem. Our framework introduces a novel network architecture that treats weights and biases as piecewise linear functions varying across layers, with the error estimator bounding the discrepancy between this discrete representation and the true continuous optimal control solution. The approach leverages dual weighted residual methodology from finite element analysis to derive computable upper bounds on the functional error. A key theoretical contribution is the derivation of explicit error bounds that decompose the total approximation error into interval-wise contributions, providing a rigorous basis for targeted architecture refinement. We demonstrate the effectiveness of our method on scientific datasets, including learning the observable-to-parameter map for the Navier-Stokes equation. Numerical results reveal that our approach consistently outperforms existing architecture adaptation methods in terms of generalization performance.
[LG-7] Higher-Order Geometric Updates for Levenberg-Marquardt Method via Riemann Normal Coordinates
链接: https://arxiv.org/abs/2607.07623
作者: Jianing Liu,Dong H. Zhang
类目: Machine Learning (cs.LG); Numerical Analysis (math.NA); Chemical Physics (physics.chem-ph); Computational Physics (physics.comp-ph); Data Analysis, Statistics and Probability (physics.data-an)
*备注: 34 pages, 3 figures
Abstract:Nonlinear least-squares optimization is central to regression, physics-informed neural networks, and other machine-learning tasks. Such problems have a natural geometric interpretation, model predictions form a manifold in data space, while the chosen parameterization can introduce parameter-effects curvature that becomes a dominant source of nonlinearity. This exposes a limitation of the Levenberg-Marquardt (LM) method, its tangent-space step is applied as a straight update in parameter coordinates. Geodesic acceleration gives a second-order correction, but its removal of parameter-effect curvature is exact only in the infinitesimal-step limit. We propose a Riemann-normal-coordinate Levenberg-Marquardt method (RNC-LM) to improve this consistency for finite optimization steps. By reformulating the geodesic equation, RNC-LM extends geodesic acceleration to arbitrary-order corrections and constructs finite-step updates with progressively higher reparameterization consistency. A line search along the resulting RNC curve controls the traveled distance while keeping the cost close to standard LM. The method eliminates the tangential component of residual acceleration order by order in a moving tangent frame, making the actual objective reduction more consistent with the linear model prediction of LM. On classical nonlinear least-squares benchmarks, RNC-LM improves convergence and robustness in curved valleys and rank-deficient problems. On a reaction-diffusion PINN failure-mode benchmark, it reduces the relative L2 error to the order of 1e-3 and recovers a physically meaningful solution. On a large-scale machine-learning potential-energy-surface fitting task, it achieves a 34-fold speedup over standard LM.
[LG-8] Asymmetric Focal Loss Improves Graph Neural Network Prediction of Drug-Drug Interactions
链接: https://arxiv.org/abs/2607.07611
作者: Faranak Hatami,Mousa Moradi
类目: Machine Learning (cs.LG); Computational Physics (physics.comp-ph)
*备注:
Abstract:Background: Graph neural networks improve computational prediction of polypharmacy side effects, but standard binary cross-entropy training allocates equal capacity to well-classified and difficult examples, potentially missing clinically significant interactions. We evaluated whether an asymmetric focal objective could improve multi-relational drug-drug interaction (DDI) prediction by emphasizing difficult positive interactions. Methods: ClinicalFocal loss was integrated into a relation-aware graph convolutional network using molecular fingerprints, physicochemical descriptors, and learned embeddings. The model was evaluated on TWOSIDES using five-fold cross-validation with identical experimental conditions (architecture, features, data partitions, hyperparameters, and random seeds) for ClinicalFocal loss and binary cross-entropy baseline. Results: ClinicalFocal loss increased accuracy from 0.699 to 0.892 (+19.3 percentage points) and F1 score from 0.700 to 0.894 (+19.4 percentage points). AUROC increased from 0.766 to 0.914, and AUCPR increased from 0.714 to 0.860. The false-negative rate decreased from 29.8% to 9.1%, while specificity increased from 69.6% to 87.5%. Overall classification error decreased from 30.1% to 10.8%, corresponding to a 64.1% relative reduction. Improvements were consistent across all five folds. Conclusions: Asymmetric focal optimization improved classification and ranking performance while achieving 90.9% recall for observed interaction triples, without modifying the underlying architecture. Loss-function design is a direct, tunable lever for improving graph-based DDI prediction.
[LG-9] Multi-Class vs. Multi-Label BERT for CVE-to-CWE Mapping: How Taxonomy Structure Shapes the Errors ICANN2026
链接: https://arxiv.org/abs/2607.07573
作者: Ana Schwengber Kelm,Christian Bockermann,Jörg Frochte
类目: Machine Learning (cs.LG); Cryptography and Security (cs.CR)
*备注: 3 figures, 7 tables, to be published at the ICANN 2026 (International Conference on Artificial Neural Networks)
Abstract:Assigning Common Weakness Enumeration (CWE) categories to Common Vulnerabilities and Exposures (CVE) records remains an important but largely manual step in vulnerability analysis. We study this task as a text classification problem and compare two modelling choices: a \emphmulti-class formulation that predicts a single CWE per CVE and a \emphmulti-label formulation that allows multiple assignments. Three transformer encoders (BERT Base, SecureBERT, and CySecBERT) are evaluated on three nested label spaces (83, 47, and 25 classes). Multi-class training achieves higher macro-F1 across all settings, although the gap to multi-label narrows from 21 to 2 percentage points as the label space shrinks. Post-hoc threshold optimisation on the multi-label side closes this gap on the 25-class setting. Confusion analysis shows that the dominant misclassification patterns follow the CWE hierarchy and are shared across all three encoders (Pearson r 0.92 ), which suggests that the error structure is driven more by taxonomy design than by encoder choice. A hierarchy-relaxed evaluation that forgives within-family confusions raises macro-F1 from \sim 81% to \sim 90%, indicating that strict metrics understate branch-level classifier quality. CySecBERT achieves the strongest results overall, with statistically significant gains concentrated in the multi-label setting.
[LG-10] Avoiding unsafe sets when training with Langevin Dynamics
链接: https://arxiv.org/abs/2607.07538
作者: Adam M. Oberman
类目: Machine Learning (cs.LG); Analysis of PDEs (math.AP); Machine Learning (stat.ML)
*备注: 16 pages, 3 figures
Abstract:Training a model with noisy gradient descent can be idealized as overdamped Langevin dynamics on the loss landscape, and a natural safety question is to bound the probability \nu_t(\mathcalA_H) = \mathbbP(Q_t \in \mathcalA_H) that the trajectory lies in a designated failure region \mathcalA_H . We study this for a smooth, strongly convex loss in d dimensions and a failure region separated from the minimizer by an energy gap. Three bounds emerge. At the end of training, the equilibrium mass \pi(\mathcalA_H) is exponentially small in d , with a complementary energy-barrier rate when the noise is small. Along the trajectory, a shape-free bound \nu_t(\mathcalA_H) \le \pi(\mathcalA_H)(1 + \sqrt\chi_0^2/\pi(\mathcalA_H),e^-mt) shows that the in-set probability relaxes to (twice) the static value after a burn-in time of order d , using only the global spectral gap m of the loss. A worked Ornstein-Uhlenbeck example shows this burn-in is necessary: an angular slice of the equilibrium shell can transiently swell by a factor exponential in d , even though its equilibrium mass is tiny. To rule such swelling out we introduce a local relaxation rate attached to the failure region, defined through the spectral measure of its centered indicator rather than a Dirichlet-form Rayleigh quotient. For geometrically isolated regions this rate exceeds the global one, shrinking the burn-in proportionally, and combined with a maximum-principle ceiling it caps the trajectory probability uniformly in time. The picture is that strong convexity sets how fast training relaxes, but the shape of the unsafe set decides whether the trajectory bulges through it on the way home.
[LG-11] Gradient-free Riemannian Langevin Sampler
链接: https://arxiv.org/abs/2607.07519
作者: Ricardo Baptista,Olivier Zahm
类目: Machine Learning (cs.LG); Statistics Theory (math.ST)
*备注:
Abstract:We address the problem of efficiently sampling multimodal probability distributions, where standard Markov Chain Monte Carlo methods often suffer from poor mixing and mode trapping. To mitigate these issues, we propose Gradient-free Riemannian Langevin Sampler (GRiLS), a novel proposal that improves exploration without requiring gradient evaluations of the target density. Our approach introduces a Riemannian metric which reshapes the local geometry in order to facilitate transitions across modes. The resulting gradient-free MCMC algorithm is particularly suitable for complex, computationally expensive targets where derivatives are unavailable or impractical. The GRiLS proposal requires knowing the mean and covariance of the target density, which we estimate using an ensemble of interacting particles. Empirical results on multimodal benchmarks demonstrate that GRiLS achieves improved mixing compared to existing gradient-based and gradient-free MCMC approaches.
[LG-12] Fast Rates for Semi-Supervised Learning via Data-Augmentation Graph Regularization
链接: https://arxiv.org/abs/2607.07513
作者: Adam M. Oberman
类目: Machine Learning (cs.LG); Statistics Theory (math.ST); Machine Learning (stat.ML)
*备注: 21 pages, 2 figures
Abstract:Self-supervised learning matches supervised accuracy from a fraction of the labels, but the labeled-sample efficiency behind this has lacked a theoretical explanation. We provide one. Data augmentation induces a similarity graph on the unlabeled data, so downstream learning on that graph is graph-Laplacian-regularized learning. We prove a fast transductive rate, O(1/n_L) in the number of labels, in place of the supervised O(1/\sqrtn_L) , by carrying the leave-one-out stability apparatus of Johnson and Zhang (JMLR 2007) over to the augmentation graph, and without the unrealistic assumptions of limit-based analyses (exact kernel, generalizing features). The bound makes augmentation quality explicit: the expected error is at most C/n_L + R_\mathrmDA(y) , where the data-augmentation alignment error R_\mathrmDA(y) is the graph-cut mass of augmentations that cross a label boundary, so good augmentations let few labels suffice. The analysis uses a streamlined loss that drops the projector, negative-sample, and orthogonality overhead of standard objectives yet still recovers the top- K ideal features in the infinite-data limit, the augmentation-kernel eigenspace studied by Zhai et al. The result explains the observed accuracy-versus-label-count curve rather than only bounding a generalization gap.
[LG-13] GIFT: Geometry-Informed Low-precision Gradient Communication for LLM Pretraining
链接: https://arxiv.org/abs/2607.07494
作者: Jieying Wang,Shuyuan Fan,Mingkai Zheng,Zhao Zhang
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG)
*备注: 12 pages, 6 figures, 3 tables
Abstract:Gradient communication is a primary scaling bottleneck in large language model (LLM) pretraining. Communicating gradients in low-precision formats, such as FP8 and NVFP4, can significantly reduce the communication volume. Existing methods quantize gradients via linear or nonlinear mappings in Euclidean space, often degrading model performance because highly anisotropic gradients incur direction-dependent distortion. We present GIFT, a geometry-informed gradient scaling method that performs low-precision communication in geometry-aware coordinates. By transforming gradients into a near-isotropic space before quantization, GIFT makes low-precision representations substantially more faithful to their high-precision counterparts. GIFT only changes the coordinate system used for low-precision gradient communication and does not change the optimizer, training recipe, communication collective, or low-precision format. We also develop a simplified geometry-aware transformation algorithm with low-rank approximation and selective application to balance the computation overhead and communication reduction. We examine the empirical convergence of GIFT using Llama-300M and Llama-600M models. Our results show that GIFT reduces the end-to-end pretraining time of Llama-600M by 7.6% on 64 NVIDIA GH200 Superchips, while improving the downstream task preservation profile over direct Euclidean FP8 communication under the same optimizer and communication path.
[LG-14] he Optimal Sample Complexity of Learning Autoregressive Chain-of-Thought
链接: https://arxiv.org/abs/2607.07423
作者: Zhiyuan Li
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 33 pages
Abstract:We prove that, in the realizable PAC setting, the sample complexity of exact-trace learning for full autoregressive Chain-of-Thought traces is upper bounded by the standard multiclass rate of the local next-token class, where this rate is governed by the Daniely–Shalev-Shwartz dimension. Under exact-trace loss, one wrong action makes the whole trace incorrect; nevertheless, for every stopping rule \mathtthalt and every pointwise \mathtthalt -halting local class \mathrmH , n_\mathrmPAC^\varepsilon,\delta(\operatornameRoll_\mathtthalt(\mathrmH))=O((\operatornameDSdim(\mathrmH)+\log(1/\delta))/\varepsilon) , with no dependence on rollout length. The dependence on \operatornameDSdim(\mathrmH) is worst-case optimal, since one-step stopping recovers ordinary multiclass learning of \mathrmH . The proof introduces parity dimension, a rollout-stable refinement of DS dimension based on even pseudo-cubes. It controls one-inclusion density via a low-coordinate spanning theorem on finite restrictions and, unlike DS dimension itself, does not increase under autoregressive rollout. We also show why this detour is necessary: DS dimension can increase under rollout.
[LG-15] Sparse Delta Memory: Scaling the State of Linear RNNs through Sparsity
链接: https://arxiv.org/abs/2607.07386
作者: Loïc Cabannes,Pierre-Emmanuel Mazaré,Gergely Szilvasy,Matthijs Douze,Maria Lomeli,Ilze Amanda Auzina,Justin Carpentier,Gabriel Synnaeve,Hervé Jégou
类目: Machine Learning (cs.LG)
*备注:
Abstract:Linear attention models allow a fixed state size and a fixed amount of compute per token. However, due to their limited state size, linear attention models fall behind in long-context recall compared to softmax-attention-based transformer architectures. Increasing the state size of linear attention improves recall performance but at the cost of higher FLOPs. In this work, we introduce Sparse Delta Memory (SDM), an architecture that scales the hidden state of gated linear RNNs to orders of magnitude higher capacity using a sparse addressing scheme. SDM extends the Gated DeltaNet architecture by replacing the dense key-value outer product with sparse reads and writes to a large explicit memory. We show that, under an isoFLOP constraint and with an identical number of parameters, a higher state memory capacity significantly improves performance on in-context learning and long-context retrieval tasks. Moreover, by learning the initial state of the SDM memory and therefore using it as a parametric memory, we show that the model further improves on a wide range of common-knowledge and reasoning tasks.
[LG-16] Generalist Vision-Language Models for Fast Radio Burst detection: a zero-shot benchmark against a specialized detector
链接: https://arxiv.org/abs/2607.07382
作者: Raiff H. Santos,Amilcar R. Queiroz,Tharcisyo S. S. Duarte,K. E. L. de Farias,Rafael A. Batista
类目: Machine Learning (cs.LG); High Energy Astrophysical Phenomena (astro-ph.HE); Instrumentation and Methods for Astrophysics (astro-ph.IM)
*备注: 24 pages, 4 figures
Abstract:Fast Radio Bursts (FRBs) are millisecond-duration radio transients whose automated detection increasingly relies on highly specialized deep learning models. These detectors achieve exceptional performance, but they require large task-specific training datasets and cannot be redefined without retraining. In this work, we evaluate whether small, open-weight, locally run generalist Vision-Language Models (VLMs) can detect FRBs in dynamic spectra under a zero-shot, prompt-only regime, with no fine-tuning and no labeled examples, returning structured decisions with a natural-language justification. From a controlled set of 3000 simulated L-band dynamic spectra containing FRBs, structured Radio Frequency Interference (RFI), and noise, we draw a balanced binary benchmark of 2000 samples and compare two such VLMs (Gemma 4 2B and 4B), sample by sample, against the state-of-the-art specialized detector SwinYNet. At the default threshold, Gemma 4 2B reaches an accuracy of 93.65%, with no statistically significant difference from SwinYNet (92.90%), while showing a significantly lower false-positive rate on structured RFI (6.4% vs. 25.0%) and no false positives on pure noise. SwinYNet retains a perfect probabilistic ranking on this benchmark (ROC-AUC of 1.0000 vs. 0.9482), a ceiling that the zero-shot VLM approaches from general-purpose pretraining alone. Rewriting the prompt alone reconfigures the same models for three-class FRB/RFI/noise classification on the full set of 3000 spectra, where they reach up to 86% accuracy without a single false FRB.
[LG-17] Mechanistic Interpretability for Neural Networks: Circuits Sparse Features and Symbolic Reasoning
链接: https://arxiv.org/abs/2607.07316
作者: Pranav Sawant,Jakub Krejčí
类目: Machine Learning (cs.LG)
*备注: 20 pages
Abstract:This article offers a comprehensive overview of mechanistic interpretability, an emerging field that seeks to reverse-engineer the internal algorithms of modern neural networks. While traditional explainable AI methods often stop at surface-level input-output correlations, this approach directly addresses the opaque “black box” nature of machine learning models, which is essential for ensuring safety and auditability in high-stakes deployments. The paper provides a detailed examination of Transformer circuit analysis, exploring how internal components like the residual stream, attention mechanisms, and induction heads drive complex tasks and in-context learning. It subsequently tackles the core challenge of superposition and polysemanticity, demonstrating how tools like Sparse Autoencoders (SAEs) and transcoders can decompose tangled network activations into distinct, human-interpretable features. Furthermore, the paper explores methods for actively controlling and modifying model behavior through steering vectors and causal interventions. Finally, it connects these mechanistic insights with neurosymbolic AI frameworks designed to translate neural representations into explicit, executable logical rules.
[LG-18] Nonlinear Bandit
链接: https://arxiv.org/abs/2607.07304
作者: Tianshuo Zheng,Ting Wu,Zhi-Hua Zhou,Keqin Liu
类目: Machine Learning (cs.LG)
*备注:
Abstract:In this paper we first study the problem of generalized linear bandit (GLB) under heavy-tailed noise. The characteristics of heavy-tailed distributions are widely observed in real-world applications such as personalized recommendation, financial markets, and medical treatments. Based on the online mirror descent (OMD) method, we propose an algorithm EHM that extends the adaptive Huber loss method (Wang et al., 2025) with one-pass update ( \mathcalO(1) computational complexity with respect to current round t and the time horizon T ), which simultaneously achieves an almost optimal regret of \widetilde\mathcalO(T^\frac11+\epsilon) where T is the time horizon. In addition, by utilizing a special property of some link function (Sawarni et al., 2025), our algorithm eliminates the need to know a commonly used parameter. Next, we study the GLB problem under the case when contextual characteristic becomes piecewise constant, and we slightly revised former algorithm to obtain the PGLB-EHM algorithm. After theoretical analysis, we prove that the regret upper bound order stays the same. Furthermore, we look deeper into a special case of nonlinear bandit (NB) and present the NB-EHM algorithm with bisection method and special restriction. Eventually we utilize the affine lifting approach and show that the general NB problem can be applied with NB-EHM to achieve a sublinear regret bound.
[LG-19] BubbleSH: A Dataset of Rising Bubbles with Deformable Interfaces
链接: https://arxiv.org/abs/2607.07275
作者: Rachna Ramesh,Kiet Bennema ten Brinke,Douwe Orij,Ivo Roghair,Vlado Menkovski
类目: Machine Learning (cs.LG); Fluid Dynamics (physics.flu-dyn)
*备注: 17 pages, 7 figures, dataset available here: see this https URL
Abstract:Bubbly flows exhibit complex multiscale dynamics, with deformable bubbles interacting through the surrounding liquid and giving rise to strongly coupled kinematic and morphological behavior. We present BubbleSH, a bubbly flows dataset consisting of transient, three-dimensional bubble-swarm dynamics obtained from high-fidelity direct numerical simulations of bubbles rising in a periodic domain. The dataset provides time-resolved bubble trajectories, velocities, and shape evolution, with bubble morphology compactly represented using spherical harmonics. Designed to be lightweight yet physically expressive, the dataset enables data-driven modeling of bubbly flow simulators where shape deformation and bubble-bubble interactions play a central role. We characterize the dataset with bubble kinematics, morphology, and interaction patterns, and introduce evaluation metrics for both trajectory and shape prediction. The sensitivity of bubble-swarm dynamics to local perturbations makes BubbleSH particularly well suited to generative models that learn distributions over possible future trajectories. We evaluate a permutationally and translationally equivariant probabilistic emulator on BubbleSH given the proposed metrics. Therefore, we establish a compact, high-fidelity dataset and a benchmark for developing and evaluating data-driven models of deformable, chaotic multiphase systems.
[LG-20] Safe Reinforcement Learning using Ideas from Model Predictive Control
链接: https://arxiv.org/abs/2607.07252
作者: Georg Schäfer,Jakob Rehrl,Stefan Huber,Simon Hirlaender
类目: Machine Learning (cs.LG); Robotics (cs.RO)
*备注: Accepted at Eurocast 2026
Abstract:Reinforcement learning (RL) enables the synthesis of control policies directly from data, making it highly appealing for complex cyber-physical systems (CPSs) and robotics. A persistent challenge, however, is ensuring strict, hard safety constraints during the active learning phase. In real-world physical systems, violating mechanical limits can cause irreversible damage, necessitating that exploration remains strictly within safe operational regions. We propose a generalized framework that combines the adaptive, high-performance nature of deep reinforcement learning (DRL) with the formal safety guarantees of model predictive control (MPC). Using a mathematical model of the system dynamics, offline MPC computations define a feasible state-action space, representing all safe combinations of system states and control inputs that guarantee constraint satisfaction. During training and deployment, the RL agent’s instantaneous actions are projected onto this globally verified feasible set via a safety filter. We systematically evaluate our generalized approach on a non-linear 1-DoF laboratory testbed, demonstrating successful exploration and stable policy convergence on physical hardware.
[LG-21] HPG-Diff: Hierarchical physics-guided diffusion with differentiable connectivity constraints for topology optimization
链接: https://arxiv.org/abs/2607.07233
作者: Jinbo Yang,Mingyue Yuan,Boyuan Zhang,Yoshifumi Kitamura,Shikai Jing
类目: Machine Learning (cs.LG); Computational Engineering, Finance, and Science (cs.CE)
*备注:
Abstract:Deep generative models offer a promising paradigm for topology optimization, enabling rapid design exploration. However, these approaches lack intrinsic physics guidance, often leading to poor generalizability across unseen boundary conditions and the formation of floating material artifacts. To address these limitations, we propose Hierarchical Physics-Guided Diffusion (HPG-Diff), a novel diffusion framework that enforces physics consistency through two synergistic mechanisms. First, we introduce a hierarchical physics-guided strategy that aligns different precomputed physics features with the denoising process, guiding material distribution toward optimal load paths to enhance generalizability. Second, we propose a floating material suppression loss as a differentiable connectivity constraint inspired by thermal conduction to improve topological connectivity. By simulating a virtual heat propagation process from load positions, this mechanism explicitly penalizes floating material during training. Quantitative evaluations demonstrate that HPG-Diff achieves average compliance errors of 0.87% (in-distribution) and 5.29% (out-of-distribution), while reducing floating material ratios to 2.90% and 2.44%, respectively. Furthermore, case studies on a 3:1 rectangular domain, including cantilever and bridge benchmarks, provide preliminary evidence that lightweight LoRA fine-tuning with a small dataset can support the adaptation of HPG-Diff to rectangular non-square domains.
[LG-22] Continual Learning With Participation Privacy: An Auditable Buffering-Aggregation Recipe ICML2026
链接: https://arxiv.org/abs/2607.07209
作者: T-H. Hubert Chan,Elaine Shi,Mengshi Zhao,Mingxun Zhou
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注: This version corrects and clarifies the independent-decomposability condition underlying the adaptive-safety result in the ICML 2026 paper, with corresponding revisions to the affected statements and proofs
Abstract:Modern federated and streaming learning systems often release intermediate models, so privacy must hold for the full trajectory under adaptive interaction. Motivated by participation privacy, we study single-edit neighboring user streams, where one insertion/deletion shifts all subsequent updates and defeats standard Hamming-neighbor continual-release analyses. We give an auditable modular recipe. A randomized buffering wrapper emits bins of size [U,2U] , reducing single-edit streams to a Hamming-style per-bin update stream with explicit backlog/delay guarantees, where U is calibrated by the privacy parameters (\varepsilon,\delta) . We then prove a certification theorem identifying when a non-adaptive Hamming-neighbor DP proof for a continual primitive lifts to adaptive inputs: the primitive must use fresh per-round randomness and have a stable one-round privacy profile under common adaptive context. Together, these ingredients yield trajectory-level (\varepsilon,\delta) -DP for single-edit streams using standard primitives (e.g., tree prefix sums), with an explicit privacy–latency link via U .
[LG-23] Geometric–Nongeometric Optimizer Calculus: A Modular Language for Reachable Gradient Methods
链接: https://arxiv.org/abs/2607.07206
作者: Zavier Li
类目: Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注: 21 pages
Abstract:Adaptive optimizers mix several mechanisms: a metric or preconditioner maps gradients to descent directions, while estimation, memory, step-size control, constraints, stochasticity, target modification, and discretization determine which directions are available and how they are used. We introduce geometric–nongeometric optimizer calculus, a modular language for auditing reachable gradient methods under explicit oracle, budget, state, and rule constraints. The geometric module is a positive cometric family that maps covectors to parameter-space directions; the nongeometric modules are information, memory, control, operator, noise, target, and discretization mechanisms. The main formal result is a direction-expressivity theorem: away from critical points, full positive-definite geometry expresses exactly the strict descent directions. We then define restricted direction residuals for admissible metric families, prove exact expressivity conditions for diagonal and block geometries, and separate this direction-level diagnostic from condition-number geometric complexity. The resulting design problem is a Pareto optimization over module budgets, not a single universal optimizer ordering. We also lift pointwise residuals to a trajectory-level residual complexity that couples direction mismatch with the variation of the explaining geometry. We include diagnostic prototypes only as evidence for the language: a high-information full-metric probe solves deterministic quadratic benchmarks to numerical precision, while a practical Muon-style PyTorch candidate gives small-scale evidence that matrix-operator updates can be audited by the calculus. The paper is a theory and benchmark-language manuscript; it does not claim large-scale optimizer state-of-the-art performance.
[LG-24] Information Allocation Dynamics in Neural Network Optimization
链接: https://arxiv.org/abs/2607.07156
作者: Zhang Gongyue,Liu Donghan,Ren Weihong,Sheng Yixuan,Wang Zhiyong,Liu Honghai
类目: Machine Learning (cs.LG)
*备注:
Abstract:Different optimizers have different update biases, but these biases are usually implicit. Existing studies mainly analyze or control such biases from the geometry of the final solution. However, how optimizer bias forms during training still lacks a clear internal mechanism. This paper proposes an information allocation dynamics perspective. It interprets optimizer implicit bias as the relative allocation of training signals between weight-like and bias-like parameter pathways. This allocation can be described and adjusted by a continuous preconditioning exponent (p). To characterize this mechanism, we first analyze the update contributions of weight and bias to the same residual signal in a minimal linear model. The weight correction term preserves input-dependent residual signals, while the bias correction term preserves the residual mean direction. They therefore correspond to different projection pathways of the residual signal. After substituting the preconditioned update into the residual update equation, the optimizer can change the relative strength of the weight correction term and the bias correction term through different preconditioning factors. Therefore, optimizer implicit bias is not only reflected in the final solution or the global training trajectory. It is also reflected in the relative write-in ratio of training signals across different parameter pathways. Overall, this paper moves the analysis of optimizer implicit bias from solution-space geometry to update dynamics during training. It reveals that the relative update allocation between weight and bias-like parameters is an important dynamical mechanism that affects parameter trajectories and generalization behavior. Subjects: Machine Learning (cs.LG) Cite as: arXiv:2607.07156 [cs.LG] (or arXiv:2607.07156v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2607.07156 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Gongyue Zhang [view email] [v1] Wed, 8 Jul 2026 08:49:23 UTC (2,541 KB)
[LG-25] Fractal KV-Cache Archives: Lossless Symbolic Storag e with In-Place Retrieval for Long-Context LLM Inference
链接: https://arxiv.org/abs/2607.07144
作者: Vladimir Gusev
类目: Machine Learning (cs.LG)
*备注: 8 pages, 3 figures. Code: this https URL
Abstract:The key-value (KV) cache dominates the memory cost of long-context autoregressive inference, and a growing body of work compresses it through quantization, eviction, or offloading. We study a complementary question: once a position’s KV state has been quantized to codebook indices, how should the resulting symbol stream be stored, and can the storage layer do more than store? A family of contractive iterated-map codes that serialize a symbol sequence into a sequence of low-dimensional real vectors is revisited, and it is shown that they form a natural archive format for a quantized KV cache with the following features. The method provides exactly the access pattern a growing cache requires. It is lossless, it runs in linear time, and supports O(1) random access and O(1) amortized append. A controlled study of the quantizer feeding this archive is conducted on GPT-2 with 1024-token contexts. Keeping a small exact window (4 attention sinks + 32 recent tokens) and archiving the rest, per-head residual vector quantization reduces the archived cache by 36-54x relative to an fp16 cache at a perplexity cost of 11-15%, and we quantify a sharp key/value asymmetry – quantizing keys is roughly 4x more damaging than quantizing values, consistent with prior low-bit KV work – and use it to allocate bits in a hybrid scheme. Finally, we show the archive is simultaneously a search index: approximate substring queries execute directly on the stored vectors, and matched context is decoded from the matched vector without ever materializing the surrounding text. We release all code; every number reproduces from a single command on a laptop CPU.
[LG-26] Distributed Sparse Interventions in Language Models
链接: https://arxiv.org/abs/2607.07128
作者: Maximilian S. Ernst(1,2,3),Lorenz Linhardt(3,4),Aaron Peikert(2),Oliver Eberle(3,4) ((1) Max Planck School of Cognition (2) Center for Lifespan Psychology Max Planck Institute for Human Development (3) Machine Learning Group Technische Universität Berlin (4) Berlin Institute for the Foundations of Learning and Data (BIFOLD))
类目: Machine Learning (cs.LG)
*备注: 29 pages, 12 figures
Abstract:Language models perform a wide range of tasks at varying levels of abstraction with the capacity to flexibly infer tasks from context, execute multiple tasks simultaneously, and select among competing tasks. To study the role of model components in task behaviour, their causal influence can be investigated through interventions. Prior work on model steering has largely focused on interventions along global directions in activation space, modeling task representations as approximately linear and additive. By studying interventions at the neuron level, we find substantial, neuron-specific nonlinear effects on model outputs that are not captured by current steering approaches. We introduce Distributed Sparse Interventions (DSI), an intervention approach that considers nonlinearities and interactions between neurons across layers to identify sparse sets of neurons that elicit task-relevant computations. Across a range of tasks, we demonstrate that DSI can activate task behaviour in instruction-tuned language models by localising and intervening on as few as 0.01% of neurons, highlighting the effectiveness of sparse, distributed interventions in the neuron basis. Additionally, adopting a set-based perspective enables computations over the identified neuron sets, offering insights into the roles of individual neurons by analysing their effects across tasks. Through sparse interventions, DSI enables fine-grained control over model behaviour, localisation of task-relevant neuron sets, and furthers our understanding of task composition.
[LG-27] A knowledge-augmented dataset of high-risk driving scenarios with LLM annotations for autonomous driving
链接: https://arxiv.org/abs/2607.07103
作者: Heye Huang,Jingguang Li,Zhiyuan Zhou,Paul Liang,Mingyu Wu,Kitae Jang,Jianqiang Wang
类目: Machine Learning (cs.LG); Databases (cs.DB)
*备注: 22 pages, 9 figures
Abstract:Safe autonomous driving requires both rapid responses to common high-risk events and deeper reasoning over rare, extreme long-tail scenarios in traffic safety. These scenarios are severely under-represented in naturalistic driving data, and existing trajectory and language-augmented datasets seldom provide high-risk event labels, semantic annotations, and verifiable safety signals. Here we present K-Risk, a knowledge-augmented dataset that combines structured driving trajectories with large language model generated semantic annotations for safety-critical driving scenarios. K-Risk integrates 20 human-driven and autonomous-vehicle trajectory datasets from Europe, China, and the United States, covering highways, urban freeways, intersections, and roundabouts. Using a unified risk-centric extraction pipeline, K-Risk curates 31,398 high-risk events, together with a 1,036-event extreme subset of near-collision cases. Each event is released as a synchronized trajectory, metadata, and language triplet containing structured scenario descriptions, abnormal-behavior notifications, and, for a representative subset, causal risk analyses and action recommendations validated through a closed-loop simulator with iterative reflection. By combining multi-dimensional risk annotations, interpretable language supervision, and verifiable decisions, K-Risk bridges structured traffic trajectories, semantic reasoning, and decision supervision, providing a standardized foundation for developing and evaluating next-generation risk-aware autonomous driving agents.
[LG-28] Structural Adversarial Attacks on Relational Deep Learning under Integrity Constraints VLDB
链接: https://arxiv.org/abs/2607.07089
作者: Alan Gany,Bogdan Cautis,Silviu Maniu
类目: Machine Learning (cs.LG); Databases (cs.DB)
*备注: To be presented at the VLDB Workshop AI for Databases
Abstract:Relational Deep Learning (RDL) has become a standard methodology for machine learning on relational databases: the database is encoded as a heterogeneous temporal graph in which tuples become nodes and primary-key to foreign-key (PK-FK) dependencies become typed edges, over which a graph neural network is trained for downstream prediction. We study the adversarial robustness of this pipeline. We consider a white-box attacker who knows how the graph is built and the model is trained, reasons about perturbations on the graph, but can only act on the upstream database, by rewiring foreign-key references while preserving the integrity constraints of the schema (foreign-key validity, the degree-one FK constraint, and functional dependencies). This restricts the attacker to a constrained, combinatorial set of admissible edits under a global perturbation budget, which is intractable to explore exhaustively and made non-additive by GNN message passing. We investigate seven attack heuristics - two random sampling baselines and five gradient-guided variants that exploit differentiable edge masks - and evaluate them on the RelBench rel-f1 benchmark. Gradient-based attacks consistently outperform random baselines on regression tasks, whereas gains on classification are smaller, which we attribute to low label-flip rates and greater local stability of classification outputs.
[LG-29] Is Randomness Necessary for Adaptive Data Analysis?
链接: https://arxiv.org/abs/2607.07085
作者: Edith Cohen,Haim Kaplan,Yishay Mansour,Shay Sapir,Uri Stemmer
类目: Cryptography and Security (cs.CR); Data Structures and Algorithms (cs.DS); Machine Learning (cs.LG)
*备注: 22 pages
Abstract:The Adaptive Data Analysis (ADA) problem formalizes the challenge of preventing false discovery and overfitting when a dataset is repeatedly reused. Formally, our input is a dataset containing n i.i.d. samples from an unknown distribution \mathcalP over a domain \mathcalX , and our goal is to answer a sequence of k adaptively chosen statistical queries with respect to \mathcalP . The main question is how many queries we can support (i.e., how large k can be), primarily as a function of the number of samples n . This question has been intensively studied and is relatively well-understood for randomized mechanisms: there are computationally efficient mechanisms that support k \approx n^2 queries, and no computationally efficient mechanism can answer k \gg n^2 queries. In this paper, we address a fundamental question: is randomness necessary for ADA? Despite a decade of work on ADA, this question remains open. A folklore observation dating back to the initial works on ADA is that randomness is not necessary when the analyst is computationally bounded. Yet, the necessity of randomness against computationally unbounded analysts has remained elusive. Our main contribution resolves this gap in the information-theoretic Random Oracle model. Perhaps surprisingly, we show that randomness is strictly necessary to answer a non-trivial number of adaptive queries: when the analyst is unbounded, any deterministic mechanism can be forced to fail after just k = \tildeO (n) queries. Comments: 22 pages Subjects: Cryptography and Security (cs.CR); Data Structures and Algorithms (cs.DS); Machine Learning (cs.LG) Cite as: arXiv:2607.07085 [cs.CR] (or arXiv:2607.07085v1 [cs.CR] for this version) https://doi.org/10.48550/arXiv.2607.07085 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-30] Prior-aware and Context-guided Group Sampling for Active Probabilistic Subsampling ICLR2026
链接: https://arxiv.org/abs/2607.07083
作者: Beomgu Kang,Hyunseok Seo
类目: ystems and Control (eess.SY); Machine Learning (cs.LG)
*备注: ICLR 2026
Abstract:Subsampling significantly reduces the number of measurements, thereby streamlining data processing and transfer overhead, and shortening acquisition time across diverse real-world applications. The recently introduced Active Deep Probabilistic Subsampling (A-DPS) approach jointly optimizes both the subsampling pattern and the downstream task model, enabling instance- and subject-specific sampling trajectories and effective adaptation to new data at inference time. However, this approach does not fully leverage valuable dataset priors and relies on top-1 sampling, which can impede the optimization process. Herein, we enhance A-DPS by integrating a deterministic (fixed) prior-informed sampling pattern derived from the training dataset, along with group-based sampling via top-k sampling, to achieve more robust optimization, method we call Prior-aware and context-guided Group-based Active DPS (PGA-DPS). We also provide a theoretical analysis supporting improved optimization via group sampling, and validate this with empirical results. We evaluated PGA-DPS on three tasks: classification, image reconstruction, and segmentation, using the MNIST, CIFAR-10, fastMRI knee, and hyperspectral AeroRIT datasets, respectively. In every case, PGA-DPS outperformed A-DPS, DPS, and all other sampling methods.
[LG-31] An Hybrid Quantum-Classical Diffusion Model for Image Generation
链接: https://arxiv.org/abs/2607.07072
作者: Qipeng Qian,Keli Deng,Yuntao Qian
类目: Machine Learning (cs.LG)
*备注:
Abstract:Quantum diffusion models provide a physics-consistent route to generative learning by formulating noising and denoising directly on quantum states. However, applying such models to classical high-dimensional data is constrained by the qubit cost of state encoding and the computational burden of simulating large density operators. We propose a scalable hybrid generative pipeline that combines a classical autoencoder for dimensionality reduction with a mixed-state quantum denoising diffusion probabilistic model (MSQuDDPM) operating in the learned latent space. The autoencoder compresses data into compact latent codes that can be embedded into a small-qubit Hilbert space, after which the quantum diffusion model learns a generative distribution over latent density operators and decodes samples back to the original domain. Algorithmically, we simplify the reverse dynamics by predicting an estimate of the clean state \rho_0 at timestep t and computing the one-step reverse update via an analytic backward propagation rule, rather than learning an explicit predictor for \rho_t-1 . We demonstrate the proposed approach on MNIST image generation and discuss how mixed-state quantum diffusion can serve as a practical backbone for hybrid quantum–classical generative modeling under realistic qubit budgets.
[LG-32] Gauge-Invariant Learnable Spectral Positional Encodings for Directed Graphs via Hermitian Block Krylov Subspaces
链接: https://arxiv.org/abs/2607.07032
作者: Jiaqing Xie,Yuxin Wang
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 8 pages main
Abstract:Spectral positional encodings (PEs) for \emphdirected graphs face two obstacles: magnetic Laplacians require an O(n^3) Hermitian eigendecomposition per potential, and their complex eigenvectors are defined only up to unitary gauge, which prior work handles with basis-invariant architectures. We propose learnable spectral PEs of the form h_\theta(A_q),R , where A_q is a normalized magnetic operator, h_\theta a learnable scalar spectral response, and R a block of random probes. Because the PE is a \emphmatrix function of the operator, it is gauge-invariant by construction. We compute it in a Hermitian block Krylov subspace from sparse matrix–vector products only, prove that k = O(\log(1/\varepsilon)) block steps suffice uniformly over heat–resolvent response families, and give a covering-number argument for why low-dimensional structured families generalize where free per-eigenvalue weights overfit. On a directed SBM whose symmetrization is uninformative by construction, direction-blind PEs stay at chance while magnetic Krylov PEs converge to the exact-eigendecomposition oracle as the depth grows. The same probes yield gauge-invariant pairwise features with 1/\sqrts Monte-Carlo error, and the undirected q=0 case improves heterophilous benchmarks over no-PE and polynomial baselines.
[LG-33] Constrained Decoding for Diffusion Language Models via Efficient Inference over Finite Automata
链接: https://arxiv.org/abs/2607.07026
作者: Meihua Dang,Stefano Ermon
类目: Machine Learning (cs.LG)
*备注:
Abstract:Constrained decoding is essential for serving LLMs, ensuring that generated outputs follow specific structures such as JSON schema-formatted function calls. Existing systems are designed for autoregressive models and assume left-to-right generation, masking out invalid next tokens at each step. Diffusion language models, however, break this assumption: they sample multiple positions simultaneously from a fully-factorized mean-field distribution at each denoising step. In this paper, we present an exact and tractable algorithm for sampling from the constrained mean-field posterior under any constraint expressible as a finite automaton. Viewing finite automata as graphical models, we obtain tractable representations of the constrained distribution that enable efficient inference. The approach guarantees constraint satisfaction by construction, supports both greedy and sampling-based decoding, and is compatible with parallel and block-wise decoding under arbitrary remasking schedules. Applying depth-reduction techniques from arithmetic circuit theory, we further reduce sampling depth from linear to logarithmic in the sequence length. Empirical evaluations on Dream-7B and LLaDA-8B show substantial accuracy gains across various tasks including function calling (xLAM, BFCL), planning (Sudoku, Countdown), text-to-SQL (Spider), and math reasoning (GSM-Symbolic), with little inference overhead relative to unconstrained decoding. For example, on BFCL-Live, our approach improves Dream-7B’s greedy decoding accuracy from 63.9% to 71.5%, and stochastic sampling accuracy from 22.3% to 69.0%, where the unconstrained baseline collapses, with under 5% wall-clock overhead.
[LG-34] Online Data Selection Is Implicit Alignment
链接: https://arxiv.org/abs/2607.07023
作者: Aoxiong Zeng,Yuxin Yang,Xiangquan Yang
类目: Machine Learning (cs.LG)
*备注:
Abstract:Supervised fine-tuning (SFT) is often treated as a capability-adaptation step, while alignment is attributed to later preference optimization or reinforcement learning. This separation is incomplete: when examples are scored and kept online during fine-tuning, the choice of which data to train on already changes the model’s behavioral preferences. We study online data selection as an implicit alignment mechanism. Given the same base model, optimizer, and selected-token budget, we compare random, loss-based, quality-based, and diversity-based online selectors and measure the behavioral drift they induce without any preference optimization. The proposed evaluation tracks helpfulness, refusal rate, verbosity, truthfulness, sycophancy, calibration, and jailbreak robustness, together with diagnostics for which behavioral modes are over-represented in the selected data. We formalize online selection as a reweighted SFT objective whose weights define an implicit preference over response styles and safety postures, so that an online scorer plays the role usually assigned to a reward model. This view predicts that high-scoring data can systematically favor longer, more assertive, more compliant, or more refusal-prone behaviors depending on how the online score is defined. Empirically, selectors that are statistically indistinguishable in task accuracy diverge sharply in refusal rate, verbosity, and sycophancy, and we show that the direction of the shift is predictable from the attribute mixture of the selected data. We introduce Alignment Drift Auditing (ADA), a controlled protocol for quantifying selection-induced behavioral movement, and Alignment-Aware Selection (AAS), a diagnostic online selector that retains data efficiency while constraining drift along safety and style axes.
[LG-35] UP: Unbounded Positive Asymmetric Optimization for Breaking the Exploration-Stability Dilemma
链接: https://arxiv.org/abs/2607.06987
作者: Chongyu Fan,Pengfei Liu,Jingjia Huang,Sijia Liu,Yi Lin
类目: Machine Learning (cs.LG)
*备注:
Abstract:Reinforcement learning (RL) has become the standard paradigm for enhancing the complex reasoning capabilities of large language models (LLMs). To achieve sample efficiency, modern RL frameworks rely on importance sampling (IS). However, these algorithms suffer from an exploration-stability dilemma. Pure IS often leads to catastrophic training instability, while standard clipping mechanisms used to mitigate this instability strictly constrain the policy update budget. By formalizing the concept of Probability Capacity (Cap), we reveal that conservative clipping structurally stifles exploration by prematurely truncating the update budget for correct but low-confidence reasoning paths. To break free from these constraints, we propose Unbounded Positive Asymmetric Optimization (UP), a universal and plug-and-play objective. UP theoretically restructures the optimization process by anchoring the policy to its current state via the stop-gradient operator. This asymmetric design unleashes unclipped, stable gradients for positive advantages to maximize exploration, while maintaining standard clipping safeguards for negative advantages to prevent training instability. Furthermore, our formulation readily extends across different optimization granularities, including token-level (GRPO, DAPO) and sequence-level (GSPO) frameworks. Extensive experiments demonstrate that UP enhances exploration capacity and achieves superior reasoning accuracy across diverse RL algorithms (DAPO, GSPO, and GRPO), model architectures (Dense, MoE, and vision-language), and training modalities (language and multimodal), validating UP as a truly universal plug-and-play enhancement for RL-based training.
[LG-36] Robust Federated Learning Under Real-World Client Churn
链接: https://arxiv.org/abs/2607.06979
作者: Dhruv Garg,Neha Lakhani,Debopam Sanyal,Myungjin Lee,Alexey Tumanov,Ada Gavrilovska
类目: Machine Learning (cs.LG)
*备注: 13 pages
Abstract:Federated Learning (FL) enables training shared models on private, on-device data, but production deployments remain constrained to slow, multi-day refresh cycles due to the complexity of coordinating massive client populations. For applications such as feed ranking, ad targeting, and personalized recommendation, model freshness: the ability to rapidly adapt to new user-local data is critical for maximizing objectives like click-through rate. This lag leaves models stale and unresponsive to volatile data distributions driven by viral trends and shifting user intent. Bridging this gap requires addressing three challenges overlooked by existing FL systems: transient client availability, dynamic data heterogeneity, and delays between model predictions and observable outcomes. We present FeLiX, an FL orchestration framework that minimizes wall-clock time-to-target accuracy on live interaction streams. FeLiX introduces three primitives: (i) streaming-aware availability tiers that leverage lightweight telemetry to identify ready clients at scale; (ii) fresh-utility selection, a dual-tier mechanism that prioritizes statistically valuable updates from devices able to meet tight refresh deadlines; and (iii) informativeness-aware, delay-robust aggregation that incorporates late, high-value updates containing ground-truth outcomes without biasing the global model toward stale distributions. Unlike prior systems that rely on unrealistic oracular knowledge of client availability, FeLiX achieves near-oracular performance in real-world settings. Across CIFAR-10, Google Speech, and realistic low-availability traces, FeLiX reduces wall-clock time-to-target accuracy by up to 2.37X while reducing communication bandwidth by 1.30X compared to state-of-the-art synchronous and asynchronous FL baselines.
[LG-37] Rethinking Multimodal Time-Series Forecasting Evaluation
链接: https://arxiv.org/abs/2607.06973
作者: Haoxin Liu,Yichen Zhou,Rajat Sen,B. Aditya Prakash,Abhimanyu Das
类目: Machine Learning (cs.LG)
*备注:
Abstract:We introduce a new context-enriched, multimodal time series forecasting benchmark, TimesX. TimesX contains a wide selection of high-quality real-world time series with diverse domains and textual contexts obtained from an automated data generation pipeline, which helps address three main issues of existing multimodal forecasting benchmarks: (1) poor generalization due to the small scale and synthetic nature of benchmark data, (2) very limited types of textual contexts in the benchmarks, and (3) an inability to mitigate data leakage in evaluation. We conduct a thorough empirical study of zero-shot multimodal forecasting approaches on TimesX. Our results suggest that many approaches that perform well on existing benchmarks may fail on TimesX. In contrast, simple ensemble methods that leverage rich textual context accompanying time-series can outperform strong baselines on TimesX.
[LG-38] Flow-ERD: Agent -type Aware Flow Matching with Entropy-Regularized Distillation for Diverse Traffic Simulation
链接: https://arxiv.org/abs/2607.06957
作者: Seulbin Hwang,Kiyoung Om,Daejung Kim,Jinhan Lee
类目: Robotics (cs.RO); Machine Learning (cs.LG)
*备注: 8 pages, 5 figures, 2 tables
Abstract:Realistic and diverse traffic simulation is essential to autonomous driving development. Yet prevailing benchmarks predominantly reward realism, and recent methods have optimized accordingly, leaving diversity underexplored. We introduce \textbfFlow-ERD, a multi-agent simulator that pursues realism and diversity jointly. Its backbone, \textbfAgent-Type Aware Flow Matching (AFM), couples flow matching’s multi-modal expressiveness with type-specific kinematic execution. It preserves fine-grained diversity while keeping motions consistent with each agent type. A second stage, \textbfEntropy-Regularized Distillation (ERD), fine-tunes the closed-loop rollout distribution with an entropy-regularized reverse-KL objective. This mitigates covariate shift while explicitly preventing collapse onto high-density modes. We evaluate Flow-ERD with a log-free diversity metric alongside standard realism scores. Flow-ERD ranks first on the WOSAC test benchmark and dominates the realism–diversity Pareto front among reproducible baselines. Our project page is available \hrefthis https URLhere.
[LG-39] Physical activities enable scalable foundation modelling for broad-spectrum health prediction
链接: https://arxiv.org/abs/2607.06954
作者: Zhenghuang Wu,Yuyao Zhu,Songlin Xu
类目: Machine Learning (cs.LG)
*备注: 18 pages, 5 figures, 6 tables
Abstract:Wearable and mobile sensing technologies have demonstrated strong potential for health inference; however, most sensor models are designed for specific disease types, limiting their transferability across different health risks. Wearable foundation models offer a more generalizable approach in diverse health risk types. Nevertheless, most existing methods rely on high-frequency raw sensor data, raising concerns about privacy, computational overhead, and scalability across devices and populations. In this paper, we propose StepFM, a foundation model built solely on step counter data for broad-spectrum health prediction. Leveraging the ubiquity and low-dimensional nature of step data, StepFM provides a practical, privacy-preserving, and computation-efficient alternative to traditional sensor-based models. We design a scalable pre-training framework that captures temporal dynamics and behavioral patterns from large-scale step sequences, enabling transfer across more than 20 health risk prediction tasks spanning diverse devices, new regions, and novel disease types. Extensive experiments demonstrate that StepFM achieves strong performance compared to existing methods while maintaining robustness across heterogeneous settings. Furthermore, our analysis reveals interpretable and generalizable relationships between physical activity patterns and various health risks, offering new insights into activity-based health modeling. Our work establishes step-based sensing as a viable foundation for scalable and real-world health monitoring.
[LG-40] Intrinsic-Noise Consolidation: A Doob-Barrier-Conditioned Diffusion Turns Analog Device Noise into a Continual-Learning Resource
链接: https://arxiv.org/abs/2607.06924
作者: Gunner Levi Howe
类目: Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
*备注: 14 pages, 9 figures. Includes a proof-of-concept training run on real BrainScaleS-2 neuromorphic silicon (hardware-in-the-loop). Code, and data
Abstract:On analog neuromorphic hardware, intrinsic device noise is normally an accuracy tax. We ask whether it can instead consolidate memories. We cast per-synapse consolidation as a Doob h-transform: condition each weight’s stochastic dynamics on never crossing a memory-critical barrier around its consolidated value. The conditioned diffusion gains an extra drift sigma^2 d/dw log h, a restoring force amplified by the noise variance itself that diverges at the barrier. We are explicit about novelty: the anchored drift -s(w-mu) our rule also contains is not ours (the limit of OUA, MESU, and EWC), and we surrender it. We claim only the conjunction of (a) the Doob barrier-conditioning as a synaptic rule, to our knowledge unclaimed (every h-transform use we found is generative modeling, none synaptic), and (b) a falsifiable prediction: increasing intrinsic noise non-monotonically improves sequential-task retention, an inverted-U that anchored-drift methods cannot produce. We pre-registered this as a go/no-go gate; it passes. On single-head Split-MNIST (8 seeds) the rule lifts retention 10.9 points at an interior optimum (paired Wilcoxon p=0.004), while matched OU/EWC/MESU anchors are monotone. Ablating the conditioning removes the effect; the optimum tracks the barrier; the inverted-U survives a second task stream and the realization where noise enters the forward pass. We then measure the intrinsic noise on real BrainScaleS-2 silicon (additive, trial-to-trial independent, tunable via on-chip averaging) and run the rule on the chip with its noise in the training loop: barrier-conditioning retains a prior task 15.6 points better than the matched control at matched average accuracy, a stability-plasticity shift, not a net-accuracy win (single seed; retention measured, energy modelled). Intrinsic analog noise thus becomes a consolidation dividend a digital accelerator must spend energy to generate.
[LG-41] Converge to Surprise: Evolutionary Self-supervised Image Clustering
链接: https://arxiv.org/abs/2607.06887
作者: Canlin Zhang,Xiuwen Liu
类目: Machine Learning (cs.LG)
*备注:
Abstract:Most self-supervised image clustering models, actually almost all deep learning approaches, are based on gradient descent: In order to calculate the loss, every optimization step requires a clearly defined target, whether a contrastive split, a masked patch or entity, an EMA-teacher output, a pseudo-label, or a differentiable information-theoretic functional. We propose a self-supervised framework that drops this requirement for image clustering. Without any prior knowledge, we have to assume that each pixel is i.i.d. according to the Principle of Maximum Entropy. Taking this as our null hypothesis H0, we define a “surprise score” that measures how unlikely the model’s output representation would be under H0. Maximizing the surprise score forces the deep learning model to reject H0 - equivalently, to discover non-random feature from data. Also, here is our fundamental assumption: a surprise score cannot, in general, be reduced to a per-step loss. Hence, we propose the “converge-to-surprise” scheme to optimize our model: an evolution-strategy (ES) outer loop, which directly maximizes the surprise score without needing its gradient, paired with a periodic gradient-descent inner loop, which uses the surprising clusters already discovered by ES as surrogate targets. On standard image benchmarks, our framework achieves new state-of-the-art results in non-parametric self-supervised image clustering - the strictest deep-clustering setting, in which the number of ground-truth classes is not given to the model.
[LG-42] Best-Arm Identification with Generative Proxy
链接: https://arxiv.org/abs/2607.06879
作者: Tianyi Ma,Hanzhang Qin,Ruihao Zhu,Jierui Zuo
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:Best-arm identification is a canonical model for data-driven decision-making, but in many applications each reward observation is costly. Motivated by the growing availability of cheap predictions from machine learning and large language models, we study fixed-confidence best-arm identification in which each costly reward pull is paired with a cheap but correlated proxy score. The marginal mean of the proxy can be estimated offline and is treated as known, whereas its correlation \rho with the reward, which governs how much the proxy helps, is unknown and must be learned online in pair with real rewards. We show that a control-variate adjustment turns this model into a heteroscedastic identification problem whose oracle sample complexity improves by residual variance 1-\rho^2 . The central difficulty is that the correlation must be learned from the same costly samples that identification consumes online, and that a plug-in estimate of the residual variance is anti-conservative and can compromise correctness. We propose PROBE (PRoxy OLS for Best-arm Exploration), a phase-elimination algorithm that directly maintains an upper certificate on the residual variance with an ordinary least squares fit, whose exact chi-square law keeps the certificate valid regardless of the unknown correlation. We prove that PROBE is \delta -PAC and attains the known-correlation oracle sample complexity up to a constant multiplicative factor and a constant additive calibration cost. The guarantee extends to the (\epsilon,\delta) -PAC setting under minimal changes to the algorithm. Numerical experiments on synthetic instances and on an auto-loan pricing replay with large language model and tabular proxies confirm that the sample savings of PROBE scale with the strength of the reward-proxy correlation, exactly as the theory predicts.
[LG-43] Generative Diffusion Models of Stochastic Graph Signals
链接: https://arxiv.org/abs/2607.06833
作者: Yiğit Berkay Uslu,Samar Hadou,Sergio Rozada,Shirin Saeedi Bidokhti,Alejandro Ribeiro
类目: Machine Learning (cs.LG); Signal Processing (eess.SP)
*备注: Submitted to and under review by IEEE Transactions on Signal Processing
Abstract:Sampling stochastic signals supported on a graph underlies many graph machine learning tasks, including recommender systems, forecasting in financial markets, and wireless network optimization. In these settings, the target signals are realizations of unknown conditional distributions. However, prevailing approaches rely mostly on intricate, application-tailored designs that often regress to a conditional mean instead of sampling from the conditional law. This paper unifies such problems as conditional graph signal generative modeling and tackles them with a single denoising diffusion framework. We learn a reverse diffusion process, parametrized by graph neural networks (GNNs), that draws graph signals conditioned directly on the graph topology and on node-feature side information. The reverse process is realized by a novel architecture, the U-Graph Neural Network (U-GNN), which generalizes the image-convolutional U-Net to graph-structured signals. The U-GNN performs multi-resolution encoder–decoder processing in which pooling and unpooling reduce to a learned node selection, expressed by nested selection matrices, and a zero-padded lifting of coarse signals back to the full node set. The graph convolutions are carried out on the original graph, with a stride that sets their hop reach, so the U-GNN bypasses explicit graph coarsening at every resolution. We demonstrate our method on two generative tasks: stock price forecasting and optimal wireless resource allocation, with extensive numerical results in both domains.
[LG-44] CaLiSym: Learning Symplectic Dynamics of Real-World Systems through Structured Canonical Lifts
链接: https://arxiv.org/abs/2607.06824
作者: Aristotelis Papatheodorou,Pranav Vaidhyanathan,Natalia Ares,Ioannis Havoutis,Gerard J. Milburn
类目: Robotics (cs.RO); Machine Learning (cs.LG)
*备注: 18 pages, 4 figures, 5 tables
Abstract:Physics-informed learning promises data-efficient and stable dynamics prediction, yet its strongest geometric guarantees have largely remained confined to closed conservative systems. This excludes many robotic systems of practical interest, where actuation, dissipation, and constraints continuously exchange energy and momentum with the environment. We introduce CaLiSym, a lightweight framework that extends exact symplectic learning to such systems by changing where the geometric prior is imposed. Rather than enforcing symplecticity on the measured physical state, CaLiSym embeds the state and its physical ports into a structured lifted canonical phase space, where the learned dynamics evolve through an exactly symplectic map. The lift is explicit and algebraic, requiring neither recurrent latent states, transformer decoders, implicit optimization, nor inference-time ODE integration. We instantiate the framework with generalized-ridge SympNet predictors and introduce GRB-SympNet, a B-spline variant that combines local approximation with exact symplectic structure. Experiments on a controlled dissipative double pendulum, a real-world quadrotor, and a contact-rich quadruped demonstrate consistent improvements in out-of-distribution autoregressive prediction while using parameter-efficient models. At the same time, the learned lifted dynamics preserve the symplectic form to numerical precision. These results show that symplectic learning can be extended beyond conservative mechanics through structured canonical lifts, enabling geometry-preserving dynamics models for real-world robotic systems.
[LG-45] On Explicit Super-Expressive Approximation for Neural Networks
链接: https://arxiv.org/abs/2607.06781
作者: Feng-Lei Fan,Ze-Yu Li,Chen-Yu Wang,Jian-Jun Wang
类目: Machine Learning (cs.LG)
*备注: 44 pages, 4 figures
Abstract:In this work, we investigate the fixed-architecture neural network approximation with explicit parameter bounds and elementary activations. While prior work demonstrated super-expressive approximation using fixed-size networks, they lack quantitative and non-asymptotic characterizations of parameter magnitude with respect to the approximation error. We resolve this issue by introducing the Chinese Remainder Theorem as a constructive encoding mechanism. For Lipschitz continuous functions on [0,1]^D , we construct a width- \max\D,4\ , depth- 5 network with explicit parameter-error trade-offs. For Hölder-smooth functions in C^r,\gamma_A\left([0,1]^D\right) , our fixed network of width \max\2D,\ D+5N+1\ and depth r + 9 achieves the parameter magnitude \mathcalP bounded by \log_2 \mathcalP=\mathcalO\bigl(\varepsilon^-2D/(r+\gamma)\log(1/\varepsilon)\bigr) . This is the dual result compared to those in the parameter-bounded and architecture-unbounded paradigm.
[LG-46] Efficient Bayesian Deep Ensembles via Analytic Predictive Inference
链接: https://arxiv.org/abs/2607.06776
作者: Sina Aghaee Dabaghan Fard,Marie Maros,Jaesung Lee
类目: Machine Learning (cs.LG); Methodology (stat.ME)
*备注:
Abstract:We introduce an efficient Bayesian deep ensemble method for predictive regression designed to enhance interpretability while maintaining competitive predictive performance and computational efficiency. Our method combines the statistical rigor of Bayesian inference with the scalability of deep ensembles, providing calibrated uncertainty estimates that enable its use not only for standalone prediction but also as a component within broader learning systems. To achieve these goals, our work relies on three key design components: (i) low-dimensional ensemble representation: predictions are expressed as a combination of a small number of trained neural predictors, enabling scalable inference whose cost depends on ensemble size rather than dataset size; (ii) closed-form Bayesian aggregation: ensemble predictions are combined using Bayesian linear regression, yielding interpretable posterior weights and calibrated uncertainty without approximate inference; and (iii) Independent ensemble training: multiple neural networks are trained separately, producing diverse predictive representations that improve robustness and uncertainty calibration. Empirical results on standard regression benchmarks demonstrate that the proposed approach achieves competitive predictive performance while maintaining reliable uncertainty estimates across settings.
[LG-47] Efficient Long-Horizon Learning for Learned Optimization
链接: https://arxiv.org/abs/2607.06772
作者: Xiaolong Huang,Benjamin Thérien,James Harrison,Eugene Belilovsky
类目: Machine Learning (cs.LG)
*备注: Meta-learning, learned optimization
Abstract:Learned optimization aims to improve upon hand-designed optimizers (e.g., Adam and Muon) by meta-learning small neural network optimizers over a distribution of tasks. While recent work has greatly advanced the architectural design and inductive biases of learned optimizers (LOs), current meta-training approaches still suffer from two main difficulties: (1) they cannot efficiently scale meta-training to long-horizon inner problems and (2) they often fail to surpass comparable hand-designed optimizers. To address these limitations, we propose Efficient Long-hOrizon (ELO) learning, an efficient meta-training algorithm that (1) reallocates redundant meta-training compute to longer failure regimes, achieving efficient long-horizon learning, and (2) enforces decoupled progressive expert supervision, providing stable meta-learning signals that additionally improve the generalization of LOs. Our empirical study evaluates ELO for meta-training both element-wise and matrix-based LOs. Across downstream language modeling (GPT-2-124M/350M on FineWeb) and image classification (ViT-B/16, ResNet-50 on ImageNet-1K) tasks, ELO substantially improves the long-unroll performance and out-of-distribution generalization of the base LOs. In particular, ELO-Celo2 consistently outperforms well-tuned AdamW across all evaluated tasks, while remaining competitive with Muon on language modeling. \textitNotably, all ELO baselines require less than 7 H100 GPU-hours for meta-training.
[LG-48] Deployment Risk Assessment Using Diff-Aware Features: A Case Study at Prime Video
链接: https://arxiv.org/abs/2607.06766
作者: Mayur Kurup,Hyunjae Suh,Swathi Vaidyanathan,Pranesh Vyas,Srinidhi Madabhushi,Yegor Silyutin
类目: oftware Engineering (cs.SE); Machine Learning (cs.LG)
*备注:
Abstract:At Amazon Prime Video, we face the critical operational challenge of managing code deployments during live events and rapid feature releases without causing service outages. Current change control approaches use blanket deployment freezes that block all changes regardless of risk, creating significant developer toil. While prior research has explored risky change predictors, these rely on developer-specific metadata or extensive historical data, raising privacy concerns and limiting applicability to new projects. We introduce a framework centered on diff-aware features, characteristics derived directly from code modifications. Our key contribution is the systematic identification of which quantitative metrics (code-level and change-level metrics) and qualitative indicators (coding style violations, change type classification) are necessary for risk prediction. We employ LLMs as multi-language feature extractors, demonstrating their effectiveness for code analysis beyond generation tasks and eliminating the need for language-specific tooling. We evaluated our framework on two datasets: Prime Video’s production environment and the public ApacheJIT dataset. Our best-performing model achieves an average recall of 0.83 and F1 score of 0.81 across both datasets for detecting risky code changes. Notably, ablation analysis reveals that change-level volume metrics (e.g., lines added/deleted) are noisy predictors, while structural code complexity provides a substantially stronger risk signal. These results demonstrate that thoughtful feature curation enables effective change risk assessment across different programming languages and organizational contexts while avoiding privacy concerns.
[LG-49] Macroeconomic Message Passing for Anticipating Foreign Exchange Regime Changes: A Deep Logical Learning Approach using Graph Tsetlin Machines
链接: https://arxiv.org/abs/2607.06719
作者: Christian Blakely,Melanie Gilmore
类目: Computational Engineering, Finance, and Science (cs.CE); Machine Learning (cs.LG)
*备注:
Abstract:This paper introduces a graph-theoretic approach for predicting market regimes in foreign exchange (FX) currency prices. Specifically, the proposed model incorporates exogenous macroeconomic variables to update localized node features via message-passing operations. Utilizing the Graph Tsetlin Machine (GraphTM) framework, we empirically demonstrate the efficacy of this approach in anticipating market regimes for the US Dollar and Japanese Yen currency pair (USD/JPY). By representing multivariate macroeconomic drivers and technical indicators as hypervectorized directed multigraphs, the GraphTM leverages structured message passing to construct deep, interpretable logical clauses capable of recognizing complex sub-graph patterns.
[LG-50] Robust Human-AI Complementarity under Uncertainty ICML2026
链接: https://arxiv.org/abs/2607.06656
作者: Yewon Byun,Bryan Wilder
类目: Machine Learning (cs.LG)
*备注: ICML 2026
Abstract:Machine learning models are often intended to augment rather than replace human decision makers, by providing information that is complementary to human judgement. Yet, in practice, human decision makers routinely fail to realize such complementary gains, even when models provide useful signal. In this work, we study how asymmetric information about the quality of information available to a human decision maker vs. an AI impacts the ability of a decision maker to extract complementary value from AI predictions. We show that a key factor is the error correlation structure between human and AI predictions. In particular, when the AI’s prediction errors are \textitnegatively correlated with those of the human, the decision maker can construct robust strategies which guarantee improvements in expected utility. We empirically investigate whether these conditions for complementarity arise in practice, using real-world forecasting benchmarks. Comments: ICML 2026 Subjects: Machine Learning (cs.LG) Cite as: arXiv:2607.06656 [cs.LG] (or arXiv:2607.06656v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2607.06656 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Yewon Byun [view email] [v1] Tue, 7 Jul 2026 17:58:06 UTC (2,001 KB)
[LG-51] Pelican-VLA 0.5: Attending Before Acting Benefits Generalization
链接: https://arxiv.org/abs/2607.06655
作者: Zeyuan Ding,Wenhai Liu,Yang Xu,Jiayu Hu,Yinda Chen,Yi Zhang,Yong Dai,Jian Tang,Xiaozhu Ju
类目: Robotics (cs.RO); Machine Learning (cs.LG)
*备注:
Abstract:In this report, we present Pelican-VLA 0.5, a unified VLA model that integrates vision-language understanding, future-frame generation, and action prediction within a single architecture. Pelican-VLA 0.5 achieves attention-level generalization: without object annotations, segmentation masks, attention supervision, or task-specific fine-tuning, its action pathway already focuses on the instruction-relevant object and contact region. This behavior persists across unseen scenes and unseen robot embodiments, and is substantially stronger than in other open-source VLA baselines. We verify that this ability originates from the learnable Reasoning Slots inserted between perception and action: by routing task-relevant visual information through a compact bottleneck, the slot interface induces manipulation-centric attention during pre-training and remains effective across different policy structures, including a MoT-style architecture.
[LG-52] Dual Attention Heads for Personalized Federated Learning in ECG Classification
链接: https://arxiv.org/abs/2607.06653
作者: Kien Le,Joseph Lindley,Quoc Bao Phan,Tuy Tan Nguyen
类目: Machine Learning (cs.LG)
*备注:
Abstract:Federated learning (FL) enables collaborative model training across institutions without sharing sensitive patient data. However, the inherent heterogeneity of electrocardiogram (ECG) data across healthcare providers presents significant technical challenges for robust classification. We propose FedDualAtt, a personalized federated learning approach that splits transformer attention heads into global and local branches. Global heads are aggregated via FedAvg to capture shared cross-site patterns, while local heads remain client-specific to adapt to institution-level recording characteristics. Experiments on FedCVD, an FL benchmark for cardiovascular disease detection, demonstrate that FedDualAtt outperforms existing FL and personalized FL methods in ECG classification tasks. Analysis of global-local head ratios reveals that different clients benefit from varying levels of architectural personalization.
[LG-53] From Jumps to Signatures: a Generative Method for Temporal Point Processes
链接: https://arxiv.org/abs/2607.06652
作者: Niels Cariou-Kotlarek,Vasileios Lampos
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 44 pages. Source code and datasets: this https URL
Abstract:Rough path signatures are a universal feature map for continuous paths and, via the expected signature, characterise path distributions. These guarantees do not directly extend to cadlag paths of Temporal Point Processes (TPPs), limiting the use of signature methods for event sequences. Furthermore, neural TPP models, including recent generative approaches, optimise per-event objectives with no global sequence-level loss, while evaluation of variable-length event sequences lacks distributional discrepancy measures. This paper proposes a common pathwise framework for addressing these limitations. We introduce the interarrival embedding, a stable, injective lift from jump paths to continuous paths of bounded variation, extending signature methods to discrete event sequences. Our theoretical contributions give rise to sigTPP, the first signature-based generative model for TPPs, trained using a path-level loss on complete trajectories. We further analyse the space of counting paths and derive three distributional discrepancies, providing mathematically justified tools for evaluating generative TPP models. Across synthetic and real-world datasets, sigTPP achieves the best average rank based on eight complementary metrics, outperforms or is within a standard error of the strongest baseline in 64% of the dataset-metric pairs, and according to a relative score, improves against every baseline by at least 19% on average.
[LG-54] Entropy-Guided Tensor Compression for Multimodal Federated Learning on Edge Devices
链接: https://arxiv.org/abs/2607.06651
作者: Quoc Bao Phan,Tuy Tan Nguyen
类目: Machine Learning (cs.LG)
*备注:
Abstract:Federated learning (FL) over mobile and edge devices increasingly involves multimodal models in which clients differ in both sensing capability and computational capacity. Existing update compression schemes typically apply uniform policies across layers and devices, without accounting for modality-specific differences in spectral structure and compressibility. We propose MESH-FL, an entropy-guided matrix product state (MPS) update-compression framework for modality-heterogeneous FL on resource-constrained devices. MESH-FL estimates the spectral entropy of each layer-wise update via truncated singular value decomposition and allocates MPS compression ranks adaptively across layers, modalities, and devices under per-client payload budgets. We show that higher spectral entropy necessitates a higher reconstruction rank under the majorization order on singular-value energy distributions. Building on this result, we prove that the proposed entropy-guided allocation solves a convex surrogate rank-allocation problem, preserves monotonicity under the exact payload model, and achieves convergence with an explicit compression-dependent error term. Experiments on a 15-node heterogeneous Raspberry Pi~4/5 cluster with modality-heterogeneous clients show that MESH-FL achieves up to 56.8\times compression while surpassing the uncompressed FedAvg baseline in final accuracy by up to 2.01%, and reduces total transmitted data to reach convergence by up to 66\times .
[LG-55] POPS: Recovering Unlearned Multi-Modality Knowledge in MLLM s with Prompt-Optimized Parameter Shaking
链接: https://arxiv.org/abs/2607.06649
作者: Zhangheng LI,Jianing Zhu,Junyuan Hong,Sungmin Eum,Shuowen Hu,Suya You,Zhangyang Wang
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注:
Abstract:Multimodal Large Language Models (MLLMs) have demonstrated impressive performance on cross-modal tasks by jointly training on large-scale textual and visual data, where privacy-sensitive examples could be unintentionally encoded, raising concerns about privacy or copyright violation. To this end, Multi-modality Machine Unlearning (MMU) was proposed as a mitigation that can effectively force MLLMs to forget private information. However, the robustness of such unlearning methods is not fully exploited when the model is published and accessible to malicious users. In this paper, we propose a novel adversarial strategy, namely Prompt-Optimized Parameter Shaking (POPS), aiming to recover the supposedly unlearned multi-modality knowledge from the MLLMs. Our method elicits the victim MLLMs to generate potential private examples via prompt-suffix optimization, and then exploits these synthesized outputs to fine-tune the models so they disclose the true private information. The experiments on the different MMU benchmarks reveal substantial weaknesses in the existing MMU algorithms. Our POPS can even achieve a near-complete recovery of supposedly erased sensitive information on the unlearned MLLMs, exposing fundamental vulnerabilities that challenge the foundational robustness of representative MMU-based privacy protections.
[LG-56] ORAN-DEFEND: Subspace Detection and Sanitization of Backdoor DRL xApps in Open RAN
链接: https://arxiv.org/abs/2607.06647
作者: Md Raihan Uddin,Fatemeh Lotfi,Tolunay Seyfi,Fatemeh Afghah
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注: 6 pages, 1 figure
Abstract:Open Radio Access Networks (O-RAN) increasingly delegate near-real-time control to deep reinforcement learning (DRL) xApps obtained from third-party vendors, creating a new supply-chain attack surface. A backdoor policy behaves optimally until an adversary injects a covert trigger into the observed key performance indicator (KPI) telemetry, at which point it issues harmful control actions that degrade quality of service (QoS). We present ORAN-DEFEND, a retraining-free wrapper that sanitizes a frozen, potentially compromised xApp by projecting each KPI window onto a safe subspace estimated from a small number of trusted clean rollouts via singular value decomposition (SVD). We establish, both analytically and empirically, a precise recovery condition: the defense succeeds if the trigger energy concentrates in the orthogonal complement of the safe subspace, and we quantify this boundary through the trigger’s \Eperp energy fraction. On the Colosseum COLORAN dataset, we evaluate four structurally distinct DRL backdoor attacks, like TrojDRL, SleeperNets, BadRL, and Q-Incept, spanning inner-loop and outer-loop poisoning regimes and demonstrate 100% return recovery and \geq99.5% defense success rate across all four when the subspace assumption holds. A geometry ablation reveals an intrinsic and previously uncharacterized limit of any linear projection defense: when the trigger collocates with the legitimate signal, the \Eperp energy fraction governs recovery monotonically, and the linear residual detector collapses to chance even while a nonlinear classifier retains perfect separability.
[LG-57] he Power of Backdoor Absorption in Community Training
链接: https://arxiv.org/abs/2607.06643
作者: Issam Seddik,Sami Souihi,Mohamed Tamaazousti,Sara Tucci Piergiovanni
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注: Technical Report, CEA-LIST. 15 pages, 5 figures, 2 tables
Abstract:Backdoor attacks severely threaten large-scale AI models. When model owners delegate training to external compute providers within a decentralized training paradigm, adversaries can craft stealthy, low-frequency triggers to inject malicious behavior while evading standard audits. Traditionally, detecting these attacks requires a full re-computation of the training steps–a prohibitive overhead that directly contradicts the owner’s resource constraints. To address this, we investigate the resilience of continuous optimization dynamics under Byzantine perturbations, where adversaries are forced to compete against a continuous influx of honest updates. Under a threat model where an adversary compromises f out of n total trainers, we quantify the minimum auditing overhead required by the model owner to probabilistically bound the attack success rate. We formalize this injection-absorption dynamic as a Discrete-Time Markov Chain (DTMC). Using this framework, we prove that the success probability of any bounded adversary asymptotically collapses to zero under a defense strategy combining natural absorption, a randomized scheduler, and lazy verification oracle. Empirical results demonstrate significant backdoor suppression with zero utility degradation even when invoking the verification oracle on merely 10% of the total training steps. This approach yields a provably sound and computationally efficient defense for safety-critical AI.
[LG-58] he Approximation Ratio for the Risk of Myopic Bayesian Active Learning for Linear Regression
链接: https://arxiv.org/abs/2607.06642
作者: Stephen Mussmann
类目: Machine Learning (cs.LG)
*备注:
Abstract:Active learning studies the fundamental question: what data should we choose to observe? The greedy algorithm in optimal experiment design is a common heuristic and also equivalent to myopic Bayesian active learning for linear regression, the common framework where long-term planning is replaced with the one-step optimal choice. In this work, we prove a first-of-its-kind approximation ratio for the greedy algorithm’s risk that is tight up to an absolute constant. The approximation ratio is linear in the maximum initial leverage score (MILS), a newly identified quantity fundamental to the greedy algorithm’s performance. Finally, we illustrate the results with simple numerical simulations.
[LG-59] UASPL: Uncertainty-Aware Self-Paced Learning with Evidential Neural Networks
链接: https://arxiv.org/abs/2607.06638
作者: Yifan Zhang,Yuxin Hu,Zhuobin Hao,Xiaozhuan Gao,Lipeng Pan
类目: Machine Learning (cs.LG)
*备注:
Abstract:Self-paced learning (SPL) is an effective learning paradigm that simulates the human learning process by progressing from easy to difficult samples based on the value of the loss function during the learning process. It has shown great potential in improving model performance and training efficiency. However, the prediction results of samples with smaller loss values are not necessarily reliable, indicating that such samples are not always simple samples for the model. Hence, this article proposes an uncertainty-aware self-paced learning based on evidential neural networks, termed UASPL, which integrates predictive reliability into sample selection through a general loss function within the Subjective Logic framework. This loss function incorporates uncertainty estimation and can be extended to different variants of SPL. Moreover, this loss function couples a sample selection preference, thereby ensuring the interpretability of the sample selection process. Finally, the experimental results on multiple datasets show that UASPL outperforms other SPL methods in terms of classification performance, interpretability, and generality. The source code is available at: this https URL.
[LG-60] Optimized Instance Alteration for Explaining and Assessing Robustness of Classifiers
链接: https://arxiv.org/abs/2607.06637
作者: Evgenii Kuriabov,David Miller,Jia Li
类目: Machine Learning (cs.LG); Methodology (stat.ME)
*备注:
Abstract:In this work, we propose a unified approach for diagnosing misclassification and assessing the robustness of black-box classifiers. Central to our method is an optimization framework that modifies an instance so that the classifier predicts a specified target label, while ensuring that the modification remains easily explainable. The objective function contains two components: an explainability-aware L_0 (XA- L_0 ) penalty that promotes sparse and interpretable modifications, and a classifier loss objective that steers the perturbed instance toward the desired output. This integrated optimization formulation is used both to identify the underlying causes of misclassification and to evaluate robustness by determining how an instance can change within a tolerance region before being reassigned to another class. To quantify robustness, we introduce the Tolerance Region Confusion Matrix (TOR-Confusion Matrix), which measures a classifier’s susceptibility by modeling the class-to-class transition probabilities induced by tolerance-bounded perturbations. We validate the proposed method on both image and tabular datasets, demonstrating its ability to jointly deliver interpretability and robustness assessment.
[LG-61] When Do Geometric Algebra Layers Beat Scalarization? A Controlled Study on SO(3)-Equivariant Vector Laws
链接: https://arxiv.org/abs/2607.06634
作者: Fabien Polly
类目: Machine Learning (cs.LG)
*备注: 10 pages, 2 figures, 4 tables. Code and data: this https URL
Abstract:Compact networks built from Clifford algebra Cl(3,0) primitives are exactly SO(3)-equivariant and learn synthetic 3D vector laws from few samples. We ask whether the geometric algebra structure itself contributes anything beyond exact equivariance. We compare against a minimal scalarization baseline: invariant dot products fed to a small MLP that outputs coefficients on the equivariant basis v_i, v_i x v_j, which is also exactly equivariant. On single-stage laws (rotation by axis-angle, cross product, central force), scalarization matches or beats the Cl(3,0) network at a fraction of the training cost, so the geometric algebra adds nothing there. On compositional targets whose computation graph nests group operations (apply R2 R1 to a point; map a local force through an orientation, then take a torque), the Cl(3,0) network beats scalarization by an order of magnitude in the low-data regime, reaching with 100 samples what the baseline needs 3000 for, and the gap survives strengthening the baseline with the triple-product invariant and 17x more parameters, external Vector Neurons and e3nn baselines, and a multiplicative coefficient network. Ablations show the required network depth tracks the rotation chain length, and scalarization falls below the constant predictor on chains of four rotations. The advantage is not composition per se: on a rotation-free nested cross product, which flattens into polynomial invariant coefficients, scalarization wins by 24x. No tested model, equivariant or not, extrapolates invariant magnitudes: on radius and separation shifts every model is worse than a constant predictor once errors are normalized. We conclude that geometric algebra layers are not a general shortcut for low-data 3D learning, but become useful precisely when the target composes group elements in depth.
[LG-62] Does Demand Response Increase Vulnerability to Cyber Attacks by Adversarial Data Modifications?
链接: https://arxiv.org/abs/2607.06632
作者: Clemens Kortmann,Eike Cramer
类目: Machine Learning (cs.LG)
*备注: 17 pages, 10 figures,
Abstract:Adversarial attacks are crafted data manipulations that aim to deteriorate the outcomes of prediction or decision-making algorithms. In the energy systems literature, adversarial attacks have been studied with a focus on problems regarding the electricity grid. Such problems include forecasting and grid state estimation, where adversarial attacks are also known as false data injection attacks. Only few studies have analyzed the potential impact that adversarial attacks have on the demand side. We analyze how manipulated price forecasts impact the decision-making in industrial demand response. To this end, we design adversarial attacks that aim to deteriorate the output of electricity price forecasting models and solve scheduling optimization problems of energy-intensive production processes using the distorted price forecasts. We make use of a generalized process model to investigate the vulnerability to adversarial attacks for a range of production scheduling problems with different levels of process flexibility. We find that adversarial attacks can erode the profits gained from demand response. However, when perturbations are limited in extent (so that they are hard to detect by the human user), demand response preserves about 90% of its financial advantage compared to steady-state process operation. Further, we find that the impact of adversarial attacks on demand response does not only depend on the magnitude of the perturbations but rather on the orientation of the adversarial perturbations. Therefore, we argue that attack analyses should explicitly incorporate the sensitivities of scheduling optimization models into the attack design to enable more rigorous assessments of decision-making under adversarial attacks.
[LG-63] When Certificates Fail: A Unified Safety Framework for Embedded Neural Interface Models
链接: https://arxiv.org/abs/2607.06630
作者: Jasmeet Singh Bindra
类目: Machine Learning (cs.LG); Signal Processing (eess.SP)
*备注: 12 pages; To be published in peer-reviewed journal
Abstract:Formal robustness certificates for embedded neural-interface models can pass while task accuracy collapses: at perturbation budget e=0.25, EEGNet classification accuracy drops by 25.7% under projected-gradient attack while the Lipschitz-style certificate remains valid for all 9 tested subjects. We argue that this gap between mathematical certification and operational safety is one instance of a broader alignment failure in neural interfaces, where training objectives diverge from user welfare. We propose a unified empirical audit framework organized around three such failures: verification insufficiency, in which certificates pass while task behavior degrades; proxy-fidelity divergence, in which task-optimized representations damage neural signal structure (a time-domain auxiliary objective reduces reconstruction MSE by 0.1132 while worsening spectral log-MSE); and latent information exfiltration, in which public-task embeddings retain private attributes (subject identity recoverable at 48.1% versus 6.7% chance). We instantiate the framework on BCI Competition IV 2a and SEED-IV using multiple deep and classical EEG decoders, official session-level validation, null controls, and paired statistical tests. The verification gap persists across EEGNet, CSP+LDA, and FBCSP+LDA, and is therefore architecture-independent. Our results establish that operational safety auditing, not certificate verification alone, is necessary for responsible neural-interface deployment.
[LG-64] STST-JEPA: Shallow-Target Spatio-Temporal Joint Embedding Prediction Architecture For EEG Self-Supervised Learning
链接: https://arxiv.org/abs/2607.06629
作者: Roy Segal,Yoni Svechinsky,Tomer Fekete
类目: Machine Learning (cs.LG); Neurons and Cognition (q-bio.NC)
*备注:
Abstract:Brain age – the age inferred from a physiological recording – is an emerging biomarker whose deviation from chronological age tracks neurological and psychiatric burden, and EEG is an attractive substrate for it because it is cheap, portable, and temporally rich. Yet EEG brain-age models must contend with cross-site montage heterogeneity, small labelled cohorts, and dominant subject-level non-stationarity, and few EEG foundation models have been shown to deliver competitive age regression across the full pediatric-to-older-adult range in which such a biomarker would actually be deployed. We introduce STST-JEPA, a self-supervised transformer for resting-state and task EEG, pretrained on 47,703 sessions spanning ages 5-81 from the this http URL and Healthy Brain Network (HBN) corpora. The model combines a latent-prediction objective - predicting masked-token representations against an EMA-of-tokenizer target - with an auxiliary signal-reconstruction term, applied to 30-second multi-channel windows under spatiotemporal block masks. A lightweight attentive probe trained on frozen pretrained embeddings achieves a best held-out-validation mean absolute error of 3.06 years (r = 0.924) for age regression on 3,367 sessions, against a predict-the-mean baseline of approximately 10 years MAE. With light task-specific fine-tuning of the model’s final layers, the same pretrained encoder achieves rank-1 placements - with the model’s native 30-second windows - on the public NeuralBench x this http URL EEG leaderboard for sex classification (balanced accuracy 0.911), age prediction (r = 0.749), and psychopathology composite regression (r = 0.215). We further show that the model’s age-prediction residual is negatively correlated with cognitive efficiency over several tasks we examined.
[LG-65] Reward Valuation in Vision Language Models: Causal Mechanisms Underlying Anhedonia
链接: https://arxiv.org/abs/2607.06626
作者: Melika Honarmand,Samin Mahdipour Aghabagher,Martin Schrimpf
类目: Machine Learning (cs.LG); Neurons and Cognition (q-bio.NC)
*备注:
Abstract:Recent Vision-Language Models capture increasingly complex aspects of human cognition. Here we ask whether this alignment extends to reward valuation, which we assess in a mechanistic framework built on clinical tests that were developed to evaluate anhedonia and motivational deficits in major depressive disorder. In the brain, anhedonia is frequently linked to dysregulation in the Nucleus Accumbens (NAc) and the broader dopaminergic reward system. While neuroimaging has localized these deficits, establishing a causal link between NAc activity and specific behavioral symptoms remains a challenge. We use these ideas from neuroscience to functionally identify reward-anticipatory units in vision language models, and test their causal role via targeted perturbations. Perturbing NAc-selective units induces behavioral effects that mirror human anhedonia: the model shifts toward low-effort, low-reward options in effort-based decision-making tasks. Crucially, our results reflect a specific deficit in reward valuation and anticipation rather than a loss of task capability: the perturbed model maintains baseline performance when reward-based choice is removed. This induced vulnerability further aligns with clinical anhedonia and motivation scales, including DARS and MAP-SR. Taken together, these results reveal reward valuation circuits in AI models that parallel those in humans.
[LG-66] Creating Power Distribution Network Layouts Using Generative Adversarial Networks and Image-Based Representations
链接: https://arxiv.org/abs/2607.06622
作者: Juan Manuel Garcia-Perez,Carlos Mateo
类目: ystems and Control (eess.SY); Machine Learning (cs.LG)
*备注: 8 pages
Abstract:Utilities increasingly rely on planning and operational tools to cope with the increased penetrations of distributed energy resources, yet the lack of realistic, openly available datasets remains a major barrier for benchmarking and comparison. Traditional test feeders, and recently proposed large-scale synthetic networks alleviate this issue but are typically based on heuristic rules and do not learn directly from data. This paper proposes a generative framework based on Generative Adversarial Networks (GANs) to create power distribution network layouts using image-based representations. The model is trained on rasterised views of distribution systems and can operate in two modes: an unconditional configuration that learns layout patterns from the training dataset, and conditional configurations that incorporate geographical context such as street maps and the spatial distribution of consumers. The methodology includes dataset preparation from Geographic Information System (GIS) sources, GAN architecture design, and the analysis of training stability and image resolution. Results from three representative cases show that the proposed approach can reproduce the topologies of low (LV), medium (MV) and high voltage (HV) feeders and align generated layouts with underlying geographical structures. At the same time, the study reveals limitations related to training stability, resolution-dependent artefacts and limits, and the absence of explicit electrical constraints. The proposed framework constitutes a data-driven complement to existing synthetic network generation methods, and could be applied to propose distribution network layouts for the electrification of new areas. This would require future extensions towards power flow, electrically validated models.
[LG-67] Fingerprint Not Blueprint: How Positional Schemes Set the Default Spectral Algebra of Attention
链接: https://arxiv.org/abs/2607.06621
作者: Li Hengyu(Institute for Solid State Physics, The University of Tokyo)
类目: Machine Learning (cs.LG); Disordered Systems and Neural Networks (cond-mat.dis-nn)
*备注: 16 pages, 5 figures, 4 tables. Code and data: this https URL
Abstract:The pre-softmax score of an attention head is a bilinear form score(i,j) = x_i^T M x_j in a learned operator M = W_q^T W_k . Because M is generally non-symmetric, hence non-normal, it has a complex eigenspectrum and non-orthogonal eigenvectors, the regime where non-Hermitian and random-matrix tools apply. We ask what this spectrum encodes, at three levels for previous-token and induction circuits. Statically, across seven pretrained models spanning three positional schemes, the strongest previous-token heads are spectrally rotational under RoPE and non-rotational, or content-like, where position enters outside QK (learned-absolute and ALiBi); the model-level separation is perfect at every top-k examined (exact permutation p=0.029 ), and zeroing the per-frequency RoPE phase Im(M_t) eliminates induction on a pre-identified previous-token head in all three RoPE models. Dynamically, over public Pythia checkpoints every head originates at the random-matrix (Ginibre) null; the rotational signature emerges with the behavior, not before it, and the population-median suppression that yields the final profile follows circuit formation, so the profile is a consolidated fingerprint, not a precursor. Causally, and at toy scale, no spectral channel is necessary: constrained two-layer training reroutes around every ban with capability intact, albeit at a significant formation delay (four pre-registered contrasts, q_BH = 0.016 ). The cost structure exposes each scheme’s default: imposing symmetry slows learned-absolute models by a factor of 2.9, whereas a RoPE head with a fully symmetric static M still routes directionally via the phase channel, impossible under absolute positions. Within the settings examined, the positional scheme sets the default spectral algebra of an attention head’s solution: a fingerprint sculpted after function, not a hard constraint upon it.
[LG-68] HiFuzz: Hierarchical Reinforcement Learning for Semantic-Aware and Adaptive CPU Fuzzing
链接: https://arxiv.org/abs/2607.06619
作者: Ya Wang,Hanwei Fan,Zhenguo Liu,Xiaofeng Zhou,Yangdi Lyu,Jiang Xu,Wei Zhang
类目: Hardware Architecture (cs.AR); Machine Learning (cs.LG)
*备注: Accepted by the 2026 IEEE International Test Conference (ITC 2026)
Abstract:Modern processor verification struggles to reach deep architectural states due to the inefficiencies of traditional mutation-based fuzzing. We propose HiFuzz, a novel hierarchical reinforcement learning framework that replaces mutation with a structured, two-layer generation process: a Program Agent for global layout and a Basic Block Agent for precise instruction filling. To overcome reward sparsity, HiFuzz integrates an adaptive coverage reward mechanism and a semantic-aware basic block encoder providing intrinsic feedback. Extensive evaluations on three real-world RISC-V cores demonstrate that HiFuzz significantly outperforms state-of-the-art fuzzers in coverage and bug detection.
[LG-69] A Quiet Failure in Calibrated Virtual Screening: Marginal Conformal Prediction Under-Covers the Minority Class and a Class-Conditional Fix Recovers It
链接: https://arxiv.org/abs/2607.06605
作者: Muhammadjon Tursunbadalov(1),Mustafojon Tursunbadalov(1) ((1) School of Science and Technology, Champions College Prep, United States)
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 16 pages, 6 figures
Abstract:Conformal prediction is being adopted in drug discovery to put an honest number on model reliability: pick an error rate alpha, and the method returns prediction sets containing the true label with probability at least 1 - alpha. We show this guarantee can be dangerous on imbalanced datasets. Across four datasets, standard (marginal) conformal prediction hits its global 90% coverage target while leaving the minority class badly exposed: realized minority coverage falls to 64.8% on blood-brain-barrier penetration and to 4.2% on clinical-trial toxicity, where the rare class is nearly abandoned. The failure is not tied to one model: a random forest, a graph network, and a frozen chemical language model all reproduce it (p 0.001 in every case), with severity tracking baseline calibration on rare labels rather than architecture. A conservation identity explains the effect: the minority’s shortfall equals the majority’s surplus amplified by the imbalance ratio, predicting the measured gap to within one point and ordering severity across datasets. The failure survives realistic scaffold splits and a second conformal score, while aggregate accuracy and overall coverage stay reassuringly high, which is exactly why it is easy to miss. Class-conditional (Mondrian) conformal prediction closes the gap on every dataset, restoring minority coverage to target for a modest increase in prediction-set size. We localize the failures to generic molecular scaffolds - plain benzene and pyridine cores occurring in both classes - propose a one-number diagnostic, and show with a cost model that abstaining on affected compounds flips a screening campaign from net-negative to net-positive utility. Our contribution is demonstrating on real chemistry how severe and invisible this known conformal-theory gap becomes under imbalance, and laying out a practical protocol restoring per-class reliability.
[LG-70] Calibration-Family Overfit: Why Trusted Sabotage Monitors Dont Transfer Across Lineages
链接: https://arxiv.org/abs/2607.06596
作者: Lucas Pinto
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注: 28 pages, 11 figures
Abstract:Trusted monitoring is a central defense in AI control: a cheaper trusted model scores an untrusted model’s actions for sabotage, and the most suspicious are audited or deferred. Such monitors are evaluated against one or two untrusted models, and the accuracy is reported as a property of the monitor. We ask whether it is partly a property of the pairing. We make the untrusted policy family the controlled axis: we fit a monitor on family A’s transcripts, apply it to family B, and decompose the cross-family AUROC into how obvious each family’s sabotage is, how capable each monitor is, and the residual own-family advantage after both are removed: the interaction. On code-backdoor transcripts the interaction is positive and survives the dominant confounds: +0.172 (95% CI [+0.158, +0.185]) on four open-weight families on a strict leak-free basis, holding across an eight-family matrix spanning four frontier vendors, blatancy-matched controls, a trained transcript-reading monitor up to 34B parameters, and a second code benchmark. The gap is geometric: each lineage’s sabotage occupies a different scoring direction, and rotating the monitor’s scoring direction toward the target lineage recovers most of the lost detection while a matched random rotation recovers nothing. At a 1% audit budget under defer-to-trusted, an off-lineage monitor catches 19% of attack attempts where an in-lineage monitor catches 41% at the same audit cost, a +21.7-point gap. Control evaluations should report cross-family transfer matrices: a single-pairing accuracy overstates the safety a monitor delivers against a lineage it was not calibrated on. We give a four-step protocol that acts on the gap, with each step a measured result.
[LG-71] Any-Dimensional Learning by Sampling
链接: https://arxiv.org/abs/2607.07680
作者: Eitan Levin,Venkat Chandrasekaran
类目: atistics Theory (math.ST); Machine Learning (cs.LG); Probability (math.PR)
*备注:
Abstract:Many machine learning models are defined for inputs of different sizes, such as point clouds containing different numbers of points, sequences of tokens of different lengths, and graphs on different numbers of nodes. Such models are trained on finitely-many examples of necessarily limited sizes. How well do these models generalize from inputs of small size to larger inputs of size not seen during training? Furthermore, evaluating such models on large inputs is often expensive. How can we sketch large inputs to obtain smaller ones on which the model takes similar values? At the heart of both questions is the need to compare inputs of different sizes and to approximate large inputs by small ones. We present a unified approach to address these questions by using random sampling maps to compare inputs of different sizes. The sampling maps we consider are generalizations of sampling with replacement, random binning, and species sampling. We characterize the application domains in which each type of sampling is appropriate in terms of the symmetries and relations between problem instances of different sizes in the domain. Our framework yields explicit generalization and sketching rates for function classes continuous with respect to a chosen notion of sampling, encompassing large families of functions defined on sequences, graphs, and tensors of different sizes. Specific examples include moment polynomials on measures, homomorphism densities and numbers of graphs, permutation-invariant transformers, and graph neural networks.
[LG-72] A Unified Detection Framework for AI-Related Content and Artifacts
链接: https://arxiv.org/abs/2607.07527
作者: Xifeng Zhang,Tao Hu,Yijie Peng,Wan Tian
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:Artificial intelligence (AI) is a double-edged sword: while it has achieved remarkable success across a wide range of domains, its deployment also calls for effective oversight and regulation, for which the detection of AI-related content and artifacts is perhaps the most direct and cost-effective approach. To this end, we propose a unified detection framework based on Mahalanobis distance scores (MDS), applicable to several important settings, including the detection of large language model (LLM) generated text, hallucination, watermark, and adversarial examples. A key component of the proposed method is to accurately characterize the positive class–such as human-generated text, factual statements, unwatermarked text, or non-adversarial samples–which requires an efficient and robust estimator of the covariance matrix of deep representations of positive samples before computing the MDS. Since the positive samples typically consist of multiple classes, and these classes may exhibit both homogeneity and heterogeneity, we develop joint estimation methods for both the casewise and cellwise minimum covariance determinant (MCD) estimators. We provide efficient optimization algorithms for both estimators and prove their convergence. We provide a reasonable definition of the breakdown point for the joint estimators and prove their corresponding high breakdown point properties. Empirical evaluations confirm the effectiveness of the proposed detection framework.
[LG-73] Statistical inverse learning and ell1-regularization
链接: https://arxiv.org/abs/2607.07468
作者: Abhishake Rastogi,Tatiana A. Bubba,Tapio Helin,Luca Ratti
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Statistics Theory (math.ST)
*备注:
Abstract:We study the recovery of sparse functions from finite, noisy, and indirect observations in the framework of statistical inverse learning. The unknown is modeled as an element of \ell^1 , and observations are generated through a possibly nonlinear forward operator A:\ell^1\to H , where H is a vector-valued reproducing kernel Hilbert space. We propose an \ell^1 -regularized empirical risk minimizer and develop a theoretical analysis of its statistical properties. Under mild assumptions, we establish almost-sure consistency and derive non-asymptotic high-probability convergence rates in both the prediction and \ell^1 reconstruction norms. The rates depend on the source smoothness parameter r , characterized by a variational source condition, and the effective dimension exponent b , describing the polynomial spectral decay of the covariance operator. We further prove matching minimax lower bounds, showing that the obtained convergence rates are optimal. To relate the theory to practical sparsity models, we consider finitely smoothing operators of the form A=G\circ S , where S is a synthesis operator, and show that approximation-space assumptions imply the required variational source conditions. In particular, we prove that membership in the approximation space k_t is equivalent to polynomial decay of the best n -term approximation error. Finally, we verify the assumptions for two representative inverse problems: reaction coefficient identification in elliptic PDEs and sparse computed tomography. For filtered Radon transforms, we derive explicit effective-dimension asymptotics, yielding concrete convergence rates for standard image models and sparsifying systems. Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG); Statistics Theory (math.ST) Cite as: arXiv:2607.07468 [stat.ML] (or arXiv:2607.07468v1 [stat.ML] for this version) https://doi.org/10.48550/arXiv.2607.07468 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Abhishake Rastogi [view email] [v1] Wed, 8 Jul 2026 14:32:08 UTC (159 KB)
[LG-74] Reliable mechanistic operator recovery with biologically-informed neural networks: principles for architecture and optimisation design
链接: https://arxiv.org/abs/2607.07425
作者: Rebecca M. Crossley,Yuan Yin,Sarah L. Waters,Ruth E. Baker
类目: Quantitative Methods (q-bio.QM); Machine Learning (cs.LG)
*备注: 63 pages, 27 figures
Abstract:Many biological processes are governed by complex dynamical mechanisms that remain incompletely understood despite increasing volumes of experimental data. Biologically-informed neural networks (BINNs) seek to address this challenge by embedding mechanistic differential equations into neural network training, enabling interpretable constitutive operators to be recovered directly from sparse and noisy observations. However, reliable operator recovery depends sensitively on network architecture, optimisation strategy, and data informativeness. Here, we present a systematic empirical study of how these factors influence mechanistic inference using BINNs applied to canonical one-dimensional advection-diffusion-reaction partial differential equation models. Across a suite of benchmark problems, we investigate how network expressivity, learning rate, loss weighting, and batch size influence optimisation behaviour and operator recovery. We show that successful mechanistic inference depends on balancing competing objectives rather than maximising any single aspect of the model or optimisation. Moderately expressive architectures outperform overly complex networks, intermediate learning rates improve optimisation stability, balanced data and PDE losses are essential for accurate operator recovery, and intermediate batch sizes provide the best compromise between computational efficiency and reproducibility. We further identify practical diagnostics for recognising common failure modes, including over-fitting, unstable optimisation, and poor mechanistic recovery when the ground truth is unavailable. Together, these findings provide evidence-based guidelines for deploying BINNs as credible tools for biological model discovery.
[LG-75] Restricted Dynamic Geometric Complexity: Certificates for Structured Preconditioning
链接: https://arxiv.org/abs/2607.07204
作者: Zavier Li
类目: Optimization and Control (math.OC); Machine Learning (cs.LG)
*备注: 31 pages
Abstract:Optimization geometrodynamics views optimizer state as evolving geometry. Its full positive-definite quadratic benchmark gives the least affine-invariant deformation needed to reduce condition number when arbitrary metrics are allowed. This paper records that benchmark in the present notation and develops restricted dynamic geometric complexity: an intrinsic certificate distance for reaching a target condition-number class when the metric is restricted to a specified family. The main proved results are monotonicity and submanifold-distance principles, diagonal and block reachability as linear matrix inequality feasibility problems, an exact two-dimensional diagonal complexity formula, and affine-invariant Kronecker projection theorems with normal equations, computable mismatch certificates, Armijo solver convergence, auxiliary self-conditioned K-target bounds, and Hessian-relative candidate certificates through an exact Kronecker Loewner-sandwich reachability condition, including a Kronecker expression threshold and a fixed-basis exact subproblem. Low-rank spectral models, curvature-proxy inflation, stochastic restricted complexity, discrete geometric length, and expression–estimation–flow–discretization accounting are presented as diagnostic interfaces rather than full optimizer characterizations. The resulting language turns structural preconditioner questions into geometric distance, reachability, and certificate problems. The repository includes deterministic toy and synthetic workflows that check diagonal expression gaps, block primal/dual certificates, Kronecker spectral width, and Hessian-relative Kronecker candidate certificates on small quadratic instances, together with low-rank spectral monotonicity.
[LG-76] Weight-Space Physics: Interpretable Hypernetworks for Lattice Quantum Field Theories
链接: https://arxiv.org/abs/2607.07127
作者: Tobias Göbel,Julian R. Ebelt,Zier Mensch,Mathis Gerdes,Miranda C.N. Cheng
类目: High Energy Physics - Lattice (hep-lat); Machine Learning (cs.LG)
*备注: 9 + 13 pages, 4 + 8 figures, 3 + 5 tables
Abstract:Lattice field theory is the workhorse of non-perturbative physics, used to simulate phenomena from the strong nuclear force to critical phenomena in materials. Its Boltzmann distributions are parametrized analytically by coupling constants, but these bare parameters are weak predictors of observables – extracting physics typically requires extensive simulation. While normalizing flows have emerged as effective samplers at fixed couplings, it remains difficult to interpret what these networks have learned. This raises a natural question: can the physics be read off directly from the flow network parameters themselves, and can those parameters be generated for unseen theories? We propose lattice field theory as a testbed for neural network interpretability: because the target physics is qualitatively well-understood and smoothly varying, it provides ideal synthetic data with known ground truth. To this end, we introduce JEPAWG, a Joint-Embedding Predictive Architecture-based Weight Generator that maps couplings directly to flow weights via a learned latent space. On a scalar theory at lattices of size 6^2 to 11^2 , the JEPAWG latent space recovers the correct intrinsic dimension of the underlying manifold, locates the phase transition, and encodes a finite-size shift aligned with the 2D Ising exponent \nu \approx 1 , allowing us to uncover physical structure by studying the network weights alone. This suggests the fascinating idea of treating the network weights as a new type of physical observable. As a generator, JEPAWG also interpolates and extrapolates to unseen couplings effectively and remains robust to weight-space discontinuities introduced by multi-seed training data, outperforming PCA, AE, and VAE baselines.
[LG-77] nsorized algorithms and scalable filtering methods for hidden Markov and factorial hidden Markov models
链接: https://arxiv.org/abs/2607.07008
作者: Roxana Barrios,Ioannis Sgouralis
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注: 23 pages, 3 figures
Abstract:A common method for the representation and analysis of time-series data is the hidden Markov model (HMM), where each observation is associated with a hidden state that evolves over time. However, many real-world systems are influenced by multiple independent factors, which are more naturally represented by factorial hidden Markov models (fHMM), where several hidden Markov chains jointly generate the observed data. Although an fHMM provides a richer and more realistic representation of many real-world systems, it can be reformulated as an equivalent HMM, but with a significantly larger state-space, leading to a severe increase in computational cost. In particular, the forward filtering algorithm, which is central to evaluation, decoding, and estimation tasks, becomes prohibitively expensive even for small systems. This work focuses on developing scalable methods for time-series analysis using tensor algebra to exploit the multidimensional structure of fHMM directly, without constructing intermediate HMM representations. Our novel filtering approach significantly improves computational performance and enables the efficient analysis of large systems and datasets, extending the scope of fHMM and providing a practical framework for data intensive applications.
[LG-78] Mathematical methods of reinforcement learning
链接: https://arxiv.org/abs/2607.06935
作者: Denis Belomestny,Alexander Gasnikov,Egor Gladin,Alexey Naumov,Artemy Rubtsov,Yuri Sapronov,Daniil Tiapkin,Nikita Yudin
类目: Optimization and Control (math.OC); Machine Learning (cs.LG)
*备注: 65 pages
Abstract:Reinforcement learning (RL) is increasingly grounded in tools from probability, optimization, and operator theory. This survey organizes the mathematical structures that underpin the design and analysis of modern algorithms in RL. We begin from Markov decision processes (MDPs) and the Bellman operators, emphasizing contraction mappings, monotonicity, and fixed-point theory that yield convergence guarantees and rates for value and policy iteration, and temporal-difference schemes. We then develop the optimization perspective: stochastic approximation and martingale methods, convex duality and the role of regularization linking mirror/proximal methods. Function approximation is treated through linear and non-linear settings, covering stabilization, error decomposition, and sample-complexity via concentration inequalities for dependent data and mixing processes. We further cover off-policy evaluation/learning, constrained RL and constrained MDPs (CMDPs). Throughout we unify algorithmic templates under common operator and variational lenses, highlighting both finite-sample bounds and asymptotic results. Our presentation is intended to provide a unified mathematical entry point for researchers in probability, optimization, and statistics interested in reinforcement learning.
[LG-79] Finding a stationary point of a stochastic convex problem
链接: https://arxiv.org/abs/2607.06883
作者: Felipe Areces,John Duchi,Malo Sommers
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注:
Abstract:We consider the problem of finding stationary points for stochastic convex optimization problems. Rather than surrogates to stationarity, such as a proximity-to-stationarity guarantee or small gradient of the Moreau envelope, we ask for a stronger notion: that the subdifferential of the objective actually contains a small element. This criterion is non-trivial, because subdifferentials of convex functions fail to converge uniformly, even in arbitrarily small neighborhoods of the optimum. Our convergence guarantees rely on dimension theory to decompose the graph of the subdifferential of a convex function, showing how stochastic sampling preserves “pieces” of these graphs, and allowing effective application of proximal-point-like methods.
[LG-80] nsor Train Diffusion: Leverag ing Low-Rank Structures for High-Dimensional Score-Based Sampling
链接: https://arxiv.org/abs/2607.06841
作者: Robert Gruhlke,Julius Berner,David Sommer,Lorenz Richter
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:Diffusion models offer a powerful framework for sampling from complex probability densities by learning to reverse a noising process. A common approach involves solving for the time-reversed stochastic differential equation (SDE), which requires the score function of the evolving sample distribution. The logarithm of this distribution’s density is governed by a Hamilton-Jacobi-Bellman (HJB) type partial differential equation (PDE). However, current methods for solving this PDE, such as PINNs or trajectory-based techniques, often suffer from long training times and significant sensitivity to hyperparameter tuning. In this work, we introduce a novel and efficient solver for the underlying HJB equation based on the functional tensor train (FTT) format. The FTT representation leverages latent low-rank structures to efficiently approximate high-dimensional functions, enabling both model compression and rapid computation. By integrating this efficient representation with a backward-in-time iterative scheme derived from backward stochastic differential equations (BSDEs), we develop a fast, robust and accurate sampling method. Our approach overcomes primary bottlenecks of existing techniques, enabling high-fidelity sampling from challenging target distributions with improved efficiency.
[LG-81] Optimization Geometrodynamics: A Framework for Dynamic Geometric Optimization
链接: https://arxiv.org/abs/2607.06723
作者: Zavier Li
类目: Optimization and Control (math.OC); Machine Learning (cs.LG); Differential Geometry (math.DG)
*备注: 25 pages, no figures
Abstract:Most gradient-based optimization methods move parameters through a fixed background geometry, even when their internal states implicitly define changing notions of length, curvature, and preconditioning. We introduce optimization geometrodynamics, a benchmark language in which optimization is a coupled evolution of a parameter trajectory, a transported distribution of particles, and a controlled time-varying Riemannian metric. The language separates invariant obstructions from improvable geometric mismatch: positive metrics preserve critical points and Morse indices, and cannot remove global geodesic-convexity obstructions, but can alter conditioning, distributional transport, and flux away from exact critical points. We introduce dynamic geometric complexity, the minimum geometric cost required to reduce an optimization difficulty observable. In the oracle benchmark model of strongly convex quadratic objectives with full positive-definite metric control, this complexity is exactly the affine-invariant distance from the relative log-spectrum to a low-condition-number set. We also analyze Hessian-matching flows, spectral Onsager relaxation, discrete exponential projection updates, gauge-invariant observables, and fixed-time local Morse-saddle flux. The paper is theory-only: its claims are formal statements with proofs, intended to provide invariants and benchmark costs against which implementable adaptive optimizers can be compared once their admissible metric families, curvature estimates, and discretization errors are specified.
[LG-82] Heat-Kernel Entropy Profiles and Geometric Effective Sample Size for Weighted Measures on Manifolds
链接: https://arxiv.org/abs/2607.06696
作者: Kisung You
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Methodology (stat.ME)
*备注:
Abstract:Weighted empirical measures on compact manifolds arise in importance sampling, particle approximations, posterior summaries, quadrature, and representation learning. Standard weight-only summaries, such as ordinary effective sample size, ignore the geometry of the support. We introduce heat-kernel entropy profiles, a multiscale summary that diffuses weighted atoms by intrinsic heat flow and tracks nonuniformity across scales. For order-two Rényi entropy, the profile is computable from pairwise heat-kernel overlaps and yields a geometric effective sample size that discounts nearby or duplicate particles while matching ordinary effective sample size for well-separated particles. We prove monotonicity, small- and large-scale asymptotics, deterministic-weight consistency, and a bounded-ratio self-normalized importance-sampling extension for compact manifolds without boundary. On spheres, the unlogged profile decomposes into spherical-harmonic energies that recover mean-direction, von Mises-Fisher-type, and Bingham-type summaries. Sphere-based experiments show that the profile reveals antipodal, girdle, multimodal, and duplicate-particle structure missed by weight-only and first-moment spherical summaries.
[LG-83] Fast determinantal sampling on general spaces and diffusion geometry DATE
链接: https://arxiv.org/abs/2607.06644
作者: Hoang-Son Tran,Pranav Gupta,Subhroshekhar Ghosh
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Probability (math.PR)
*备注: Preliminary version - to be updated
Abstract:Determinantal point processes have recently emerged as a kernel-based alternative to standard independent sampling for constructing efficient minibatches, coresets, and other compact representations of large-scale datasets. In particular, sampling mechanisms based on DPPs are believed to demonstrate better approximation properties compared to classical i.i.d. samplers, even at the scale of the exponent. One of the key strengths of DPP based samplers is that they can be deployed over very general spaces, in contrast to more classical sampling methods beyond i.i.d. which tend to work in very well-structured settings, principally Euclidean spaces. In this work, we establish explicit rate guarantees for determinantal sampling in spaces that extend far beyond known Euclidean setups, focusing on spectral kernels obtained from eigenspaces of naturally associated Laplacian and other Markov diffusion operators. This includes, in particular, Riemannian manifolds and weighted networks. In determinantal sampling from compact Riemannian manifolds, we establish sampling rates that automatically pick up the intrinsic dimensionality d_\textint of the underlying manifold. In the setting of networks, we investigate DPP-based samplers on the celebrated k-nearest neighbour graphs, as well as weighted random geometric graphs, and demonstrate a similar improved dependence on the intrinsic dimensionality of the data. Overall, our approach achieves guarantees of \big(\textsample size\big)^-\frac12-\frac12d_\textint that match known rates on Euclidean spaces of comparable dimension. In terms of techniques, we connect to the celebrated Weyl’s Law for manifold spectra, and leverage tools from the theory of Markov diffusions and Dirichlet forms as well as certain ingredients from the theory of pseudodifferential operators, which could be of independent interest in this area.
[LG-84] rajectory Inference of Human Aging from Cross-Sectional DNA Methylation Data
链接: https://arxiv.org/abs/2607.06583
作者: Chandan Gupta,Syed Haider,Pietro Liò
类目: Quantitative Methods (q-bio.QM); Machine Learning (cs.LG)
*备注: 13 pages, 6 figures
Abstract:DNA methylation (DNAm) serves as one of the most robust molecular biomarkers of biological aging. While conventional epigenetic clocks accurately predict chronological age from high-dimensional CpG profiles, they treat aging as a static regression task, meaning they can only output a single score rather than simulating how an entire profile continuously changes over time. To reconstruct these continuous dynamics, we frame lifelong human epigenetic aging as a trajectory inference problem across discrete age snapshots derived from widely available cross-sectional data. We introduce a two-stage computational pipeline: first, an age-regularized Variational Autoencoder (VAE) maps high-dimensional CpG profiles onto a chronologically ordered latent manifold while preserving a generative decoder bridge back to the original methylation space. Second, we model the continuous movement across this latent space via Regularized Unbalanced Optimal Transport (RUOT) that unifies deterministic drift, random diffusion, and non-conservative mass changes. By resolving this RUOT formulation using the DeepRUOT framework, our model fluidly accommodates population-level density shifts like survivorship bias and cellular attrition without requiring rigid biological priors. Evaluated on a large-scale, 80-year pan-tissue dataset, our model demonstrates robust distribution interpolation and uncovers a prominent late-life surge in the learned growth field that mathematically captures the variance expansion driven by stochastic epigenetic drift. Finally, by decoding continuous latent paths back to individual CpG sites, we reconstruct and empirically verify distinct biological aging archetypes, offering a rigorous, generative paradigm for simulating human molecular aging.
[LG-85] Recovering Candidate Circadian Regulators of Arrhythmic Pituitary Hormone Genes Using Reliability-Weighted Magnetic Laplacian with rwMagLap
链接: https://arxiv.org/abs/2607.06579
作者: Shabnam Sodagari,Nick Jasperson
类目: Quantitative Methods (q-bio.QM); Machine Learning (cs.LG)
*备注:
Abstract:We study how to recover candidate circadian-clock regulators of pituitary hormone genes that are important for women’s health but do not show a clear 24-hour rhythm in bulk tissue, aiming to nominate clock-linked regulatory targets that could inform future chronopharmacologic and chronotherapeutic strategies. We propose \textbfrwMagLap, which builds a graph on rhythmic backbone genes. For each edge, we combine 24-hour fit quality with peak-time phase, represented as a complex unit-circle value, yielding a Hermitian adjacency matrix and a magnetic Laplacian. We insert arrhythmic hormone genes, treated as anchors, by a reliability-weighted nearest-neighbor projection. The projected anchor-neighbor weights are pooled into a soft teleport distribution, and complex personalized PageRank then ranks rhythmic backbone genes by the magnitude of their PageRank scores. In pituitary data, we find that all 11 women’s-health anchors are arrhythmic. Even so, we find that the top-50 list is 7.95\times enriched for the 13-gene KEGG circadian set (7 of the 8 set genes in the 454-gene backbone; corrected Benjamini-Hochberg (BH) p_\mathrmBH=4\times10^-6 ) and 4.54\times enriched for the 111-gene Reactome set (8 of 16 genes; p_\mathrmBH=1.6\times10^-4 ), while a phase-blind real-valued baseline recovers none. We recover candidates through reliability weighting and phase-aware seeding rather than through magnetic propagation. The magnetic phase adds a different capability: it represents temporal order. On pituitary backbone, the magnetic embedding recovers measured peak-time order of connected pituitary genes with accuracy 0.971 , while q=0 , i.e., no magnetic charge, is at chance.
[LG-86] Value of Information under Imprecise Probabilities: Decision-Rule-Specific Values and Fixed-Measure Envelopes on a Credal Set
链接: https://arxiv.org/abs/2607.06570
作者: Rowan Iskandar
类目: Machine Learning (stat.ML); Information Theory (cs.IT); Machine Learning (cs.LG); Probability (math.PR)
*备注:
Abstract:Value-of-information (VOI) analysis is usually conducted under a single probability measure. However, in practice, the available evidence often pins the measure down only to a set. Consequently, under a set of probability measures, VOI requires different formulations. First, we explicate a rule-specific VOI that fixes a decision rule for acting under imprecision (such as Gamma-maximin) and measures what the information is worth to a decision maker who uses that rule. Second, we derive a fixed-measure envelope that evaluates the classical VOI functional over all admissible precise measures. We formalize this distinction and explicate its consequences for the expected perfect, partial, and sample information. The expected value of perfect information is concave over the credal set. Hence, when the set is generated by finitely many measures, its lower envelope endpoint is obtained exactly from the generators, while its upper endpoint may be interior and is computed by a finite linear program. The Gamma-maximin value, in contrast, can exceed the entire envelope, so a rule-specific value is not recovered from the envelope’s endpoints. A continuity bound limits how much the VOI can change as the measure varies, and we identify when the partial- and sample-information endpoints can still be obtained from the generators. Because the single-measure VOI must itself be estimated, the procedure we give combines standard estimators for it with a search over the credal set. By using a worked decision problem, we show how the two quantities separate conclusions that hold across every admissible measure from conclusions that depend on one unidentified choice of measure.
[LG-87] owards a Pseudo-Labeling Workflow for Celltype-Classification from Explanted Brain Slice Recordings
链接: https://arxiv.org/abs/2607.06569
作者: Cora Jostock,Jonas Ort,Henner Koch,Gregor Schiele,Andreas Erbslöh
类目: Quantitative Methods (q-bio.QM); Machine Learning (cs.LG)
*备注: 4 pages, 5 figures, Biomedizische Technik 2026 (accepted)
Abstract:This paper proposes an unsupervised workflow to pseudo-label extracellular spikes from human brain slice MEA recordings into two putative cell types: pyramidal cells and interneurons. Here, the raw data from the data acquisition system is used and processed. The pipeline for pre-processing includes bandpass filtering, threshold–based spike detection, frame alignment and normalization. In the ML workflow, dimensionality reduction (PCA, t-SNE, UMAP), clustering (GMM, k-means). To achieve an online system, template matching and OSort under varying curation strictness is also considered. All pipelines are evaluated by different cluster quality with within-cluster Pearson correlation, Silhouette score, and Calinski-Harabasz index. Applying stricter curation improves separation at some cost to inclusivity.
附件下载
