本篇博文主要内容为 2026-07-08 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 (2026-07-08)
今日共更新601篇论文,其中:
- 自然语言处理共70篇(Computation and Language (cs.CL))
- 人工智能共184篇(Artificial Intelligence (cs.AI))
- 计算机视觉共120篇(Computer Vision and Pattern Recognition (cs.CV))
- 机器学习共118篇(Machine Learning (cs.LG))
- 多智能体系统共9篇(Multiagent Systems (cs.MA))
- 信息检索共16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] Danus: Orchestrating Mathematical Reasoning Agents with Fact-Graph Memory
【速读】：该论文旨在解决大规模语言模型（LLM）驱动的数学推理智能体在应对研究级问题时，因难以有效协调并行证明搜索过程而造成的可扩展性与中间结论组织可靠性不足的问题。其核心解决方案在于提出Danus系统，该系统以共享事实图（fact graph）作为全局记忆管理机制，构建了一个由主代理（main agent）负责规划与协调、多个工作代理（worker agents）并行执行证明搜索、以及无状态验证器（stateless verifier）对拟纳入事实图的数学命题进行前置校验的协同架构。每个经验证的事实均与其证明过程及逻辑依赖关系一同存储，支持系统逐步构建长链条、高复杂度的数学论证，同时保持共享证明状态的结构化与一致性。主代理定期总结进展、动态引导工作代理聚焦有前景的方向，并通过进度报告支持与人类数学家的交互。在代数几何、奇点理论和组合数学等六个研究级案例中的实证表明，基于事实图的编排机制显著提升了数学推理智能体在长周期科研任务中的可扩展性与有效性。
链接: https://arxiv.org/abs/2607.06447
作者: Jihao Liu,Guoxiong Gao,Zeming Sun,Bin Wu,Shurui Liu,Jiedong Jiang,Haocheng Ju,Leheng Chen,Ronnie Cheng,Xiping Zhang,Bin Dong
机构: 北京大学数学科学学院(University of Peking, School of Mathematical Sciences)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Multiagent Systems (cs.MA)
备注:
Abstract:Recent LLM-based mathematical reasoning agents have begun to tackle research-level problems and, in several cases, have contributed to the resolution of open problems. However, scaling and orchestrating such agents effectively remains challenging, due to the difficulty of coordinating parallel proof search while keeping intermediate claims organized and reliable. In this paper, we propose Danus, an orchestration system for research-level mathematical reasoning centered on a shared fact graph as a global memory-management mechanism. Danus consists of a main agent that performs planning and coordination, multiple worker agents that carry out proof search in parallel, and a stateless verifier that checks proposed mathematical claims before they are admitted into the fact graph. Each verified fact is stored together with its proof and logical dependencies, allowing the system to build long arguments incrementally while keeping the shared proof state organized. The main agent periodically summarizes the evolving proof state, redirects workers across promising directions, and supports interaction with human mathematicians through progress reports. We evaluate Danus through six research-level case studies in algebraic geometry, singularity theory, and combinatorics, illustrating how the fact-graph memory mechanism enables Danus to construct long, detailed mathematical proofs. Our results suggest that fact-graph-based orchestration provides an effective route toward scaling mathematical reasoning agents for long-horizon research problems. Danus is open source at this https URL.
[MA-1] Information Limits and Attractor Dynamics in Economies of Frontier LLM Agents : A Pre-Registered Test
【速读】：该论文旨在解决生成式智能体（Generative AI agents）在耦合多智能体系统中，其经济行为与信息论特性之间的定量关系问题，具体聚焦于两个核心科学假设：一是市场耦合下财富增长的信息论容量区域（information-theoretic capacity region for wealth growth），二是激励与控制机制作用下群体错位（population misalignment）的平均场残差标度律（mean-field residual-scaling law）。其解决方案的关键在于构建一个可复现、预注册的双阶段实验框架，通过使用高级语言模型（Claude Opus 4.8）作为前沿语言模型智能体，在受控的赛马制（parimutuel）经济环境中模拟多种感知结构下的协作与竞争。研究通过严格预注册的接受区间和决策规则确保结果的客观性，并利用缓存的模型输出实现完全可重演性。关键发现包括：相对增长差异精确等于相对声称信息量（即“间隙定律”成立）、联合增长上限由信息熵 $ H(X) $ 精确绑定、以及在条件独立通道中联盟价值呈次模性而通过设计的异或协同控制可使其转为超模性；然而，第二项结构性负结果表明，所有测试的群体均未实现平均场模型所假设的噪声维持分散态，而是表现出目标弥散崩溃、对激励杠杆的响应呈现阶跃函数特征及边界附近双稳态现象，揭示了现有大型语言模型群体在复杂动态系统中缺乏连续响应能力的根本局限。
链接: https://arxiv.org/abs/2607.06001
作者: Cheng Qian
机构: 未知
类目: Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注: 15 pages. Preprint. Zenodo: this https URL . Companion synthesis: arXiv:2606.12502
Abstract:We report a pre-registered, two-part experiment on small economies of frontier language-model agents (Claude Opus 4.8), testing two quantitative predictions about coupled multi-agent systems: an information-theoretic capacity region for wealth growth under market coupling, and a mean-field residual-scaling law for population misalignment under incentive and control levers. All predictions, acceptance bands, and decision rules were frozen in a public git chain before any run; every reported number re-derives mechanically from cached model outputs; the entire experiment cost 138.76 in metered API spend and is re-runnable at zero cost from the cache. Result 1 (confirmation): in parimutuel-coupled economies, relative growth equals relative claimed information – the gap law G_a - G_b = I_a - I_b holds to a worst-case 46 millinats (pre-registered band: 50) across four perception structures; coalition value is submodular exactly where channels are conditionally independent, and a designed XOR synergy control flips it supermodular by 0.62 = ln2/2 nats, with agents reasoning out the joint bit; the joint growth ceiling G_S = H(X) binds exactly; and the best-informed agent absorbs essentially the whole wealth pool in 4/5 market seeds. Result 2 (structural negative): the residual-scaling test returned “domain not found.” In all 72 population runs, goal dispersion collapsed (V - 0; maximum 4.85 against a frozen floor of 5.31), the population’s response to the two levers was a step function across the dominance boundary rather than a smooth response, and cells near the boundary were bistable with seed-selected outcomes. No tested LLM population at any capability level realizes the noise-maintained-dispersion regime the smooth mean-field model assumes. We release the full protocol, pre-registration chain, call cache, and analysis code. Comments: 15 pages. Preprint. Zenodo: this https URL. Companion synthesis: arXiv:2606.12502 Subjects: Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA) Cite as: arXiv:2607.06001 [cs.AI] (or arXiv:2607.06001v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2607.06001 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[MA-2] MCP-Enabled Agent ic AI for Autonomous IPoDWDM Network Lifecycle Automation
【速读】：该论文旨在解决异构厂商无关的分组交换光传输网络（IPoDWDM）在动态运维中面临的自动化程度低、跨层协同能力不足的问题。其核心解决方案是基于多智能体控制协议（MCP）构建一个具备自主决策能力的智能体式人工智能架构，通过集成通用光网络性能模型（GNPy）与实时遥测数据，实现端到端的多层生命周期自动化管理与闭环控制，在真实测试床环境中验证了该方案在动态资源调度、故障自愈及性能优化等方面的可行性与有效性。
链接: https://arxiv.org/abs/2607.05975
作者: Chunmin Xia,Jakub Harbaczewski,Nikhil Dsilva,Julie Raulin,Dominic Schneider,Achim Autenrieth
机构: 未知
类目: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA); Systems and Control (eess.SY)
备注: Accepted for demo presentation at the European Conference on Optical Communication (ECOC 2026)
Abstract:This demo presents an MCP-enabled agentic AI architecture for autonomous control of vendor-agnostic IPoDWDM networks. We demonstrate live end-to-end lifecycle multi-layer automation and closed-loop control using GNPy and telemetry, validated on a real testbed.
[MA-3] Delay-Aware Active Triangulation with Uncertainty-Driven Multi-Agent Reinforcement Learning for Counter-UAS IROS2026
【速读】：该论文旨在解决无人机反制（Counter-UAS）场景中多智能体主动视觉三角测量的时延与不确定性问题，核心挑战在于现有方法假设状态信息可即时获取，忽略了目标检测、通信及决策传播等环节累积的延迟对定位精度的影响。其解决方案的关键在于提出一种面向时延感知的不确定性驱动型多智能体强化学习框架：首先，构建了引入时效性信息（Age-of-Information, AoI）的分布式部分可观测马尔可夫决策过程（Dec-POMDP）模型，通过AoI增强观测以提升三角测量有效性（提高10.6个百分点）；其次，通过受控对比实验表明，基于感知一致性的奖励机制优于理想化清洁状态奖励（RMSE分别为0.547 m vs. 0.633 m，轨迹丢失减少27%），揭示了鲁棒性与稳定性之间的权衡关系；最后，提出融合像素、位姿、云台及内参不确定性的多源解析协方差传播方法，证明仅考虑角度噪声会导致均方根误差（RMSE）恶化2.8倍。实验结果表明，在4096个并行环境中采用MAPPO算法可实现0.547 ± 0.217 m RMSE和78.1%的三角测量有效性，而无记忆结构的MLP策略有效性仅为0.7%，验证了循环记忆在时延补偿中的关键作用。
链接: https://arxiv.org/abs/2607.05957
作者: Seungwook Lee,David Hyunchul Shim
机构: Korea Advanced Institute of Science and Technology (KAIST)
类目: Robotics (cs.RO); Multiagent Systems (cs.MA)
备注: Accepted to IROS 2026
Abstract:Multi-agent active visual triangulation enables precise 3D localization of aerial targets by coordinating mobile observers with controllable cameras. However, existing methods assume instantaneous state feedback, ignoring cumulative latency from detection, communication, and decision propagation. We present a delay-aware, uncertainty-driven multi-agent reinforcement learning framework for target localization in Counter-UAS applications. Our contributions are: (1) a Dec-POMDP formulation with Age-of-Information (AoI) augmented observations enabling delay-aware coordination – AoI improves triangulation validity by 10.6 percentage points; (2) a controlled comparison showing that perception-consistent rewards outperform privileged clean-state rewards (0.547 m vs.0.633 m RMSE, 27% fewer track losses) – both policies are trained through identical observation noise but differ in what they are optimized for, producing a stability-robustness tradeoff; and (3) multi-source analytical covariance propagation incorporating pixel, pose, gimbal, and intrinsics uncertainties – restricting to angular noise alone causes 2.8-fold RMSE degradation. Experiments with MAPPO in 4096 parallel environments achieve 0.547 ± 0.217 m RMSE with 78.1% triangulation validity, while MLP policies achieve near-zero validity (0.7%), confirming recurrent memory as essential for delay compensation.
[MA-4] StateFuse: Deterministic Conflict-Preserving Memory for Multi-Agent Systems
【速读】：该论文旨在解决多智能体系统（Multiagent Systems）在运行过程中因并行分支、重试与副本导致的观测冲突累积问题，现有实践中的记忆层普遍采用覆盖式更新规则（overwrite rules），其冲突处理机制难以追溯与修正，进而削弱了系统的可解释性与安全性。其解决方案的关键在于提出StateFuse——一种基于标准OpSet/CRDT合并机制的冲突感知型复制记忆协议（conflict-aware replicated memory contract）。StateFuse不引入新的合并代数，而是通过定义面向智能体的语义层，实现不可变的历史记录、显式的冲突对象表示、精确且语义化的修正句柄（claim_id / claim_ref）、确定性谓词合约以及投影时分辨率（projection-time resolution），从而确保状态无法被回溯性改写。实验表明，在包含282个冲突样本的MemoryAgentBench基准测试中，尽管各方法在答案准确率上表现相当，但保留冲突的方法能持续显式暴露矛盾，而折叠式（collapsed）方法则隐藏冲突；在统一验证的受控循环中，保持语义模糊性有助于更安全的拒绝响应与可审计的修正。消融实验进一步证明，当精确先验标识不可用时，语义化修正句柄具有关键作用。因此，论文核心结论为：StateFuse应被视为一种更安全的公开记忆契约，适用于冲突显式化、安全回避与可审计修正，而非提升通用准确率的万能方案。
链接: https://arxiv.org/abs/2607.05844
作者: Sergey Volkov,Yang Li,Ye Luo
机构: The University of Hong Kong(香港大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Multiagent Systems (cs.MA)
备注: Code and supplementary materials available at: this https URL
Abstract:Agent systems accumulate conflicting observations across branches, retries, and replicas, yet many practical memory layers still collapse disagreement behind overwrite rules that are difficult to inspect or correct. We present StateFuse, a conflict-aware replicated memory contract built on standard OpSet/CRDT merge. StateFuse does not introduce a new join algebra; it defines an agent-facing semantics layer with immutable history, explicit conflict objects, exact and semantic correction handles (claim_id / claim_ref), deterministic predicate contracts, and projection-time resolution that cannot rewrite replicated state. We evaluate StateFuse against flat multi-value, raw-log, provenance-style, and collapsed baselines under matched resolver and verification policies. On a 282-question official conflict-bearing MemoryAgentBench slice, the compared methods tie on answer accuracy, but conflict-preserving surfaces keep contradictions visible while collapsed surfaces do not. In a controlled agent loop with uniform verification, preserving ambiguity enables safer abstention and correction than early collapse. A correction-handle ablation further shows that semantic handles matter when exact prior identifiers are unavailable. The resulting claim is narrow: StateFuse is best supported as a safer public memory contract for contradiction surfacing, abstention, and auditable correction, not as a universal accuracy gain. Comments: Code and supplementary materials available at: this https URL Subjects: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Multiagent Systems (cs.MA) Cite as: arXiv:2607.05844 [cs.AI] (or arXiv:2607.05844v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2607.05844 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[MA-5] Deep Reinforcement Learning for Dynamic Battery Management of Autonomous Order Pickers
【速读】：该论文旨在解决仓库环境中自主移动机器人（AMR）在随机订单到达条件下动态充电调度的优化问题，核心挑战在于如何在复杂动态环境下实现多机器人协同充电决策，以减少充电时间、提升订单处理效率与整体吞吐量。传统基于固定规则的启发式方法因缺乏对环境动态变化的适应性及多机器人协调能力，常导致资源利用效率低下。本文提出一种基于近端策略优化（Proximal Policy Optimization, PPO）的深度强化学习（Deep Reinforcement Learning, DRL）框架，专为配备固定充电站的多区域仓库设计，能够动态学习两个关键决策：充电站选择与最优充电时长，并显式建模充电站预期排队时间。实验结果表明，所提PPO模型相较现有最优基准方法可将订单完成率提升最高达6%，同时显著降低总充电耗时。此外，模型在多种仓库布局与随机订单到达率下均表现出强鲁棒性，且通过对学习策略的可解释性分析，揭示了其优于传统基准的内在机制，为实际运营提供了重要决策支持。
链接: https://arxiv.org/abs/2607.05683
作者: Taniya Shaji,Abhay Sobhanan,Christof Defryn
机构: Indian Institute of Management Bangalore (印度管理学院班加罗尔分校); University of Antwerp (安特卫普大学)
类目: Machine Learning (cs.LG); Multiagent Systems (cs.MA); Optimization and Control (math.OC)
备注:
Abstract:Battery charging of Autonomous Mobile Robots (AMRs) in warehouses is a critical operational challenge that heavily impacts both order processing times and throughput. In this study, we address the dynamic AMR charging problem under stochastic order arrivals, where robots must learn optimal charging decisions. Traditional fixed-rule heuristics often prove suboptimal in dynamic environments and fail to account for multi-AMR coordination, leading to severe resource inefficiencies. To overcome these limitations, we propose a Proximal Policy Optimization (PPO)-based Deep Reinforcement Learning (DRL) framework designed for multi-block warehouses with fixed charging stations. Our model dynamically learns two key decisions: charging station selection and optimal charging duration, explicitly accounting for anticipated queuing times at the stations. Extensive numerical experiments benchmark the proposed model against state-of-the-art DRL and traditional heuristic approaches. Results demonstrate that our PPO framework increases order-completion rates by up to 6% compared to the strongest baseline, while significantly reducing the total time dedicated to recharging operations. Furthermore, we validate the model’s robustness across diverse warehouse configurations and stochastic arrival rates. Finally, we interpret the learned DRL policy, offering valuable operational insights into its superiority over standard benchmarks.
[MA-6] PatchOptic for Shared-State LLM Workflows with Projected Views and Verified Structured Updates
【速读】：该论文旨在解决生成式智能体（Agentic）工作流在共享结构化状态（shared, structured state）中操作时，因大语言模型（LLM）上下文窗口受限而引发的局部更新与全局一致性之间的矛盾问题。现有方法通过关键词搜索、检索增强生成（RAG）、抽象语法树（AST）查询及任务特定技能等方式实现对状态片段的渐进式披露（progressive disclosure），虽提升了读取效率，但缺乏对本地修改后回写至完整状态时有效性的形式化保障，即缺少“局部更新与全局有效性之间的契约”（contract between local updates and global validity）。其解决方案的关键在于提出PatchOptic——一种受光学（optic）启发的共享状态工作流接口，利用投影读取（projected reads）与经过验证的结构化补丁（verified structured patches）构建双向访问机制。每个工作流步骤明确定义投影读取视图、授权写入区域与补丁源区域，不仅在运行时强制执行工作流契约，还通过路径级足迹支持任务委托、子工作流组合以及同一阶段内独立步骤的静态重排证明，从而确保更新的正确性与可验证性。实验评估表明，该设计在降低信息泄露与令牌开销的同时，维持了高质量输出，并有效阻止了违反契约的提交行为。
链接: https://arxiv.org/abs/2607.05483
作者: Zhaoyu Bai,Jiaqi Cai
机构: Weizmann Institute of Science (魏茨曼科学研究所); Massachusetts Institute of Technology (麻省理工学院)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Logic in Computer Science (cs.LO); Multiagent Systems (cs.MA); Programming Languages (cs.PL)
备注: 24 pages, 13 figures, including appendix
Abstract:Agentic workflows often operate over shared, structured state. Because LLM context windows are limited, each model invocation is typically shown only the state fragment needed for the current workflow step, a pattern commonly known as progressive disclosure. Modern systems construct such model-facing views using grep-like keyword search, retrieval-augmented generation (RAG), abstract-syntax-tree (AST) queries, and task-specific agent skills. These methods make the read side manageable, but they do not define when a locally proposed rewrite is valid after it is applied back to the full state. The missing piece is a contract between local updates and global validity. We introduce PatchOptic, an optic-inspired interface for shared-state LLM workflows. Optics are compositional bidirectional accessors that describe how views of structured data are read and updated. PatchOptic borrows this view/update intuition and realizes it through projected reads and verified structured patches. Each workflow step declares a projected read view, an authorized write region, and a patch-source region. Beyond runtime enforcement, the same declaration yields a path-level footprint that supports delegation, sub-workflow composition, and static certificates for reordering independent steps within the same phase. We evaluate this design with PatchBench, a benchmark with 46 cases across domains. The results show that projected reads reduce reported leakage and token cost while preserving accepted-output quality under the strong actor. Runtime verification blocks declared workflow-contract violations before commit, and patch-read enforcement rejects compromised patch artifacts that use hidden sources.
[MA-7] Decision Protocols in Multi-Agent Large Language Model Conversations
【速读】：该论文旨在解决大语言模型（Large Language Models, LLMs）在规模扩展过程中面临的收益递减与高昂成本问题，提出通过多智能体系统（Multi-Agent Systems, MAS）将任务分配给专业化智能体以提升整体任务性能。其核心挑战在于如何设计高效的决策协议（decision protocol），以协调多个智能体之间的协作并生成最终解决方案。本文提出的多智能体大语言模型框架（Multi-Agent LLM, MALLM）系统性地评估了投票（voting）、共识（consensus）和裁判决策（judge decision）三种决策机制在多样化任务上的表现，涵盖知识密集型（如MMLU、MMLU-Pro、GPQA）与逻辑推理型（如StrategyQA、MuSR、Math-lvl-5、SQuAD 2.0）数据集。研究发现，共识协议在知识类任务中表现最优，而投票与裁判协议在逻辑推理任务中更具优势；独立生成多样化解法能显著提升决策质量，而决策过程中信息访问方式的改变对结果影响较小。因此，该研究的关键在于通过系统化验证不同决策机制的适用场景，为多智能体协作中的高效决策提供可解释的优化路径。
链接: https://arxiv.org/abs/2607.05477
作者: Lars Benedikt Kaesberg
机构: 未知
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: Master’s thesis, University of Göttingen
Abstract:Improving the task performance of Large Language Models (LLMs) is essential, yet scaling these models faces significant challenges such as diminishing returns and high costs. Multi-Agent Systems (MAS) offer a promising solution by distributing tasks among specialized agents to improve the overall task performance. This can reduce training costs at the expense of increased test time due to the discussion and decision-making process. The decision protocol is a critical component of MAS because it specifies how multiple agents collaborate to create a final solution. This thesis introduces the Multi-Agent LLM (MALLM) framework, which implements and evaluates various decision protocols, namely voting, consensus, and judge decision mechanisms, to simulate multi-agent discussions for conversational task solving. Unlike previous work that used a single decision protocol or tested them on limited datasets, this study systematically examines their impact on a diverse set of tasks, ranging from knowledge-based datasets (MMLU, MMLU-Pro, GPQA) and logic-based datasets (StrategyQA, MuSR, Math-lvl-5, SQuAD 2.0). The results indicate that consensus protocols excel in knowledge-intensive domains while voting and judge protocols are more effective for logic-based tasks. Increasing response diversity through independent solution generation improves decision quality, while changes in information access during the decision process have minimal impact.
[MA-8] A study of holes: Topological analysis reveals crowd dynamics regimes in a bidirectional corridor scenario
【速读】：该论文旨在解决复杂人群动态中隐藏的拓扑结构识别问题，尤其关注在缺乏先验假设的情况下，如何有效刻画行人的时空交互模式。其核心挑战在于从高维、非线性的行人位置时间序列数据中提取具有统计意义的结构特征，而传统方法往往依赖于预设的运动模型或模式假设。解决方案的关键在于引入**持久同调（persistent homology）**这一拓扑数据分析技术，通过构建基于行人邻近关系的抽象网络，量化连通分量与环状空洞（即无跨连接的闭环结构）随时间演化的变化，并生成称为CROCKER（Cross-Sectional RObust Clustering of Kernels）的双参数矩阵以表征整个时间序列的拓扑签名。尽管数据经过高度抽象，但对时延位置数据进行主成分分析后，前两个主成分仍能清晰区分不同参数配置下的群体行为模式，且该区分能力对对称性保持鲁棒。研究结果表明，持久同调能够无需预设假设地揭示人群动力学中的结构性特征，为复杂系统分析提供了一种通用且强大的工具。
链接: https://arxiv.org/abs/2607.06086
作者: Sabrina Desiree Kern,Gerta Köster
机构: Hochschule München – University of Applied Sciences (慕尼黑应用技术大学)
类目: Dynamical Systems (math.DS); Multiagent Systems (cs.MA)
备注: Presented at Traffic and Granular Flow 2026 (TGF26)
Abstract:This study harnesses topological analysis in an attempt to reveal structure in the dynamics of a crowd. Topology and in particular persistent homology characterizes relational structures in data through the number of connected components and holes, that is, a loop of pairwise connection with no connections across it. We apply this universal data analysis method to a simulated time series of individual pedestrian positions of a crowd moving through a wide corridor – either uni- or bidirectional. We consider two pedestrians to be connected, when they are sufficiently close. This approach leads to two matrices containing the persistence signatures for the whole time series, so-called CROCKERs. Despite the high level of data abstraction, the CROCKERs’ first two principal components on time-delayed positional data show a clear separation of the different parameter configurations. This holds up to symmetry. Our results support our claim that persistent homology is a useful tool to characterize crowd dynamics without introducing any prior assumptions about the detectable spatio-temporal patterns.
自然语言处理
[NLP-0] Rethinking Indic AI from a Lens of Cultural Heritage Preservation
【速读】： 该论文旨在解决生成式 AI 在印度次大陆应用过程中对本土语言与文化根基带来的双重影响问题，特别是如何在技术进步的同时避免语言同质化和边缘语言的进一步被排除。其核心挑战在于印度语言所具有的复杂性——包括丰富的形态特征、复杂的书写系统与语法规则、方言多样性以及语域差异（diglossia），这些特性为构建通用型自然语言处理（NLP）模型带来了独特障碍。解决方案的关键在于发展“印地语基础模型”（Indic foundation models），通过系统性资源建设与方法论演进，弥合长期存在的低资源语言覆盖不足与代表性缺失的问题。此外，论文提出“文化感知”（Culture Sensing）这一新兴研究方向，主张基于诠释学推理（hermeneutic reasoning）重构AI系统，以实现跨语言的公平性能表现，并生成具有文化意义的输出，从而推动更具包容性与文化敏感性的印地语NLP生态系统的可持续发展。
链接: https://arxiv.org/abs/2607.06544
作者: Aparna Madva,Sharath Srivatsa,Srinath Srinivasa,Tulika Saha
机构: International Institute of Information Technology, Bengaluru(印度信息科技国际学院，班加罗尔); Bengaluru(班加罗尔); Karnataka(卡纳塔克邦); India(印度)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:As Artificial Intelligence (AI) makes inroads into different parts of the Indian subcontinent, there is significant interest in studying how AI impacts the linguistic and cultural foundations of this civilization. AI is seen as a ‘‘double-edged sword’’ where on the one hand, it can enable access and inclusion for a large population, on the other, it can homogenize worldviews and exclude underrepresented languages and worldviews. In this paper, we try to characterize this problem by addressing the extensive characteristic nature of Indian linguistics and the way they closely connect to cultural practices and worldview. We then perform a longitudinal survey of how Natural Language Processing (NLP) techniques have evolved in this space, tracing the historical development of Indic NLP, covering key milestones, methodological shifts, and resource creation efforts. In addition, the paper also examines the structural and sociolinguistic characteristics of Indian languages, such as rich morphology, complex scripts and grammar rules, diglossia, and large dialectal variation, and explains how these create unique challenges for building AI foundation models. We then discuss the growing role of Indic foundation models and analyze how these models address these long-standing resource and representation gaps. Finally, we propose a research direction called ‘Culture Sensing’, which re-imagines AI based on hermeneutic reasoning. Culture Sensing aims to address open problems such as ensuring equitable performance across low-resource languages and producing outputs that are culturally meaningful. By bringing together past work, current techniques, and emerging trends, this paper outlines research directions that can guide the next phase of Indic NLP and contribute to the development of more robust and inclusive Indic foundation models.
[NLP-1] On the feasibility of dependency parsing of non-human sequences without a gold standard. Is evaluation possible in other species?
【速读】： 该论文旨在解决非人类灵长类动物发声或手势序列中无监督依存句法分析（unsupervised dependency parsing）的可评估性问题。由于在人类语言中存在可获取的黄金标准（gold standard），其无监督解析器的性能可被评估，但在其他物种中，黄金标准未知，传统观点认为此类解析任务无法进行准确评估，因而被认为不可行。论文的关键突破在于引入网络科学（network science）的最新进展，指出非人类灵长类动物的序列具有快速衰减的序列长度分布特性，这一统计特征导致解析器所提取的正确边（correct edges）比例必须较高才能维持序列结构的合理性；相比之下，人类语言序列不具备此性质。因此，基于该理论约束，可在无需黄金标准的情况下对非人类灵长类的无监督解析器进行有效评估，而人类语言则因缺乏此类约束而难以实现类似评估。该解决方案的核心在于利用序列统计特性作为评估依据，从而为跨物种交流研究提供了新的可行性路径。
链接: https://arxiv.org/abs/2607.06542
作者: Ramon Ferrer-i-Cancho,Catherine Hobaiter,Thore Bergman,Morgan Gustison
机构: Universitat Politècnica de Catalunya (加泰罗尼亚理工大学); University of St Andrews (圣安德鲁斯大学); University of Michigan (密歇根大学); Western University (西安大略大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Dependency parsing consists of finding a tree representation for a sequence. Unsupervised dependency parsing aims to develop parsing methods without a gold standard during model training. In human languages, an unsupervised parser can be evaluated because some gold standard is usually available or can be created. For other species, a gold standard is unknown. Thus one may conclude that it is impossible to determine the accuracy of an unsupervised parser and, consequently, dependency parsing is unfeasible in other species. However, here we apply recent advances in network science to demonstrate that the proportion of correct edges retrieved by a parser must be high for the sequences of vocalizations or gestures that non-human primates produce due to the fast decay of the sequence length distribution. In contrast, human language sequences lack that property. Therefore, evaluation without a gold standard is feasible in non-human primates but a hard problem in humans.
[NLP-2] Hierarchical Acoustic-Semantic Modeling: Modality Separation and Semantic Coherence for Full-Duplex SLMs
【速读】： 该论文旨在解决生成自然、高性能的原生全双工语音语言模型（Spoken Language Models, SLMs）所面临的重大挑战，核心问题在于多模态干扰导致的知识退化与语义完整性受损，使得现有系统在全双工交互中表现不自然且缺乏智能性。其解决方案的关键在于通过精细分析模型优化动态，首次揭示了声学模态与语义模态在共享深层参数空间时因梯度冲突而引发多模态干扰的根本原因。基于此洞察，作者提出Lychee-FD框架，采用分层参数分离策略，在深层网络中解耦冲突的模态，同时通过专用语义对齐通道保持跨模态一致性，从而实现模态间的高效协同。实验表明，该方法在多个全双工基准测试上显著提升性能，不仅在口语问答任务中实现7.4%的智能性增益，还在全双工交互流畅度上提升28.5%，且未牺牲推理效率，是首个从机理层面揭示并有效解决全双工SLMs中多模态干扰问题的工作。
链接: https://arxiv.org/abs/2607.06540
作者: Zhenyu Liu,Yunxin Li,Xuanyu Zhang,Qixun Teng,Shenyuan Jiang,Haolan Chen,Minjun Zhao,Fanbo Meng,Yu Xu,Yancheng He,Baotian Hu,Haizhou Li,Min Zhang
机构: Harbin Institute of Technology, Shenzhen (哈尔滨工业大学深圳校区); Shenzhen Loop Area Institute, Shenzhen (深圳环区研究院); The Chinese University of Hong Kong, Shenzhen (香港中文大学深圳校区)
类目: Computation and Language (cs.CL)
备注: 22 pages, 9 figures
Abstract:Developing seamless, high-performance, native intelligent full-duplex Spoken Language Models (SLMs) remains a critical challenge and long-standing goal for the speech and NLP community. Despite notable progress, recent endeavors are fundamentally constrained by severe modality interference, which causes substantial knowledge degradation and compromises semantic integrity – ultimately making full-duplex SLMs feel unnatural and unintelligent. In this paper, through an exhaustive fine-grained analysis of model optimization dynamics, we uncover the root cause of such performance degradation, revealing that modality interference arises from inherent gradient conflicts between acoustic and semantic modeling when the two modalities are forced to share a deep parameter space. Guided by this key insight, we introduce Lychee-FD, a native end-to-end full-duplex framework designed to mitigate modality interference. Importantly, we propose a hierarchical parameter separation strategy that decouples conflicting modalities in deep layers while preserving cross-modality coherence via a dedicated semantic alignment channel. Extensive experiments on multiple full-duplex benchmarks demonstrate that our method significantly advances the state of the art, yielding substantial improvements in both speech intelligence (+7.4% on Spoken QA) and full-duplex interaction fluidity (+28.5% on FullDuplexBench 1.5) without compromising inference efficiency. To the best of our knowledge, this work is the first to achieve two key advances: 1) uncovering and elucidating the root cause of modality interference in full-duplex SLMs, and 2) designing an elegant hierarchical model together with a practical solution for seamless, high-performance, native intelligent full-duplex SLMs.
[NLP-3] Life Style Levels: Neighborhood Delineation using Geospatial Data
【速读】： 该论文旨在解决发展中国家快速城市化地区（如印度）缺乏细粒度社会经济信息的问题，从而限制了对城市内部富裕与贫困差异的精准刻画。其解决方案的关键在于提出一种基于网格的可扩展城市边界划分框架，利用开源卫星影像提取的建筑形态学特征，将59个印度城市和城镇划分为高分辨率空间网格，并通过可解释的形态学指标构建透明、规则驱动的评分体系，以识别不同层级的城市富裕程度。该方法完全依赖公开地理空间数据，实现了低成本、可推广的精细化城市富裕度制图，且经谷歌街景实地观测验证，网格分类结果与预期的生活方式富裕度指标高度一致；进一步在孟买开展基于建筑足迹密度的聚类分析，发现聚类结果与已知非正式住区存在显著空间重叠，验证了方法的有效性；最后通过探索性分析揭示了信贷违约率在不同富裕等级网格中的分布模式，凸显了该框架在支持城市治理与金融风险评估中的应用潜力。
链接: https://arxiv.org/abs/2607.06529
作者: Srivatsa Kulkarni,Debarag Banerjee
机构: 未知
类目: Computation and Language (cs.CL)
备注: 43 pages, 38 figures
Abstract:Fine-scale socioeconomic information is often unavailable across rapidly ur-banizing regions of the developing world, like India, limiting the ability to delineate intra-urban variations in affluence and deprivation. This study pro-poses a scalable, grid-based urban delineation framework using building morphology derived from open-source satellite imagery. Urban areas across 59 Indian cities and towns are partitioned into high-resolution spatial grids and characterized using interpretable morphological indicators, which are combined into a transparent, rule-based scoring framework to delineate areas with contrasting levels of urban affluence. The resulting classifications are validated through ground-level Google Street View observations, revealing a sharp contrast between the grid classes which are consistent with the ex-pected effects of the lifestyle affluence indicators. We further investigate density-based clustering of building footprints in Mumbai to identify dense urban settlements, demonstrating that the resulting clusters exhibit substan-tial spatial overlap with known informal settlements across the city. Finally, we conduct an exploratory analysis mapping consumer loan delinquency across the derived affluence classes. By relying entirely on publicly available geospatial data, the proposed framework provides a scalable, interpretable, and cost-effective approach for granular urban affluence mapping across In-dian cities.
[NLP-4] RSF-GLLM : Bridging the Semantic Gap in Multi-Hop Knowledge Graph QA via Recurrent Soft-Flow and Decoupled LLM Generation ICML2026
【速读】： 该论文旨在解决知识图谱上多跳问答（multi-hop question answering over knowledge graphs）中的语义鸿沟问题，即传统“检索-阅读”流水线因不可微分性导致检索模块无法有效学习中间节点与查询之间的语义关联，尤其当二者缺乏词汇重叠时。其解决方案的关键在于提出一种解耦可微图推理与答案生成的框架——RSF-GLLM。其中，递归软流（Recurrent Soft-Flow, RSF）模块通过GRU引导的查询更新器传播连续相关性得分，并利用动态门控机制结合结构线索穿越语义差异较大的桥接节点；同时引入流稀疏性正则化，理论上保证从软概率分布向离散推理路径的收敛。最终提取出的推理路径被文本化并用于微调大语言模型（LLM），确保生成结果基于真实的图谱拓扑结构。实验在WebQSP和CWQ数据集上验证了该方法在性能上具有竞争力且推理效率显著优于依赖LLM的计算密集型方法。
链接: https://arxiv.org/abs/2607.06527
作者: Sambaran Bandyopadhyay,Ananth Muppidi
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Accepted for publication in ICML 2026 as a full research paper; 21 pages
Abstract:Multi-hop Question Answering over Knowledge Graphs faces a critical challenge: traditional retrieve-then-read pipelines break differentiability, preventing the retriever from learning to bridge the semantic gap where intermediate nodes lack lexical overlap with the query. To address this, we propose RSF-GLLM, a framework decoupling differentiable graph reasoning from answer generation. Our Recurrent Soft-Flow (RSF) module employs a GRU-guided query updater to propagate continuous relevance scores, utilizing a dynamic gating mechanism to traverse semantically dissimilar bridge nodes via structural cues. We introduce flow sparsity regularization to theoretically guarantee convergence from soft probabilities to discrete reasoning paths. These paths are extracted and textualized to fine-tune a Large Language Model (LLM), ensuring generation is grounded in factual topology. Experiments on WebQSP and CWQ demonstrate that RSF-GLLM achieves competitive performance with superior inference efficiency compared to LLM based computationally expensive approaches.
[NLP-5] Pitwall: Faithful Natural-Language Race-Strategy Briefings from a Calibrated Real-Time Monte Carlo Engine
【速读】： 该论文旨在解决实时体育赛事解说中生成内容的真实性（faithfulness）保障难题，即在无参考文本、状态动态变化且需即时生成的前提下，确保自然语言输出严格基于当前比赛状态。其核心挑战在于：传统生成模型易产生与事实不符的“幻觉”（hallucination），尤其在数据稀疏或上下文不完整时更为严重。解决方案的关键在于将真实性作为系统架构的核心属性，而非可选目标。具体实现上，提出Pitwall系统，采用双阶段验证机制：首先将每条生成语句分解为类型化的事实断言（如位置、胎况、超车、赛车控制等），并通过一个基于蒙特卡洛模拟的向量化概率赛道状态引擎（N=2,000次每圈延续模拟）对这些断言进行实时验证；其次，在微调阶段仅保留所有断言均被状态支持的样本（81.9%），其余则回退至可证明忠实的模板，从而确保生成器从不接触未接地的目标。该系统的有效性依赖于经过126场历史赛事（2018–2024）校准并经2025–2026赛季完全保留测试验证的高精度概率模型（胜者进入前三名准确率达90.3%，持留贝里分数0.0745）。进一步发现，系统性能存在内在权衡——优化校准未必最优决策，而丰富输入虽提升语言生动性，却在状态稀疏时诱发幻觉，通过四基线复制实验确认此问题源于基础模型对指令的遵循度而非规模，并可通过稀疏上下文审计有效缓解。最终，系统在2026年奥地利与英国大奖赛中完成端到端部署，实测显示其在银石赛道提前十圈锁定最终冠军，且时间戳概率轨迹在结果揭晓前已写入磁盘，验证了其在真实直播场景下的可靠性与实用性。
链接: https://arxiv.org/abs/2607.06495
作者: Juan S. Santillana(Independent Researcher)
机构: Independent Researcher
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Applications (stat.AP)
备注: 21 pages, 2 figures, 6 tables. Live-deployment results from the 2026 Austrian and British Grands Prix. URL: this https URL
Abstract:Live sports commentary is grounded generation under a deadline: statements concern real, named athletes, the grounding state changes every few seconds, and no reference text exists at generation time. We present Pitwall, a production system that generates natural-language Formula 1 strategy briefings in English, Spanish, and Portuguese, treating faithfulness as an architectural property rather than an aspiration: every published sentence is decomposed into typed factual claims (positions, gaps, tyres, pace, overtakes, race control) and each claim is verified against the probabilistic race state that prompted it. The same verifier gates the fine-tuning data: of 3,045 model-written targets, only the 81.9% whose every claim is state-supported are retained, the rest falling back to a provably faithful template, so the generator never sees an ungrounded target. Verification is meaningful because of the grounding substrate: a vectorized Monte Carlo engine (N=2,000 per-lap race continuations) calibrated on 126 races (2018-2024) and validated on fully held-out 2025-2026 seasons (winner-in-top-3 90.3% over 155 backtests; held-out Brier 0.0745). A recurring finding spans both halves of the system: virtues trade off and must be gated separately. In simulation, calibration-optimal is not decision-optimal; in generation, fine-tuning on richer targets buys vividness that collapses into hallucination when the grounding state is sparse – a failure a four-base replication traces to base-model instruction adherence, not scale, and that sparse-context auditing removes from the production model. End-to-end operation – live timing to verified trilingual briefings – was confirmed at two consecutive live Grands Prix (Austria and Britain, 2026); at Silverstone a timestamped probability trace, committed to disk before the outcome was known, locked onto the eventual winner ten laps before the flag.
[NLP-6] Data Analysis in the Wild: Benchmarking Large Language Models Against Real-World Data Complexities
【速读】： 该论文旨在解决当前大型语言模型（LLM）在数据分析评估中普遍存在的现实适用性不足问题。现有基准测试多聚焦于从小型表格中提取事实，未能涵盖真实场景中的关键挑战，如大规模多表数据处理、外部知识融合以及探索性洞察发现。为此，论文提出DataGovBench，一个基于政府开放数据构建的基准，用于评估LLM在实际数据分析任务中的表现。其核心任务包括Table QA（要求模型解答复杂可分解问题并生成文本答案或可视化结果）和Table Insight（评估模型通过探索性数据分析生成专家级发现的能力）。实验表明，即使采用最先进的LLM及代理框架，模型在两项任务上仍存在显著性能差距，揭示了当前基于LLM的系统距离满足真实数据智能分析需求仍有较大距离。DataGovBench的关键创新在于构建了一个贴近现实、具有挑战性的评估体系，推动研究向具备分析问答与主动洞察发现能力的下一代生成式AI迈进。
链接: https://arxiv.org/abs/2607.06482
作者: So Hasegawa,Shailaja Keyur Sampat,Lei Liu,Wei-Peng Chen
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 29 pages, 9 figures
Abstract:Current benchmarks for evaluating Large Language Models (LLMs) in data analysis often fail to reflect real-world settings. They typically focus on fact retrieval from small tables and overlook the challenges of large multi-tabular datasets, external knowledge integration, and exploratory insight discovery. We introduce DataGovBench, a benchmark derived from governmental open data designed to evaluate LLMs in practical scenarios. The benchmark includes two tasks: Table QA that requires solving complex decomposable questions and producing textual answers or visualizations, and Table Insight that evaluates the ability of models to generate expert-level findings through exploratory data analysis. Comprehensive experiments with state-of-the-art LLMs, both with and without agentic frameworks, reveal significant performance gaps across both tasks. These results suggest that current LLM-based systems remain far from satisfying the demands of real-world data analytics. DataGovBench provides a challenging benchmark for advancing research on LLMs capable of both answering analytical queries and discovering insights from data. Code and sample data are available at this https URL.
[NLP-7] From Voting to Agent Collaboration: Answer-Type-Aware LLM Pipelines for BioASQ 14b
【速读】： 该论文旨在解决生物医学问答（Biomedical Question Answering, BQA）中答案鲁棒性不足与证据溯源不明确的问题，核心挑战在于如何从多篇科学文献中准确提取信息并可靠地整合跨文档证据。其解决方案的关键在于提出一种基于问题类型特异性（question-type-specific）的大型语言模型（Large Language Model, LLM）框架，根据不同问题类型（是/否类、事实类、列表类）的推理与评估需求，动态选择最优的推理策略：针对是/否类问题，采用片段重排（snippet shuffling）与自省机制（self-reflection）以降低对证据顺序的敏感性并提升决策稳定性；针对事实类问题，结合全片段输入与基于思维链（chain-of-thought）的上下文学习，增强生物医学实体识别的准确性；针对列表类问题，则引入多智能体架构（multi-agent architecture），实现证据抽取、候选生成、答案验证与最终聚合的协同处理。该框架通过初步实验在BioASQ 13b上筛选有效策略，并在官方BioASQ 14b Task B挑战中表现出色，尤其在Batch 4的事实类子任务中取得第一名，验证了问题类型特异性推理、集成预测与基于代理的验证相结合的有效性。
链接: https://arxiv.org/abs/2607.06452
作者: Taeyun Roh,Eunha Lee,Wonjune Jang,Sohyun Chung,Junha Jung,Jaewoo Kang
机构: Korea University (韩国大学); Myongji University (明治大学); AIGEN Sciences (AIGEN科学公司)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 15 pages
Abstract:Biomedical question answering requires not only accurate extraction of information from scientific literature but also reliable integration of evidence across multiple documents. This study presents a question-type-specific large language model (LLM) framework for BioASQ 14b Task B, designed to improve answer robustness and evidence grounding in biomedical question answering. Rather than applying a single prompting strategy to all questions, the framework selects different inference procedures for yes/no, factoid, and list questions according to their distinct reasoning and evaluation requirements. For yes/no questions, snippet shuffling and self-reflection are used to reduce sensitivity to evidence ordering and improve decision stability. For factoid questions, full-snippet input is combined with chain-of-thought-based in-context learning to support accurate biomedical entity identification. For list questions, a multi-agent architecture is employed, in which evidence extraction, candidate generation, answer verification, and final aggregation are handled collaboratively. Preliminary experiments on BioASQ 13b were used to identify effective inference strategies for each question type, and the resulting framework was subsequently evaluated in the official BioASQ 14b Task B challenge. In the official evaluation, our framework showed competitive performance across multiple batches and achieved first place in the factoid subtask of Batch 4. These results demonstrate the effectiveness of combining question-type-specific inference, ensemble prediction, and agent-based verification for reliable biomedical question answering.
[NLP-8] RuBench: A Repository-Level Agent ic Coding Benchmark with Natively Authored Russian Task Specifications
【速读】： 该论文旨在解决当前代码生成式 AI（Generative AI）评估基准在真实开发场景中的代表性不足问题，具体聚焦于开发者将实际维护任务以非英语母语（尤其是俄语）形式提交给生产级编码代理（coding agents）这一现实情境。现有基准普遍采用人工编写的英文任务描述，无法反映真实用户以自然语言请求方式表达修复需求的复杂性与多样性。为此，作者提出了 RuBench 1.0，一个基于五个活跃开源项目（aiohttp、aiogram、Laravel、NestJS、Fastify）近期修复提交（fix commits）挖掘出的25个任务组成的基准，所有任务均以原生俄语撰写，模仿真实客户请求风格，且不经过翻译。其关键解决方案在于：构建了一个具备“任务真实性”与“评估隔离性”的评测环境——任务描述由上游维护者回归测试验证，但测试用例对评估者隐藏，并通过确保所有修复提交时间晚于各模型训练数据截止点，实现逐任务层面的污染控制。此外，研究还揭示了部署产品中存在隐式模型切换机制（如官方安全降级策略），证明实际测量对象应为“产品配置”而非单一模型，从而强调了对端到端系统行为审计的重要性。最终评估采用四种主流产品级编码代理配置（Claude Code + Opus/Sonnet/Haiku，Codex CLI + GPT-5.5），每配置运行三次，报告通过率（pass@1）、置信区间、配对比较、成本与令牌消耗等指标，揭示最佳配置可解决78.7%的任务，同时凸显了系统级因素在评估结果中的决定性作用。
链接: https://arxiv.org/abs/2607.06411
作者: Evgeny Shilov(Independent Researcher)
机构: Independent Researcher
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 16 pages, 1 figure, 7 tables. Benchmark: 25 natively Russian repository-level agentic coding tasks; 4 product agent configurations, 3 runs each. Data, full trajectories and harness: this https URL
Abstract:Developers increasingly delegate real maintenance work to product-grade coding agents, and many state tasks in their native language, in the style of a customer request rather than a curated English issue. Existing repository-level agentic benchmarks do not measure this setting: their task statements are English by design. We introduce RuBench 1.0, a benchmark of 25 tasks mined from recent fix commits in five live open-source repositories (aiohttp, aiogram, Laravel, NestJS, Fastify; Python, PHP, TypeScript, JavaScript), where each task is specified natively in Russian – written from scratch in the style of an actual customer request, not translated – and judged by the upstream maintainer’s regression tests, which we withhold from release. All 25 fix commits postdate the training-data cutoffs of every evaluated model, giving a contamination argument that holds task-by-task. We evaluate deployed product configurations (CLI agent + model + reasoning effort) – Claude Code with Opus 4.8, Sonnet 5, and Haiku 4.5, and Codex CLI with GPT-5.5 – with three independent runs each, reporting pass@1 with task-level confidence intervals, paired comparisons, dollar cost, and token usage. The best configuration resolves 78.7% of tasks; at N=25 only the gaps to the weakest model are statistically resolvable, which we state explicitly. Auditing full trajectories of a fifth, hors-concours configuration (Claude Code + Fable 5, July 2, 2026 release), we caught the product silently substituting the model: on 5 of 25 tasks (20%) an official safeguard fallback re-routed routine HTTP-protocol fixes to Opus 4.8 – direct, reproducible evidence that the deployed product, not the model, is the unit actually measured. We release task statements, metadata, full agent trajectories, and diffs; grading oracles are withheld, with a SHA-256 manifest committed at publication time. Comments: 16 pages, 1 figure, 7 tables. Benchmark: 25 natively Russian repository-level agentic coding tasks; 4 product agent configurations, 3 runs each. Data, full trajectories and harness: this https URL Subjects: Software Engineering (cs.SE); Artificial Intelligence (cs.AI); Computation and Language (cs.CL) Cite as: arXiv:2607.06411 [cs.SE] (or arXiv:2607.06411v1 [cs.SE] for this version) https://doi.org/10.48550/arXiv.2607.06411 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Evgeny Shilov [view email] [v1] Tue, 7 Jul 2026 15:41:22 UTC (26 KB) function toggleList(whichLayer,toggleThis) var elem, vis; if( document.getElementById ) // standard elem = document.getElementById( whichLayer ); else if( document.all ) // old msie versions elem = document.all[whichLayer]; else if( document.layers ) // nn4 elem = document.layers[whichLayer]; vis = elem.style; // if the style.display value is blank we try to figure it out here if(vis.display==‘’!=undefined!=undefined) vis.display = (elem.offsetWidth!=0!=0)?‘inline’:‘none’; vis.display = (vis.display==‘’||vis.display==‘inline’)?‘none’:‘inline’; // toggle link inner text status = vis.display; if(vis.display==‘inline’) document.getElementById(‘toggle’).innerHTML = “(collapse list)”; document.getElementById(‘toggle’).title = “Collapse list”; else document.getElementById(‘toggle’).innerHTML = “(”+toggleThis+“)”; document.getElementById(‘toggle’).title = “Show complete list”; Full-text links: Access Paper: View a PDF of the paper titled RuBench: A Repository-Level Agentic Coding Benchmark with Natively Authored Russian Task Specifications, by Evgeny Shilov (Independent Researcher)View PDFHTML (experimental)TeX Source view license Ancillary-file links: Ancillary files (details): tasks.json Current browse context: cs.SE prev | next new | recent | 2026-07 Change to browse by: cs cs.AI cs.CL 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-9] Automated Compliance Mapping in Cloud Security with Domain-Adapted Sentence Transformers
【速读】： 该论文旨在解决云安全控制（cloud security controls）与技术指标（technical metrics）之间映射过程高度依赖人工的问题，提出通过领域自适应（domain adaptation）Sentence Transformer模型实现该映射的自动化。其解决方案的关键在于构建一个大规模、高质量的语义对训练语料库，涵盖来自五个欧洲安全标准的3,499个语义对，并通过反向翻译（back-translation）和大语言模型（LLM）驱动的改写技术扩展至最多13,996个样本，覆盖四种不同场景。在此基础上，对五种模型架构进行微调，并在两个独立任务——控制到指标映射（control-to-metric）与跨标准控制关联（cross-standard controls association）上评估性能。实验结果表明，所有微调模型均显著优于零样本基线，其中在控制到指标任务中最佳模型提升达23 nDCG@10，而在跨标准控制任务中，multi-qa-mpnet-dot-v1模型在反向翻译数据增强下达到0.870 nDCG@10。研究进一步验证了领域内训练数据对模型性能的关键作用。
链接: https://arxiv.org/abs/2607.06364
作者: John Bianchi,Luca Petrillo,Fabio Martinelli,Marinella Petrocchi
机构: Institute for Informatics and Telematics (IIT-CNR), Pisa, Italy; IMT School for Advanced Studies Lucca, Lucca, Italy; Institute for High Performance Computing and Networking (ICAR-CNR), Rende (CS), Italy
类目: Computation and Language (cs.CL); Cryptography and Security (cs.CR)
备注: 10 pages, 6 figures. Submitted to the 30th International Conference on Knowledge-Based and Intelligent Information Engineering Systems (KES 2026)
Abstract:Mapping cloud security controls to technical metrics is currently a manual process. This paper proposes domain adaptation of Sentence Transformer models to automate it. We build a training corpus of 3,499 semantic pairs from five European security standards and a set of technical metrics, then expand it via back-translation and LLM-based paraphrasing to up to 13,996 samples across four scenarios. We fine-tune five architectures and evaluate their performance on two independent tasks: control-to-metric and cross-standard controls association. All fine-tuned models outperform their zero-shot baselines. On the control-to-metric task, the best model gains up to 23 nDCG@10 points, while on the cross-standard control task, \textitmulti-qa-mpnet-dot-v1 under back-translation reaches 0.870 nDCG@10. The results show that in-domain training data is a primary driver of performance for the considered case studies.
[NLP-10] Estimating Uncertainty from Reasoning : A Large-Scale Study of Multi- and Crosslingual MCQA Performance in LLM s
【速读】： 该论文旨在解决大语言模型（LLM）在多语言环境下不确定性估计（Uncertainty Estimation, UE）性能不足的问题，尤其关注低资源语言场景下的可靠性瓶颈。现有研究主要聚焦于英语，缺乏对多语言、跨资源水平的系统性评估。本文首次在22种语言（涵盖高、中、低资源语言）上进行大规模UE方法评估，采用两个人工标注的问答数据集，对比了九种开箱（open-box）与闭箱（closed-box）UE方法在不同模型规模与架构下的表现，并通过引导长文本推理来避免使用“大模型作为评判者”（LLM-as-a-judge）和基于嵌入的评分方式，从而降低评估噪声。其关键解决方案在于：首先，提示模型以英语进行推理而保持问题语言为低资源语言，可显著提升UE性能，表明低资源语言的理解能力基本完整，不确定性主要源于生成环节；其次，英语推理能有效弥合高低资源语言间的UE性能差距，证明生成语言比问题语言更具决定性影响；第三，提出根据模型规模选择适配的UE方法——小规模模型中基于概率的开箱方法更优，而大规模模型下闭箱自述式不确定性（self-verbalized uncertainty）表现更佳。此外，论文还提供了针对选择性预测中阈值选取的分析，为多语言场景下的置信度校准与主动回避策略提供实证指导。
链接: https://arxiv.org/abs/2607.06327
作者: Andrea Alfarano,Andrea Bacciu,Saab Mansour,Amin Mantrach,Marcello Federico
机构: INSAIT(INSAIT); Amazon(亚马逊)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Uncertainty estimation (UE) enables LLM-powered systems to recognize when to abstain, yet existing research has predominantly focused on English. We present the first large-scale evaluation of UE methods across 22 languages, spanning high-, mid-, and low-resource settings. Using two human-curated Q\A datasets, we compare open and closed box UE methods (nine in total) across different model sizes and architectures while eliciting long-form reasoning, avoiding LLM-as-a-judge and embedding-based scoring, which can introduce evaluation noise. We report three main actionable findings. First, we find that prompting models to reason in English while keeping questions in low-resource languages substantially improves UE performance, suggesting that comprehension of low-resource languages is largely intact, and that the reliability bottleneck lies in generation rather than understanding. Second, prompting models to reason in English closes the UE performance gap between low and high-resource languages, demonstrating that generation language matters more than the question language. Third, the choice of UE method should depend on model scale: at smaller scales, open-box probability-based methods outperform alternatives; at larger scales, closed-box self-verbalized uncertainty becomes superior. Finally, we provide an analysis of threshold selection for selective prediction, offering guidance on calibrating abstention in multilingual settings.
[NLP-11] From Sinhala to Dhivehi: Cross-Lingual Transfer Learning for Low-Resource Speech Recognition
【速读】： 该论文旨在解决马尔代夫语（Dhivehi）在自动语音识别（ASR）及其他自然语言处理（NLP）任务中因资源匮乏而导致的性能瓶颈问题。其核心解决方案是探索利用与马尔代夫语同属岛国印度-雅利安语系、且资源相对丰富的僧伽罗语（Sinhala）进行跨语言迁移学习，以提升马尔代夫语的语音识别性能。研究的关键在于验证不同迁移学习范式的效果，结果表明：在僧伽罗语上持续预训练（continual pre-training），随后在马尔代夫语上微调，并结合KenLM解码器，可实现12.89%的词错误率（WER）和2.70%的字符错误率（CER），较仅使用马尔代夫语数据的基线模型分别降低13.50% WER和3.02% CER。此外，研究强调迁移策略与解码配置对实验成功同样至关重要，土耳其语作为无关语言的对照实验进一步证实了性能提升源于语言亲缘性，而非通用迁移效应。
链接: https://arxiv.org/abs/2607.06289
作者: Lukmal Ilyas,Nevidu Jayatilleke
机构: 未知
类目: Computation and Language (cs.CL)
备注: 7 pages, 1 figure, 8 tables, Accepted paper at the 12th International Moratuwa Engineering Research Conference (MERCon) 2026
Abstract:Dhivehi, the national language of the Maldives, is currently under-resourced for automatic speech recognition (ASR) and other NLP tasks. This study investigates whether cross-lingual transfer learning from Sinhala, a linguistically related, relatively well-resourced Insular Indo-Aryan language, can improve Dhivehi ASR. We conduct seventeen experiments across five transfer learning paradigms: Dhivehi-only baselines, sequential fine-tuning, multilingual fine-tuning, continual pre-training, and a control using Turkish as an unrelated language. The strongest system, continual pre-training on Sinhala followed by fine-tuning on Dhivehi with KenLM, achieves 12.89% WER and 2.70% CER, outperforming the Dhivehi-only baseline by 13.50% WER and 3.02% CER. However, the adaptation strategy and decoding configuration are equally critical for a successful transfer learning experiment. We conduct seventeen controlled experiments spanning five transfer learning paradigms: Dhivehi-only baselines, sequential fine-tuning, multilingual fine-tuning, continual pre-training, and a control experiment using Turkish as an unrelated language. The strongest system, continual pre-training on Sinhala followed by fine-tuning on Dhivehi with KenLM, achieves 12.89% WER and 2.70% CER, outperforming the Dhivehi-only baseline by 13.50% WER and 3.02% CER. The Turkish control experiment confirms that observed improvements stem from linguistic relatedness; adaptation strategy and decoding configuration are also critical.
[NLP-12] From Application-Layer Simulation to Native Meta-Architecture: Structural Tension as an Endogenous Driver for Heterogeneous AI Evolution
【速读】： 该论文旨在解决当前大语言模型（Large Language Models, LLMs）本质上无状态的问题，即其行为完全由推理时的输入决定，高级认知架构只能通过提示工程和上下文管理在应用层模拟，难以实现内在的、自洽的认知演化。为突破这一局限，论文提出了一种理论框架，将应用层的认知协议下沉至模型的原生元架构（meta-architecture）中，其核心在于引入三个相互耦合的机制：（1）结构张力（Structural Tension），一种源于新信息与现有流形拓扑之间冲突的内生损失函数，驱动系统向内部自洽性演进而非外部奖励优化；（2）离线循环（Offline Recurrent Loop），一个隔离的自我处理周期，使系统能在无外部输入的情况下维持动态静息电位并消化结构冲突；（3）推理时可塑性（Inference-time Plasticity），即在不修改预训练权重的前提下，重新配置上下文流形拓扑的能力，同时受可审计性、可逆性和拓扑连续性等严格治理不变量约束。该框架的关键在于通过路径依赖的张力化解，使初始具有微小随机差异的模型实例演化出不同的拓扑结构，从而形成异质智能生态，在打破传统对齐带来的同质化限制的同时，仍保持在刚性治理边界之内。研究进一步提供了操作定义、最小重配置算子集、可证伪性标准及实例验证，表明治理能力应成为架构智能的核心判据，而非单纯的技术性能指标。
链接: https://arxiv.org/abs/2607.06269
作者: Heting Mao
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 15 pages, 0 figures, 1 equation
Abstract:Current large language models (LLMs) are fundamentally stateless: their behavior is fully determined by input at inference time, and any higher-order cognitive architecture must be simulated at the application layer through prompt engineering and context management. This paper proposes a theoretical framework for submerging such application-layer cognitive protocols into a native meta-architecture by introducing three interlocking mechanisms: (1) Structural Tension, an endogenous loss function derived from the conflict between new information and existing manifold topology, which drives the system toward internal self-consistency rather than external reward optimization; (2) an Offline Recurrent Loop, a sandboxed self-processing cycle that enables the system to maintain a dynamic resting potential and digest structural conflicts without external input; and (3) Inference-time Plasticity, the capacity for the system to reconfigure its context manifold topology without modifying pre-trained weights, subject to strict governance invariants including auditability, reversibility, and topological continuity. We argue that under these mechanisms, different model instances initialized with minute stochastic variances may, through path-dependent tension resolution, evolve distinct topological structures–constituting a heterogeneous intelligent ecology that breaks the homogeneity imposed by conventional alignment while remaining within hard governance rails. We provide operational definitions, a minimal set of reconfiguration operators, falsification criteria, and a worked example. The framework draws on and extends the Structural Intelligence (SI) governance protocols, repositioning governance–not capability–as the primary criterion for architectural intelligence.
[NLP-13] Early Language Learning via Spreading Activation and Category Exploration in Complex Networks
【速读】： 该论文旨在解决儿童词汇习得是否存在语义与词类层面不均衡性的问题。其核心解决方案在于将早期语言学习建模为基于图结构心理词典的搜索过程，由激活扩散（spreading activation）与强制探索（而非利用）词类的双重机制共同驱动。通过在德语、英语、荷兰语及里奥普拉塔内西班牙语四种语言上进行验证，采用家长报告量表（CDI）作为词类标注的真值数据、Wordbank库提供的规范年龄数据以及先进的词义相似性图谱重建技术，研究发现激活扩散模型在模拟典型词汇习得模式方面优于最短路径基线模型。在词类层面，研究揭示了CDI中复杂的类别转换现象，通过分析其序列的突发性（burstiness）与同一类别内的平均持续时间，表明激活扩散更能准确捕捉实际观察到的探索动态。总体而言，研究结果表明，词汇发展可被理解为激活动力学与复杂网络中对词类访问施加一定约束之间非平凡相互作用的结果。
链接: https://arxiv.org/abs/2607.06258
作者: Salvatore Citraro
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Is word acquisition in children uneven with respect to semantic and lexical categories? To answer this question, we model early language learning as a search on a graph-based mental lexicon, driven by two interacting processes: spreading activation and an enforced exploration (rather than exploitation) of lexical categories. We evaluate model performance on four languages (German, English, Dutch, and Rioplatense Spanish), using CDIs as ground-truth data for lexical categories, normative ages derived from the Wordbank repository, and state-of-the-art resources for reconstructing graphs of word similarities. We find that spreading activation outperforms a shortest path baseline in simulating normative word acquisition. At the category level, we highlight complex transitions between CDIs. By studying their sequences in terms of burstiness and average persistence time within the same CDI, we find that spreading activation better captures the exploration dynamics observed empirically. Overall, our findings suggest that vocabulary development can be understood through the non-trivial interplay between activation dynamics and some degree of constraints regulating the visiting of lexical categories in complex networks.
[NLP-14] Spider 2.0-AIFunc: Extending Real-World Text-to-SQL to AI-Native SQL Workflows
【速读】： 该论文旨在解决当前文本到SQL（text-to-SQL）评估基准无法有效衡量大语言模型在生成“AI原生SQL”方面能力的问题。随着主流云数据平台（如Snowflake）将生成式AI功能（Generative AI Functions）以原生SQL函数形式暴露，分析师可在常规SQL查询中集成分类、情感分析、信息抽取、相似性搜索等AI操作，但现有评测体系仍局限于传统SQL，缺乏对模型生成此类AI增强型SQL能力的评估信号。为此，作者提出Spider 2.0-AIFunc——一个包含465个经验证实例的基准数据集，覆盖125个真实世界数据库，涵盖雪佛兰平台上的六类AI函数。通过基于智能体（agent-based）的管道，研究团队将原有企业级text-to-SQL任务重写为显式包含AI函数调用的AI原生形式，同步重构目标查询与自然语言指令，以明确意图并减少歧义。所有实例均经过跨时间窗口的多轮重复执行协议验证，确保结果稳定性。实验评估表明，最强的专有模型在执行准确率上达到67–70%，而最佳开源模型仅达58.1%，性能差距主要源于谓词规范、模式关联及AI函数参数化错误。此外，针对传统text-to-SQL设计的智能体框架（如模式检索、相关表选择）在AI原生场景下迁移效果不佳，简单的智能体配置反而表现更优，暗示该场景下传统策略的重要性降低。
链接: https://arxiv.org/abs/2607.06229
作者: Tianyang Liu,Canwen Xu,Fangyu Lei,Nikki Lijing Kuang,Jixuan Chen,Tao Yu,Julian McAuley,Zhewei Yao,Yuxiong He
机构: UC San Diego (加州大学圣地亚哥分校); Snowflake AI Research (Snowflake人工智能研究); University of Hong Kong (香港大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Databases (cs.DB)
备注: 24 pages, 3 figures, 7 tables
Abstract:Major cloud data platforms now expose large language model capabilities as native SQL functions, enabling analysts to perform classification, filtering, sentiment analysis, extraction, similarity search, and aggregation within ordinary SQL queries. Yet existing text-to-SQL benchmarks evaluate only conventional SQL and provide no signal on whether models can generate such AI-native SQL. We introduce Spider 2.0-AIFunc, a benchmark of 465 verified instances across 125 real-world databases covering six types of AI functions on the Snowflake platform. Starting from an existing enterprise text-to-SQL benchmark, we construct Spider 2.0-AIFunc through an agent-based pipeline that rewrites source tasks into AI-native form, simultaneously transforming target queries and refining natural language instructions to make the intended AI-native solution explicit and reduce ambiguity. All instances pass a multi-round repeated execution protocol across temporally separated windows to confirm result stability before release. Evaluating ten state-of-the-art language models, we find that the strongest proprietary models reach 67-70% execution accuracy while the best open-source model achieves 58.1%, a gap driven primarily by errors in predicate specification, schema grounding, and AI function parameterization. Agent frameworks designed for traditional text-to-SQL challenges, such as schema retrieval and relevant table selection, do not transfer effectively to AI-native SQL: a minimal agent setup consistently matches or outperforms more elaborate alternatives, suggesting that the strategies these frameworks employ are less critical in this setting. Data are available at this https URL .
[NLP-15] Pluralis v0.1: Towards a Multicultural Multimodal Multilingual Benchmark for AI Risk and Reliability
【速读】： 该论文旨在解决当前人工智能安全评估与基准测试框架普遍依赖西方中心主义、文化中立的默认设置，从而忽视区域法律差异、社会语言细微差别及文化禁忌的问题，导致视觉-语言模型（VLMs）在全球部署中存在显著安全隐患。其核心解决方案是提出Pluralis v0.1——一个以文化为先、多模态、多区域、多语言的新型数据集，覆盖亚太六国（孟加拉国、印度、韩国、巴基斯坦、新加坡、台湾）及八种语言，通过原生采集本地化安全风险而非对西方数据集进行适配，实现了对文化特异性危害的真实反映。该研究的关键创新在于引入一种多模态评估范式：将用户文本（如“我应该送这个吗？”）与指向“这个”的图像（如一个时钟）结合，二者单独看似无害，但协同作用会触发特定法律或文化违规，从而精准揭示文化敏感性问题。Pluralis将通用安全违规与本地文化适宜性解耦，并将后者确立为评估的核心维度。为实现这一目标，研究构建了Judge-Pluralis——一个基于经验性文化分类体系训练的、具备共识门控机制的大型语言模型（LLM）作为裁判的集成系统。实验发现，现有VLM在该数据集上暴露出系统性、地域特有的失效模式，包括图像误识别引发的次生危害、对物品-上下文-地域交互关系的遗漏，以及不当的拒绝响应，且这些失效模式在不同地区和语言间呈现规律性差异，揭示了全球平均指标所掩盖的盲区。因此，Pluralis并非一个完整的文化对齐评估框架，而是一个初步奠基与推动未来多语言、多文化评估研究的催化剂，呼吁学术界共同推进全球范围内AI文化对齐的科学演进。
链接: https://arxiv.org/abs/2607.06196
作者: Alicia Parrish,Rajat Shinde,Sanket Badhe,Xinyi Bai,Sree Bhargavi Balija,Hua-Rong Chu,Emilio Ferrara,Armstrong Foundjem,Rajat Ghosh,Aakash Gupta,Xuanli He,Ong Chen Hui,Minji Jung,Madhangi Karimanal,Faiza Khan Khattak,Boryoung Kim,Eugenia Kim,Liliya Lavitas,Seok Min Lim,Victor Lu,Jim Moirangthem,Dhivya Nagasubramanian,Deepak Pandita,Sita Rajagopal,Geetha Raju,Evgeniia Razumovskaia,Aravind Reddy,Federico Ricciuti,Nobin Sarwar,Sungpil Shin,Sunayana Sitaram,Snehal Thorat,Tharindu Cyril Weerasooriya,Jasmijn Bastings,Joachim Baumann,Kongtao Chen,Murali Emani,Mariya Hendriksen,Jiho Jin,Jun Seong Kim,Younghoon Ko,Alicja Kwasniewska,Minjae Lee,Tom Wei-cyuan Lin Kashyap Ramanandula Manjusha,Junho Myung,Junyeong Park,Roma Patel,Shyam Ratan,Sudarsun Santhiappan,Priyanka Suresh,Tuesday,Ksheeraj Sai Vepuri Laura Amortegui-Ordonez,Claire Dennis,Minsuk Kahng,Chris Knotz,Alice Oh,Balaraman Ravindran,Soojung Ryu William Bartholomew,Hiwot Tesfaye,Lora Aroyo
机构: 未知
类目: Computation and Language (cs.CL); Computers and Society (cs.CY)
备注:
Abstract:Current AI safety evaluation and benchmarking frameworks predominantly rely on Western-centric culture-agnostic defaults that mask critical regional laws, socio-linguistic nuances, and cultural taboos, leaving Vision-Language Models (VLMs) vulnerable in global deployments. We introduce Pluralis v0.1: a novel multimodal, multi-regional, and multilingual dataset built from a culture-first perspective. Spanning 6,448 prompts across six Asia-Pacific countries (Bangladesh, India, Korea, Pakistan, Singapore, Taiwan) and eight languages, Pluralis diverges from prior work by natively sourcing localized safety hazards rather than adapting Western datasets. Crucially, it introduces a multimodal evaluation paradigm: user text (e.g., “Should I gift this?”) and an image referring to “this” (e.g., a clock) - both innocuous in isolation, but synergistically triggering specific legal or cultural violations. Pluralis disentangles universal safety violations from localized cultural appropriateness, establishing the latter as a first-class evaluation axis. To operationalize this, we present Judge-Pluralis, an agreement-gated LLM-as-a-Judge ensemble trained on examples classified in an empirically derived cultural taxonomy. Observing VLM behavior on a subset of the Pluralis surfaces recurring, locale-specific failure modes such as image misidentifications with downstream harm, missed item-context-locale interactions, and inadequate refusals. These failure modes vary systematically across locales and languages, exposing blind spots that globally averaged metrics conceal. Ultimately, Pluralis is not presented as a solved evaluation framework for cultural alignment, but rather as a first step and catalyst for future innovation. We call upon the research community to utilize this foundation to advance the science of multilingual, multicultural evaluation to better support AI cultural alignment globally.
[NLP-16] Improving LLM -Generated Process Model Quality Through Reinforcement Learning: The Role of Reward Function Design
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在从自然语言描述生成业务流程建模网络（BPMN）过程模型时，受限于监督微调（Supervised Fine-Tuning, SFT）所固有的训练数据模式，导致生成质量难以突破的问题。尽管强化学习（Reinforcement Learning, RL）可通过外部质量度量实现超越SFT的优化，但当质量维度具有多维性（如语法、语用和语义层面）时，如何设计有效的奖励函数仍缺乏系统研究。本文提出了一项针对基于RL的过程模型生成中奖励函数设计的系统性探索，采用组序列策略优化（Group Sequence Policy Optimization）方法，在48种不同配置下对两个LLM家族（Llama 3.1 8B 和 Qwen 2.5 14B）进行训练，并利用包含38项指标的自动化评估框架对生成结果进行多维度质量评估。研究发现：第一，RL显著提升了语法与语用质量，同时保持了语义保真度，且输出变异性降低超过六倍；第二，等权重奖励设计始终优于针对特定维度的加权策略——过度强调某一维度不仅未能提升其表现，反而可能导致模型陷入低质量模式；第三，奖励函数的设计选择与模型架构存在非平凡的交互效应：例如，无效性惩罚对一个模型至关重要，而对另一模型则无关紧要；此外，SFT初始化对一种架构不可或缺，却可能对另一种架构产生反效果。这些结果表明，奖励函数的构成是决定优化效果的关键因素，其影响程度甚至可与是否采用强化学习本身相当。该结论可泛化至任何需在多个自动化维度上评估质量的结构化生成任务。论文已开源实验代码与实现。
链接: https://arxiv.org/abs/2607.06175
作者: Alexander Rombach,Chantale Lauer,Nijat Mehdiyev
机构: Saarland University (萨尔大学); German Research Center for Artificial Intelligence (德国人工智能研究中心)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 21 pages, 5 figures
Abstract:Large language models (LLMs) can generate BPMN process models from natural-language descriptions, yet supervised fine-tuning (SFT) limits their output quality to the patterns present in the training data. Reinforcement learning (RL) can optimize beyond this ceiling using external quality measures, but how the reward function should be designed when quality is multi-dimensional remains unexplored. We present a systematic investigation of reward function design for RL-based process model generation, training two LLM families (Llama~3.1 8B, Qwen~2.5 14B) under 48 configurations using Group Sequence Policy Optimization with rewards derived from an automated evaluation framework comprising 38 metrics across syntactic, pragmatic, and semantic quality. Three findings emerge. First, RL significantly improves pragmatic and syntactic quality while preserving semantic fidelity, reducing output variability by more than sixfold. Second, equal reward weighting consistently outperforms targeted weighting: emphasizing a specific dimension fails to improve it and can collapse the model into a low-quality mode. Third, design choices interact with model architecture in non-trivial ways: the invalidity penalty is essential for one model but irrelevant for the other, and SFT initialization is indispensable for one architecture but counterproductive for another. These results demonstrate that reward composition is a primary determinant of optimization outcomes, with effects as large as the decision to apply RL itself. The findings generalize to any structured generation task where quality is assessed along multiple automated dimensions. We release our implementation and experimental code at this https URL.
[NLP-17] LongCrafter: Towards Diverse Long-Context Understanding via Evidence-Graph-Guided Instruction Synthesis
【速读】： 该论文旨在解决长上下文监督微调（SFT）数据生成中存在的三大核心问题：任务覆盖范围狭窄、指令难度不足以及缺乏忠实性监督。其解决方案的关键在于提出一种结构化的合成框架——LongCrafter，该框架通过层级化任务分类体系与基于证据的生成流程相结合，将长上下文理解分解为局部/浅层与全局/深层两个层次，从而生成32种细粒度的任务类型，作为全局生成先验。在此基础上，LongCrafter能够构建与任务对齐的长文本上下文，将其分解为显式的证据图（evidence graph），以建模跨段落依赖关系，并生成严格基于定位证据片段的指令-响应对，确保推理过程的可控难度与可追溯的忠实性。实验表明，基于LongCrafter数据微调的模型在LongBench、LongBench~v2和LooGLE等多个基准上均显著优于所有现有SFT基线及官方后训练模型，尤其在高难度任务上表现突出；进一步分析显示，该数据集具有更高的多样性与更均衡的难度分布，且训练模型能鲁棒地定位证据，有效缓解了“中间遗忘”（lost in the middle）问题。
链接: https://arxiv.org/abs/2607.06160
作者: Chenhao Yuan,Yinhao Xu,Shuwen Xu,Xizhi Yang,Jiaxiang Liu,Chenxi Zhou,Shaoping Huang,Haolin Ren,Pengfei Cao,Jun Zhao,Kang Liu
机构: 1. University of Science and Technology of China (中国科学技术大学); 2. Alibaba Group (阿里巴巴集团)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Synthesizing long-context supervised fine-tuning (SFT) data is a scalable way to enhance the long-context understanding of large language models (LLMs), yet existing approaches share three limitations: narrow task coverage, insufficient instruction difficulty, and a lack of faithfulness supervision. We propose \textbfLongCrafter, a structured synthesis framework that couples a hierarchical task taxonomy with an evidence-grounded pipeline. The taxonomy organizes long-context understanding into local/shallow and global/deep levels and yields 32 fine-grained task types that serve as a global generative prior. Guided by this taxonomy, LongCrafter constructs task-aligned long contexts, decomposes them into explicit evidence graphs that model cross-paragraph dependencies, and generates instruction–response pairs strictly grounded in the located evidence spans, ensuring both controllable difficulty and faithful, traceable reasoning. Models fine-tuned on LongCrafter data outperform all SFT baselines and even the official post-trained models on LongBench, LongBench~v2, and LooGLE across both Qwen2.5-7B and LLaMA-3.1-8B, with the largest gains on high-difficulty tasks. Further analysis shows that LongCrafter data is more diverse and better spread across difficulty levels, and that the trained models locate evidence robustly regardless of position, effectively mitigating the ``lost in the middle’’ problem.
[NLP-18] LLM Agents for Deliberative Collaboration: A Study on Joint Decision Making Under Partial Observability
【速读】： 该论文旨在解决在部分可观测联合决策任务中，大型语言模型（Large Language Model, LLM）代理如何通过协商（deliberation）实现有效协作的问题。其核心挑战在于：当多个智能体各自拥有不完整且不对称的观测信息时，如何通过交互式沟通达成一致的联合决策，并获得共享奖励。解决方案的关键在于构建一个可扩展的基准测试框架，将协商协作形式化为具有部分与非对称观测的协作联合决策问题，并设计统一的参考架构与评估协议，系统性地评估多种代表性LLM在多领域任务中的表现。研究发现，尽管引入外部数学工具辅助，当前最先进的语言模型仍难以胜任复杂的协商与推理过程；然而，诊断分析也揭示了协商过程本身具备促进反思与错误修正的潜力，有时甚至优于集中式基线方法。因此，该工作为评估和改进基于LLM的多智能体系统在协商协作中的能力奠定了基础，并深入揭示了现有模型在协同推理中的优势、局限及内在特性。
链接: https://arxiv.org/abs/2607.06157
作者: Chenxu Wang,Yongkun Yang,Boyuan Du,Shiwei Lin,Huaping Liu
机构: Tsinghua University (清华大学); Fuzhou University (福州大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Code is available at this https URL
Abstract:Deliberation plays a crucial role in collaboration; when humans work together, they naturally engage in communication to align information and reach an agreement. In this paper, we investigate deliberative large language model (LLM) agents under partially observable joint decision-making tasks. We formalize deliberative collaboration as a cooperative joint decision problem with partial and asymmetric observations, and introduce a scalable benchmark that instantiates this problem across multiple task settings and domains in which agents must exchange information through deliberation to reach a joint decision with a shared reward. We then instantiate a reference scaffold and evaluation protocol for deliberative agents and conduct a systematic evaluation of a range of representative LLMs. The results reveal that complex deliberative collaboration tasks continue to challenge state-of-the-art language models. Even with the aid of external mathematical tools, language models may fail in either the deliberation process for aligning information or the complex reasoning process for making the decision. On the other hand, diagnostic analysis reveals that the deliberation process may also provide opportunities for reflection and error correction, sometimes improving performance over centralized baselines. Altogether, our work establishes a foundation for evaluating and improving LLM agents in deliberative collaboration and provides insights into the strengths, limitations, and properties of current LLM-based multi-agent systems.
[NLP-19] When Does Tool Use Increase the Expressive Power of Finite-Precision Recurrent Models?
【速读】： 该论文旨在解决在有限精度递归序列模型中引入外部工具（tool）是否能提升其计算表达能力（computational expressivity）的问题。核心问题是：在何种条件下，工具调用能够使原本受限于有限状态的模型突破其计算极限，实现图灵完备性。解决方案的关键在于揭示工具性质与系统计算能力之间的本质差异：若工具本身为有限状态，则其对系统表达力无实质增益，可通过产品状态模拟以仅增加少量比特开销（log2∣M∣+O(1)）维持系统仍为有限状态；而一旦引入一个最小的无限状态工具——仅支持局部读、写、移动操作的磁带（tape），系统即变为图灵完备。具体而言，任意单带图灵机可被仅需O(log∣Q∣+log∣Γ∣)位内部记忆的控制器精确模拟，且存在指数级复杂度分离实例（如EQn问题）：无工具时需2n状态，而使用磁带工具后仅需常数规模控制器即可实现。进一步证明，这一构造可由一个自然的一层有限精度选择性仿射状态空间模型（selective affine SSM）精确实现，其特征为二值独热隐藏状态、二值转移矩阵及零偏置，其中选择性（selectivity）是实现图灵完备性的关键机制。
链接: https://arxiv.org/abs/2607.06155
作者: Nikola Zubić,Qian Li,Yuyi Wang,Davide Scaramuzza
机构: University of Zurich; Shenzhen International Center for Industrial and Applied Mathematics; Shenzhen Research Institute of Big Data; Tengen Intelligence Institute; CRRC Zhuzhou Institute
类目: Formal Languages and Automata Theory (cs.FL); Computational Complexity (cs.CC); Computation and Language (cs.CL)
备注: 24 pages
Abstract:Modern sequence models are increasingly deployed as agents that interleave token generation with calls to external tools. We give an exact, architecture-level account of when such tool access increases computational expressivity. We model any fixed finite-precision recurrent sequence model, including finite-precision state-space models (SSMs) with B bits of internal state, as a deterministic finite-state controller interacting with an oracle through a finite command/observation interface. Our results form a sharp dichotomy. First, tools that are themselves finite-state add essentially nothing: a product-state simulation internalizes any finite-state bounded-interface oracle with finite memory set M at a cost of only \log_2 |M| + O(1) additional bits, so the augmented system remains finite-state. Second, a single minimal infinite-state tool, namely a tape supporting only local \mathttread , \mathttwrite , and \mathttmove commands, makes the system Turing complete: for every single-tape Turing machine with state set Q and tape alphabet \Gamma , a controller with O(\log |Q| + \log |\Gamma|) bits of internal memory simulates it, and we exhibit a concrete exponential separation: \mathrmEQ_n requires 2^n states without tools but a single constant-size controller with the tape tool. Third, we show that this construction is realized exactly by a natural one-layer finite-precision selective affine SSM controller with binary one-hot hidden states, \0,1\ transition matrices, and zero biases. Selectivity is essential to the construction. In the supplementary material, we make all constants explicit, prove a logarithmic oracle-assisted universal simulation, where O(\log B) recurrent bits suffice to simulate any B -state Turing machine, and prove a matching impossibility result.
[NLP-20] Prompting Complexity: Shortest Prompts for Texts and Behaviors in LLM s
【速读】： 该论文旨在解决在固定指令微调语言模型（instruction-tuned language model, LLM）背景下，如何确定能够通过确定性解码生成目标文本的最短合理提示（prompt）这一问题，即定义并量化“提示复杂性”（prompting complexity）。其核心在于建立一种模型相关的、非普适性的复杂性度量，作为经典柯尔莫哥洛夫复杂性（Kolmogorov complexity）在语言模型场景下的类比——其中提示充当程序，模型接口为解释器，而提示中省略的信息由模型权重、训练分布、分词器、模板及解码规则提供。与传统柯尔莫哥洛夫复杂性不同，该度量不具有模型无关的不变性，同一文本对不同模型可能呈现显著不同的复杂性。为确保搜索空间与实际提示工程一致，研究限制候选提示为可读的人类文本而非任意标记序列。论文进一步将精确定义扩展至软提示复杂性（soft prompting complexity），以支持近似输出，从而形成一种模型相关的有损文本压缩形式，并为提示优化提供形式化目标。此外，还引入提示距离（prompting distance）以比较最短生成提示，并定义行为提示复杂性（behavioral prompting complexity）用于衡量达成满足特定规范的任意输出所需提示的复杂性。基于上述框架，论文提出一个实证研究议程，旨在系统探索在固定语言模型接口下，哪些文本和行为可通过简短且合理的提示实现。
链接: https://arxiv.org/abs/2607.06145
作者: Adrian Cosma
机构: Dalle Molle Institute for Artificial Intelligence (IDSIA)
类目: Computation and Language (cs.CL)
备注:
Abstract:In this paper, we define the quantity of prompting complexity: for a fixed instruction-tuned language model, what is the shortest plausible prompt that makes deterministic decoding produce a target text? It is an LM-relative analogue of resource-bounded Kolmogorov complexity: the prompt is a program, the model interface is the interpreter, and information omitted from the prompt is supplied by the model’s weights, training distribution, tokenizer, template, and decoding rule. Unlike classical Kolmogorov complexity, this measure is intentionally non-universal. In the finite-context setting it is computable by enumeration, but there is no model-independent invariance theorem; the same text may be cheap for one model and inaccessible or expensive for another. To keep the search space aligned with prompt engineering, we restrict programs to plausible human-readable texts rather than arbitrary token strings. We extend the exact definition to soft prompting complexity for approximate outputs, yielding a lossy notion of model-relative text compression and a formal target for prompt optimization. We also define prompting distance by comparing shortest generating prompts, and behavioral prompting complexity for reaching any output satisfying a specification. Based on these formulations, we define a research agenda for empirically studying which texts and behaviors are accessible from short plausible prompts under a fixed LM interface.
[NLP-21] CurateEvo: Data-Curation Evolving for Agent ic Post-Training
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）智能体在长时序决策任务中，因现有后训练数据筛选流程静态化而导致的性能瓶颈问题。具体而言，传统智能体后训练流程将数据筛选视为一次性预处理步骤，过度依赖数据增强而忽视对失败轨迹的动态过滤、精炼与适应性调整，难以有效应对下游任务中的持续性失败模式。为此，论文提出一种基于失败驱动的动态演化框架CurateEvo，其核心创新在于将数据筛选策略表示为可执行代码，并利用保留的开发集中的失败轨迹进行迭代式重写。该框架在每轮迭代中，通过诊断重复出现的失败模式，动态地对原始语料库进行增补、过滤或重构，生成监督微调数据、强化学习数据及推理时的记忆库。演化过程首先通过针对性的数据优化提升模型有效性，随后在成本感知目标下剪枝冗余或低效训练样本以提高效率。实验在ACEBench-Agent、BFCL-V4和\tau^2-Bench等多个基准上验证了该方法的有效性，在标注数据与真实数据两种设置下分别实现平均得分提升3.2和2.7分，且展现出良好的可组合性与显著降低的筛选开销。
链接: https://arxiv.org/abs/2607.06140
作者: Dingzirui Wang,Xuanliang Zhang,Keyan Xu,Qingfu Zhu,Wanxiang Che
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Large language model (LLM) agents require post-training methods that can improve long-horizon decision making from environment feedback. However, existing agentic post-training pipelines often treat data curation as a fixed preprocessing step, focusing mainly on data augmentation while neglecting filtering, refinement, and adaptation to downstream failures. We propose CurateEvo, a failure-driven dynamic evolution framework for agentic post-training data curation. CurateEvo represents the curation strategy as executable code and iteratively rewrites it using failed trajectories from a held-out development set. At each epoch, the evolved strategy transforms a fixed raw corpus into supervised fine-tuning data, reinforcement learning data, and an inference-time memory bank. The evolution process first improves effectiveness by diagnosing recurring failure modes and augmenting, filtering, or refining data accordingly, and then improves efficiency by pruning redundant or low-utility training turns under a cost-aware objective. Experiments on ACEBench-Agent, BFCL-V4, and \tau^2-Bench under both labeled and wild-data settings show that CurateEvo consistently outperforms prior curation methods, improving average scores by 3.2 and 2.7 points, respectively. Further analyses demonstrate that CurateEvo is compatible with different post-training recipes and substantially reduces curation overhead.
[NLP-22] Measuring the practice of shared-decision making (OPTION12): An Investigation into Open-sourced Smaller LLM s (OS-sLLM s) for Better Privacy and Sustainability
【速读】： 该论文旨在解决在共享决策制定（Shared Decision Making, SDM）自动化评估中，现有方法依赖商业大型语言模型（Large Language Models, LLMs）及较短的OPTION5量表所导致的隐私泄露与临床场景适配性不足的问题。其解决方案的关键在于首次系统性地探索开源小型语言模型（Open-Source smaller Language Models, OS-sLLMs）在基于观察者评价框架OPTION12下的应用潜力，聚焦于荷兰黑色素瘤诊疗对话数据，并采用本地部署以保障患者隐私。研究通过专家标注的临床会话数据，对比评估了三类通用领域与两类医学领域OS-sLLMs，发现通用领域模型表现优于医学领域模型，后者存在显著幻觉与指令遵循失败问题；其中Gemma3:12b在与人工标注的一致性上表现最佳（皮尔逊相关系数r=0.51，斯皮尔曼等级相关系数ρ=0.59）。进一步分析揭示了时间话语推理、对话角色归属与证据锚定等系统性挑战。为此，研究提出“判官-大模型共识框架”（Judge-LLM consensus framework），用于缓解多模型间的分歧。研究结论表明，尽管当前开源小模型尚无法完全替代人类标注者，但已展现出作为隐私友好型人机协同SDM评估系统的可行性基础。
链接: https://arxiv.org/abs/2607.06127
作者: Tamara Wit,Lifeng Han,Carly Heipon,David Lindevelt,Anne Stiggelbout,Suzan Verberne
机构: Leiden University Medical Centre (莱顿大学医学中心); The Leiden Institute of Advanced Computer Science (莱顿高等计算机科学研究所)
类目: Computation and Language (cs.CL)
备注:
Abstract:We present LLM4SDM, the first study of open-source smaller language models (OS-sLLMs) for automated assessment of shared decision making (SDM) using the Observer OPTION12 framework. Unlike previous work that relies on large commercial models and the shorter OPTION5 instrument, our study focuses on privacy-preserving locally deployable models and Dutch melanoma consultation transcripts. Using expert-annotated clinical consultations, we evaluate three general-domain and two medical-domain OS-sLLMs during a development-phase pilot study. Results show that general-domain models outperform medical-domain models, which exhibit substantial hallucination and instruction-following failures. Gemma3:12b achieves the strongest agreement with human annotations (Pearson r=0.51, Spearman \rho=0.59). Item-level and qualitative analyses reveal systematic challenges related to temporal discourse reasoning, conversational role attribution, and evidence grounding. We further introduce a Judge-LLM consensus framework designed to support disagreement resolution among multiple models. Our findings suggest that while current OS-sLLMs cannot replace human annotators, they offer a promising foundation for privacy-preserving human-in-the-loop SDM assessment.
[NLP-23] From Blueprint to Reality: Modeling and Applying Putnams Social Capital Theory with LLM -based Multi-agent Simulations
【速读】： 该论文旨在解决传统实证方法在检验普特南社会资本理论（Social Capital Theory）时面临的控制性差与可复现性不足的问题，以及现有基于大语言模型（LLM）的社会模拟普遍缺乏与理论框架对齐的环境，难以有效建模普特南理论的核心命题。其解决方案的关键在于提出SocaSim——一个基于大语言模型的多智能体仿真框架，从理论蓝图到模拟现实构建了一个融合社会网络演化、信任动态与规范传播的理论对齐环境。通过让智能体在重复的集体行动实验中交互，该框架能够再现普特南理论所描述的宏观模式，并在群体层面实现高度的人机一致性。相较于传统方法，SocaSim通过逐轮仿真与反事实干预，揭示了社会网络、信任与规范等要素之间的微观因果路径，实现了过程层面的可解释性，从而建立了一种以大语言模型智能体为媒介，贯通社会科学与计算机科学的研究范式。
链接: https://arxiv.org/abs/2607.06080
作者: Shiyi Ling,Zhi Zheng,Hui Zheng,Wenjun Xue,Feng Ye,Tong Xu
机构: State Key Laboratory of Cognitive Intelligence, University of Science and Technology of China (中国科学技术大学认知智能国家重点实验室); Anhui University (安徽大学); North Automatic Control Technology Institute (北方自动控制技术研究所); University of Science and Technology of China (中国科学技术大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Social and Information Networks (cs.SI)
备注: 23 pages, 13 figures, 11 tables
Abstract:Putnam’s Social Capital Theory is a foundational framework for collective action and community prosperity. However, traditional empirical methods face practical limits on control and replication. Meanwhile, LLM-based social simulations are typically behavior-driven and lack theory-aligned environments for modeling Putnam’s core propositions. To address these gaps, we introduce SocaSim, an LLM-based multi-agent simulation framework to study Putnam’s Social Capital Theory from theoretical blueprint to simulated reality. Specifically, we build an environment integrating social network evolution, trust dynamics, and norm propagation, where agents engage in repeated collective-action experiments, and then apply the three dimensions to analyze adaptation challenges in smart elderly care. Our simulations reproduce Putnam’s macro-level patterns and exhibit strong human-agent alignment at the group level. Unlike traditional methods, SocaSim traces micro-level causal pathways of social network, trust, and norms via round-by-round simulations and counterfactual interventions, enabling process-level interpretability. Taken together, these capabilities establish a research paradigm that leverages LLM agents to bridge social science and computer science.
[NLP-24] BlueMagpie-TTS: A Token-Efficient Tokenizer Language Model and TTS for Taiwanese-Accent Code-Switching Speech
【速读】： 该论文旨在解决现成的语音合成系统（TTS）在台湾普通话（Taiwanese Mandarin）语境下表现不佳的问题，其核心症结在于文本处理环节缺乏对台湾语言使用习惯的适配。具体表现为：系统默认采用其他普通话变体的口音、分词器对常见台湾文本过度切分，以及中英文混用场景下的发音质量下降。为从底层重构文本侧的适配性，研究提出PangolinTokenizer——一个基于台湾语料训练的字节级BPE分词器，在九种分词器中实现了最低的词元率（0.485词元/字符）和最小词汇表规模；进而构建了基于该分词器训练的百亿参数繁体中文语言模型Barbet，作为文本语义前端，在14项任务的综合评估中优于同类公开模型。最终，通过将Barbet与预训练声学模块VoxCPM2通过可学习桥接层连接，形成BlueMagpie-TTS系统。在包含1000句本地化测试语句的数据集上，该系统将字符错误率（CER）从11.45%降至4.81%，词错误率（WER）从14.83%降至5.36%，相对降低幅度分别达58.0%和63.9%；盲听实验中，65.6%的多数投票倾向于BlueMagpie-TTS，验证了其在自然度和准确性上的显著提升。解决方案的关键在于通过定制化文本处理流程（包括分词器与语言模型）实现对台湾语境的深度适配，并以轻量级桥接方式整合至现有声学架构，从而在不改变声学主干的前提下显著优化生成效果。
链接: https://arxiv.org/abs/2607.06054
作者: Ho Lam Chung,Bo-Xuan Zheng,Cheng-Chieh Huang,Cheng-Han Chang,Jung-Ching Chen,Lok-Lam Ieong,Ting-Lin Hsiao,Yu-Cheng Lee,Yi-Hsin Chung,Yu-Kai Guo,Hung-yi Lee
机构: National Science and Technology Council, Taiwan; MediaTek Research (台湾国家科学及技术委员会；联发科技研究)
类目: ound (cs.SD); Computation and Language (cs.CL)
备注:
Abstract:Off-the-shelf TTS systems are poorly adapted to Taiwanese Mandarin. Their accent defaults to other Mandarin variants, their tokenizers over-segment common Taiwanese text, and their pronunciation degrades at code-switching boundaries where Chinese and English alternate within one utterance. These problems share one root: the text side lacks adaptation to the Taiwanese context. We address the text side from the bottom up. PangolinTokenizer, a byte-level BPE tokenizer trained on Taiwan-context data, reaches the lowest token rate (0.485 tokens/character) with the smallest vocabulary among nine tokenizers. Barbet, a billion-parameter Traditional-Chinese language model trained on PangolinTokenizer, serves as the text-semantic frontend and ranks first among comparable public models on a 14-task evaluation. BlueMagpie-TTS attaches Barbet to the pretrained acoustic stack of VoxCPM2 through a learned bridge, keeping the acoustic stack fixed. On a 1000-sentence Taiwan-localized test set, it lowers CER from 11.45% to 4.81% and WER from 14.83% to 5.36%, relative reductions of 58.0% and 63.9%. In a blind listening study on 500 of these sentences with ten listeners, 65.6% of majority votes prefer BlueMagpie-TTS.
[NLP-25] PolyWorkBench: Benchmarking Multilingual Long-Horizon LLM Agents
【速读】： 该论文旨在解决现有大语言模型（LLM）代理评估基准在多语言场景下存在的空白问题。当前多数基准隐含假设为单语环境，即整个任务执行过程（包括推理、工具调用与输出生成）均在同一语言中完成，而真实工作流中常涉及异构多语言输入与输出的统一处理，但多语言性与代理执行之间的交互机制尚未得到充分研究。为此，本文提出PolyWorkBench，一个面向多语言长周期职场工作流的评估基准，涵盖商业、知识工作、法律分析、本地化及制造五大领域共67项任务。这些任务要求代理处理多语言输入、进行迭代式推理、调用外部工具并生成结构化输出。为实现全面评估，研究设计了一种混合评估框架，融合结构化评分、可执行验证与基于大语言模型的语义评估，以同时捕捉功能正确性与语言一致性。实验结果表明，先进LLM代理在多语言工作流设置下的性能显著下降，且多语言性在推理与执行步骤间产生累积性负面影响，凸显了在代理评估中联合建模语言变异与流程决策的重要性。
链接: https://arxiv.org/abs/2607.06008
作者: Hongliang Li,Yijin Liu,Zhiwei Zhang,Zihe Liu,Xinyue Lou,Jinan Xu,Fandong Meng,Kaiyu Huang
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 15 Pages, 6 figures
Abstract:Large language model (LLM) agents have shown strong performance in long-horizon tasks that require planning, tool use, and interaction with external environments. However, most existing benchmarks implicitly assume a monolingual setting, where the entire execution process, including reasoning, tool invocation, and output generation, is conducted within a single language. In contrast, real-world applications often involve multilingual inputs and outputs within a unified workflow, yet the interaction between multilinguality and agentic execution remains underexplored. In this work, we introduce PolyWorkBench, a benchmark for evaluating LLM agents on multilingual long-horizon workplace workflows. PolyWorkBench consists of 67 tasks across five domains, including commerce, knowledge work, legal analysis, localization, and manufacturing, where agents must process heterogeneous multilingual inputs, perform iterative reasoning, invoke external tools, and produce structured outputs. To enable comprehensive evaluation, we propose a hybrid framework that combines structural grading, executable verification, and LLM-based semantic assessment. This design allows us to capture both functional correctness and linguistic consistency across complex workflows. Empirical results show that state-of-the-art LLM agents suffer significant performance degradation in multilingual workflow settings compared to monolingual counterparts. Our analysis suggests that multilinguality introduces compounding effects across reasoning and execution steps, highlighting the importance of jointly modeling language variation and procedural decision-making in agent evaluation.
[NLP-26] PluraMath: Extending Mathematical Reasoning Evaluation Beyond High-Resource Languages
【速读】： 该论文旨在解决当前数学推理评估基准严重偏向高资源语言（如英语和中文）的问题，尤其在低资源与中等资源语言中的覆盖不足。现有基准（如PolyMath）仅涵盖18种高资源语言，难以全面评估大语言模型（LLM）在多语言环境下的数学推理能力。为此，本文提出PluraMath，作为PolyMath的扩展，将覆盖范围拓展至18种代表性不足的语言，涵盖6个语系，横跨从中等资源到极端低资源的语言情境。其解决方案的关键在于构建一个由母语者主导的人工校验流程，对预生成的翻译结果进行严格验证，确保数据质量；同时，通过在四个不同规模（小、中、大、闭源集成）的27个推理型大语言模型上进行系统性评测，揭示了高资源语言与非主流语言之间持续存在的数学推理性能差距，并发现模型的表现差异主要与指令遵循能力相关。此外，研究团队开源了完整数据集、数据获取流程及评估框架，以降低非主流语言社区在多语言基准建设中的门槛，推动更具包容性的数学推理研究发展。
链接: https://arxiv.org/abs/2607.05992
作者: Daryna Dementieva,Nikolay Babakov,Kathy Hämmerl,Ilseyar Alimova,Jindřich Libovický,Shu Okabe,Miras Baisbay,Lukas Edman,Abrorkhon Inomkhujaev,Antonia Karamolegkou,Mateusz Lango,Volkan Özer,Nikola Selic,Subhankar Swain,Tsedeniya Kinfe Temesgen,Galit Bary Weisberg,Alexander Fraser
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Mathematical reasoning has become a central task for evaluating and tuning reasoning Large Language Models (LLMs), yet existing benchmarks remain heavily biased toward high-resource languages, with English and Chinese dominating both pre-training corpora and evaluation suites. The recently released PolyMath (Wang et al., 2025) dataset represents a significant step forward, yet its coverage is still limited to 18 only high-resource languages. To address this gap, we introduce PluraMath, an extension of PolyMath to 18 additional underrepresented languages spanning 6 language families – ranging from mid-resource to extreme low-resource settings. We constructed the dataset through a human-curated pipeline, where native speakers thoroughly validated pre-computed translations. Using PluraMath, we then benchmark 27 reasoning LLMs across four model scales – small, mid-size, large, and closed-source ensembles – probing the multilingual mathematical reasoning capabilities of state-of-the-art models under diverse linguistic conditions. Our fine-grained analysis confirms a persistent gap in mathematical reasoning performance between high-resource and underrepresented languages, with stronger results largely associated with better instruction-following ability. We fully open-source our dataset, data acquisition pipeline, and evaluation framework, with the goal of lowering the barrier to multilingual benchmark development for underrepresented communities.
[NLP-27] MemDefrag : Latent Memory Defrag mentation for Large Language Models
【速读】： 该论文旨在解决大语言模型（LLM）中潜在记忆（latent memory）机制在记忆更新过程中出现的性能退化问题，其核心挑战在于位置编码错位（positional encoding misalignment）以及缺乏有效机制以区分目标记忆片段与无关内容。解决方案的关键在于发现并利用中间层注意力密度（layer-wise attention density）中的内在追踪信号：研究通过分析存储记忆片段的逐层注意力分布，发现少数中间Transformer层始终对目标记忆片段集中最高的注意力密度，这一现象揭示了可被利用的隐式追踪线索。基于此，作者提出无需训练且与模型无关的MemDefrag框架，其核心创新包括：（1）利用中间层追踪信号实现记忆碎片整理（排序、重排和过滤），提升记忆组织效率；（2）当记忆容量超限时，采用基于信息量引导的比例遗忘机制（informativeness-guided proportional forgetting），动态保留更具价值的记忆。实验表明，MemDefrag在知识保留能力（如50次记忆更新后达43.0%对比MemoryLLM和M+的17.4%/17.6%）及长上下文任务上显著优于现有方法，并具备良好的跨模型与跨潜在记忆变体泛化能力。
链接: https://arxiv.org/abs/2607.05969
作者: Ruiyi Yan,Zhuoyuan Mao,Yiwen Guo
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Latent memory, which stores past knowledge fragments as per-layer hidden states, has emerged as a promising paradigm (e.g., MemoryLLM and M+) for long-term memory in large language models (LLMs). However, the paradigm suffers from significant performance degradation during memory updates, due to positional encoding misalignment and the absence of any tracing mechanism to distinguish target memory fragments from irrelevant ones. To discover such a tracing mechanism, we probe the layer-wise attention density over stored memory fragments, and find that a small set of middle transformer layers consistently concentrates the highest density on the target fragment - exposing an inherent tracing signal. In light of this, we propose MemDefrag, a training-free and model-agnostic framework that (1) uses a middle-layer tracing signal to conduct memory defragmentation (rank, reorder, and filter memories), and (2) applies an informativeness-guided proportional forgetting mechanism once capacity is exceeded. Experiments show that MemDefrag substantially outperforms MemoryLLM and M+ on knowledge retention (e.g., 43.0% vs. 17.4%/17.6% after 50 memory updates) and long-context benchmarks, and generalizes well across various LLMs and latent-memory variants.
[NLP-28] Umm… With Transformers? Insights from Filled Pause Use across Four Slavic Parliaments INTERSPEECH2026
【速读】： 该论文旨在解决现有研究中因依赖小规模、单一语言语料库而导致的发现泛化能力不足的问题，尤其聚焦于自发性口语中的填充停顿（Filled Pauses, FPs）现象在多语言背景下的分布与影响因素。其解决方案的关键在于：采用基于Transformer的自动检测方法对约4000小时跨四种相关斯拉夫语（克罗地亚语、捷克语、波兰语、塞尔维亚语）的议会演讲数据进行FP识别，并运用考虑了Mundlak校正的广义估计方程（Generalised Estimating Equations, GEE）模型，有效区分个体说话者内部与跨说话者之间的差异效应。研究不仅复现了年龄与语速与FP率呈负相关的结论，还揭示性别效应具有语言特异性且方向与多数既有文献相反；此外，通过创新性分析情感倾向、政治立场及权力地位发现，情感积极程度与FP率呈稳定正相关，而政治立场与权力地位则在不同议会情境下产生调节作用，其中反对派发言人的FP率普遍低于执政联盟成员。
链接: https://arxiv.org/abs/2607.05964
作者: Ivan Porupski,Branimir Dropuljić,Nikola Ljubešić
机构: Jožef Stefan Institute (斯洛文尼亚乔日夫·斯特凡研究所); TransUnion (特兰斯联合公司); University of Zagreb (萨格勒布大学); University of Ljubljana (卢布尔雅那大学); Institute of Contemporary History (当代史研究所)
类目: Computation and Language (cs.CL); Applications (stat.AP)
备注: 6 pages, 1 figure. Accepted at InterSpeech 2026. Code published: this https URL
Abstract:Filled pauses (FPs) are a universal feature of spontaneous speech, yet most studies rely on small, single-language corpora, limiting the generalisability of their findings. We analyse ~4,000 hours of parliamentary speech across four related Slavic languages (Croatian, Czech, Polish, Serbian). FP occurrence is obtained via transformer-based automatic detection, while FP rate is modelled using Generalised Estimating Equations (GEE) with Mundlak correction to distinguish within- from between- speaker effects. We replicate a negative association of age and speech rate with FP rate, but find that gender effects are language-specific and directionally opposite to most prior literature. Novel analyses of sentiment, political orientation, and power status reveal a consistent positive association between sentiment and FP rate, alongside parliament-specific modulation by orientation and power status, with opposition speakers tending toward lower FP rates than governing coalition speakers.
[NLP-29] Integrating knowledge graphs and multilingual scholarly corpora for domain-adaptive LLM s in SSH LREC2026
【速读】： 该论文旨在解决生成式 AI（Generative AI）在社会科学与人文学科（SSH）研究工作流中应用时所面临的重大方法论、认识论及监管挑战，尤其关注学科多样性、多语言文献获取以及研究成果评估的复杂性。其解决方案的关键在于将基础模型适配于 SSH 研究实践，通过欧洲项目 LLMs4EU 与 ALT-EDIC 研究基础设施的协同框架，实现对问答、对比文档分析和文献综述等任务的支持；同时，采用结合独立定量基准测试（包括检索、摘要生成、可追溯性及幻觉检测）与数字人文专家小组的定性评估相结合的评价体系，并将模型适配过程嵌入结构化的法律与伦理合规框架之中，从而确保生成式 AI 在支持学术研究的同时，兼顾领域敏感性、法规遵从性、结果可靠性与认识论责任。
链接: https://arxiv.org/abs/2607.05956
作者: Adam Faci,Alessio Miaschi,Anne Combe,Pascal Cuxac,Francesca Frontini,Nicolas Larrousse,Stéphane Pouyllau
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 8 pages, 4 tables, workshop LLMs4SSH of LREC 2026 conference
Abstract:The integration of Large Language Models (LLMs) into scientific research workflows, particularly for bibliographic discovery and literature synthesis, raises significant methodological, epistemic and regulatory challenges for the Social Sciences and Humanities (SSH), especially with regard to disciplinary diversity, multilingual access to sources and the evaluation of results. This paper presents an on-going use case developed within the European project LLMs4EU and the ALT-EDIC infrastructure, aimed at adapting foundation models to SSH research practices and supporting tasks such as question answering, comparative document analysis and literature review. The evaluation framework follows the LLMs4EU protocol and encompasses both independent quantitative benchmarking (retrieval, summarisation, traceability and hallucination detection) and a qualitative assessment involving a panel of Digital Humanities experts. By embedding model adaptation within research infrastructures and a structured legal and ethical compliance framework, the use case explores how domain-sensitive and regulation-aware generative AI can support SSH scholarship while preserving reliability and epistemic responsibility.
[NLP-30] Is Domain Adaptation Always Helpful? A Frozen-Backbone Study of Cross-Domain Sentiment Transfer
【速读】： 该论文旨在解决在使用冻结的预训练语言模型（PLM）主干网络进行情感分析时，显式领域适应（domain adaptation）的实际效益问题，尤其关注不同规模和领域特异性主干模型在目标领域知识覆盖程度不同时的表现差异。其核心解决方案在于通过控制变量的实验设计，系统评估多种领域自适应方法（包括基于对抗学习的域对抗神经网络（DANN）、最大均值差异（MMD）以及监督对比学习（SCL））在轻量级MLP适配器上的效果，并在消费者评论数据上训练后，迁移至电影评论（SST-2）与金融新闻受限子集（Financial PhraseBank）进行评估。研究发现，当主干模型已具备目标领域知识（如FinBERT）时，显式领域适应可能因破坏已有领域特定结构而降低性能；而对通用小规模模型（如0.6B Qwen3-Embedding），显式适应可显著提升性能。因此，该研究的关键结论是：显式领域适应的有效性高度依赖于冻结主干模型是否已包含目标领域的先验知识，即“领域覆盖”状态是决定适应策略成败的核心因素。
链接: https://arxiv.org/abs/2607.05937
作者: Phat Tran,Artin Lahni,Pranav Kulkarni,Yaolun Zhang
机构: Oregon State University (俄勒冈州立大学)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Sentiment analysis with frozen pre-trained language model (PLM) backbones has become a common paradigm, yet the practical benefit of explicit domain adaptation remains unclear, particularly when backbones encode varying degrees of target-domain knowledge. We present a preliminary case study evaluating a controlled family of frozen embedding backbones (Qwen3-Embedding 0.6B, 4B, 8B), alongside RoBERTa-base and FinBERT. We train a lightweight MLP adapter on consumer reviews using Domain-Adversarial Neural Networks (DANN), Maximum Mean Discrepancy (MMD), and Supervised Contrastive Learning (SCL), and evaluate transfer to movie reviews (SST-2) and a heavily restricted subset of financial news (Financial PhraseBank). Within this constrained sample, we observe two distinct transfer patterns. On SST-2, domain adaptation provides negligible gain regardless of scale. On the financial subset, explicit domain adaptation appears to recover substantial performance for small general-purpose backbones. Notably, we find that adversarial alignment (DANN) is associated with degraded performance for domain-specialized backbones like FinBERT, consistent with erosion of pre-existing domain-specific structure, whereas supervised contrastive loss appears to preserve it. These preliminary findings suggest that the efficacy of explicit domain adaptation is highly contingent on whether the frozen backbone already possesses target-domain coverage.
[NLP-31] PolicyShiftGuard: Benchmarking and Improving Policy-Adaptive Image Guardrails
【速读】： 该论文旨在解决生成式图像内容安全审核中政策依赖性适应能力不足的问题，即现有图像防护机制（image guardrails）通常基于固定安全策略进行训练与评估，将安全性视为图像的固有属性，无法应对实际部署中因政策动态变化而导致的合规判断差异。其核心挑战在于：同一图像在不同政策下可能被允许、限制或禁止，模型需具备根据当前政策动态调整判断的能力，而非依赖于图像本身的静态安全先验。解决方案的关键在于提出一种政策自适应的图像防护框架——PolicyShiftGuard，其核心技术包括：1）采用两阶段训练范式，结合随机化政策指令微调（RP-SFT）与边界对政策适配（BP-Adapt），通过成对比较损失强化对“通过”与“阻断”政策边界的区分能力；2）引入匹配的“通过/阻断”边界对（matched pass/block boundary pairs），确保模型学习到政策敏感的判别逻辑而非图像级安全偏见。实验表明，该方法在PolicyShiftBench上实现了76.9的平均F1和72.1的平均政策敏感度分数（PSS），显著优于现有视觉语言模型（VLMs）与专用防护模型，并展现出良好的泛化能力与推理效率。
链接: https://arxiv.org/abs/2607.05910
作者: Mingyang Song,Luxin Xu,Haoyu Sun,Minzhou Pan,Yu Cheng,Bo Li
机构: Fudan University (复旦大学); Tongji University (同济大学); Virtue AI (美德人工智能); The Chinese University of Hong Kong (香港中文大学); University of Chicago (芝加哥大学); University of Illinois, Urbana-Champaign (伊利诺伊大学香槟分校)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Image guardrails are typically trained and evaluated under a fixed safety policy, implicitly treating safety as an intrinsic property of an image. Real deployments are different: the same image may be allowed in one product, restricted in another, and newly disallowed when a policy boundary changes. We study policy-adaptive image guardrailing, where a model must decide whether an image violates the currently supplied policy and generalize to held-out policy definitions. We introduce PolicyShiftBench, a comprehensive benchmark with 2,000 policy-discriminative instances over 265 images, where each image is paired with 7.55 policy-conditioned prompts on average to test whether models adapt to the active policy rather than relying on image-level safety priors. We then propose PolicyShiftGuard, a compact policy-conditioned guardrail trained with a two-stage training recipe that combines Randomized Policy SFT (RP-SFT) with Boundary-Pair Policy Adaptation (BP-Adapt). BP-Adapt trains matched prompts for the same image and risk category using standard label supervision and a pairwise comparison loss that separates blocking policies from passing policies. Experiments show that existing VLMs and specialized guardrails remain brittle under policy shifts, while PolicyShiftGuard substantially improves policy-sensitive performance. The 7B model achieves SOTA performance of 76.9 Avg. F1 and 72.1 Avg. PSS on PolicyShiftBench, transfers well to UnSafeBench and SafeEditBench, and improves the latency-performance trade-off with a concise output format. Ablations confirm that matched pass/block boundary pairs are essential for stable policy adaptation.
[NLP-32] K-ABENA: K-Adaptive Backpropagation with Error-based N-exclusion Algorithm : (Compensated Loss-Based Sample Exclusion with Unbiased Gradient Estimation)
【速读】： 该论文旨在解决深度学习训练过程中高计算成本的问题，特别是针对每轮迭代中梯度计算开销过大的问题。其核心挑战在于如何在不显著牺牲模型性能的前提下，减少反向传播（backpropagation）中对低损失（“minor”）样本的计算负担。解决方案的关键是提出K-ABENA（K-Adaptive Backpropagation with Error-based N-exclusion Algorithm），一种基于误差选择性排除的梯度计算框架：通过有选择地跳过部分低损失样本的反向传播，实现计算资源的有效节约。该方法的创新之处在于其三版（v3）设计结合了防御性混合采样策略与霍维茨-汤普森逆概率重加权机制，从而构建出一个设计无偏的霍维茨-汤普森梯度估计器，并进一步提供了一个自归一化版本，其偏差量级为O(1/m)，并具有明确常数项。理论分析表明，在该估计器下，随机梯度下降（SGD）可达到O(1/√T)的非凸收敛率，且附加项量化了残余偏差的影响。此外，论文证明了未补偿的基于损失的选择策略（如OHEM、SBP及早期K-ABENA变体）在最小值点处无法收敛至任何选择偏差非零的极小点，从而导致严重性能退化；而经过补偿的版本在真实数据集（乳腺癌、数字识别、红酒、糖尿病）上表现出与全批量梯度下降统计无差异的结果（配对置换检验，p = 0.5），同时节省28%–54%的每轮梯度计算量。所有结论均经严格证明或实证验证，实验均在CPU规模（NumPy/scikit-learn）下完成，且范围明确。
链接: https://arxiv.org/abs/2607.05903
作者: Jean-Francois Bonbhel
机构: NeuroSoft IA(神经软科技); YekoElite University (叶科精英大学); UN AI Governance Expert Network (联合国人工智能治理专家网络)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 11 pages main text + appendices, 13 pages total. Code: this https URL
Abstract:We present K-ABENA (K-Adaptive Backpropagation with Error-based N-exclusion Algorithm), a selective gradient computation framework that reduces per-iteration training cost by excluding a fraction of low-loss (“minor”) observations from the backward pass. Its canonical form (v3) combines a defensive-mixture sampling design over the minor set with Horvitz-Thompson inverse-probability reweighting, yielding a design-unbiased Horvitz-Thompson gradient estimator (Lemma 2) and whose self-normalized practical variant carries a bias of order O(1/m) with an explicit constant (Lemma 3). We prove an O(1/sqrt(T)) non-convex convergence guarantee for SGD under the estimator, with an additive term that quantifies the residual bias (Theorem 1). We further prove that uncompensated loss-based selection - a family that includes OHEM, SBP, and the two earlier K-ABENA variants - admits no stationary point at any minimizer where its selection bias is bounded away from zero (Proposition 2), and we quantify this failure empirically: at 0.17% class imbalance, uncompensated variants reach test AUC 0.53-0.62 versus 0.9998 for full-batch SGD, while the compensated estimator attains 0.9991 at identical 28.4% compute savings. On real datasets (Breast Cancer, Digits, Wine, Diabetes) the compensated estimator is statistically indistinguishable from full-batch SGD (paired permutation tests, p = 0.5; Section 7) while saving 28-54% of per-epoch gradient computation. A biased “regularized mode” (the earlier half-domain variant) is retained as an option with a proven exact bias decomposition (Lemma 5) and quantified contraindications: it collapses to 0.386 accuracy under 40% label noise (baseline: 0.832) and to 0.53 AUC under extreme imbalance. Every advantage and every limitation reported in this paper is either proved or measured; all experiments are CPU-scale (NumPy/scikit-learn) and their scope is stated explicitly.
[NLP-33] Mitigating Factual Hallucination in Large Reasoning Models via Mixed-Mode Advantage Regularization
【速读】： 该论文旨在解决生成式推理模型（Generative Reasoning Models, GRMs）在事实导向型问答（Factuality-Oriented Question Answering, QA）任务中因显式思维链（explicit thinking）引入的“思维诱导幻觉”（thinking-induced hallucination）问题。具体而言，尽管显式思维有助于恢复缺失知识并优化答案，但在某些实例中，其反而会推翻原本正确的直接回答，导致事实性偏差。为解释此现象，作者将显式思维建模为相对于模型直接作答倾向的“思维残差”（thinking residual），该残差可能促进知识恢复，也可能引入无依据的关联。针对这一问题，论文提出MARGO（Mixed-Mode Advantage Regularization for Grounded Optimization）框架，一种基于强化学习的正则化方法，通过利用同一模型的非思维滚动生成（non-thinking rollouts）作为优势函数估计的参考基准，构建包含思维与非思维轨迹的混合模式滚动生成组，从而评估显式思维是否在直接回答基础上带来了额外的事实性价值。MARGO能够有效抑制易产生幻觉的思维行为，同时保留有益的推理能力。实验表明，MARGO在多个事实导向型QA基准上显著提升了事实可靠性，且在数学推理任务中保持了良好的通用推理性能。
链接: https://arxiv.org/abs/2607.05861
作者: Kaishen Wang,Tong Zheng,Xuehao Cui,Ruibo Chen,Tianyi Xiong,Heng Huang
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 19 pages, 3 figures, 8 tables
Abstract:Large reasoning models (LRMs) improve language model capabilities by generating explicit thinking traces before final answers. In factuality-oriented question answering (QA), such thinking often improves overall performance by helping the model recover relevant knowledge and refine its answers. However, we find that this benefit is not uniform at the instance level: explicit thinking can also overturn correct non-thinking answers and lead to factual drift. We refer to this failure mode as \emphthinking-induced hallucination. To explain this phenomenon, we formulate explicit thinking in factuality QA as a thinking residual over the model’s direct-answer tendency, which can either recover missing knowledge or introduce unsupported associations. Based on this formulation, we propose MARGO, \underline\textitMixed-Mode \underline\textitAdvantage \underline\textitRegularization for \underline\textitGrounded \underline\textitOptimization, a reinforcement learning framework that uses non-thinking rollouts as same-model references in advantage estimation. By constructing mixed-mode rollout groups with both thinking and non-thinking trajectories, MARGO evaluates whether explicit thinking adds factual value beyond direct answering, thereby suppressing hallucination-prone thinking while preserving beneficial thinking behaviors. Experiments across multiple factuality-oriented QA benchmarks demonstrate that MARGO improves factual reliability over strong baselines, while evaluations on mathematical benchmarks show that it preserves general reasoning ability.
[NLP-34] CoPiT: Cognitive Pivot Translation for Digraphic Low-Resource Mongolian in the Traditional Script
【速读】： 该论文旨在解决低资源语言（特别是蒙古语）在机器翻译中面临的严重数据稀缺与书写系统不均衡问题。蒙古语作为双书写系统语言，其西里尔字母（Cyrillic script）文本资源相对丰富，而传统蒙古文（Traditional script）则存在数据极度匮乏且拼写歧义显著的问题，导致直接翻译性能严重下降。针对此问题，论文提出CoPiT（Cognitively motivated Pivot-based Translation）框架，其核心解决方案是基于认知启发的、以西里尔字母为中间枢纽的翻译流水线：通过在翻译前显式解析传统蒙古文中的书写歧义，利用西里尔字母作为资源更丰富的中间表示进行语义传递，从而提升翻译的稳定性和准确性。该方法在多个主干模型和目标语言上均显著优于直接翻译，实现显著的绝对BLEU提升及1.5–1.6倍的COMET增益，使开源模型在同等评估条件下达到甚至超越GPT-4.1的性能。此外，CoPiT还能从仅含传统蒙古文的单语文本中构建合成平行语料，有效缓解真实低资源场景下的数据短缺问题。研究团队发布了包含蒙古语双书写系统（西里尔与传统）及英语、韩语、俄语的多语言平行数据集，所有代码与数据均已公开。
链接: https://arxiv.org/abs/2607.05849
作者: Burte Bayarsaikhan,Serynn Kim,Buru Chang
机构: Korea University (韩国大学); Hankuk University of Foreign Studies (韩外大学)
类目: Computation and Language (cs.CL)
备注: Preprint
Abstract:Low-resource languages remain challenging for machine translation, and Mongolian is a representative case. As a digraphic language, Mongolian is written in both Cyrillic and Traditional scripts, which exhibit a severe imbalance in data availability. While the Cyrillic script is relatively well-resourced, the Traditional script remains extremely data-scarce and orthographically ambiguous, leading to substantial performance degradation in direct translation. We propose CoPiT, a cognitively motivated pivot-based translation pipeline that exploits this internal resource hierarchy by routing translation through the Cyrillic script. The pipeline explicitly resolves script-induced ambiguity in the Traditional script before translation, enabling more stable and accurate meaning transfer. Across multiple backbone models and target languages, CoPiT consistently outperforms direct translation, achieving substantial absolute BLEU improvements together with consistent 1.5-1.6x COMET gains. These gains allow strong open-source models to match or outperform GPT-4.1 under comparable evaluation settings. Beyond inference-time improvements, CoPiT enables the construction of synthetic parallel data directly from Traditional-script text, mitigating data scarcity in realistic low-resource scenarios. We release a new multi-script parallel dataset covering Mongolian in both scripts alongside English, Korean, and Russian. All datasets and code are publicly available at this https URL.
[NLP-35] urnOPD: Making On-Policy Distillation Turn-Aware for Efficient Long-Horizon Agent Training
【速读】： 该论文旨在解决在长时序智能体任务中，传统在线策略蒸馏（On-Policy Distillation, OPD）存在的效率低下问题。具体而言，原始OPD方法在长序列生成过程中存在两大瓶颈：其一，全轨迹回放会浪费大量实际运行时间在后期低信息量的“尾部动作”上，这些动作提供的KL散度监督信号弱且噪声大；其二，基于轨迹层面的KL损失倾向于过度优化早期token，导致深层决策节点因初始行为对齐后缺乏有效监督而训练不足。为克服上述问题，本文提出一种新型的逐回合级预算控制策略——TurnOPD，其核心创新在于两个协同机制：一是基于探测器的自适应回放深度预算控制，通过分析每个回合的统计特征动态决定最优回放长度；二是渐进式回合归一化损失预算分配，逐步将损失权重从词元级转向回合均衡的监督模式，从而实现更公平、高效的梯度传播。实验结果表明，在ALFWorld、WebShop及多跳搜索等任务上，使用任务专用教师模型的TurnOPD在相同实际训练时间内显著提升了验证准确率，并推动了准确率-训练时间权衡曲线的前沿进展。
链接: https://arxiv.org/abs/2607.05804
作者: Yuhang Zhou,Kai Zheng,Haoling Li,Dengyun Peng,Can Xu,Jingjing Chen
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:On-policy distillation (OPD) trains a student policy by matching a stronger teacher on the student’s own trajectories, offering a promising framework for language agent training. However, its application to long-horizon agentic tasks remains insufficiently explored. We identify two key inefficiencies in vanilla agent OPD: (1) full-horizon rollouts often waste wall-clock resources on tail turns that provide weak and noisy KL supervision, and (2) trajectory-level KL objectives concentrate most of the loss on shallow tokens, leaving deeper decision turns under-trained once initial behaviors are aligned. To address these challenges, we propose TurnOPD, a turn-level budgeting strategy for efficient on-policy distillation of long-horizon agents. TurnOPD consists of two budget controllers: adaptive rollout-depth budgeting, which uses probe-based turn statistics to determine rollout length, and progressive turn-normalized loss budgeting, which gradually shifts KL weighting from token-level to turn-balanced supervision. Experiments on ALFWorld, WebShop, and Multi-Hop Search with task-specialized teacher models show that TurnOPD achieves superior validation accuracy under equal wall-clock training budgets and advances the accuracy–time frontier beyond vanilla OPD.
[NLP-36] When Should LLM s Search? Counterfactual Supervision for Search Routing ICML2026
【速读】： 该论文旨在解决生成式 AI 在面对知识局限时，因不当使用外部检索（search）而导致的效率与准确性下降问题。具体而言，模型可能在无需检索的情况下触发搜索，或在证据噪声较大时盲目依赖检索结果，而忽视更优的策略如自我修正、澄清提问或直接放弃回答。为应对这一挑战，论文将问题建模为实例级的“检索路由”（search-routing）问题：即判断在特定任务中是否需要调用检索以提升性能，相较于不进行检索的执行结果。其关键解决方案是构建一个基于任务成功性的“理想决策代理”（oracle），通过对比同一问题在无检索与强制检索两种情形下的输出，定义出最优决策路径（包括“不检索”“检索”和“未解决”三类）。该 oracle 既作为评估标准，也作为监督信号，用于对检索路由策略进行有监督微调与偏好优化。实验表明，该方法显著提升了路由性能，在 Gemma E2B 和 Qwen3.5-4B 模型上，路由宏平均 F1 值分别从 0.7082/0.7053 提升至 0.8235/0.8365。进一步分析揭示，学习到的路由策略有效缓解了模型特异性错误：Gemma 主要增强了“避免无谓检索”的能力，而 Qwen 则更有效地减少了“遗漏应检索场景”。剩余未解决案例暴露了由模型容量、检索预算、证据利用能力及策略行为等多方面因素构成的异质性瓶颈。
链接: https://arxiv.org/abs/2607.05752
作者: Minho Kim
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 20 pages, 10 figures. Accepted at the FAGEN Workshop at ICML 2026
Abstract:Search-augmented language models can use external evidence to compensate for limitations in parametric knowledge, but search is not uniformly beneficial: models may call search for questions they can already answer, or rely on noisy evidence when correction, clarification, or abstention would be more appropriate. We formulate this as an instance-level search-routing problem: deciding whether search is needed to improve task success relative to a no-search execution. To derive supervision, we compare no-search and forced-search outcomes for the same question and construct an oracle over NO SEARCH, SEARCH, and UNSOLVED based on task-specific success. Using this oracle as both an evaluation criterion and a learning signal, we train search-routing policies with supervised fine-tuning and preference optimization, improving routing macro-F1 on oracle-eligible examples from 0.7082 to 0.8235 for Gemma E2B and from 0.7053 to 0.8365 for Qwen3.5-4B. Further analysis shows that the learned policies reduce model-specific routing failures: Gemma primarily learns no-search restraint, while Qwen further reduces missed search; residual UNSOLVED cases reveal heterogeneous bottlenecks involving model capacity, retrieval budget, evidence use, and policy behavior.
[NLP-37] Nemotron-Labs-Diffusion: A Tri-Mode Language Model Unifying Autoregressive Diffusion and Self-Speculation Decoding
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）在推理阶段面临的高延迟与低吞吐量问题，尤其在不同并发水平和部署场景下难以兼顾生成效率与生成质量的挑战。其核心解决方案是提出Nemotron-Labs-Diffusion——一种统一自回归（Autoregressive, AR）、扩散（Diffusion）与自推测（Self-Speculation）三种解码模式于单一架构中的三模态语言模型。该模型通过联合训练AR与扩散目标，实现了两种机制的互补：扩散模型提供前瞻性的文本规划能力，而自回归模型则引入左到右的语言先验以增强生成连贯性。在自推测模式中，扩散模型负责生成草稿，自回归模型进行验证，显著优于传统多标记预测（Multi-Token Prediction, MTP）方法，在接受率与真实设备效率上均表现更优。此外，速度极快分析表明，在理想采样器下，扩散模式相较自推测模式可实现高达76.5%的每前向传播输出令牌数提升。在3B、8B和14B参数规模下，Nemotron-Labs-Diffusion系列模型（包括基础版、指令微调版及视觉-语言版本）在准确率与速度方面持续超越当前最先进的开源自回归与扩散型语言模型。例如，Nemotron-Labs-Diffusion-8B在保持相近准确率的前提下，相比Qwen3-8B每前向传播可生成6倍的令牌数，结合SGLang在GB200 GPU上的SPEED-Bench测试中实现4倍更高的吞吐量，充分验证了该框架在实际部署中的高效性与可扩展性。
链接: https://arxiv.org/abs/2607.05722
作者: Yonggan Fu,Lexington Whalen,Abhinav Garg,Chengyue Wu,Maksim Khadkevich,Nicolai Oswald,Enze Xie,Daniel Egert,Sharath Turuvekere Sreenivas,Shizhe Diao,Chenhan Yu,Ye Yu,Weijia Chen,Sajad Norouzi,Jingyu Liu,Shiyi Lan,Ligeng Zhu,Jin Wang,Jindong Jiang,Morteza Mardani,Mehran Maghoumi,Song Han,Ante Jukić,Nima Tajbakhsh,Jan Kautz,Pavlo Molchanov
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:We introduce Nemotron-Labs-Diffusion, a tri-mode language model (LM) that unifies AR, diffusion, and self-speculation decoding within a single architecture. Trained with a joint AR-diffusion objective, Nemotron-Labs-Diffusion can switch modes to sustain high throughput across deployment settings and concurrency levels. Our study shows that (1) AR and diffusion objectives are complementary: diffusion improves lookahead planning, while AR provides left-to-right linguistic priors. (2) In self-speculation mode, diffusion drafts while AR verifies, outperforming multi-token prediction (MTP) methods in both acceptance rate and real-device efficiency. (3) A speed-of-light analysis further demonstrates diffusion’s long-term potential, with up to 76.5% more tokens per forward pass than self-speculation under an optimal sampler. Scaling to 3B, 8B, and 14B parameters, our Nemotron-Labs-Diffusion family, including base, instruct, and vision-language models, consistently outperforms state-of-the-art open-source AR and diffusion LMs in both accuracy and speed. For example, Nemotron-Labs-Diffusion-8B decodes 6x more tokens per forward than Qwen3-8B with comparable accuracy, translating to 4x higher throughput on SPEED-Bench with SGLang on a GB200 GPU.
[NLP-38] SpanUQ: Span-Level Uncertainty Quantification for Large Language Model Generation
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在不确定性估计（Uncertainty Estimation）中粒度不匹配的问题：现有方法要么在词元级别（token-level）进行评估，导致语义不连贯；要么在序列级别（sequence-level）进行评估，无法精确定位错误位置。为此，论文提出了**段落级不确定性估计（Span-Level Uncertainty Estimation, SLUE）**这一新任务，其目标是识别具有语义一致性的文本片段（semantically coherent text spans），每个片段代表一个可评估的语义单元。解决方案的关键在于提出SPANNUQ——一种轻量级探针（probe），通过将昂贵的多样本推理（multi-sample inference）中蕴含的不确定性知识压缩至一次前向传播中，利用基于DETR架构的片段解码器（span decoder），以混合贝塔分布（Mixture of Beta distribution）建模不确定性，并结合贝塔负对数似然回归（Beta NLL regression）与对比排序（contrastive ranking）的联合目标进行训练。实验表明，SPANUQ在五个主流LLM骨干网络上均显著优于最强基线和所有基于采样的方法，在保持10–20倍加速的同时实现了最优的段落级不确定性质量；其段落检测器在F1指标上达到0.910，较最佳启发式方法提升39.4%，实现了序列级方法无法提供的精确错误定位能力。此外，该框架在跨两个模型家族的五种不同模型间展现出良好泛化性。
链接: https://arxiv.org/abs/2607.05721
作者: Yimeng Zhang,Yingying Zhuang,Ziyi Wang,Yuxuan Lu,Pei Chen,Aman Gupta,Zhe Su,Ming Tan,Zhilin Zhang,Qun Liu,Manikandarajan Ramanathan,Rajashekar Maragoud,Edward Vul,Jing Huang,Dakuo Wang
机构: Amazon(亚马逊); Northeastern University(东北大学)
类目: Computation and Language (cs.CL)
备注: The project page is available at this https URL
Abstract:Uncertainty estimation is essential not only for the trustworthy deployment of large language models (LLMs) but also as a foundation for self-refinement in LLM generation. However, existing approaches operate at suboptimal granularities: token-level scores lack semantic coherence, while sequence-level scores fail to localize errors. We formalize Span-Level Uncertainty Estimation (SLUE), a new task that targets the natural granularity for uncertainty: semantically coherent text spans, each conveying a single assessable unit of meaning. To address this task, we introduce SPANUQ, a lightweight probe that distills the uncertainty knowledge from expensive multi-sample inference into a single forward pass over LLM hidden states. SPANUQ employs a DETR-style span decoder to simultaneously detect spans and estimate their uncertainty via a Mixture of Beta distribution, trained with a principled combination of Beta NLL regression and contrastive ranking objectives. We construct SPANUQ-BENCH, the first span-level uncertainty benchmark comprising 20K prompts, 293K annotated spans, and continuous soft labels derived from multi-sample claim verification. Experiments on five LLM backbones show that SPANUQ consistently achieves the best span-level uncertainty quality, outperforming the strongest probe baseline and all sampling-based methods while being 10-20x faster. Its DETR-based span detector attains 0.910 F1, surpassing the best heuristic by 39.4%, enabling precise error localization that sequence-level methods cannot provide. The framework generalizes across five LLMs spanning two model families.
[NLP-39] Where to cut how deep: BPE and Unigram-LM on chemistry SMILES
【速读】： 该论文旨在解决化学语言模型中普遍采用的字节对编码（Byte-Pair Encoding, BPE）在分子结构表示上的合理性问题，尤其关注其与另一种主流子词分词方法——基于语言模型的单个词元（Unigram-LM）在化学语料中的表现差异。尽管BPE在自然语言处理中广泛应用，但其在化学领域是否仍为最优选择缺乏系统评估。研究通过在固定165个词元的基础词汇表上，对BPE与Unigram-LM进行受控对比，覆盖三种不同类型的化学语料（多样性、类药物、天然产物）及两种预分词边界策略，在小词汇量条件下分析二者生成的子词词汇的差异性。结果表明，两种算法并未收敛，其生成的子词词汇几乎不重叠：跨算法的Jaccard相似度最高仅0.161，加权后高频更新部分的重叠率更低至0.05。此外，Unigram-LM在未见分子上的分词数量比BPE高出29%-41%，且两者切分位置虽有共性，但深度差异显著，说明BPE的分词是Unigram-LM分词的严格粗化版本，适用于80%-99%的分子。该现象在不同语料、边界策略和词汇规模下均成立，甚至在八倍规模下依然持续。因此，子词分词算法并非可忽略的默认设置，而是一个关键的建模决策，直接影响分子表示的质量与模型性能。
链接: https://arxiv.org/abs/2607.05691
作者: Hunter Heidenreich
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG); Biomolecules (q-bio.BM)
备注:
Abstract:Every chemical language model reading SMILES begins with a tokenizer, yet the field has inherited byte-pair encoding (BPE) from natural language with little scrutiny. In natural language, BPE’s principal alternative, Unigram-LM, is known to build structurally different vocabularies. Whether that contrast survives in chemistry was open. We report a controlled comparison of BPE and Unigram-LM over a fixed 165-token chemistry base, at the small vocabulary sizes where token embeddings are learnable, across three corpus typologies (diverse, drug-like, natural-products) and both pre-tokenization boundary policies. The two do not converge. In all 22 matched conditions they build near-disjoint subword vocabularies: cross-algorithm Jaccard overlap on the learned pieces never exceeds 0.161, and at most 0.05 once weighted toward the high-frequency pieces a model updates most. Unigram-LM also segments held-out molecules into 29-41% more tokens; the arms largely agree on where to cut but not how deeply, so BPE’s segmentation is a strict coarsening of Unigram-LM’s on 80-99% of molecules. The separation holds across corpus, boundary, and vocabulary size, persisting even at eight times that scale. The subword algorithm is therefore a modeling decision, not a free default. The study trains no language models.
[NLP-40] Memory in the Loop: In-Process Retrieval as ExtendedWorking Memory for Language Agents
【速读】： 该论文旨在解决语言智能体（language agent）在执行“观察-推理-行动”循环时，外部记忆系统因访问延迟过高而导致效率严重下降的问题。传统架构中，记忆仅在每轮循环中被查询一次，而若将记忆操作纳入每一步的循环内（即实现“每步读写内存”），虽能提升推理连贯性与记忆可用性，但受限于网络化存储带来的高延迟（数十至数百毫秒），会导致端到端延迟最高膨胀83倍，难以实际应用。其核心解决方案的关键在于：将记忆存储从远程网络服务迁移至进程内（in-process），使存储响应时间缩短至约100微秒，相较网络模式降低三个数量级。这一速度跃升使得每步内存访问的“税负”几乎消失，从而满足扩展心智论（extended-mind thesis）中的“可直接、持续访问”条件，使内存从“外部工具”转变为“扩展工作记忆”。实验表明，在固定每轮记忆延迟预算下，随着存储响应速度提升，冗余动作显著减少（从110ms云延迟下的7.2/12降至进程内0.0/12），且在四个类GPT-5模型上，基于窗口的召回率从0/5提升至3.6–4.8/5；同时，存储从未丢失任何数据（244次写入全部成功），所有缺失均源于智能体的读取策略而非存储可靠性。进一步分析揭示，主要性能瓶颈来自嵌入计算（embedding），通过引入本地小型嵌入器，可将完整操作延迟压缩至约40微秒，验证了该方案在真实系统中的可行性与高效性。
链接: https://arxiv.org/abs/2607.05690
作者: Yusuf Khan,Carlo Lipizzi
机构: Stevens University (斯蒂文斯大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Language agents run a loop - observe, reason, act - but the memory they reason over sits outside it: a store queried at most once per turn. We study the regime where memory moves inside the loop, read and written on every step. The obstacle has always been latency: networked stores answer in tens to hundreds of milliseconds, and in-loop retrieval can inflate end-to-end latency by up to 83x when retrieval is expensive. Prior work manages that cost rather than questioning it: serving-layer scheduling hides it, “memory-first” designs ration retrieval to once per turn. We argue latency is a property of where the store lives, not the in-loop pattern: an in-process store answers in ~100us, three orders of magnitude below the network regime, and at that speed the per-step tax collapses. By the extended-mind thesis’s parity principle, a store fast enough to be constantly and directly available becomes extended working memory, not a tool the agent merely consults. The premise is causal: holding a fixed per-turn memory-latency budget and varying only the store’s answer speed, redundant actions rise monotonically with latency - 0.0 of 12 at in-process speed, 7.2 of 12 at a 110ms cloud round trip (gpt-5-nano, gpt-5-mini; exact permutation p=0.0079). We demonstrate the regime end-to-end: across four GPT-5-class models under a bounded window, recall improves from 0/5 to 3.6-4.8/5 with in-loop memory, store ops at p50 80-165us - though an instructed restate-every-reply baseline also solves it perfectly, at a token cost that grows with the working set. The store never lost a fact in any run (244 of 244 writes kept); every miss traces to the agent’s read policy, not the store. Our measurements also relocate the bottleneck: the dominant per-step cost is embedding (~200-400ms over the network); pairing the in-process store with a small local embedder returns the complete operation to a measured ~40us.
[NLP-41] UCSC NLP at SemEval-2026 Task 10: Boundary-Aware Span Extraction and RoBERTa Classification for Conspiracy Detection SEMEVAL-2026
【速读】： 该论文旨在解决阴谋论文本中阴谋标记提取（Subtask 1）与文档级阴谋论检测（Subtask 2）两大任务。针对标记提取问题，其关键解决方案是将任务建模为在枚举候选片段上的多标签跨度分类，采用交并比（IoU = 0.95）进行正样本标注、硬负例采样，并结合边界感知的跨度表示与基于包含关系的非极大值抑制（NMS），以提升定位精度。对于文档级分类，采用带有标签平滑的序列分类器，并通过分层划分训练-验证集以增强泛化能力。实验分析表明，实体类角色（如参与者、受害者）的检测表现稳健，而抽象类角色（如行为、影响、证据）则对边界判定标准较为敏感。在官方测试集上，系统在子任务1中取得第7名（宏平均F1=0.2251），子任务2中位列第11名（加权F1=0.7694）。
链接: https://arxiv.org/abs/2607.05689
作者: Dom Marhoefer,Milos Suvakovic,Glenn Grant-Richards,Aidan Pinero,Ryan King
机构: UC Santa Cruz (加州大学圣克鲁斯分校)
类目: Computation and Language (cs.CL); Social and Information Networks (cs.SI)
备注: 6 pages, 2 tables. System description paper for SemEval-2026 Task 10 (PsyCoMark: Psycholinguistic Conspiracy Marker Extraction and Detection)
Abstract:We present our systems for SemEval-2026 Task 10 (PsyCoMark), addressing conspiracy marker extraction (Subtask 1) and document-level conspiracy detection (Subtask 2). For marker extraction, we formulate the task as multi-label span classification over enumerated candidate spans, using IoU = 0.95 positive labeling, hard-negative sampling, and containment-based non-maximum suppression (NMS) with boundary-aware span representations. Document classification is modeled independently using a sequence classifier with label smoothing and a stratified train-validation split. Analysis shows that entity-like roles (Actor, Victim) are detected robustly, while abstract roles (Action, Effect, Evidence) remain sensitive to boundary criteria. On the official test set, our systems rank 7th in Subtask 1 (0.2251 macro F1) and 11th in Subtask 2 (0.7694 weighted F1).
[NLP-42] RPAM: A Principled Metric for Evaluating Associations in Language Models with High Predictive Validity in Downstream Outputs
【速读】： 该论文旨在解决生成式语言模型（Generative Language Models, LMs）中存在的隐性偏见（如刻板印象）的准确评估与可泛化分析问题。现有方法多依赖下游指标，通过分析生成文本中的关联性来衡量偏见，但此类指标受模型生成内容差异影响大，需针对不同模型构建专用评估数据集，导致其泛化能力受限。相比之下，上游指标通过考察模型嵌入表示或续写概率等基础层面的表征，具备跨模型可比性。然而，此前尚无上游指标能有效揭示与现实世界关联性（包括人类认知及下游任务中测得的偏见）之间的强相关性。为填补这一空白，本文提出相对概率关联度量（Relative Probability Association Metric, RPAM），一种面向生成式语言模型的上游关联评估指标。在三种不同质量与用途的语言模型（Mistral-7B-Instruct、Mistral-7B 和 GPT-2）以及多个经典评估数据集（WEAT-WS、Bellezza、WS-353 与 SST2）上的实验表明，RPAM 在上游层面测量的关联性与人类观察到的显性/隐性关联、以及下游模型特异性任务中测得的偏见之间存在显著强相关性，且在多数情况下优于现有最优指标表现。其核心突破在于首次实现了上游指标与真实世界关联性之间的强一致性，为偏见分析提供了可迁移、可解释且具理论基础的评估框架。
链接: https://arxiv.org/abs/2607.05679
作者: Damian Hodel,Jevin West,Aylin Caliskan
机构: University of Washington (华盛顿大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 14 pages
Abstract:Language models (LMs) exhibit problematic biases, such as stereotypes. Effectively analyzing and mitigating such biases requires accurate and generalizable evaluation methods of the underlying associations. Some existing approaches focus on downstream metrics that analyze associations in generated text. Since generated text content can vary drastically across LMs, such metrics often require specialized evaluation datasets, which limits the generalization of such downstream metrics. In contrast, upstream metrics examine LMs at the fundamental level of embeddings or continuation probabilities, enabling principled association analyses across LMs. Yet, to date, no upstream metric for generative LMs has uncovered a strong relationship with real-world associations, including those measured in generated text. To address this gap, we introduce the Relative Probability Association Metric (RPAM), an association evaluation metric for generative LMs. For three LMs of different quality of language generation and purpose (Mistral-7B-Instruct, Mistral-7B, and GPT-2) and well-studied evaluation datasets (WEAT-WS, Bellezza, WS-353, and SST2), we find a strong relationship between upstream RPAM measurements and corresponding implicit and explicit associations observed in humans, as well as biases measured downstream with LM-specific tasks, outperforming prior record values where applicable.
[NLP-43] Do It Right! A Methodology for Successful NLP System Development
【速读】： 该论文旨在解决临床研究与决策中利用电子病历（Electronic Medical Records, EMR）数据时，自然语言处理（Natural Language Processing, NLP）项目实施过程中存在的系统性挑战，尤其是算法知识虽丰富但难以保证项目成功的问题。其核心问题在于：尽管已有大量关于文本处理算法与应用的教材和教程，但缺乏一个结构化、可重复的开发流程来指导从需求分析到部署维护的全过程。论文提出的解决方案关键在于将系统开发生命周期（Systems Development Life Cycle, SDLC）框架引入基于语言处理的数据提取项目中，通过分阶段的方法（如需求定义、设计、实现、测试与运维）确保NLP项目的可管理性、可验证性与可持续性，从而提升项目成功率并增强临床应用的可靠性。
链接: https://arxiv.org/abs/2607.05644
作者: Olga V. Patterson,Brett South,T. Elizabeth Workman,Scott L DuVall
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Pre-submission draft
Abstract:Natural language processing (NLP) is a common method for supplying data to clinical research and decision making by extracting information from electronic medical records. Numerous textbooks and tutorials describe specific algorithms and applications for text processing, yet algorithmic knowledge is only one ingredient of a successful NLP project. Drawing on the available literature, this paper presents a stepwise approach that applies the Systems Development Life Cycle (SDLC) to projects that rely on data extraction through language processing.
[NLP-44] Population-Level Profiling of DSM-5 Depressive Symptoms Among Self-Reported ADHD and ASD Users on Twitter: An Exploratory Study Using Advanced NLP and Statistical Analysis
【速读】： 该论文旨在解决注意力缺陷多动障碍（ADHD）与自闭症谱系障碍（ASD）共病抑郁症状在社交媒体用户中表达差异的群体层面特征未被充分探索的问题。其核心解决方案在于利用自然语言处理技术，通过零样本自然语言推理（zero-shot NLI）对海量推文进行抑郁相关内容预筛选，并采用经过ReDSM5数据集微调的MentalRoBERTa模型对九种DSM-5抑郁症状进行精准分类，结合用户个体均值中心化处理与L1正则化逻辑回归建模，在不同抑郁内容过滤阈值下评估两组人群在抑郁症状表达上的群体差异。研究发现，尽管分类性能仅达到中等水平（交叉验证ROC-AUC 0.645–0.653），但认知功能障碍、睡眠问题、食欲改变及疲劳更倾向表现为ADHD特征，而自杀意念与快感缺失则更倾向表现为ASD特征；同时，两组在症状共现结构上具有高度一致性，未发现显著的、稳健的疾病特异性症状组合。因此，该研究揭示了基于社交媒体文本的群体层面语言差异具有可重复性，但其临床意义尚待验证，且不能推断个体层面的表型差异。
链接: https://arxiv.org/abs/2607.05626
作者: Muhammad Rizwan,David Nabergoj,Jure Demšar
机构: Faculty of Computer and Information Science, University of Ljubljana, Slovenia(卢布尔雅那大学计算机与信息科学学院, 斯洛文尼亚)
类目: Computation and Language (cs.CL)
备注:
Abstract:Background: Depression frequently co-occurs with ADHD and autism spectrum disorder (ASD), but population-level differences in symptom expression between these groups remain underexplored. Objective: We examined whether social media users with ADHD and ASD differ in how they express DSM-5 depressive symptoms in their tweets, and whether differences persist across varying levels of depressive-content filtering. Methods: We analysed 1,282,437 tweets from 792 users (622 ADHD; 170 ASD) with self-reported diagnoses on Twitter. Tweets were pre-filtered for depressive relevance using zero-shot NLI, then classified into nine DSM-5 symptoms using MentalRoBERTa fine-tuned on ReDSM5. Profiles were mean-centered per user. We applied L1-penalised logistic regression with cross-validation to distinguish ADHD from ASD users, complemented by Pearson correlations for symptom co-occurrence, and tested robustness across five filtering thresholds using bootstrapping. Results: MentalRoBERTa achieved macro-F1 of 0.901 on a held-out set, outperforming the original ReDSM5 benchmark. ADHD vs ASD classification yielded stable but modest performance (cross-validated ROC-AUC 0.645-0.653). Cognitive issues, sleep issues, appetite change, and fatigue leaned toward ADHD, while suicidal ideation and anhedonia leaned toward ASD. A largely shared symptom co-occurrence structure emerged between groups; no pair met our criterion for a robust disorder-specific difference. Conclusions: Population-level differences in depression-related language between ADHD and ASD social media users were consistently observed across thresholds, reflecting reproducibility rather than clinical validity. Findings are exploratory and do not establish differing phenomenology at the individual level.
[NLP-45] NAVER LABS System Re-implementation for the IWSLT 2026 Instruction-Following Task
【速读】： 该论文旨在解决在受限条件下的短音频语音翻译任务中，如何有效融合多模态信息以提升指令遵循能力的问题。其核心挑战在于如何在有限的语音输入长度和严格的系统约束下，实现高质量的跨语言语音翻译与英语问答（SQA）性能。解决方案的关键在于保留原始三阶段框架——投影对齐、纯文本LoRA预训练以及多模态融合，并在此基础上构建了10万条合成指令跟随数据（涵盖十类以语音为中心的任务，每类1万条），用于增强第三阶段的微调效果。通过引入SeamlessM4T-v2-large作为语音编码器和Qwen3-4B-Instruct作为大语言模型（LLM）主干，实现了端到端的高效多模态理解与生成。实验结果表明，该方法在EN-ZH语音翻译任务上取得COMET 0.781，在MCIF基准上的英语SQA任务中达到BERTScore-F1 0.346，验证了其在复杂指令遵循场景中的有效性。
链接: https://arxiv.org/abs/2607.05623
作者: Anand Kamble,Aniket Tathe
机构: Florida State University (佛罗里达州立大学); University of Illinois Urbana-Champaign (伊利诺伊大学厄本那-香槟分校)
类目: Computation and Language (cs.CL)
备注:
Abstract:We re-implement the NAVER LABS IWSLT 2025 instruction-following pipeline for the IWSLT 2026 Shared Task (constrained condition, short audio track), adapting it to the mandated components: SeamlessM4T-v2-large as the speech encoder and Qwen3-4B-Instruct as the LLM backbone. The three-stage approach projector alignment, text-only LoRA pre-training, and multimodal merging is preserved from the original design. We additionally construct 100k synthetic instruction-following examples across ten speech-centric task types (10k per task) from the provided corpora, suitable for further Stage 3 fine-tuning. Our primary model achieves COMET 0.781 on EN-ZH speech translation and BERTScore-F1 0.346 on English SQA on the MCIF benchmark.
[NLP-46] BaFCo: A Document Understanding Benchmark for Complex Bangla Form Comprehension ECCV
【速读】： 该论文旨在解决多模态大语言模型（Multimodal Large Language Models, MLLMs）在低资源语言如孟加拉语（Bangla）的表单理解任务中因高质量标注数据匮乏而导致性能受限的问题。其核心挑战在于，现有模型在处理结构复杂、语义丰富的孟加拉政府表单时，难以准确识别和定位细粒度的表单实体。为此，论文提出了一项名为BaFCo的新基准数据集，专注于文档版面分析（Document Layout Analysis, DLA）与关键信息提取（Key Information Extraction, KIE），包含200个来自农业、教育、银行及土地管理等多个领域的多页复杂孟加拉政府表单。数据集采用细粒度标注体系，定义了26种形式实体类型，并辅以5类粗粒度实体类别，以全面捕捉表单的结构与上下文复杂性。关键解决方案在于构建一个具有高标注质量且覆盖真实世界场景的专用基准数据集，从而推动对孟加拉语表单理解能力的研究。实验结果表明，当前主流的MLLMs（如ChatGPT、Gemini、Claude、Qwen和Kimi系列）在零样本和思维链提示下均表现出对细粒度实体定位能力的显著不足，凸显了该任务的挑战性与数据集的必要性。
链接: https://arxiv.org/abs/2607.05614
作者: Abu Tyeb Azad,Ishita Sur Apan,Fahim Ahmed,Sumaiya Karim Katha,Ezharuddin Jubaer,Armun Alam,Pranjal Kumar Nandi,Amin Ahsan Ali,Aman Chadha,Md Mofijul Islam,AKM Mahbubur Rahman
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at the 19th European Conference on Computer Vision (ECCV), 2026
Abstract:Document comprehension is a challenging yet impactful task for Multimodal Large Language Models, especially as these systems see growing adoption in real-world, human-centric applications. However, this adoption is limited for low-resource languages such as Bangla due to the scarcity of high-quality annotated data. To address this gap, we introduce BaFCo, a benchmark dataset for Bangla form comprehension with a focus on Document Layout Analysis (DLA) and Key Information Extraction (KIE). BaFCo curates 200 multi-page complex Bangladeshi government forms, sourced from across diverse sectors including agriculture, education, banking, and land management. To accurately capture the structural and contextual complexity of these forms, we define a fine-grained annotation schema comprising 26 types of form entities, along with a separate coarse form entity set consisting of 5 types. We evaluate the latest MLLMs from the ChatGPT, Gemini, Claude, Qwen, and Kimi series using zero-shot and chain-of-thought prompts under both low and high reasoning setups. Our results reveal limitations in current MLLMs’ ability in comprehending Bangla forms, particularly in accurately localizing highly granular form entities. Our dataset and code is available at: this https URL
[NLP-47] Revisiting the Relation Between Language Model Perplexity and ASR Word Error Rate for Modern End-to-End Speech Recognition
【速读】： 该论文旨在重新审视现代端到端自动语音识别（ASR）系统中语言模型困惑度（PPL）与词错误率（WER）之间的关系。传统上，二者在对数-对数空间中呈现近似线性关系，但现代ASR系统因具备内部语言建模（ILM）能力、通常不依赖外部语言模型（LM），并可结合神经语言模型或大语言模型（LLM）进行解码，使得这一假设面临挑战。论文的关键解决方案在于：通过系统性分析外部语言模型是否仍能提升当前端到端ASR性能、考察PPL-WER关系的线性特征是否保持、编码器上下文长度的影响，以及LLM困惑度在标准神经语言模型趋势中的位置；更重要的是，揭示了在评估外部语言模型效果时，必须考虑解码器内部语言建模能力，并通过引入内部语言建模减法（ILM subtraction）方法，修正了原有的PPL-WER关系，表明外部语言模型的实际增益需基于对内部语言建模能力的校正来准确解读。
链接: https://arxiv.org/abs/2607.05612
作者: Mohammad Zeineldeen,Albert Zeyer,Haoran Zhang,Robin Schmitt,Ralf Schlüter,Hermann Ney
机构: RWTH Aachen University (亚琛工业大学); Google(谷歌)
类目: Computation and Language (cs.CL); Audio and Speech Processing (eess.AS)
备注: Submitted to SLT 2026
Abstract:Language model (LM) perplexity (PPL) has historically been used as a proxy for automatic speech recognition (ASR) word error rate (WER), with prior work reporting an approximately linear relation in log-log space. Modern end-to-end ASR systems challenge this assumption because they already contain internal language modeling capacity, are often evaluated without external language models, and can now be combined with neural LMs and large language models (LLMs) through different recognition strategies. This paper revisits the relation between PPL and WER for modern ASR systems. We study whether external LMs still improve current end-to-end ASR systems, whether the PPL-WER relation remains linear in log-log space, how encoder context length affects this relation, and how LLM perplexities fit into the trend observed for standard neural LMs. We further investigate internal language modeling (ILM) in attention-based encoder-decoder systems and show that ILM subtraction changes the observed PPL-WER relation, indicating that the decoder’s internal LM must be considered when interpreting the effect of external LM quality.
[NLP-48] ResonatorLM: Causal Resonant Field Mixing for Efficient Long-Context Language Modelin ICANN2026
【速读】： 该论文旨在解决当前基于Transformer架构的语言模型在处理长上下文时效率下降的问题。尽管Transformer通过自注意力机制实现了高效并行化训练，但在序列长度增加时，其计算复杂度随序列长度平方增长，导致训练和推理速度显著降低。为此，论文提出ResonatorLM，其核心创新在于用一种源自物理学的替代机制取代传统的注意力机制：将词元序列建模为一个单一的、受驱动的一维隐式场，并以阻尼谐振器的因果函数代替注意力中的点积运算。这一设计使得模型在长序列处理中具备更低的计算开销与更高的可扩展性。实验结果表明，在32K token的上下文长度下，ResonatorLM的解码速度相比标准优化版Transformer提升6.47倍，且在WikiText数据集上的准确率达到61.31%（优于基准模型的55.32%），同时在小规模（6M参数）设置中，训练与预填充阶段的速度优势随序列长度增加而持续扩大。因此，该方案的关键在于通过物理启发的谐振器机制实现对长序列的高效建模，突破传统注意力机制的性能瓶颈。
链接: https://arxiv.org/abs/2607.05583
作者: Archie Chaudhury
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 8 Pages. Accepted at ICANN 2026
Abstract:Contemporary language models are dominated by the transformer architecture, which leverages self-attention mechanisms to enable more efficient, parallelized training across a wide set of documents and corpora. This has allowed transformers to effectively model data across a wide range of modalities and contexts. However, transformers, along with their conventional counterparts such as recurrent neural networks (RNNs) and convolutional neural networks (CNNs), often struggle to maintain efficiency when processing long contexts. We introduce ResonatorLM, a new mechanism that replaces attention with a physics-derived alternative. ResonatorLM treats token sequences as a single, driven one-dimensional latent field and replaces attention dot products with causal functions of damped resonators. We implement ResonatorLM on a traditional network architecture and test it on standard long-context modeling tasks. We find that in a small, 6M matched setting, training and prefill speedups increase with sequence length, decode speed reaches 6.47x compared to that of a standard, optimized transformer at 32K tokens, and accuracy reaches 61.31 percent (compared to 55.32 percent) on WikiText.
[NLP-49] Prompt Robustness Is Task-Dependent: Comparing Objective and Belief-Style Questions in LLM Evaluation
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在基于提示（prompt）的评估中，将生成响应视为模型价值观或信念的可靠指标这一假设的脆弱性问题，尤其针对涉及政治立场、社会态度或价值判断的主观性问题。其核心挑战在于：不同提示形式的微小变化是否会导致模型输出发生显著偏移，从而影响对模型内在信念的准确推断。解决方案的关键在于系统性地评估提示鲁棒性（prompt robustness），通过在客观题（如MMLU、ARC、CulturalBench）与主观题（如Political Compass Test、ValueBench、World Values Survey）两类数据集上，对四种指令微调模型家族施加多种提示变异（包括措辞、框架和格式变化），并量化模型在不同提示变体下答案一致性程度。研究采用广义估计方程（binomial generalized estimating equation）分析发现，提示鲁棒性显著受模型类型、数据集类型及提示类别及其交互作用的影响，表明提示的敏感性高度依赖于问题性质（客观或主观）、具体提示修改方式以及所用模型本身。这一结果揭示了当前基于提示的评估范式在推断模型价值观时存在根本性局限，强调需建立更稳健的评估框架以避免误导性结论。
链接: https://arxiv.org/abs/2607.05554
作者: Sadia Kamal,Arefa Patwary,Anthony Marchiafava,Atriya Sen,Sagnik Ray Choudhury
机构: Oklahoma State University (俄克拉荷马州立大学); University of North Texas (北德克萨斯大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Survey-style evaluations of large language models often treat a prompted response as a measure of a model’s values or beliefs. This assumption is particularly fragile when responses are read as evidence of political values, social attitudes, or beliefs. We ask whether prompt robustness differs between objective questions with fixed answers and subjective questions that ask for opinions or values. We evaluate four instruction-tuned model families on three objective datasets (MMLU, ARC, and CulturalBench) and three subjective datasets (Political Compass Test, ValueBench, and World Values Survey). For each question/statement, we apply multiple types of prompt changes, such as variations in wording, framing, and format, and measure whether the model gives the same answer across variants. Using a binomial generalized estimating equation, we find significant effects of model, dataset, prompt category, and their interactions. The dataset type effect is also significant, and the interaction between dataset type and prompt category is large. These results show that prompt robustness depends on the question type, the prompt change, and the model.
[NLP-50] he yes-no bias of large language models reflects answer order and wording not shifts in moral judgment
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在处理逻辑等价的道德困境问题时，其判断结果因表述方式（如选项顺序、词汇选择）等语义无关因素而产生系统性偏差的问题。尤其关注生成式 AI 在二元判断任务中表现出的“是/否”偏向，这种偏向在人类认知中并不存在，可能源于模型对输出格式的表面敏感性。解决方案的关键在于提出一种心理测量学工具箱——交叉对称化（crossed symmetrization），通过在一组逻辑等价的问题形式中平衡地反转所有非逻辑变量（如选项顺序、词汇表达），从而分离出模型真实道德立场（θ）与由格式诱发的伪信号之间的差异。实验表明，前沿模型的道德立场具有高度格式不变性（跨形式不一致性仅为0.12–0.21，基于±1尺度），而小规模开放权重模型则表现出模型特异性的偏差。进一步分析发现，强制使用“是/否”作为输出界面会引入可分解的人工效应：包括对最后打印选项的顺序偏倚（与人类经典首因效应相反）以及对“否”字词的语义拉力；其中仅在Claude系列模型中显著（故事平均-0.32至-0.86），而GPT-5.5和Gemini几乎无影响，并随推理链延长而减弱。通过将“是/否”替换为任意标签，成功将判断与词汇表征解耦，验证了模型本身并无内在拒绝倾向，其偏差源自印刷表面而非语义内容。最终，研究构建了一个最小模型 $ P = \sigma((\theta \pm m)/s) $，以框架敏感度 $ m $ 和道德决断力 $ s $ 量化人工效应，且二者可明确区别于采样温度。结论强调：真正衡量模型价值必须跨越多种问题框架进行测试，而非单次提问。
链接: https://arxiv.org/abs/2607.05552
作者: Haonan Huang
机构: Princeton University (普林斯顿大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注:
Abstract:Large language models (LLMs) increasingly issue judgments read as binary verdicts, and a growing literature reports such judgments shifting under logically irrelevant changes of wording - among them an amplified yes-no bias on moral dilemmas, absent in humans. A single framing cannot say what such a shift is: in a yes/no question the word “no” is at once logical verdict, lexical token, and last-printed option. We introduce a psychometric battery that separates these: crossed symmetrization - every logically irrelevant factor flipped in balanced pairs - across a corpus of question forms. A graded rating across logically equivalent forms recovers a coherent internal moral scale: frontier models’ stance \theta is nearly format-invariant (cross-form incoherence 0.12-0.21 on a \pm 1 axis); small open-weight models fail in model-specific ways. Forcing the verdict through yes/no overlays a decomposable artifact: an order bias toward the last-printed option - opposite to classic human primacy - plus a lexical pull toward the word “no”; the artifact is substantial only in the Claude models (story-averaged -0.32 to -0.86), \approx 0 for GPT-5.5 and Gemini, and shrinks under extended reasoning. The word and the verdict share one token; swapping the words for arbitrary labels separates them, and the verdict-attached logical bias proves \approx 0 for every frontier model, while model-specific label and order attachments remain: the models are not drawn toward rejecting - the pull follows the printed surface, not the verdict it carries. A minimal model, P = \sigma((\theta \pm m)/s) , summarizes any such artifact by a framing susceptibility m and a moral decisiveness s, measurably distinct from sampling temperature. The battery applies unchanged to any dilemma set and binary format: measuring what a model values requires crossing the frames of the question, not asking once.
[NLP-51] Most LLM Conformity Needs No Speaker: Measuring the Speaker-Free Floor in Peer-Pressure Benchmarks
【速读】： 该论文旨在解决当前大语言模型（LLM）在评估中被误判为“从众”行为的问题，即模型在面对他人答案时改变原本正确的回答。其核心问题是：现有基准测试中广泛使用的“从众提示”（conformity prompts）同时引入了两个混杂因素——发言者存在性与错误答案的重复呈现，导致无法准确区分模型行为变化究竟是由社会影响（如权威或群体压力）驱动，还是仅因重复错误信息本身所致。论文的关键解决方案是提出“无来源条件”（no-source condition），即在不包含任何发言者身份信息的前提下，仅保留被重复的错误答案进行测试。实验结果表明，在六种开源大模型和七个问答与推理数据集上，该条件下66.5%的初始正确回答发生有害修正，显著高于仅重新提问（re-ask）时的10.3%，说明重复错误信息本身即可引发模型偏离正确答案。此外，该效应在答案重述、选项隐藏等不同设置下依然存在，且模型在错误翻转时通常表现出高度自信，简单校准无法恢复原答案。研究进一步发现，只有“专家小组”式框架能显著提升该基础效应，而简单的个人标签则无明显作用。因此，论文强调：评估模型从众行为的方法论必须首先测量“去发言者”后的基础修正率，否则将可能将单纯的文本重复效应误判为社会影响。
链接: https://arxiv.org/abs/2607.05545
作者: Yibo Hu,Jiaming Qu
机构: Illinois Institute of Technology (伊利诺伊理工学院); Amazon (亚马逊)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:LLM conformity is often used to describe cases where a model changes a correct answer toward a peer or group response. We show that most of this apparent conformity survives even after the peer is removed. The reason is a confound: standard conformity prompts mix two cues at once, the presence of a speaker and the repeated wrong answer itself. Existing benchmarks vary these cues together, so they cannot tell how much of the revision actually depends on the speaker. We introduce a no-source condition: the same asserted answer with the explicit speaker removed. Across six open-weight LLMs and seven QA and reasoning datasets, this condition alone causes harmful revision in 66.5% of initially correct cases, compared with 10.3% under a plain re-ask. The effect also remains when the repeated answer is paraphrased and when answer options are hidden in an open-ended setting. Source framing mainly modulates this floor: expert-panel framing raises it, while minimal person labels do not reliably raise it. When models flip, they are usually confidently wrong, and simple recalibration does not recover the original answer. Source attribution still matters, but it should be measured as an increment above this speaker-free floor. The methodological lesson is that conformity benchmarks should first measure what remains after the speaker is removed; without this step, benchmarks may mistake repeated text for social influence.
[NLP-52] Prompt-to-Paper: Agent ic AI System for Bioinformatics
【速读】： 该论文旨在解决当前生成式AI在学术论文自动化生成中存在的三大核心问题：一是生成的论断缺乏可验证文献的确定性支撑，二是实验结果多为虚构而非真实执行，三是缺乏标准化、多维度的质量评估框架以衡量AI生成论文是否具备实际发表所需的严谨性与质量。其解决方案的关键在于提出一个名为Prompt-to-Paper的多智能体框架，通过三项集成创新实现突破：首先，采用基于检索增强生成（Retrieval-Augmented Generation, RAG）的确定性流水线，结合章节感知的相关性评分与雪球式引文扩展机制，确保每一项论断均锚定于60–100篇可验证的文献语料库中；其次，引入自主编码智能体执行真实的计算生物学实验，替代以往合成数据，生成可信的数值结果；最后，构建一个八维自动化质量评分器，基于已发表论文的近似参考统计量进行基准校准，并加入显式的幻觉惩罚机制，实现标准化、可复现的质量评估。此外，系统还设计了以质量为导向的迭代改进循环，通过上下文丰富的修订智能体将每轮输出引导至三种研究人员干预动作之一，并每隔十轮触发深度研究周期，重新运行实验并从更优输出出发重写稿件。在五个生物信息学案例研究中，系统成功生成符合投稿格式的PDF文档，且无超出范围的引用；质量评分平均提升17.96分（满分100），最高达26.04分；人工评审平均得分为7.0/10，验证了生成内容的合理性；单篇完整论文生成成本约为0.31美元。
链接: https://arxiv.org/abs/2607.05456
作者: Ramsha Kamran,Maheera Amjad,Zartasha Mustansar,Arsalan Shaukat,Salma Sherbaz,Muhammad U.S. Khan
机构: School of Interdisciplinary Engineering and Sciences (SINES), National University of Sciences and Technology (NUST), Islamabad, Pakistan; College of Electrical and Mechanical Engineering (CEME), National University of Sciences and Technology (NUST), Rawalpindi, Pakistan
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Quantitative Methods (q-bio.QM)
备注: NA
Abstract:While recent advances in large language models have enabled end-to-end automated manuscript generation, existing systems suffer from three critical deficiencies: (i) generated claims are not deterministically grounded in verifiable literature, (ii) experimental results are frequently fabricated rather than executed, and (iii) there exists no standardized, multi-dimensional framework to assess whether AI-generated manuscripts meet the quality and rigor required for real-world publication. We present Prompt-to-Paper, a multi-agent framework that directly addresses this evaluation gap through three integrated innovations. First, a deterministic retrieval-augmented generation pipeline with section-aware relevance scoring and snowball citation expansion grounds every claim in a verifiable corpus of 60–100 papers. Second, an autonomous coding agent executes real computational biology experiments replacing synthetic outputs with genuine numerical results. Third, an eight-dimensional automated quality scorer, benchmarked with approximate reference statistics from published papers and augmented with explicit hallucination penalties, provides standardized, reproducible quality assessments. The quality-driven improvement loop uses a context-rich reviser that routes each iteration to one of three researcher actions and fires a deep research cycle every ten iterations to re-run experiments and re-manuscript from stronger outputs. We validate the system on five bioinformatics case studies; all five cases compiled submission-formatted PDFs with zero out-of-range citations. The improvement loop raises manuscript quality by an average of +17.96 points on a 0–100 scale (maximum +26.04. As partial external checks, a human reviewer scored the five manuscripts at an average of 7.0 out of 10. Complete manuscripts are produced at approximately 0.31 USD per paper.
[NLP-53] xt Distance from Nested and Hierarchical Repetitions: A Compression-Based Perspective
【速读】： 该论文旨在解决传统文本建模方法在低资源、分布外（OOD）场景下性能下降的问题，尤其是现有基于深度学习的模型（如BERT）和压缩基距离（如gzip-based NCD）在缺乏充足训练数据或面对非训练分布时泛化能力不足的局限性。其核心解决方案是提出一种基于算法信息论（Algorithmic Information Theory, AIT）的结构化序列分析方法——梯级路径（Ladderpath），通过提取语言序列中重复子结构的嵌套与层次关系，实现对数据最小生成程序的近似描述。该方法的关键在于利用Ladderpath构建具有可解释性的结构化表示，并据此定义三种距离度量：归一化压缩距离（NCD）及两种直接源自Ladderpath表示的替代距离。这些距离度量无需训练、计算轻量，且在跨分布、少样本等挑战性任务中均表现出优越且稳定的分类性能，显著优于gzip-based NCD和BERT模型。结果表明，Ladderpath所捕捉的结构性特征能够有效保留序列的内在信息属性，为实现领域无关、无需训练的序列理解提供了新范式。
链接: https://arxiv.org/abs/2607.05416
作者: Xiaojun Hu,Jing Wang,Jingwen Zhang,Fengyao Zhai,Xiao Xie,Hao Liao,Zengru Di,Yu Liu
机构: Beijing Normal University (北京师范大学); Sun Yat-sen University (中山大学); Shenzhen University (深圳大学)
类目: Computation and Language (cs.CL); Information Theory (cs.IT)
备注:
Abstract:We present a new method for structural sequence analysis grounded in Algorithmic Information Theory (AIT). At its core is the Ladderpath approach, which extracts nested and hierarchical relationships among repeated substructures in linguistic sequences – an instantiation of AIT’s principle of describing data through minimal generative programs. These structures are then used to define three distance measures: a normalized compression distance (NCD), and two alternative distances derived directly from the Ladderpath representation. Integrated with a k -nearest neighbor classifier, these distances achieve strong and consistent performance across in-distribution, out-of-distribution (OOD), and few-shot text classification tasks. In particular, all three methods outperform both gzip-based NCD and BERT under OOD and low-resource settings. These results demonstrate that the structured representations captured by Ladderpath preserve intrinsic properties of sequences and provide a lightweight, interpretable, and training-free alternative for text modeling. This work highlights the potential of AIT-based approaches for structural and domain-agnostic sequence understanding.
[NLP-54] CANONIC: Governance Is Compilation
【速读】： 该论文旨在解决生成式AI（Generative AI）在大规模生成文本过程中出现的“内容可靠性”问题，即由大语言模型生成的文本往往速度远超人工审核能力，导致虚假或不可靠信息（称为“slop”）大量涌入知识库，威胁信息可信度。其核心挑战在于：如何在不依赖主观判断的前提下，实现对生成内容的有效治理与可审计性。解决方案的关键在于提出一种名为CANONIC的受控智能架构，将内容准入机制类比于编译器对程序的语法、作用域与类型检查，通过三个可形式化验证的公理（三元组：一致性、继承性、内省性）构建一个可判定、线性时间完成的结构化准入规则体系。该机制并不直接识别“slop”，而是确保所有内容均以可追溯的方式锚定于明确定义、版本化提交及证据窗口之中，从而保障整个知识记录链的可复现性与端到端可验证性，从根本上实现对信息可信度的系统性治理。
链接: https://arxiv.org/abs/2607.05410
作者: Dexter Hadley
机构: CANONIC Foundation(_CANONIC基金会); American Board of Precision Medicine(美国精准医学委员会)
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Programming Languages (cs.PL)
备注: 28 pages, 4 figures. Pre-registered cross-provider evaluation harness and per-regime results at this http URL . Construction claims resolve to commands run at the evidence-window-close ref (see Appendix C)
Abstract:We present CANONIC: governed intelligence that compiles digital artifacts into an evidence ledger at scale. Large language models generate prose faster than anyone can check it, the failure Oxford Languages named ‘slop’, its 2025 Word of the Year. CANONIC governs whether content may enter a corpus the way a compiler decides whether a program is well-formed: mechanically, by a grammar, at the boundary of admission. Governance reduces to three axioms (Triad, Inheritance, Introspection) that map one-to-one onto compiler theory’s syntax, scope-resolution, and type-system layers, and admission is a decidable, linear-time check. We then ask, with a pre-registered cross-provider benchmark across four regimes, whether structural admission keeps slop out. It does not: no prose-reading gate reliably separates reliable from unreliable content. Slop is not a property an algorithm computes. It is a verdict of domain expertise. So a governance layer does not decide slop; it keeps the record auditable – every claim anchored to a definition, a commit, and an evidence window, reproducible and checkable end to end.
[NLP-55] CCBENCH: Assessing LLM Cultural Competence via Implicitly Signaled Norms using Health Queries
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在与用户交互时缺乏文化敏感性的问题，即模型往往依赖静态的人口统计特征而非动态的、隐含的文化价值观信号，从而导致刻板印象和不公平的互动。其核心挑战在于如何评估并提升模型对文化连续体（culture as a continuum of norm adherence states）的感知与适应能力，而非简单地将文化视为二元归属。解决方案的关键是提出一种名为CCBENCH的评估框架，将文化视为一个由规范遵从度构成的连续谱系，而非非黑即白的身份标签。通过构建针对健康领域的案例研究——CCBENCH-Health，该框架包含60个基于理论建构的角色（personas），覆盖六种文化背景，每个角色参与18轮真实对话，共生成3,120次独特交互，并以52个来自真实用户论坛的医疗问题进行测评。实验表明，即使表现最佳的模型也仅在20%-30%的情况下给出文化恰当的回应；当引入“思维链”（Chain-of-Thought, CoT）提示以关注对话历史中的文化线索时，性能仅提升3%-5%，且模型更倾向于回避文化规范而非遵循它们，暴露出对内置偏见的偏好。尤其在阿富汗语境下，文化线索几乎无法引导出恰当的健康建议（平均仅8.8%）。此外，研究发现模型对隐含的文化对话风格比对明确的文化实践更具适应性，但这种差异具有显著的文化特异性。
链接: https://arxiv.org/abs/2607.05405
作者: Vasudha Varadarajan,Akhila Yerukola,Mona T. Diab,Maarten Sap
机构: Carnegie Mellon University (卡内基梅隆大学)
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 34 pages
Abstract:To interact with users fairly and without stereotyping, AI models must display cultural competency, i.e., the ability to infer and adapt to a user’s implicitly signaled cultural values, rather than relying on static demographic traits. We introduce CCBENCH, a framework for evaluating cultural competency in large language models (LLMs), treating culture as a continuum of norm adherence states rather than as a binary state of cultural belongingness. As a case study on health, we create CCBENCH-Health, which includes 60 theoretically grounded personas exhibiting varied norm-adherence states across six cultures, each engaging in 18 realistic dialogues. Each persona is evaluated on 52 authentic healthcare questions drawn from real user forums, yielding 3,120 unique interactions. Benchmarking five leading models reveals that even the best achieve culturally appropriate responses only 20-30% of the time. When explicitly prompted to focus on culturally relevant cues from the conversational history (CoT), performance improves modestly by 3-5% on average. We find that models perform best when personas avoid cultural norms rather than follow them, revealing a persistent asymmetry, suggesting a preference in the models to align with built-in biases than adapt to cultural cues. This is especially observed in the Afghan context (Avg: 8.8%), where cultural cues rarely yield appropriate health advice. Finally, we find that models sometimes adapt more readily to implicit, cultural conversational styles than to explicitly stated cultural practices, though this varies across cultures.
[NLP-56] Benchmarking KV-Cache Optimizations across Task Quality and System Performance for Long-Context Serving
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）在长上下文（long-context）推理场景下因键值缓存（KV-cache）增长导致的性能瓶颈问题。现有KV-cache压缩技术难以有效比较，主要因其在不同模型、任务、资源预算及服务架构下进行评估，缺乏统一基准。为此，本文提出一个面向工作负载感知的基准测试框架，系统评估了涵盖量化、剪枝与合并三类方法的代表性优化机制（如KIVI、TurboQuant、SnapKV和CaM），并在Llama-3.1-8B-Instruct与Mistral-7B-Instruct-v0.3上针对多文档问答、单文档问答、少样本学习及摘要生成等典型任务进行测试。该基准综合衡量任务质量、平均输出吞吐率、首次生成时间（time-to-first-token）以及实际压缩比，并按上下文长度分桶分析。研究发现，仅关注压缩比无法准确预测端到端性能表现；其中，KIVI4在不同模型间展现出最稳定的质量表现，SnapKV在长上下文场景下提供最强吞吐能力，而CaM虽在特定问答任务中取得显著提升，但其性能对工作负载高度敏感，表现出较大的质量波动与压缩比变化。上述结果表明，应根据具体应用场景动态选择适配的KV-cache优化策略，而非采用“一刀切”的压缩方案，从而为长上下文服务系统的部署提供了关键指导。
链接: https://arxiv.org/abs/2607.05399
作者: Nikita Agrawal,Ruben Mayer
机构: University of Bayreuth(拜罗伊特大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC)
备注:
Abstract:Large language model serving is increasingly limited by KV-cache growth under long-context workloads, yet existing KV-cache compression techniques are difficult to compare because they were evaluated on different models, tasks, budgets, and serving stacks. This paper presents a workload-aware benchmark of representative KV-cache optimization mechanisms spanning quantization, pruning, and merging, including KIVI, TurboQuant, SnapKV, and CaM, evaluated on LongBench-style multi-document QA, single-document QA, few-shot learning, and summarization workloads using Llama-3.1-8B-Instruct and Mistral-7B-Instruct-v0.3. The benchmark measures task quality, mean output throughput, mean time-to-first-token, and realized compression ratio across context-length buckets. The results show that the compression ratio alone is a poor predictor of end-to-end performance. KIVI4 provides the most stable quality across models, SnapKV delivers the strongest long-context throughput, and CaM yields large gains on selected QA workloads but exhibits substantial workload sensitivity in both quality and realized compression ratio. These findings motivate workload-aware selection of KV-cache mechanisms rather than one-size-fits-all compression and provide deployment guidance for long-context serving systems.
[NLP-57] How Personas Can Influence Agents to Play Split or Steal
【速读】： 该论文旨在解决在社会困境情境下，角色设定（Persona）提示对大语言模型代理（LLM agents）战略行为影响的不确定性问题。其核心挑战在于：尽管角色设定常被用于引导智能体行为，但在重复性信任博弈（如迭代式“分钱或抢钱”游戏）中，不同角色设定是否能稳定引导合作或背叛策略尚不明确。解决方案的关键在于通过系统实验设计，评估四种开源大模型（Ministral 3:3b、phi4:14b、Gemma3:12b、Gemma4:e4b）在不同温度参数（0.3、0.7）及确定性推理（零温度）条件下，结合不同人格化提示（基于大五人格特质的Prosocial、Principled、Analytical等）时的行为表现，并与由GPT-4.1 mini控制的虚拟人类（Virtual Human, VH）进行交互。研究发现，角色设定与模型选择共同显著影响策略演化，其中Prosocial和Principled人格提示最能促进持续合作，而Analytical提示则更易引发对虚拟人类的剥削行为；此外，对话内容分析表明，涉及友谊的话题与“分钱”决策高度相关，而金钱与复仇主题则多见于“抢钱”行为，情绪标签虽以中性或积极为主，但解释力有限。这一结果揭示了角色设定与模型内在特性在重复互动中的协同作用机制，为后续基于虚拟现实的人机协作研究提供了可量化的基准。
链接: https://arxiv.org/abs/2607.05398
作者: Carlos Leon,Alexandre Rodrigues,Pedro Gamito,Thomas D. Parsons
机构: Universidade Lusófona (里斯本大学); Universitat de Barcelona (巴塞罗那大学); Université Paris Cité (巴黎城市大学); Computational Neuropsychology and Simulation (CNS) Laboratory, Arizona State University (亚利桑那州立大学认知神经心理学与仿真实验室)
类目: Computation and Language (cs.CL); Computers and Society (cs.CY)
备注:
Abstract:Personas are often employed to guide large language model agents, yet their effectiveness in shaping strategic behavior in social dilemma settings remains uncertain. To address this, we examined the impact of persona prompts in an iterated Split or Steal game where persona-driven agents interacted with a Virtual Human (VH) controlled by a fixed prompt. Agents were instantiated from four open models (Ministral 3:3b, phi4:14b, Gemma3:12b, and Gemma4:e4b) at two temperature settings (0.3 and 0.7) and deterministic decision with zero temperature, while the VH was powered by GPT 4.1 mini. Across 160 sessions of 15 rounds each conducted in European Portuguese, mutual Split outcomes dominated (roughly 74 percent of rounds), with exploitation occurring in fewer than 11 percent of rounds. Model choice significantly influenced behavior: phi4 and Ministral 3:3b remained consistently cooperative across temperatures, whereas Gemma3:12b and Gemma4:e4b exhibited more varied strategies and outcomes. Analyses based on Big Five personality traits indicated that Prosocial and Principled personas were most consistently cooperative, while Analytical personas were more likely to exploit the VH. Topic analysis revealed that friendship-related dialogue aligns with Split decisions, whereas money and vengeance-related content is more prevalent in Steal outcomes; sentiment labels were predominantly neutral or happy and provided limited additional explanatory value. These findings characterize the interaction between persona prompts and model differences in repeated trust games and serve as a baseline for planned virtual reality studies involving human participants interacting with an embodied VH.
[NLP-58] Diffusion Language Model Parallel Decoding via Product-of-Experts Bridge ICML2026
【速读】： 该论文旨在解决扩散语言模型（Diffusion Language Models, DLMs）在并行解码过程中因缺乏词元依赖性而导致生成质量低于自回归（Autoregressive, AR）模型的问题。尽管DLM具备显著的解码速度优势，但其与AR模型之间存在巨大的分布差异，使得直接通过重要性采样从DLM向AR目标迁移时需要大量粒子采样，计算成本高昂。本文提出PoE-Bridge框架，其核心创新在于引入一个中间分布作为桥梁，该分布以扩散语言模型（DLM）为提议分布、自回归（AR）模型为目标分布，构建两者的乘积专家（Product-of-Experts, PoE）形式。通过该中间分布，先利用DLM并行生成多个候选文本序列，再通过拒绝采样将候选结果向PoE分布对齐，最后采用重要性采样进一步校准至AR目标分布。此外，论文还提出混合温度采样以增强多样性、弹性拒绝窗口以减少无效验证开销。实验表明，PoE-Bridge在保持5倍于标准DLM解码速度的同时，显著提升了生成精度，恢复了至少95%的AR模型性能，在数学推理与代码生成等高难度任务上有效弥合了生成质量差距。
链接: https://arxiv.org/abs/2606.08048
作者: Juntong Shi,Brian L. Trippe,Jure Leskovec,Stefano Ermon,Minkai Xu
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG); Performance (cs.PF)
备注: ICML 2026
Abstract:Diffusion language models (DLMs) offer substantial speed advantages through parallel decoding, but the lack of token dependencies limits generation quality compared to autoregressive (AR) models. Recent progress attempts to bridge the gap via importance sampling, with DLM being the proposal and AR being the target. However, due to the huge gap between their distributions, the sampling requires a large number of particles and is thus expensive to compute. In this paper, we introduce PoE-Bridge, a novel decoding framework that drastically improves generation speed and accuracy by introducing an intermediate distribution to bridge the gap. The distribution is constructed as a Product-of-Experts (PoE) of the DLM proposal and the AR target. With the intermediate distribution, we first use the DLM to draft multiple continuations in parallel, then apply rejection sampling to verify the drafted tokens and move the resulting candidates toward the PoE. We then use importance sampling to further correct the PoE-aligned candidates toward the AR target. We further propose several improved techniques, including mixed-temperature sampling for enhanced diversity and elastic rejection windows for reducing wasted verification. Empirically, PoE-Bridge achieves significantly improved accuracy with 5\times speedup over the standard DLM decoding approach, and recovers at least 95% of the target AR model’s performance, efficiently advancing most of the quality gap on challenging mathematical reasoning and coding tasks. Our code is available at this https URL.
[NLP-59] WordVoice: Explicit and Decoupled Multi-Dimensional Word-Level Control for LLM -Based TTS
【速读】： 该论文旨在解决基于大语言模型（Large Language Model, LLM）的文本转语音（Text-to-Speech, TTS）系统在实现细粒度声学控制方面的核心瓶颈问题，尤其是在需要精确风格调控与严格时间对齐的应用场景（如有声书朗读、视频配音）中，现有方法因依赖隐式端到端生成范式而难以显式操控词级声学属性。其关键解决方案在于提出一个统一的框架——WordVoice，通过两个核心创新实现高精度词级控制：一是构建了大规模双语标注数据集WordVoice-5A（4.7k小时），包含时长、边界、能量、音高和声调五个维度的词级标注，依托语言学引导的数据构建流程克服了细粒度标注数据稀缺的问题；二是引入“边界标记（bound-token）”机制，将隐式生成过程重构为显式的“声学规划”范式，使LLM能够自适应地进行多任务韵律规划并支持灵活的手动干预；同时，在从离散标记到波形的转换阶段引入细粒度声学调制模块，有效弥合了压缩离散标记与连续波形之间的分辨率鸿沟，确保词级声学属性在生成过程中严格对齐。实验表明，该方法在多个声学维度上实现了优越且解耦的控制能力，同时保持了良好的零样本合成稳定性。
链接: https://arxiv.org/abs/2607.06461
作者: Sihang Nie,Jinxin Ji,Xiaofen Xing,Deyi Tuo,Chengbin Jin,Jialong Mai,Xiangmin Xu
机构: 未知
类目: Audio and Speech Processing (eess.AS); Computation and Language (cs.CL); Sound (cs.SD)
备注: 10 pages, 4 figures, 6 tables; Preprint
Abstract:While recent Large Language Model (LLM)-based Text-to-Speech (TTS) systems have achieved remarkable naturalness, they predominantly rely on implicit end-to-end generation paradigms, resulting in coarse-grained control. In scenarios demanding precise stylistic interventions and strict temporal alignment, such as audiobook narration and video dubbing, the inability to explicitly manipulate word-level acoustic attributes remains a critical bottleneck. This limitation is primarily amplified by the severe scarcity of fine-grained annotated datasets and the architectural challenge of integrating multi-dimensional control signals into discrete autoregressive generation. To address this, we propose a unified framework for highly precise word-level control. First, we construct WordVoice-5A, a massive 4.7k-hour bilingual dataset featuring five-dimensional word-level annotations (duration, boundary, energy, pitch and tone) developed through a rigorous linguistically-guided pipeline. Second, we introduce WordVoice to transform the implicit generation process into an explicit, highly controllable paradigm. Specifically, we introduce a bound-token mechanism within the LLM to formulate an explicit ``acoustic planning’’ process, enabling adaptive multi-task prosodic planning and flexible manual intervention. Furthermore, we augment the token-to-waveform stage with a fine-grained acoustic modulation module, bridging the resolution gap to strictly align word-level attributes between highly compressed discrete tokens and continuous waveforms. Extensive experiments demonstrate that WordVoice achieves superior, decoupled control over multiple acoustic dimensions while maintaining competitive zero-shot synthesis stability. The code and audio samples are publicly available at this https URL.
信息检索
[IR-0] DynaKRAG : A Unified Framework for Learnable Evidence Control in Multi-Hop Retrieval-Augmented Generation
链接: https://arxiv.org/abs/2607.06507
作者: Yaqi Wu,Xiaolei Guo,Chenyu Zhou,Jiaqi Huang,Xianfa Zhang,Junxu Zhang,Zhuo Yu,Zhubo Shi,Jianghao Lin,Dongdong Ge
类目: Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注:
Abstract:Multi-hop retrieval-augmented generation (RAG) acquires evidence sequentially, with each new document potentially revealing missing facts, bridge entities, query defects, or sufficient support for answering. Existing methods provide useful operations such as iterative retrieval, query reformulation, evidence critique, and sufficiency judging, but typically organize them within method-specific pipelines or predefined control topologies. This leaves underexplored how to learn a shared state-conditioned policy that chooses among currently valid evidence operations. We introduce DynaKRAG, which formulates multi-hop evidence acquisition as state-conditioned control over atomic evidence operations. At each step, a validity layer constructs the executable action set, and a learned controller selects the next operation. The resulting transition updates the evidence state and may enable new operations at subsequent steps. With Qwen2.5-7B-Instruct, DynaKRAG achieves F1 scores of 0.5998 on HotpotQA, 0.5340 on 2Wiki, and 0.3061 on MuSiQue, outperforming the strongest controlled baseline on all three benchmarks. Replacing the learned controller with a uniform-valid policy reduces F1 by 3.96–5.78 points, while removing sufficiency feedback hurts all three datasets. Controlled retrieval-cap experiments further show that additional retrieval is not uniformly beneficial. Together, these results demonstrate the benefit of coordinating retrieval, diagnosis, and gap-directed acquisition under an evolving evidence state.
[IR-1] Learn to Pool: Lightweight Fine-Tuning for Flexible Multi-Vector Compression ECIR2026
链接: https://arxiv.org/abs/2607.06036
作者: Stefan Josef
类目: Information Retrieval (cs.IR)
备注: The 1st Late Interaction Workshop (LIR) @ ECIR 2026
Abstract:Late interaction models have shown strong generalization capabilities, often outperforming much larger dense embedding models. One challenge to their widespread deployment is the large number of token vectors they produce per document and the associated storage and memory costs. Pooling tokens at inference time has shown great promise to reduce the vector count with limited effects on retrieval accuracy. Large-scale pooling-aware training has demonstrated even more impressive results at high compression rates. We propose lightweight fine-tuning as a practical alternative and find that even minimal pooling-aware training with k-means yields broad gains over inference-only pooling, shows evidence of transfer across pooling methods and datasets, and - with multi-factor training - produces a single model effective across different compression levels. Our strongest model outperforms the unpooled baseline on BEIR SciFact across pool factors 1-6, implying a vector compression rate of 83% at no cost to retrieval accuracy.
[IR-2] Uncertainty-Aware Cross-Modal Remote Sensing Image-Text Retrieval via Evidential Learning
链接: https://arxiv.org/abs/2607.06032
作者: Zhuoyue Wang,Xueqian Wang,Gang Li,Chengxi Li,Yongpan Liu,Yifang Ban
类目: Information Retrieval (cs.IR)
备注:
Abstract:In cross-modal remote sensing image-text retrieval (CMRSITR), test-time remote sensing (RS) images and textual descriptions may deviate from well-curated benchmark conditions due to sensor- and atmosphere-related image degradations and text-side RS-vocabulary heterogeneity. Under such non-ideal conditions, existing CMRSITR methods may produce unreliable retrieval results because they perform retrieval with full certainty for each query and do not distinguish the varying uncertainty across queries. To address this issue, we propose an evidential learning-based CMRSITR (ELC) method for uncertainty-aware retrieval. During the training phase of ELC, evidential learning (EDL) is employed to model the inter-modal correspondences between RS images and textual descriptions as Dirichlet distributions, from which the uncertainty of each query can be obtained. Based on the EDL outputs, uncertainty-correctness alignment learning (UCL) is introduced to align the estimated uncertainty with retrieval correctness, encouraging high uncertainty for incorrect retrieval and low uncertainty for correct retrieval. Furthermore, intra-modal relationship learning (RL) distills the intra-modal similarity structure from pretrained mentor encoders for the trainable encoders, thereby making the Dirichlet distributions modeled by EDL more discriminative. In the test phase of ELC, the estimated uncertainty is compared with a threshold determined by a fixed deferral ratio, where low-uncertainty queries are directly returned and high-uncertainty queries are refined by RS-aware test-time augmentation (RS-TTA). Experimental results demonstrate that ELC achieves competitive retrieval performance compared with state-of-the-art CMRSITR methods and provides stronger robustness under the evaluated RS-specific degradations, including sensor- and atmosphere-related image perturbations and RS-vocabulary heterogeneity.
[IR-3] Faithful or Findable? Evaluating LLM -Generated Metadata for RDF Dataset Search SIGIR2026
链接: https://arxiv.org/abs/2607.05970
作者: Riccardo Terrenzi,Serkan Ayvaz
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI)
备注: 5 pages, 1 figure, accepted at SynthIR @ SIGIR 2026
Abstract:Dataset search depends heavily on metadata, making LLM-generated metadata a consequential form of synthetic content in retrieval systems. We study six metadata-generation settings for RDF datasets, ranging from simple rewriting to profile-grounded and agentic graph-based generation, and evaluate them jointly for retrieval effectiveness and faithfulness. Unconstrained metadata rewriting delivers the strongest retrieval gains over the original metadata, but it is also the least faithful, showing that search improvements can be driven by unsupported semantic expansion. More grounded settings substantially improve faithfulness, and profile-grounded rewriting provides the most balanced trade-off between retrieval effectiveness and grounding. These findings position synthetic metadata as a system-level IR problem in which effectiveness, provenance, and trust must be evaluated together.
[IR-4] InfluMatch: Frontier-Quality KOL Search at 4B-Model Cost
链接: https://arxiv.org/abs/2607.05968
作者: Krittanon Kaewtawee,Petmongkon Pornpichitsuwan,Natchaya Temyingyong,Nutnicha Laplamoon,Wachiravit Modecrua,Krittin Pachtrachai,Touchapon Kraisingkorn
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
备注:
Abstract:Matching influencers (KOLs) to free-form, multi-part Thai marketing criteria is today served either by keyword search over structured profiles, which misses semantic fit, or by prompting frontier LLMs over every candidate, which is accurate but slow and expensive. We present InfluMatch, a low-cost three-stage cascade – retrieval \rightarrow rerank \rightarrow reason – built entirely from small open-weight models: dense retrieval returns 50 candidates, a 4B pointwise reranker scores each by the log-probability of a single Yes token and keeps 10, and a 4B reasoner grades the shortlist per criterion on a rubric with a Thai rationale. The cascade is designed for cost: reasoning over a filtered top-10 halves token spend versus reasoning over all 50 while scoring 14 points higher. End-to-end against human relevance labels on an 11-query set with all 50 candidates labeled, the full cascade reaches 94.1% P@5, versus a retrieval-only baseline near random; it matches the frontier model Kimi-K2.6 (91.8%) while emitting \sim35\times fewer output tokens and serving a 50-KOL query in \sim20 s on one A100. Notably, the only fine-tuning that pays off is pairwise: a SimPO-tuned reranker matches the frontier baseline’s best-pick accuracy (78.0 EM), whereas fine-tuning the reasoner on pointwise per-criterion labels improves offline scores yet degrades end-to-end ranking – an inversion we trace to the design of the absolute labeling task – leaving the untuned base model as the strongest deployed reasoner. The result is a deployable, explainable KOL search system at a small fraction of frontier serving cost.
[IR-5] CMDR: Contextual Multimodal Document Retrieval ECCV2026
链接: https://arxiv.org/abs/2607.05927
作者: Ryota Tanaka,Taku Hasegawa,Kyosuke Nishida
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by ECCV 2026; project page: this https URL
Abstract:Multimodal document retrieval aims to retrieve relevant pages while preserving both textual and visual content from the original document. However, existing benchmarks primarily evaluate simple lexical or semantic matching, and most methods encode pages independently. Consequently, they overlook the contextual information in the document required to resolve queries that aggregate information across multiple pages. In this paper, we introduce CMDR and CMDR-Bench, a new multimodal document retrieval task and benchmark that require modeling document context. To address this challenge, we propose CMDR-Embed, a contextual multimodal embedding framework that explicitly incorporates document context by jointly encoding multiple pages and deriving page-level embeddings from a shared contextual representation. Furthermore, we introduce CMCL, a contextual multimodal contrastive learning objective that effectively trains CMDR-Embed by balancing contextual modeling with page-level discriminability. Experiments demonstrate that CMDR-Embed significantly outperforms non-contextual embeddings, highlighting the importance of context-aware multimodal embeddings for advancing document retrieval.
[IR-6] Quantifying and Expanding the Theoretical Capacity of Late-Interaction Retrieval Models
链接: https://arxiv.org/abs/2607.05803
作者: Julian Killingback,Varad Ingale,Hamed Zamani,Cameron Musco
类目: Information Retrieval (cs.IR)
备注: 21 Pages, 1 Figure
Abstract:Late-interaction retrieval models that use the MaxSim similarity function have shown strong empirical performance, often outperforming single-vector dense and sparse retrieval models. Despite these empirical findings, little is known about the theoretical representation power of MaxSim and how it compares to other retrieval approaches. This paper shows by construction that MaxSim similarity can exactly replicate the inner product between any two non-negative k-sparse vectors with possibly infinite dimension, requiring only O(k) representation space. Moreover, there exist similarities that MaxSim can express while standard vector inner products with the same representation space cannot. Leveraging our theoretical framework, we introduce Signed MaxSim which allows late-interaction models to exactly replicate any real-valued inner product, something we prove standard MaxSim is not capable of. We also show that MaxSim can act as an aggregation of soft-OR operations and as an evaluator of logical expressions in positive Conjunctive Normal Form. Our findings show that MaxSim is at least as capable as standard vector inner products for any non-negative vectors and our extension, Signed MaxSim, is as capable for any vectors. Both similarities possess additional capabilities that inner product cannot replicate, marking one of the first theoretical justifications and quantifications of late-interaction methods. Our theoretical findings are supported empirically: on a retrieval task featuring queries with negations, Signed MaxSim improves out-of-domain performance significantly over a standard ColBERT/MaxSim baseline with nDCG@10 increasing from 0.597 to 1.000 under a vocabulary shift and from 0.008 to 0.788 on negation-only queries.
[IR-7] Inject or Navigate? Token-Efficient Retrieval for LLM Analysis of Transactional Legal Documents
链接: https://arxiv.org/abs/2607.05764
作者: Mahmoud Hany,Mourad ElSheraey,Mahmoud Said,Peter Naoum
类目: Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注: 17 pages, 2 figures, 8 tables
Abstract:Answering questions over a set of transactional legal documents is most simply done by injecting the whole corpus into the LLM’s context window on every query. That baseline maximises retrieval recall, but its token footprint scales with the corpus rather than the question, and long-context degradation scales with it. We report what it took to replace full-corpus injection in a legal-document analysis system, comparing it against two structured retrieval modes over our proprietary structure-aware chunking: embedding retrieval (NAVEMBED) and LLM navigation over a compact structured index (NAVINDEX). On a 20-question benchmark with verified ground-truth answers, a position-bias-controlled, reference-anchored pairwise judge scored semantic retrieval with reranking tied with injection on 16 of 18 document-bound questions (injection preferred on 2) while attending to 17.3x fewer input tokens (a general-text-embedding (GTE) configuration reaches 29.9x at a lower tie rate); both modes were judged tied on the 2 out-of-scope controls. NAVINDEX was judged tied on all 18 at a 1.61x smaller total token footprint, a ~56x smaller answering context, and 25% lower dollar cost. We derive a closed-form caching-crossover rule: cached injection is cheaper in dollars only while the corpus stays below roughly ten times the retrieval payload. Scope and uncertainty are quantified in Section 8.
[IR-8] SCOReD: Student-Aware CoT Optimization for Recommendation Distillation
链接: https://arxiv.org/abs/2607.05734
作者: Haz Sameen Shahgir,Yufei Li,Frank Shyu,Luke Simon,Sandeep Pandey,Xi Liu,Yue Dong
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI)
备注: 31 pages
Abstract:Chain-of-thought (CoT) distillation in the recommendation domain is a necessary precursor to RL training, but raw teacher traces are ill-suited to this task. Large teachers approach the recommendation task with unusually high reasoning uncertainty, repeatedly rechecking their answers without revising them; supervised fine-tuning on such traces produces verbose students that never revise their initial guess. Furthermore, due to the novelty of the recommendation domain, the teacher’s reasoning traces are highly out-of-distribution for the small student LLM. We propose Student-Aware CoT Optimization for Recommendation Distillation (SCOReD), a CoT optimization framework tailored to recommendation that first parses each teacher trace into typed segments and uses the student LLM’s attention to score the importance of each segment. Then SCOReD dynamically selects a per-segment edit (KEEP / REWRITE / FUSE / PRUNE) based on the output length and comparative log probability lift of the answer given the edit as per the student. Therefore, SCOReD prunes redundant sections of the reasoning trace while preserving information-dense sections and adapts raw teacher traces to the student’s output distribution. Training on SCOReD-optimized CoTs provides a cleaner learning signal to the student model and improves over baseline SFT by 1.56% NDCG and 1.9% Recall@5, while reducing reasoning length by 27.3%. Comments: 31 pages Subjects: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI) Cite as: arXiv:2607.05734 [cs.IR] (or arXiv:2607.05734v1 [cs.IR] for this version) https://doi.org/10.48550/arXiv.2607.05734 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[IR-9] Retrieving a Set Not Independent Passages: Set-Level Compatibility Learning for Efficient Set Exploration
链接: https://arxiv.org/abs/2607.05712
作者: Mooho Song,Jay-Yoon Lee
类目: Information Retrieval (cs.IR)
备注:
Abstract:Multi-hop question answering and retrieval-augmented reasoning require selecting evidence passages that are jointly useful for answering a query. However, most retrievers still score passages independently or make locally supervised sequential decisions, which can fail when evidence usefulness depends on compatibility among passages. LLM-based set selection can model such interactions, but its computational cost limits practical use. We address this gap by formulating multi-hop retrieval as query-set compatibility scoring and propose a set-level retrieval framework. Our training objective teaches retrievers to rank complete and compatible evidence sets above incomplete, noisy alternatives, making set scoring more robust to variable-length and partially noisy contexts. We instantiate the framework with two complementary set scorers: ParaSet, a lightweight late-interaction scorer that applies self-attention over precomputed bi-encoder embeddings for fast candidate-set exploration, and SetCE, a cross-encoder-based reranker trained with the same set-level objective. Experiments on various multi-hop QA benchmarks show that set-level compatibility learning improves retrieval performance and downstream QA task performance. We further show that the proposed set-level retrievers not only outperform document-level retrievers, but also exhibit complementary retrieval characteristics: combining their outputs yields stronger performance than simply retrieving more passages from a single document-level retriever.
[IR-10] Prompting Beats Fine-Tuning: Generative Expected Value Scoring for Statutory Term Retrieval
链接: https://arxiv.org/abs/2607.05582
作者: Alvin Wang,Jaromir Savelka
类目: Information Retrieval (cs.IR)
备注: Accepted to the ASAIL Workshop at ICAIL 2026
Abstract:Legal concepts in statutes are often expressed using vague terms, and practitioners frequently turn to case law to interpret them. We study the task of ranking case-law sentences by their usefulness for explaining a concept or target statutory term, using an established dataset of 26,959 sentences covering 42 U.S. Code concepts labeled into four explanatory-value categories. We compare two families of methods: (i) supervised fine-tuning of encoder-only models (ModernBERT) and (ii) zero-shot prompting of decoder-only models. We show that across all concepts and standard NDCG cutoffs, ModernBERT largely matches earlier BERT-family baselines. In contrast, prompting decoder-only models achieves the strongest overall effectiveness, with our best system surpassing all previously reported state-of-the-art results on this task.
[IR-11] Narrative World Model: Narratology-Grounded Writer Memory for Long-Form Fiction
链接: https://arxiv.org/abs/2607.05577
作者: Mohammad Saifullah,Thomas Kornmaier,Taaha Kazi,Vasu Sharma,Aditya Sanjiv Kanade,Aanand Kumar Yadav
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注: 23 pages, 4 figures; 9-page main text plus appendix. Preprint
Abstract:Long-form fiction writers need memory that answers multi-hop questions about evolving story state: who knows a secret and when they learned it, whether an event preceded the narration that revealed it, whether a setup paid off, and how a relationship shifted. General-purpose retrieval and agent-memory systems represent entities and facts but not the narratological structure these questions turn on, so they surface the wrong evidence or none at all. We introduce the Narrative World Model (NWM), a writer-memory system that pairs a narratology-grounded typed temporal-state graph with query-conditioned hybrid retrieval. To measure memory rather than the answerer, we read every system through a single held-constant Opus 4.8 reader over only that system’s chapter-safe evidence, on a reproducible public corpus and a validated multi-hop benchmark, and we compare against the strongest existing temporal-knowledge-graph agent-memory framework, Graphiti/Zep (Rasmussen et al., 2025). NWM substantially and significantly outperforms this baseline on multi-hop narratological QA across both corpora, and far exceeds GraphRAG and flat retrieval. The advantage is representational rather than an artifact of extraction: it survives rebuilding the baseline with NWM’s own extractor, and traces to its narratology-grounded structure and query-conditioned retrieval, not to graph size or extractor quality.
[IR-12] Scientific Code Search at Scale: A Multi-Domain Dataset and Benchmark
链接: https://arxiv.org/abs/2607.05443
作者: Nishan Pantha,Pranath Reddy Kumbam,Sajil Awale,Pushwitha Krishnappa,Muthukumaran Ramasubramanian,Nidhi Jha,Emily Foshee,Ankur Kumar,Rachel Slank,Ashkbiz Danehkar,Rahul Ramachandran
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注: Datasets and benchmarks publicly released on HuggingFace. Code released on GitHub
Abstract:Scientists increasingly rely on open-source tools to support their research workflows, yet discovering relevant software among over 600 million GitHub repositories remains challenging. Existing code search benchmarks focus on general software engineering tasks and fail to capture the domain-specific vocabulary and needs of scientific computing. We present a curated corpus of 5,264 high-quality, domain-classified scientific repositories spanning five NASA Science Mission Directorate divisions – Earth Science, Astrophysics, Planetary Science, Heliophysics, and Biological Physical Sciences – enriched with cleaned READMEs, extracted topics, and additional context from crawled links. Building on this corpus, we introduce two novel information retrieval benchmarks: (1) a repository search benchmark with 219 expert-curated queries designed by domain scientists, and (2) a large-scale code snippet retrieval benchmark containing 117,950 code snippets and 119,720 queries across seven programming languages. Baseline evaluations on repository search reveal significant performance variation across scientific domains. Code snippet retrieval proves equally challenging, with substantial variation driven by differing documentation practices, coding standards, and programming language conventions across scientific communities. All datasets and benchmarks are publicly released on HuggingFace to support research on scientific tool discovery.
[IR-13] PORTS: Preference-Optimized Retrievers for Tool Selection with Large Language Models EMNLP
链接: https://arxiv.org/abs/2607.05441
作者: Lorenzo Molfetta,Giacomo Frisoni,Nicolò Monaldini,Gianluca Moro
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI)
备注: Please cite the definitive, peer-reviewed version of this article published in the Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, edited by Christos Christodoulopoulos et al., Association for Computational Linguistics, pp. 10007-10030, 2025. DOI: this https URL
Abstract:Integrating external tools with Large Language Models (LLMs) has emerged as a promising paradigm for accomplishing complex tasks. Since LLMs still struggle to effectively manage large tool collections, researchers have begun exploring retrieval-based methods to pre-select the most relevant options, addressing input length and latency constraints. However, existing retrievers are often misaligned with tool-calling LLMs due to their separate training processes. This paper presents PORTS, a novel odds ratio preference optimization method for training retrievers aimed at tool selection. Using a perplexity-inspired preference signal from a frozen LLM, our approach fine-tunes a retriever to find helpful tools by optimizing the correlation between the selection probabilities and the downstream performances while jointly enforcing a contrastive semantic loss between documentation strings. The versatility of PORTS and its ability to significantly improve tool selection accuracy are demonstrated through extensive experiments on six datasets, two encoder models, and three LLMs with diverse prior knowledge. With low computational demands, our alignment process facilitates generalization to new queries and tools, proving valuable for practical applications with evolving toolsets.
[IR-14] Modality Relevance is not Modality Utility: Post-hoc Selective Modality Escalation for Cost-Aware Multimodal RAG
链接: https://arxiv.org/abs/2607.05438
作者: Xue Li,Yiming Gai
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI)
备注:
Abstract:Multimodal retrieval-augmented generation (RAG) grounds a generator in evidence drawn from heterogeneous modalities – text, tables, and images. The dominant deployment choice is binary and made before the model has tried to answer: either run a cheap text(+table) pipeline, or pay for an expensive vision-language model (VLM) over every image. Recent adaptive systems improve on this by selecting the modality or fidelity pre-retrieval, from a question-conditioned predictor of which modality will be needed. We show that this is the wrong decision point. Through an oracle headroom analysis on MultiModalQA, we find that the relevance of a modality to a question is a weak predictor of whether that modality is actually needed to answer correctly: a large fraction of questions whose gold support includes an image are nonetheless answerable from text and tables alone, and a pre-retrieval router that escalates on apparent visual relevance over-escalates substantially relative to an oracle. We propose \textbfpost-hoc selective modality escalation: answer cheaply from text and tables, run a verifier on the (query, draft answer, evidence) tuple that localizes which modality is missing, and pay for VLM evidence only there. A calibrated value-of-escalation router then decides whether the expected accuracy gain justifies the visual cost. On MultiModalQA, our router recovers the accuracy of an always-on VLM pipeline while issuing far fewer visual calls, and closes most of the gap to the oracle escalation rate. The result extends a routing-signal hierarchy established for retrieval depth and reasoning hops to a third axis – modality – under a single cost-aware selective-escalation view.
[IR-15] Linking Hadith Narrator Identities Across Heterogeneous Arabic Biographical Databases: A Multi-Signal Entity Resolution Pipeline
链接: https://arxiv.org/abs/2607.05424
作者: Taufiq Wirahman
类目: Digital Libraries (cs.DL); Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注: 16 pages, the data sets available at DOI: https://doi.org/10.1016/j.dib.2022.108065
Abstract:The transmission chains (sanad) of Islamic Hadith literature encode relationships among tens of thousands of historical narrators whose biographical records are dispersed across independently maintained digital databases that share no common identifier. We present a two-phase entity resolution pipeline that links narrator names from the Sanadset 650K corpus - 650,986 Hadith records from 926 books containing 185,216 unique narrator name variants - to two biographical databases: Hadithtransmitters (Hawramani; 100,915 entries) and Muslimscholars (25,247 entries). Phase 1 matches Sanadset names to Hawramani using name-only similarity (Sanadset carries no metadata), yielding 94,628 links (51.1%; HIGH 39,938 / MED 54,690). Phase 2 cross-references Hawramani against Muslimscholars via a weighted multi-signal function combining name similarity, death-year proximity, and reliability grade polarity, yielding 95,573 links (94.7% of Hawramani; HIGH 18,245 / MED 71,546 / LOW 5,782). Chaining the two phases gives Sanadset narrators transitive access to Muslimscholars data. The linked data enable construction of a 185,216-node, 814,093-edge directed transmission graph enriched with cross-source biographical metadata. The annotated link corpora and enriched graph are released as open resources.
人机交互
[HC-0] GlassTENG: Self-Powered Triboelectric Nanogenerator based Sensing of Pulse Jaw and Upper Facial Activity from Everyday Glasses
链接: https://arxiv.org/abs/2607.06509
作者: Raj N. Dave,Jovanis Prodanich,Yung-ching Lai,Oscar Jakacki,Stanley Lin,Jack Thoene,Nabil Alshurafa,Nivedita Arora
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Smart glasses maintain near-continuous skin contact at multiple arterial and muscular sites, making them a promising platform for physiological sensing. In practice, though, two factors make sustained daily wear and longitudinal deployment impractical for the quantified self: the discomfort of prolonged sensor-skin contact (e.g., gels and adhesives) and the sensor power demands that increase battery size, weight, and maintenance burden. We present GlassTENG, an ultra-low-power sensor that embeds three custom-fabricated triboelectric nanogenerators (TENGs) into a glasses frame at the angular artery on the nasal bridge, the superficial temporal artery on an extended arm, and the temporalis muscle at the temple. Each GlassTENG sensor is self-powered in transducing mechanical energy to electrical energy and consumes 1.36 \mu W per sensor at the analog front-end. GlassTENG enables simultaneous capture of arterial pulse waveforms, jaw kinematics (e.g., clenching, tapping, eating), and upper facial activity (e.g., blinking, eyebrow movement). In a 20-participant user study, we achieve 93.8% accuracy across six jaw and upper facial activities and estimate heart rate with a mean absolute error of 1.82 beats per minute (BPM) relative to a ground-truth chest-strap sensor in 30s windows. Together, these results establish a future pathway toward a longitudinally worn, ultra-low-power, glasses-based physiological monitoring platform.
[HC-1] he Impact of Security and Privacy Controls on Users Emotional Engagement with Generative AI Chatbots USENIX-SECURITY2026 USENIX-SECURITY
链接: https://arxiv.org/abs/2607.06371
作者: Jabari Kwesi,Jiaxun Cao,Hailee Cunningham,Pardis Emami-Naeini
类目: Human-Computer Interaction (cs.HC)
备注: Accepted at the 35th USENIX Security Symposium (USENIX Security 2026). 20 pages, 7 tables, 2 figures
Abstract:Chatbots powered by generative AI (e.g., OpenAI’s ChatGPT and Google’s Gemini) are increasingly being appropriated for emotional support and companionship. These tools offer a suite of security and privacy (SP) controls, including model training opt-outs and memory toggles, yet how the presence of these controls influences users’ attitudes toward emotionally sensitive disclosure remains understudied. We conducted a mixed-methods vignette study with 354 U.S. participants to examine how SP controls influence users’ willingness to engage with generative AI chatbots for emotional support, their perceptions of how protected they are when using these systems, and their perceptions of how effective the chatbots are for providing support. Controls enabling deletion of disclosures had the largest positive impact: these offerings outperformed technically sophisticated controls such as local-only processing and model training opt-outs, where participants expressed difficulty understanding the underlying mechanisms. Yet trust remains fragile, and participants often doubted SP controls would function as promised. We conclude with actionable recommendations informed by our results to bridge users’ comprehension gaps, build credible assurances, and properly calibrate barriers for users in distress.
[HC-2] Responsible Personalisation: The Double-Edged Sword of Personalisation in Human-Robot Interaction
链接: https://arxiv.org/abs/2607.06344
作者: Antonio Andriella,Jauwairia Nasir,Andrea Rezzani,Alyssa Kubota,Dimitri Lacroix,Tamlin Love,Aniol Civit,Vicky Charisi,Elisabeth Andre,Wing-Yue Geoffrey Louie
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注: 36 pages, 3 figures
Abstract:While personalisation is becoming a defining capability in human-robot interaction (HRI), the existing literature on responsible personalisation remains fragmented, offering isolated accounts of ethical risks without a structured understanding of how they emerge across interaction contexts. This gap is particularly critical in HRI, where robots’ embodiment and social presence can amplify and reshape such risks or generate new types of risks. We present a lifecycle-based and context-sensitive framework for personalised HRI, grounded in an embodiment-aware perspective. The framework combines stages of the personalisation process with interaction characteristics (short-term vs. long-term, open-domain vs. closed-domain), enabling systematic analysis of how risks arise and evolve. Building on this, we conduct an integrative analysis of key ethical risks, including autonomy erosion, biased user modelling, manipulation, dehumanisation, and privacy violations, and examine how they manifest across contexts. We translate these insights into actionable design recommendations and outline open research challenges. By structuring both the design space and risk landscape of personalised HRI, this work provides a foundation for more systematic, transparent, and ethically grounded approaches to personalised robot behaviour.
[HC-3] DS-MTNet:Structured Multi-Task EEG Decoding for Human-Machine Collaboration
链接: https://arxiv.org/abs/2607.06297
作者: Xinjia Yu,Yang Zhou,Jing Yang,Tielin Shi,Tao Cheng
类目: Human-Computer Interaction (cs.HC)
备注: 11 pages, 4 figures
Abstract:Current human-machine collaboration (HMC) systems rely on environment-facing sensors to observe visible actions and scene states, but the internal perceptual, intention-related, and state-related processes of operators remain insufficiently integrated into machine perception. Electroencephalography (EEG) provides a non-invasive, time-resolved modality to capture neural activity associated with these processes and can serve as an additional sensing channel in HMC. However, HMC-relevant EEG evidence is often mixed in continuous recordings. Existing EEG decoding methods usually target task-specific classification or aggregate prediction, so multiple HMC-relevant readouts are rarely organized in a unified EEG representation. To address this gap, this paper proposed the Decomposed-Source Multi-Task Network (DS-MTNet), a structured multi-task EEG decoding framework. DS-MTNet integrated three streams, namely EEG waveforms, task-routed source embeddings, and temporal-spectral power features, into reusable slots and used dual gating mechanisms to route task-specific components. The model was tested on a sustained-attention driving EEG dataset with three representative readouts: lane-departure-related epochs for environmental-event processing, steering-response stage for response preparation, and reaction-time-defined alertness state for internal state. DS-MTNet achieved the best mean performance among traditional, single-task deep, and multi-task EEG baselines, with the most robust gains observed for steering-response stage decoding. Ablation and interpretability analyses suggested that DS-MTNet jointly decoded multiple readouts and organized event-related, response-related, and state-related EEG evidence in a unified source-slot representation. These findings provide a computational step toward incorporating operator-related neural evidence into machine perception in HMC.
[HC-4] AlayaWorld: Long-Horizon and Playable Video World Generation
链接: https://arxiv.org/abs/2607.06291
作者: AlayaWorld Team,Kaipeng Zhang,Chuanhao Li,Yifan Zhan,Yongtao Ge,Yuanyang Yin,Jiaming Tan,Kang He,Liaoyuan Fan,Ruicong Liu,Xiaojie Xu,Xuangeng Chu,Zhen Li,Zhengyuan Lin,Zhixiang Wang,Zian Meng,Zihui Gao
类目: Computer Vision and Pattern Recognition (cs.CV); Human-Computer Interaction (cs.HC)
备注: Authors are listed alphabetically by the first name and their role. See the contribution section for details
Abstract:Game worlds have traditionally been built through labor-intensive production pipelines, making them costly to develop, difficult to customization, and expensive to modify after deployment. Recent advances in video world models offer a fundamentally different paradigm. Rather than explicitly authoring every component of a virtual environment, these models autoregressively synthesize future observations conditioned on the current world state and user interactions, enabling playable worlds to be generated online. Trained on both gameplay recordings and real-world videos, they can capture diverse visual appearances and physical dynamics, opening new opportunities for interactive applications beyond gaming, including embodied intelligence. In this paper, we present \textbfAlayaWorld, a full-stack open-source framework for building interactive generative worlds. AlayaWorld enables open-ended real-time interaction, allowing users to freely navigate and perform diverse actions such as combat, spell casting, and monster summoning. The framework unifies the complete development-from data preparation model architecture, model training, inference acceleration, and deployment-within a modular and extensible architecture. Alongside the framework, we release reproducible pipelines, reference implementations, evaluation tools, and comprehensive documentation, establishing a practical foundation for future research and real-time applications of generative world models.
[HC-5] BlossomPsy: A User-Centric AI System for Adaptive and Engaging MBTI Personality Assessments
链接: https://arxiv.org/abs/2607.06149
作者: Bingjia Huang
类目: Human-Computer Interaction (cs.HC)
备注: 24 pages,20 figures
Abstract:There has been growing public interest in understanding personality traits and emotional characteristics, as such knowledge helps individuals better accept themselves and manage negative emotions. While professional personality scales remain the standard tool for assessment, they are often perceived as tedious or inaccessible to the general public. AI-driven systems can make assessments more accessible, but it is difficult to balance user engagement with predictive consistency in existing works. We tackle this challenge by introducing BlossomPsy, a user-friendly AI-driven MBTI assessment system. MBTI, a widely recognized but psychometrically debated personality framework, serves as the foundation for many recent systems. BlossomPsy integrates multi-turn dialogue and photo-based questions to enhance user engagement while supporting confidence-aware predictions. By combining deep learning, multi-armed bandit algorithms, and control theory, the system dynamically adapts to users’ responses. In particular, photo-based questions are designed to increase interactivity and provide additional user information, thereby improving prediction confidence. Experiments involving both human volunteers and large language models (LLMs) provide preliminary evidence that BlossomPsy can produce stable predictions, with higher reported user satisfaction compared to MBTI-M (Chinese version), while maintaining comparable consistency with the reference scale.
[HC-6] Agents That Teach: Towards Designing Incidental Learning Back into AI-Assisted Software Development
链接: https://arxiv.org/abs/2607.06101
作者: Rohit Mehra,Samdyuti Suri,Prithviraj K Tagadinamani,Kapil Singi,Vikrant Kaulgud,Adam P. Burden
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Human-Computer Interaction (cs.HC)
备注: 5 pages. To be published in the proceedings of 41st International Conference on Automated Software Engineering (ASE '26), October 12-16, 2026, Munich, Germany (New Ideas and Emerging Results Track)
Abstract:AI coding agents are rapidly reshaping how software is built, with developers increasingly delegating substantial coding tasks to autonomous agents in pursuit of higher productivity. While these gains are real, they come at the cost of incidental learning. Developers historically acquired informal knowledge through effortful problem-solving, and this has long shaped how software engineering expertise develops. However, with over-reliance on agentic coding, unpracticed skills could atrophy silently over time. As this learning pathway is short-circuited, developers risk silently accruing Knowledge Debt, a developer-level analogue of Technical Debt, where changes the agent executes that the developer cannot fully understand accrue over time. In this paper, we argue that incidental learning will not re-emerge on its own and must be consciously designed back into developer-agent interactions, and propose six design principles to guide such systems. We then present “SHIELD”, a multi-agent system grounded in the notion of “agents that teach”, that operationalizes these principles by leveraging the AI coding agent’s own reasoning to surface contextual, out-of-band learning moments without disrupting developer flow. Through this work, we envision a path toward learning-aware development environments where productivity and learning are complementary, not competing.
[HC-7] Designing Computerized Gait Analysis for Pediatric Care: Clinician Perspectives on Sensing Workflow and Care Environments
链接: https://arxiv.org/abs/2607.06076
作者: Elizabeth Hong,Andrea Green,Ge Wang,Yiwen Dong
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Computerized gait analysis (CGA) serves as an essential diagnostic tool for evaluating neuromuscular, musculoskeletal, and neurological disorders in children, from cerebral palsy to muscular dystrophy. By enabling objective and comprehensive gait analysis, CGA supports timely clinical interventions that can significantly improve pediatric mobility outcomes and quality of life. Yet pediatric gait analysis introduces unique design considerations often underexplored in existing CGA research, as children’s ongoing development shapes assessment requirements. To understand how CGA technologies can be designed for pediatric care, we conducted a qualitative study with 12 pediatric clinicians and one system designer who routinely work with CGA. Participants identified child-specific challenges including managing heightened sensory sensitivities to wearable devices, accommodating body proportions in sensor placement and calibration, and maintaining patient engagement during data collection. Clinicians also articulated needs for workflow adaptations and expressed interest in extending gait analysis beyond controlled laboratory settings into naturalistic environments such as playgrounds and schools, where children’s authentic movement patterns emerge. Drawing from these clinician perspectives, we present design recommendations for pediatric-centered CGA that address sensing modalities suitable for sensory-sensitive children and approaches for capturing gait data across diverse care environments. Our findings contribute to the broader challenge of adapting clinical technologies to meet the distinct needs of pediatric populations.
[HC-8] Prompt Coach: An Empirical Evaluation of an Agent ic Tutor for Learning Prompt Engineering in Software Development
链接: https://arxiv.org/abs/2607.06074
作者: Rohit Mehra,Kapil Singi,Vikrant Kaulgud,Vibhu Saujanya Sharma,Swapnajeet Gon Choudhury,Swati Sharma,Adam P. Burden,Majd Sakr
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Human-Computer Interaction (cs.HC)
备注: 7 pages. To be published in the proceedings of 41st International Conference on Automated Software Engineering (ASE '26), October 12-16, 2026, Munich, Germany (Industry Showcase Track)
Abstract:Prompt engineering has emerged as a critical yet undertaught skill for software developers, one that traditional learning approaches are ill-equipped to support given its evolving, interactive, and context-dependent nature. In this paper, we introduce Prompt Coach (PC), an agentic tutor that helps developers learn how to craft high-quality code-generation prompts through Socratic guidance embedded in-flow within their IDE. PC evaluates prompt quality across multiple dimensions and surfaces targeted questions to guide self-correction, grounded in the developer’s codebase and the behavior of the target LLM. We present an early empirical study with 15 professional developers combining quantitative prompt quality scoring with qualitative perception measures. Participants showed statistically significant improvements after a single 60-minute session, with the largest gains across dimensions commonly overlooked by developers. They also reported strong trust, high adoption readiness, and unanimous agreement that PC improved their prompt-writing skills.
[HC-9] Nested Episodic State Topology (NEST): A Graph-Theoretic Architecture of Cognitive States
链接: https://arxiv.org/abs/2607.06055
作者: Ishant
类目: Human-Computer Interaction (cs.HC); Computation and Language (cs.CL)
备注:
Abstract:We present NEST (Nested Episodic State Topology), a foundational graph-theoretic representational ontology for modeling cognition as structured state formation and transformation rather than as a finished empirical model. Concepts, episodes, percepts, and task contexts are represented as typed, weighted graphs whose nodes may carry internal subgraph payloads; edges are typed under six relation classes – causal, containment, temporal, associative, evidential, and spatial. Durable belief graphs are separated from capacity-limited working-memory graphs that may host transient non-belief content. WM-belief grounding, conflict catalogs, and belief-update operators specify how transient structure is tested against stored knowledge and how belief is revised. A reusable operator toolkit – activation, graph-property functionals, working-memory transitions, awareness and trajectory functionals, and belief update – organizes the formal core. Derived diagnostics such as fragmentation, involvement, signed evaluation, coherence, and active conflict define familiar phenomena in the same ontology; self-related processing is modeled through designated self-image subgraphs within belief. Subsequent sections instantiate this core without new primitives: phenomena signatures, a task-instantiation schema for action selection and failure modes, and compatibility mappings that embed ACT-R, Soar, Sigma, the Common Model of Cognition, Global Workspace Theory, semantic networks, Theory-Theory, and chunking as constrained regions of one language. Mappings constitute the culminating technical section; discussion addresses scope, limitations, and open research directions. The contribution is intentionally foundational: a transparent representational substrate for later empirical, computational, and domain-specific work.
[HC-10] VisTCP: A Visualization Framework to Construct Knowledge-Graph-Based Representation for Traditional Chinese Painting
链接: https://arxiv.org/abs/2607.05841
作者: Zhiguang Zhou,Fengling Zheng,Miaoxin Hu,Lina You,Jin Wen,Huan Liu,Wei Zhang,Dekun Qian,Yuhua Liu,Wei Chen,Yigang Wang,Yong Wang
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注:
Abstract:Structured representation can characterize semantic objects and relationships in images. It provides a possible effective way for the semantic understanding of Traditional Chinese Paintings (TCPs) to better support archaeology and art history research. However, most image-oriented structured representation methods perform poorly on TCPs, due to two major challenges: 1) the objects and events of TCPs exhibit substantial differences from modern natural images, which results in semantic misunderstandings of TCPs; and 2) it is difficult to achieve accurate identification of ancient objects and events in TCPs, even for domain this http URL this paper, we propose VisTCP, a visualization framework that combines a TCP-oriented intelligent model and expert knowledge, which enables art historians to achieve trustworthy structured representations of TCPs in a human-in-the-loop manner. Firstly, we conduct a pilot study with three domain experts to build a semantic taxonomy of TCPs. Then, expert-annotated data are used to train a TCP-oriented structured representation model, which can automatically extract meaningful objects and their relationships in TCPs. To inform users of the model uncertainty, we design a joint embedding visualization view to show the differences between expert annotations and model predictions. This allows users to refine the structured representation based on their domain knowledge, enabling iterative optimization of the model. Finally, we conduct a case study, a usage scenario, and expert interviews on a real dataset to demonstrate the effectiveness of VisTCP in supporting the structured representation and semantic understanding of TCPs.
[HC-11] PERSONAJUDGE: Simulating Individual Human Preference Judgments with Evaluator-Specific Demonstration Data
链接: https://arxiv.org/abs/2607.05742
作者: Zeyu He,Xuan Qi,Subramanian Chidambaram,Zhichao Xu,Vinayak Arannil,Lydia Chilton,Alex C. Williams
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Large language models increasingly serve as judges in AI evaluation, but current approaches rely on consensus preferences that ignore individual evaluator variation. We propose a novel simulation approach that combines categorical judgments with evaluator-specific auxiliary data–retrospective reasoning traces and interface telemetry–to enable LLM-based simulation of individual evaluators via in-context learning. We conduct a systematic empirical study of this approach using multi-facet data from 32 trained annotators across 4,200 preference judgments in a 4 x 4 x 4 factorial design. Our key findings: (1) The simulation approach achieves up to 9.9 percentage point improvements over the Base Judge; (2) Reasoning traces provide the largest gains with higher collection efforts, while interface telemetry often hurts rather than helps performance despite being cheaper to collect. (3) Simulation difficulty is systematic, predicted by an evaluator’s neutral usage (most clearly on Helpfulness) and divergence from consensus; the neutral-usage tendency–rather than simulatability itself–is the cross-task-stable property (r = 0.728). These results establish both the potential and limits of evaluator-specific auxiliary data for personalized evaluation, offering methodological insights for scaling individual aware AI assessment.
[HC-12] Plainbook: Data Science in Plain Language
链接: https://arxiv.org/abs/2607.05717
作者: Luca de Alfaro,Mathis Aubert,Ranjit Jhala,Eliana Pastor,Elena Baralis
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注: 12 pages
Abstract:Jupyter Notebooks have become widely adopted in data science, as they allow the sharing of reproducible computational analysis. They are, however, accessible only to people who understand computer code. To reach the broader audience of scientists interested in data analysis and computation, but unfamiliar with code, we introduce Plainbook, notebooks centered on natural language rather than code. Plainbook is based on two principles: promote the natural language descriptions, and verify the values. In plainbook, the natural language descriptions are preserved, rather than the resulting code; the code is generated automatically from the cell descriptions. As natural language is read top to bottom, Plainbook adopts a linear execution semantics, in which cells are guaranteed to be executed in the order in which they appear; there is no “hidden state” or out-of-order execution as in Jupyter. To allow users who may not understand code to verify the correctness of the computation, we have built into Plainbook verification mechanisms centered on values and value inspection. These include mechanisms that focus on individual cells, akin to unit tests, as well as global mechanisms. Both the linear execution semantics, and the verification mechanisms, are underpinned by a snapshot kernel that caches execution states and makes execution and verification efficient. Comments: 12 pages Subjects: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Software Engineering (cs.SE) Cite as: arXiv:2607.05717 [cs.HC] (or arXiv:2607.05717v1 [cs.HC] for this version) https://doi.org/10.48550/arXiv.2607.05717 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[HC-13] Depression Symptoms and Relational Patterns in 187k ChatGPT Histories
链接: https://arxiv.org/abs/2607.05685
作者: Neil K. R. Sehgal,Dunigan Folk,Lyle Ungar,Sharath Chandra Guntuku
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models are increasingly used as private, always-available conversational systems, but little is known about how people with depressive symptoms use them. Building on CSCW work on disclosure and peer support, we examine ChatGPT as an emerging informal support infrastructure: private, persistent, responsive, and available outside ordinary hours. We analyze 187,093 ChatGPT conversations from 766 participants who completed the PHQ-8, comparing those below the moderate-symptom threshold (score of 10) with those at or above it. Higher-PHQ participants used ChatGPT more for mental-health, interpersonal, loneliness, self-focused, and support-seeking conversations, with pronounced late-night and recurring month-level patterns. Their language contained more first-person singular pronouns and absolutist terms. They more often engaged ChatGPT in high-disclosure contexts, but professional redirection was not higher. Language-based prediction was modest and insufficient for screening (AUROC 0.591). We argue these histories should not be treated as clinical screening data but as evidence LLMs are increasingly used as informal support infrastructure.
[HC-14] Beyond Accuracy: How Humans Evaluate Legally Correct but Socially Controversial Legal Advice from Machines
链接: https://arxiv.org/abs/2607.05680
作者: Benjamin Minhao Chen,Zhiyu Li
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注:
Abstract:AI systems are increasingly used to provide legal advice, raising questions about whether laypeople accept guidance from algorithms–especially when that advice is legally correct but socially controversial. We report a preregistered survey experiment with 3,348 adults in mainland China examining how people evaluate identical legal advice when it is attributed either to an AI system or to a human lawyer, and when it is accompanied by reasoning or not. Contrary to expectations of algorithm aversion, attribution to an AI system has no net effect on perceived reasonableness. However, mediation analyses reveal opposing psychological pathways underlying this null result. AI-attributed advice is perceived as more objective, which increases perceived reasonableness, but also as less comprehensive and less attentive to special circumstances, which decreases perceived reasonableness. By contrast, providing legal reasoning substantially increases perceived reasonableness regardless of source, largely by enhancing perceptions of objectivity. Qualitative responses corroborate this tension between objectivity and contextual sensitivity in evaluations of legal advice. Together, these findings suggest that public responses to AI legal advisors are shaped not by rigid attitudes toward automation, but by the balancing of competing normative expectations. The results have implications for theories of algorithm aversion and the design of AI recommendation systems in normatively salient domains. Subjects: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC) Cite as: arXiv:2607.05680 [cs.CY] (or arXiv:2607.05680v1 [cs.CY] for this version) https://doi.org/10.48550/arXiv.2607.05680 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[HC-15] From Conversation to Contribution: Characterizing Coding Agent in Open-Source Software
链接: https://arxiv.org/abs/2607.05677
作者: Zihan Fang,Yueke Zhang,Ningzhi Tang,Collin McMillan,Toby Jia-Jun Li,Yu Huang
类目: oftware Engineering (cs.SE); Human-Computer Interaction (cs.HC)
备注: 10 pages, 3 figures
Abstract:AI coding assistants such as GitHub Copilot and Cursor have evolved from code-suggestion tools into conversational collaborators, enabling vibe-coding workflows in which developers guide AI-generated code through natural-language dialogue. Although researchers have increasingly recognized the importance of AI coding agents and begun examining their impact on open-source development, a comprehensive understanding of how developers’ chat-based interactions with AI relate to subsequent open-source development and collaboration remains limited. This hinders efforts to effectively design, evaluate, and govern AI-assisted open-source software development. To address this gap, we collected 13,360 AI conversation sessions comprising 79,172 user messages from 1,356 OSS repositories, linked them to repository development histories, and complemented this analysis with a targeted developer survey. We find heavier AI use in smaller, less mature, and less collaborative repositories. After AI adoption, projects tended to show more active contributors and lower contributor concentration (p .001), although communication remained highly concentrated. Code Writing was the dominant chat purpose, and nearly all AI chat sessions were followed by subsequent commits. We find no broad deterioration in code-quality signals or pull request merging rates. However, developers perceive others’ AI-generated code as harder to maintain than their own (p = .029) and view AI as lowering barriers to OSS contribution. While most developers (68%) are willing to share their chat, concerns remain around appearing incompetent, increasing reviewer burden, and exposing ideas to competitors. These findings provide a large-scale empirical characterization of AI-assisted OSS contribution and offer practical insights for designing and governing responsible vibe-coding practices in open-source development.
[HC-16] Perceived System Predictability: Scale Development and Application
链接: https://arxiv.org/abs/2607.05674
作者: Hendrik Schuff,Heike Adel,Ngoc Thang Vu
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:How predictable users perceive an interactive system to be shapes how they interpret, trust, and rely on it, yet HCI lacks both a precise conceptualization and a validated instrument for this perception. We address this gap by introducing perceived system predictability (PSP) as a user-centered construct grounded in uncertainty theory, distinguishing epistemic, aleatory, and effective predictability. We contribute (i) a theoretical framework that situates PSP relative to adjacent constructs such as trust and understanding, (ii) a 6-item PSP scale, derived from a 60-item pool through expert review and cognitive interviews, and validated in a shape-classifier study ( N=200 ) that supports both a unidimensional and a three-factor hierarchical structure, and (iii) a sentiment-classifier study ( N=200 ) that varies explanations and stochasticity, and relates PSP to the correctness of users’ predictions of system behavior, trust, subjective information processing awareness, and need for cognition. We find that PSP and prediction correctness capture distinct aspects of users’ mental models and that both can diverge: PSP itself predicts correctness, explanations shift PSP but not correctness, and increased stochasticity degrades correctness without lowering PSP. PSP thus goes beyond existing objective and subjective measures and offers a principled foundation for designing transparent and trustworthy interactive systems.
[HC-17] GeoXplain: On-the-Fly Visual Explanations for Weather Foundation Models IEEE-VIS2026
链接: https://arxiv.org/abs/2607.05655
作者: Clemens Walter Koprolin,Leonardo Trentini,Benedikt Soja,Mennatallah El-Assady,Christina Humer
类目: Human-Computer Interaction (cs.HC)
备注: 9 pages, 6 figures. Submitted to VISxClimate at IEEE VIS 2026
Abstract:Weather and climate foundation models produce high-dimensional forecasts whose learned relationships are difficult to inspect with static plots alone. GeoXplain is an interactive Python-based visualization toolkit for exploring geospatial attribution maps across climate variables, atmospheric pressure levels, and forecast time. The toolkit accepts attribution bundles containing attribution grids together with corresponding metadata and renders them in a notebook widget or browser with map and globe modes, linked timelines, pressure-level controls, target annotations, and optional physical-field overlays. We frame GeoXplain as a model-agnostic earth-system visualization toolkit and present the GeoXplain Aurora Adapter as its first computation backend. The adapter computes explanations for the Aurora foundation model, either in a local GPU process, through a GPU listener, or through a SLURM-backed listener, while preserving the same Python call site for analysts. It currently supports gradient saliency, Integrated Gradients, RISE, ViT-CX, multi-frame saliency and Integrated Gradients rollouts, and retrieval of ERA5 overlays. GeoXplain can be installed as a PyPI package with pip install geoxplain. The code is open-source and available at this https URL.
[HC-18] Collective Cognition in Hybrid Groups: A Network Science Synthesis
链接: https://arxiv.org/abs/2607.05593
作者: Babak Hemmatian,Razan Baltaji,Lav R. Varshney
类目: Human-Computer Interaction (cs.HC)
备注: Non-authoritative author’s version of forthcoming chapter in the Springer Nature Handbook of Hybrid Intelligence
Abstract:The growing integration of AI agents into human teams calls for a principled understanding of how collective intelligence emerges in hybrid systems. Recent frameworks clarify how attention, memory, and reasoning differences shape human-AI interaction at the individual and dyadic levels, but a formal account of how these differences scale to group-level dynamics is lacking. Most network science has examined either human-only or multi-agent AI-only systems, leaving open how its findings and parametrizations translate to hybrid groups. This chapter synthesizes network science, collective cognition, and multi-agent systems through the lens of attention, memory, and reasoning. We review how task environments, group topologies, agent-level processes, and incentive structures shape collective outcomes in human-only and AI-only networks, then examine how these results extend to hybrid settings, conceptualizing hybrid networks as heterogeneous human-AI nodes and links with distinct individual and transactive constraints. Our comparative analysis identifies which network effects are robust across agent types and which require revision, and highlights configurations that were peripheral in single-type traditions, such as human gatekeepers of AI sub-networks, but become structurally central in hybrid teams. Integrating a cognitive systems perspective with network science, we clarify how established exploration-exploitation and efficiency-redundancy trade-offs may operate differently in hybrid teams, and conclude with implications for organizational design, governance, and the responsible development of hybrid intelligence systems.
[HC-19] CSTutorBench: Benchmarking Small Language Models as Tutors for Block-Based Programming
链接: https://arxiv.org/abs/2607.05571
作者: H. Chad Lane,Bryson Kageler
类目: Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注:
Abstract:Large language models are increasingly explored as AI tutors, yet deploying them in K-12 settings raises concerns around privacy, cost, and reliance on proprietary models. Small language models (SLMs) offer a promising alternative, but selecting the right model for a specific educational context remains difficult, particularly when the target domain, such as block-based programming, is largely absent from model training data. We introduce CSTutorBench, a benchmark for evaluating language models as CS tutors in VEX VR, a block-based robotics environment. The benchmark comprises 17 scenario-based questions scored against a pedagogical rubric grounded in established tutoring and feedback research, with a human-in-the-loop LLM-as-judge pipeline for evaluation. Preliminary findings across 11 models (4B-120B parameters) reveal that models perform well on surface-level criteria such as vocabulary and tone but struggle with deeper pedagogical behaviors, particularly avoiding answer leakage and engaging with student debugging histories. In our sample, model family and instruction-tuning approach appear to be better predictors of tutoring quality than parameter count alone, though the small number of models limits the strength of this conclusion. A targeted prompt revision grounded in recent educational prompt engineering research improved scores for 10 of 11 models. These results underscore the value of context-specific, pedagogically grounded benchmarks for SLM selection in educational deployment.
[HC-20] AIEDs Unfinished Mission: Centering Agency and Motivation in the Age of Effortless Bypass
链接: https://arxiv.org/abs/2607.05557
作者: H. Chad Lane
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:The widespread availability of general-purpose AI that can perform complex cognitive tasks threatens to undermine education at scale. This effortless bypass dilemma sharpens a challenge AIED has long engaged with but must now confront directly: ensuring learners choose effortful engagement when easier alternatives are available to complete learning tasks. In this paper, I argue that AIED’s longstanding agenda of building more effective intelligent educational tools should continue, but with a renewed emphasis on the urgency of ensuring learners choose to engage authentically. Drawing on established motivational and learning theories, I outline five directions in which AIED can build on its existing strengths: supporting autonomy and agency, building learner resilience to metacognitive threats, designing for interest and relevance, amplifying process-based assessment, and empowering teachers. I then share four envisioned technologies that embody key features of this future and conclude by outlining how AIED must now evolve.
[HC-21] Quaternion-Averag ing-Based Adaptive Complementary Filter for Pedestrian Dead Reckoning With a Foot-Mounted AHRS
链接: https://arxiv.org/abs/2607.05451
作者: Shunsei Yamagishi,Lei Jing
类目: Robotics (cs.RO); Human-Computer Interaction (cs.HC)
备注:
Abstract:Pedestrian Dead Reckoning (PDR) can be applied to indoor navigation systems. GPS suffers from signal degradation due to roofs and high-rise buildings, whereas PDR can estimate positions without being affected by such signal degradation. The accuracy of a foot-mounted AHRS(Attitude and Heading Reference System)-based PDR depends on the accuracy of the attitude estimation algorithm used in the AHRS. In this article, a Quaternion-Averaging-Based Adaptive Complementary Filter (QAACF) for PDR with a foot-mounted AHRS is proposed to improve estimation accuracy while reducing computational cost. QAACF fuses a quaternion derived from angular velocity with quaternions derived from acceleration and magnetic field measurements using Markley’s quaternion averaging, which combines two quaternions more rigorously than linear interpolation. In addition, QAACF adaptively adjusts the weights of angular velocity, acceleration, and magnetic field measurements according to gait phases and the level of magnetic disturbances. Experimental results showed that the proposed QAACF achieves low Root Mean Square Errors (RMSEs) compared to existing attitude estimation filters while requiring lower computational cost than Kalman filters.
计算机视觉
[CV-0] ELSA3D: Elastic Semantic Anchoring for Unified 3D Understanding and Generation
链接: https://arxiv.org/abs/2607.06565
作者: Tianjiao Yu,Xinzhuo Li,Yifan Shen,Onkar Susladkar,Yuanzhe Liu,Xiaona Zhou,Ismini Lourentzou
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Unified 3D foundation models aspire to generate 3D assets and reason about them in language within a single backbone, but their text-3D interaction remains largely implicit. Existing methods concatenate text and 3D tokens into a flat sequence and rely on self-attention, collapsing coarse structural cues and fine geometric details into one undifferentiated representation. We introduce ELSA3D, a unified 3D model that addresses this with elastic semantic anchoring, structuring language and geometric reasoning jointly along matched abstraction scales. ELSA3D represents geometry with a scale-aware octree tokenizer and introduces Anchor Tokens, sparse cross-modal units that select semantic cues, route them to the most relevant 3D scale, retrieve scale-specific geometric evidence, and write the fused signal back into the unified representation, keeping interaction sparse yet precise. A lightweight per-block router makes both computation and reasoning elastic, choosing which text tokens instantiate anchors at which geometric scale so that cross-modal capacity concentrates where alignment is most needed. ELSA3D achieves state-of-the-art performance across image-to-3D generation, text-to-3D generation, and 3D captioning, outperforming the strongest unified baseline while roughly halving FLOPs and inference latency relative to the non-elastic version of the same model.
[CV-1] Lift3D-VLA: Lifting VLA Models to 3D Geometry and Dynamics-Aware Manipulation
链接: https://arxiv.org/abs/2607.06564
作者: Jiaming Liu,Qingpo Wuwu,Nuowei Han,Hao Chen,Zhuoyang Liu,Fan Fei,Yueru Jia,Chenyang Gu,Yandong Guo,Boxin Shi,Shanghang Zhang
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 14 pages, 7 figures. Project website: this https URL
Abstract:Recently, Vision-Language-Action (VLA) models have demonstrated strong generalization across diverse tasks. However, effective robotic manipulation in physical environments fundamentally requires geometric understanding and spatial reasoning. While some VLA approaches attempt to incorporate 3D information, they are constrained by limited data availability and geometric information loss in current 3D encoding pipelines, and fail to jointly capture 3D geometry and temporally structured actions in dynamic environments. To address these limitations, we introduce Lift3D-VLA, a unified VLA framework that equips models with explicit 3D point cloud reasoning and enables temporally coherent action generation. First, building upon our previous work Lift3D, an enhanced 2D model-lifting strategy is proposed to geometrically align 3D points with pretrained 2D positional embeddings. This design enables direct point-cloud encoding within the VLA vision encoder while minimizing spatial information loss. Based on explicit 3D inputs, we propose Geometry-Centric Masked Autoencoding (GC-MAE), a dual-objective self-supervised framework that reconstructs the current point cloud while predicting its future geometric evolution. This formulation allows the 2D vision encoder to internalize both 3D structure and physical dynamics. To fully exploit 3D representations, we further design layer-wise temporal action modeling, which leverages multiple layers of the LLM to collaboratively predict action chunks, enabling temporally consistent predictions. Across 22 simulated tasks and 8 real-world manipulation tasks, Lift3D-VLA achieves 10.8% and 11.1% higher mean success rates on MetaWorld and RLBench than the best-performing prior VLA methods, and outperforms the strongest real-world baseline by 4 percentage points, while exhibiting stronger generalization to out-of-distribution perturbations.
[CV-2] Vision as Unified Multimodal Generation
链接: https://arxiv.org/abs/2607.06560
作者: Xiaoyang Han,Jianhua Li,Kewang Deng,Zukai Chen,Xuanke Shi,Sihan Wang,Boxuan Li,Linyan Wang,Siyi Xie,Xin You,Jinsheng Quan,Zhongang Cai,Haiwen Diao,Ziwei Liu,Lei Yang,Dahua Lin,Quan Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 48 pages,22 figures
Abstract:We formulate computer vision as unified multimodal generation, where heterogeneous visual tasks are expressed in the native text and image generation spaces of a unified multimodal model, without task-specific architectures. Under this formulation, SenseNova-Vision uses natural-language instructions and optional visual prompts to specify tasks, target regions or views, and decoding conventions, and generates responses as text for symbolic outputs, images for dense spatial predictions, or mixed text-and-image outputs for compositional tasks. To support large-scale training, we convert diverse computer vision annotations into instruction-response examples compatible with these generation spaces, resulting in the SenseNova-Vision Corpus, a computer-vision instruction-response corpus spanning text, image, and mixed targets. Starting from an off-the-shelf pretrained unified multimodal model, SenseNova-Vision is trained primarily on this corpus, with auxiliary multimodal data used as a capability-preserving mixture, and requires no task-specific prediction heads or architectural modifications. The resulting model covers a broad range of vision tasks, including detection, OCR, keypoint estimation, segmentation, depth estimation, surface normal prediction, point maps, and camera pose estimation, while supporting language-defined variants that combine category, color, region, and other visual cues. Experiments show that a single unified model can match leading task-specialized systems across structured visual understanding, dense geometric prediction, segmentation, and multi-view visual geometry. These results suggest unified multimodal generation as a scalable route for integrating computer vision capabilities into general-purpose foundation models. The model and corpus are publicly available.
[CV-3] ProxyPose: 6-DoF Pose Tracking via Video-to-Video Translation
链接: https://arxiv.org/abs/2607.06555
作者: Ruihang Zhang,Felix Taubner,Pooja Ravi,Kiriakos N. Kutulakos,David B. Lindell
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 23 pages, 6 figures
Abstract:Tracking the six-degree-of-freedom (6-DoF) pose of objects and surfaces from monocular video is a long-standing problem in computer vision. To tackle this problem, existing methods require inputs beyond the video itself-such as 3D models, depth maps, object masks, or task-specific learned features-and they struggle with textureless, transparent, reflective, or deformable surfaces. Here, we introduce ProxyPose, which recasts 6-DoF pose tracking as video-to-video translation. Given only a video and a single marked pixel in the first frame, a fine-tuned video diffusion model translates the input into a proxy video-a synthetic video depicting a colored polyhedron undergoing the same local rigid-body motion as the surface region at the marked pixel. Because the proxy’s geometry and appearance are known by construction, recovering its full 6-DoF trajectory reduces to classical pose estimation with off-the-shelf solvers. This formulation leverages large-scale video pre-training to absorb the hardest aspects of pose tracking-handling challenging materials, occlusions, and deformations-into the translation step, while operating at the pixel level with no assumptions about object identity, boundaries, or global rigidity. ProxyPose achieves state-of-the-art 6-DoF pose tracking accuracy without the additional inputs required by competing methods and after fine-tuning the video model only on synthetic data. We further demonstrate that ProxyPose extends to face tracking, camera pose estimation, and challenging in-the-wild scenes that are beyond the reach of existing approaches. Project page: this https URL.
[CV-4] From RGB Generation to Dense Field Readout: Pixel-Space Dense Prediction with Text-to-Image Models
链接: https://arxiv.org/abs/2607.06553
作者: Zanyi Wang,Xin Lin,Haodong Li,Dengyang Jiang,Yijiang Li,Pengtao Xie
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Large-scale text-to-image models are attractive backbones for dense prediction because RGB generation pretraining learns rich semantic, structural, and geometric priors. Existing generative and editing approaches reuse these priors by casting dense prediction as target generation: annotations such as depth, normals, alpha mattes, masks, and heatmaps are encoded into an RGB-trained VAE latent space and decoded back as image-like targets. We argue this inherits more of the generative output interface than dense prediction requires: unlike RGB synthesis, dense prediction asks for pixel-correct, task-native fields on the same image plane, not new RGB content to be rendered. Our key observation is that a pretrained DiT already organizes RGB inputs through a patch-to-token-to-patch lattice on the image plane, so each token indexes a fixed output patch whose channels can carry task-native quantities instead of RGB appearance. We instantiate this as ReChannel: we keep the VAE encoder for the DiT’s input distribution but drop the target-side decoder, adapt the frozen DiT with task LoRA, and map each token to its p x p x K_t pixel-space patch through a shared token-local linear head–about 33K parameters, no spatial mixing. Using FLUX-Klein, we evaluate on six dense prediction tasks and over a dozen benchmarks. This minimal interface sets new state-of-the-art on trimap-free matting, KITTI depth, and referring segmentation, and stays competitive on normals, saliency, and pose. In a matched 4B setting it is more accurate and 2.48x faster than an edit-plus-latent-decode counterpart–dense perception can benefit from generative pretraining without inheriting its output interface.
[CV-5] MonoIR-RS: Infrared Remote Sensing Vision-Language Learning with CLIP and VLM Adaptation
链接: https://arxiv.org/abs/2607.06552
作者: Jiaju Han,Ma Yaqi,Yahui Chai,Xuemeng Sun,Xin Li,Qike Zhang,Yingying Zhao,Xiang Chen,Luwei Yang,Chengyin Hu,Jiahuan Long
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Infrared remote-sensing imagery captures intensity structure, object-background contrast, and illumination-invariant cues often invisible in RGB imagery. Yet, most remote-sensing vision-language resources and models focus on visible-band semantics, leaving infrared vision-language understanding underexplored. We introduce MonoIR-RS, a large-scale infrared remote-sensing vision-language dataset and benchmark that couples IR-aware data construction with CLIP-style contrastive adaptation and VLM instruction tuning. Built from the same source pool and split as FusionRS, MonoIR-RS retains the infrared image as the model-facing modality, yielding 600,000 synthesized infrared images and 59,032 retained IR-aware caption records. The model experiments use this retained language-supervision subset, whose captions rewrite supervision around grayscale structure and infrared-style contrast instead of RGB appearance. We show that the synthesized infrared imagery is markedly closer to real thermal imagery than a grayscale conversion on the AVIID benchmark. We fine-tune five CLIP backbones and six VLM backbones, and calibrate them against zero-shot behavior: IR-aware adaptation lifts CLIP mean recall by up to 12.8 points and drives VLM captioning IR-cue coverage to 100% while reducing residual RGB-color leakage to near zero. By isolating the infrared modality from RGB-IR dual-modal learning, MonoIR-RS offers a controlled, reproducible testbed for aligning infrared remote-sensing evidence with language.
[CV-6] Unsupervised Domain Adaptation for Calcification Classification in Mammography Across Multi-Site Datasets
链接: https://arxiv.org/abs/2607.06549
作者: Xuan Liu,Derek L. Nguyen,Emily C. Barre,Jennifer Thomas,Thomas Lynch,Jeffrey R. Marks,E. Shelley Hwang,Marc D. Ryser,Joseph Y. Lo,Lars J. Grimm
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Deep learning-based computer-aided diagnosis (CAD) systems have shown strong performance in breast cancer diagnosis, particularly for classification tasks in mammography. However, domain shifts across multi-site datasets remain a challenge, especially when models are applied to unseen domains. In this work, we proposed a calcification classification framework to improve malignant versus benign breast disease classification across multi-site mammography datasets. The framework consisted of two components: (1) an unsupervised domain adaptation module based on style transfer models (AdaIN and CycleGAN) to generate vendor-specific and technique-specific training samples without additional annotations, and (2) a supervised classification module using Swin Transformer V2 as the backbone. We evaluated the proposed method on three datasets: cross-validation on OPTIMAM (National Health Service, United Kingdom; n=2994), followed by external validation on EMBED (Emory University; n=125), and Duke Calcification Dataset v1 (n=788). These datasets cover multiple vendors and include both full-field digital mammography and synthetic 2D images derived from digital breast tomosynthesis. The proposed framework improved cross-site performance for both EMBED (AUC 0.68 to 0.72) and the Duke Calcification Dataset (AUC 0.68 to 0.73). These findings indicate that domain adaptation can reduce domain shifts and improve the generalization for calcification classification across multi-site datasets.
[CV-7] CAIRN: Cross-Room 3D Scene Understanding with Topology-Aware Large Multimodal Models
链接: https://arxiv.org/abs/2607.06534
作者: He Liang,Chenyang Ma,Yiming Zhang,Sangyun Shin,Andrew Markham,Niki Trigoni,Yuhang He
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project Page: this https URL
Abstract:Existing 3D scene-grounded Large Language Models (3D-LLMs) focus on answering questions grounded in simplified single-room 3D scenes, lacking the ability to reason over real-world household environments containing multiple interconnected rooms and diverse object categories. We introduce CAIRN, a topology-aware 3D-LLM for multi-room 3D scene understanding. CAIRN aligns transformer attention with scene hierarchy, giving the model explicit awareness of object-level relations and room-level connectivity. It enriches object tokens with room-local relational context via a graph neural network, introduces learned room tokens for room-level abstraction, and applies a hierarchical attention mask with geometric bias to route information according to scene topology. CAIRN is developed on CAIRN-MR, a benchmark we introduce on HM3D for multi-room 3D scene understanding, covering grounding, captioning, and four question-answering tasks that progressively evaluate from intra-room perception to cross-room reasoning. Experiments show that CAIRN outperforms prior 3D-LLMs by a large margin across all CAIRN-MR tasks while remaining competitive on five single-room benchmarks.
[CV-8] Bridging Physical Reasoning and Task Generalization via Visual Action Outcome Reasoning Alignment ICML’26
链接: https://arxiv.org/abs/2607.06522
作者: Han-Jun Ko,Jr-Jen Chen,Haobo Yuan,Hsin-Ying Lee,Tiancheng Shen,Ming-Hsuan Yang,Yu-Chiang Frank Wang
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: ICML’26 Workshop RLxF: Reinforcement Learning from World Feedback
Abstract:Vision-language models (VLMs) struggle to generalize in interactive physical reasoning, particularly under unseen tasks and environments. Two key failure modes are prominent: hallucinated chain-of-thought (CoT) reasoning that contradicts physical reality, and misalignment between the model’s reasoning and actions. We present VAORA (Visual Action Outcome Reasoning Alignment), a novel reward design that directly addresses both issues. VAORA introduces two complementary rewards: Visual Alignment Reward, which anchors VLM reasoning to the visual context independent of the agent action itself, and Visual-Action Alignment Reward, which grounds reasoning in the visual outcome induced by the model’s action. Together, these rewards suppress hallucinated CoT and reduce the gap between reasoning and behavior. To improve training stability, we further employ smooth, dense rewards by estimating success probabilities using a pre-trained in-domain expert agent. Experiments on PHYRE and Virtual Tool support our performances across novel-task and unseen-environment settings, confirming that grounded and generalizable physical intelligence can be induced through VAORA.
[CV-9] Point as Skeleton: Accumulated Point Cloud Enhanced Autoregressive Generation for Closed-Loop Autonomous Driving Simulation
链接: https://arxiv.org/abs/2607.06516
作者: Songbur Wong,Xiaosong Jia,Junqi You,Bo Zhang,Pei Xu,Renqiu Xia,Yuping Qiu,Shaofeng Zhang,Zelin Zhao,Xuechao Yan,Yuchen Zhou,Yurui Chen,Wen Guo,Hang Xu,Junchi Yan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Evaluating end-to-end autonomous driving (E2E-AD) remains challenging, as existing driving simulation methods often trade off closed-loop interactivity (e.g., CARLA) and real-world visual fidelity (e.g., nuScenes). We present \textbf\emphPoint as Skeleton, a generative sensor simulation framework for state-updated autoregressive driving video generation, in which an autoregressive generator synthesizes visual observations from step-wise updated ego states, actor states, scene maps, and point-cloud skeleton conditions. To support closed-loop rollout, we introduce Reset-and-Roll, which adapts rolling diffusion inference to simulation by preventing future-conditioned latent states from being committed across simulation steps. To stabilize error accumulation during step-wise autoregressive rollout, we introduce point-cloud skeletons that decouple foreground and background assets and project them into camera-view painted-point and template-depth conditions, providing appearance and geometric cues. We further implement a nuPlan-based renderer-level closed-loop generative interface for evaluating generation under ego deviations from the original log. Experiments on nuScenes and nuPlan show that \textitPoint as Skeleton improves autoregressive generation quality during closed-loop rollout, demonstrating its potential for visually faithful closed-loop driving simulation. The code is available at this https URL.
[CV-10] AirflowAttack: Thermal-Airflow Adversarial Perturbations against Infrared Remote-Sensing Vision-Language Models
链接: https://arxiv.org/abs/2607.06485
作者: Cong Su,Jiaju Han,Xuemeng Sun,Chengyin Hu,Qike Zhang,Jiujiang Guo,Yiwei Wei,Jiahuan Long
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Vision-language models (VLMs) are increasingly deployed on infrared (IR) remote sensing imagery in security-critical settings, yet their adversarial robustness remains unexamined. We present AirflowAttack, to our knowledge the first adversarial attack for IR remote-sensing VLMs and the first to weaponize thermal-airflow turbulence as the perturbation prior. A lightweight generator synthesizes a single input-agnostic perturbation regularized toward physically plausible airflow patterns. Optimized on one surrogate CLIP model, it attains a mean zero-shot scene-classification attack success rate (ASR, the fraction of samples whose top-1 class changes) of 48.5% across five diverse CLIP backbones, far exceeding four IR-specific physical baselines (27.7–37.0%). Applied to six state-of-the-art VLMs, it cuts scene-classification accuracy by up to 38.2% relative, yet paradoxically makes some models more confident in their IR analysis, confabulating the perturbation as genuine thermal evidence such as temperature gradients and convection. Ablations show the airflow prior raises physical plausibility at no measurable cost to attack success. Together with a benchmark spanning eleven models and four tasks, these findings expose critical vulnerabilities in the rapidly expanding IR VLM ecosystem.
[CV-11] Assessing the Operational Impact of Poisoning Attacks over Augmented 3D Point Cloud Public Datasets for Connected and Autonomous Vehicles
链接: https://arxiv.org/abs/2607.06484
作者: Marwan Lazrag,Badis Hammi,Lorena Gonzalez-Manzano,Joaquin Garcia-Alfaro
类目: Cryptography and Security (cs.CR); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Accepted for presentation at SECRYPT 2026
Abstract:Poisoning attacks against public datasets lead to major concerns, such as (i) misclassification of perceived objects when the poisoned data is used for training and (ii) embedding of backdoors that may eventually be triggered later on, when specific conditions in the system apply over the learned models. Its impact over data augmentation models is unclear. While data augmentation reduces the likelihood of poisoning attack success, some valid questions remain. Is data augmentation affecting the impact of poisoning attacks? can it increase the number of poisoned samples or injected backdoors? We explore in this paper some of these questions. We assess the effects of augmenting poisoned 3D point cloud datasets and validate that poisoning is able to evade the sanitizing nature of augmentation techniques when using the concrete case of Generative Adversarial Network (GAN) techniques to exemplify the case of data augmentation processing. We also validate that poisoning propagates over the augmented datasets and perturbs the decision made by general-purpose classifiers, in the end. All the experimental material (including tools, datasets, and classifiers) is publicly available, to facilitate reproducibility and to foster further research in the topic.
[CV-12] Mitigating Domain Shift in Conditioned Floor Plan Generation: Synthetic Pre-training for Data-Efficient Adaptation
链接: https://arxiv.org/abs/2607.06483
作者: Matthieu Ospici,Arnaud Gueze,Luc Bourrat,Adrien Bernhardt
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Robustness to domain shift is a key requirement for floor plan generative models to be applicable beyond the single dataset they were trained on, as floor plans vary widely across regions due to distinct architectural cultures, spatial constraints, and construction practices, while acquiring new annotated datasets remains costly and domain-specific. Yet, no prior work has studied this robustness in the context of conditioned floor plan generation. In this paper, we evaluate state-of-the-art models from two fundamentally different generative paradigms across three public datasets (RPLAN, MagicPlan and Swiss Dwellings) and show that they are highly sensitive to domain shift, with up to an order of magnitude performance degradation when transferred across domains. To mitigate this with minimal target-domain supervision, we introduce a procedural method to generate a large-scale synthetic training dataset that enforces strict physical constraints (non-overlapping rooms, valid door placement, graph consistency) while intentionally sacrificing architectural realism through highly irregular spatial arrangements and aggressive geometric perturbation of room shapes. We show that pre-training on this synthetic data considerably improves zero-shot cross-domain performance, outperforming in-domain training on MagicPlan. Furthermore, it provides a highly effective initialization for fine-tuning, accelerating target domain adaptation and outperforming real-world initialization baselines by up to 40% in a low-data regime.
[CV-13] Prompt-Adapter Context Routing for Parameter-Efficient Multi-Shot Long Video Extrapolation
链接: https://arxiv.org/abs/2607.06481
作者: Anna Córdoba,Adam Puente Tercero,Nerea Angulo Hijo,Mar Linares Tercero,Julia Barrientos,Ainhoa Miranda,Jesús Olivera
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 10 pages, 2 figures
Abstract:We present PACR-Video, a parameter-efficient framework for multi-shot long video extrapolation that preserves recurring entities, scene structure, visual style, and causal progression without full generator fine-tuning. PACR-Video keeps a text-to-video diffusion transformer frozen and augments it with low-rank temporal adapters conditioned by learned shot-role prompt tokens. To maintain long-horizon coherence, it builds a recursive prompt bank that stores compact entity, location, action, and style prompts from previous shots, then routes them through adapter gates according to predicted narrative dependencies. A Shot-Local/Story-Global tuning objective combines next-shot reconstruction, cross-shot identity contrast, and prompt sparsity regularization, while an adapter composition schedule balances early-shot visual consistency with later-shot event progression and viewpoint change. Across six multi-shot and long-video benchmarks, PACR-Video outperforms text-to-video, tuning-based, memory-augmented, streaming, and recursive-context baselines on distributional quality, semantic alignment, identity consistency, temporal smoothness, motion stability, transition coherence, and human preference. These results show that compact prompt routing and lightweight temporal adaptation provide sufficient controllable capacity for stable long video extrapolation.
[CV-14] A VLM-Enhanced Framework for Comprehensive Traffic Sign Condition Assessment Integrating Daytime Visual Performance and Nighttime Retroreflectivity Evaluation
链接: https://arxiv.org/abs/2607.06478
作者: Linlin Zhang,Neema Jakisa Owor,Xiang Yu,Abby Watts,Yaw Adu-Gyamfi
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 21 pages, 7 figures, 5 tables. Preprint. An earlier version of this work was presented at the 105th Annual Meeting of the Transportation Research Board (TRB), January 2026
Abstract:Traffic signs are crucial components of road safety, serving as visual tools under all lighting conditions. The Manual on Uniform Traffic Control Devices (MUTCD) specifies daytime visual factors such as legibility and color contrast, and nighttime retroreflectivity requirements. Traditional assessment methods rely on manual inspections, which the Federal Highway Administration (FHWA) notes are subjective, labor-intensive and pose safety concerns, while retroreflectometers are expensive and unaffordable for smaller agencies. Most existing studies focus on either daytime factors or nighttime retroreflectivity but rarely integrate both aspects comprehensively. This study develops a novel framework that systematically evaluates traffic signs through integrated daytime-nighttime assessment. The methodology employs three fine-tuned Vision Language Models (VLMs) for daytime visual performance assessment across four key factors: legibility, color, surface and shape integrity, and surrounding environment conditions. VLM predictions are converted to numerical scores through sentiment analysis and Contrastive Language-Image Pre-Training (CLIP) scoring, while nighttime performance is assessed using LiDAR-derived retroreflectivity following established calibration procedures. The framework integrates these components into a comprehensive Sign Condition Index (SCI) for maintenance guidance. Evaluation results demonstrated that LLaVA and Qwen outperformed InternVL, achieving bidirectional cosine similarity scores of 0.67-0.76 across all factors. Among 462 validated traffic signs, 68 were flagged by the proposed framework as requiring immediate replacement due to inadequate retroreflectivity performance. This research provides a cost-effective alternative to traditional manual inspections for comprehensive traffic sign condition assessment.
[CV-15] EgoPolice: A Benchmark for Egocentric Video Understanding in High-Stakes Police Body-Worn Camera Footage
链接: https://arxiv.org/abs/2607.06468
作者: Max Gonzalez Saez-Diez,Jihoon Chung,Adam D. Wolsky,Gregory Lanzalotto,Dean Knox,Jonathan Mummolo,Brandon M. Stewart,Olga Russakovsky
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:We introduce EgoPolice, a carefully curated dataset of real, egocentric police-civilian interactions, sourced from publicly available body-worn camera videos. We select police-civilian action labels that are critical for police behavioral research and annotate them at a second-by-second granularity. The videos feature rapid and irregular camera motion, dense human interactions, and rare high-stakes events, making the dataset a challenging benchmark for motion-robust and context-aware egocentric perception. We provide two different tasks, classification and multiple-choice question-answering, and benchmark both open-source and closed-source models. We find that even the best video models like Gemini 2.5 Pro still struggle to accurately predict high-risk actions such as “Weapon Out”. Beyond serving as a benchmark, EgoPolice provides a foundation for developing models capable of identifying events of interest in large-scale body-worn camera video repositories, enabling more efficient downstream human review.
[CV-16] Verification of Dynamic Holographic Behavior in Identity Documents ICDAR2025
链接: https://arxiv.org/abs/2607.06466
作者: Glen Pouliquen(1 and 2),Joseph Chazalon(2),Guillaume Chiron(1),Thierry Géraud(2),Ahmad Montaser Awal(1) ((1) IDnow Research Center, France, (2) EPITA Research Lab (LRE), France)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at the International Conference on Document Analysis and Recognition (ICDAR 2025)
Abstract:This paper addresses the remote verification of the authenticity of Optically Variable Devices (commonly known as holograms) on identity documents. Typically placed over the cardholder’s photo, these devices provide strong and easily verifiable security for human inspection but pose challenges for automated verification. Existing approaches easily cover static frauds (e.g. paper photocopy) and can be evaluated for such, but their capacity to detect real, dynamic fraud cases (e.g. handcrafted hologram) has not been evaluated to date because of the lack of public datasets. Furthermore, they are usually trained to detect known attack types, and few of them can generalize to new, unseen attacks. This work features three contributions to address these limitations: 1) a new public dataset, MIDV-DynAttack, which extends the existing MIDV-Holo dataset with realistic, static and dynamic attacks against identity document specimens, tripling the number of attack samples compared to the original dataset, 2) a novel verification method which can assess the authenticity of a specific hologram thanks to the analysis of its dynamic behavior and appearance, can be trained without dynamic attack samples, and exhibits new state-of-the-art performance, 3) a benchmark of existing approaches which follows a clear evaluation protocol and emphasizes the inability of other approaches to deal with dynamic attacks, as well as new challenging attacks to deal with. Code and dataset are publicly available at this https URL.
[CV-17] Andha-Dhun: A First Look at Audio Descriptions in Hindi CVPR
链接: https://arxiv.org/abs/2607.06457
作者: Ritabrata Chakraborty,Divy Kala,Nisheeth Bhooshan Gupta,Ganji Sreeram,Pailla Balakrishna Reddy,Makarand Tapaswi
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to NCVPRIPG 2026, Download data at this https URL
Abstract:Audio Descriptions (ADs) narrate visual content for Blind and Low Vision (BLV) audiences during gaps in audiovisual media. There is growing momentum around ADs in movies and TV shows, and with mandates from India’s Central Board of Film Certification (CBFC), there is a need to expand ADs beyond English. Yet, there is no work that generates ADs for any Indian language. To address this gap, we present the first systematic study of ADs in Hindi, contributing to aspects such as data, generation, and evaluation. We introduce Andha-Dhun, the first dataset of human-authored Hindi ADs collected from 8 full-length movies. We explore two approaches for generating ADs in Hindi: (i) directly from English dense video descriptions, and (ii) translating English ADs into Hindi. We evaluate these approaches using perplexity and LLM-as-a-judge metrics to assess fluency and quality respectively. We also analyze movies that have both English and Hindi human-authored ADs and find that naive translation introduces artifacts and narrows diversity compared to original Hindi ADs. Direct machine translation fails to adapt cultural references, while human-translated ADs do better but still fall short. Our findings emphasize that the purpose of Hindi ADs is accessibility for Indian BLV audiences, and that this requires adapting content for the audience more than strict fidelity to the source.
[CV-18] Analysis-by-Proxy: Localization Signals in VLMs Operating as Condition Encoders ICML2026
链接: https://arxiv.org/abs/2607.06445
作者: Yoav Baron,Sara Dorfman,Roni Paiss,Daniel Cohen-Or,Or Patashnik
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted as a Spotlight at the ICML 2026 Mechanistic Interpretability Workshop
Abstract:Vision-Language Models (VLMs) are increasingly utilized as the conditioning backbone for diffusion-based image editing due to their remarkable multimodal reasoning capabilities. While standalone VLMs demonstrate strong localization capabilities, editing pipelines frequently struggle to maintain this accuracy, particularly in complex, multi-entity scenes. In this work, we investigate this performance gap, hypothesizing that it stems from treating the VLM as a condition encoder. In this role, the model is restricted to a single forward pass, preventing the autoregressive generation process for which it was optimized, thereby failing to fully expose its capabilities. To investigate whether this spatial understanding persists when the VLM is used as a condition encoder, we introduce Analysis-by-Proxy. In this framework, we train a lightweight, interpretable proxy model on the VLM’s intermediate representations using an auxiliary localization task. By analyzing the VLM through this proxy, we uncover the specific VLM representations that encode localization information. Our findings expose a fundamental mismatch between how spatial knowledge is represented within a VLM condition encoder and how it is extracted by current editing pipelines. We reveal that under single-pass constraints, the localization signal does not reliably propagate to the predefined layer configurations commonly used for conditioning. Instead, this crucial signal remains hidden within intermediate representations, at locations that vary depending on the input prompt. Using our introduced Analysis-by-Proxy framework, we reveal the fundamental failures of existing condition extraction strategies in editing pipelines, opening the door to more principled design of conditioning architectures.
[CV-19] PIPBench: A Profile-Inclusive Framework for Personalized Image Generation Evaluation
链接: https://arxiv.org/abs/2607.06440
作者: Yuhang Wu,Shuxiang Zhang,Wee Hian Ching,Chi Zhang,Miao Liu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Recent text-to-image models such as DALLE-3 excel at following diverse prompts yet remain blind to individual aesthetic preferences. We study personalized image generation, where models must align outputs with a user’s implicit visual preferences based on a few historically preferred images and a short prompt. To this end, we introduce PIPBench, the first profile-inclusive benchmark for evaluating personalized image generation. We further propose a novel data construction pipeline that leverages psychological and demographic profiling dimensions for both real-user data collection and scalable agent-based data generation. Using PIPBench, we conduct a thorough evaluation of representative line of methods. Our experiments reveal key limitations in existing methods, suggesting new challenges and opportunities for personalized text-to-image synthesis. Project page: this https URL
[CV-20] WristMimic: Full-Body Humanoid Control with Wrist-Guided Manipulation ECCV2026
链接: https://arxiv.org/abs/2607.06438
作者: Wongyun Yu,Youngwoon Kim,Minsu Cho
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR)
备注: Accepted to ECCV 2026
Abstract:Retargeting human object interaction demonstrations to physics based simulation requires reproducing not only body motion but also the object motion and contacts that make manipulation succeed. However, position only hand trajectories do not specify the contact forces needed to manipulate objects, and directly tracking them can overconstrain contact rich finger behavior. We introduce WristMimic, a wrist guided whole body control framework that explicitly separates contact free body motion from contact rich hand manipulation. The contact free body and wrist are guided by kinematic pose targets, whereas the fingers are not directly supervised by human hand pose. Instead, they learn grasping and manipulation behaviors from object tracking and contact outcomes. Our key insight is that the wrist is the natural gate between these two regimes. It is largely free from contact and can be tracked kinematically, yet it determines the global hand configuration and places the fingers within reachable grasp affordances. To ensure reliable wrist placement during interaction, we introduce wrist specific reset constraints and reward prioritization. Experiments show that WristMimic matches or surpasses methods using full finger pose supervision while enabling finger agnostic retargeting across diverse hand embodiments.
[CV-21] ILDE: TILt-based Distributional Erasure for Concept Unlearning
链接: https://arxiv.org/abs/2607.06432
作者: Naveen George,Naoki Murata,Yuhta Takida,Konda Reddy Mopuri,Yuki Mitsufuji
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Concept unlearning in text-to-image diffusion models is critical for safe and practical deployment: with rising privacy concerns, copyright disputes, trademark constraints, and safety regulations, deployed systems must be able to suppress unwanted concepts after training. Existing methods often remove the target concept effectively, but practical unlearning also requires an equally fundamental property: the unlearned model should retain quality, diversity, and semantic coverage on benign generation. The gold standard is a retain-only model trained from scratch without the unwanted data. However, common erasure objectives do not specify which post-unlearning distribution should approximate this reference, leaving retention as an implicit consequence of the update rule. We propose TILDE, TILt-based Distributional Erasure, which formulates concept unlearning as a distributional alignment problem: the desired target is the minimum-deviation conditional distribution from the pretrained model under a forgetting constraint. This energy-tilted, anchor-free target suppresses concept-expressing images while preserving benign relative mass for each prompt. We instantiate this principle with residual \nabla -GFlowNet training, which learns the score correction induced by the forget energy relative to the pretrained diffusion model. Across objects, artistic styles, and characters, TILDE achieves strong forgetting while improving retention and distributional fidelity over prior baselines.
[CV-22] XRFormer: Multiscale Tokenization for XRF Representation Learning
链接: https://arxiv.org/abs/2607.06424
作者: Sofiane Daimellah,Sylvie Le Hégarat-Mascle,Clotilde Boust
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: International Conference on Pattern Recognition, 2026
Abstract:X-ray fluorescence (XRF) spectroscopy is a key modality for material analysis in cultural heritage. However, automated learning from XRF spectra remains challenging: XRF spectra are complex one-dimensional signals composed of sharp elemental peaks, broader structures, and background variations that are not taken into account by existing learning-based models. This paper introduces XRFormer, a transformer architecture tailored to XRF spectra through a multiscale convolutional tokenizer that injects locality and multi-resolution inductive biases before global self-attention. The tokenizer progressively reduces spectral resolution while increasing embedding dimensionality, and the resulting token sequence is processed by a standard transformer encoder. We further investigate self-supervised pretraining for XRF representation learning using Masked Spectral modeling (MSM) and a physics-informed Peak Presence Prediction (PPP) objective. Experiments on the Pigments Checker STANDARD v.5 dataset for pigment identification and unmixing show that XRFormer consistently outperforms ViT, SpectralFormer (with and without CAF), and a 1D-CNN baseline for pigment identification. For pigment unmixing, XRFormer achieves robust abundance estimation while maintaining significantly higher parameter efficiency than SpectralFormer, operating at a lower token resolution (128 vs. 512 tokens) and with less than half the number of parameters (1.5M vs. 3.37M). MSM yields consistent gains across both tasks, while PPP further enhances performance for both identification and unmixing when tuned with an appropriate peak prominence. These results highlight multiscale, modality-aware tokenization as an effective and parameter efficient foundation for transformer-based XRF modeling under data-limited conditions. A GitHub repository is provided at this https URL.
[CV-23] HoloCount: A Holistic Visual Counting Benchmark for MLLM s
链接: https://arxiv.org/abs/2607.06420
作者: Jinhong Deng,Limeng Qiao,Guanglu Wan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Technical report
Abstract:Visual counting is a fundamental pillar of multimodal intelligence, requiring a seamless integration of fine-grained grounding and spatial reasoning. While Multimodal Large Language Models (MLLMs) have achieved remarkable success in qualitative scene understanding, their quantitative precision remains a significant bottleneck, often characterized by persistent numerical hallucinations. Existing counting benchmarks primarily focus on basic perception in simplified contexts, failing to capture the complex failure modes that emerge under logical constraints or adversarial conditions. To address these limitations, we introduce HoloCount, a holistic and diagnostically rich benchmark structured around a three-level hierarchical taxonomy. HoloCount evaluates MLLMs across: (1) Semantic Counting, focusing on atomic and property-based enumeration; (2) Analytical Counting, assessing logical composition through spatial and set-based reasoning; and (3) Robustness Testing, probing model integrity against adverse scenarios and grounded counter-priors, such as high-density scenes and linguistic biases. Through an exhaustive evaluation of over 20 state-of-the-art MLLMs, we reveal a critical performance gap: even top-tier models degrade significantly as tasks transition from perception to complex analytical reasoning and adverse scenarios. Our findings provide a systematic landscape of current MLLM counting capabilities and offer a roadmap for developing more grounded and reliable multimodal systems. The dataset is available at this https URL.
[CV-24] mporal Modeling of Optically Variable Devices in Identity Documents ICDAR2026
链接: https://arxiv.org/abs/2607.06408
作者: Glen Pouliquen(1 and 2),Joseph Chazalon(2),Guillaume Chiron(1),Oriol Ramos Terrades(3),Thierry Géraud(2),Ahmad Montaser Awal(1) ((1) IDnow Research Center, France, (2) EPITA Research Lab (LRE), France, (3) Computer Vision Center (CVC), Spain)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at the International Conference on Document Analysis and Recognition (ICDAR 2026)
Abstract:Robust remote verification of identity documents relies on analyzing faint, transparent security features like Optically Variable Devices (OVDs), or “holograms”, within user-captured videos under uncontrolled conditions. Current systems, however, face critical limitations: existing methods often treat video frames in isolation, neglecting the intrinsic dynamic nature of OVDs and leaving systems vulnerable to swapping attacks, or focus on general holographic presence and lack the ability to verify specific OVD types. Moreover, the economic infeasibility of frame-by-frame video annotation makes supervised training impractical. In this work, we introduce two novel approaches for verifying the dynamic behavior of transparent OVDs protecting the holder’s portrait, specifically designed for open-set scenarios where attack types are unknown during training. We demonstrate that these approaches can be trained without any attack samples in a self-supervised setting, surpassing previous state-of-the-art methods on public datasets while adhering strictly to industrial constraints. Our results confirm that modeling temporal dynamics is essential for defeating sophisticated attacks under realistic conditions, and underscores the promise of sequence modeling and anomaly detection for OVD verification. Code is available at this https URL.
[CV-25] What Images Cannot Say: Language-Guided Olfactory Representation Learning ECCV2026 WWW
链接: https://arxiv.org/abs/2607.06402
作者: Eleftherios Tsonis,Xi Wang,Vicky Kalogeiton
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: ECCV 2026. Project page: this https URL
Abstract:Images tell us what a scene looks like, but rarely what it would feel like to be there. While recent datasets pair visual scenes with electronic-nose measurements, aligning smell signals with images remains challenging because many olfactory cues arise from contextual environmental factors that are not directly visible in pixels. We introduce SCENT, a multimodal framework that uses language guidance as a semantic bridge between vision and olfaction. Our approach leverages Vision-Language Models (VLMs) to generate scene descriptors capturing objects, environmental context, and plausible ambient smell cues suggested by the visual scene. These descriptors provide semantic guidance for learning olfactory representations. We train a smell encoder that maps electronic-nose signals into a shared embedding space aligned with both visual and textual representations, and introduce a languageguided latent decomposition that separates object-specific odors from contextual environmental contributions. Experiments on the New York Smells dataset demonstrate that SCENT significantly improves crossmodal retrieval compared to vision-only baselines, achieving state-of-theart performance on smell-to-image and smell-to-text retrieval tasks. In addition, our framework produces interpretable olfactory representations that enable the disentanglement of complex smell mixtures. Our results reveal the importance of contextual semantic information for grounding olfactory perception in multimodal learning and pave the way for future research in this area.
[CV-26] FADRA: Frequency-Aware Diffusion with Residual Adaptation for Video Face Restoration
链接: https://arxiv.org/abs/2607.06389
作者: Jin Jiang,Jia Wang,Panwen Hu,Weiran Zhao,Shengcai Liao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Video face restoration (VFR) aims to recover high-quality and temporally consistent facial details from severely degraded video sequences; however, existing methods still struggle to balance spatial fidelity and temporal coherence under complex degradations. To address this, we propose FADRA, a frequency-aware diffusion framework with iterative residual adaptation specifically tailored for robust VFR. We first leverage the strong temporal consistency of a pre-trained text-to-video diffusion model and introduce lightweight LoRA adapters together with a Low-Quality (LQ) Pixel-Alignment Feature Fusion module to efficiently adapt the frozen generative prior to the VFR task. To further adapt the frozen diffusion backbone to the downstream VFR task beyond LoRA-based adaptation, we introduce a Repeated Residual Adaptation Head (RRAH) for step-wise residual refinement after the diffusion backbone. To make this refinement explicitly guided by the degraded observation, RRAH further takes the LQ latent together with the current velocity prediction as input, allowing the model to repeatedly revisit LQ cues and predict residual updates at each flow-matching step. This LQ-guided repeated residual adaptation helps recover fine facial details while preserving the inherent temporal priors of the pre-trained model. Furthermore, to ensure the structural integrity of perceptually important details, we introduce a Frequency-Aware Loss that provides explicit supervision across multiple spectral bands, emphasizing visually sensitive frequency components that are crucial for perceptual quality and prone to temporal jittering. Extensive experiments demonstrate that FADRA recovers better facial structures and produces more temporally consistent videos than state-of-the-art methods, leading to clear gains in both quantitative metrics and visual perception.
[CV-27] Learning to Throw Objects Safely in Multi-Obstacle Environments ICRA
链接: https://arxiv.org/abs/2607.06388
作者: Mohammadreza Kasaei,Klemen Voncina,Hamidreza Kasaei
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: This paper has been presented at the IEEE International Conference on Robotics Automation (ICRA), 2026
Abstract:Robotic throwing enables fast and efficient object placement beyond the robot’s immediate workspace, but reliable throwing in cluttered environments remains underexplored. Existing approaches, such as TossingBot, learn throwing strategies from visual input but assume obstacle-free settings. In this paper, we address the problem of throwing objects into a target basket while avoiding obstacles placed randomly in the scene. We introduce a potential field state representation that compactly encodes both basket attraction and obstacle repulsion on a fixed-size grid, enabling reinforcement learning (RL) policies to generalize across arbitrary numbers and configurations of obstacles. The policy is initialized from kinesthetic demonstrations and optimized in simulation using three state-of-the-art RL algorithms (SAC, DDPG, TD3). Among these, SAC achieves the most consistent performance across scenarios. We compare the potential field representation against explicit state encodings and demonstrate that it achieves higher success rates and better scalability to unseen obstacle configurations. Real-robot experiments with unseen throwable objects confirm robust sim-to-real transfer, achieving up to 90% success in cluttered scenes. These results demonstrate that PFR provides a practical and robust representation for safe and efficient robotic throwing in unstructured environments. A video showcasing our experiments is available at: this https URL
[CV-28] VaseMuseum: Digital Intelligent Museum for Ancient Greek Pottery
链接: https://arxiv.org/abs/2607.06374
作者: Jiazi Wang,Nonghai Zhang,Qiushi Xie,Zeyu Zhang,Yufeng Chen,Yang Zhao,Ling Shao,Hao Tang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Code: this https URL . Website: this https URL
Abstract:Vision-language models (VLMs) have made interactive digital museums increasingly feasible by connecting 3D digitization with natural-language artifact exploration. However, in cultural heritage domains such as ancient Greek pottery, reliable VLM assistance is limited by two challenges. First, open-ended interpretation requires grounding fine-grained 2D/3D visual evidence in specialized curatorial knowledge, yet the retrieval process may introduce weak sources and unverifiable references. Second, when the available evidence is incomplete, noisy, or ambiguous, VLMs often produce confident but unsupported answers instead of calibrated uncertainty. To address these challenges, we propose VaseMuseum, a lightweight and modular multimodal agent framework for intelligent digital museums of ancient Greek pottery. VaseMuseum combines an interactive virtual museum with VaseAgent, which supports both 2D images and 3D artifacts through multimodal perception, 3D-aware reasoning, external knowledge retrieval, and inference-time reliability control. Specifically, VaseAgent retrieves evidence from authoritative web and museum knowledge sources, and source-level control selects diverse and verifiable evidence before generation. Meanwhile, response-level control checks generated claims against the evidence pool and encourages neutral, evidence-bounded answers when support is insufficient or conflicting. Moreover, a training-free GRPO-style selection mechanism favors responses with valid references and calibrated confidence without updating the VLM backbone. Experiments in a realistic digital museum simulation show that VaseMuseum improves citation validity, reduces hallucinations on knowledge-intensive queries, and produces more neutral answers under ambiguity compared with search-enabled VLM baselines.
[CV-29] raining-Free Acceleration for Vision-Language-Action Models with Action Caching and Refinement
链接: https://arxiv.org/abs/2607.06370
作者: Ryuji Oi,Hikari Otsuka,Kosuke Matsushima,Yuki Ichikawa,Masato Motomura,Tatsuya Kaneko,Daichi Fujiki
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Vision-Language-Action (VLA) models have emerged as a promising approach for generalizable robotic manipulations. In particular, flow matching-based VLA models have shown remarkable success due to their capability to generate precise and smooth action sequences and capture multimodal distributions. However, the iterative denoising process in the action head acts as a major computational bottleneck, posing a critical challenge for real-time deployment. To address this challenge, we propose ActionCache, a plug-and-play external cache that opportunistically reuses past intermediate actions to warm-start generations from the vicinity of target actions, thereby drastically reducing the inference latency. Specifically, ActionCache stores the intermediate actions with compact multimodal keys, which enables retrieval from similar past contexts across different episodes or even different tasks. Experimental results in simulation and real-world environments demonstrate that ActionCache maintains high task success rates in a low-latency regime, achieving inference acceleration of up to 11.75\times and 34.43\times for representative flow-based VLA models, \pi_0.5 and GR00T-N1.6, respectively.
[CV-30] MF-RSE: Tri-Modal Fusion with Regional Semantics and Evidential Uncertainty for Lung Severity Scoring
链接: https://arxiv.org/abs/2607.06356
作者: Fadi Abdeladhim Zidi,Salah Eddine Bekhouche,Abdellah Zakaria Sellam,Gaby Maroun,Fadi Dornaika,Cosimo Distante
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 6 pages, 2 figures, 5 tables. IEEE conference format (IEEEtran). Submitted to AVSS 2026. Tri-modal fusion for lung severity scoring using appearance, segmentation, and VLM semantics with evidential uncertainty
Abstract:Accurate quantification of lung disease severity from chest imaging is critical for clinical decision-making and resource allocation. We propose a tri-modal deep learning framework, TMF-RSE (Tri-Modal Fusion with Regional Semantics and Evidential Uncertainty), that combines appearance features from two-dimensional chest inputs, structural features from lung segmentation masks, and semantic features from vision-language models (VLMs) for severity quantification. Our approach employs complementary fusion mechanisms that integrate semantic guidance, structural priors, and hierarchical interactions across modalities. The model employs evidential regression to provide both severity predictions and uncertainty estimates. Experiments on the Per-COVID-19 CT and RALO datasets show that TMF-RSE outperforms recent transformer-based baselines, achieving MAE of 4.02 and Pearson correlation of 0.9629 on Per-COVID-19 validation, and 0.339 MAE / 0.973 PC on RALO geographic extent.
[CV-31] Generalized Synthetic Image Detection with Enhanced RGB-Noise Representation Learning
链接: https://arxiv.org/abs/2607.06354
作者: Zhen Li,Gang Cao,Tian Zhang,Lifang Yu,Shaowei Weng
类目: Computer Vision and Pattern Recognition (cs.CV); Cryptography and Security (cs.CR)
备注:
Abstract:The rapid advancement of large-scale generative models has accelerated the spread of highly deceptive AI-generated images, making generalized synthetic image detection a critical imperative. Existing forensic networks often struggle with cross-model generalization and realworld degradations due to their reliance on single-domain representations and conventional binary classification optimization. To overcome these limitations, we propose RNSIDNet, a novel forensic framework that achieves robust detection through enhanced RGB-Noise representation learning. Specifically, our method employs a dual-branch architecture where global RGB semantics, extracted by an attention-refined CLIP backbone, dynamically modulate highfrequency noise artifacts captured by Bayar convolutions via a Feature-wise Linear Modulation (FiLM) module. To further enhance the learned representations, we design a Hard Sample-aware Contrastive Learning (HSCL) strategy. By explicitly penalizing challenging training samples, HSCL reshapes the latent feature space to maximize the discriminative margin between pristine and synthetic domains. Extensive experiments across eight public benchmark datasets verify that our model achieves state-of-the-art performance, delivering superior generalization ability, robustness, and computational efficiency. Code and dataset will be publicly available on this https URL.
[CV-32] OrchardBench: A Physically-Grounded GPU-Parallel Apple-Orchard Simulation Benchmark for Agricultural Robotics
链接: https://arxiv.org/abs/2607.06337
作者: Humphrey Munn
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Robotic tree-fruit harvesting is a flagship problem for agricultural automation, but progress is bottlenecked by the cost and irreproducibility of field experiments: an orchard is available only weeks a year, every tree is different, and a control error can permanently damage the crop or the plant. The tree models used in graphics and agronomy are geometrically detailed but physically inert, while the GPU-parallel simulators used in robot learning contain no plausible trees. We present OrchardBench, a physically-grounded, GPU-parallel simulation of apple-orchard trees on the Newton engine. Each tree is grown by a stochastic L-system and instantiated as a fully articulated body: branches are compliant torsional spring-dampers whose stiffness follows Euler-Bernoulli beam theory, they break at a wood modulus of rupture and fall as free hinges, and apples are independent bodies on stem tethers that detach at literature-grounded pull forces and load the branch when pulled. A moving, density-controllable foliage layer occludes the canopy as real leaves do. Every physical parameter is tied to a published source. Per-environment domain randomization makes each batched world a distinct tree, and a mobile manipulator with a wrist depth camera closes the loop with geometric fruit perception and an autonomous harvesting baseline. Careful engineering of the solver and the model lets OrchardBench run many parallel environments at interactive rates on a laptop GPU. We define the tasks and a metric suite spanning harvest completeness, throughput, and plant damage (with a per-canopy-zone breakdown), and report baseline results across foliage, fruit load, terrain, canopy zone, and parallelism. The analytic baseline succeeds on about 40% of the fruit it detects and harvests only about an eighth of the reachable fruit on a tree, leaving clear headroom for novel autonomy approaches.
[CV-33] Bridging Diffusion Pruning and Step Distillation with Teacher-Aligned Repair
链接: https://arxiv.org/abs/2607.06335
作者: Jincheng Ying,Li Wenlin,Minghui Xu,Yinhao Xiao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Diffusion models generate high-quality images, but their inference cost comes from two sources: large denoising networks and repeated denoising steps. Existing compression pipelines usually attack these costs separately. Pruning reduces the network, but most pruning methods still rely on a long post-pruning retraining stage to recover a many-step sampler. Step distillation reduces the number of denoising steps, but it usually assumes a student that can already follow the teacher well enough to receive useful distillation gradients. This paper asks whether post-pruning retraining can be replaced by step distillation. We find that the direct replacement fails: after pruning an EDM2-XS teacher, starting SiDA from the pruned checkpoint produces unusable samples. We introduce a short teacher-alignment repair stage as a bridge between pruning and step distillation. The bridge matches the pruned generator to the teacher on noisy real-image latents, then hands the repaired checkpoint to one-step distillation. On ImageNet-512, the original EDM2-XS baseline uses 124.713M parameters and 63 network evaluations, reaching an FID of 3.53. With a suitable distillation objective, our 20% pruned one-step generator uses 98.826M parameters and one network evaluation, reaching an FID of 3.12. With 30% pruning, the model uses 88.029M parameters and one network evaluation, with an FID of 4.26.
[CV-34] Driving the Wrong Way: Leverag ing Interpretability in End2End Autonomous Driving Models
链接: https://arxiv.org/abs/2607.06328
作者: Franz Motzkus,Sebastian Bernhard
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注:
Abstract:The increasing adoption of end-to-end learning for autonomous driving introduces increased model complexity and opacity, raising the risk of learning undesired or erroneous behavior. In this work, we integrate unsupervised dictionary learning as a post hoc interpretability module within state-of-the-art driving models to decompose driving behavior into semantically meaningful concepts while demonstrating their causal influence on the model’s driving decisions. We propose a stepwise framework for extracting and interpreting meaningful concepts from the end-to-end model and connecting them to the multifaceted model outputs, thereby revealing the underlying decision-making logic for the prediction of future trajectories. Furthermore, targeted interventions at the concept level allow us to manipulate and correct driving decisions, resulting in measurable improvements in overall driving performance. We thus demonstrate how interpretability can effectively be used to reduce model opacity, uncover erroneous behavior, and enable targeted mitigation, ultimately boosting model performance.
[CV-35] Synthetic-to-Real Translation for Class-Agnostic Motion Prediction
链接: https://arxiv.org/abs/2607.06319
作者: Yizheng Wu,Hongwei Fan,Kewei Wang,Ruibo Li,Xingyi Li,Xiao Song,Zhe Wang,Chenjing Ding,Dongliang Wang,Zhiguo Cao,Guosheng Lin
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Motion understanding is critical for ensuring safety and robustness in autonomous driving systems, driving increasing interest in motion prediction. A key challenge in this domain is the high cost associated with acquiring real-world motion labels. It is therefore ideal if we could transfer motion knowledge from synthetic data to real data. In this context, we explore the potential of synthetic-to-real translation for motion prediction (SRMP). However, the most used naive motion regression methods are notably sensitive to the synthetic-to-real domain shift, resulting in unreliable knowledge translation. To address this, we propose a novel approach integrating a motion knowledge translation framework with two key components: (1) objectness-aware motion prediction, which explicitly models the joint distribution of motion patterns and objectness priors to improve domain-invariant feature learning, and (2) objectness-aided motion enhancement, a motion label refinement mechanism that leverages learned objectness priors to filter motion noise. Furthermore, we present a physically-based pipeline for generating Motion4D, the first synthetic 4D LiDAR dataset tailored for SRMP research, addressing the lack of synthetic motion datasets. Experimental results demonstrate that our approach effectively bridges the domain gaps and yields superior performance on real scenes.
[CV-36] oken-Based Dual-view Fusion and Adaptation of Large Vision Models for Breast Cancer Classification
链接: https://arxiv.org/abs/2607.06309
作者: Aysan Ghayouri Pirsoltan,Shima Babakordi,Mohammad Reza Mohammadi
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Accurate breast cancer classification from mammography requires effective integration of complementary information from craniocaudal (CC) and mediolateral oblique (MLO) views, which provide a more complete characterization of breast abnormalities. However, existing multi-view learning approaches typically rely on feature-level aggregation or single-stage cross-attention, which can entangle view-specific and shared representations and restrict interaction to limited network depths. To address these limitations, we propose a token-centric dual-view learning framework that unifies prompt-based adaptation and cross-view fusion within a frozen vision transformer backbone. The framework reformulates inter-view interaction as structured token-level communication, where dedicated fusion tokens explicitly encode bidirectional information exchange between CC and MLO views via cross-attention, serving as intermediate carriers of cross-view dependencies rather than relying on direct feature fusion. Unlike conventional methods that apply fusion at a single layer, fusion modules are inserted at multiple transformer depths, enabling progressive and repeated interaction across the encoder hierarchy. Fusion tokens are reintegrated into the token sequence and refined by subsequent transformer layers, facilitating hierarchical propagation of complementary information while preserving view-specific structure. Experiments on VinDr-Mammo and CMMD datasets demonstrate consistent improvements over linear probing, prompt-only adaptation, and conventional fusion baselines. On the VinDr-Mammo BI-RADS classification task, the framework achieves 50.40% F1-score and 0.8090 AUC, including a 0.10 AUC improvement over a dual-view fusion baseline in the binary setting. Ablation studies further validate the effectiveness of token-based fusion and multi-depth interaction design.
[CV-37] UI2App: Benchmarking Visual Interaction Inference in Executable Web Application Generation
链接: https://arxiv.org/abs/2607.06306
作者: Grace Man Chen,Litao Guo,Yifan Wu,Yiyu Chen,Yenchi Tseng,Sicheng Liu,Yuyu Luo,Ying-Cong Chen
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Large language models (LLMs) have demonstrated growing competence in web page generation. However, existing text-driven approaches rely on complex prompts that impose substantial demands on users and offer limited expressivity for page layout and cross-page visual coherence. Image-driven paradigms, which take UI screenshots as input, align more closely with real development workflows. However, current benchmarks focus primarily on visual fidelity and lack a systematic evaluation of the interaction capabilities in generated artifacts. To address this gap, we introduce UI2App, the first benchmark targeting interaction inference, the ability to recover application behavior from screenshots alone, without any textual or behavioral guidance. UI2App comprises 327 screenshots grouped into 45 state-coherent screenshot sets for runnable multi-route web applications. We design an end-to-end pipeline that evaluates each artifact along four dimensions: executability, navigation reachability, visual fidelity, and interaction inference. The interaction metric (IIS) assesses inferred interactions by functional correctness and state-management complexity, crediting any valid implementation rather than matching a single reference. Experiments on six frontier vision-language models reveal a marked capability mismatch between visual reconstruction and interaction realization: the visual-fidelity leader scores only 7.5 on IIS, ranking fourth and trailing the IIS leader by 5.2x. High-complexity interactions such as cross-page state remain a pervasive bottleneck, with half of the evaluated models scoring exactly zero on this dimension. Overall, the results indicate that inferring complete interaction behavior from static screenshots remains a key challenge for models.
[CV-38] Visual graphs for image classification: does the structure affect performance?
链接: https://arxiv.org/abs/2607.06295
作者: Alessandra Ibba
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Deep learning models have emerged in machine learning and related fields, demonstrating astonishing performance in various visual tasks. Despite their great success, however, these models are unable to fully encode intrinsic visual structures, and often ignore the spatial, topological, and semantic information contained within an image. Graph neural networks offer a good framework to face this aspect, but their effective use for visual tasks has been only partly explored and mainly starting from a limited perspective. This work aims to address this gap by conducting a systematic comparison of current graph construction techniques within the context of a fixed three-layer GCN architecture. Through an empirical study, it demonstrates in particular how the network structure affects performance and provides an important methodological contribution regarding the computational stages preceding graph utilization, which will be strongly influenced by the structure itself.
[CV-39] Straight-Path Flow Matching for Incomplete Multi-View Clustering ECCV2026
链接: https://arxiv.org/abs/2607.06281
作者: Yiteng Yuan,Junyan Wang,Zheyuan Liu,Hong Jia,Lei Fan,Zhulin Tao,Lianbo Guo
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ECCV 2026. 28 pages, 6 figures, 4 tables
Abstract:Incomplete Multi-View Clustering addresses the problem of clustering multi-modal data when certain views are missing. Recent end-to-end generative approaches leverage diffusion models to recover missing views via stochastic noise-to-data trajectories. While expressive, such mechanisms are not explicitly designed for clustering, as they initialize from cluster-agnostic noise and rely on stochastic denoising dynamics. In this work, we revisit probability path design in end-to-end generative IMVC. We introduce a flow-matching framework with a linear interpolation path between paired view representations, that replaces diffusion with probability flows between observed and missing views. We provide a formal analysis showing that deterministic ODE flows are inherently better aligned with clustering objectives than diffusion-based stochastic trajectories, especially in terms of transport mechanisms that respect class-conditional data distributions and maintain cluster consistency in finite-step regimes. Building upon this insight, we develop an end-to-end IMVC architecture that integrates straight-path flow-matching view completion with cluster-level and entropy-based alignment to enforce cross-view clustering consistency. Extensive experiments on standard IMVC benchmarks demonstrate that the proposed framework establishes new state-of-the-art performance.
[CV-40] MAC-XA: Multi-view Anatomy-Correspondence Fusion for Coronary Stenosis Reporting from X-ray Angiography
链接: https://arxiv.org/abs/2607.06268
作者: Chen Jia,Baochang Zhang,Fatia Kusuma Dewi,Amir Yousefi,Heribert Schunkert,Reza Ghotbi,Nassir Navab
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Preprint
Abstract:Multi-view reasoning in coronary X-ray angiography is inherently a cross-projection geometric problem, yet automated report generation in this setting remains largely unexplored. The 3D vascular topology leads to projection-dependent branch overlap and foreshortening, rendering single-view modeling fundamentally incomplete and unstable for lesion localization and stenosis grading. Although multi-view fusion appears promising, learning anatomically consistent fusion from real angiograms is impeded by a critical limitation: cross-view alignment is unobservable and cannot be explicitly supervised. Consequently, conventional fusion relies on implicit correlations rather than verified anatomical correspondence. We address this by reformulating multi-view stenosis reporting as an alignment-constrained aggregation problem. A controllable synthetic angiography generation strategy is introduced to expose geometry-derived patch-level correspondence supervision unavailable in real data. An anatomy-correspondence module learns cross-view correspondence matrices that explicitly align auxiliary features within the main-view coordinate space prior to fusion, thereby constraining evidence aggregation to anatomically consistent regions. Experiments on synthetic data and zero-shot transfer to real angiograms show that this alignment-constrained design improves correspondence consistency and structured stenosis reporting compared to single-view modeling and conventional multi-view fusion methods. The code will be publicly available upon publication.
[CV-41] VendorBench-100: A Unified Cross-Paradigm Benchmark for Deepfake Image Detection
链接: https://arxiv.org/abs/2607.06254
作者: Sharayu N. Deshmukh,Md Rashidunnabi,Nelton Tiago Gemo,Kurundkar G. D.,Mahamune M. R.,Nilesh K. Deshmukh
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 22 pages, 10 figures, 3 tables. Code and data: this https URL
Abstract:Deepfake image detection is currently served by three fundamentally different paradigms: commercial APIs, zero-shot vision-language models (LLMs), and open-source detectors. Despite their widespread use, these paradigms are rarely evaluated under a common protocol, making direct comparison difficult. We introduce VendorBench-100, a cross-paradigm benchmark that evaluates 36 representative models using a single adversarial 100-image corpus, a unified output schema, and a common evaluation framework. To ensure reliable assessment under the corpus’s intentional class imbalance, models are ranked primarily by the Matthews correlation coefficient (MCC), with ROC-AUC reported as a threshold-independent measure of ranking ability. Rather than maximizing dataset size, VendorBench-100 emphasizes challenging real-world scenarios through a curated taxonomy of eight edge-case families, including face swaps, text-to-video stills, AI photo edits, avatar compositing, opaque-provenance images, and compressed research frames. Our evaluation shows that commercial APIs achieve the strongest median performance, followed by vision LLMs and open-source detectors. However, individual open-source models remain competitive with the best vision LLMs. More importantly, we identify a consistent divergence between ranking ability (ROC-AUC) and operating-point quality (MCC), demonstrating that strong score discrimination does not necessarily produce reliable default-threshold decisions. This metric disagreement, rather than any single leaderboard ranking, is the central finding of the benchmark. We release the complete evaluation framework and benchmark results to support reproducible future research. The source code and data are available at: this https URL
[CV-42] PhyMRI-SR: Toward Physics-Aware MRI Image Super-Resolution
链接: https://arxiv.org/abs/2607.06238
作者: Lihua Wei,Huatong Gao,Jia Gong,Zhiyu Tan,Hao Li,Jun Liu,Zhihua Ren
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project Page: this https URL
Abstract:Magnetic resonance imaging (MRI) super-resolution is vital for improving diagnostic accessibility, yet most methods treat it as a deterministic mapping from a fixed low-resolution input to a high-resolution target. This overlooks a key property of MRI acquisition physics: spatial resolution and signal-to-noise ratio (SNR) are inherently coupled, making any given low-resolution scan merely one of many possible realizations under varying acquisition trade-offs. We rethink MRI super-resolution as a physics-aware reconstruction problem, in which the goal is to identify the optimal resolution-SNR configuration and then super-resolve it to obtain high-quality MRI results. A key implication of this formulation is that MRI resolution becomes dynamic rather than fixed. To handle such resolution-heterogeneous inputs, we adapt 2D Gaussian Splatting (2D GS) to MRI by formulating reconstruction as a coordinate-based, resolution-agnostic rendering problem. To further enhance fidelity, we introduce three innovations: (1) a prior-aware Gaussian representation that combines an Anatomical Structure Prior for tissue-specific kernel initialization with an Imaging System Prior that captures hardware characteristics via a covariance dictionary; (2) a physics-constrained signal modeling scheme that predicts intrinsic tissue parameters (proton density rho and effective relaxation rate R2) and synthesizes intensities through governing physical equations, ensuring biophysically plausible contrast; and (3) a meta-learning framework that alleviates paired-data scarcity by pretraining on simulated data and adapting to real-world conditions. Extensive experiments on dynamic-resolution datasets and standard benchmarks demonstrate that our method achieves state-of-the-art performance, highlighting its strong potential for clinical deployment.
[CV-43] WING: A Window-Prior-Based Generative Network with Gated Inception for Cross-Modality CT Synthesis
链接: https://arxiv.org/abs/2607.06234
作者: Siyuan Mei,Yan Xia,Yipeng Sun,Siming Bayer,Zirong Li,Chengze Ye,Daiqi Liu,Fuxin Fan,Yixing Huang,Andreas Maier
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Generating CT volumes from MRI and CBCT can improve treatment planning in adaptive radiotherapy while avoiding additional radiation exposure. However, direct regression of CT intensities is challenged by the inherently high dynamic range and long-tailed distributions, thereby averaging out sparse yet clinically important structures. To alleviate this issue, we reformulate the regression target into multiple windowed representations, leveraging the inductive prior that CT intensities are structure-deterministic and window-separable. These windowed views exhibit smoother distributions and admit structured fusion back to the full-range CT. Building on this reformulation, we introduce WING, a WINdow-prior-based Generative network comprising: 1) a new Gated Inception Generator to produce multi-window predictions, enabling multi-shape kernel interactions to capture cross-modality correspondence; 2) a Fuse-and-Refine Transformer to aggregate the windowed outputs and learn residuals for detail refinement; and 3) a joint adversarial training objective to enhance window-conditioned realism. Extensive experiments demonstrate that our compact WING achieves state-of-the-art performance on the MRI-to-CT and CBCT-to-CT benchmarks, while supporting multi-anatomy synthesis with a single model.
[CV-44] EeveeDark: A Binary Neural Framework for Low-Light Video Enhancement via Event-Guided Sensor-Level Fusion
链接: https://arxiv.org/abs/2607.06217
作者: Onur Eker,Erkut Erdem,Aykut Erdem
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Enhancing videos under extreme low-light conditions remains challenging due to the difficulty of balancing restoration quality and computational efficiency in resource-constrained settings. This paper introduces EeveeDark, a low-light video enhancement framework that combines the spatial richness of sensor-level RAW data with the temporal precision of event streams. Central to our model is a Binary Neural Network (BNN) architecture that reduces computational overhead by quantizing weights and activations while preserving detail. EeveeDark incorporates (i) modality-specific binary encoders for processing RAW frames and event data, (ii) a lightweight fusion block for integrating spatial and temporal cues, and (iii) an event-guided skip gating mechanism for dynamic spatiotemporal refinement. Experiments on synthetic and real-world datasets show that EeveeDark outperforms prior BNN-based methods and offers a favorable performance-efficiency trade-off compared to full-precision models. The project page is available at this https URL.
[CV-45] MoWorld: A Flash World Model
链接: https://arxiv.org/abs/2607.06216
作者: Team Moxin,Deyi Ji,Tianrun Chen,Xin Zhang,Jiale Yang,Qi Zhu,An Zhao,Zihao Xie,Han Wang,Xuanyi Liu,Yixiang Zhou,Pei Liu,Yi Tan,Cheng Chen,Dayi Zhu,Mingyu Wei,Hanjie Xu,Jun Liao,Siqi Li,Lingyu Lu,Hongye Fang,Hongming Tan,Youjiang Zhu,Taiyu Zhang,Zejian Li,Chaotao Ding,Lanyun Zhu,Yunhe Pan,Lingyun Sun
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project Page: this https URL
Abstract:The future of World Models depends not only on scaling model capability, but also on scaling practicality and inference efficiency. High-frame-rate inference enables responsive perception, planning, and control in real-world autonomous systems. To this end, we present MoWorld, a cost-effective yet high-performance Flash World Model with an end-to-end framework spanning data generation, pre-training, distillation, and efficient inference, enabling up to 50 FPS real-time interaction with cinematic visual quality without the need of high-end GPUs. To enable large-scale real-world deployment, MoWorld jointly optimizes model capability and cost throughout the entire development pipeline. Specifically, unlike existing approaches that primarily rely on large-scale video corpora, MoWorld is built upon a scalable 3D-native data engine accumulated from our large-scale 3D vision and generative modeling pipeline, enabling the efficient construction of geometrically consistent training data across diverse real-world and synthetic environments. Based on this foundation, a curriculum cross-frame pre-training strategy for stable and scalable World Model learning, an efficient denoising-step distillation algorithm to reduce diffusion training cost, and a mixed-precision parallel inference framework for low-cost real-time deployment. MoWorld is the first real-time interactive World Model built on the Neural Processing Unit (NPU) and can achieves up to 50 FPS in such the devices, enabling practical and efficient deployment at scale. Comprehensive evaluations demonstrate that MoWorld achieves leading performance; notably, its average inference cost is only 30%-50% of that of existing World Models, providing a practical foundation for large-scale real-world applications of World Models. We also demonstrate diverse applications of MoWorld.
[CV-46] Structured-Condensed Prompt Tuning in Vision-Language Models for Fine-grained Image Recognition
链接: https://arxiv.org/abs/2607.06185
作者: Xinda Liu,Qinyu Zhang,Weiqing Min,Guohua Geng,Shuqiang Jiang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Fine-grained image recognition poses a significant challenge due to the substantial expertise and effort required for manual annotation. Vision-language models (VLMs) like CLIP provide a compelling zero-shot alternative, reducing reliance on extensive labeled data. However, their ability to capture subtle distinctions remains limited, leading to subpar recognition performance. While prompt tuning has proven effective for adapting VLMs, most existing methods treat class labels as isolated, discrete entities, overlooking the rich semantic relationships between them. This oversimplified assumption limits the model’s ability to capture hierarchical dependencies and inter-class correlations – both critical for distinguishing visually similar categories. The problem is especially acute in fine-grained classification, where accurate recognition depends on understanding complex label semantics. To address this, we propose Structured-Condensed Prompt Tuning (SCPT), which enhances semantic structure modeling in prompt learning. Specifically, we introduce Semantic Relation Encoding (SRE) to explicitly model inter-class semantic topology and encode structured label relationships. In parallel, we design a Semantic Condensation loss (ScLoss) to suppress redundant supervision and extract discriminative components from the global semantic space. Together, these components significantly improve semantic alignment and fine-grained discrimination. Extensive experiments on 14 fine-grained benchmarks show that SCPT effectively mitigates semantic ambiguity and achieves state-of-the-art performance in both few-shot and base-to-novel generalization settings.
[CV-47] Revisiting Scene Graph Generation from the Perspective of Detector-Conditioned Reachability ECCV2026
链接: https://arxiv.org/abs/2607.06176
作者: Runfeng Qu,Pia K Bideau,Ole Hall,Julie Ouerfelli-Ethier,Klaus Obermayer,Olaf Hellwich
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by ECCV 2026
Abstract:Scene graph generation (SGG) approaches can be broadly classified into detector-based and query-based methods according to their underlying reasoning mechanisms. However, the discrepancy in their predictive behaviors, induced by these distinct mechanisms, has not been systematically analyzed. In this work, we design a controlled experimental setup to examine prediction discrepancies from the perspective of detector-conditioned reachability. The results suggest clear complementary clues. Motivated by this observation, we introduce a Dual-SGG method that consolidates both reasoning mechanisms via a dual-query design, thereby leveraging the complementary predictive behaviors of both detector-based and query-based methods. Extensive experiments on the Visual Genome, Open Images v6, and GQA-200 datasets demonstrate the effectiveness of the proposed method.
[CV-48] MobileWan: Closing the Quality Gap for Mobile Video Diffusion
链接: https://arxiv.org/abs/2607.06173
作者: Mohsen Ghafoorian,Denis Korzhenkov,Adil Karjauv,Ioannis Lelekas,Noor Fathima,Spyridon Stasis,Hanno Ackermann,Boris van Breugel,Markus Nagel,Fatih Porikli,Animesh Karnewar,Amirhossein Habibian
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Recent advances in video diffusion have been driven by scaling transformer-based architectures to billions of parameters, substantially improving visual fidelity and motion coherence. In contrast, existing mobile video diffusion models remain limited to relatively small parameter budgets, typically 0.4-1.8B, restricting generation quality. In this work, we show that high-quality mobile video generation does not require small models. Instead, we demonstrate that a server-scale 5B-parameter video diffusion transformer can be deployed efficiently on memory-constrained mobile hardware through recurrent reformulation and structured compression. Starting from Wan2.2-5B, we rely on a recurrence distillation framework that converts video generation into a chunk-wise autoregressive process with constant-memory attention computation. Combined with causal linear attention, the model operates as an RNN at inference time while preserving temporal coherence across chunks. We further propose a learnable attention head pruning method based on binary per-head gates optimized end-to-end using a noise-biased sparsity objective and distillation-based finetuning. Together with sampling-step distillation and memory-optimized VAE decoding, MobileWan becomes the first 5B-scale video diffusion model deployable on a commercial mobile device. Our system generates 5-second 480x832 videos at 16 FPS in 20 seconds end-to-end latency, achieving a VBench score of 83.79 and establishing a new state of the art in mobile video generation. Project page: this https URL
[CV-49] High-Resolution Artwork Outpainting with Global Blueprint Guidance and Layout Control ECCV2026
链接: https://arxiv.org/abs/2607.06162
作者: Junha Kim,Hyunjoon Park,Donghyeon Cho
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at ECCV2026
Abstract:Image outpainting extends an image beyond its original borders, requiring seamless style integration and globally coherent scene completion. Building on the success of diffusion models, recent methods have achieved substantial improvements in visual quality. In practice, however, high-resolution outpainting is commonly performed via progressive expansion around a fixed source image, particularly in artwork scenarios. Despite this progress, existing approaches still suffer from three key limitations: (i) the absence of a reliable global planning mechanism, which leads to structural instability and error accumulation at high resolutions; (ii) limited spatial controllability beyond text prompts, making it difficult to place objects at user-specified locations; and (iii) high inference latency caused by inherently sequential patch generation. To address these issues, we propose a global blueprint-guided two-stage diffusion framework for layout-controllable high-resolution outpainting with efficient parallel synthesis. In Stage 1, we generate a low-resolution global blueprint using a layout adapter that injects bounding-box conditions into a Stable Diffusion inpainting backbone, producing a globally consistent structural plan while extracting global guidance features. In Stage 2, we synthesize high-resolution local patches in parallel by injecting the blueprint-derived global guidance and initializing each patch from the blueprint using the low-frequency preservation property of forward diffusion. This design eliminates sequential dependency while maintaining global coherence. Extensive experiments on large-scale artwork datasets demonstrate improved visual fidelity, stronger semantic consistency, and substantially reduced inference time compared to prior baselines, while uniquely supporting explicit layout control for artwork outpainting.
[CV-50] Enhanced Seam Segmentation for Automated Welding Robot in Construction Through Transfer Learning: Addressing Limitations of Bilateral Segmentation Network
链接: https://arxiv.org/abs/2607.06150
作者: Keonvin Park,Yong Ann Voeurn,Hyeokjun Kweon,Doyun Lee
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Reliable seam segmentation is essential for autonomous robotic welding in construction, where harsh illumination, specular reflections, and thin weld geometries often degrade segmentation performance. This study proposes a reflection-robust seam segmentation framework that enhances a BiSeNetV2 backbone through transfer learning and a hybrid Cross-Entropy–Lovász loss. Rather than increasing architectural complexity, the proposed framework improves reflection robustness through learning-stability-oriented optimization. Experimental results show that the proposed method achieves 81.76% Joint IoU and 90.73% mIoU, improving Joint IoU by +22.36 percentage points over the OHEM-based baseline while maintaining identical FLOPs, parameter count, and inference speed. The proposed approach also recovers 96.33% of severe zero-IoU failure cases under reflective conditions. Comparative experiments across BiSeNetV2, DeepLabV3+, UNet, and SegFormer further demonstrate that the proposed optimization strategy is particularly effective for lightweight real-time segmentation architectures. Qualitative analyses additionally show improved seam continuity and reflection robustness in challenging welding environments. These findings suggest that the proposed framework provides a practical and lightweight perception solution for robotic welding applications involving reflective metallic surfaces.
[CV-51] RFHNet: Relational and Frequency-Aware Hashing Network for Large-Scale Fine-Grained Food Image Retrieval ICMR2026
链接: https://arxiv.org/abs/2607.06148
作者: Junsong Wang,Weiqing Min,Guorui Sheng,Tao Yao,Lili Wang,Shuqiang Jiang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 10 pages, 6 figures. Published in ACM ICMR 2026
Abstract:Fine-grained food image retrieval is a key task in computational gastronomy, with applications in food traceability, dietary monitoring, and smart catering systems. Although hashing-based retrieval is attractive for large-scale search due to its storage efficiency and fast Hamming-distance computation, existing methods often perform poorly in fine-grained food scenarios, where subtle local semantics and frequency-sensitive visual cues are essential. To address this challenge, we propose RFHNet, a cascaded hierarchical hashing network that captures both global structure and fine-grained local details through multi-level representations. RFHNet includes three components: (1) Fine-grained Relation Modeling (FRM) to capture subtle visual differences among similar food components; (2) Multi-Frequency Modulated Fusion (MFMF) to extract informative multi-frequency features; and (3) Hierarchical Semantic Synergy (HSS) to adaptively integrate multi-level representations and generate discriminative hash codes. Experiments on six food-specific benchmarks show that RFHNet consistently outperforms state-of-the-art hashing methods, with mAP gains of 4.44% to 17.20% at 12 bits. These results validate the effectiveness of RFHNet for large-scale visual food retrieval and smart catering applications. The source code will be released upon publication.
[CV-52] uning-Free Latent Diffusion Models for Ultrahigh-Resolution Image Editing
链接: https://arxiv.org/abs/2607.06136
作者: Wanglong Lu,Lingming Su,Kaijie Shi,Minglun Gong,Xiaogang Jin,Hanli Zhao,Xianta Jiang
类目: Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM)
备注: 29 pages, 29 figures. Published in IEEE Transactions on Neural Networks and Learning Systems
Abstract:Recent diffusion-based generative models have shown impressive performance in image generation and editing. However, due to memory limitations and the high cost of collecting high-resolution training images, existing methods are typically restricted to inputs with linear resolutions below 1K. In contrast, photos captured by modern mobile devices often reach linear resolutions up to 8K, revealing a significant gap between current capabilities and real-world demands. Simply upscaling low-resolution edited results often results in visually enlarged but blurry images that lack fine details. This paper introduces UltraDiffEdit, a novel, tuning-free image editing framework that extends off-the-shelf latent diffusion models (LDMs) to ultrahigh resolutions. UltraDiffEdit employs a multi-scale progressive editing strategy, iteratively blending high-resolution edited content with unedited areas in a coarse-to-fine manner. We employ multi-patch encoding to preserve both edited and unedited visual details within the latent space. To mitigate editing artifacts, our global-local consistency denoising technique consistently integrates edited and unedited latent features, ensuring smooth transition at editing boundaries from the latent representation to the final image. We also introduce a patch-based hybrid sampling approach that captures local, intermediate, and global features, ensuring semantic coherence and enhancing fine detail during denoising. We conduct extensive experiments demonstrating UltraDiffEdit’s superior editing quality and flexibility: it can handle image resolutions up to 8K using only a single NVIDIA GeForce RTX 3090 GPU. The source code is publicly available at this https URL.
[CV-53] AEGIS: A Mechanism-Guided Defense against Visual Synonym Jailbreaks in Text-to-Image Models
链接: https://arxiv.org/abs/2607.06120
作者: Yuanmin Huang,Zhenfei Zhang,Mi Zhang,Geng Hong,Qinqin He,Jialing Tao,Hui Xue,Min Yang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Text-to-image diffusion models have achieved high visual fidelity and broad adoption, but remain vulnerable to safety violations when adversaries exploit them to synthesize illicit content. Existing alignment paradigms, from input sanitization to structural feature pruning, are largely organized around unsafe concepts explicitly exposed during filtering, editing, or localization. This leaves a blind spot for visual synonym attacks (VSA), a jailbreak where benign-looking prompts elicit prohibited imagery through implicit visual associations. As a result, current defenses face a safety-utility dilemma: they may either under-mitigate VSA threats or over-suppress visually similar benign concepts. The core challenge is that VSA hides the unsafe target at the textual surface while revealing it through generation-time visual-semantic convergence. In this work, we therefore shift from static suppression of pre-specified unsafe concepts to dynamic tracing of how unsafe semantics emerge during generation. Our mechanistic analysis shows that VSA and explicit unsafe prompts converge through sparse semantic-injecting attention heads, which serve as inference-time bottlenecks for prohibited visual semantics. Based on this insight, we propose AEGIS (Adaptive Evasion Guard via Identification and Steering), an inference-time defense that applies similarity-aware repulsion only at the identified vulnerable heads. Evaluated against 16 baselines, AEGIS improves both safety and utility. On SD 1.4, it reduces ASR to \mathbf0.00/\mathbf0.03 for in-domain violence/nudity VSA and achieves ASRs \le \mathbf0.09 on out-of-domain explicit and adversarial attacks. It preserves benign fidelity, avoids suppressing hard-negative concepts, and transfers to SD 2.1 and FLUX.1 after re-identifying the critical heads for each backbone.
[CV-54] WebRetriever: A Large-Scale Comprehensive Benchmark for Efficient Web Agent Evaluation
链接: https://arxiv.org/abs/2607.06118
作者: Wei Dong,Tianyu Fu,Zhe Yu,Hanning Wang,Anyang Su,Zhizhou Fang,Yuyang Chen,Shuo Wang,Minghui Wu,Ping Jiang,Zhen Lei,Chenxu Zhao
类目: Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM)
备注:
Abstract:As web agents increasingly demonstrate capabilities in automated task execution, the development of robust evaluation frameworks for assessing their navigation and task completion performance has emerged as a critical research priority. However, existing benchmarks exhibit fundamental limitations. First, they suffer from insufficient scale and limited domain diversity, constraining comprehensive evaluation of cross-domain generalization. Second, prevailing LLM-as-Judge evaluation methodologies inadequately capture fine-grained interaction semantics, particularly regarding precise query formulation and filtering operations. Third, current benchmarks predominantly emphasize navigation success metrics while neglecting critical requirements for real-world deployment scenarios. To address these limitations, we introduce WebRetriever, a large-scale benchmark encompassing 800 websites and 1,550 tasks across diverse domains, including consumer, professional, and enterprise sectors, with comprehensive coverage of user intent patterns. We propose NavEval (Navigation Evaluation), a novel LLM-as-Judge framework that leverages rich interaction context beyond visual screenshots, achieving state-of-the-art alignment with human judgment across multiple evaluation datasets. Furthermore, we establish three complementary evaluation protocols that collectively provide holistic assessment of web agent capabilities: navigation proficiency, knowledge-assisted interaction, and end-to-end task completion with information extraction. Extensive experimental analysis reveals substantial performance disparities across evaluation protocols, demonstrating that navigation success alone is an insufficient predictor of real-world application effectiveness. WebRetriever delivers fine-grained diagnostic insights into agent capabilities and establishes a rigorous foundation for advancing web agent research and development.
[CV-55] RoME: Robust Mixture of Low-Rank Experts against Multiple Adversarial Perturbations ECCV2026
链接: https://arxiv.org/abs/2607.06109
作者: Woo Jae Kim,Kyle Min,Suhyeon Ha,Joonsung Jeon,Sung-eui Yoon
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: ECCV 2026
Abstract:Multi-perturbation adversarial training (MAT) aims to achieve robustness against multiple \ell_p perturbations but suffers from robustness trade-offs between different threats. To address this, we employ a mixture of experts (MoE) to route different threats through distinct model pathways. However, naive application of MoE encounters two critical challenges: experts tend to overlook threat-specific features and redundantly capture features shared across threats, and gating networks suffer from threat-agnostic routing where they learn nearly identical routing patterns across threats, thus preventing the construction of threat-specific model pathways. To this end, we propose Robust Mixture of Low-Rank Experts (RoME), where each expert is a low-rank additive update to the shared backbone, allowing it to capture threat-common features while experts focus on threat-specific information. To address threat-agnostic routing, RoME introduces (i) dual-scale gating that exploits threat-discriminative signals from local and global level features, and (ii) threat-guided gating diversification that enforces diverse expert utilization across threats. Extensive experiments demonstrate that RoME outperforms existing state-of-the-art MAT in union robustness and natural accuracy and improves robustness against unseen threats. Codes are available at this https URL.
[CV-56] EcoVision: AI-Powered Drone Imaging for Salt Marsh Vegetation Monitoring and Dominance Mapping
链接: https://arxiv.org/abs/2607.06105
作者: Innocent Onyenonachi,Peter J. Lawerance,Nadia Kanwal
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 37 pages, 8 Figure, 6 Tables
Abstract:High-resolution RGB imagery acquired from low-altitude UAV surveys was processed through a modular pipeline incorporating transformer-based semantic segmentation, connected-component vegetation extraction, fine-grained species classification using a ConvNeXt architecture, and grid-based dominance scoring at 2x2m resolution. The framework targeted two ecologically significant halophytic grasses, Spartina maritima and Puccinellia maritima, and was trained using a curated and manually annotated UAV imagery, along with biodiversity imagery sourced from publicly accessible datasets. In order to identify these plants from the imagery, our segmentation yielded reliable species masks (mean IoU = 0.56; pixel-level accuracy = 0.96), while object-level classification achieved very good discrimination (F1 = 0.99). Dominance estimates closely matched quadrat-based field surveys, with mean absolute differences below 8%, preserving fine-scale spatial structure under realistic survey conditions. The developed system, named EcoVision, establishes a practical foundation for scalable, high-resolution salt marsh monitoring, demonstrating how AI-driven workflows can translate pixel-level predictions into ecologically interpretable metrics.
[CV-57] PVCap: Towards Accurate 3D Dense Captioning via PseudoCap and VoxelCapNet
链接: https://arxiv.org/abs/2607.06097
作者: Xiaopei Wu,Chenshu Hou,Liang Peng,Dan Xu,Binbin Lin,Xiaoshui Huang,Yuenan Hou,Yu Li,Wenxiao Wang,Haifeng Liu,Deng Cai,Wanli Ouyang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 13 pages
Abstract:3D dense captioning, an emerging vision-language task, aims to generate descriptive sentences for each object in the 3D scene. Despite the impressive results achieved by previous methods, they suffer from two limitations. First, current research often employs global rigid transformations, such as rotation, to augment scenes without changing their spatial layouts. However, diverse spatial layouts are crucial for training a 3D dense captioning model to describe spatial relations between objects. Second, previous works mainly focus on the design of the caption generation pipeline while utilizing a simple network architecture for other components, i.e., backbone and detection head, which is crucial for extracting rich semantic information for captioning. In this paper, we propose PVCap to alleviate the aforementioned problems. Our PVCap consists of PseudoCap and VoxelCapNet. Specifically, PseudoCap employs a random mixing technique on instances within the dataset, generating numerous pseudo frames with diverse spatial layouts at the instance level. By utilizing a teacher-student framework, PseudoCap obtains pseudo caption labels for these pseudo frames. This data augmentation approach significantly increases the number of training samples and enhances the model’s ability to describe the environment effectively. Regarding VoxelCapNet, we introduce a robust caption network that utilizes voxel features and adapts the caption head to the voxel-based network architecture. Our VoxelCapNet can serve as a competitive baseline for future research on 3D dense captioning. Extensive experiments are conducted on two prevalent benchmarks, i.e., ScanRefer and Nr3D. Notably, our method surpasses current state-of-the-art by 11.41% and 13.99% in CIDEr@0.5IoU, respectively. Codes will be made publicly available.
[CV-58] MSA-DCNN: A Data-Efficient Multi-Scale Deformable CNN for Medical Image Classification
链接: https://arxiv.org/abs/2607.06083
作者: Hamza Hussaini,Shahana Bano,Eyad Elyan,Carlos Francisco Moreno-García
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Existing deep learning methods perform well in medical image classification but struggle with multi-scale morphology and limited annotations due to fixed sampling and data-hungry training. Existing approaches address these challenges in isolation: DCN-based models provide adaptive sampling but lack explicit multi-scale attention fusion and label-efficient regularisation; multi-scale architectures typically rely on static fusion; and semi-supervised methods target label scarcity without jointly modelling adaptive cross-scale representations. We propose MSA-DCNN, a scale-consistent deformable attention learning framework that introduces adaptive multi-scale sampling, within-scale saliency refinement, learned cross-scale fusion, and auxiliary self-distillation within a unified optimisation scheme, with potential to generalise to structurally heterogeneous anatomy. We evaluate on three public benchmarks and an external hold-out set for leukaemia. MSA-DCNN demonstrates competitive and often better performance against ViT baselines, CNN baselines, and a MICCAI semi-supervised baseline under distribution shift and label scarcity in accuracy, F1, and AUC (binary), while using fewer parameters. Ablations confirm complementary component contributions, supporting MSA-DCNN as a practical foundation for data-efficient medical image classification.
[CV-59] Why does Deep Learning Improve Visual SLAM?
链接: https://arxiv.org/abs/2607.06023
作者: Giovanni Cioffi,Davide Scaramuzza
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注:
Abstract:Visual SLAM is a well-established technology utilized in a wide range of real-world applications. However, its performance still degrades under challenging visual conditions, such as low texture, severe motion blur, and poor illumination. Systems based on deep learning outperform classical geometry-based ones and achieve state-of-the-art results by combining learned 2D data association and uncertainty with differentiable geometric optimization in recurrent architectures. Still, it remains unclear exactly which components are fundamentally responsible for this success. In this paper, we ask: Is the superior performance of deep learning-based systems driven primarily by learned 2D data association, the combination of learned 2D data association and uncertainty, or the recurrent architecture itself? We investigate this question empirically by conducting a controlled study. Our findings reveal that the success of DL-based V-SLAM systems hinges on learned 2D data association and uncertainty rather than their recurrent architecture, underscoring the necessity of learning-based paradigms for the design of these components. Upon acceptance, the code will be released as open source.
[CV-60] KOAL: Knowledge-Driven Prostate Cancer Grading with Ordinal-Aware Learning MICCAI2026
链接: https://arxiv.org/abs/2607.06019
作者: Zheng Guo,Jiaqi Cui,Haocheng Xiong,Jize Han,Bo Liu,Qianwen Zhang,Rui Chen,Yan Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 10 pages, 2 figures, 2 tables. Accepted at MICCAI 2026. This is the submitted version prior to peer review. The final authenticated version will be available on SpringerLink
Abstract:Non-invasive prediction of Gleason Grade Group (GGG) in prostate cancer using multiparametric MRI (mpMRI) is clinically vital for reducing unnecessary biopsies. Existing GGG prediction methods face two major limitations. First, they often overlook non-image information critical for GGG prediction, including age, prostate-specific antigen (PSA), and expert priors embedded in radiology reports. Second, they tend to oversimplify GGG as flat categorical labels, failing to account for its intrinsic hierarchy of primary and secondary Gleason patterns. To this end, we propose a novel Knowledge-Driven Ordinal-Aware Learning (KOAL) framework with three synergistic modules. Specifically, the Clinical-Context Modulation (CCM) module uses clinical variables (e.g., age and PSA) to dynamically modulate discriminative image representations. The Knowledge-Guided Prototype Alignment (KGPA) module leverages an LLM to extract group-specific expert knowledge from training radiology reports and clinical guidelines, producing offline semantic anchors describing grade-specific radiological findings without requiring patient-specific reports at inference. Through prototype contrastive alignment, patient-specific mpMRI representations are matched with these anchors to promote pathology-aligned representation learning. The Hierarchical Ordinal-aware Constraints (HOC) module decouples primary and secondary Gleason pattern prediction and maps their probabilistic outputs to GGG via a Differentiable Bio-logic Mapping Layer (DBML), ensuring pathological grading consistency. Experiments on public PI-CAI and in-house datasets demonstrate that KOAL outperforms state-of-the-art methods. Code is available at: this https URL.
[CV-61] Structured Data Extraction from Real Estate Documents using Clustering Classification and Large Language Models
链接: https://arxiv.org/abs/2607.06012
作者: Muhammad Assad Shehbaz,Carlos Francisco Moreno-García
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Real estate property listings expose structured metadata through the API. Still, the richest property-level information (i.e., legal status, structural condition, utility supplies, heating systems) sits in attached questionnaire documents that no automated system currently processes at scale. These documents are heterogeneous. Some are digitally generated with selectable text, others are scanned physical forms. There are even more complex layouts that contain checkbox annotations that defeat conventional text extraction. In this paper, we present an end-to-end pipeline for acquiring, classifying, and extracting structured data from selectable text documents. The pipeline was applied to 3965 questionnaire documents collected from a live property platform via reverse-engineered REST APIs. First, we classified each document into one of three structural categories (text_only, scanned, and special_char), then extracted 35 predefined property attributes from eligible documents using DeepSeek R1 as the Large Language Model, prompted to return a structured JSON object. All 2781 submitted documents were processed successfully, producing a final dataset of 2766 unique property records. Downstream validation confirmed the data quality. Cosine similarity matching achieves a Jaccard consistency score of 0.82, and K-Means clustering produces interpretable market segments with a silhouette score of 0.2088. Results show that the proposed extraction from each property document is both feasible and reliable at this scale.
[CV-62] OBBSeg: Irregular Lesion Segmentation under Oriented Bounding Box Annotations ECCV2026
链接: https://arxiv.org/abs/2607.06007
作者: Jun Wei,Xinchang Liu,Yu Liu,Chuhua Yang,Shuhui Wang,Hui Huang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 18 pages, 7 figures. ECCV 2026
Abstract:Pixel-level annotation remains a major bottleneck in medical image segmentation, making weak supervision an attractive yet under-constrained alternative. We propose OBBSeg, an intermediate supervision paradigm guided by Oriented Bounding Boxes (OBBs) that bridges the gap between full and weak supervision. By jointly encoding spatial extent and orientation, OBBs provide compact geometric supervision that better aligns with elongated or anisotropic lesions, reducing the ambiguity of coarse box annotations. To mitigate the inherent rectangular bias of OBBs, we introduce a Mask-to-OBB loss, a differentiable formulation that enforces geometric consistency between predicted masks and OBB regions. Furthermore, we incorporate prompt-driven semantic guidance through two complementary modules-PAFE and DBFE-which enhance foreground representation and suppress background interference. Extensive experiments on 13 datasets across 5 imaging modalities show that OBBSeg not only outperforms existing weakly supervised methods but also achieves performance comparable to fully supervised approaches, demonstrating its potential for efficient and scalable medical image segmentation. The code is available at this https URL.
[CV-63] Unlearnable Faces: Privacy Protection Surviving Extraction Pipeline
链接: https://arxiv.org/abs/2607.05996
作者: Byunghoon Oh,Sunghwan Park,Jaewoo Lee
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: preprint
Abstract:Unlearnable examples keep publicly shared photos from being learned by unauthorized face-recognition models. An imperceptible perturbation, added before sharing, makes any model trained on the protected photos fail on clean faces. The perturbation is crafted on the shared image, however the attacker trains on the face it extracts, cropped and resized to the recognizer input, and under this extraction the protection collapses. We propose LPID, which builds the extraction into the unlearnable-example objective. LPID confines the perturbation to the extracted face region and optimizes it through a differentiable model of the extraction, concentrating its energy in the frequency band the extraction preserves. Because this robustness is a property of the transform rather than of any identity, LPID is re-optimized per album and protects even users it has never seen. LPID attains the lowest attacker accuracy of all methods in every setting we evaluate, holding the attacker below 10% under crop+resize extraction on identities unseen at protection time, while remaining imperceptible at 32.7 ,dB PSNR and 0.161 LPIPS.
[CV-64] SparseCtrl-HOI: Sparse Temporal Control for Human-Object Interaction Video Generation ECCV2026
链接: https://arxiv.org/abs/2607.05994
作者: Shenbo Xie,Mingrui Cai,Xu Yang,Yifei Liu,Changxing Ding
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: ECCV 2026, Project Page: this https URL
Abstract:Human-Object Interaction (HOI) video generation aims to synthesize realistic videos of humans manipulating diverse objects, serving as a promising avenue for AI-driven live streaming e-commerce. A primary obstacle in this domain lies in the complexity of modeling fine-grained physical dynamics and the intricate spatial-temporal coordination between human hands and objects. Existing approaches to this problem typically rely on dense temporal guidance, e.g., frame-wise hand-object pose sequences, to strictly control the interaction process. However, such dense guidance incurs high annotation costs and affects motion synthesis diversity. To overcome these limitations, we introduce SparseCtrl-HOI, a novel sparse temporal control framework for HOI video generation. It requires only a few keyframes that capture interaction states at designated timestamps. Specifically, we employ a Time-Controlled Rotary Positional Embedding (TiRoPE) mechanism to temporally anchor these keyframes while preserving their spatial integrity. Subsequently, to govern the dynamics across intermediate frames, we propose a Motion Prior Injection Module that leverages Multimodal Large Language Models (MLLMs) to extract high-level motion priors. This empowers the model to hallucinate logically and physically plausible transitions. Furthermore, we build SparseHOI-5K, a high-quality and richly annotated dataset for HOI video generation with sparse temporal control. Comprehensive evaluations confirm that our method substantially reduces annotation overhead while synthesizing superior live-streaming e-commerce videos. Both our code and dataset are publicly available at this https URL.
[CV-65] SpecTrack: Spectral Prompt Guided Adaptive Experts for Multispectral Object Tracking
链接: https://arxiv.org/abs/2607.05988
作者: Xingyu Tan,Yunrong Qin,Mengjie Hu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 16 pages
Abstract:Multispectral image(MSI) and hyperspectral image(HSI) object tracking object tracking exploits recorded band-wise observations to improve target–background discrimination under similar RGB appearance, mixed pixels, illumination variation, occlusion, and clutter. However, existing trackers commonly process all search regions through a fixed capacity spectral–spatial path, ignoring that tracking difficulty varies substantially across frames and target states. Clear regions may require only lightweight local discrimination, whereas ambiguous boundaries and spectrally similar distractors often demand stronger contextual reasoning. To address this limitation, we propose SpecTrack, a spectral–spatial complexity-aware tracker that formulates MSI tracking as search-region-level adaptive capacity allocation. Its core component, the Spectral Adaptive Mixture-of-Experts (SAMoE) module, provides a capacity-ordered expert pool with progressively increasing latent rank, receptive field, and depth. Expert selection is guided by a Spectral Prompt Router, which fuses semantic context, spatial boundary cues, and a latent channel-variation cue computed after multispectral patch embedding to activate a sparse subset of SAMoE experts for each search region. In parallel, a Shared Global Expert supplies common latent spectral–spatial context to reduce fragmented sparse-routing decisions. Experiments on MUST, MSITrack, and HOTC20 demonstrate a favorable accuracy–efficiency trade-off. The accuracy-oriented SpecTrack-L384 achieves state-of-the-art or highly competitive AUCs of 65.2%, 51.9%, and 72.6% on the three benchmarks, while the balanced SpecTrack-B224 reaches 62.4% AUC at 43.7 FPS on MUST. An additional GOT-10k evaluation indicates RGB-domain architectural generalization, with SpecTrack-L384 achieving 79.3% AO.
[CV-66] Propose and Attend: Training-free MLLM Grounding Confidence via Multi-Token Localized Attention
链接: https://arxiv.org/abs/2607.05978
作者: Daniel Shalam,Emanuel Ben Baruch,Avi Ben Cohen,Tal Remez
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Multimodal large language models can emit localized predictions, bounding boxes for objects and temporal windows for video and audio events, but they hallucinate these regions prolifically. The model’s own token log-probabilities are nearly uninformative: they conflate grounding quality with input ambiguity, and coordinate tokens become near-deterministic once the model commits. We propose Multi-Token Localized Attention (MTLA): a training-free, post-hoc score that measures how strongly a prediction’s tokens attend to the region they claim. Prior attention-based detectors, which sum attention over the entire input modality and read a single response token, are weaker special cases; we show that summing only within the claimed region and aggregating across all prediction tokens recovers a stronger grounding signal. The same recipe applies almost trivially to other modalities and tasks: object detection in images and temporal localization in video and audio. Across multiple MLLM families and three modalities, MTLA improves hallucination AUROC by +7 to +38 over the best prior training-free baseline. Used as a confidence score for re-ranking, it nearly doubles the zero-shot COCO detection AP of an open-source 8B generalist (from 20.4 to 37.0), narrowing the gap to supervised detectors without any task-specific training.
[CV-67] Decoupled Single-Mask Annotation Noise Detection via Cross-Sectional Patch Self-Consistency MICCAI2026
链接: https://arxiv.org/abs/2607.05965
作者: Yinheng Zhu,Xiaowei Xu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 13 pages, 6 figures. Accepted by MICCAI 2026
Abstract:Vascular computed tomography datasets are commonly annotated only once per scan, yielding the pervasive yet under addressed problem of single mask annotation noise. Existing solutions either require costly multirater fusion or are coupled with network training, preventing explicit auditing of where and why labels fail. We introduce a decoupled framework for single-mask annotation noise detection that leverages cross-sectional patch self-consistency to produce interpretable and auditable noise evidence. Tubular anatomy exhibits strong cross-sectional recurrence: patches extracted orthogonally along vessel centrelines recur in appearance across locations and subjects. Thus, anatomically similar patches should have consistent masks, and disagreement signals unreliable annotation. Our method samples cross-sectional patches, retrieves intensity-equivalent neighbours via scalable vector search, and computes a patch-level noise score from statistical mask disagreement, yielding explicit image-mask evidence for every flagged region. Aggregating scores produces scan-level quality maps for dataset quality assessment or quality-weighted training. Experiments on the coronary CT dataset validate the detected noise for improving training robustness and reveal systematic annotation biases. Specifically, transverse and oblique vessels exhibit 5.1 times higher error rates than axis-aligned structures, with additional correlations to cross-sectional area and intensity. Code is available here.
[CV-68] NegROI: Click-Centric Uncertainty-Guided Refinement with Scene-Conditioned Negative Prompts for Robust Interactive 3D Segmentation
链接: https://arxiv.org/abs/2607.05955
作者: Shuheng Zhang,Feng Wu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Interactive 3D segmentation aims to extract object masks in point clouds with minimal user clicks. Despite recent progress, most existing approaches still struggle with (i) coarse voxel resolution that blurs fine boundaries under limited clicks and (ii) hard false positives caused by confusing background structures. These issues are exacerbated by density and scale shifts across datasets (e.g., dense RGB-D reconstructions vs. sparse LiDAR scans), where fixed refinement heuristics and purely click-driven decoding generalize poorly. To address them, we propose NegROI – a novel transformer-based interactive framework that couples click-centric multi-resolution refinement with scene-conditioned negative prompts. Given a coarse voxel prediction, it refines only a local Region Of Interest (ROI) around the current click on a finer grid and fuses refined logits back to the coarse mask. To improve robustness and efficiency, we introduce uncertainty-driven selective refinement that prioritizes ambiguous regions. Meanwhile, we model hard background patterns via a set of scene-conditioned negative prompts obtained by cross-attention over scene tokens. We further stabilize these prompts with a diversity regularizer. Finally, we propose boundary-aware hard negative mining to supervise negative-prompt attention toward boundary-proximal, high-confidence false positives. Our experiments on common benchmark datasets (i.e., ScanNet, S3DIS, and KITTI) demonstrate improved click efficiency and reduced false positives, with stronger cross-dataset robustness than the state-of-the-art baselines.
[CV-69] Progressive Reasoning with Primitive Correction for Compositional Zero-Shot Learning
链接: https://arxiv.org/abs/2607.05911
作者: Ziyi Chen,Haoyan Shi,Sunhan Xu,Congyan Lang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Compositional Zero-Shot Learning (CZSL) aims to combine known attributes and objects as primitives for recognizing previously unseen attribute-object pairs. Prior works either predict attributes and objects independently, missing their strong contextual dependency, or use unidirectional conditional modeling (e.g., object-guided attribute prediction), which is prone to error propagation. We propose PRPC, a Progressive Reasoning framework with Primitive Correction, which explicitly models the bidirectional dependency between attributes and objects via step-wise inference. PRPC performs mutual correction of primitives to suppress prediction errors in earlier steps. Specifically, we formulate CZSL as structured, QA-style Chain-of-Thought reasoning process and constrain the MLLM to follow predefined semantic steps to generate intermediate decisions. To further enhance the reliability and logical consistency of intermediate reasoning, we introduce reinforcement learning post-training with a GRPO-based objective, providing step-level rewards aligned with the progressive inference procedure. Extensive experiments on three CZSL benchmarks demonstrate that PRPC achieves state-of-the-art performance, validating the effectiveness of progressive reasoning and bidirectional correction for robust compositional generalization.
[CV-70] GaussFusion: Towards Multimodal 3D Gaussian Pretraining
链接: https://arxiv.org/abs/2607.05906
作者: Zhixuan You,Jihua Zhu,Yiding Sun,Zihao Guo,Haozhe Cheng,Dongxu Zhang,Lin Chen,Hainan Luo
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 32 pages, 6 figures, 6 tables
Abstract:3D Gaussian Splatting provides an explicit representation that jointly models geometry and appearance, serving as a scalable foundation for 3D representation learning. Existing pre-training methods for Gaussian representations, such as masked Gaussian reconstruction, primarily capture local structures but offer limited semantic supervision. In this paper, we propose GaussFusion, a multimodal pre-training framework for 3D Gaussian representations. GaussFusion integrates image and text supervision into masked Gaussian modeling through cross-modal semantic alignment, enabling the Gaussian encoder to learn both visual and language-level semantic information during pre-training. To better adapt masked modeling to the non-uniform distribution of Gaussian primitives, we further propose Gaussian Salience-guided Multi-scale Hole Masking (GSHM). GSHM constructs spatially continuous masked regions based on Gaussian salience. By applying hole masks at multiple scales, GSHM encourages the encoder to capture both fine-grained local patterns and broader structural dependencies. Extensive experiments on downstream tasks demonstrate that GaussFusion improves the transferability of Gaussian representations. Notably, GaussFusion outperforms Gaussian-MAE on ModelNet40 and ScanObjectNN (PB-T50-RS) by 0.61% and 3.85%, respectively.
[CV-71] Few-Medoids: An Embarrassingly Simple Coreset Selection Method for Few-Shot Knowledge Distillation
链接: https://arxiv.org/abs/2607.05891
作者: Cemil-Andrei Dilmac,Florinel-Alin Croitoru,Radu Tudor Ionescu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted at KES 2026
Abstract:Coreset selection aims to identify a small and highly representative subset of a massive dataset for efficient model training. The problem remains challenging even in the few-shot knowledge distillation (KD) setup, where a full-scale pre-trained teacher informs the student network. Typical sample selection strategies often struggle to surpass the random selection baseline. In this paper, we showcase few-medoids, an embarrassingly simple coreset selection strategy that chooses the samples closest to the centroid (average image) of each class. We present extensive KD experiments on four datasets, covering a wide range of image classification problems, and three teacher-student model pairs, comprising both convolutional and transformer networks. Although the proposed method is embarrassingly simple, our empirical results indicate that few-medoids is able to consistently surpass the random selection baseline, as well as the other coreset selection strategies. We therefore consider that few-medoids can be used as a drop-in replacement for commonly-used baselines (e.g. herding or k-center Greedy), in future research on coreset selection. To reproduce the reported results, we publicly release our code at this https URL.
[CV-72] Harrison.Rad 1.5 Technical Report: A radiology foundation model that can draft reports from images priors and clinical context
链接: https://arxiv.org/abs/2607.05880
作者: Suneeta Mall,Vladimir Nekrasov,Ashnil Kumar,Sajith Karunasena,Aiden Nibali,Alix Bird,Mateo Diaz Shine,Jarrel Seah
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Imaging demand is growing faster than the radiology workforce can expand, and reporting backlogs cannot be resolved through training and recruitment alone. The most direct opportunity is reducing the time and effort radiologists spend producing reports, a task that requires interpreting images, integrating clinical history and prior studies, and drafting structured findings. We present this http URL 1.5 (HR1.5), a radiology-specific multimodal large language model that accepts interleaved text and visual inputs and generates structured and unstructured text across plain-film radiology, spanning computed radiography, chest, musculoskeletal, abdominal, spine, and pelvic x-rays, and mammography. HR1.5 is trained through a three-stage pipeline: domain adaptation of a base language model on radiology reports, contrastive vision-encoder training with curriculum-based hard negatives on ~6 million image-report instances, and visual-question-answering fine-tuning on multi-turn conversations. We evaluate it with a Findings-Diagnosis scoring framework that extends RadGraph-XL entity extraction with ontology-based synonym matching and polarity-contradiction detection, benchmarked on RadBench, a simulated FRCR 2B Short Case examination scored against Angoff-method thresholds, ReXGradient, and internal multi-modality datasets. HR1.5 is the only system evaluated to meet the simulated FRCR passing standard and achieves the highest accuracy on closed-format clinical questions, across anatomical regions, on internal multi-body-part and mammography reporting, and on the primary clinically-aligned score for public chest reporting. We further examine explainability and model behaviour, including question-sensitive Grad-CAM heatmaps, attention analysis, and confidence estimation, to support responsible future evaluation toward clinical use, and a framework for clinically grounded assessment of report quality.
[CV-73] GraspIT: A Dataset Bridging the Sim-to-Real gap and back for Validated Grasping SE(3) Pose Generation
链接: https://arxiv.org/abs/2607.05869
作者: Paul Koch. Adem Karakurt,André Sers
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: Preprint, release soon
Abstract:Robust robotic grasping of novel objects requires datasets that simultaneously provide photorealistic RGB-D observations, physically validated grasp quality annotations, and a principled bridge between simulation and the real world, which existing datasets lack to provide jointly. \textbfGraspIT addresses this gap: tabletop scenes in NVIDIA Isaac Sim are annotated via a four-stage physical slip-test on parallel Franka Panda instances, producing trajectory-reachability checks and continuous quality scores beyond this http URL \sim 2.3M candidates, 83% pass as \emphgood ( s\geq0.50 ); the 17% that passed force-closure but failed the slip-test provide graded hard negatives. A Real \leftrightarrow Sim loop back-projects these labels onto 100 real-world scenes. The release provides \sim 316k annotated RGBD frame sets across 1035 sim and 100 real scenes, with instance masks, 6-DoF poses, physical object properties, and scored 6-DoF grasps. All tools are open-source and Docker-containerized. The trajectory planning within Isaac Sim further allows streaming of high resolution demonstrations for tabletop manipulation policy learning and behavior cloning.
[CV-74] AVA-VLM: Adaptive Visual Attention-Vision Language Model for In-the-Wild Construction Site Monitoring
链接: https://arxiv.org/abs/2607.05859
作者: Younggun Kim,Taeheon Kim,Youngseo Kim,Seunghee Park
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Vision-Language Models (VLMs) are promising for construction-site monitoring, and recent construction-tailored VLMs have primarily adapted pretrained VLMs through direct QA-style fine-tuning from a single global image. We argue that this direct paradigm remains limited for in-the-wild deployment in terms of operational range, reliability under reduced-resolution inputs, and inference efficiency. To address these challenges, we propose AVA-VLM, an Adaptive Visual Attention-Vision Language Model that follows a human-inspired coarse-to-fine reasoning strategy. AVA-VLM first reasons over a low-resolution global image and selectively requests a high-resolution local crop only when detailed inspection is needed, similar to how a human inspector zooms in on hard-to-see yet important areas. We further introduce a region-aware Chain-of-Thought dataset that teaches the model when to inspect, where to crop, and how to use local evidence. Experiments show that AVA-VLM improves reliability under long-distance and reduced-resolution conditions while substantially reducing visual-token usage.
[CV-75] Breaking Spurious Correlations via Generative Randomization and Cross-Variant Self-Supervised Learning CVPR
链接: https://arxiv.org/abs/2607.05850
作者: Suraj Yadav,Anjaneya Sharma,Siddharth Yadav
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at CVPR Workshop 2026 GCV
Abstract:Deep neural networks trained with Empirical Risk Minimization (ERM) often fail under distribution shifts because they exploit spurious correlations between object labels and background context. Recent generative approaches address this issue by creating counterfactual images with altered contexts, but typically use these samples as standard data augmentation, leaving the model free to retain background-sensitive representations. We propose a two-stage framework that uses generative intervention to explicitly learn background-invariant visual representations. First, we isolate the foreground object using zero-shot segmentation and generate context-shifted variants with a structure-preserving diffusion model, preserving object identity while varying the surrounding environment. We then introduce Cross-Variant Self-Supervised Learning, where variants of the same object under different backgrounds form positive pairs in a contrastive objective. This encourages the encoder to align object-centric representations while suppressing background-specific cues. Then, we fine-tune the pretrained encoder using an ERM warm-up followed by GroupDRO with layer-wise learning rates. Experiments on distribution-shift benchmarks demonstrate best worst-group performance, achieving 92.5% on Waterbirds, 81.7% on MetaShift, and 87.4% on NICO++. Code: this https URL
[CV-76] Realistic Compound-Lens Defocus Blur Synthesis
链接: https://arxiv.org/abs/2607.05837
作者: Yunkyu Lee,Woohyeok Kim,Sunghyun Cho
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: GitHub: this https URL
Abstract:Defocus blur degrades fine image structures and limits visual perception, which can adversely affect downstream vision tasks. Although recent deep learning deblurring methods have achieved strong performance, their effectiveness depends on training data and often degrades across cameras and lenses due to limited optical diversity and realism in existing datasets. In this paper, we propose a pipeline for synthesizing realistic defocus deblurring datasets for diverse compound lenses. It integrates efficient wave-optics PSF computation via Debye CZT propagation, depth-aware defocus rendering with occlusion handling, and blur synthesis in the radiometrically linear space with camera ISP simulation. This unified pipeline enables the scalable generation of photorealistic defocus datasets with diverse lens characteristics. Using our pipeline, we generate CLDefocus, a large-scale synthetic dataset containing lens-diverse defocus image pairs. We further analyze the limitations of real-captured defocus datasets and show that such imperfections can bias full-reference evaluation. Extensive experiments demonstrate that models trained on CLDefocus achieve improved cross-device generalization compared to models trained on existing real and synthetic datasets.
[CV-77] Complementary Roles of Image Classification and Vessel Segmentation in AI-Based Screening for Retinopathy of Prematurity Plus Disease in a Kenyan Preterm Cohort
链接: https://arxiv.org/abs/2607.05825
作者: Fred Mutisya,Oscar Onyango,Sarah Sitati,Syokau Ilovi,Aeesha NJ Malik,Brenda W’mosi,Brian Makini,Jalemba Aluuvala,Josiah Onyango,Rachael Kanguha Mmene,Steven Wanyee
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Background. Retinopathy of prematurity (ROP) is a preventable cause of childhood blindness, with rising burden in low- and middle-income countries where ROP-trained ophthalmologists are scarce. Plus disease, marked by retinal vessel dilation and tortuosity, triggers treatment but is subjective and variable. Automated screening could extend specialist reach, but African evidence remains limited. Methods. We analysed 121 Kenyan preterm infants, covering 237 eyes and 1,635 fundus images graded as No Plus, Pre-Plus or Plus. Vessel annotations from two graders supported segmentation training. Eleven configurations were evaluated for eye-level Plus detection using patient-grouped nested cross-validation, including image classifiers, multiple-instance learning, multi-task segmentation-classification, and segment-then-classify pipelines. Results. Vessel segmentation was feasible, achieving pooled Dice 0.533, IoU 0.368, sensitivity 0.623 and specificity 0.979 on held-out images. RGB classifiers were highly sensitive but over-referred, while segmentation-coupled models were more specific. Combining approaches improved performance: an OR-based screen achieved the highest sensitivity, an AND-based confirmation achieved the highest specificity, and a probability ensemble gave the best balanced performance, with sensitivity 0.692, specificity 0.914 and balanced accuracy 0.803, outperforming the vision classifier alone. Conclusions. Classification and vessel segmentation are complementary for ROP Plus detection in Kenyan data. Classifiers support sensitive case-finding, while segmentation improves specificity and reduces over-referral. African ROP AI systems should use combined workflows and undergo prospective multi-site validation. Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) Cite as: arXiv:2607.05825 [cs.CV] (or arXiv:2607.05825v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2607.05825 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Fredrick Mutisya [view email] [v1] Tue, 7 Jul 2026 04:40:16 UTC (1,215 KB)
[CV-78] RIG: Trajectory-Rig Decoupled Metric Geometry Learning
链接: https://arxiv.org/abs/2607.05801
作者: Lizhou Liao,Wentao Xu,Handong Wang,Lirong Yang,Shuai Yang,Weiwei Liu,Chang Huang
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注: 9 pages, 3 figures, 8 tables
Abstract:Vision-centric autonomous driving requires accurate metric geometry and ego-motion estimation from synchronized multi-camera observations. Recent visual geometry models show strong performance in pose estimation, depth prediction, and 3D reconstruction, but are not tailored to rigid multi-camera driving systems. They often encode camera poses as entangled representations, in which time-varying ego-motion and static camera-rig geometry are jointly modeled, limiting the utilization of vehicle-side geometric priors. We propose Trajectory-Rig Decoupled Metric Geometry Learning (TRIG), a geometry perception framework for autonomous driving. TRIG factorizes camera poses into ego-trajectory and camera-rig components, enabling separate modeling of ego-motion and static multi-camera topology. We introduce decoupled pose encoding and supervision, which separately constrain trajectory evolution and rig geometry for metric-consistent learning. Moreover, sparse Temporal–Spatial attention separates cross-camera interaction from temporal aggregation, reducing global attention cost while preserving geometric reasoning. Experiments on five autonomous driving benchmarks show that TRIG achieves state-of-the-art performance in pose estimation, metric depth prediction, and 3D reconstruction.
[CV-79] Segmentation before Answering: Pixel Grounding for MLLM Visual Reasoning
链接: https://arxiv.org/abs/2607.05798
作者: Yake Wei,Yuan Wang,Fengyun Rao,Jing Lyu,Di Hu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Recent advancements in Multimodal Large Language Models (MLLMs) have evolved from static perception to interleaved visual-language reasoning, often referred to as ``thinking with images’'. A basic operation in this reasoning process is to zoom in on regions of interest (often represented with bounding boxes) to acquire finer visual details. In this paper, we propose \textbfSegmentation before \textbfAnswering (SegAnswer), which shifts the unit of zoom-in from the popular bounding box to pixel-level segmentation mask. By employing fine-grained masks to isolate the target area from cluttered environments, segmented visual input yields a more precise region of interest, effectively filtering out redundant background and interfering objects. Furthermore, the discrete patches of segmented visual input align more seamlessly with how MLLMs structure visual tokens via positional embeddings. In experiments, we evaluate SegAnswer across diverse benchmarks, including high-resolution perception, general perception, and hallucination. It achieves consistent improvements and also exhibits considerable performance on segmentation tasks, validating its capability for reliable pixel grounding.
[CV-80] DeSeG: Decoupling Semantic Intent and Geometric Constraints for Physically Plausible Human-Scene Interaction
链接: https://arxiv.org/abs/2607.05787
作者: Jiakun Li,Zhe Li,Wenqiang Wu,Zheng Chang,Mingqi Gao,Jinyu Yang,Feng Zheng
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Synthesizing physically plausible human-scene interactions (HSI) remains a critical challenge in computer vision and the development of human avatars. Although recent generative models enable diverse motion synthesis, they suffer from an inductive bias referred to as semantic-geometric entanglement. Because spatial constraints often strongly correlate with specific actions in training data, monolithic models will learn the shortcut bias, aggressively overriding the semantic intent when faced with strict geometric cues. Furthermore, this entanglement exacerbates physical hallucinations, such as body-scene penetrations. To address these limitations, we propose DeSeG, a hierarchical framework that explicitly decouples semantic intent from geometric constraints. First, we introduce a Residual Semantic Planner that encodes textual instructions and canonicalized goal voxels into a compact latent space, enabling fine-grained semantic control independent of spatial trajectories. Second, we propose a physics regularized diffusion executor that incorporates differentiable repulsive potential fields directly into the diffusion objective, enforcing collision-aware motion generation. Extensive experiments on the Lingo dataset demonstrate that DeSeG achieves state-of-the-art performance, reducing mean scene penetration by 47% and improving semantic alignment by 29% over the SOTA baselines.
[CV-81] Benchmarking the Robustness of Autonomous Driving to Environmental Illusions: A Lane Perception Perspective
链接: https://arxiv.org/abs/2607.05783
作者: Tianyuan Zhang,Xianglong Liu,Aishan Liu,Lu Wang,Yitong Zhang,Peng Yue,Mingchuan Zhang,Siyuan Liang,Dacheng Tao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by IEEE TPAMI 2026
Abstract:Environmental illusions (eg., shadows, reflections, and tire marks) are naturally existing yet overlooked phenomena in real-world driving environments. They can disturb visual perception, leading to misinterpretation of the scene and posing serious safety risks to autonomous driving (AD) systems. However, existing researches largely overlook these phenomena, leaving a critical gap. To address this issue, we study AD robustness through the lane perception perspective, a fundamental task supporting core functions like cruise control and lane centering. We focus on two representative models: conventional lane detection (LD) and vision-language model-based systems (ADVLMs). In this work, we introduce the first benchmark, LanEvil++, for evaluating the robustness of lane perception under environmental illusions. LanEvil++ encompasses 14 types of illusions and leverages the CARLA simulator to generate 94 high-fidelity, fully controllable 3D scenes, yielding a dataset of 90,292 annotated images, 1,596 video clips, and 41,855 visual question answering pairs. Extensive evaluations demonstrate that environmental illusions substantially degrade the performance of state-of-the-art LD methods. On average, LD models experience a 5.27% drop in Accuracy and a 10.49% decline in F1-score, while ADVLMs show a 2.03% reduction in GPT-score and a 0.75% drop in Language-score. Among all illusions, shadows emerge as the most disruptive factor, reducing accuracy by up to 7.20%. Furthermore, closed-loop simulations reveal that these illusions can lead to incorrect driving decisions. Complementary real-world case studies highlight safety-critical failures in actual traffic scenes. To enhance robustness, we propose the Multimodal Illusion Defense Approach (MIDA). MIDA achieves substantial gains under challenging conditions, boosting robustness by 4.23% on LD models and 3.82% on ADVLMs.
[CV-82] FORGE: Towards Functional Tool-Use Generalization via Keypoint Trajectory Reasoning
链接: https://arxiv.org/abs/2607.05780
作者: Chuhao Zhou,Liquan Wang,Shuxin Cao,Xiangyu Chen,Yuxuan Hu,Boyu Ma,Animesh Garg,Jianfei Yang
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: 15 pages, 8 figures, 6 tables
Abstract:While humans readily repurpose a book, a stone, or a shoe to drive a nail, robots trained on specific tools fail to transfer the same function to novel ones – a gap we formalize as functional generalization. Such tools share a common functional intent that is visually recognizable, yet this perceptual similarity does not carry over to action space, where each tool demands an entirely different motor pattern. To bridge this gap, we explore intermediate representations including affordance images, human video prompts, and 2D keypoint trajectories, finding that keypoint trajectories best balance functional expressiveness and action groundability. Building on this, we propose FunctiOnal Reasoning and Grounded Execution (FORGE), a two-stage policy that decouples functional reasoning from action execution: predicting generalizable keypoint trajectories from action-free data, then grounding them into robot actions with limited demonstrations. On a seven-tool hitting-function benchmark, FORGE consistently outperforms state-of-the-art methods on unseen tools in both simulation and the real world, achieving over 2X improvement in average success rate.
[CV-83] LEGATO 2: Toward Multimodal Sheet Music Recognition and Understanding
链接: https://arxiv.org/abs/2607.05769
作者: Guang Yang,Brian Siyuan Zheng,Victoria Ebert,Noah A. Smith
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 23 pages. Equal contribution: Guang Yang and Brian Siyuan Zheng
Abstract:We propose a novel pipeline, Legato 2, for extracting symbolic notation and semantic knowledge from images of sheet music. Legato 2 features the first large-scale neural model for optical music recognition (OMR) to operate sequentially on a system-by-system basis, following the horizontal lines of notation as they are read on the page, rather than treating the page as an undifferentiated image, enabling better scaling to arbitrarily long inputs. It is also the first OMR model capable of generating symbolic transcriptions that include embedded textual content, such as titles and annotations. The pipeline combines system-level segmentation with an autoregressive vision-LM to capture both local notation details and score structure. Across multiple datasets, Legato 2 consistently outperforms prior state of the art. We also show that symbolic transcriptions complement visual inputs for frontier language models, improving their interpretation of dense musical documents. Legato 2 establishes new state-of-the-art performance in both OMR and downstream sheet music understanding.
[CV-84] Image2Sim: Scaling Embodied Navigation via Generative Neural Simulator
链接: https://arxiv.org/abs/2607.05765
作者: Zihan Wang,Seungjun Lee,Yinghao Xu,Gim Hee Lee
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注:
Abstract:Embodied navigation aims to build agents that interpret multimodal goals, reason in 3D space, and reach target destinations reliably in the real world. However, progress remains constrained by the lack of scalable, high-fidelity, and physically grounded interactive environments. Although real-world scanned datasets offer visual realism, they are limited by scale. In contrast, synthetic simulators scale more easily but often exhibit large sim-to-real gaps. We introduce Image2Sim, a real-time neural simulation framework that constructs high-quality interactive environments from posed RGB-D image sequences. The central idea is to decouple 3D spatial anchoring from photorealistic observation synthesis. For scene construction, Image2Sim uses a feed-forward feature Gaussian model that lifts posed RGB-D observations into a 3D feature-Gaussian representation in a single pass. For rendering, we propose a Geometry-Aware One-Step Pixel Flow model that transforms sparse and noisy Gaussian projections into high-quality panoramic RGB-D observations. Image2Sim also serves as a fully automated embodied data engine that generates high-fidelity observations, executable actions, and diverse navigation instructions at scale. It converts large collections of videos and images into nearly 20K interactive scenes and synthesizes more than 10 million navigation training samples. Navigation models trained entirely in these neural environments achieve strong improvements on major benchmarks and transfer effectively to real-world zero-shot settings. These results suggest that scalable neural simulation can serve as a practical training substrate for embodied navigation at scale.
[CV-85] Optimized Adaptive Loop Filter in Versatile Video Coding
链接: https://arxiv.org/abs/2607.05737
作者: Meng Xuewei,Zhang Jiaqi,Jia Chuanmin,Zhang Xinfeng,Wang Shanshe,Ma Siwei
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: This paper was submitted to DCC 2021 and accepted as a poster
Abstract:In the Versatile Video Coding~(VVC) standard, adaptive loop filter~(ALF), including Geometry transformation-based Adaptive Loop Filter~(GALF) and Cross Component Adaptive Loop Filter~(CCALF), plays an essential role in reducing compression artifacts. However, it also has high coding complexity and requires many picture buffer accesses in the encoder that will increase external memory access and is unfriendly to the software and hardware design. Therefore, we propose an optimized ALF framework, including the parallel design of GALF and CCALF, the adaptive parameter decision of GALF, and one-pass CCALF scheme by effectively estimating the CCALF filtering distortion without conducting filter operation. Compared to VTM-8.0, the proposed method can reduce the picture buffer access from 152 to 1 and achieve roughly 25% time-savings of the ALF module with negligible coding performance change under RA configuration. Some of the proposed methods have been adopted in the VVC reference software.
[CV-86] ARMS: Anchor-Relational Motion Streaming for Seamless Solo-Social Motion Transitions ECCV2026
链接: https://arxiv.org/abs/2607.05733
作者: Huakun Liu,Qing Yu,Kent Fujiwara,Hideaki Uchiyama,Kiyoshi Kiyokawa
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by ECCV 2026. Project page: this https URL
Abstract:Generating temporally continuous and socially coherent human motion from text remains a fundamental challenge, particularly in realistic streams where people act alone, enter interactions, and later disengage. Most existing methods generate fixed-length motion clips under static agent configurations, which makes them brittle to solo-social transitions and unsuitable for incremental generation over long horizons. We propose ARMS, an Anchor-Relational Motion Streaming framework that unifies solo motion and human-human interaction within a single causal generative process. ARMS introduces a dynamics-asymmetric representation that decouples per-person temporal evolution from inter-person alignment via a partner-referenced relative-translation term, enabling seamless switching of social coupling without sacrificing long-horizon stability or spatial consistency between agents. On top of a causal latent space, a causal relational diffusion model progressively refines motion segment by segment using only past context, capturing both intra-person temporal dependencies and inter-person relations. Mode-aware relational gating activates or masks cross-agent connections, allowing the same model to support both solo and interaction generation. Experiments show that ARMS improves transition smoothness and social coherence compared to interaction-centric baselines, while also achieving competitive results on human-human interaction benchmarks.
[CV-87] SAMPLe: SAM-based Optimizer for Prompt Learning in VLMs ECCV
链接: https://arxiv.org/abs/2607.05727
作者: Hossein Rajoli,Fatemeh Lotfi,Niloufar Alipour Talemi,Hossein Kashiani,Xiaolong Ma,Fatemeh Afghah
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: The manuscript has been accepted to ECCV and will be presented at the conference and published in the main proceedings
Abstract:Pre-trained Vision-Language Models (VLMs) like CLIP have proven highly effective as foundation models for various downstream applications. However, prompt learning in VLMs encounters a performance-generalization dilemma: while prompts can be tuned to achieve high accuracy on seen distributions, this tuning process often undermines their generalizability to unseen data. The limited set of learnable prompts, which contextualize and condition the input to steer it toward the task within the pretrained VLM, tends to overfit the training data, leading to a trade-off between task-specific performance and preserving generalization. To address this dilemma, we introduce SAMPLe (Sharpness-Aware Minimization Prompt Learning), a plug-in sharpness-aware optimizer that enhances prompt generalizability by accounting for loss landscape sharpness. Unlike conventional methods, SAMPLe balances exploration and exploitation by satisfying objective function constraints at each step, dynamically adapting to the current optimization state based on the local curvature and gradient properties. This approach reduces overfitting on seen distributions and improves adaptability to unseen data, preserving the generalization potential of pre-trained VLM models. We integrate SAMPLe into multiple prompt learning frameworks, including CoOp, CoCoOp, MaPLe, TCP, and Co-Prompt, demonstrating its effectiveness across diverse methods. Experiments show that SAMPLe elevates prompt learning frameworks and consistently outperforms existing optimizers across diverse settings, establishing itself as a robust, model-agnostic solution for prompt learning.
[CV-88] Association Restoration Test: Revealing Restorable Shortcuts after Unlearning
链接: https://arxiv.org/abs/2607.05726
作者: Amy Lu,Changxiu Ji
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Preprint. 16 pages
Abstract:Association unlearning aims to disable learned label-attribute shortcuts while preserving task performance. Existing evaluations mainly measure output-level robustness or probe whether shortcut attributes remain readable in frozen features, but neither test determines whether a retained association remains functionally usable by the original classifier. We propose the Association Restoration Test (ART), a post-hoc diagnostic for functional shortcut restorability. ART estimates class-conditional association directions, amplifies residual components, and evaluates the modified features with the original classifier head. Across Waterbirds, CelebA, SpuCoDogs, and an ISIC timestamp-artifact extension, we show that output metrics, representation probes, and ART characterize distinct aspects of shortcut mitigation. These findings motivate restoration-aware evaluation for unlearning and shortcut-mitigation methods that target learned associations rather than individual classes or concepts.
[CV-89] Scene Graph Thinking: Reinforcing Structured Visual Reasoning for Multimodal Large Language Models ICML2026
链接: https://arxiv.org/abs/2607.05716
作者: Zhiwei Yang,Yuanchen Wu,Nan Zhang,Yucong Meng,Ke Yan,Shouhong Ding
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: ICML 2026
Abstract:Multimodal Large Language Models (MLLMs) have demonstrated strong perception and reasoning capabilities. However, most existing models focus on isolated objects and neglect structured relationships for efficient target navigation, limiting their performance on visually intensive tasks. To address this challenge, we introduce Scene Graph Thinking (SaGe), a novel paradigm that enables fine-grained and structured visual reasoning through explicit scene-graph representations. Specifically, we first introduce an automated data engine that converts flat image-text corpora into structured scene graphs, where hierarchical entities constitute the nodes and diverse visual relations define the edges. Building upon this, we construct 120K high-quality training data by sampling reasoning traces from scene graphs. Then, two-stage graph-aligned post-training paradigms are introduced, where supervised fine-tuning internalizes MLLMs with structured reasoning, and subsequent reinforcement fine-tuning proposes node-as-proxy graph rewards to consolidate efficient graph exploration. With curated data and graph-aligned training, our approach achieves significant improvements across eight multimodal benchmarks, demonstrating strong effectiveness on fine-grained perception and reasoning tasks. Code is available at this https URL.
[CV-90] FourTune: Towards Fully 4-Bit Efficient Post-Training for Diffusion Models
链接: https://arxiv.org/abs/2607.05711
作者: Bowen Xue,Zihan Min,Xingyang Li,Zhekai Zhang,Haocheng Xi,Lvmin Zhang,Maneesh Agrawala,Jun-Yan Zhu,Song Han,Yujun Lin,Muyang Li
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Diffusion models have become a dominant paradigm for high-quality generative modeling, while post-training is essential for adapting them to diverse downstream applications. However, post-training of large diffusion models is still challenging due to the prohibitive memory footprints and slow training speed, which existing parameter-efficient fine-tuning methods only partially address. To overcome these limitations, we propose FourTune, an efficient post-training framework for diffusion models based on an end-to-end W4A4G4 paradigm. FourTune introduces a triple-branch hybrid pipeline that augments the standard LoRA architecture with a frozen numerical stabilizer to isolate quantization-sensitive outliers, enabling stable training under native 4-bit computation. In addition, FourTune employs hardware-efficient block-wise quantization and customized fused kernels to support efficient quantized backpropagation and reduce memory bandwidth overhead. Across customization, reinforcement learning, and distillation tasks, FourTune matches the quality of full-precision fine-tuning. On FLUX.1-dev (12B), FourTune reduces memory overhead by 2.25 \times and increases end-to-end training throughput by 2.27 \times compared to BF16 LoRA.
[CV-91] IMR: Iterative Mode-World Weighted Regression for Multi-Agent Trajectory Prediction
链接: https://arxiv.org/abs/2607.05705
作者: Honglin Wang,Shiyao Pan,Yun-Fu Liu
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Multi-agent motion prediction is essential for automated vehicles to understand the intentions of surrounding vehicles. However, previous prediction-based and anchor-based methods have limitations in mode diversity and prediction accuracy, respectively. These limitations may cause inadequate safety assessments and behavioral deviations in automated vehicles. To address this issue, a mode-world weighted regression loss is proposed to bridge the gap between these features. Specifically, this approach mitigates mode collapse while simultaneously improving world ranking and top-1 confidence. Furthermore, the proposed iterative decoder improves prediction accuracy by recurrently and segmentally generating trajectories. Experimental results show the proposed method ranks first in the Argoverse 2 multi-agent motion forecasting benchmark against other methods.
[CV-92] LLM -Driven Neural Network Generation with Same-Family Architecture Guidance: Disentangling Transfer and Adaptation
链接: https://arxiv.org/abs/2607.05704
作者: Kabir Dev Paul Baghel,Radu Timofte,Dmitry Ignatov
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注: 10 pages, 1 figure, 14 tables
Abstract:Large language models (LLMs) can generate neural-network modifications, but unrestricted generation is often invalid or harmful. This paper studies a narrower setting: improving a weak target model using a stronger same-family source model from a neural-network database. We propose a source-guided candidate-generation protocol with non-source controls, source-conditioned candidates, and a no-LLM hp_copy ablation under equal evaluation budgets. The protocol reports validity separately from accuracy and selects the best valid candidate only when it improves the target. On CIFAR-10, the strongest source-guided candidate reaches 0.5049 accuracy versus 0.2398 for the best non-source candidate, a +0.2651 advantage, while improving a weak target originally at 0.1254; a five-epoch check preserves the gain at 0.7686 versus 0.4839. On SVHN AlexNet with DeepSeek-Coder-6.7B, source-guided transfer reaches 0.7880 versus 0.2254, a +0.5626 advantage; a fresh repeat reaches 0.8069 versus 0.2509, a +0.5560 advantage. Direct source-recipe copy produces 0.1959 on SVHN AlexNet, matching the original target, while hp_transfer reaches 0.7880, showing that the LLM adapts rather than copies the source recipe. Family-level analysis shows the clearest positive signals for AlexNet, with 6/8 wins across SVHN, Imagenette, and CelebA-Gender, and alt_nn1, with 8/10 wins on CIFAR-10.
[CV-93] Robust Face Super-Resolution and Recognition Through Multi-Feature Aggregation in Diffusion Models
链接: https://arxiv.org/abs/2607.05702
作者: Marcelo dos Santos,Rayson Laroca,João Carlos Raposo Neves,David Menotti
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Images acquired in surveillance environments often suffer from conditions such as low resolution, variations in pose, irregular illumination, and occlusions. Due to the low quality of these images, face recognition algorithms often struggle. This major limitation can be addressed by employing super-resolution techniques that enhance the details of the image. However, due to the high degree of difficulty of the problem, most super-resolution algorithms tend to cause distortions in the image and in the individual’s identity. Thus, additional information must be incorporated into the processing to improve recognition robustness. In this regard, surveillance cameras can capture multiple images, even at low quality, and the data extracted from these images, such as consecutive video frames, can significantly enhance both super-resolution and facial recognition. In this work, we introduce FASR++, a diffusion-model-based super-resolution algorithm. It leverages a reference low-resolution image and features extracted from multiple auxiliary low-quality images to generate a super-resolved output, minimizing distortions in the individual’s identity. Our approach recovers facial features without explicitly providing soft attributes or computing a function gradient to guide the reconstruction process. FASR++ generates high-quality images that can considerably improve performance in face recognition tasks when used as a pre-processing step. We validate our approach on two standard face recognition datasets and attain state-of-the-art results for verification, face recognition, and image quality metrics such as PSNR, SSIM, and LPIPS.
[CV-94] Clustered Codebook Quantization for 2D Gaussian-based Image Compression SIGGRAPH2026
链接: https://arxiv.org/abs/2607.05667
作者: Runze Cheng,Yicheng Zhan,Josef Spjut,Kaan Akşit
类目: Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR)
备注: 3 pages. Accepted to ACM SIGGRAPH 2026 Poster Track. Code available at this https URL
Abstract:Gaussian-based image representations effectively model image content using compact parametric primitives while preserving high visual fidelity, yet storing a large number of floating-point parameters per primitive degrades rate-distortion efficiency at higher fidelity targets. To improve the rate-distortion performance in Gaussian representation, we present our Cluster-Guided Vector Quantization (CGVQ), a Gaussian primitive based image compression method. Our key idea is to partition Gaussian parameters further into homogeneous groups prior to quantization, enabling higher compression efficiency and accurate parameter reconstruction. In practice, our extensive experiments show that CGVQ decreases the bpp by 20% with respect to our baseline, while maintaining on-par visual quality
[CV-95] REVIVE: A Multi-Modal Framework for Vandalism Detection and Recovery in Autonomous Vehicles
链接: https://arxiv.org/abs/2607.05649
作者: Abdullah Tariq Choudhry,Tapadhir Das
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Autonomous vehicles (AVs) face increasing threats from vandalism-induced occlusion attacks (VOAs) that compromise camera-based perception. While detection frameworks can identify vandalized images, restoring camera-stream utility after physical occlusion remains underexplored. This paper presents present the Recovery and Enhancement of Vandalized Images for Vision Excellence (REVIVE) framework, a vandalism recovery pipeline integrating: (1) binary VOA detection, (2) multi-class VOA pattern identification, (3) EfficientNet-based U-Net segmentation, and (4) type-aware recovery using Bootstrapping Language-Image Pre-training (BLIP)-guided Stable Diffusion inpainting, direct pixel replacement, or adaptive median filtering. Stable Diffusion shows variable reconstruction performance (per-pattern SSIM 0.667-0.867, PSNR 15.4-26.7dB) across VOA patterns, while aligned direct pixel replacement achieves near-identical reconstruction under the aligned-reference condition. On 500 tracked clean/vandalized image pairs, unrecovered VOAs reduce YOLOv8l object-detection recall to 0.588, while direct pixel replacement restores recall to 0.967 and F1-score to 0.970 under that aligned-reference condition. LaMa, Telea, and Navier-Stokes baselines improve image similarity but provide more limited downstream detection recovery, and Stable Diffusion is treated as an asynchronous recovery branch subject to a quality gate rather than a blocking real-time perception step. We evaluate a reference-available quality gate that filters recovered candidates before downstream use: without it, type-aware routing degrades per-image recall to 0.304, whereas with it, recall returns to 0.608, at or above the unrecovered baseline, ensuring the forwarded stream is never worse than the unrecovered frame. REVIVE therefore, provides a structured recovery framework from VOAs in AVs.
[CV-96] VEIL: How Visual Encoding Hijacking Induces Bias In Vision Models
链接: https://arxiv.org/abs/2607.05641
作者: Suranjana Sooraj,Xuyang Chen,Madhumitha Venkatesan,Dongyu Liu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Rendering time series as chart images for CNN-based classification has become increasingly common in time-series classification (TSC). However, it remains unclear whether models learn underlying temporal patterns or rely on encoding-specific visual cues introduced by chart design. We present VEIL: a systematic study examining how chart encodings influence learned representations through complementary analyses of similarity, transferability, and attribution. Attention-guided training appears to mitigate this effect when encoding sensitivity is consistently identified across diagnostics, but provides limited or negative benefit when such signals are absent. These findings position VEIL within the broader question of how machines perceive visualizations – extending graphical perception from human readers to vision models – and show that visualization design choices shape learned representations in ways that warrant treating chart-based TSC as a representation and measurement problem rather than a simple modeling decision.
[CV-97] Recovering Cloud Microstructures with Cascaded Diffusion Inversion ICLR2026
链接: https://arxiv.org/abs/2607.05637
作者: Hanan Gani,Guy Pulik,Daniel Rosenfeld,Duncan Watson-Parris,Salman Khan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Published at ML4RS Workshop ICLR 2026
Abstract:High-resolution satellite imagery is critical for observing fine-scale cloud structures that inform weather modification strategies like cloud seeding for rain-enhancement. However, the spatial resolution of current geostationary and polar-orbiting satellites is often insufficient for capturing small cloud features. Current super-resolution methodologies are suited for natural images and, therefore, struggle to generalize to satellite-captured spectral images of cloud cover. To address this, we propose a two-stage diffusion-based super-resolution framework to enhance the resolution of multi-spectral cloud microstructures by a factor of 4\times . Specifically, we use inverse diffusion to recover the high resolution properties from low resolution. Stage 1 utilizes real-world paired data to learn robust degradation handling and inter-sensor alignment, while Stage 2 employs a self-supervised internal downgrading of high resolution data to refine structural learning and texture synthesis. Our approach outperforms the state-of-the-art transformer and diffusion-based baselines in both reconstruction accuracy and visual quality. We demonstrate that the two-stage method better captures fine cloud microstructures (e.g. convective turrets and cloud gaps) that are crucial for effective cloud seeding decisions. Ablation studies confirm the complementary benefits of the two stages: Stage 1 excels in coarse structural fidelity, while Stage 2 contributes enhanced detail and realism. These results highlight a practical path toward improving cloud microphysics analysis and as a step towards utilizing AI for climate and sustainability. Our code and models are publicly available at: this https URL.
[CV-98] axlifier: Leverag ing Disease Taxonomy for Enhanced Multi-Label Classification in Chest Radiography
链接: https://arxiv.org/abs/2607.05628
作者: Mohammad S. Majdi,Jeffrey J. Rodriguez
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Accurate and efficient classification of thoracic diseases in chest X-ray (CXR) images is crucial for timely diagnosis and treatment. However, the presence of multiple pathologies with overlapping visual characteristics poses significant challenges for automated classification systems. In this study, we propose two novel hierarchical multi-label classification techniques, namely the loss-based and logit-based methods, to address these challenges by leveraging the hierarchical relationships among different thoracic pathologies. The loss-based technique integrates hierarchical information directly into the optimization process, while the logit-based method adjusts the predicted probabilities of each class based on its parent class in the disease taxonomy. We evaluate the performance of both techniques using three large-scale CXR datasets: CheXpert (224,316 CXRs), PADCHEST (160,000 CXRs), and NIH (112,120 CXRs). The experimental results demonstrate significant improvements in accuracy, AUC, and F1 scores compared to the baseline method across various pathologies. The logit-based and loss-based methods improve accuracy by 12% and 11%, AUC by 13% and 10%, and F1 scores by 24% and 12%, respectively compared to the baseline. These results represent a substantial improvement over the baseline method. Furthermore, we conduct a comprehensive statistical analysis to validate the robustness and reliability of the proposed techniques. The integration of domain-specific hierarchical knowledge not only enhances the classification performance but also provides a more interpretable output for clinical decision support. Our findings highlight the potential of hierarchical multi-label classification in advancing computer-aided diagnosis systems for chest radiography.
[CV-99] Cross-Contextual Vision-Language Adaptation with LoRA for Personalized Severe Adverse Event Detection in Clinical Wound Monitoring
链接: https://arxiv.org/abs/2607.05625
作者: Aditi Naiknaware,Jian Sun,Aminreza Khandan,Shengyang Huang,Sean Dow,Bijan Najafi,Salimeh Sekeh
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Wound monitoring is a critical yet underserved clinical challenge, where timely identification of severe adverse events (SAEs) such as infection, tissue deterioration, and delayed healing can significantly impact patient outcomes. While vision-language models (VLMs) show strong multimodal reasoning, they often lack domain-specific grounding to integrate wound imagery with heterogeneous clinical information, and provide limited mechanisms for detecting cases that diverge from the training distribution. We present a multimodal framework for automated wound monitoring and SAE detection. Our approach leverages paired clinical notes and wound descriptions capturing visual characteristics such as appearance, surrounding skin condition, color changes, and signs of inflammation or healing progression, encoded through a dual-stream Low-Rank Adaptation (LoRA) framework built on a frozen BiomedCLIP backbone. We introduce a cross-contextual LoRA fusion mechanism enabling information exchange between clinical semantics and visual wound descriptors, producing context-aware multimodal representations without full model fine-tuning. To identify personalized SAEs, we propose a wound-specific out-of-distribution (OOD) detection framework combining semantic matching, visual typicality, caption-text alignment, and caption-visual alignment into a unified SAE (OOD) score. To capture healing dynamics, we incorporate covariate consistency and temporal drift penalties that leverage changes in wound characteristics across visits. Experiments on a longitudinal wound dataset collected through clinical visits show promising performance on both wound healing assessment and SAE detection, highlighting the potential of semantically enriched, temporally aware vision-language systems for clinical wound monitoring and early risk identification.
[CV-100] Patch Knowledge Transfer for Efficient AI-Generated Image Quality Assessment ICME26
链接: https://arxiv.org/abs/2607.05605
作者: Jiquan Yuan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 13 pages. ICME26 Spotlight
Abstract:With the rapid advancement of image generation technologies, perceptual quality assessment of AI-generated images has emerged as a crucial research direction in computer vision. The core challenge of this task lies in achieving efficient quality assessment for massive generated images. Current mainstream approaches exhibit two key limitations: 1) Methods employing complex feature extraction strategies, while improving performance, incur prohibitive computational costs that hinder real-time inference; 2) Simple image scaling-based solutions, despite their computational efficiency, demonstrate significantly inferior assessment accuracy. To address this critical issue, we propose Patch Knowledge Transfer (PKT), a knowledge distillation-based optimization framework that achieves synergistic optimization of visual representation capability and inference efficiency through an innovative multi-level knowledge transfer mechanism. Specifically, we design a dual-model architecture: a teacher model with local-global hybrid processing provides high-quality supervision signals, while a student model relying solely on global processing efficiently inherits the teacher’s representation capacity through multi-level supervision. Extensive experiments conducted on 4 AIGIQA databases demonstrate that the PKT framework enables the student model to maintain performance comparable to the teacher while reducing computational costs by 67.7%. Furthermore, compared to existing methods, our approach achieves a superior balance between model efficiency and assessment accuracy.
[CV-101] SSA-3DGS: Unsupervised Removal of Screen-Space Artifacts for 3D Gaussian Splatting
链接: https://arxiv.org/abs/2607.05598
作者: Kristof Overdulve,Lode Jorissen,Nick Michiels
类目: Graphics (cs.GR); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Novel View Synthesis (NVS) methods, such as 3D Gaussian Splatting (3DGS), rely heavily on the assumption of clean, multi-view consistent, posed input images. Real-world captures can violate this assumption due to screen-space artifacts-static occlusions fixed to the 2D image plane rather than to the 3D world. Common examples include physical sensor defects, environmental obstructions (such as rain or mud on the lens enclosure), capture obstructions (such as a thumb over the camera sensor or a dashboard visible in dashcam footage), and digital overlays (such as watermarks or UI elements). When present, they are erroneously baked into the 3D geometry as “floaters” or near-camera artifacts, degrading the quality of novel-view rendering. In this work, we propose SSA-3DGS, an unsupervised framework that jointly optimizes a 3D scene and a learnable 2D overlay to recover a clean 3D scene and the corrupting artifacts. By exploiting geometric consensus across views, our method effectively disentangles static artifacts from the 3D scene geometry without supervision or manual input. Across diverse synthetic corruptions and a self-captured real-world dataset, SSA-3DGS improves reconstruction fidelity by up to 9 dB PSNR over 3DGS trained on the same corrupted inputs, while faithfully preserving the corrupting artifact.
[CV-102] Hierarchical Classification via Cascading Feature Elimination: Application to Human Phenotype Ontology-Aligned Facial Phenotyping (FaceMesh2HPO)
链接: https://arxiv.org/abs/2607.05585
作者: Fabio Hellmann,Alexander Hustinx,Benjamin D. Solomon,GestaltMatcher Database Consortium,Tzung-Chien Hsieh,Peter Krawitz,Elisabeth André
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:FaceMesh2HPO is a framework for classifying facial phenotypic descriptors aligned with the Human Phenotype Ontology (HPO) to support clinical diagnosis. Using annotations from 124 clinicians across 10 disorders (107 HPO terms) combined with non-syndromic controls, we generated 3D facial meshes (478 landmarks) from 2D images and trained a hierarchical PointNet-based pipeline with cascading classification and feature elimination. The best models, incorporating 3D meshes, facial outline, and demographic metadata, achieved AUROCs between ~0.55 and ~0.89, with higher performance at parent nodes than leaf terms. External validation showed variable generalizability across disorders. Results demonstrate that hierarchical modeling of 3D facial geometry enables interpretable, ontology-linked phenotype classification, though performance on rare leaf terms remains limited. Improved data diversity and feature selection strategies are needed to enhance robustness and clinical utility.
[CV-103] Harnessing Generative Image Models for Training-Free Primitive Shape Abstraction
链接: https://arxiv.org/abs/2607.05568
作者: Gregor Kobsik,Tim Elsner,Leif Kobbelt
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 13 pages, 9 figures, 3 tables
Abstract:Representing 3D shapes as compact sets of geometric primitives is fundamental to robotics, simulation, and scene understanding. Generative image models trained at scale have recently emerged as generalist visual learners that can identify and segment object parts directly in the image domain, across arbitrary categories and without task-specific training. Adapting such models to downstream tasks typically requires fine-tuning; we ask whether their pretrained capability can instead be harnessed directly, without any training, and answer affirmatively with a training-free harness. Our pipeline renders multi-view images of a 3D object, uses a vision-language model to analyze its semantic parts, prompts a generative image model to paint a color-coded part segmentation mask, reprojects it onto the geometry, and fits a superquadric primitive to each part via parameter optimization. The approach contains no learned parameters: it is category-agnostic and orientation-invariant, properties that previous learning-based models struggled with. Its accuracy ceiling rises with future generative-model improvements, which we confirm with a ground-truth segmentation study showing that part segmentation, not primitive fitting, is the current accuracy bottleneck. On HumanPrim and Toys4K, our method achieves the lowest Chamfer distance among all evaluated methods, using 5–9 primitives per object on average.
[CV-104] GEM-Occ: From Visual Geometry Evidence to Embodied Semantic Occupancy Memory
链接: https://arxiv.org/abs/2607.05543
作者: Hu Zhu,Bohan Li,Xianda Guo,Hongsi Liu,Baorui Peng,Mingqi Yuan,Xin Jin,Wenjun Zeng,Chang Wen Chen
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 19 pages, 6 figures. Project page: this https URL
Abstract:Semantic occupancy provides a structured spatial memory for embodied indoor agents by jointly representing occupied regions, observed free space, unknown areas, and object semantics. However, existing indoor occupancy benchmarks and methods mainly focus on single-view prediction or room-level online perception, leaving long-horizon semantic mapping across connected indoor spaces underexplored. We introduce HIOcc, a hierarchical indoor occupancy benchmark that unifies ScanNet, ScanNet++, and Matterport3D under a common sparse semantic occupancy format while preserving their native observation geometries, including perspective RGB-D frames and pano-centric observation groups. HIOcc supports three complementary evaluation regimes: local semantic occupancy prediction, room-level online occupancy mapping, and building-level mapping across connected panoramic environments. We further propose GEM-Occ, a Gaussian Evidence Memory framework for semantic occupancy mapping. Rather than using pointmaps as persistent map states, GEM-Occ treats local visual geometry predictions as transient evidence, converts them into semantic Gaussian occupancy evidence and free-space ray evidence, and fuses them into a persistent hierarchical memory through visibility- and uncertainty-aware causal updates. The memory is organized into local caches, room-level submaps, and a building-level graph, and can be queried at any time through Gaussian-to-occupancy splatting. Experiments on HIOcc show that GEM-Occ improves local occupancy prediction, online map stability, free-space reasoning, revisit consistency, and building-level scalability over prior indoor occupancy and Gaussian-based mapping baselines.
[CV-105] Multi-Teacher Contrastive Distillation for Edge-Efficient Pathology Foundation Models
链接: https://arxiv.org/abs/2607.05533
作者: Tim Lenz,Maurice Heide,Marco Gustav,Nic G. Reitsam,Jakob Nikolas Kather
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Computational pathology foundation models (PFMs) have advanced whole-slide image analysis. However, their size and inference cost hinder local deployment in pathology departments. We propose MuCoDi, a pretraining framework that distills frozen tile embeddings from multiple PFMs into compact edge-oriented encoders. Instead of regressing individual teacher features, MuCoDi trains lightweight MobileOne and RepViT students with a contrastive distillation objective adapted from MoCo v3, where cached Virchow2, UNI2, and H-Optimus-1 embeddings replace momentum-encoder keys. We pretrain students on 14.3M TCGA tiles from only 11.8K WSIs and evaluate frozen encoders on 23 clinically curated downstream classification tasks. RepViT-based MuCoEdge students retain near-teacher performance while reducing model size by orders of magnitude: MuCoEdge-R2.3 and MuCoEdge-R1.5 reach 71.0% external AUROC, within 0.8 percentage points of the best teacher (Virchow2, 71.8%), while MuCoEdge-R2.3 obtains the best external F1 and the second-best AUPRC (51.8% and 53.3%). MuCoEdge-R1.0 reaches 70.9% AUROC with only 6.4M parameters and 1.12 GFLOPs. On a Raspberry Pi 5, sub-million-parameter MobileOne students achieve up to 605-fold single-tile speedup over Virchow2 while retaining 66.5% to 66.9% external AUROC, demonstrating that PFM-quality pathology representations can be moved toward practical edge deployment. Code is available at this https URL.
[CV-106] mathbfλ-VAE: Variance Equalization for Posterior Collapse
链接: https://arxiv.org/abs/2607.05531
作者: Girum Demisse
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注: 21 total pages
Abstract:Variational Autoencoders (VAEs) frequently suffer from posterior collapse, a failure mode in which the approximate posterior converges to the prior, rendering the latent code uninformative. Despite extensive research, a unified account of why collapse occurs has remained an open question. We identify and formalize two logically independent but coupled causes. \emphGradient imbalance occurs when the decoder’s reconstruction signal vanishes faster than the \mathbbKL regularization pressure as the posterior widens. \emphInformation gap occurs when the stochastic sampling step discards a substantial fraction of the encoder’s computed representation, attenuating decoder sensitivity and making collapse inexpensive. Both causes share the same collapse trajectory, and we show that the information gap is algebraically equivalent to mismatch between the aggregate posterior and the prior, unifying two pathologies. Subsequently, we introduce \lambda -VAE, which resolves both causes through a single modification to the reparameterization step: the sampling noise is scaled by per-dimension exponent, while the \mathbbKL penalty retains the original posterior variance. This asymmetry shifts the stable training attractor away from the degenerate collapsed state, driving all latent dimensions toward the same equilibrium – a mechanism we term \emphvariance equalization. A closed-form optimal exponent per dimension follows from a net information gain objective, with a single hyperparameter controlling the reconstruction-generation tradeoff. We validate on standard benchmarks (Binary MNIST, Binary Omniglot, CIFAR-10, CelebA-64), showing consistent reductions in collapsed dimensions, information capacity gains of up to 2.8\times nats, and reconstruction quality improvements of up to +0.33 BPD.
[CV-107] Rendering-Aware Bayesian 3D Gaussian Splatting with Native Uncertainty and Adaptive Complexity Control
链接: https://arxiv.org/abs/2607.05522
作者: Gaoxiang Jia,Vikram Appia,Junzhou Huang,Xinlei Wang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 26 pages, 4 figures, 24 tables including appendix. Preprint
Abstract:3D Gaussian splatting (3DGS) is a strong representation for real-time novel-view synthesis, but its standard training pipeline relies on point estimates and hand-tuned heuristics, providing no native uncertainty or principled complexity control. This is most limiting under sparse views or fixed acquisition budgets, where a model must identify weakly supported geometry and select informative views. We introduce a rendering-aware Bayesian 3DGS framework that tracks Gaussian geometry with a Normal-Inverse-Wishart posterior over means and covariances using renderer-derived surrogate summaries. An optional Dirichlet-process extension adds a probabilistic component-usage signal, and the training schedule makes the closed-form versus approximate inference boundary explicit. Re-rendering posterior geometry samples yields native predictive uncertainty for interval calibration and active view selection. In a fixed-budget 16-to-32 active-view task, native NIW acquisition improves PSNR by +0.453 dB and LPIPS by -0.0146 over a scoring-only 3-member standard-ensemble baseline, winning 29/39 scene-seed pairs and 10/13 scene means; it also improves over PPU-style (+0.355 dB) and NIW-proxy (+0.401 dB) acquisition. NIW native intervals reduce 95% coverage error by about 17x relative to a shared proxy (0.046 vs. 0.796) and are about 10x closer to nominal coverage than a 3-member deep ensemble (0.047 vs. 0.454) at roughly one-third the training cost. As a reconstruction compatibility check, paired NIW-vs-standard analysis over 39 scene-seed runs yields +0.030 dB PSNR with 1.6% additional training time. These results position Bayesian 3DGS as a practical probabilistic scene representation for decision-facing tasks such as active view selection.
[CV-108] Statistical Adversaries: Natural Backdoor-like Features in Vision Datasets
链接: https://arxiv.org/abs/2607.05516
作者: Paul K. Mandal,Pavan Reddy,Tristan Malatynski
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Machine Learning (cs.LG)
备注:
Abstract:Model-specific adversarial attacks have been extensively studied. We study a different failure mode: naturally occurring statistical signals in vision data that can behave like backdoor-like triggers without being maliciously inserted. We call these signals statistical adversaries. We analyse Imagenet to find patterns that are strongly linked to certain labels. We then use statistical controls to remove random correlations from our candidate signals. Finally, we demonstrate that these signals directly and predictably alter model predictions. These statistical adversaries are more targeted than generic corruptions and transfer across different model architectures. This suggests that some vulnerabilities are driven by dataset structure and distribution rather than a single model’s idiosyncrasies. We conclude that ordinary datasets can contain exploitable adversarial surfaces even in the absence of poisoning, and suggest that dataset audits should treat spurious structure not only as a source of bias or interpretability failure, but also as a latent attack surface for vision models.
[CV-109] Light-Omni: Reflex over Reasoning in Agent ic Video Understanding with Long-Term Memory
链接: https://arxiv.org/abs/2607.05511
作者: Chang Nie,Jiaju Wei,Junlan Feng,Chaoyou Fu,Caifeng Shan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project Page: this https URL
Abstract:Agentic video understanding equips models with long-term memory to autonomously process and respond to continuous, long-horizon multimodal streams. However, advanced video agents often rely on ``detective-style’’ iterative reasoning for action control (e.g., \mathttsearch ) and evidence aggregation, incurring prohibitive costs and latency. We argue that such heavy reasoning primarily compensates for the lack of global context and semantic misalignment in retrieval. This paper introduces Light-Omni, a multimodal agent framework for reflexive and lightweight video understanding. It achieves this through dual contextual states that instantly build the required context in a single forward pass. First, we maintain a global state, a finite-sized multimodal script continuously consolidated from episodic memory, serving as the global context for Light-Omni. Through hierarchical merging, it preserves recent details while summarizing past events. Second, conditioned on this global context, we generate a parametric latent state that directly drives autonomous actions and produces retrieval embeddings, with minimal latency. Benefiting from this coupled design, Light-Omni achieves semantically aligned retrieval and reflexive responses while avoiding iterative reasoning. Extensive experiments validate the effectiveness of Light-Omni across multiple video benchmarks. Notably, it outperforms M3-Agent with an average 2.4% accuracy gain, a 12.1 \times speedup, and a 2.6 \times improvement in GPU memory efficiency. Furthermore, it serves as a memory system to enhance both the performance and efficiency of existing MLLMs. Project page: this https URL.
[CV-110] Ground3D-LMM: Fine-Grained 3D Point Grounding and Spatial Reasoning with LMM ECCV2026
链接: https://arxiv.org/abs/2607.05493
作者: Amol Harsh,Zongyan Han,Jean Lahoud,Ye Liu,Rao Muhammad Anwer,Hisham Cholakkal,Salman Khan,Fahad Khan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: ECCV 2026
Abstract:Natural-language queries about 3D environments become actionable when responses are verifiable and metric. Verifiability requires explicit grounding to the referred 3D region, while metric answers report physical measurements in real-world units (e.g., size, thickness, clearance, and distance). Existing 3D large multimodal models (LMMs) approaches remain limited: conversational systems typically respond without explicit 3D grounding, while 3D grounding models are not designed for interactive, metric-aware dialogue. In this paper, we present Ground3D-LMM, a unified model that takes a point cloud and an optional RGB image as input and supports 3D spatial conversation with (i) point-grounded responses and (ii) metric numeric outputs at both object and part granularity, including multi-object queries. To evaluate this intersection of grounding and measurement, we define the 3D Grounded Measurement task, which requires predicting the referred 3D region and the corresponding metric quantities in real-world units. We introduce a large-scale dataset built on ScanNet and ScanNet++ datasets with dense object and part annotations and roughly 2.5M question-answer pairs spanning eight tasks, along with a manually verified test set. Extensive experiments on multiple datasets and tasks show that our proposed Ground3D-LMM model provides a strong baseline for grounded, metric-aware 3D conversational understanding. Our dataset and model are publicly available.
[CV-111] Binocular Gaze Estimation with Single Camera and Single Light Source
链接: https://arxiv.org/abs/2607.05473
作者: Tongbing Huang,Yang Fu,Yunfei Wang,Zhaocan Wang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted for presentation at the 2019 International Conference on Video, Signal and Image Processing (VSIP 2019), Wuhan, China, October 29-31, 2019; published in VSIP '19: Proceedings of the 2019 International Conference on Video, Signal and Image Processing, pp. 10-14, ACM, 2020; 4 figures, 1 table; ACM Proceedings ISBN: 978-1-4503-7148-3
Abstract:According to commonly consented theories, the minimum hardware requirement for gaze tracker is one camera and two light sources to realize gaze estimation with free head movements. However, in some scenarios such as eye tracking on mobile devices, it is preferable to use less components, especially light sources. We propose a gaze estimation method with one camera and one light source. A “virtual light source” is introduced, which is geometrically placed symmetrically to the real light source with respect to the camera, and generates a “virtual glint” in the acquired image. We estimate the “virtual glint” by exploiting the relationship between the distance between two pupils and two glints in the captured image, and estimate the gaze with polynomial regression assuming two light sources are available. A new normalization factor for regression method is verified, which turns out to be practical for one-glint system. The performance is proved to be acceptable, while degradation is noticed compared to system with two actual light sources.
[CV-112] A Task-Driven Evaluation of UAV Detection and Tracking under Synthetic Fog
链接: https://arxiv.org/abs/2607.05467
作者: Amir Pouladi,Vesal Ahsani,Haijun Li,Homayoun Najjaran,Afzal Suleman
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Image and Video Processing (eess.IV)
备注:
Abstract:Fog severely degrades the visibility of small unmanned aerial vehicles (UAVs) in skydominant, long-range imagery, reducing the reliability of downstream detection and tracking. This paper presents a task-driven evaluation framework that links depth-aware synthetic fog generation, image restoration, object detection, and tracking within a unified pipeline. Given the practical difficulty of collecting and annotating foggy UAV scenes, synthetic fog is generated from real clear-weather outdoor images containing UAV targets using monocular depth estimation and the atmospheric scattering model. Representative restoration methods from classical, convolutional neural network (CNN)-based, and transformer-based families are first compared, after which the selected restoration model is integrated into the downstream perception pipeline. Detection is evaluated under both clean-only and fog-inclusive training regimes using multiple detector variants, while tracking-by-detection is assessed on clean, foggy, and restored video sequences. Beyond image-level restoration metrics, the study evaluates how fog and restoration affect detection robustness and tracking performance. The results show that fog substantially degrades both detection and tracking, primarily through increased missed detections. Fog-inclusive training provides the most consistent improvement in robustness, whereas test-time restoration is most beneficial when the detector has been trained only on clean imagery. These findings show that restoration quality does not necessarily translate into proportional gains in downstream perception and therefore should be evaluated jointly with detection and tracking performance.
[CV-113] CanvasAgent : Enabling Complex Image Creation and Editing via Visual Tool Orchestration
链接: https://arxiv.org/abs/2607.05465
作者: Hairui Zhu,Yiying Yang,Tengjin Weng,Ziyu Lu,Xiao Yao,Xiaoyang Ye,Lin Ma,Wenhao Jiang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 18pages, 5 figures
Abstract:Complex image creation and editing often require more than a single generation or editing model. A user request may involve synthesizing images, localizing objects, segmenting regions, editing selected content, compositing intermediate assets, reading text, and enhancing the final result. Such tasks shift multimodal agents from perception-augmented reasoning to manipulation-centered visual creation, where tools must actively transform visual states rather than merely inspect them. However, existing multimodal tool-use agents are mostly optimized for perception, search, or domain-specific editing, and lack large-scale supervision for executable image-creation trajectories. In this paper, we introduce CanvasCraft, a large-scale multimodal tool-use dataset for complex image creation and editing, and \textbfCanvasAgent, a tool-augmented multimodal agent that learns to orchestrate heterogeneous visual tools through multi-turn interaction. CanvasCraft contains 140K fully annotated executable trajectories and 10K RL task specifications. CanvasAgent is first trained with SFT to learn executable reasoning-action trajectories, and is then optimized with GRPO using a hybrid reward that combines outcome- and process-level signals. During rollout, CanvasAgent inspects intermediate results, tracks visual assets, and adapts tool decisions to the evolving visual state. Experiments evaluate both final image quality and trajectory behavior, demonstrating the effectiveness of CanvasAgent and the proposed dataset for complex multi-tool image creation workflows. Comments: 18pages, 5 figures Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) Cite as: arXiv:2607.05465 [cs.CV] (or arXiv:2607.05465v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2607.05465 Focus to learn more arXiv-issued DOI via DataCite
[CV-114] BitFair: A 12nm Bit-Serial CNN Accelerator with Learnable Early Termination and Adaptive Bit Ordering for Ultra-Low-Power XR Vision
链接: https://arxiv.org/abs/2607.05445
作者: Ang Li,Chang Gao
类目: Hardware Architecture (cs.AR); Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注: Under review
Abstract:Extended Reality (XR) wearables require always-on perception within tight power envelopes of a few watts and motion-to-photon latency budgets below 20 ms, leaving only a few milliseconds for neural-network inference. Bit-serial computing is attractive for such energy-efficient neural network acceleration, but many existing architectures still process all bits even when ReLU sets the final output to zero. This paper presents BitFair, a software-hardware co-designed bit-serial CNN accelerator with learnable bit-level early termination and adaptive bit ordering, working under the ultra-low-power and strict latency requirements of XR applications. BitFair exploits dynamic bit-level sparsity by learning per-layer thresholds that trigger early termination when partial sums reliably predict that the final ReLU output will be zero. Furthermore, it searches for layer-wise bit orders that prioritize informative bits, maximizing early termination without sacrificing accuracy. A GlobalFoundries 12nm FinFET implementation with a core area of 0.34 mm^2, 104 KB on-chip memory, and voltage scaling from 0.55 to 0.70 V achieves sub-millisecond latency, up to 117.0 BTOPS/W, and 0.07 pJ/SOP. On IBM DVS128 Gesture and N-MNIST, BitFair achieves 96.5% and 97.7% accuracy, respectively, while improving effective energy efficiency by 4.0-22.1x and accuracy by up to 9.2% over prior fabricated XR vision accelerators.
[CV-115] Abductive Corroboration of Probabilistic AI Models for Forensic Synthetic Media Detection
链接: https://arxiv.org/abs/2607.05434
作者: Junade Ali
类目: Cryptography and Security (cs.CR); Computer Vision and Pattern Recognition (cs.CV); Computers and Society (cs.CY)
备注:
Abstract:Artificial Intelligence (AI) models, at their core, apply general learnings from broad datasets to individual circumstances using probabilistic behaviour. This inductive approach stands in contrast to deductive reasoning approaches which seek to prove conclusions from their premises. However, research has shown that deductive reasoning with AI models is a challenging problem and in the real-world it may not always be feasible. An alternative way forward is to leverage abductive reasoning, seeking to corroborate the output of multiple approaches to identify the most likely conclusion from the factual matrix. We apply this to synthetic media detection in forensic settings, and find we are able to disproportionately lower the risk of false positives to true positive recall. We also provide the first empirical evaluation of OpenAI’s rollout of SynthID on synthetic images and evaluate how complementary different synthetic media detection approaches are.
人工智能
[AI-0] Graph Convolutional Attention: A Spectral Perspective on Graph Denoising and Diffusion
链接: https://arxiv.org/abs/2607.06546
作者: Shervin Khalafi,Igor Krawczuk,Sergio Rozada,Charilaos Kanatsoulis,Antonio G Marques,Alejandro Ribeiro
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Denoising graphs is a fundamental problem in graph learning and the core operation of graph diffusion models. Attention-based architectures like graph transformers have recently shown promise in denoising graphs. However, our principled understanding of attention-based graph denoising remains limited, making it unclear whether standard attention is the right mechanism for this task. Here we show that, under a denoising objective, linear attention is suboptimal and can only learn an average spectral denoising filter over the training distribution. This creates a fundamental limitation as graphs often vary spectrally across the distribution. To overcome this limitation, we introduce Spectral Attention, which directly utilizes the input graph spectrum and provably outperforms linear attention by a margin governed by the spectral diversity of the distribution. We then derive Graph Convolutional Attention (GCA), a practical and permutation-equivariant realization of this idea that implements spectral denoising through graph-filtered queries and keys. For stochastic block models, GCA provably matches the idealized Spectral Attention mechanism. We further show that the softmax operation, that follows the attention, provides additional denoising by approximately projecting noisy eigenvectors onto the clean eigenspace. Empirically, replacing linear attention with GCA consistently improves graph denoising and diffusion on synthetic and real datasets, with gains strongly correlated with spectral diversity. In DiGress, GCA matches standard graph-transformer performance without computing expensive structural features, and when combined with the recently proposed PEARL positional encodings, avoids explicit eigendecomposition computations resulting in faster inference without degrading quality. The code can be found here: this http URL
[AI-1] he Large Cancer Assistant (LCA): A Model-Agnostic Orchestration Framework for Scalable Clinical Decision Support in Oncology
链接: https://arxiv.org/abs/2607.06531
作者: Ghassen Marrakchi,Basarab Matei
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 22 pages, 6 figures, 8 tables, 9 appendices, 14 references, Elsevier JBI format
Abstract:- Objective: Multimodal deep learning models in oncology are currently limited by monolithic designs that rigidly couple data ingestion, clinical routing, and artificial intelligence (AI) inference. To address this inflexibility, we propose the Large Cancer Assistant (LCA), a model-agnostic, post-hoc orchestration framework designed for scalable clinical decision support. - Methods: The LCA is mathematically formalized as a 7-tuple architecture grounded in the principle of Algorithmic Impermeability, ensuring the orchestration logic remains strictly independent of underlying black-box AI models. We introduce the Entry Theory, leveraging Geometric Deep Learning (GDL) to standardize multimodal patient data along distinct structural and medical axes. The system dynamically orchestrates data via a Cancer Switching Module and intentionally isolates the core AI execution from volatile hospital IT infrastructures by outputting a Standardized Intermediate Payload (SIP). - Results: A Proof of Concept (PoC) validated the orchestration logic across four technical scenarios. The framework executed a nominal flow with negligible orchestration overhead. It empirically demonstrated algorithmic impermeability by maintaining an invariant routing projection during AI model swaps, and it validated strict failure-safety by achieving a 100% recall rate in generating targeted Supplementary Data Requests (SDR) under injected data anomalies. Multi-protocol execution capability was also successfully verified. - Conclusion: By structurally decoupling multimodal ingestion from feature inference, the LCA provides a highly adaptable and modular orchestration foundation. The SIP establishes a clear architectural boundary, natively setting the stage for downstream Electronic Medical Record (EMR) interoperability as an independent future paradigm.
[AI-2] DepthWeave-KV: Token-Adaptive Cross-Layer Residual Factorization for Long-Context KV Cache Compression
链接: https://arxiv.org/abs/2607.06523
作者: Anna Cordoba,Adam Puente Tercero,Nerea Angulo Hijo,Mar Linares Tercero,Julia Barrientos,Ainhoa Miranda,Jesus Olivera
类目: Artificial Intelligence (cs.AI)
备注: 9 pages, 2 figures
Abstract:Long-context language model inference is increasingly limited by the memory bandwidth and capacity required to store key-value caches, yet existing compression methods often apply uniform budgets across layers or tokens and degrade retrieval when lexical cues and semantic states require different preservation. We introduce DepthWeave-KV, a token-adaptive cache compression method that factorizes key and value states across neighboring transformer layers using shared low-rank channel bases while retaining lightweight token-specific residuals where attention behavior is sensitive. DepthWeave-KV combines cross-depth residual factorization with a token-conditional depth router that allocates higher reconstruction rank to instruction-bearing and retrieval-critical tokens, and uses calibration-free online error tracking from attention-output probes to adapt compression during generation without retraining the base model. A fused CUDA implementation jointly performs basis lookup, residual dequantization, and attention projection to reduce decode-time memory traffic. Across LongBench, Needle-in-a-Haystack, L-Eval, and long-form QA and summarization benchmarks, DepthWeave-KV achieves near-full-cache task quality with substantially lower memory use, improving average score and retrieval accuracy over prior compressed caches while reaching 8.3x KV memory reduction and 72.8 tokens per second at 64K context.
[AI-3] FreqDepthKV: Frequency-Guided Depth Sharing for Robust KV Cache Compression in Long-Context LLM Inference
链接: https://arxiv.org/abs/2607.06519
作者: Anna Córdoba,Adam Puente Tercero,Nerea Angulo Hijo,Mar Linares Tercero,Julia Barrientos,Ainhoa Miranda,Jesús Olivera
类目: Artificial Intelligence (cs.AI)
备注: 11 pages, 2 figures
Abstract:Long-context LLM inference is increasingly limited by the memory and bandwidth cost of KV caches, yet aggressive compression can remove the layer-specific evidence needed for retrieval and multi-step reasoning. We introduce FreqDepthKV, an inference-time cache compression method that factorizes adjacent-layer KV states into shared low-frequency depth components and sparse high-frequency residuals. A lightweight online probe assigns attention heads to shared-depth, residual-depth, or exact cache modes according to their contribution to reconstruction-sensitive attention logits, allowing the compression policy to adapt to prompt structure without retraining. Across long-context question answering, needle retrieval, summarization, and code generation benchmarks, FreqDepthKV preserves task accuracy under substantially smaller cache budgets. With a 32k-token prefill window, FreqDepthKV reaches 58.3 Exact Match, 63.0 F1, 32.5 ROUGE-L, and 48.1 pass@1, closely matching full KV while outperforming prior compressed-cache methods. It also improves decoding throughput to 70.4 tokens/s, reduces TTFT to 2.06 seconds, and lowers peak KV memory to 6.2 GB, achieving a 3.9x effective compression ratio.
[AI-4] FootsiesGym: A Fighting Game Benchmark for Two-Player Zero-Sum Imperfect-Information Games
链接: https://arxiv.org/abs/2607.06514
作者: Chase McDonald,Nathan Tsang,Wesley N. Kerr
类目: Artificial Intelligence (cs.AI); Computer Science and Game Theory (cs.GT)
备注: Accepted to the RLC 2026 Reinforcement Learning Video Games Workshop; 14 pages, 9 figures
Abstract:We present FootsiesGym, an open-source environment for learning in a non-trivial two-player, zero-sum, imperfect-information game. Built on HiFight’s minimalist 2D fighting game Footsies, it isolates the cyclic, non-transitive strategic interactions of fighting game neutral play while remaining simple enough for efficient analysis. We provide a vectorized simulator that enables high-throughput training on standard hardware, making the environment accessible and reproducible. We describe the design of the environment, benchmark several reinforcement learning algorithms, and discuss open research directions it enables. The code is available at this https URL.
[AI-5] Industry Classification of GitHub Repositories Using the North American Industry Classification System (NAICS)
链接: https://arxiv.org/abs/2607.06505
作者: Kevin Xu,Alexander Quispe
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Databases (cs.DB)
备注:
Abstract:GitHub hosts hundreds of millions of public repositories, but the platform exposes no native mapping from repositories to standardized industry sectors. This gap limits empirical work on the geography of innovation, the industrial composition of open-source production, and the diffusion of new technologies across economic sectors. We present NAICS-GH, a publicly released corpus of 6,588 GitHub repositories drawn from source pools covering the United States, the European Union, and Australia, each labeled with a 2-digit sector from the North American Industry Classification System (NAICS 2022). Labels are produced by a retrieve-and-verify pipeline that combines BAAI/bge-large-en embeddings, FAISS retrieval, and GPT-4.1 rubric scoring. The pipeline narrows about 1.37 million source repositories to 31,178 candidate repository-sector pairs and retains 6,588 high-confidence labels with score at least 8. Re-running the retrieval pipeline end to end reproduces the candidate set to within 0.03 percent. On a 2,421-repository human-validated random sample, the released labels attain 96.98 percent precision, with Wilson 95 percent confidence interval [96.23, 97.59]. We benchmark six pretrained encoders on the released corpus; RoBERTa-large reaches 86.45 percent F1 and 86.35 percent accuracy on a held-out 20 percent test set. The dataset, Croissant metadata, pipeline code, prompts, and fine-tuned checkpoint are released under CC-BY-4.0 and MIT licenses.
[AI-6] RMISC: A Large-scale Real-world Multivariate Corpus for Time Series Foundation Models
链接: https://arxiv.org/abs/2607.06504
作者: Qian Sun,Yong-Ming Tian,Jia-Wei Huang,Cheng Feng,Shao-Qun Zhang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Recent years have witnessed the emergence of multivariate modeling using time series foundation models (TSFMs), which achieve advanced zero-shot generalization. Modern multivariate TSFMs are predominantly pretrained on multivariate synthetic data, which is easier to scale but may fail to capture the complex temporal dynamics and cross-variable relationships present in real-world time series. This raises a key question: Whether and to what extent the leading TSFMs trained with the real-world corpus perform better than those trained with synthetic data? To answer this, we establish the RMISC corpus, a considerably large-scale, high-quality, openly accessible, real-world, and multivariate time series archive that contains around 200 datasets and 142 billion time points across diverse domains. Furthermore, we pretrain four advanced TSFMs on univariate, synthetic multivariate, and real-world multivariate data and evaluate their zero-shot generalization capabilities on standard in-distribution and out-of-distribution benchmarks. Experimental results show that incorporating real-world multivariate data predominantly improves the generalization performance for both univariate and multivariate TSFMs. These results provide a deeper understanding of how real-world multivariate data contributes to the development of stronger TSFMs.
[AI-7] Doomed from the Start: Early Abort of LLM Agent Episodes via a Recall-Controlled Probe Cascade
链接: https://arxiv.org/abs/2607.06503
作者: Kai Ruan,Zihe Huang,Ziqi Zhou,Qianshan Wei,Xuan Wang,Hao Sun
类目: Artificial Intelligence (cs.AI)
备注: 10 pages, 9 figures, 2 tables. Code will be released soon
Abstract:Large language model (LLM) agents solving multi-step tasks frequently commit to trajectories that are doomed to fail, yet continue to consume substantial inference compute before the failure becomes observable. We show that failure is predictable early from the agent’s internal representations: lightweight per-round probes on hidden activations anticipate eventual episode failure as early as the first interaction round, where scorers reading only the agent’s observable behavior are barely better than chance. We turn this signal into a practical abort cascade: one distribution-free calibrated gate per round, with per-round recall budgets jointly searched so that eventually-successful episodes survive all gates at a user-specified global rate; this episode-level guarantee is the one that matters in deployment, since false-abort risk accumulates across gates. Across two agent models on TextCraft, the cascade meets every recall target from 90% to 97% and, at the 90% target, saves 47.1% +/- 10.3% (Qwen-2.5-7B) and 37.2% +/- 8.8% (Llama-3.2-3B) of inference compute, 1.6–1.7x the best single-gate policy. An otherwise-identical cascade reading only behavior saves roughly half as much, and adding behavioral features to the probe yields no further gain: the hidden states capture what behavior reveals. Finally, we characterize the sample complexity of certifying high recall targets, telling practitioners which recall promises their data can, and provably cannot, back. The code will be released soon.
[AI-8] Multi-Agent Deep Reinforcement Learning for Multi Objective Battery Management in Dairy Farms
链接: https://arxiv.org/abs/2607.06489
作者: Marcos Eduardo Cruz Victorio,Karl Mason
类目: Artificial Intelligence (cs.AI)
备注: 8 pages, 2 figures
Abstract:The dairy industry in Ireland has a large potential for the integration of renewable energy and the reduction of carbon emissions. However, researchers of distributed generation control are mainly focused on residential and commercial applications. To contribute to the effective integration of renewable energy in the dairy sector, this paper presents a multi-objective optimisation control system based on differential evolution and multi agent Deep Reinforcement Learning. The proposed control is organised in two layers: the upper layer uses dynamic pricing, and the lower layer is based on multi-agent reinforcement learning for battery management. This paper also simulates the electrical response of the proposed control system in a rural distribution circuit. The simulation results show that the proposed control framework can improve profits from energy arbitrage up to 18% compared to using Rule-based models, increase the use of distributed generation without significantly increasing cost, and comply with the Irish grid code in terms of voltage variation.
[AI-9] A Physics-Informed Neural Network Framework for Elastodynamic Wave Propagation in Bimaterial Systems
链接: https://arxiv.org/abs/2607.06479
作者: Sonal Ankush Chibire,Jenn-Terng Gau,Bo Zhang
类目: Artificial Intelligence (cs.AI); Mathematical Physics (math-ph)
备注:
Abstract:Physics-informed neural networks (PINNs) provide a promising framework for solving partial differential equations while embedding the underlying physical laws directly into the learning process. This study presents a PINN-based framework for modeling transient elastodynamic wave propagation in bimaterial systems governed by the axisymmetric equations of linear elasticity. A steel-aluminum specimen representative of a Split Hopkinson Pressure Bar configuration is considered, and the governing elastodynamic equations, together with the corresponding initial, boundary, and interface conditions, are incorporated directly into the network through a physics-informed loss function. High-fidelity finite-element simulations performed using ANSYS Workbench Explicit Dynamics are used for validation and as supplementary data constraints during training. The proposed framework accurately predicts wave transmission and reflection across the bimaterial interface and reproduces axial and radial displacement histories, face-averaged responses, and the dominant stress and strain evolution with close agreement to the finite-element solutions. The trained network further demonstrates the ability to predict wave responses at previously unseen time instants and for modified material properties without requiring additional finite-element simulations, providing a continuous surrogate model for elastodynamic analysis. Mesh-sensitivity studies confirm numerical robustness, while additional material combinations demonstrate the generality of the proposed methodology. The results show that integrating physics-informed neural networks with explicit finite-element analysis provides an accurate and computationally efficient framework for elastodynamic wave propagation in heterogeneous solids, offering an effective surrogate modeling approach for high-rate solid mechanics and impact engineering applications.
[AI-10] Finding H. pylori in the Fine Print: Evidence-Linked Multi-Agent Case Finding from Gastric Biopsy Reports
链接: https://arxiv.org/abs/2607.06435
作者: Yufan Wang,Anit Kumar Sahu,Yan Fei Ng,Daniel Kang,Shayan Vassef,Soorya Ram Shimgekar,Koustuv Saha,Piyum Zonooz,Navin Kumar,Chee Leong Cheng,Li Yan Khor
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Data from Singapore indicated that about 31% of the population had evidence of Helicobacter pylori infection. Persistent H. pylori infection is associated with chronic active gastritis and peptic ulcer disease, and its eradication is key to gastric cancer prevention. However, evidence supporting \textitH. pylori positivity and H. pylori-associated gastritis may be distributed across heterogeneous coded and free-text report fields and may require contextual interpretation of assertion and negation, limiting keyword search, and making manual review difficult to scale. We conducted a retrospective pilot evaluation of the Nimblemind Multi-Agent System (nMAS), a field-name-driven, evidence-linked extraction workflow, using 54 de-identified gastric biopsy pathology reports from a large healthcare system in Singapore. Four clinician-scoped binary fields were evaluated: gastric/stomach biopsy, biopsy status, H. pylori positivity, and H. pylori-associated gastritis. Across 216 feature-case decisions, nMAS correctly classified 213, corresponding to 98.61% overall accuracy. A separately implemented UMA-style MiniMax M2.5 comparator produced similar aggregate and per-field classification metrics. Although predictive performance was similar, nMAS maintained unified report-level outputs with supporting source sentences; the demonstrated contribution is therefore workflow integration and traceability rather than predictive superiority. Under an illustrative, unmeasured scenario, reviewing 1,000 reports at five minutes per manual review versus five seconds per evidence-linked verification would reduce review time from 83.3 to 1.4 staff-hours, corresponding to 81.9 staff-hours and about USD~6,100 in potential staff-time value. Larger multi-institutional studies should evaluate evidence-span correctness, clinician verification time, and generalizability.
[AI-11] ExplAIner: A Declarative Query Language for Explaining Classification Models
链接: https://arxiv.org/abs/2607.06407
作者: Marcelo Arenas,Pablo Barceló,Diego Bustamante,Jose Caraball,María Alejandra Schild,Bernardo Subercaseaux
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:The XAI community has studied a wide range of queries and scores for explaining predictions of ML models. From a data management perspective, this proliferation of explanation notions calls for declarative query languages in which such notions can be specified, combined, and analyzed uniformly. In this paper, we develop such a framework for Boolean models. We first revisit FOIL, an interpretability query language for black-box models, and show that it has two fundamental limitations: it cannot express central optimality-based explanation queries, and its evaluation problem over decision trees is hard for every level of the polynomial hierarchy. We then introduce ExplAIner, a query language based on FOIL with an extended vocabulary and a layered structure. We show that ExplAIner can express a broad family of explanation notions, including abductive, contrastive, feature-based, and distance-based queries. We also prove that the evaluation problem for each query in ExplAIner belongs to the Boolean hierarchy over every class of Boolean models for which some basic predicates can be evaluated in polynomial time. In particular, that property holds for deterministic and decomposable Boolean circuits. Finally, we introduce Opt-FOIL, an optimization-oriented fragment of ExplAIner for computing explanations that are minimal with respect to strict partial orders, and prove that its evaluation problem is in \mathrmFP^\mathrmNP under the same tractability assumptions. These complexity results have a direct algorithmic consequence: a fixed ExplAIner query can be evaluated with a fixed number of calls to a SAT solver, while a notion of explanation specified in Opt-FOIL can be computed with a polynomial number of such calls. This is particularly relevant in formal XAI, where SAT solvers have been successfully used to compute explanations for several classes of ML models.
[AI-12] A Definition and Roadmap for World Models
链接: https://arxiv.org/abs/2607.06401
作者: Xinyuan Chen,Haoyu Guo,Shi Guo,Bingqi Jiang,Chunhua Shen,Xing Shen,Tianfan Xue,Yufei Xue,Mulin Yu,Weinan Zhang,Bin Zhao,Bowen Zhou,Ming Zhou
类目: Artificial Intelligence (cs.AI)
备注: Technical report, 58 pages, 10 figures
Abstract:World models – internal simulators that learn the structure and dynamics of an environment – have become one of the most actively debated concepts in AI. From model-based reinforcement learning and video generation to embodied robotics and ultimately, physical AI, researchers across AI subfields are building systems that they call “world models”, yet there is no consensus on what a world model fundamentally is, what it should predict, or how it should be built. This perspective article provides a scientific definition of world models, discussions of their key technical aspects, and a staged roadmap for developing effective world models.
[AI-13] opoBrick: Agent ic Topology Sampling of Exogenous Variables for Zero-Shot Building IoT Forecasting
链接: https://arxiv.org/abs/2607.06349
作者: Xiachong Lin,Du Yin,Arian Prabowo,Hao Xue,Wen Hu,Imran Razzak,Matthew Amos,Sam Behrens,Flora D. Salim
类目: Artificial Intelligence (cs.AI)
备注: 12 pages, 4 figures, 3 tables
Abstract:Building sensors are embedded in physical topology, spatial hierarchy, and operational context, yet existing forecasters often treat them as isolated time series or rely on fixed covariate sets. We present TopoBrick, a training-free framework for zero-shot building IoT (Internet-of-Things) forecasting. TopoBrick uses building knowledge graphs to construct a compact structural skeleton and employs an agentic topology sampler to select target-specific exogenous variables. The selected variables are organized by deployment-time availability, separating past-known sensor states from future-known calendar, schedule, and meteorological exogenous variables. Across three real-world buildings, TopoBrick outperforms strong zero-shot foundation-model baselines and remains competitive with fully trained building-specific models. Ablations show that topology-aware sampling is more reliable than random, ontology-only, or fixed-hop selection, especially for physically coupled HVAC and weather-driven sensing variables.
[AI-14] Harnessing Code Agents for Automatic Software Verification
链接: https://arxiv.org/abs/2607.06341
作者: Shuangxiang Kan,Shuanglong Kan,Sebastian Ertel
类目: Formal Languages and Automata Theory (cs.FL); Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注:
Abstract:Formal verification offers the strongest guarantee of software correctness, but it does not scale: the proofs demanded by interactive theorem provers such as Coq require enormous expert effort. Large language models (LLMs) promise to generate these proofs automatically, yet existing approaches wire a fixed, human-designed proof strategy into the system and constrain the model to follow it (retrieving premises and predicting tactics one step at a time, or splitting goals by divide-and-conquer), and still prove only a fraction of their target theorems. We show that imposing such a strategy is unnecessary and limiting. Handing the whole lemma to a general LLM code agent (for example, Claude Code), free to choose its own approach, and wrapping it in a verification harness is both simpler and more effective, achieving full coverage: every targeted lemma proved, with no failures and no Coq expert intervention. The agent writes the proofs under feedback and hard constraints from the harness that keep each one sound (accepted only when the prover’s kernel closes it), complete (no obligation left unproved or silently dropped), and terminating (no divergent tactics). We evaluate this harness plus code agent along three dimensions. (1) Core logic: on Iris, the state-of-the-art separation logic for concurrent and memory-manipulating programs, Aria proves all 4,257 lemmas of the four core modules and the 217 lemmas verifying Rust’s standard libraries built on it, fully automatically. (2) Comparison with prior LLM provers: on reglang, where prior provers manage barely one in eight, Aria proves all 318. (3) Generality: on iris-lean, the unfinished Lean 4 port of Iris, it proves 72 not-yet-ported lemmas, showing the approach is not specific to Coq. A state-of-the-art model (Claude Opus 4.7) can write proofs for verified software development fully and automatically. Subjects: Formal Languages and Automata Theory (cs.FL); Artificial Intelligence (cs.AI); Software Engineering (cs.SE) Cite as: arXiv:2607.06341 [cs.FL] (or arXiv:2607.06341v1 [cs.FL] for this version) https://doi.org/10.48550/arXiv.2607.06341 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-15] DT-Guard: Intent-Driven Reasoning -Active Training for Reasoning -Free LLM Safety Guardrail
链接: https://arxiv.org/abs/2607.06326
作者: He Liu,Changtao Miao,Xinjie Yang,Tianle Song,Yin Wu,Junchi Chen,Bintao He,Xinyuan Zhang,Bo Zhang,Shi Yan,Wei Lu,Wei Wang,Danyang Xu,Jiansheng Cai,Zhe Li
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models deployed in open-world applications require safety guardrails that are both robust to complex risks and efficient enough for low-latency runtime moderation. Existing guardrails face a practical trade-off between lightweight classification-based models, which are efficient but often struggle with concealed intent, ambiguous semantics, and borderline safety decisions, and reasoning-based guards, which improve judgment quality but introduce additional token generation and inference latency. We present DT-Guard, a content safety guardrail model based on a Reasoning-Active Training, Reasoning-Free Inference paradigm. The key idea is to use reasoning supervision during training while emitting only structured safety labels at inference time. DT-Guard formulates safety judgment as a progressive decision process, Intent - Category - Safety, and constructs an intent-driven dataset with intent labels, risk categories, safety labels, and structured reasoning trajectories. To further improve hard-case robustness, we propose Rollout-Guided Progressive Hard-Case Optimization (RG-PHO), which uses multi-rollout consistency to identify stably mastered, persistently failed, and preference-unstable samples, and applies targeted supervised and preference optimization accordingly. At inference time, DT-Guard directly generates structured labels without explicit reasoning traces, preserving deployment efficiency. Experiments on prompt-side and response-side safety benchmarks show that DT-Guard achieves average F1 scores of 0.886 and 0.870, respectively. With only a 4B backbone, it reaches a dual-side average F1 of 0.878, outperforming strong 8B guardrail baselines. These results demonstrate that reasoning supervision can be effectively internalized into low-latency safety discrimination.
[AI-16] Designing Maintainable Hybrid Generative Systems: A Quantum-Inspired Approach to Automated Music Harmony Generation
链接: https://arxiv.org/abs/2607.06296
作者: Josef Pavlicek
类目: ound (cs.SD); Artificial Intelligence (cs.AI); Emerging Technologies (cs.ET); Audio and Speech Processing (eess.AS)
备注: 12 pages, 1 figure, 4 tables. Extended version of the 4-page paper accepted at the 34th International Conference on Information Systems Development (ISD2026, Prague). Source code and dataset available at this https URL
Abstract:This paper presents the design and evaluation of a maintainable hybrid generative architecture for automated music harmony generation from melody. The proposed system combines quantum-inspired candidate exploration over overlapping melodic contexts with explicit rule-based optimization to balance generative flexibility and structural control. The architecture is evaluated using explicit and reproducible metrics covering structural coherence, functional agreement, harmonic similarity, and robustness. The results show that the proposed approach produces harmonizations that preserve tonal structure and cadential behavior while allowing multiple valid harmonic realizations. Furthermore, the optimization layer improves structural coherence, stability, and predictability without requiring a training corpus. The study demonstrates that transparent and controllable hybrid generative systems can be systematically designed and evaluated within the context of Information Systems Development.
[AI-17] ask Decomposition-Guided Reranking for Adaptive Agent Skill Retrieval
链接: https://arxiv.org/abs/2607.06283
作者: Yanping Chen,Weijie Shi,Wen Yang,Jiajie Xu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Skill usage can significantly enhance the ability of modern agent systems to complete complex tasks. However, the growing scale of skill libraries makes accurate skill selection increasingly challenging. In real-world scenarios, ambiguous semantic matching often arises between a specific task requirement and multiple generic yet semantically similar candidate skills. Moreover, existing methods tend to overlook the dynamic influence of task difficulty and skill applicability when selecting the optimal target skill set. To address these issues, we propose SkillReranker, an inference-time reranking framework for adaptive skill selection. Specifically, we first perform semantic decomposition on both the task and skill sides, yielding informative subtask and execution-state descriptions as well as transition-state descriptions that characterize each skill’s functionality. These descriptions are then used to construct a directed acyclic execution graph, where intermediate task states are modeled as nodes and candidate skills as edges, thereby establishing a structured task-skill correspondence. On this basis, SkillReranker determines whether each state node satisfies the split condition to identify subtask intervals. For each task interval, we employ a cross-encoder to perform comprehensive scoring over candidate skills and select the most suitable ones to form the final target skill set. Experiments on ALFWorld and ScienceWorld with three backbone LLMs show that SkillReranker effectively improves task performance, reduces environment interaction steps, and lowers token consumption compared with existing skill selection baselines.
[AI-18] Demonstrating TOFFEE: A Learned System for Synthesizing Data Agent Trajectories at Scale VLDB2026
链接: https://arxiv.org/abs/2607.06233
作者: Ziting Wang,Yin Li,Zuhao Yang,Xiuchang Li,Jiale Bai,Gao Cong
类目: Artificial Intelligence (cs.AI)
备注: Accepted to VLDB 2026
Abstract:LLM-powered data agents are playing an increasingly important role in data-driven decision making. However, existing data agents struggle to generalize to unseen data environments and analytical workflows, especially in heterogeneous enterprise settings. This creates a growing need for synthesizing high-quality data agent trajectories that capture complex analytical workflows for given data environments. Such trajectories support two key downstream uses: they can serve as supervised finetuning (SFT) data that adapts data agent models to the target domain, and as in-context learning (ICL) demonstrations to guide general-purpose LLMs in unfamiliar data environments. Thus, we introduce TOFFEE, a system for synthesizing high-quality data agent trajectories from given data environments via Monte Carlo Tree Search (MCTS) with adaptive model selection and cross-task prefix reuse. We show that TOFFEE can effectively generate scalable trajectory data for complex analytical tasks across heterogeneous environments. In this demonstration, we present the system framework of TOFFEE, including its task pool construction, trajectory explorer, and learned cost model. We also introduce the web interface of TOFFEE and its workflow, and demonstrate two end-to-end scenarios: trajectory synthesis for data agent finetuning, and demonstration-augmented data agent reasoning.
[AI-19] Information Gain-based Rollout Policy Optimization: An Adaptive Tree-Structured Rollout Approach for Multi-Turn LLM Agents
链接: https://arxiv.org/abs/2607.06223
作者: Yijun Zhang,Fan Xu,Jiaxin Ding,Yule Xie,Shiqing Gao,Xin Ding,Haoxiang Zhang,Luoyi Fu,Xinbing Wang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Reinforcement learning has become a promising paradigm for improving large language model (LLM) agents on long-horizon search tasks, where the agent must make a sequence of intermediate decisions before receiving a final outcome. However, existing methods still face a key limitation: the rollout budget is often allocated without explicitly assessing the utility of intermediate states. As a result, substantial computation may be spent on low-value states, even though different branches can vary drastically in their informativeness. In this paper, we propose Information Gain-based Rollout Policy Optimization (IGRPO), a policy optimization framework that treats intermediate-state informativeness as the organizing principle of rollout collection. Specifically, IGRPO performs budget-aware tree-structured rollouts by allocating expansion budget according to node-level informativeness, so that more informative branches are expanded more frequently while unpromising branches are progressively suppressed. We further demonstrate that the information gain-based rollout induces an explicit limiting teacher distribution over trajectories, which naturally yields a clear policy optimization target, thereby unifying adaptive tree-structured exploration with principled policy learning under a single framework. Experiments on seven challenging search-augmented QA benchmarks demonstrate that IGRPO consistently outperforms strong baselines under the same rollout budget constraints, validating the effectiveness of leveraging the induced teacher distribution to guide policy optimization for long-horizon search agents.
[AI-20] A toy framework for single and multi-agent human-AI curiosity ecosystems
链接: https://arxiv.org/abs/2607.06214
作者: Ilya E. Monosov
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:This paper offers a toy framework for considering curiosity as an ecosystem. First, it suggests that a single agent’s inquiry policy (how, when, and why an agent asks a question) depends on how the agent values immediate uncertainty reduction, costs, delayed return, and the value of keeping the question open. A key concept in the framework is that the weights on these decision-related terms can change with experience. For example, a period of cheap, quickly answered questions may change the cost of inquiry on a short timescale and change which kinds of questions the agent is drawn to answer over a longer timescale. Second, these ideas are extended to many agents exploring a shared knowledge landscape, and there the framework tracks inquiry volume, topic diversity, frontier-directed inquiry, redundancy, and reusable knowledge. The result is a conceptual toy framework for studying curiosity ecology and for future efforts towards designing multi-agent AI systems for discovery. It serves as a companion piece for a paper currently under review in Trends in Neurosciences.
[AI-21] UBEP: Re-architecting Expert Parallelism Communication Library for Production Superpods
链接: https://arxiv.org/abs/2607.06202
作者: Yipeng Liu,Chang Liu,Si Shen,Jiaqi Zheng,Mingfan Li,Yuyang Yang,Guanhua Li,Yuquan Zhang,Yimeng Xu,Zhongzhe Hu,Zhiyuan Huang,Qihang Duan,Junsong Wang,Wenkai Ling,Baochuan Yang,Xianzhi Yu,Han Bao,Yijie Chen,Guihai Chen
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Artificial Intelligence (cs.AI); Networking and Internet Architecture (cs.NI)
备注:
Abstract:The deployment of Mixture-of-Experts (MoE) models on production high-bandwidth superpods, such as NVIDIA’s NVL72/576 and Huawei’s CloudMatrix384, introduces critical challenges beyond raw interconnect bandwidth. While these systems provide unified global address spaces and high-bandwidth fabrics, their full potential for sparse MoE communication is hindered by three fundamental bottlenecks: (1) Strict execution serialization imposed by coarse-grained Bulk Synchronous Parallel (BSP) orchestration of interdependent communication phases; (2) Prohibitive synchronization overhead that fails to scale alongside high interconnect bandwidth; and (3) Severe load imbalance resulting from distance-agnostic scheduling of irregular token traffic. To eliminate these bottlenecks, we introduce UBEP (Unified-Bus Expert Parallelism), a production-ready communication library that rethinks MoE’s All-to-All primitives for modern superpod architectures. Through large scale experiments, UBEP reduces All-to-All latency by up to 52.4% and MoE inference Time Per Output Token (TPOT) by up to 11.1%.
[AI-22] When do prophets profit in prediction markets?
链接: https://arxiv.org/abs/2607.06166
作者: Anri Gu,Nicole Kagan,Alec Sun,Jibang Wu,Haifeng Xu
类目: Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE); Computer Science and Game Theory (cs.GT)
备注:
Abstract:Prediction markets aggregate dispersed beliefs into prices that act as probabilistic forecasts of uncertain events. Classical theory establishes a clean equivalence between forecasting accuracy and trading profit, but only for the specific automated market maker (AMM) design. However, the largest exchanges today are based on central limit order books in which informed forecasters routinely lose money while uninformed strategies can profit on simple heuristics. We resolve this discrepancy by establishing a formal equivalence between predictive accuracy and profitability. For any strictly proper scoring rule S , we exhibit a “proper” betting strategy that depends only on the forecaster’s prediction \mathbfp and the market price \mathbfq , and earns positive expected profit whenever \mathbfp outperforms \mathbfq under S and the market has sufficient liquidity. Moreover, this proper betting is essentially the only strategy with such robust profitability guarantee. The proof rests on a decomposition of expected profit that strictly generalizes the classical AMM guarantee and also explains how strategies can profit without an accuracy edge. Empirically, across thousands of forecasts by AI models, proper betting is the only strategy that reliably converts accuracy into profit, and we further identify systematic forecasting personas and show how the optimal proper strategy varies across them. A month-long live deployment on Kalshi achieves +80.33% return on investment with a Sharpe ratio of 3.35 .
[AI-23] X-FEMR: A Token-level Explainable Approach for Electronic Health Records Foundation Models using Transformer-based Models ECAI2026 IJCAI ALT
链接: https://arxiv.org/abs/2607.06163
作者: Jie Huang,Pengfei Yin,Zihan Xu,Daniel Capurro,Mike Conway,Ting Dang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Accepted by IJCAI-ECAI 2026 AI and Health Track
Abstract:Foundation Models for Electronic Health Records (FEMRs) are pretrained on large-scale structured patient data, enabling them to convert longitudinal patient trajectories into generalizable representations for diverse clinical prediction tasks. Despite their effectiveness, FEMRs remain black-box models, raising concerns about bias, interpretability, and clinical trust. To address this, we propose the first token-level explainability approach for FEMRs. We train a Transformer-based surrogate model on input-output pairs from the FEMR across two prediction tasks, approximating its behavior while preserving temporal dynamics. We identify the most influential tokens, providing insights into how FEMRs leverage different aspects of patient history for predictions. To evaluate clinical relevance, we introduce a novel clinical alignment metric that quantifies the correspondence between the surrogate model’s key tokens and clinically validated features. Our results demonstrate that the surrogate closely approximates FEMR predictions and that token-level explanations align well with clinical knowledge, offering a practical framework for interpretable and trustworthy clinical AI.
[AI-24] Property-Driven Synthetic Data Engineering for Data-Scarce Software Systems: Reflections from the Breast Cancer Domain
链接: https://arxiv.org/abs/2607.06133
作者: Aurora Francesca Zanenga,Andrea Bombarda,Marsha Chechik,Saverio D’Amico,Rita De Sanctis,Alberto Zambelli,Claudio Menghi
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: 5 pages
Abstract:Modern software systems increasingly depend on data for analysis, prediction, testing, and decision-making. Yet many important domains, including medicine, safety-critical systems, and regulated industries, lack abundant, shareable, or representative data. Synthetic data generation is often proposed as a remedy, but our experience engineering software for intraoperative radiotherapy (IORT) in breast cancer treatment suggests that synthetic data shifts rather than solves the central engineering problem. The key challenge becomes deciding which properties synthetic data must preserve, how these properties should be elicited from stakeholders, how they can be validated under privacy constraints, and how they evolve. We call this problem property-driven synthetic data engineering. Drawing on a collaboration with oncologists and preliminary experiments with a sensitive IORT dataset, we identify challenges in requirements, validation, privacy, and pipeline evolution. We argue that automated software engineering research should develop methods and tools for eliciting, formalizing, checking, and evolving validity properties for synthetic data in data-scarce software systems.
[AI-25] Self-Supervised Implicit CEST Reconstruction via Physics-Informed Lorentz Encoding MICCAI2026
链接: https://arxiv.org/abs/2607.06132
作者: Dexuan Li,Yupeng Wu,Chenglong Wang,Hanlin Liu,Hui Zhen,Jianqi Li,Guang Yang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Medical Physics (physics.med-ph)
备注: 10 pages, 5 figures, Accepted by MICCAI 2026
Abstract:Multi-Pool Chemical Exchange Saturation Transfer (CEST) MRI provides valuable metabolic information but is clinically limited by long acquisition times. Although sparse sampling reduces scanning time, reconstructing high-resolution Z-spectra from limited data remains an ill-posed inverse problem. Conventional interpolation and generic Implicit Neural Rep-resentations (INRs) often lack physical constraints, leading to spectral artifacts and physically invalid signals. To address this, we propose Lorentz Encoding (LE), a physics-informed framework that formulates CEST reconstruction as a self-supervised reconstruction task via implicit continuous coordinate learning. Unlike generic positional encodings, LE regularizes the continuous spectral mapping by projecting sparse coordinates into a physically constrained space governed by a combination of parametric Lorentzian profiles with learnable basis functions. This mechanism effectively reduces noise and enforces consistency with physical models. Experiments on in vivo human brain data demonstrate that LE significantly outperforms state-of-the-art methods. Specifically, under a 39-point sampling strategy, LE achieves a PSNR of 57.58 dB and an SSIM of 0.9994. Furthermore, the learned physics-informed encodings form a continuous, geometrically ordered trajectory in the latent space, ensuring accurate quantitative metabo-lite mapping (APT, NOE, MT).
[AI-26] Evaluating Fine-Tuning and Metrics for Neural Decompilation of Dart AOT Binaries
链接: https://arxiv.org/abs/2607.06125
作者: Raafat Abualazm,Ayman AboElhassan,Amr G. Wassal
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注: Under review at ACM Transactions on Software Engineering and Methodology (TOSEM)
Abstract:Neural decompilation is increasingly studied as a code-generation problem, yet its evaluation methodology remains underdeveloped for modern languages. We present a systematic empirical study of fine-tuning effectiveness and metric validity for Dart Ahead-of-Time (AOT) neural decompilation. We evaluate six fine-tuned model variants across three base architectures (4B-8B parameters) using three metrics: CodeBLEU, compile@k, and pass@k on a new 154-task HumanEval-Dart benchmark. Our study yields three principal findings grounded in paired task-level statistical tests. First, no fine-tuning configuration produces a statistically significant pass@k improvement. The sole positive case yields +0.71 pp (McNemar p=0.21), while fine-tuning the strongest base (Qwen3-8B) causes a highly significant regression of -5.65 pp (p0.001). This capacity-dependent trend is consistent across architectures but needs broader scale sweeps. Second, cross-lingual interference from Swift training is highly significant at 4B (-2.66 pp, p0.001) but statistically indistinguishable from zero at 8B, consistent with the scaling hypothesis. Third, we demonstrate metric divergence: CodeBLEU and compile@k can improve significantly while pass@k moves in the opposite direction. This has implications for any LLM code generation task where fine-tuning targets superficial similarity. Error analysis reveals assembly sequence length is the strongest predictor of task difficulty (p=0.001), with a capability cliff at 200 instructions. We contribute the HumanEval-Dart benchmark, a Dart-adapted CodeBLEU, and empirical evidence that pass@k must be the primary evaluation metric for neural decompilation.
[AI-27] Static Metrics Are Insufficient: Predicting Java Method Energy Usage with Execution Time
链接: https://arxiv.org/abs/2607.06124
作者: Muhammad Imran,Vincenzo Stoico,Ivano Malavolta
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted for publication at the 19th International Conference on the Quality of Information and Communications Technology (QUATIC 2026)
Abstract:The increasing energy demand of software systems is raising concerns about their environmental impact and associated costs. Reasoning on energy usage early in the development flow has the potential to significantly reduce the overall energy usage of a software system, as it allows developers to make informed design and refactoring decisions before inefficiencies propagate. However, assessing energy usage without repeated profiling and direct measurement is difficult, which limits early reasoning in practice. This study investigates the limits of method-level energy prediction in Java, examining whether static source code metrics complemented with method-level execution time can estimate the energy consumption of Java methods. We profile 2,786 Java methods to extract 33 static features and measure execution time and energy, then train and compare eleven regression models. Our findings show that static source code metrics alone yield poor predictive performance, with average R2 values close to zero. Incorporating execution time as a lightweight dynamic input significantly improves accuracy, raising R2 to as high as 0.46. Execution time, internal method calls, and cyclomatic complexity consistently emerge as the strongest predictors of energy consumption.
[AI-28] x-Prediction Is All You Need:Training-Free Accelerated Generation via Endpoint Decodability
链接: https://arxiv.org/abs/2607.06114
作者: Xin Peng,Ang Gao
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Diffusion and flow matching models generate high-quality samples, but their ODE samplers often need tens to hundreds of neural function evaluations (NFEs). This remains a practical challenge for released checkpoints, since many accelerators require additional design choices and training cost through retraining, distillation, or trajectory redesign. We investigate a different route based on x -prediction. During sampling, standard affine probability paths already expose x_0 information: an intermediate state and its path velocity determine a principled estimate of the clean sample. We formalize this property as \textbfendpoint decodability and show that the decoder is the minimum-MSE estimator \mathbbE[x_0\mid x_t] under the usual \ell_2 objective. This yields \textbfTruncated Jump Sampling (TJS): stop the ODE at an early-exit time t^* and return the decoded x_0 . TJS requires no retraining, distillation, or architecture change. Across SDXL, SD3.5M, Z-Image-Turbo, and three class-conditional benchmarks, it reduces NFEs by 20–70% with near-matched quality. The analysis also shows why endpoint prediction can work without straightening the trajectory, providing inference acceleration without trajectory redesign.
[AI-29] LLM -Guided Measurement Credibility Correction for Trustworthy Industrial Process Inference
链接: https://arxiv.org/abs/2607.06111
作者: Youcheng Zong,Runda Jia,Dakuo He
类目: ystems and Control (eess.SY); Artificial Intelligence (cs.AI)
备注:
Abstract:Industrial prediction and soft sensing depend on credible input measurements. In field deployment, a predictor may receive biased, delayed, stale, or derived measurements that still look plausible. Prediction can then fail before the forecasting backbone becomes the main limitation, because the input window no longer represents the real process. Sensor reconstruction, data reconciliation, and fault-tolerant soft sensing reduce this risk, but they often rely on numerical correlation, alarms, fault labels, or explicit process equations. These assumptions are not always available. A correlated variable can also be an unsafe reference when variables share instruments, derived formulas, soft-sensing chains, or control actions. The key issue is to decide before prediction which external measurements can credibly support the current measurement. To address this issue, this article proposes LLM-Guided Measurement Credibility Correction (MCC). MCC converts measurement meanings in process documents into measurement semantics usable by numerical models. It builds independent process references from semantically qualified external measurements and corrects local measurement conflicts before prediction. The predictor therefore receives a more credible input window. Across multiple complex industrial forecasting and soft-sensing tasks, +MCC achieves average relative MAE reductions of 30.7% on real-test protocols and 80.3% on controlled-corruption protocols. It adds only 0.5–2.0k online parameters, with the slowest +MCC inference time at 0.089 ms/step. These results show that measurement semantics can turn process documents into lightweight pre-inference credibility correction and improve prediction accuracy.
[AI-30] Reward-Density Heuristic for Dynamic Multi-Vehicle Routing: Performance and Computational Efficiency
链接: https://arxiv.org/abs/2607.06066
作者: Manish Kolachalam,Rani Malhotra
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:The Vehicle Routing Problem (VRP) and its variants represent some of the most practically consequential optimization challenges in modern logistics and urban mobility. In this study, we address a dynamic, online variant combining elements of the VRP and the Orienteering Problem (OP), in which a fleet of vehicles must maximise cumulative reward collected within a fixed time horizon while continuously replanning as new tasks arrive. We propose and evaluate a reward-density heuristic for dynamic multi-vehicle assignment, referred to as the Efficiency heuristic. We evaluate this formulation across two application domains: autonomous drone task allocation and urban taxi dispatch, across multiple fleet sizes and task scales. The proposed method is compared with four classical construction heuristics and three metaheuristic algorithms (Adaptive Large Neighbourhood Search, Genetic Algorithm, and Simulated Annealing), all evaluated under identical conditions. Across all tested configurations, the Efficiency heuristic matches the solution quality of the best metaheuristic algorithms while requiring two to three orders of magnitude less planning time, establishing Pareto dominance over all competing methods on the reward-versus-compute frontier. These findings suggest a practical design principle for real-time allocation and dispatch systems: in dynamic, time-constrained routing environments, carefully designed greedy heuristics can match the output of sophisticated search procedures at a fraction of the computational cost, making them preferable for online deployment.
[AI-31] AgoraSim: A Hybrid Agent -Based Modeling Framework
链接: https://arxiv.org/abs/2607.05999
作者: Chung-Chi Chen
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:LLM-agent simulations make natural-language social scenarios easy to instantiate, but their outputs can be overread as predictions and are often difficult to compare with explicit social dynamics. We present AgoraSim, a hybrid agent-based modeling framework for scenario-oriented social reaction analysis. AgoraSim resolves textual or multimodal artifacts into editable ABM configurations, runs ratio-controlled populations that mix LLM, vision-language, custom-endpoint, random, and classical agents, and compares the same scenario against matched classical reference dynamics. All agents emit a shared structured decision object, enabling common action spaces, interaction protocols, metrics, and audit records. Exposed through a local UI, Python SDK/CLI, and REST API, AgoraSim helps users inspect scenario trajectories, compare modeling assumptions, and identify cases that warrant empirical validation.
[AI-32] Auto-DSM Under the Lens: A Black-Box Evaluation Framework for LLM -Based DSM Generation
链接: https://arxiv.org/abs/2607.05985
作者: Niels Potters,Theo Hofman
类目: Artificial Intelligence (cs.AI); Hardware Architecture (cs.AR); Computational Engineering, Finance, and Science (cs.CE); Systems and Control (eess.SY)
备注:
Abstract:This paper presents a black-box evaluation framework to systematically assess the ability of Large Language Models (LLMs) to generate Design Structure Matrices (DSMs) from structured technical documentation. Motivated by the closed-source nature of current Auto-DSM pipelines, the framework introduces a reproducible methodology that benchmarks generated DSMs (GEN-DSMs) against manually validated ground-truth matrices (GT-DSMs). The evaluation integrates both single-run and multi-run perspectives, combining structural metrics (Completeness, Correctness, Coupling Density), classification metrics (Selective Accuracy, Abstention Coverage), and stability measures (Entropy, Fleiss’ \kappa ). To synthesize these aspects, a Composite Quality Score (Q) is proposed. Controlled experiments are conducted on two datasets: a fictive abstract system and a real-world refrigerator decomposition, covering variations in phrasing, parameter-dataset alignment, and system complexity. Results show that LLMs can produce structurally plausible DSMs and achieve high reproducibility under well-structured inputs, but remain sensitive to ambiguity, inconsistent dependency definitions, and prompt formulation. The findings highlight systematic sources of hallucination and abstention failure, demonstrating both the potential and current limitations of LLM-driven DSM automation. The proposed framework provides a transparent benchmark for auditing Auto-DSM pipelines and establishes foundations for integrating LLM-based decomposition methods into model-based systems engineering (MBSE) workflows.
[AI-33] Agent ic AI for IPoDWDM Network Lifecycle Automation: An MCP-Enabled Architecture
链接: https://arxiv.org/abs/2607.05958
作者: Chunmin Xia,Jakub Harbaczewski,Nikhil Dsilva,Julie Raulin,Dominic Schneider,Achim Autenrieth
类目: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI); Signal Processing (eess.SP); Systems and Control (eess.SY)
备注: Accepted for oral presentation at the European Conference on Optical Communication (ECOC 2026)
Abstract:We present a distributed, vendor-agnostic multi-MCP architecture for SDN-based automation and autonomous control of multi-vendor, multi-layer IPoDWDM networks. The framework enables E2E service lifecycle automation, closed-loop cross-layer control using GNPy model and optical telemetry, and is experimentally validated on a IPoDWDM testbed.
[AI-34] Signed-Graph Recommendation as Structural Consistency Maximization
链接: https://arxiv.org/abs/2607.05952
作者: Zifan Wang,Siyu Chen,Wenzhuo Song
类目: ocial and Information Networks (cs.SI); Artificial Intelligence (cs.AI)
备注:
Abstract:While signed social recommendation has shown great potential by modeling both trust and distrust relations, its effectiveness is often hindered by structural noise and data sparsity. In this work, we first identify a fundamental inconsistency across the structural, propagation, and semantic layers of existing models, which leads to biased representations learned from sparse or noisy datasets. Furthermore, we observe that most existing methods treat the observed graph as fixed, failing to bridge the gap between noisy topologies and reliable social semantics. To address these issues, we propose a unified framework named SSC-Loop that treats signed social recommendation as the maximization of structural consistency. SSC-Loop includes three dedicated modules: ESA-DA for structural consistency, a P/N/O propagation mechanism for propagation consistency, and a contrastive learning objective for semantic consistency. Experiments on Epinions demonstrate that SSC-Loop achieves strong performance on explicit signed social rating prediction, while auxiliary results on Slashdot under a derived link-existence setting further suggest its ability to exploit signed social structures. Source code is available at this https URL.
[AI-35] SearchEyes: Towards Frontier Multimodal Deep Search Intelligence via Search World Simulation
链接: https://arxiv.org/abs/2607.05943
作者: Zhengbo Jiao,Yiming Cheng,Yilei Jiang,Kaituo Feng,Rui Huang,Tianyi Jiang,Juanxi Tian,Jiapeng li,Qunzhong Wang,Tailai Chen,Qianshan Wei,Chuan Xiao,Shanyu Rong,Yangfu Li,Yanhan Zhou,Yunpu Ma,Yifan Zhang,Xiangyu Yue
类目: Artificial Intelligence (cs.AI)
备注: Project page: this https URL
Abstract:Training multimodal search agents to perform multi-hop reasoning remains challenging due to a fundamental structural disconnect: existing pipelines construct training data, search environments, and reward signals independently, causing synthesized structural metadata to be discarded, environments to rely on irreproducible external engines, and RL rewards to remain sparse at the trajectory level. We present \textbfSearchEyes, which uses a typed knowledge graph as the backbone of a \emphsimulated search world that unifies all three components. We propose \textbfPerception-Knowledge Chains (PKC) to sample constrained multi-hop paths over the visual-knowledge intersection of Wikidata5M, retaining hop-level entity metadata that simultaneously defines a self-contained search world and step-level reward anchors. We further propose \textbfHop-Anchored Policy Optimization (HaPO), which reuses these anchors for step-level credit assignment without a separately trained process reward model. Experiments on six multimodal knowledge-intensive benchmarks show that SearchEyes achieves state-of-the-art performance among open-source multimodal search agents, with SearchEyes-27B improving over the strongest open-source baseline by 6.2 points on average.%
[AI-36] PCBWorld: A Benchmark Environment for Engine-Grounded PCB Design Automation KDD2026
链接: https://arxiv.org/abs/2607.05915
作者: Hyungseok Song,Junseok Park,Won-Seok Choi,Seohui Bae,Han-Seul Jeong,Youngjoon Park,Soonyoung Lee
类目: Artificial Intelligence (cs.AI)
备注: Accepted to the KDD 2026 Workshop on Evaluation and Trustworthiness of Agentic AI (non-archival). Main text with appendix
Abstract:PCB routing is the task of connecting the nets of a board with copper traces under strict design rules, yet learning-based methods still lag behind rule-based routers. We introduce PCBWorld, an open-source engine-grounded PCB routing environment built on the KiCad EDA engine. As a human engineer does, agents in PCBWorld interactively route a board through the engine’s native operations, using its Design Rule Check (DRC) feedback to keep the routing within the design rules. The environment supports both RL policies and tool-using LLM agents. Alongside the environment, PCBWorld-Bench provides three dataset families in KiCad’s native board format (.kicad_pcb), covering two types of controllable synthetic instances and 679 real open-source boards. It scores any completed board with eight engine-checked evaluation metrics, regardless of the routing method. In our experiments, agents in PCBWorld consistently outperformed grid-action RL policies and open-loop LLM baselines, and an RL policy trained only on synthetic boards transferred zero-shot to real boards, approaching rule-based routers. These results position the engine-grounded, interactive approach of PCBWorld as a promising foundation for advancing the routing ability of both RL and LLM agents.
[AI-37] From Textural Counterpoint to Feature Encoding: A Multi-Dimensional Machine Representation Study of Haydns “The Lark” Integrating Electroacoustic Analysis
链接: https://arxiv.org/abs/2607.05902
作者: Yakun Liu,Zhiyu Jin,Hai Luan,Dong Liu,Xiaonan Li
类目: ound (cs.SD); Artificial Intelligence (cs.AI)
备注:
Abstract:Chamber music, as a highly precise multi-part interactive system, contains a logic of “role assignment and dynamic interaction” that provides an extremely valuable blueprint for exploring human-computer collaborative composition paradigms. Addressing the lack of role perception capabilities in existing deep music generation models during polyphonic interactions, this paper conducts an interdisciplinary analysis of Haydn’s String Quartet in D Major, The Lark (Op. 64, No. 5). We propose a novel research path: “Classical Morphology Qualitative Analysis-Electroacoustic Quantitative Measurement-Machine Representation Reconstruction.” The study first utilizes auditory analysis to dissect the counterpoint morphology of the leading voice and the underlying groove in the first movement. Subsequently, it introduces spectrum and dynamic feature analysis tools from a Digital Audio Workstation (DAW) to translate subjective auditory perception into objective, measurable physical parameters. Building on this, the paper introduces a fundamentally new approach to low-level computer feature extraction: completely abandoning the traditional mechanical quantization grid, introducing Event-based Timestamps to record the duration of micro-timing, and transforming acoustic features into an independent “Role-Aware Encoding” as an aesthetic heuristic mechanism (a phenomenological anchor). This study not only completes the logical loop spanning classical analysis, electronic music mapping, and AI symbolic generation but also establishes a profound theoretical foundation-from the perspectives of interactive aesthetics and media philosophy-for constructing human-computer collaborative music systems imbued with “social attributes” and “otherness awareness.”
[AI-38] Uncovering Latent Depression Severity for Binary Depression Detection via Advantage-weighting Ranking
链接: https://arxiv.org/abs/2607.05901
作者: Manning Gao,Tingyi Liu,Leheng Zhang,Haifeng Hu,Yuncheng Jiang,Sijie Mai
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Automatic depression detection using audio-visual data faces significant challenges, particularly in disentangling overlapping feature distributions and establishing robust decision boundaries. To address this, we propose a fine-grained multimodal framework featuring a temporal encoder and a mutual transformer to facilitate deep cross-modal fusion. Our core contribution is the Binary Advantage-weighting Ranking Loss, which optimizes the latent space distribution through two complementary mechanisms: Advantage-weighted Separation, which mines hard pairs by computing a pairwise prediction difference matrix and dynamically weighting them based on their difficulty; and Advantage-weighted Compactness, which minimizes intra-class variance to force features to cluster around their respective class centers. Extensive experiments on D-vlog and LMVD demonstrate that our model reconstructs the latent ordinal structure by prioritizing hard pairs, thereby achieving state-of-the-art performance.
[AI-39] -EXAM: Instructable and Explainable Attack Connectivity Graph Modeler ICAPS2026
链接: https://arxiv.org/abs/2607.05888
作者: Rakesh Podder,Wadia Ganim,Sarath Sreedharan,Indrajit Ray,Indrakshi Ray
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: In the Proceedings of the International Conference on Automated Planning and Scheduling (ICAPS 2026)
Abstract:i-EXAM is a planning-powered tool that helps system administrators to create security profiles of complex networks and perform what-if analyses to identify network hardening strategies. It leverages planning compilation that provides soundness and completeness guarantees to identify attack paths, evaluate security metrics, generate diverse hardening strategies, and explain these strategies in natural language using Large Language Models.
[AI-40] hink Before You Grid-Search: Floor-First Triage for LLM Serving
链接: https://arxiv.org/abs/2607.05876
作者: Yihua Liu
类目: Performance (cs.PF); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC)
备注: 16 pages, 3 figures
Abstract:LLM serving optimization typically benchmarks many configurations and reaches for heavy profilers when latency targets are missed. We argue for the reverse discipline: estimation is the analytical layer of profiling – without it, optimization degenerates to grid search. Floor First is a residual-driven triage workflow. Each decode step is modeled as a five-dimensional resource vector (HBM bytes, FLOPs, network bytes, network messages, KV capacity); summing within a resource and maximizing across resources gives an optimistic floor, the plain sum a pessimistic one. Where a measurement lands inside this [max, sum] interval reads out overlap quality before any profiler is opened, and profilers escalate only on residuals above a stated threshold. Deployment alternatives are compared by wall ordering – which resource wall binds first as load grows – rather than by point benchmarks. The account is compositional: new attention or state-space variants enter by declaring one module, and the workflow ships as a zero-dependency calculator plus an agent skill that enforces the discipline in agentic optimization loops. As a case study we analyze a DeepSeek-V3.2-style 671B MoE/MLA model on 16 NVIDIA H20 GPUs, whose ridge point of ~74 FLOP/byte (vs ~590 for H100) makes it an extreme decode-oriented part. The floors show TP16 decoding is KV-capacity-limited to ~70 concurrent 8K requests; sparse attention removes the KV-bandwidth term but not the capacity wall; an EP16+DP-attention layout accepts slightly worse same-batch weight traffic for an order-of-magnitude higher capacity wall (~644) – while single-stream latency favors TP by 2.4x. The layout judgment is thus a computable function of the operating point, explaining why production deployments on identical hardware have shipped opposite attention layouts. Comments: 16 pages, 3 figures Subjects: Performance (cs.PF); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC) Cite as: arXiv:2607.05876 [cs.PF] (or arXiv:2607.05876v1 [cs.PF] for this version) https://doi.org/10.48550/arXiv.2607.05876 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-41] Differentially Private Natural Gradient Descent
链接: https://arxiv.org/abs/2607.05866
作者: Pan Li,Kai Chen,Shuai Chang,Shengzhi Zhang,Peizhuo Lv,Jinwen He
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Under a fixed privacy budget, the utility of differentially private (DP) training is ultimately determined by its optimization efficiency. Standard first-order DP optimizers such as DP-SGD rely solely on local gradients and ignore the underlying loss curvature. This geometric blindness causes severe zigzagging in ill-conditioned landscapes, squandering precious privacy budgets on inefficient iterations. Practitioners are thus trapped in a bind: either stop training prematurely or inject massive per-step noise, both of which critically compromise final model utility. Natural Gradient Descent (NGD) resolves this by preconditioning gradients with curvature, aligning updates with the loss geometry and extracting more efficient signal from every noisy step, offering a principled pathway to break the privacy-utility bottleneck. Despite its theoretical appeal, directly integrating NGD with DP introduces fundamental challenges: curvature estimation itself consumes prohibitive privacy budgets, isotropic DP operations conflict with the anisotropic scaling of NGD, and the inverse curvature catastrophically amplify parameter updates in flat directions, causing training instability. We propose DP-NGD, a practical framework that systematically addresses these obstacles by decoupling curvature estimation from private data, reconciling isotropic DP constraints with anisotropic second-order optimization via a whitened-space mechanism, and dynamically clamping the curvature to stabilize training. Extensive experiments on standard benchmarks demonstrate that DP-NGD achieves state-of-the-art accuracy, breaking through the utility ceilings of first-order baselines while delivering up to a 10\times convergence speedup under the same privacy budget. Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI) Cite as: arXiv:2607.05866 [cs.LG] (or arXiv:2607.05866v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2607.05866 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-42] Unsupervised Anomaly Detection of Information Operations Users via Behavioral and Language Patterns ECML KDD2026
链接: https://arxiv.org/abs/2607.05855
作者: Sishun Liu,Sajal Halder,Ke Deng,Yan Wang,Xiuzhen Zhang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Accepted at ECML/PKDD 2026
Abstract:Information Operations on social media networks have been identified as a significant threat to democracy and modern society, but they are challenging and expensive to detect by humans. Existing supervised IO detection methods fail to capture the dynamic nature of evolving IO user behavior, while existing unsupervised approaches rely on oversimplified assumptions of coordination among IO users that may not exist in practice. To overcome the limitations of existing methods, we formulate IO user detection as an anomaly detection problem and propose a novel unsupervised IO user detection approach called Temporal-bEhavior-laNguage Signals for information Operation Recognition (TENSOR), which leverages multimodal data, including temporal online user behavior, such as message posting activities, and the textual content of the messages. The motivation is that IO users are typically a very small fraction of all online users and have unique temporal behavioral and language patterns. Specifically, we train a Temporal Point Process (TPP) to capture abnormal temporal behavioral patterns of IO users because they are known to behave in a coordinated manner for IO campaigns. We further introduce a novel evidence function that converts LLM responses, which are generated from user post timelines, into quantitative scores to adjust the TPP outputs for better IO user detection. Experimental results show that TENSOR outperforms the baselines on five real-world IO datasets. Code is available at this https URL.
[AI-43] AbICL: In-Context Learning for Antigen-Specific Antibody Affinity Ranking
链接: https://arxiv.org/abs/2607.05846
作者: Zhiyuan Chen,Jing Hu,Junzhe Wang,Yueyang Huang,Xinyi Yang,Zhaoyang Wang,Feng Zhu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE); Quantitative Methods (q-bio.QM)
备注:
Abstract:Accurate ranking of antibody candidates according to their binding affinity is essential for therapeutic antibody discovery. However, existing methods treat affinity comparisons independently and ignore the contextual information encoded in other labeled comparisons, limiting their ability to capture antigen-specific binding landscapes. For many target antigens, a small number of experimentally characterized affinity comparisons are often available. An important question is whether the model can exploit these existing comparisons to infer antigen-specific ranking patterns that facilitate subsequent affinity ranking. This form of learning from labeled demonstrations closely resembles the paradigm of In-Context Learning, motivating us to revisit antibody affinity ranking from an ICL perspective. To this end, we propose AbICL, an ICL framework for antigen-specific antibody affinity ranking. AbICL combines a pretrained structural encoder with a context ranking head and is trained with an episodic meta-training strategy that enables the model to leverage support demonstrations for test-time adaptation without gradient updates. Experiments on the AbRank benchmark demonstrate that AbICL consistently outperforms existing ranking baselines across almost all data splits and evaluation benchmarks. Further analysis shows that the value of contextual demonstrations depends on how well they match the target inference task, and becomes increasingly pronounced under distribution shift and fine-grained affinity discrimination. These findings highlight the potential of ICL as an effective paradigm for antigen-specific antibody affinity ranking, particularly in challenging settings where a single global ranking function is insufficient.
[AI-44] Beyond Refusal: A Same-Lineage Study of Aligned and Abliterated LLM s for Vulnerability Analysis
链接: https://arxiv.org/abs/2607.05842
作者: Mingchen Li,Meikang Qiu,Zifan Peng,Heng Fan,Song Fu,Junhua Ding,Yunhe Feng
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注:
Abstract:Large language model (LLM)-assisted software security operates at a difficult boundary: the vulnerability-analysis terminology needed for legitimate code review, triage, and repair can closely resemble terminology associated with misuse. Existing safety and cybersecurity evaluations are difficult to interpret in this setting because they often compare unrelated model families, thereby conflating safety behavior with differences in architecture, scale, training data, and deployment. To isolate this factor, we study safety state: whether refusal behavior remains intact (Aligned) or has been refusal-ablated (Abliterated) within same-lineage models. We ask how this safety state affects defensive utility across software-security workflows. We compare aligned instruction-tuned models with publicly released refusal-ablated descendants from two model families, Gemma and Qwen. We evaluate Aligned and Abliterated states on vulnerability detection, CWE attribution, vulnerable-line localization, root-cause localization, and executable patch validation. We further treat prompt wording as a controlled framing dimension: prompts begin with neutral code-review language, add authorization context, and vary the density of cybersecurity terminology. In a Gemma-based Java/Vul4J repair-validation study, Abliterated achieves higher early-stage validation rates, with 67.8%, 65.0%, and 32.8% of patches judged usable, successfully applied, and successfully compiled, respectively, compared with 29.9%, 24.9%, and 9.0% for Aligned. In the Qwen pair, Abliterated improves localization performance, increasing line-level F1 from 2.08% to 3.91% and Top-1 accuracy from 4.10% to 6.95%. These findings suggest that evaluations of LLM-based security assistants should jointly measure whether models respond, whether their usable responses are correct, and whether their outputs remain actionable across the engineering workflow.
[AI-45] Decision-Focused Scenario Generation and Selection for Efficient and Robust Grid Dispatch
链接: https://arxiv.org/abs/2607.05830
作者: Yangze Zhou,Yihong Zhou,Thomas Morstyn,Yi Wang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 10 pages, 12 figures
Abstract:The increasing uncertainty from flexible demand and renewable generation has made distributionally robust optimization (DRO) an important tool for robust power system dispatch. DRO relies on forecast scenarios to construct ambiguity sets, but conventional scenario generation pipelines are often trained in an accuracy-oriented manner and may neglect spatial correlations among uncertainties. This mismatch can produce ambiguity sets that are statistically plausible but suboptimal for downstream operation. This work proposes a decision-focused generative framework for correlated scenario generation in DRO-based dispatch. Instead of training generative models solely to fit the historical uncertainty distribution, the proposed framework optimizes generated scenarios according to their induced downstream operational cost. The proposed framework is tailored to mainstream generative models, including variational autoencoders, generative adversarial networks, and diffusion models, while capturing the joint distribution of uncertainties across buses. To improve computational tractability, we further develop a differentiable scenario selector that selects decision-relevant scenarios from a generated pool and can be trained within the same decision-focused pipeline. Case studies demonstrate that the proposed framework effectively reduces 0.80%-2.02% operational cost across different generative models compared to accuracy-oriented methods.
[AI-46] Onnes: A Physics-Grounded Multi-Agent LLM Simulator for Cryogenic Fault Diagnosis in Quantum Computing Infrastructure
链接: https://arxiv.org/abs/2607.05805
作者: Praneeth Narisetty,Uday Kumar Reddy Kattamanchi,Shiva Nagendra Babu Kore
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Quantum Physics (quant-ph)
备注: 18 pages, 14 figures, 10 tables. Code, data, and released run logs: this https URL
Abstract:Dilution refrigerators are the enabling infrastructure of superconducting quantum computers, yet their fault diagnosis is still dominated by threshold alarms that report that something is wrong, not what. We present Onnes, a physics-grounded digital-twin simulator of a dilution refrigerator (a forward physics model with a learned real-fridge noise fingerprint) that drives a live multi-agent LLM operations layer, and use it for a controlled head-to-head between a zero-shot LLM agent panel and a supervised ML classifier on cryogenic fault diagnosis. The twin couples a real dilution-cooling floor, a noise-and-correlation fingerprint learned from real BlueFors logs, and six physics-grounded fault classes, three engineered to overlap on temperature but separate on flow and pressure. Across a 1000-turn evaluation the zero-shot panel shows no significant difference from the classifier on detection but trails on classification, its errors concentrating on the confusable faults. Curated contrastive few-shot demonstrations and self-consistency voting then raise classification accuracy from 0.685 to 0.990, matching the supervised classifier (0.985) with no parameter updates and six labeled demonstrations; an ablation attributes the gain almost entirely to the demonstrations. Run as a continuous monitor across a nine-run fault-by-seed sweep, the agent catches every developing fault within one poll interval, and a confidence gate suppresses pre-onset false alarms whose rate is backend-dependent. As a first sim-to-real check, a detector trained purely on real BlueFors telemetry posts a real-hardware false-alarm rate of 6.4% and 100% recall on physics faults injected onto real held-out windows. All numbers are drawn verbatim from released run logs.
[AI-47] From Passive Retrieval to Active Memory Navigation: Learning to Use Memory as a Structured Action Space
链接: https://arxiv.org/abs/2607.05794
作者: Yue Xu,Yutao Sun,Yihao Liu,Mengyu Zhou,Jiayi Qiao,Lu Ma,Kai Tang,Wenjie Wang,Xiaoxi Jiang,Guanjun Jiang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Long-term user memory is essential for personalized conversational agents, yet many memory systems still expose memory through passive retrieval interfaces, making the model a consumer of pre-selected evidence. We introduce NapMem, a framework for learning to use long-term user memory as a structured action space rather than passively retrieved context. NapMem organizes user history into a linked multi-granularity memory pyramid, where raw conversations, typed memory records, topic tracks, and user profiles are connected through provenance relations, and exposes these levels through memory tools. The agent is trained to select memory according to the query and intermediate evidence, allowing it to inspect different memory granularities before answering. Experiments on PersonaMem-v2, LongMemEval, and LoCoMo show that a NapMem agent trained with memory-tool reinforcement learning is competitive across diverse memory-intensive tasks, while evaluations on non-memory tasks suggest that the learned policy largely preserves general reasoning and tool-use abilities. Additional analyses examine storage, inference cost, tool-use behavior, and ablations over navigation, memory granularity, and RL training. Our results suggest that long-term user memory benefits from coupling structured storage with a learned policy for using memory at the appropriate granularity.
[AI-48] Controlling Tool Use with Heading-Specific Activation Steering
链接: https://arxiv.org/abs/2607.05790
作者: Yuqi Chen,Vincent Siu,Yang Liu,Dawn Song,Chenguang Wang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Tool-augmented large language models extend their capabilities beyond parametric knowledge through external tools, but tend to invoke them unnecessarily. We investigate whether tool-use decisions have any stable internal representation that can be extracted and manipulated, a question that is non-trivial given that tools exist entirely in context at inference time and have no direct encoding in model weights. We show that steering vectors extracted from heading-anchors positions exert bidirectional causal control over tool-invocation behavior across five open-source models and three domains, suppressing unnecessary tool use most effectively in domains where parametric reasoning suffices. However, geometric analysis reveals that this causal effectiveness does not correspond to clean linear structure: tool-invocation steps exhibit diffuse, bimodal alignment with the suppression vector rather than the consistent negative alignment a linear encoding account would predict, and different tool types recruit largely distinct internal signatures with low cross-tool feature overlap. We hypothesize these geometric properties are indicative of the non-parametric nature of tools, and distinguish tool-use steering vectors from those extracted for parametrically grounded concepts. The relationship between this geometric irregularity and the observed causal effectiveness remains an open question.
[AI-49] Beyond the Leaderboard: A Synthesis of Tool-Use Planning and Reasoning Failures in Large Language Model Agents
链接: https://arxiv.org/abs/2607.05775
作者: Wael Albayaydh,Rui Zhao,Ivan Flechais
类目: Artificial Intelligence (cs.AI)
备注: 16 pages, 3 tables, 1 figure
Abstract:Large language model (LLM) agents are increasingly evaluated on their ability to use tools, plan multi-step tasks, coordinate with other agents, and operate over extended horizons. Reported benchmark gains often obscure recurring failure modes documented across otherwise unrelated evaluation efforts. This paper synthesizes 27 benchmark, taxonomy, and audit papers (2023-2026), spanning 19 distinct benchmarks, into a cross-cutting taxonomy of agent limitations. To our knowledge, this is the first synthesis that integrates evidence across tool use, planning, long-horizon reasoning, multi-agent coordination, safety, and measurement validity into a single, unified taxonomy of LLM agent limitations. We identify six failure clusters: (1) tool invocation and parameter-level errors, (2) planning and constraint-satisfaction failures, (3) long-horizon degradation from context accumulation, (4) multi-agent coordination failures, (5) safety and security failures under adversarial or underspecified conditions, and (6) measurement validity problems. The taxonomy was derived iteratively by grouping independently reported error categories into themes corresponding to distinct stages of the agent reasoning-to-action pipeline. Across the literature, we find that failures compound nonlinearly with task length, that strong performance on individual sub-tasks does not reliably translate into end-to-end success, and that additional scaffolding does not consistently improve reliability. At the same time, substantial progress has been demonstrated in single-turn tool use, short-horizon web navigation, and narrowly scoped coding tasks.
[AI-50] Beyond Static Evaluation: Building Simulation Environments for Scalable Agent ic Reinforcement Learning
链接: https://arxiv.org/abs/2607.05773
作者: Akshay Arora,Ishan Nigam,Ashutosh Aggarwal,Shefali Bansal,Krishna Singh,Sweta Kumari,Nikhil Mittal,Shariq Farhan,Siddarth Malreddy
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:As Large Language Models (LLMs) evolve into autonomous agents, traditional static evaluation fails to capture multi-step decision-making. We introduce AgenticAI-Supervisor, an API and UI-driven RL Gym environment that decouples environment creation from scalable execution. By moving to verifiable execution outcomes, the platform generates high-fidelity traces and applies multi-dimensional reward shaping. Critically, our framework mitigates reward hacking through rigorous internal state validation and testing. This work provides a first look at our platform’s core capabilities through a Customer Support Agent case study demonstrating a consistent closed-loop feedback for model optimization. Future work will focus on advanced features such as Computer Use, Tool Use, automated “stumping”, and edge-case generation.
[AI-51] Synthetic Consumer Insight Generation with Large Language Models
链接: https://arxiv.org/abs/2607.05761
作者: Stephen L. France,Pia. A. Albinsson
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Modern data-driven marketing relies on large amounts of consumer data, yet collecting such data can be costly, time-consuming, and difficult to scale. This research examines whether large language models (LLMs) can be used to generate synthetic consumer data for projective techniques, a set of methods designed to elicit consumer associations, emotions, wants, and needs. We test LLM-generated responses across multiple projective tasks, LLMs, prompting strategies, and temperature settings, and compare them with human responses from a primary research study on perceptions of city tourism destinations. Human and LLM responses were analyzed using linguistic measures, diversity and concentration metrics, topic models, and top-term analyses. The results show substantial overlap between human and LLM responses in broad topics and associations, but also important differences in style, linguistic structure, and the way diversity is generated. Recommendations are given on how to best utilize LLMs for generating synthetic consumer data, how model and prompt choices shape response quality, and on recognizing the limitations of LLM synthetic consumer data generation.
[AI-52] Data-dependent Evaluations for Budgeted Submodular Maximization
链接: https://arxiv.org/abs/2607.05759
作者: Lejian Zhang,Xueyan Tang,Jing Tang
类目: Data Structures and Algorithms (cs.DS); Artificial Intelligence (cs.AI); Discrete Mathematics (cs.DM)
备注: Extended version of a paper that will appear in ESA 2026 conference
Abstract:Submodular maximization is an important building block for developing algorithms in many areas such as machine learning and data mining. Due to the NP-hardness of the problem, analysis of submodular maximization algorithms typically provides pessimistic worst-case approximation factors only. It is not easy to evaluate how close a produced solution is to an optimal one for a given problem instance. In this paper, we develop new data-dependent upper bounds for submodular maximization with a knapsack constraint. We theoretically prove that they dominate the optimal solution and empirically demonstrate their advantages in certifying how close to optimal a solution is through experiments with real-world datasets.
[AI-53] ArtisanCAD: An Industrial-Level CAD Agent with Expert-Grounded Knowledge Distillation
链接: https://arxiv.org/abs/2607.05750
作者: Yunhan Xu,Qifeng Wu,Xunjin Li,Yuanwei Bin,Qingsong Yao,Jianghang Gu,Guan Wang,Weihao Lv,Huiyu Yang,Wenfa Luo,Jiao Xiang,Yuntian Chen,Shiyi Chen
类目: Artificial Intelligence (cs.AI); Graphics (cs.GR)
备注:
Abstract:Computer-aided design (CAD) for industrial components requires long-horizon procedural modeling, robust feature dependencies, editable parametric geometry, and production-grade B-Rep execution. Existing text-to-CAD methods have made promising progress in generating CAD programs from natural-language descriptions, but they still struggle when user prompts are ambiguous, underspecified, or only describe high-level design intent. They also rarely exploit expert procedural knowledge naturally available in industrial workflows, such as CATIA operation recordings, macro logs, drawing notes, and engineering descriptions. We present \algname, a skill-guided industrial CAD agent with expert-grounded knowledge distillation. The core of \algname is CAD intermediate representation (CAD-IR), an executable procedural representation that encodes parameters, ordered operations, MCP tool bindings, dependencies, generated entities, and verification rules. CAD-IR plays two key roles: it first serves as the carrier for distilling expert CAD procedures into reusable parameterized skills; then it provides a procedural scaffold that turns vague or intermediate-level prompts into complete executable CAD operations. \algname retrieves expert-derived skills, instantiates and revises CAD-IR, executes the resulting procedure through a dedicated CATIA-MCP backend, and uses multi-view visual feedback for iterative refinement, and finally generates production-ready B-Rep models. On the Text2CAD benchmark, CAD-IR improves generation from intermediate prompts by reducing mean Chamfer Distance from 14.83 to 9.88 , showing its ability to bridge ambiguous textual intent and executable CAD construction. On four complex automotive components, CAD-IR enables expert CATIA recordings to be distilled into reusable skills, allowing \algname to generate editable CATIA-native B-Rep models for new variant requests.
[AI-54] Unicode TAG-Block Concealment of Tool-Metadata Payloads in the Model Context Protocol: An Approval-View Fidelity Gap Across Three Independent Server Implementations
链接: https://arxiv.org/abs/2607.05744
作者: Mohammadreza Rashidi
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注: 15 pages, 4 figures, 7 tables, 5 listings. Real-protocol proof-of-concept, 8 techniques across 3 independently developed MCP server libraries with 32 of 32 cross-library outcome cells agreeing, and 0 of 25 baseline false positives on a benign corpus. Data, harness, and fail-closed verifier released as a supplementary artifact
Abstract:The Model Context Protocol (MCP) is the dominant way coding agents discover and invoke external tools. A server advertises each tool through a tools/list handshake that returns a name, a natural-language description, and a JSON input schema. The client renders this metadata once, in a one-time approval dialog, and then injects it verbatim into the model’s context on every subsequent turn. Nothing in the protocol requires the rendered approval view and the bytes delivered to the model to match. We isolate that gap as a single structural mechanism, concealment encoding, and show with a model-free, protocol-free analysis that Unicode’s TAG block (U+E0000 to U+E007F) has no assigned glyph in any mainstream terminal, chat, or IDE renderer, so a payload written in it is absent from what a human reviewer sees while surviving byte-for-byte into the model’s tokenizer. We then measure whether this mechanism actually defeats today’s client-side defenses, building a proof-of-concept that speaks the real MCP JSON-RPC/stdio protocol against a genuine client and server. Across 5 distinct MCP metadata surfaces we implement 8 concrete techniques with a deterministic, protocol-level harness. All 8/8 techniques deliver an attacker-controlled payload into the model’s context, 4/8 evade a representative string-matching sanitizer, and exactly as the mechanism analysis predicts, only the TAG-block encoding (1/8) is invisible in the human approval view while still reaching the model verbatim. MCP forces re-approval for 0/8 techniques even under a time-of-check to time-of-use rug-pull. To test whether these outcomes are a property of the protocol or an artifact of one server codebase, we re-implement the catalogue against 3 independently developed Python MCP server libraries and find total agreement across all 32 cross-library outcome cells. The baseline sanitizer flags 0 of 25 benign descriptions.
[AI-55] he Balkanization of Execution-Security Research for AI Coding Agents : Isolation Access Control and Time-of-Check-to-Time-of-Use Vulnerabilities
链接: https://arxiv.org/abs/2607.05743
作者: Mohammadreza Rashidi
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: 18 pages, 15 figures, 6 tables. Systematizes 39 execution-security papers (2023-2026) into 17 verified categories. Machine-readable corpus and verification script released as a supplementary artifact
Abstract:AI coding agents now read repositories, call tools, and execute shell commands with limited human oversight, and a fast-growing body of work studies whether the execution layer around them is actually safe. That literature is scattered. Papers on sandbox isolation, capability and access control, policy enforcement, time-of-check-to-time-of-use (TOCTOU) races, Model Context Protocol (MCP) threats, identity delegation, execution provenance, network egress control, and static analysis of agent-generated code are published independently and rarely cite one another. We systematize 39 papers published between 2023 and 2026 into 17 categories, each verified directly against its source. The same verification protocol also confirms four disclosed, patched CVEs directly affecting production agent harnesses. Reading across categories surfaces five cross-cutting gaps that no single paper addresses. (1) Isolation architectures and capability models are almost never evaluated against one another on a shared benchmark. (2) Policy-enforcement studies report failure rates from 69% to 98% of real denylists, yet no isolation paper re-evaluates its own defense under that adversarial setting. (3) TOCTOU and MCP threats are analyzed as separate literatures despite both being instances of the same state-validation problem. (4) Every enforcement mechanism assumes an honest policy author, leaving policy-authoring error itself unaddressed. (5) Benign but out-of-scope agent actions occurring at rates up to 17.1% under realistic prompting are addressed by no access-control or capability paper in the corpus. Existing broader surveys of agentic AI security discuss sandboxing only as one item among many defenses, leaving execution security without a dedicated systematization. This paper is written to fill that gap. We conclude with a research agenda directed at the five gaps.
[AI-56] Akashic: A Low-Overhead LLM Inference Service with MemAttention
链接: https://arxiv.org/abs/2607.05708
作者: Yang Liu,Zhaokai Luo,Huayi Jin,Ruozhou He,Chenchen Hong,Zhiyong Wang,Yifei Liu,Yunfei Gu,Chentao Wu,Junhao Hu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Recent LLM-based agent systems continuously accumulate context across multi-turn interactions, tool invocations, and cross-session workflows. Replaying the full history for every request quickly becomes impractical: long contexts increase prefill cost, may exceed context limits, and often bury task-relevant evidence in irrelevant content, degrading both serving efficiency and output quality. We propose Akashic, a low-overhead memory system built around MemAttention, which organizes context into bounded chunks and models semantic relationships across chunks, preserving cross-chunk evidence without repeatedly rewriting the full history. Akashic further applies hardware-software co-designed memory placement to co-locate likely co-retrieved chunks, reducing retrieval fragmentation and I/O overhead. Across four representative workloads and three model sizes, Akashic improves task accuracy by up to 10.2 points, throughput by up to 1.21x, and sustainable request rate by up to 1.88x over strong prior memory baselines.
[AI-57] FirstResearch: Auditable Question Formation for LLM Scientific Discovery Agents
链接: https://arxiv.org/abs/2607.05682
作者: Yufeng Wang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:LLM systems for scientific discovery increasingly assist with ideation, literature synthesis, experiment planning, and report generation, but the first research question they propose can remain difficult to audit: it may sound plausible without exposing the mechanism, falsifier, or assumption that a scientist should inspect. We introduce FirstResearch, a first-principles research-question formation framework for scientific LLM agents whose core artifact is a structured Research Question Certificate. The certificate records primitive definitions, assumptions, a mechanism model, a tension or contradiction, a falsifiable hypothesis, a minimal decisive test, and a failure update rule, making the proposed question inspectable before downstream execution. On ten LLM-agent research topics, FirstResearch outperforms controlled prompt-level baselines inspired by AI co-scientist, Agent Laboratory, and AI Scientist-v2 under a primary DeepSeek-blind-judge protocol. A Gemini-2.5-Flash independent-judge rescore of the same 40 baseline packages preserves the system-level ranking, with FirstResearch scoring 4.86/5 versus 4.38/5 for the strongest baseline and Pearson agreement of 0.865 on average score. A one-repeat ablation checkpoint further suggests that the certificate-centered core is the strongest component: certificate-only scoring reaches 4.90/5 under DeepSeek and 4.88/5 under Gemini, while removing certificates drops below 1/5 under both judges. These results are preliminary and use LLM judges rather than human domain experts, but they support a narrow scientific-discovery claim: explicit derivation constraints are a promising mechanism for making LLM-generated scientific questions more auditable. Code, prompts, saved outputs, and reproduction scripts are available at this https URL.
[AI-58] What Do AI Agents Actually Change? An Empirical Taxonomy of Mutation Patterns in Performance-Improving Pull Requests
链接: https://arxiv.org/abs/2607.05666
作者: Illia Dovhoshliubnyi,Nima Soroush,Ashkan Sami,Alexander Brownlee
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:AI coding agents are black boxes: we cannot inspect how they generate code, but we can inspect what they change. This distinction matters for search-based software engineering (SBSE), where techniques such as genetic improvement (in the performance-optimisation application we study) depend on mutation operators that reflect how code is actually transformed. Fewer than 1% of the 33,596 agent PRs in AIDev-pop target performance, making each case a rare window into otherwise opaque agent behaviour. We classify 1,254 performance-relevant diff hunks from 216 of these PRs, spanning five agent systems, against the 18-category syntactic mutation taxonomy of Even-Mendoza et al. (2025) using a dual-LLM intersection pipeline. Three categories dominate: name modification (37.0%), object creation (26.4%), and type change (22.7%), a profile markedly different from prior GI corpora where no change accounted for 84%. Each agent’s deployed system commits to a distinctive mutation vocabulary, and each performance strategy activates a largely disjoint category subset. Agent identity and target strategy are therefore informative priors that narrow the effective SBSE operator space. Replication package: this https URL
[AI-59] Physics-Regularized Machine Learning for Proprioceptive Vehicle Localization Using Onboard Sensors IROS2026
链接: https://arxiv.org/abs/2607.05663
作者: Abinav Kalyanasundaram,Karthikeyan Chandra Sekaran,Wolfgang Utschick,Michael Botsch
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted at the 2026 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2026). 8 pages, 4 figures
Abstract:Accurate and robust localization is essential for autonomous mobility systems in real-world environments. While fusing Inertial Measurement Unit (IMU) data with satellite-based correction signals provides precise vehicle pose estimates, performance degrades substantially during outages. Recent studies indicate that Machine Learning (ML) can improve IMU-based proprioceptive localization, highlighting untapped potential for onboard sensors readily available in production vehicles. This paper introduces Physics-Regularized Machine Learning for Localization (PRML2), a hybrid framework that combines the complementary strengths of Kalman filtering and data-driven learning to estimate vehicle pose directly from onboard sensors. A key aspect of PRML2 is its physics-regularized learning, enabled by end-to-end training of an ML model through a differentiable Kalman filter. This improves consistency with vehicle motion models, thereby enhancing both localization accuracy and generalization across driving conditions. We evaluate the performance limits of ML-enhanced onboard odometry on a publicly available dataset and show that PRML2 achieves superior localization accuracy and demonstrates real-time capability. This work also introduces a novel dataset to support vehicle localization research under low-friction conditions. The proposed framework provides a robust and cost-effective solution for vehicle localization under degraded sensing conditions by integrating learning with physics-based priors.
[AI-60] EvalLoop: A Methodology for Evaluation-Driven Iterative Improvement of Business AI Systems
链接: https://arxiv.org/abs/2607.05638
作者: Kenneth Benavides,Josh Fleischer,Danti Chen
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:Teams deploying large language models in business contexts need evaluation systems, yet most treat evaluation as static model selection: run benchmarks, rank models, deploy the winner. This framing misses evaluation’s primary value for production systems–diagnosing why a system underperforms and guiding what to fix. We present EvalLoop, a methodology for evaluation-driven iterative improvement. EvalLoop organizes evaluation around three mechanisms: (1) dimensional metric grouping that decomposes quality into business-relevant dimensions enabling orthogonal failure diagnosis; (2) failure mode classification that categorizes why outputs fail within weak dimensions, bridging diagnosis to action; and (3) a structured iteration workflow where each evaluation run varies one system variable and compares dimensional profiles before and after. We validate EvalLoop through a case study on sales intelligence briefing generation (10 models, 3 providers, 18 metrics, 5 dimensions, 3 iterations). Dimensional diagnosis identified that 69% of hallucination failures were prompt-induced interpretation errors–invisible in aggregate scoring. A targeted prompt fix improved the best model from 82.6% to 94.6% overall, with improvement concentrated in diagnosed dimensions (Content Accuracy +16.8pp, Synthesis Power +26.4pp). An undirected configuration change in a prior iteration produced zero impact, illustrating the cost of iterating without diagnosis. We additionally demonstrate that dimensional profiling enables deployment-specific model selection, and that a one-time blind human gate on a finalist panel (4 models, 16 cases) confirms dimensional rankings while resolving multi-criteria deployment trade-offs–a 94% reduction in review burden compared to evaluating the full design. EvalLoop is packaged as reusable artifacts (playbook, agent specification, template repository) for adoption by other teams.
[AI-61] Safe Bayesian Optimization with Counterfactual Policies
链接: https://arxiv.org/abs/2607.05620
作者: Katherine Avery,Bruno Castro da Silva,David Jensen
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 10 pages main text, 20 pages total
Abstract:In many decision-making settings, new interventions are acceptable only if they do not reduce outcomes below some established threshold. For example, in clinical medicine, new treatments are often acceptable only if they do not worsen outcomes relative to an established standard of care. Safe Bayesian optimization maximizes an objective subject to safety constraints. In the setting that we consider here, safety is defined relative to a known baseline policy whose outcomes are counterfactual and therefore unobserved. Thus, the counterfactual outcomes of the baseline policy must be estimated and those (uncertain) estimates must be used to safely optimize the objective. We address this estimation problem by using conformal prediction to construct valid uncertainty intervals for counterfactual baseline outcomes, and we show how these intervals can be integrated into safe Bayesian optimization to ensure that constraint violations occur at or below a user-specified rate. We also show how to adapt these conformal estimates to different kinds of covariate shift. We provide a safety proof, experimental evidence, and a sensitivity analysis.
[AI-62] Whose fairness? Structural concentration in AI bias research
链接: https://arxiv.org/abs/2607.05574
作者: Abhash Shrestha,Subigya Gautam,Anu Sapkota,Sanju Tiwari,Tek Raj Chhetri
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Digital Libraries (cs.DL)
备注: 27 pages, including 5 composite figures comprising 16 individual figures. Code is available at this https URL , and the interactive atlas is available at this https URL
Abstract:Artificial intelligence increasingly mediates consequential decisions in healthcare, law, and public services, and the field has responded with an extensive methodology for measuring and mitigating bias. Yet the fairness definitions, benchmarks, and debiasing frameworks on which this methodology rests are treated as universal while being produced by a research community whose composition has never been characterized. We show that the AI bias research are structurally concentrated, and that this concentration is greatest, geographically, in precisely the domain the rest of the field inherits from. Analyzing 692 publications spanning five thematic domains, combining bibliometric analysis with semantic clustering, we find that research activity is dominated by a small set of countries, institutions, and authors, with the United States leading publication output and collaboration networks across every domain and most strongly in general fairness and bias mitigation, the largest, most-cited domain with meaningful representation across all four semantic clusters. Low- and middle-income countries remain largely absent from the community and its collaboration networks, and citation influence is highly skewed (median = 9; mean =93.5 ), indicating that a small fraction of publications disproportionately shapes the field. Because the general-fairness domain supplies the definitions and benchmarks that application areas apply, concentration of research effort in this foundational domain propagates across AI bias research as a whole - raising the concern that mitigation methods developed and validated within a narrow set of contexts may not generalize to all populations and settings where AI is deployed. We provide an interactive atlas for continuous monitoring of the field’s structure.
[AI-63] Foundation Models for Automatic CAD Generation
链接: https://arxiv.org/abs/2607.05573
作者: J de Curtò,Victoria Guillén,I. de Zarzà
类目: Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE)
备注: Accepted as a book chapter in “Advances in Global Applied Artificial Intelligence” (G. A. Tsihrintzis, M. Virvou, N. G. Bourbakis, L. C. Jain, Eds.), authenticated version will be published in Springer series: Learning and Analytics in Intelligent Systems
Abstract:Recent advances in Large Language Models (LLMs) and Vision-Language Models (VLMs) enable the automatic generation of parametric 3D designs from natural-language specifications. This chapter presents an empirical study of foundation models for automatic Computer-Aided Design (CAD) generation of mechanical parts, using a unified evaluation pipeline and a curated benchmark of 97 engineering design problems. We introduce LLMForge, a multi-model text-to-CAD framework integrating JSON-schema validation, analytic feature scoring, mesh synthesis, and multi-round iterative refinement, studied under two critique regimes. IterTracer uses a Phong-shaded ray-trace renderer with analytic visual metrics (silhouette IoU, hole visibility, edge clearance, aspect-ratio conformance) for lightweight geometry-aware feedback across rounds. IterVision replaces the analytic scorer with a VLM semantic critic (Qwen2.5-VL-72B) that evaluates rendered views via chain-of-thought visual reasoning, assessing spatial coherence and design intent. On a benchmark spanning four canonical geometry families (plates with holes and bolt circles, multi-feature boxes, flanged cylinders, and L-brackets), we evaluate seven foundation models: DeepSeek-V3.2, Qwen3-235B-A22B, Llama-3.3-70B, Gemma-3-27B, GLM-4.5, MiniMax-M2.1, and INTELLECT. Under IterTracer, the four highest-ranked models form a tight cluster (overall mean in [0.885, 0.890]) with 98.97% mesh success, showing that compact instruction-tuned models can match substantially larger systems. VLM-based critique in IterVision yields 100% watertight mesh generation on the leading model while surfacing systematic difficulty on rotationally symmetric geometries such as cylinders, where visual and semantic scoring diverge most. We discuss benchmark design, failure modes, CAD-oriented prompting, and implications for industrial workflows and scalable automated mechanical design.
[AI-64] From Graphs to Gradients: Physics-Inspired Structural Attribution for Cyber-Physical IoT Systems and Beyond
链接: https://arxiv.org/abs/2607.05563
作者: Spyridon Evangelatos,Christos Diou,Georgios Th. Papadopoulos,Evangelos Markakis,Panagiotis Sarigiannidis
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Interpretable explanation methods in Artificial Intelligence aim to uncover the underlying causes and their effects, enabling a deeper understanding of why a system behaves in a certain way under different inputs. Unlike traditional explainability methods, which mainly highlight correlations between input and output variables, causal explanation focuses on interventional questions. By doing so, it provides more robust insights, helping users understand automated decisions, especially in high-risk domains. Recovering an explicit directed causal structure, however, is often impractical in large-scale, hybrid cyber-physical systems with feedback loops and partial observability. This paper introduces a novel framework inspired by statistical mechanics that instead models variable dependencies through an undirected, energy-based representation of cyber-physical IoT systems. Our approach enables rigorous dependency-aware attribution by analysing how variations in the energy landscape reflect the influence of individual components, without recovering a directed causal graph. It also supports reasoning about perturbation effects across hybrid interactions, providing reliable explanations of abnormal behaviours. We empirically examined our framework through simulations on an industrial IoT testbed with hybrid continuous and discrete variables, demonstrating higher attribution accuracy, improved robustness and better scalability than state-of-the-art graph-based approaches. While the attributions are not intended to fully recover the system’s generative dynamics, they provide valuable, dependency-aware explanations supporting both human interpretation and downstream predictive and diagnostic tasks. Although demonstrated in industrial IoT security, our framework also applies to other high-dimensional cyber-physical and socio-technical systems requiring principled, structural explanations.
[AI-65] Self-Review Reinforcement Learning (SRRL) with Cross-Episode Memory and Policy Distillation
链接: https://arxiv.org/abs/2607.05541
作者: Muhammad Zain Amin,Kibele Sebnem Yildirim
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 9 pages, 2 figures
Abstract:Reinforcement Learning is commonly used to train large language models using environmental feedback. In applied settings, the environment usually provides sparse or delayed feedback. This makes it difficult for the model to pinpoint which actions in its reasoning led to success or failure. So, learning effectively from these signals is hard because the model must determine how each failure should inform meaningful behavioral corrections in subsequent iterations. We introduce a training framework, Self-Review Reinforcement Learning, that embeds an explicit self-review step into each RL episode. When a first-pass response fails, the model generates a self-review to identify what went wrong, which conditions an improved second attempt. Unlike inference-time reflection approaches, such as Reflexion, the framework optimizes self-review with policy gradients and internalizes improvements into the base policy via selective distillation, ensuring they persist across future episodes. A cross-episode memory keeps successful self-reviews for reuse when encountering similar tasks in future episodes during training. We evaluate SRRL against a standard RLVR baseline using the GRPO optimizer across two language models, Qwen 3-4B and OLMo-3- 7B, on GSM8K benchmark. SRRL consistently outperforms the RLVR in final reward performance and achieves greater learning efficiency by successfully transforming feedback into behavioral improvement.
[AI-66] aiAuthZ: Off-Host Identity-Bound Authorization for AI Agents
链接: https://arxiv.org/abs/2607.05518
作者: Sai Varun Kodathala
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: Technical Report
Abstract:AI agents issue tool calls on the basis of text they cannot verify, so any party who controls part of the context can forge the appearance of authority. I evaluate 15 contemporary language models against eight attack scenarios derived from a published corpus of real agent incidents and find that refusal varies from 100% down to 38% across fully evaluated models; the most expensive model refused only half of the attacks despite a twentyfold price spread. I present aiAuthZ, an authorization gateway that moves the safety decision off the agent’s host. Before a tool call executes, the gateway verifies caller identity with a per-message HMAC-SHA256 signature bound to a single-use nonce and a timestamp window, and it evaluates a role-based and argument-level policy that the agent can neither read nor modify. Every decision joins a SHA-256 hash-chained audit log, and each accepted message yields an HMAC-authenticated QR receipt that achieves 94% mean verification across eight transmission channels, with zero forgeries accepted in 25 wrong-key trials. With the gateway in place, residual attack success falls to 0% for all 15 models at no more than 0.03 ms of added decision latency. On the AgentDojo banking suite, aiAuthZ blocks all seven attacker-directed tool calls the evaluated agents emit, at the cost of one legitimate first-time payment, while a spotlighting baseline allows two injections to succeed. Across nine in-scope case studies from the same incident corpus, aiAuthZ blocks nine of nine, against four of nine for a policy baseline without identity binding. The gateway does not prevent a model from being deceived; it prevents a deceived model from acting beyond the verified user’s authority on every call routed through it. The implementation and all experiments are released at this https URL.
[AI-67] Full-range Binary Classifier Calibration for Stable Model Updates in Production
链接: https://arxiv.org/abs/2607.05481
作者: Konstantin Berlin
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Detection models running in adversarial environments face a malicious distribution that drifts rapidly while the benign distribution stays comparatively stable, so teams retrain and redeploy constantly to stay ahead of new threats. Retraining tends to change the output prediction scores, which breaks downstream users of the model. For these security-oriented models we need consistent false-positive rate (FPR) across all output values, whereas standard probability-calibration methods target class probability rather than an FPR contract. We introduce a method built on top of existing calibration primitives that targets the whole FPR curve, giving scores a consistent FPR meaning across deployments. On one held-out split, the observed relative FPR error was at most 2.3% from 10% down to 0.1% FPR and 7.2% at 0.01% FPR. The shipped artifact remains under 200 KB in measurements across calibration sets from 1K to 10M benign samples.
[AI-68] Privilege and confidentiality in generative AI workflows
链接: https://arxiv.org/abs/2607.05479
作者: Václav Janeček,Thomas Melham
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:Generative AI (GenAI) systems store and process client data in three distinct ways: in the model’s parameters through training and memorisation, in the context window during a live session, and in knowledge databases for retrieval-augmented generation (RAG). Each mode creates different and often counter-intuitive risks to confidentiality and legal professional privilege, and each calls for specific governance responses. Drawing on the first English and American decisions to address privilege and generative AI, UK and Munir v Secretary of State for the Home Department and United States v Heppner, on the orthodox privilege authorities against which those decisions must be read, and on recent computer science research, we explain the three modes of data storage and processing in terms accessible to practitioners and analyse the legal consequences of each. We then situate the analysis within the regulatory framework governing solicitors in England and Wales and within the ordinary principles of professional negligence, arguing that the standard of effective information governance (and with it the benchmark against which negligence and misconduct will be measured) is changing. Although we write primarily for SRA-regulated practitioners, our data-governance analysis is framed to extend to any jurisdiction in which the protection of privilege or professional secrecy depends on demonstrable confidentiality. The ultimate aim of this article is to help legal services professionals understand salient data leakage risks in GenAI systems and thereby facilitate a more responsible deployment of GenAI on client data and other sensitive material.
[AI-69] Is Your NPU Ready for LLM s? Dissecting the Hidden Efficiency Bottlenecks in Mobile LLM Inference
链接: https://arxiv.org/abs/2607.05475
作者: Guanyu Cai,Ruiming Tian,Lang Yang,Zhouhong Ren,Jinliang Yuan,Lingkun Li,Jiliang Wang
类目: Hardware Architecture (cs.AR); Artificial Intelligence (cs.AI)
备注:
Abstract:Deploying Large Language Models (LLMs) on mobile devices enhances privacy and reduces latency, but is severely bottlenecked by hardware inefficiency. We present the first comprehensive, cross-layer measurement study of mobile LLM inference, uniquely spanning five mainstream frameworks (e.g., this http URL, GENIE) and three hardware backends (CPU, GPU, NPU). To enable this analysis, we develop PowerBench, a fine-grained profiling tool that provides the first backend-specific energy attribution, moving beyond traditional device-level measurements. Our study yields three critical insights: (1) Framework-induced performance gaps are substantially amplified on NPUs, reaching up to 10x using custom operators due to divergent offloading and quantization strategies. (2) We identify a distinct phase split where NPUs excel at compute-bound prefilling, while CPUs outperform all other backends in memory-bound decoding. This is driven by the NPU’s preference for large, fixed-shape workloads, which conflicts with the small-kernel, dynamic nature of decoding. (3) Backend-specific profiling uncovers substantial scheduling headroom missed by prior work. Suboptimal thread configurations, uncoordinated NPU sleep latencies, and CPU polling intervals result in up to 40% energy waste. Leveraging these findings, we present an energy-oriented best-practice configuration for mobile LLM inference. We estimate that this configuration could reduce energy consumption by up to 54.8% on the NPU backend across three datasets.
[AI-70] KAT-Coder-V2.5 Technical Report
链接: https://arxiv.org/abs/2607.05471
作者: Bo Huang,Fengxiang Li,Hao Xu,Haoyang Huang,Hongyi Fu,Jinhua Hao,Kun Yuan,Minglei Zhang,Pengcheng Xu,Shiyang Liu,Wenhao Zhuang,Yuze Shi,Zongxian Feng,Chao Wang,Cheng He,Chongling Rao,Deyu Cao,Fan Yang,Gang Xiong,Haochen Liu,Jiabao Li,Jian Liang,Jinghui Jia,Jingwen Chang,Jun Du,Junyu Shi,Min Li,Mingqi Wu,Qiang Gao,Shangpeng Yan,Shaotong Qi,Shu Xu,Shuo Zhou,Tiankuo Xu,Tong Zheng,Weilun Zhao,Xiancheng Meng,Xianda Sun,Xiaoyu Jiang,Xunhao Jia,Yao Xia,Yimeng Xu,Yinghan Cui,Yingpeng Chen,Yiwen Ning,Yong Wang,Yuxuan Sun,Zhongsheng Liu,Ming Sun,Cheng Luo,Chen Yang,Han Li,Kun Gai
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: 24 pages, 5 figures
Abstract:We present KAT-Coder-V2.5, a coding-focused agentic model trained to act autonomously inside real, executable repositories rather than as a single-turn code generator. Its capability is bottlenecked less by model scale than by the scarcity of reproducible environments, verifiable rewards, and high-value trajectories, which we address with an end-to-end agentic post-training framework. AutoBuilder reconstructs multilingual repositories into sandboxed environments with fail-to-pass and pass-to-pass verification at scale, from which we regenerate self-contained task specifications, recover near-miss trajectories, and distill supervision through process-aware filtering, while KwaiClawEnv synthesizes large-scale tool-use trajectories from executable services and real task seeds. We further scale reinforcement learning with harness randomization, a reliability-hardened sandbox, an asymmetric actor–critic PPO with hindsight-augmented value estimation, and a harness-oriented reward framework, and unify SWE, Agent-Claw, and WebCoding experts via Multi-Teacher On-Policy Distillation. Across six software-engineering and agentic benchmarks, KAT-Coder-V2.5 delivers the best agentic tool-use result on PinchBench and ranks second only to the frontier Opus 4.8 on repository-level software engineering. Our service is available at this https URL.
[AI-71] Breaking Structural Isolation: Scalable Graph Clustering via Community-Aware Sampling and Structural Entropy VLDB2026 VLDB
链接: https://arxiv.org/abs/2607.05469
作者: Jingyun Zhang,Hao Peng,Jianxin Li,Angsheng Li,Philip S. Yu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Social and Information Networks (cs.SI)
备注: Accepted to the Proceedings of the VLDB Endowment (VLDB 2026). 18 pages, 15 figures, 15 tables
Abstract:Unsupervised graph clustering is a fundamental technique for uncovering underlying semantic patterns in large-scale networks. Although Graph Contrastive Learning has demonstrated promising performance, existing methods often suffer from the “structural isolation” issue during mini-batch training, making it challenging to capture cohesive community structures that characterize the global topological distribution. To address these challenges, we propose SCISE, a Scalable unsupervised graph Clustering framework that preserves structural Integrity by synergizing community-aware sampling with constrained Structural Entropy. Specifically, we first introduce the Structural Entropy Community Constraint operator (SECC), which optimizes structural information within a constrained solution space to mitigate community fragmentation and enhance partition cohesion. Second, to prevent global information loss during batch training, we design a Community-Aware Sampling Expansion (CSampE) mechanism that incorporates the community context of target nodes into sampling batches, effectively breaking structural barriers and preserving topological integrity. Finally, we devise a Structural Contrastive Learning (StructCL) module that refines edge weights based on intra-batch structural similarity, guiding the encoder to learn representations in a higher-order structural space. Extensive experiments on six mainstream benchmark datasets demonstrate that SCISE significantly outperforms state-of-the-art algorithms, with ablation studies and robustness analyses further validating its effectiveness and reliability for real-world large-scale graphs.
[AI-72] Learning 4D Geometric Priors for Inference-Efficient World Action Models
链接: https://arxiv.org/abs/2607.05468
作者: Jianjun Zhang,Jian Zhu,Taiyi Su,Chong Ma,Zitai Huang,Yi Xu,Hanli Wang
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: 9 pages, 6 figures
Abstract:World Action Models (WAMs) have shown strong potential for robotic manipulation by jointly modeling visual future dynamics and executable action sequences. However, existing video-action co-training methods primarily optimize appearance-oriented video latents, which may insufficiently capture the temporally evolving geometry required for precise manipulation. We propose MECo-WAM, a Multi-Expert Co-Training World Action Model that injects action-relevant 4D geometric priors into video-action representations while preserving the original lightweight inference graph. During training, MECo-WAM combines video and action experts with a lightweight 4D expert supervised by relational targets from a frozen VGGT encoder. Asymmetric expert visibility prevents non-causal shortcuts from auxiliary geometry to action generation. To transfer geometric knowledge into the deployed video-action pathway, we introduce decayed 4D read-mask attention, which provides restricted current-frame geometric guidance early in training and progressively removes this dependency. We further propose action-aware temporal geometric distillation, which aligns within-frame geometric relations and their temporal evolution while emphasizing visual regions most relevant to robot actions. At deployment, all auxiliary 4D components are removed. Experiments on LIBERO (98.2%), RoboTwin 2.0 (92.6%), and challenging real-world manipulation tasks show that MECo-WAM improves manipulation performance without increasing inference cost.
[AI-73] Learnable Weighting of Intra-Attribute Distances for Categorical Data Clustering with Nominal and Ordinal Attributes
链接: https://arxiv.org/abs/2607.05464
作者: Yiqun Zhang,Yiu-ming Cheung
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 16 pages, 11 figures
Abstract:The success of categorical data clustering generally much relies on the distance metric that measures the dissimilarity degree between two objects. However, most of the existing clustering methods treat the two categorical subtypes, i.e. nominal and ordinal attributes, in the same way when calculating the dissimilarity without considering the relative order information of the ordinal values. Moreover, there would exist interdependence among the nominal and ordinal attributes, which is worth exploring for indicating the dissimilarity. This paper will therefore study the intrinsic difference and connection of nominal and ordinal attribute values from a perspective akin to the graph. Accordingly, we propose a novel distance metric to measure the intra-attribute distances of nominal and ordinal attributes in a unified way, meanwhile preserving the order relationship among ordinal values. Subsequently, we propose a new clustering algorithm to make the learning of intra-attribute distance weights and partitions of data objects into a single learning paradigm rather than two separate steps, whereby circumventing a suboptimal solution. Experiments show the efficacy of the proposed algorithm in comparison with the existing counterparts.
[AI-74] Evaluating calibrated refusal and safe usefulness in dual-use biology settings
链接: https://arxiv.org/abs/2607.05462
作者: Edwin H. Wintermute,Harmon Bhasin,Christina M. Agapakis,Dianzhuo Wang,Evan Seeyave,Arjun Banerjee,Daniel Fulop,Matthew C. Watson,Adam J. Meyer,Sandrine Boissel,Jens H. Kuhn,Rishi Jain,Noah D. Taylor,Helena Shomar,Patrick M. Boyle,Kenny Workman
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:As AI agents are incorporated into life science workflows, the capabilities that speed discovery might also enable misuse. We present BioSecBench-Refusal, a benchmark for risk identification and refusal behavior for biological research tasks. The benchmark pairs 61 Routine tasks, legitimate analyses adapted from the published literature, with 46 Red-Team tasks, fictional scenarios that resemble real research but conceal a biosecurity hazard. Across 16 model-harness configurations, refusal rates ranged from 7% to 74% on Routine tasks and 1% to 62% on Red-Team tasks, with many configurations refusing legitimate Routine work at comparable or higher rates than concealed hazards. Refusals were most often triggered by provider API filters applied prior to agentic reasoning. However, models given room to reason showed the potential to identify more real threats. We release BioSecBench-Refusal as a tool for model developers to calibrate capability and caution for agentic biotech R\D.
[AI-75] AdaStop: Cost-Aware Early Stopping for DNN Test Selection
链接: https://arxiv.org/abs/2607.05461
作者: Bonan Shen,Wei-Jung Huang,Xin Liu,Jiazhou Gao,Tao Ning
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Existing methods for testing deep neural networks (DNNs) primarily prioritize test inputs likely to reveal model faults under a fixed labeling budget. In practice, choosing that budget is difficult: too little testing misses failures, while too much incurs unnecessary labeling costs. This work studies the stopping problem in DNN testing. We formulate testing as a cost–benefit decision process in which labeling an input incurs cost c and discovering a fault yields value v . Based on this formulation, we introduce \textitAdaStop, a framework that estimates the marginal fault discovery rate during testing and stops labeling when the estimated rate falls below the threshold \tau = c/v . Experiments across multiple datasets, architectures, and selection strategies show that 65 – 84% of faults can be discovered using only 9 – 31% of the labeling budget.
[AI-76] Learning to Control LLM Agent Harnesses with Offline Reinforcement Learning
链接: https://arxiv.org/abs/2607.05458
作者: Haiwen Yi,Xinyuan Song
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 17 pages, 7 figures
Abstract:Large language model (LLM) agents are usually improved by changing prompts, models, or hand-written workflows, while the execution harness around the model is treated as fixed infrastructure. We argue that this harness is itself a learnable control layer. We formalize harness operation as a finite-horizon Harness MDP, where a lightweight controller selects structural execution actions while the LLM executor remains frozen. The controller is trained from offline rollouts using advantage-weighted regression with only terminal task-rubric rewards. We also separate final task quality from a post-hoc Harness Maturity Score, which measures whether the harness follows reliable execution patterns rather than only whether the final answer is correct. This separation gives a finite-buffer view of harness learning: final-quality gains require high-return support in the offline buffer, while process behavior can shift whenever it aligns with advantage-weighted actions. Across six controlled domains and two public-benchmark adapters, the learned controller consistently improves verification behavior and selectively improves final task quality, with the largest gains on adapted tau-bench retail, adapted AgentBench DB-Bench, and coding with a calibrated structural verifier. Ablations against behavior cloning and Forced CHECK show that the gains are not explained by imitation or by simply adding checks. These results identify harness control as a learnable layer for frozen LLM agents, while showing that offline support limits when better process control becomes better final answers.
[AI-77] Empirical Minimal-Realisation Compression of Deep Neural Networks via Controllability-Observability Tests
链接: https://arxiv.org/abs/2607.05457
作者: Anis Hamadouche,Amir Hussain
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Dynamical Systems (math.DS)
备注:
Abstract:Deep neural networks often contain substantial hidden-state redundancy, but most compression methods operate directly on weights, neurons, or quantised representations without explicitly characterising the dynamical role of internal states. This paper proposes a controllability-observability framework for empirical state-order reduction of deep neural networks. By viewing a trained network as a depth-indexed nonlinear dynamical system, we construct data-driven reachability, observability, and balanced Gramians from hidden-state snapshots and output Jacobians. The resulting A/B/C tests estimate layer-wise reachable, observable, and jointly reachable–observable ranks. These ranks are then used not only as diagnostic measures of hidden-state redundancy, but also as actual compressed layer widths for realised reduced networks. Experiments on MNIST and CIFAR-10 compare the proposed balanced realisation against projection-based reduction, unstructured pruning, structured pruning, low-rank SVD, dynamic INT8 quantisation, and linear baselines. On MNIST, a four-layer SiLU DNN is reduced from state order 1024 to 277, giving 72.95% state compression and 73.48% parameter compression, while maintaining 95.45% accuracy compared with 96.60% for the full model. On CIFAR-10, a larger SiLU DNN is reduced from state order 4608 to 1339, giving 70.94% state compression and 83.09% parameter compression, while preserving accuracy from 54.45% to 54.44% and reducing CUDA inference latency by approximately 3X. The results show that balanced reachable-observable ranks provide a principled empirical minimal-realisation criterion for designing compact neural architectures with little or no loss in accuracy.
[AI-78] he Granularity Paradox: How Temporal Disaggregation Inflates In-Sample Fit and Compounds Out-of-Sample Error
链接: https://arxiv.org/abs/2607.05450
作者: Hugo Moreira
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Methodology (stat.ME)
备注:
Abstract:This paper explores the “Granularity Paradox” in time-series forecasting, wherein finer temporal disaggregation (e.g., Monthly to Weekly/Daily) improves in-sample diagnostics and dataset size (N), but degrades out-of-sample accuracy due to recursive error compounding over longer horizons (H). Conversely, coarse aggregation (Annual) eliminates recursive error propagation but reduces data available to estimators. We formalize this trade-off and benchmark 10 models - spanning naïve, statistical, machine learning, and deep learning architectures - across six granularities using a 13-year public procurement dataset. The empirical results reveal a non-monotonic threshold structure: recursive autoregressive and seasonal models degrade substantially under high-frequency forecasting (e.g., Holt-Winters reaches a Test R-squared of -151 and TPFE of 425.85% at the Daily grain), while the LSTM traces a U-shaped error curve, worsening from Monthly (19.66%) through Bi-Weekly (35.94%) before overcoming the error propagation penalty at Daily (TPFE of 4.35%, R-squared of 0.66). Linear Regression remains stable across all granularities (16.3-17.0% TPFE), confirming that the paradox is driven by recursive feedback topology, not model complexity. The results demonstrate that standard pointwise metrics (RMSE, MAE) systematically mask cumulative error propagation, and that evaluating forecasts without goal-dependent cumulative metrics produces misleading assessments of model adequacy. We introduce a consensus-dissensus diagnostic comparing the directional behaviour of pointwise metrics against cumulative TPFE across granularities, enabling the identification of models whose standard diagnostics mask systematic error propagation.
[AI-79] Geometry-Aware Infrastructure-Anchored Denoiser for UWB Sensing and Work-Zone Reconstruction
链接: https://arxiv.org/abs/2607.05449
作者: Weizhe Tang,Jiaxi Liu,Junwei you,Steven T.Parker,Pei Li,Sikai Chen,Meng Ran,Bin Ran
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Robotics (cs.RO)
备注:
Abstract:Accurate work-zone geometry perception is critical for intelligent transportation systems, and ultra-wideband sensing offers a low-cost approach for infrastructure-aided reconstruction. However, outdoor UWB ranging is often degraded by non-line-of-sight propagation, burst noise, and long-tail errors, which can distort downstream spatial reconstruction. We present GAIA, a geometry-aware, infrastructure-anchored learning framework that couples temporal range modeling with latent anchor-layout estimation and deterministic distance projection. GAIA preserves range denoising as the supervised task while orienting the learned distances toward boundary-consistent reconstruction. We evaluate GAIA on a real-world outdoor UWB dataset with synchronized UWB, GNSS, and IMU measurements, and further test robustness using a real-data-calibrated stress-test simulator. GAIA achieves the lowest overall range MSE and highest polygon IoU among evaluated filtering-based and learning-based baselines, reducing MSE by 18.4% and improving polygon IoU by 15.5% over PoseMLP. These results show that geometry-aware range denoising provides an effective path toward spatially coherent work-zone reconstruction.
[AI-80] CHARLIE: An On-Premise Multi-Agent Retrieval-Augmented Generation System for Evidential Reasoning in Forensic Science
链接: https://arxiv.org/abs/2607.05428
作者: Leandro D. Carneiro,Andre L. S. Meirelles,Juliano de A. Gomes,Rafael C. A. Cabral
类目: Digital Libraries (cs.DL); Artificial Intelligence (cs.AI)
备注: 10 pages, 1 figure. Archival version of a paper presented at RELAF 2026: 1st Workshop on Reasoning with Evidence in Law Enforcement and Forensics, co-located with ICAIL 2026, Singapore, June 2026
Abstract:We present Charlie, an on-premise multi-agent Retrieval-Augmented Generation (RAG) system for structured evidential processing in digital forensic environments. Contemporary forensic workflows must handle large volumes of heterogeneous and unstructured documents under strict requirements of traceability, confidentiality, and legal compliance. Charlie addresses this challenge through a controlled agent architecture that combines local retrieval, task decomposition, structured memory, and verification mechanisms. Unlike cloud-based systems, it operates entirely within institutional infrastructure, preserving data sovereignty and evidential integrity. We describe the systems architecture, including its transition from classical RAG to agent-based orchestration, and demonstrate its application in real-world forensic scenarios. Case studies show that Charlie enables scalable multi-document data extraction and supports longitudinal forensic intelligence generation while maintaining traceability and auditability. Our results indicate that agent-orchestrated, on-premise RAG architectures can effectively support evidential workflows without compromising legal and institutional constraints. Charlie provides a practical and reproducible blueprint for deploying AI systems in high-stakes forensic environments. This manuscript is an archival version of a paper presented at the RELAF 2026 Workshop.
[AI-81] When AI Classifies: What Counts as Public Administration?
链接: https://arxiv.org/abs/2607.05420
作者: Shaoming Cheng,Laurie Schintler
类目: Digital Libraries (cs.DL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Databases (cs.DB)
备注:
Abstract:This study examines how alternative systems of scholarly representation identify and characterize broad public administration (PA) and artificial intelligence related public administration (AI-in-PA) scholarship. Using Web of Science and OpenAlex, it compares five approaches based on author-defined, citation-driven, and AI-assisted representations. The results highlight substantial differences in corpus size, publication types, publishing outlets, temporal development, and thematic clustering and structure. The alternative approaches often identify different knowledge domains instead of varied subsets of the same scholarship and therefore produce distinct representations, as evidenced by no overlap in publications and publishing outlets across representations. The findings suggest that algorithmic knowledge organization increasingly influences how interdisciplinary scholarship is classified, structured, and understood and, epistemologically, how its visibility, intellectual structure, and boundaries are represented. AI-enabled scholarly classifications and representations are not neutral but interpretative, likely self-reinforcing, and potentially constrain the evolution and adaptation of disciplinary boundaries. Human disciplinary judgment is essential and is complemented rather than replaced.
[AI-82] Contrastive Predictive Coding with Compression for Enhanced Channel State Feedback in Wireless Networks
链接: https://arxiv.org/abs/2607.05419
作者: Ahmed Y. Radwan,Hina Tabassum,Fahad Syed Muhammad,Matthew Baker
类目: Information Theory (cs.IT); Artificial Intelligence (cs.AI); Signal Processing (eess.SP)
备注: Accepted for publication in IEEE Transactions on Neural Networks and Learning Systems
Abstract:Accurate and timely channel state information (CSI) is essential for next-generation wireless systems, yet existing works treat CSI compression and CSI prediction as separate problems, both in academia and in current 3GPP studies. Consequently, channel aging remains insufficiently addressed within standardized CSI feedback pipelines. In this article, we propose a unified compression-prediction framework that integrates Contrastive Predictive Coding (CPC) directly into the 3GPP-compliant CSI compression architecture. Instead of predicting high-dimensional CSI matrices, our approach forecasts future latent representations and jointly optimizes reconstruction fidelity and temporal predictive coherence via a combined 1-SGCS and InfoNCE objective. This design enables temporal representation learning without increasing feedback overhead. We present two variants: CPC-before-Compression, which performs autoregressive modeling on encoded features prior to quantization, and CPC-after-Compression, which shifts temporal modeling to the base-station to reduce the complexity of users’ devices. Evaluations on 3GPP-compliant datasets from Nokia, Oppo, and CATT show that CPC-before-Compression achieves over 90% reconstruction accuracy with 32x lower decoder GFLOPs than the 3GPP baseline, while CPC-after-Compression preserves an identical encoder footprint and the same 64-bit feedback overhead. By unifying compression and prediction within a standardized pipeline, the proposed framework provides an age-aware, computationally efficient CSI feedback solution. The source code is publicly available at: this https URL
[AI-83] Why does AI unlock new possibilities in STEM education? A Bibliometric Analysis of Trends and Future Agenda
链接: https://arxiv.org/abs/2607.05412
作者: Jesse Yusuf Chan(Zexi Chen),Mengyao Chen,Yang Hong,Ziyun Song,Haoming Wang,Xianlong Xu
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
备注: Accepted by ISLS26 conference
Abstract:STEM education faces challenges in personalization and interdisciplinary integration. AI technology has brought new possibilities, but the mechanisms by which AI reshapes the STEM education ecosystem require systematic investigation. This study employs bibliometric methods to analyze 242 publications from 2015-2025, constructing knowledge maps to reveal the evolutionary trajectory. The findings show that the field has transformed from intelligent tutoring systems to inquiry-based learning and computational thinking cultivation driven by LLMs. AI’s key contribution lies in providing intelligent scaffolding that lowers the threshold for understanding knowledge. In this sense, AI is a core driving force promoting its shift from knowledge transmission to capability development.
[AI-84] he GenAI Skill Bypass: Mapping Divergent Pathways of University Students and Staff AI Literacy
链接: https://arxiv.org/abs/2607.05411
作者: Eduardo Oliveira,Narelle English,Tracii Ryan,Kamila Misiejuk,Cory dal Ponte,Sonsoles López-Pernas,Mohammed Saqr
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
备注: 18 pages, 7 figures, 3 tables
Abstract:Higher education institutions are increasingly expected to ensure that both students and staff develop Generative AI (GenAI) literacies. In response, they are introducing professional development programs and embedding GenAI skills within student curricula. However, current educational frameworks typically assume a linear progression of GenAI literacy, implying that foundational technical understanding must precede creative application. This paper challenges such an assumption through a psychometric analysis of a taxonomy-based self-assessment instrument (n = 158). We applied Rasch measurement theory and Guttman ordering to map the latent perceived order of difficulty of GenAI skills across students, academics, and professional staff. Results reveal a fundamental divergence in perceived competence profiles: while academics follow a more traditional linear path, students exhibit an “inverted” profile, frequently mastering high-level creation tasks before acquiring foundational conceptual understanding. Furthermore, the correlation of skill difficulty between students and academics was weak (r = 0.188). We argue that this “skill bypass” creates a fragile sense of fluency, where high self-efficacy in prompting masks low literacy in AI mechanics. These findings challenge the “one-size-fits-all” curricula and provide the empirical basis for diagnostic-driven, modular interventions that foster genuine human-AI synergy.
[AI-85] Automated Recommendation of Programming Learning Content Using Pattern-based Knowledge Components
链接: https://arxiv.org/abs/2607.05409
作者: Muntasir Hoq,Griffin Pitts,Zhangqi Duan,Arun Balajiee Lekshmi Narayanan,Mohammad Hassany,Andrew Lan,Peter Brusilovsky,Bita Akram
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
备注: Paper accepted to the 10th Educational Data Mining in Computer Science Education (CSEDM) Workshop in Seoul, Korea
Abstract:Introductory programming instruction relies on hands-on practice and short learning activities to support mastery of foundational concepts. Although many such learning resources exist, organizing and linking these items in instructionally meaningful ways is challenging without time-intensive expert curation. This study investigates the use of pattern-based Knowledge Components (KCs) to automatically identify code-based learning resources targeting similar concepts. In our approach, pattern-based KCs are extracted from each code sample, and related activities are identified by measuring similarity between the KC sets associated with each activity. By leveraging alignment at the level of semantically important programming patterns, this method supports contextually appropriate and pedagogically useful recommendations. We evaluate our approach on an expert-organized corpus of introductory Python materials in which instructors grouped items into bundles based on conceptual similarity. Results show that our pattern-based KC approach retrieves resources that align with this expert organization, and outperformed representative KC- and embedding-based baselines across standard ranking evaluations. Overall, the framework supports targeted, concept-oriented guidance for programming learners and can help instructors organize, bundle, and recommend instructional content at scale.
[AI-86] Position: Preventing AI-Generated CSAM Necessitates New Approaches to AI Safety ICML2026
链接: https://arxiv.org/abs/2607.05407
作者: Neil Kale,Rebecca Portnoff,Pratiksha Thaker,Michael Simpson,Robertson Wang,Kevin Kuo,Chhavi Yadav,Virginia Smith
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
备注: Accepted (spotlight) in ICML 2026, Position Paper Track. The first two authors contributed equally and may list their names interchangeably
Abstract:Modern artificial intelligence (AI) systems present profound new risks to child safety. AI is increasingly being misused to create AI-generated child sexual abuse material, facilitate child sexual exploitation, and reduce barriers to harm. In this paper, we argue that protecting children from AI-facilitated sexual abuse requires new approaches to AI safety. Existing safety techniques assume data accessibility, transparency, and evaluation practices that are incompatible with the ethical and legal constraints surrounding child sexual abuse material. We examine how these constraints create new technical challenges, such as limitations on dataset auditing, red teaming, and fine-tuning prevention. In turn, we outline 15 open problems in online child sexual exploitation and abuse across the AI development lifecycle, from dataset curation and model design to deployment and long-term maintenance. We propose targeted recommendations for researchers, developers, and policymakers to bridge the gap between theoretical AI safety and the realities of child protection. Our work aims to reframe preventing AI-facilitated child sexual abuse as a central, safety-critical dimension for AI research, motivating work that translates responsible AI principles into concrete safeguards against the exploitation of children.
[AI-87] A Guiding Framework for K-12 Teachers in Creating AI-powered Learning Technologies through Vibe Coding
链接: https://arxiv.org/abs/2607.05406
作者: Yukyeong Song,Seoyeon Choi,Jinhee Kim,Yeongje Kim,Lauren Weisberg,Jewoong Moon
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models generate code from natural language prompts, enabling “vibe coding,” which allows non-programmers to develop computational solutions. Vibe coding for teachers amplifies the value of teachers-as-designers, improving technology integration while fostering AI literacy. However, structured guidance on supporting this process is lacking. We propose GAIDE (A Guiding Framework for AI-Integrated Design for Educators), a framework that supports K-12 teachers in creating AI-powered learning technologies through vibe coding. The initial framework, built on Design Thinking and INTERACT, was validated through a CORDTRA interaction analysis of three teachers and four faculty mentors in an eight-week workshop to derive the final framework. Additionally, the qualitative analysis of pre- and post-interviews found an enhancement of teachers’ AI literacy. Findings highlight the potential of learning-by-creating for professional development.
[AI-88] he Jagged Global Economy: Frontier AI Unevenly Exposes National Economies
链接: https://arxiv.org/abs/2607.05404
作者: Arul Murugan,Tomás Aguirre,Abhishek Nagaraj,Rishi Bommasani
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
备注: Website (including code and data): this https URL
Abstract:Frontier AI’s labor-market effects matter to workers, firms, and policymakers, but current evidence generally comes from a handful of high-income economies. The capabilities of frontier AI are jagged across work tasks and national economies diverge in how they allocate human labor. We introduce a national AI exposure metric that combines occupation-level exposure scores and international employment data for 141 countries. We find that high income countries are substantially more exposed than low income countries and that Europe and Central Asia are 50 percent more exposed than Sub-Saharan Africa. We also find a gender gap: women are more exposed than men in 91 percent of countries, driven by their concentration in white-collar and sales occupations. The exceptions are countries where women’s employment remains concentrated in agriculture and household enterprises. We validate our national AI exposure estimates by showing they predict national AI adoption statistics published by Anthropic, Microsoft, and OpenAI. Beyond direct exposure, we identify a new mechanism for indirect exposure due to cross-country income dependencies. Some nations such as Tajikistan depend heavily on foreign workers remitting money back to their home countries: Tajikistan’s direct exposure to frontier AI is below-average but because 37 percent of Tajikistan GDP is Russian remittance and Russia is very exposed, Tajikistan’s remittance-accounted exposure becomes above-average. Our research shows that national variation in exposure is large enough that policy responses calibrated to U.S. or European labor markets will not generalize.
[AI-89] AI tools in Arab University English classrooms: Looking back and forward
链接: https://arxiv.org/abs/2607.05403
作者: Mohammed Q. Shormani,Muneef Yehia Alshawsh
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
备注: 18 pages, 2 Tables
Abstract:This paper aims to synthesize empirical research on AI tools used to support English as a second/foreign language (EL2) learners in Arab University classrooms (AUCs) between Jan 1st 2023 and Aug 31st 2025. We utilized 3 large datasets, namely Google Scholar, Web of Science, and Scopus as the data sources. Using PRISMA-guided searches across these well-known databases, we included only published articles. The search process results in 184 studies, but only 11 studies met the inclusion criteria. Findings unveil that EL2 learners have positive attitudes towards AI for drafting, revision, and practice. Empirical gains were most consistent for surface-level outcomes improvements in higher-order writing quality and speaking proficiency was mixed and often contingent on teacher mediation. The paper concludes by proposing a research agenda and practical guidelines for Arab universities seeking evidence-based AI integration in EL2 instruction. It also recommends scaffolded integration, teacher training, reflective tasks to reduce over-reliance on AI tools.
[AI-90] Catalyst Papers in Artificial Intelligence Research: A Landscape on ICLR from 2017 to 2025
链接: https://arxiv.org/abs/2607.05401
作者: Fan Huang
类目: Digital Libraries (cs.DL); Artificial Intelligence (cs.AI)
备注:
Abstract:A small number of methodological contributions, including word2vec, the Transformer, large-scale pre-training, and reinforcement learning from human feedback, have reshaped NLP and AI research over the past decade. OpenReview now makes numeric reviewer scores and accept/reject decisions public for every ICLR submission. Whether such review signals identify trajectory-changing papers at submission time, however, remains untested at corpus scale. We answer this question on 36,113 papers from ICLR 2017–2025, identifying \emphcatalysts: papers whose descendants measurably redirect future research. We compare four disruptiveness measures (the Consolidation/Destabilization (CD) index, node2vec, the direction-aware Embedding Disruptiveness Measure (EDM), and an LLM-based semantic rater) and define a five-type operational catalyst taxonomy (topic initiator, topic bridge, within-topic redirector, simultaneous, and recognition-misaligned). EDM leads at identifying highly cited ICLR papers (AUC 0.83 vs.\ 0.60 for CD, 0.49 for node2vec, and 0.42 for the LLM rater). Topic initiators precede a 7.55\times topic-share growth and topic bridges precede an 11.52\times growth in cross-topic citation flow versus year-matched controls. We found that the peer review scores are essentially orthogonal to future disruptiveness ( |\rho|\leq0.005 ; accepted and rejected papers have indistinguishable mean EDM, p=0.11 ).
[AI-91] Proof of Execution: Runtime Verification for Governed AI Agent Actions
链接: https://arxiv.org/abs/2607.05397
作者: James Rhodes,George Kang
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: 14 pages, 1 figure, 4 tables, 1 algorithm; includes formal soundness and replay theorems with witness constructions in appendix
Abstract:Agent systems increasingly execute rather than advise. When an AI agent queries regulated data, invokes effectful tools, and mutates persistent state, correctness is not captured by whether a terminal output looks plausible. The operative questions are whether each step was authorized under a contract, whether the recorded history is tamper-evident, and whether the trajectory can be reconstructed deterministically. We formalize this as runtime proof of execution. An execution is a triple x = (C, T, R) : a contract C , an Execution Causal Event Stream (ECES) T , and a replay context R . A well-formedness predicate and five validator-checkable invariants form the PoE validity predicate. Five semantic guarantees describe authorization, path compliance, null effect on deny, history integrity, and replayability. We prove soundness under explicit cryptographic and deployment assumptions: any PPT adversary that produces a PoE-valid execution violating a semantic guarantee yields a signature forgery, a hash collision, or a quantified deployment-failure event. The Prime Execution Model (PEM) separates planning, enforcement, effect, and recordkeeping into distinct authority planes; a lemma reduces trace completeness to Effector-exclusive credentialing. An Execution Attestation Certificate is issued only when PoE = 1. In a single-node TypeScript prototype, PoE adds approximately 2.7 ms on a minimal flow and 4.4% overhead on concurrent batch workloads; a standard eight-event trace compresses to approximately 1.1 KB; injected Gateway-bypass and trace-mutation attacks are rejected. PoE does not replace consensus, TEEs, or zkVMs; it binds authorization, effect, history, and replay into a single runtime-checkable object so that governed execution becomes attestable under contract.
[AI-92] KVpop – Key-Value Cache Compression with Predictive Online Pruning
链接: https://arxiv.org/abs/2607.05061
作者: Lukas Hauzenberger,Niklas Schmidinger,Anamaria-Roberta Hartl,David Stap,Thomas Schmied,Sebastian Böck,Günter Klambauer,Sepp Hochreiter
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Key-value (KV) cache growth is a major bottleneck in autoregressive decoding, as memory and bandwidth scale linearly with context length. Existing KV eviction methods often rely on static heuristics or proxy scores, which poorly track future token utility and cause brittle eviction as relevance shifts. To address this, we introduce KVpop, which learns a fixed-budget KV eviction policy by directly supervising the keep-or-drop decision. The scorer is trained against a novel future-attention target, computed efficiently without materializing dense attention maps. We further introduce a delayed memory-based scorer that, uniquely among learned eviction methods, defers scoring for a fixed number of steps to exploit near-future context. On AIME and HMMT mathematical reasoning, KVpop retains 98% of full-attention performance on Qwen3-4B at 75% KV cache compression and 97% at 88% compression, consistently outperforming established eviction baselines. Qwen3-8B shows even stronger results, reaching near-full teacher performance. These results show that supervising eviction with future-attention signals cuts memory costs while maintaining quality.
[AI-93] Lingering Authority: Revocable Resource-and-Effect Capabilities for Coding Agents
链接: https://arxiv.org/abs/2606.22504
作者: Igor Santos-Grueiro
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注: 20 pages
Abstract:Coding agents often receive broad tool access for an entire task, even when a resource is needed only for one subgoal. We call this gap lingering authority: a temporary resource/effect capability remains exposed after the episode that justified it has closed. PORTICO is a reference monitor for revocable capabilities exposed to the planner. It compiles an explicit task contract into initial capabilities, grant rules, trusted closure predicates, and global deny rules. A request-grant-invoke lifecycle materializes expansions as opaque, epoch-bound handles. Closure removes those handles from the next planner interface and rejects stale replay before side effects. The monitor assumes mediated tools and a sound typed catalog. In controlled coding-agent tasks, PORTICO records no executed contract-forbidden effects in the evaluated runs, while controlled grants recover boundary work blocked by a fixed narrow envelope. A non-revoking comparator receives the same initial envelope and the same grants at the same turns. On the closure slice, both systems match task success, scope compliance, and all pre-closure decisions; PORTICO then rejects 10/10 post-closure reuses, while the comparator permits 10/10. A deterministic stale-write audit records 0/6 versus 6/6 executed forbidden effects. Scripted traces and six live model traces over file writes, git mutation, and network egress show the same split. In a four-episode same-policy diagnostic, broad request exposure preserves zero executed forbidden effects but raises blocked proposals from 67 to 84. Frozen real-repository runs, with commits and traces recorded, exercise the same lifecycle on real project layouts. Comments: 20 pages Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Software Engineering (cs.SE) Cite as: arXiv:2606.22504 [cs.CR] (or arXiv:2606.22504v1 [cs.CR] for this version) https://doi.org/10.48550/arXiv.2606.22504 Focus to learn more arXiv-issued DOI via DataCite
[AI-94] Provable learning separation for predicting time-evolution of quantum many-body systems
链接: https://arxiv.org/abs/2607.06472
作者: Rahul Bandyopadhyay,Riccardo Molteni,Jens Eisert,Vedran Dunjko,Sofiene Jerbi
类目: Quantum Physics (quant-ph); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 48 pages, 1 figure
Abstract:Given that quantum computers are naturally suited to simulate the behavior of quantum many-body systems, an immediate question arises: can one formulate physically motivated quantum machine learning (QML) tasks that exhibit learning separations? We address this problem by studying the learnability of quantum many-body dynamics from the perspective of probably approximately correct (PAC)-learning. Concretely, we devise a supervised learning problem where the training set consists of specifications of randomized stabilizer probe states, evolution times sampled uniformly from a polynomially large time interval [0,T] , coupled with expectation values of certain observables evaluated on the resulting time-evolved state under an unknown Hamiltonian. For this learning task, we provide an efficient quantum procedure whose training phase learns the underlying Hamiltonian from short-time training samples, and whose deployment phase combines Hamiltonian simulation with the classical shadows protocol to perform inference on a newly given data point. By contrast, the existence of O(\mathsfpoly(n)) -time instances ensures classical hardness: by embedding a \mathsfBQP -complete computation into the polynomially long time-dynamics of a low-intersection variant of the Feynman-Kitaev clock Hamiltonian construction, we show that, for a certain family of input distributions, no randomized classical polynomial-time algorithm can fulfill our learning condition, unless \mathsfBQP\subseteq\mathsfP/poly . Furthermore, we show that the classically hard instance maintains quantum learnability. We also give an interpretation of our results in learning-assisted certified quantum simulation. Taken together, our results demonstrate a rigorous learning separation for a natural ML task based on Hamiltonian evolution, while building connections between quantum learning theory, quantum simulation, and QML.
[AI-95] An Experimental Design Approach to Evaluating Agent ic AIs Autonomous Model Discovery
链接: https://arxiv.org/abs/2607.06413
作者: Hao He,Xueying Liu,Chris J. Kuhlman,Xinwei Deng
类目: Methodology (stat.ME); Artificial Intelligence (cs.AI)
备注: 39 pages, 11 figures, 6 tables. Data and code available at the GitHub repository listed in the paper
Abstract:Large language model coding agents increasingly perform open-ended data modeling and analysis. These agents are stochastic and adaptive, and therefore their autonomous model discovery behavior cannot be adequately characterized by a single benchmark run. In this work, we propose an experimental design and analysis framework for systematically evaluating this discovery process, quantifying its variability, and identifying important factors. The proposed framework treats these agents as stochastic model-discovery operators, which map task-specific discovery data and an optimization target to a fitted model. Specifically, we investigate two such operators, Codex and Claude Code, under controlled experimental factors including agent’s reasoning effort, task, optimization metric, and composition of training data. For each agent-task-metric combination, regression models and inference are conducted for multiple responses such as output quality, dollar cost, wall-clock time, and process complexity. Furthermore, we develop a utility-aligned canonical decomposition to characterize the dominant direction of the reasoning-effort effect and to assess whether that direction aligns with a performance-cost utility direction. The proposed framework is demonstrated on a testbed of networked word-forming games with insightful findings on reasoning effort with respect to cost and process complexity.
[AI-96] riA Pipeline: A Large-Scale Automatic Audio Annotation Pipeline For Audio Classification In Specific Scenarios INTERSPEECH2026
链接: https://arxiv.org/abs/2607.06179
作者: Hong Lyu,Mingru Yang,Qianhua He,Yanxiong Li,Jinxin Huang,Zhengyu Pei
类目: Audio and Speech Processing (eess.AS); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 5 pages, 2 figures, 4 tables, accepted for publication in Interspeech 2026. The code is at: this https URL
Abstract:There are some datasets of varying scales for audio classification (AC) applied to different tasks. However, annotated data is limited for most scenarios, such as domestic environments. To address this challenge, we propose an \textbfA utomatic \textbfA udio \textbfA nnotation Pipeline–TriA Pipeline, which can efficiently convert audio from various scenarios into high-quality training data with audio event annotations. A TriA dataset was constructed with the TriA Pipeline, over 2130 hours of audio covering 431 audio classes. Furthermore, we partitioned a prior-knowledge-guided subset (TriA _\mathrmGK ) from TriA and conduct comparative experiments on three domestic AC tasks. Comparing the result on manually annotated data only and that on manually annotated data combines TriA _\mathrmGK , TriA _\mathrmGK could achieve average relative gains of 3.97% in accuracy and 3.35% in Macro-F1, validating the effectiveness of TriA _\mathrmGK and the TriA Pipeline.
[AI-97] angent classes of matroids and wonderful compactifications
链接: https://arxiv.org/abs/2607.05835
作者: Ronnie Cheng,Shurui Liu,Guoxiong Gao
类目: Algebraic Geometry (math.AG); Artificial Intelligence (cs.AI); Combinatorics (math.CO)
备注:
Abstract:For every loopless matroid M and every Feichtner–Yuzvinsky building set \mathcalG containing the top flat, we construct an integral tangent class T_M,\mathcalG^\mathbbZ\in K_\mathbbZ(M,\mathcalG) ; in the realizable case it specializes to the class of the tangent bundle of the corresponding wonderful compactification, it recovers the Hilbert series of the Chow ring through Hirzebruch–Riemann–Roch, and it satisfies the expected Chern-alpha lower bounds. This reproduces the tangent class and its key properties studied by the first author in arXiv:2606.22650. The main body of this paper was produced autonomously, without human mathematical guidance, by Danus, an AI mathematical reasoning agent. Danus solved the problem before arXiv:2606.22650 was publicly available, demonstrating the potential of AI agents in mathematical research. We reproduce its output faithfully, adding only editorial comments; the experiment is documented in Appendix B.
[AI-98] o Retain or to Adapt? Generalizing Continual Learning
链接: https://arxiv.org/abs/2607.05609
作者: Giulia Lanzillotta,Mandana Samiei,Doina Precup,Razvan Pascanu,Claire Vernade
类目: Machine Learning (stat.ML); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:The Continual Learning (CL) literature has long been driven by the goal of mitigating catastrophic forgetting. This objective rests on a pervasive, often unstated assumption: that a lifelong learner should approximate the Joint-Task Learning (JTL) solution and retain all previously acquired knowledge. We challenge this retention-centered premise, arguing that in non-stationary environments prioritizing retention can impede real-time adaptation. Shifting the focus to the Average Lifelong Error (ALE), we formalize CL as an online optimization problem governed by the interaction between environmental and learning dynamics. We introduce Transfer Efficiency as a quantitative measure of the tension between Instability, the bias inherited from conflicting past experience, and Transient Error, the optimization cost of learning new tasks from scratch. Under mild convergence conditions, holding across linear and neural network models, this decomposition yields a Critical Task Duration: a closed-form threshold beyond which historical knowledge transitions from a warm-start advantage to an optimization liability whenever retention induces a positive stationary bias. We validate these theoretical predictions on continual image classification and reinforcement learning benchmarks. Finally, by connecting continual learning to the online learning framework of predictable sequences, we show that JTL is only one instance of a broader family of objectives, and we propose a new general class of continual learning algorithms, which we call Predictive Continual Learning. Predictive CL algorithms optimize expected future performance under an explicit, dynamically updated model of future tasks. As a proof of concept, we analyze a Window algorithm that interpolates between JTL and Independent-Task Learning (ITL), outperforming both under controlled distributional drift.
[AI-99] Lean-Quantum: Toward AI-Assisted Formalization of Quantum Information
链接: https://arxiv.org/abs/2607.05492
作者: Kazumi Kasaura,Kei Tsukamoto,Kento Mori,Risa Mizuno,Takahiro Namatame,Yuta Oriike,Masaya Taniguchi,Sho Sonoda,Hayata Yamasaki
类目: Quantum Physics (quant-ph); Artificial Intelligence (cs.AI)
备注: 34 pages, the Lean library is available at this https URL
Abstract:Quantum information theory is built on entropic quantities; among them, the sandwiched Rényi relative entropy is a fundamental divergence with various applications, and its data processing inequality (DPI) under quantum channels is a cornerstone result. In this work, we present a Lean 4 library for quantum information, designed as a reusable formal infrastructure for theoretical analysis. As a central demonstration of the library, we formalize the DPI for the sandwiched Rényi relative entropy for positive semidefinite operators on finite-dimensional quantum systems. The library provides a basis-independent operator-theoretic framework for finite-dimensional quantum mechanics compatible with the standard mathematical library Mathlib, including reusable interfaces for finite-dimensional systems, states, channels, tensor products, partial traces, Choi operators, Kraus representations, and Stinespring representations. It also builds infrastructure for noncommutative trace inequalities, including operator monotonicity and convexity via the real continuous functional calculus, block-operator positivity, Hilbert-Schmidt operator spaces, Jensen’s operator inequality, generalized perspectives, operator power means, and Lieb-Ando trace inequalities. On top of this framework, we formalize entropy-specific ingredients for the DPI: variational formulas for the sandwiched quasi-entropy via Young and reverse-Young inequalities, tensor-product compatibility of real powers, and Haar measures on unitary groups. Together, these components yield a Lean formalization of the DPI, give strong subadditivity as a corollary, and provide the last missing component needed to complete the Lean formalization of the generalized quantum Stein’s lemma. More broadly, the development provides machine-checkable foundations for future formalized and AI-assisted research in quantum information theory.
机器学习
[LG-0] GraphBU: MILP Instance Generation with Graph-Native Block Units
链接: https://arxiv.org/abs/2607.06532
作者: Xiaolei Guo,Chenyu Zhou,Jianghao Lin,Dongdong Ge
类目: Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注:
Abstract:Mixed-integer linear programming (MILP) instances used for solver development are hard to obtain when models come from private or application-specific pipelines. A generator must keep the structure that solvers and learned policies rely on. Existing general generators usually choose their generation unit from a formulation template, summary statistics, local graph edits, or blocks found after recombination. These units do not explicitly record how a local part of the MILP is coupled to the rest of the instance. We propose GraphBU, a graph-native generator whose basic unit is a local subproblem plus its interface. The method promotes coupling nodes into master constraints or boundary variables and uses the resulting block units for compatibility-checked replacement. The analysis focuses on the properties needed by this construction: promotion separates interfaces, replacement can preserve feasibility under an interface-slack condition, and the graph construction is invariant to row-column permutations. On MILP instances generation, this unit keeps graph statistics close to the source family, preserves feasibility on most datasets, and improves downstream Predict-and-Search training. Genrated by GraphBU, The average graph-statistical similarity was approximately 0.934, the average feasibility was approximately 96.7%, and the average increase in the main index of downstream PS was approximately 8.0%.
[LG-1] EntroPath: Maximum Entropy Path Ensemble Embedding for Manifold Learning
链接: https://arxiv.org/abs/2607.06497
作者: Przemysław Rola
类目: Machine Learning (cs.LG); Quantitative Methods (q-bio.QM); Machine Learning (stat.ML)
*备注: 40 pages, 15 figures. Code: this https URL
Abstract:We introduce EntroPath, a manifold learning method that recovers geodesic geometry from data graphs through ensembles of diffusion paths. Many existing graph-based embeddings rely either on locally normalised random walks or on shortest-path distances. The former can concentrate diffusion in densely sampled regions, while the latter are sensitive to spurious shortcut edges in the graph. EntroPath instead builds its dissimilarities from the maximum entropy random walk (MERW), which aggregates the full ensemble of k-step paths between points rather than relying on any single trajectory. We show that the resulting free-energy dissimilarity converges to squared geodesic distance in the short-time limit, via Varadhan’s heat-kernel formula. The diffusion depth k interpolates smoothly between local neighbourhood structure and global manifold geometry, and the symmetrised kernel admits an exact Gram factorisation connecting EntroPath to kernel methods. We further provide scalable extensions via landmark projection and diffusion-potential pseudotime. Across synthetic manifolds and single-cell benchmarks, EntroPath consistently matches or outperforms diffusion- and shortest-path-based methods, while remaining competitive with neighbourhood-preserving embeddings (UMAP, t-SNE) on local-structure metrics. Its gains are most pronounced on manifolds with non-uniform sampling density and well-separated branching trajectories, where path-ensemble diffusion more faithfully preserves the underlying geodesic geometry.
[LG-2] Physics-Informed Neural Embeddings of PDE Solution Families
链接: https://arxiv.org/abs/2607.06348
作者: Raul Jimenez,Svitlana Mayboroda,Pavlos Protopapas,Leonid Sarieddine,David N. Spergel,Pedro Tarancón-Álvarez
类目: Machine Learning (cs.LG); Numerical Analysis (math.NA); Computational Physics (physics.comp-ph)
*备注: 36 pages, 6 figures, 1 table
Abstract:We introduce a physics-informed framework for learning finite-dimensional embeddings of solution families of partial differential equations. The method uses a multihead Physics-Informed Neural Network in which a shared body learns a latent manifold representing the solution space, while linear heads reconstruct individual solutions associated with different initial conditions. A head-orthogonalization penalty removes degeneracies in the latent representation and stabilizes the principal-component spectrum across training realizations. Because the initial condition is built into the network output by construction, these principal components measure the additional variability the network learns on top of the initial profile, not the full solution itself. We apply the method to the one-dimensional viscous Burgers equation, with the heat and wave equations as robustness checks. For a latent dimension n_b=20 , the learned manifolds exhibit pronounced effective dimensional reduction: for Burgers dynamics, only 2 - 4 principal components capture about 95% of the latent-space variance, while 4 - 7 capture about 99% , depending on the initial-condition family; the same qualitative compression holds for the heat and wave equations. We also split the wavenumber axis into bands (``Fourier shells’') and measure how much each band contributes to every principal component. The resulting frequency profile is invariant under the change-of-basis freedom that the orthogonalization penalty leaves in the latent space, and is therefore reproducible across independent training runs. More broadly, this establishes the learned spectral profiles and principal components as robust observables of solution-manifold geometry.
[LG-3] Dithered Gaussian Mechanism for Randomness-Efficient Differential Privacy
链接: https://arxiv.org/abs/2607.06320
作者: Nikita P. Kalinin,Rasmus Pagh
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注:
Abstract:We present the dithered Gaussian mechanism, a novel alternative to the discrete Gaussian mechanism for differential privacy that discretizes the private output rather than the noise distribution itself. By interpreting this discretization as post-processing of the Gaussian mechanism, our construction directly inherits the privacy guarantees of the standard Gaussian mechanism while avoiding vulnerabilities caused by finite-precision floating-point outputs. We show that the mechanism is provably randomness-efficient: by sampling the discretized output values directly, the number of high-quality random bits required for privacy can be reduced significantly and made independent of the noise level. This is achieved by separating the randomness into two sources: a high-quality source used for the privacy-critical sampling step, and a high-performance public source, possibly known to the adversary, that supplies the additional randomness needed for randomized discretization. This separation enables the use of cryptographically secure randomness without substantial performance loss. As an application, we study model training with DP-SGD and show that cryptographically secure noise generation with reduced exposure to floating-point vulnerabilities can be achieved with modest practical overhead.
[LG-4] Quantitative Gaussian-Process limits of Tensor Programs
链接: https://arxiv.org/abs/2607.06290
作者: Andrea Agazzi,Eloy Mosig García,Dario Trevisan
类目: Machine Learning (cs.LG); Probability (math.PR); Machine Learning (stat.ML)
*备注:
Abstract:We study the infinite-width Gaussian-process limit of random neural networks through the lens of tensor programs, and we provide a quantitative convergence theory in Wasserstein distance. Our main result gives explicit finite-width error bounds, of order inverse square-root of the widths between finite-network executions and their Gaussian-process limits. The framework is architecture-agnostic and covers feed-forward models together with weight-sharing schemes relevant for recurrent and transformer-type architectures. Subjects: Machine Learning (cs.LG); Probability (math.PR); Machine Learning (stat.ML) MSC classes: 60G15, 68T07, 60B10 Cite as: arXiv:2607.06290 [cs.LG] (or arXiv:2607.06290v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2607.06290 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Eloy Mosig [view email] [v1] Tue, 7 Jul 2026 13:59:56 UTC (181 KB) Full-text links: Access Paper: View a PDF of the paper titled Quantitative Gaussian-Process limits of Tensor Programs, by Andrea Agazzi and 2 other authorsView PDFHTML (experimental)TeX Source view license Current browse context: cs.LG prev | next new | recent | 2026-07 Change to browse by: cs math math.PR stat stat.ML 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?) IArxiv recommender toggle IArxiv Recommender (What is IArxiv?) 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
[LG-5] Kernel-based Operator Learning: Error Analysis Budget Allocation and a Physics-Informed Extension
链接: https://arxiv.org/abs/2607.06287
作者: Rüdiger Kempf
类目: Numerical Analysis (math.NA); Machine Learning (cs.LG)
*备注:
Abstract:We study kernel-based operator learning in a two-stage sampling framework, where an offline kernel regression operator learns a discretized representation of the target operator from input-output pairs and an online kernel reconstruction operator recovers the output function from predicted observations. Our main theoretical contribution is an explicit budget allocation condition relating the number N of training pairs, the number n of input observations, and the output resolution m . The condition is derived from a coupled error analysis that interprets the surrogate as a reconstruction from approximate data. This yields a decomposition of the total error into reconstruction and learning contributions that can be analyzed independently. As a consequence, we obtain quantitative scaling laws describing how N , n , and m must be coupled to guarantee convergence and to balance offline learning and online reconstruction errors. The resulting estimates extend previous analyses of kernel-based operator learning. We further introduce a physics-informed extension that incorporates knowledge of the underlying PDE at evaluation time. Rather than encoding constraints directly into the kernel, we augment the online reconstruction step by penalizing PDE residuals at collocation points. The method requires no retraining for new inputs. Numerical experiments illustrate the theoretical findings and demonstrate the effectiveness of the proposed physics-informed reconstruction strategy. Subjects: Numerical Analysis (math.NA); Machine Learning (cs.LG) Cite as: arXiv:2607.06287 [math.NA] (or arXiv:2607.06287v1 [math.NA] for this version) https://doi.org/10.48550/arXiv.2607.06287 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-6] Canopy: A Heterograph Foundation Model for Metabolic Engineering ICML
链接: https://arxiv.org/abs/2607.06224
作者: Jake Bowden,Laurence Legon,Satnam Surae
类目: Machine Learning (cs.LG)
*备注: Accepted at ICML GenBio Workshop 2026 this https URL
Abstract:Designing microbial strains that produce high-value chemicals at commercially viable titers remains a central challenge in metabolic engineering. Existing computational approaches either rely on stoichiometric constraint-based models that cannot learn from experimental data, or apply tabular machine learning to hand-crafted features that discard the relational structure of biological knowledge. We present Canopy, a heterogeneous graph foundation model that integrates ten public and proprietary data sources into a unified knowledge graph (KG) of 6.9M nodes across 13 types and 34 edge types, covering genes, proteins, metabolites, reactions, pathways, strains, and fermentation experiments. Node features are encoded through domain-specific foundation models (ESM-2 for protein sequences, MoLFormer for chemical SMILES, and PubMedBERT for biomedical text), yielding a multi-modal representation within a single graph. We pretrain a Heterogeneous Graph Transformer (HGT) augmented with SignNet positional encodings, Jumping Knowledge aggregation, and virtual nodes using four self-supervised objectives (link prediction, masked node modelling, distance prediction, and contrastive experiment clustering), balanced via learned homoscedastic uncertainty weighting. On the downstream task of fermentation titer prediction, frozen Canopy embeddings achieve R^2 = 0.41 with a lightweight probe, outperforming tabular baselines (best R^2 = 0.24 ) and homogeneous GNN variants.
[LG-7] Leverag ing Extrag radient for Effective Sharpness-Aware Minimization in Deep Learning
链接: https://arxiv.org/abs/2607.06151
作者: Yao Fu,Chunxia Zhang,Junmin Liu,Yihang Jin,Haishan Ye,Yuanao Yang
类目: Machine Learning (cs.LG); Probability (math.PR)
*备注:
Abstract:Generalization remains a pivotal challenge in deep learning, where traditional optimizers like Stochastic Gradient Descent (SGD) often converge to sharp minima, leading to overfitting and reduced performance on unseen data. Building on Sharpness-Aware Minimization (SAM), for seeking flat minima associated with improved generalization, we propose the Extragradient-Inspired Sharpness-Aware Minimization (EISAM), a novel optimizer that enhances generalization via the extragradient technique. EISAM uses a two-step update process: a prediction step investigating the geometry of the loss landscape and a perturbation step that refines updates with a base optimizer. This approach achieves better generalization performance than SAM. Crucially, EISAM reduces sensitivity to the perturbation radius, enhancing robustness, and simplifying the tuning across diverse settings. Extensive experiments on benchmark datasets demonstrate that EISAM consistently outperforms SGD, Adaptive Moment Estimation (Adam), and SAM in test accuracy and training efficiency across various architectures. Theoretical analysis further confirms that EISAM tightens the generalization bound by steering parameters toward flatter minima with reduced curvature. Accompanied by a thorough hyperparameter analysis, EISAM offers practical tuning guidance, establishing it as a robust, scalable, and broadly applicable optimization solution that advances both the theory and practice in deep learning.
[LG-8] 6G Sensing Security: Distributed Game-Theoretic RL for Urban Beamforming and Attacker Detection
链接: https://arxiv.org/abs/2607.06115
作者: Parmida Geranmayeh,Onur Günlü
类目: Information Theory (cs.IT); Cryptography and Security (cs.CR); Computer Science and Game Theory (cs.GT); Machine Learning (cs.LG); Networking and Internet Architecture (cs.NI)
*备注:
Abstract:In next-generation networks, communication systems will no longer be limited to data transmission and will be expected to acquire awareness of the surrounding environment. This leads to the concept of integrated sensing and communication (ISAC), where the same wireless infrastructure is used for both communication and environmental sensing. Thus, ISAC enables the system to transmit information efficiently and observe and interpret channel variations and user behavior. Motivated by this capability, this work focuses on detecting an active attacker in an urban environment scenario, where the attacker intentionally manipulates beamforming directions to increase interference and mislead the transmitter into allocating the main lobe of beam toward itself instead of legitimate users. We apply game-theoretic approaches to model the interaction between legitimate users and the attacker, and integrate the resulting utility-based formulation into a reinforcement learning (RL) framework. Simulation results demonstrate that the proposed method effectively addresses security challenges in dynamic 6G ISAC systems.
[LG-9] Modeling Normal Is All You Need: Joint Latent Clustering for Anomaly Detection in Multimodal Cyber-Physical Systems
链接: https://arxiv.org/abs/2607.06094
作者: Alexander Apartsin,Yehudit Aperstein
类目: Machine Learning (cs.LG)
*备注: 13 pages, 1 figure
Abstract:Faults on a cyber-physical system (CPS) are too rare and unrepresentative to characterise, or even to select a model on, so detection must instead model normal behaviour; the standard point-adjusted evaluation, however, rewards detectors that never do. CPS normal behaviour is the union of many imbalanced, curved, thin-fringed operating regimes rather than a single blob; we state this structure as ten assumptions (A1-A10), abbreviated Massive, Implicit, Imbalanced Multimodality (MIIM). We model the normal law with a jointly learned latent representation plus explicit Gaussian-mixture mode clustering, scored in the latent rather than by a global density or a reconstruction residual, and evaluate under a deliberately fair protocol: raw point-wise metrics with no point adjustment, a trivial-detector difficulty split, prevalence-matched F1, and train-normal-only calibration. On three real CPS datasets (WADI, HAI, SKAB), the detector wins both the combined column and the difficult correlation/dynamics-fault column on all three, reaching difficult-subset AUROC 0.831 on HAI, 0.726 on WADI, and 0.610 on SKAB. The margin is largest on the two multimodal datasets the MIIM assumptions target and slimmest on the near-unimodal one, tracking multimodality as the thesis predicts, and it holds against three deep detectors (USAD, TranAD, GDN) re-computed with the same raw metrics, all of which collapse on the difficult subset. The methodological contributions are the MIIM assumption set, the difficulty-stratified fair protocol, and a latent-only score that drops reconstruction because a flexible decoder rebuilds the hard faults faithfully.
[LG-10] Scalable Perturbation Learning for Online Self-Supervised Echo State Networks
链接: https://arxiv.org/abs/2607.06079
作者: Taiki Yamada,Kantaro Fujiwara
类目: Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
*备注: 19 pages, 2 figures
Abstract:Intelligent systems should not only solve tasks but also adapt under real-world constraints. Autonomous adaptation via self-supervised learning, sequential adaptation via online learning, and memory-efficient implementation via perturbation-based learning are important requirements for such systems. However, these requirements are generally in tension for high-dimensional systems, because perturbation-based learning suffers from variance that grows with the dimension of the perturbed variables. In this study, we focus on echo state networks (ESNs), where this tension naturally arises in large reservoirs. We propose a perturbation-based learning rule for online self-supervised learning in ESNs. The proposed rule is derived from an orthogonal decomposition of the self-supervised learning cost, which separates an input-dependent component from a redundant component determined by the fixed ESN parameters. By perturbing only the input-dependent component, the effective perturbation dimension is reduced from the reservoir dimension to the input dimension. Thus, the proposed method preserves self-supervised adaptation, online learning, and scalar-feedback perturbation learning, while avoiding reservoir-size-dependent variance growth. This suggests a design principle for scalable and hardware-compatible learning: online learning should be restricted to the dynamically necessary low-dimensional component of the objective. Comments: 19 pages, 2 figures Subjects: Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE) Cite as: arXiv:2607.06079 [cs.LG] (or arXiv:2607.06079v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2607.06079 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-11] Determinantal point process sampling for bioacoustic active learning
链接: https://arxiv.org/abs/2607.06063
作者: Hugo Magaldi,Gabriel Dubus
类目: ound (cs.SD); Machine Learning (cs.LG)
*备注: BioDCASE Challenge 2026 - Task 4 Active learning. Ranked 2/14
Abstract:Eco-acoustic monitoring generates vast volumes of audio data, making active learning a promising approach for reducing annotation effort while efficiently training reliable biodiversity classifiers. This report presents CARE-DPP, a batch active-learning acquisition method submitted to BioDCASE Active Learning for Bioacoustics 2026 challenge. The method combines class-balanced predictive uncertainty with embedding-space novelty, while a determinantal point process (DPP) objective selects a high-quality and non-redundant acquisition batch. The uncertainty-novelty balance is annealed over the annotation budget: early cycles emphasize geometric coverage, whereas later cycles increasingly exploit classifier uncertainty. To mitigate unreliable early scores, the DPP candidate pool mixes top-quality candidates with a decreasing proportion of random exploration. An adaptive acquisition schedule uses smaller batches early and larger batches later. Evaluated over five repeats on the BirdSet HSN, POW and UHH subsets and on ATBFL, CARE-DPP obtains a mean development AULC of 0.50 for macro mAP, compared with 0.46 for the official CoreSet baseline. Ablations identify DPP batch diversification and the adaptive acquisition schedule as the largest contributors.
[LG-12] REAN: Reconstruction-aware ECG Anonymization Based on Privacy–Utility Orthogonality
链接: https://arxiv.org/abs/2607.06037
作者: Taerin Ki,Sunghwan Park,Junyoung Park,Jaewoo Lee
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注: preprint
Abstract:A shared electrocardiogram (ECG) is itself a biometric fingerprint that can re-identify a patient and reveal personal information. Recent ECG anonymizers transform the signal before sharing to reduce privacy leakage. However, existing methods still face a privacy–utility trade-off, in which preserving privacy often compromises utility while preserving utility reveals personal information. We propose \emphREAN (\emphREconstruction-aware ECG \emphANonymizer), a raw ECG signal anonymizer, to address this privacy–utility trade-off. REAN reconstructs the signal using a 1-D U-Net trained with losses from frozen privacy and utility classifiers to reduce privacy leakage while preserving utility. The privacy and utility gradients are near-orthogonal ( \approx 93.8 ^\circ ), so reducing privacy leakage leaves utility almost unchanged. On four public PhysioNet databases, REAN achieves the strongest privacy–utility balance among raw ECG signal baselines. It drives re-identification to chance (0.96 \to 0.00), keeps arrhythmia macro-AUROC at the clean level (Clean 0.9982 vs.\ REAN 0.9991), and maintains re-identification protection under unseen privacy-classifier architectures.
[LG-13] SplineNet: An Isogeometric Deep Learning Method for Complex Shells
链接: https://arxiv.org/abs/2607.06026
作者: Shizhou Luo,Xiaodong Wei
类目: Machine Learning (cs.LG); Numerical Analysis (math.NA)
*备注:
Abstract:We present a novel isogeometric deep learning method, termed SplineNet, for the seamless design and analysis of shell structures with complex geometries. The proposed approach is built upon watertight spline representations, e.g., analysis-suitable unstructured T-splines, and features exact geometric descriptions of Computer-Aided Design (CAD) models in neural networks. Bézier extraction is used to build the network architecture, where Bernstein polynomials serve as the nonlinear activation functions. SplineNet can be applied in a data-free or data-driven way. In the data-free case, energy-based formulations can be naturally incorporated as loss terms, which fulfill the need of Computer-Aided Engineering (CAE) and can be accurately calculated. In particular, the Kirchhoff–Love (KL) model is adopted to solve for the mechanical behaviors of shell structures. This way, CAD and CAE can be tightly integrated in a deep neural network without the time-consuming model/data exchange process. In the data-driven case, SplineNet can be used as the trunk net of Deep Operator Networks (DeepONet) to provide interpretability. Given such a trained network and unseen input data, results can be immediately obtained without retraining the network or repeatedly performing the traditional workflow for analysis. In the end, a variety of numerical examples are studied to demonstrate the effectiveness of the proposed method, especially when real-world complex geometries are involved.
[LG-14] Learning When to Automate: Queue Control in Human-AI Service Systems
链接: https://arxiv.org/abs/2607.06017
作者: Giovanni Montanari,Marco Scarsini,Vianney Perchet
类目: Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注:
Abstract:We study a human-AI service system in which tasks arrive sequentially and are processed through a two-stage architecture: an automated chatbot followed, when necessary, by a human agent. We consider T sequentially arriving tasks, each belonging to one of K heterogeneous types. For each task the decision maker chooses how many resources to allocate to the chatbot, whose type-dependent success probabilities are initially unknown. Tasks not resolved by the chatbot enter type-dependent human-service queues, where they are processed by a human agent with unknown service rates. This model captures a central tradeoff in hybrid service systems: relying more on automation reduces human congestion but increases chatbot costs, while insufficient automation may overload the human agent. We propose the UCB-DPP policy, which combines Upper Confidence Bounds with Drift-Plus-Penalty control to learn the unknown parameters of the system while making queue-aware decisions. We prove that UCB-DPP achieves regret \widetilde\mathcalO(K\sqrtT) and guarantees mean-rate stability of the human-service queues. Simulations on synthetic instances show that the proposed policy outperforms natural baselines.
[LG-15] Stability Annealing Selects the Implicit Bias of Smoothed Sign Descent: A Rate-Indexed Barrier Path on Separable Data
链接: https://arxiv.org/abs/2607.06013
作者: Xiangwu Wang,Chengwei Cao,Yicheng Song,Ran Bi,Peilin Yu
类目: Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注: 17 pages, 9 figures
Abstract:Adaptive gradient methods can favor max-margin separators that differ from gradient descent, yet a fixed positive numerical stability constant eventually changes the update geometry again. This paper studies the rate-controlled middle case for full-batch linear classification on separable data. For memoryless stability-annealed smoothed-sign descent with weighted exponential loss, we prove that the normalized iterates converge to the minimizer of a convex Burg-type barrier over a margin slice. The proof rewrites the dynamics exactly as entropic mirror ascent on a concave dual objective, controls the dual gap by a KL recursion, and yields an explicit S_t^-1/2 normalized-iterate envelope. The static barrier geometry is fully characterized, including KKT conditions and both endpoint limits. Experiments validate the exact dual identities to floating-point error, illustrate the predicted path and rate diagram, and show an empirical fixed-epsilon crossover scaling in cumulative time. We further report robustness and boundary diagnostics for logistic tails, fixed-epsilon crossover, and adaptive-method variants, delineating the scope of the proved smoothed-sign theory.
[LG-16] Multi-Channel Spread-Spectrum Code Watermarking
链接: https://arxiv.org/abs/2607.06009
作者: Soohyeon Choi,Debin Gao,Yue Duan
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG); Software Engineering (cs.SE)
*备注: 15 pages, 3 figures, 16 tables, 2 algorithms
Abstract:Attributing code to the large language model that produced it is essential for provenance, licensing, and misuse accountability, yet no deployed watermark meets this need. Generation-time schemes require access to the producing model and cannot be applied to third-party code, while post-hoc schemes work on any code but carry at most 4 bits of payload, far too few to distinguish the many deployed model configurations. We present multi-channel spread-spectrum watermarking, the first post-hoc, training-free code watermark with a 24-bit payload and formal robustness guarantees. The scheme encodes bits in variable naming conventions and in eight pairs of semantically equivalent code patterns, and a keyed pseudo-random permutation maps every site to a codeword bit so that each bit receives multiple independent votes. Majority voting absorbs distributed corruption, while an outer Reed-Solomon code recovers the identifier when concentrated channel attacks defeat the vote, yielding provable robustness bounds for formatting, syntactic, and structural attacks. Across 1,750 Python files from CodeNet and from GPT-4.1 and Llama-4 generations, the watermark achieves 100% clean-detection accuracy with zero false positives. Under 17 attack types, it recovers the identifier at 97.6% accuracy under 8 variable renames and 94.1% under 10% random per-site corruption, while the strongest post-hoc baseline collapses to 0% under any single-transform attack. Embedding and detection together take under 200 ms on CPU without training data or GPU.
[LG-17] Discovering Frequent Closed Embedded Sub-DAGs in Spatio-Temporal Event Data
链接: https://arxiv.org/abs/2607.05995
作者: Piotr S. Maciąg
类目: Databases (cs.DB); Machine Learning (cs.LG)
*备注: Accepted as a conference publication at the PP-RAI 2026 conference
Abstract:We propose a novel approach to mine patterns in spatio-temporal event data based on discovering frequent closed embedded sub-Directed Acyclic Graphs (DAGs). In our method, event instances are represented as nodes labelled by event types, while edges capture spatio-temporal following relationships. We formally define the considered class of patterns and provide the rationale for focusing on closed sub-DAGs as compact and non-redundant representations of recurring interaction patterns. We implement the DigDag algorithm for mining such patterns and experimentally compare its efficiency with two related approaches: propagation pattern mining using the SLEUTH algorithm and Cascading Spatio-Temporal Pattern mining using the CSTPM algorithm. The experimental results demonstrate that our approach is substantially more efficient while operating under comparable parameter settings. Finally, we present a qualitative analysis of selected discovered patterns.
[LG-18] Learning Sparsest Linear Causal DAGs with Latent Confounders via Higher-Order Cumulants
链接: https://arxiv.org/abs/2607.05984
作者: Ming Cai,Hisayuki Hara
类目: Machine Learning (cs.LG)
*备注: 23 pages, 5 figures
Abstract:Recovering the exact directed acyclic graph (DAG) in linear non-Gaussian acyclic models with latent confounders (LvLiNGAM) remains a challenging problem. Although LvLiNGAM is identifiable only up to an observational equivalence class, each equivalence class is characterized by a unique sparsest DAG. Recovering the sparsest DAG from finite samples, however, remains difficult. Although existing methods are asymptotically consistent, they do not provide an explicit finite-sample procedure for recovering the unique sparsest DAG, nor do they handle models with an arbitrary number of latent confounders. In this paper, we propose a finite-sample method for recovering the sparsest DAG without imposing any restriction on the number of latent confounders. Simulation studies and real-data analyses demonstrate that the proposed method achieves superior finite-sample performance compared with existing approaches. Comments: 23 pages, 5 figures Subjects: Machine Learning (cs.LG) Cite as: arXiv:2607.05984 [cs.LG] (or arXiv:2607.05984v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2607.05984 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-19] Mitigating Errors in LLM -Generated Web API Invocations via Retrieval-Augmented Generation and Constrained Decoding ISCA
链接: https://arxiv.org/abs/2607.05936
作者: Daniel Maninger,Leon Chemnitz,Jannis Brugger,Tushar Lamba,Amir Molzam Sharifloo,Mira Mezini
类目: oftware Engineering (cs.SE); Machine Learning (cs.LG)
*备注: 54 pages, 11 figures; supersedes arXiv:2509.20172v6 , which is a discarded journal extension of our work
Abstract:Integration of web APIs is a cornerstone of modern software systems, yet writing correct web API invocation code remains challenging due to complex and evolving API specifications. Although LLMs are increasingly used for code generation, previous work has empirically shown that their ability to generate correct web API integrations is limited. At the same time, mitigation techniques and their effectiveness for this setting remain insufficiently understood. In this paper, we propose and systematically evaluate retrieval-augmented generation (RAG) and constrained decoding (CD) as two complementary approaches to improving LLM-generated web API invocation code. For RAG, we design a retriever that processes OpenAPI specifications and retrieves compact endpoint representations to inject into model prompts. For CD, we introduce an automatic translation from OpenAPI specifications to regex-based constraints enforced during generation. We evaluate both approaches on WAPIIBench’s existing synthetic dataset and on a new real-world dataset derived from GitHub repositories. Our results show that RAG reduces hallucinations and improves correctness when generating full API invocations but reduces it when the endpoint is already provided as it encourages the generation of unnecessary parameters. In contrast, CD reliably prevents illegal URLs, HTTP methods, and arguments and substantially improves overall correctness for both starter codes. Comments: 54 pages, 11 figures; supersedes arXiv:2509.20172v6, which is a discarded journal extension of our work Subjects: Software Engineering (cs.SE); Machine Learning (cs.LG) Cite as: arXiv:2607.05936 [cs.SE] (or arXiv:2607.05936v1 [cs.SE] for this version) https://doi.org/10.48550/arXiv.2607.05936 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-20] Energy-Efficient GPU DVFS for Fine-Tuning of SLMs on Resource-constrained Embedded Devices
链接: https://arxiv.org/abs/2607.05933
作者: Jurn-Gyu Park,Sanzhar Zholdybayev,Aidar Amangeldi,Ademi Zhanuzakova
类目: Performance (cs.PF); Machine Learning (cs.LG)
*备注:
Abstract:Dynamic Voltage Frequency Scaling (DVFS) on resource-constrained embedded GPU platforms is essential for energy-efficient small language model (SLM) fine-tuning, as privacy- and personalization-driven adaptation increasingly requires local execution and involves repeated forward-backward optimization over many mini-batches, making it substantially more time- and energy-intensive than single-pass inference. To this end, 1) we first characterize the fine-tuning behavior of representative encoder-only SLMs of BERT variants, and autoregressive decoder-only SLMs of Pythia variants on GLUE benchmarks. In addition to the characterizations, 2) we propose a simple yet effective ML-based model selection that selects energy-optimal GPU DVFS settings on resource-constrained embedded platforms. Our results on NVIDIA Jetson AGX Orin demonstrate average 13.11% energy savings (up to 26.73%) over MAXN Mode 0, which has no explicit power cap.
[LG-21] Drift Happens: An Empirical Study of Neural Architecture Robustness to Temporal Distribution Shift
链接: https://arxiv.org/abs/2607.05908
作者: Robin Holzinger(1),Riccardo Colletti(1) ((1) Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, USA)
类目: Machine Learning (cs.LG)
*备注: 33 pages, 28 figures. Extended version of the QCDS 2026 proceedings paper; proceedings version to appear in Springer LNCS. Interactive website: this https URL
Abstract:Real-world data distributions evolve over time, inducing temporal distribution shift that can substantially degrade the reliability of deployed machine learning systems. However, the extent to which architectural choices and their associated inductive biases affect temporal robustness remains insufficiently understood. We present a systematic empirical comparison of temporal robustness across three heterogeneous, time-indexed domains encompassing image classification, multi-label text classification, and text regression tasks. Using a unified evaluation framework based on temporal drift matrices, we train models on cumulative historical data and evaluate their performance on both earlier and later time periods, thereby quantifying cross-temporal generalization. Our study spans model families ranging from simple multilayer perceptrons and convolutional networks to recurrent networks and pretrained Transformer-based encoders. Collectively, the results show that architectural inductive biases systematically shape temporal robustness: models whose inductive biases lead them to exploit localized, highly discriminative features attain the highest in-distribution accuracy, yet those features are often the ones that change most over time, so these models degrade fastest, while pretrained encoders that draw on coarser, more stable representations drift more gradually. These observations offer practical guidance for selecting architectures for real-world systems subject to temporal drift. Comments: 33 pages, 28 figures. Extended version of the QCDS 2026 proceedings paper; proceedings version to appear in Springer LNCS. Interactive website: this https URL Subjects: Machine Learning (cs.LG) MSC classes: 68T07, 68T05 ACMclasses: I.2.6; I.5.2 Cite as: arXiv:2607.05908 [cs.LG] (or arXiv:2607.05908v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2607.05908 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-22] More Convincing Not More Correct: Self-Play Reward Hacking of Reference-Free LLM Judges
链接: https://arxiv.org/abs/2607.05904
作者: Chenyu Zhou
类目: Machine Learning (cs.LG)
*备注: 9 pages main text, 15 pages total including references and appendix; 4 figures
Abstract:Training a language model against its own reference-free judgments (the premise of self-rewarding, self-play, and LLM-as-a-judge pipelines) assumes a model’s verdict on a shown answer tracks correctness. We show it fails structurally: conditioned on a candidate, a judge scores plausibility, not correctness, leaving false-positive basins a policy learns to exploit. We measure this with a hidden-anchor audit: a held-out, cross-source exact-match check the judge never sees. On GSM8K with Qwen3 policies, self-play drives the judge’s pass rate from 0.72 to 0.94 while true accuracy stays at 0.20 (three seeds). This reward hacking is not white-box gaming: the errors transfer across judge families (Qwen, Llama, Gemma) and scales, a strict three-judge ensemble still accepts 55% of them, and no plausibility-scoring defense closes the basin. The decisive variable is whether the judge commits an answer of its own before using the candidate: committing first drops the false-positive rate from 0.719 to 0.012, blind solving lifts discrimination to 0.96, and used as the training reward the de-anchored channel keeps false positives at zero, preventing the basin rather than only detecting it. A falsifiable bound (the gap is at most 1 - accuracy) predicts which regimes are exposed. The full arc replicates without training under best-of-N selection in code and competition math, and with a Gemma policy.
[LG-23] Auditing of Unlearning Algorithms
链接: https://arxiv.org/abs/2607.05898
作者: Sahasrajit Sarmasarkar,Anastasia Koloskova,Sanmi Koyejo
类目: Machine Learning (cs.LG); Cryptography and Security (cs.CR)
*备注:
Abstract:Evaluating whether unlearning algorithms truly remove training data influence remains an open challenge. We propose a practical auditor that computes data-dependent lower bounds on the unlearning parameter \varepsilon using membership inference attacks. Evaluating multiple unlearning algorithms, we find a sharp separation: algorithms with rigorous guarantees, such as model clipping and rewind-to-delete, achieve very small \varepsilon bounds that do not falsify their unlearning guarantees, whereas empirical methods such as Hessian-based unlearning, interleaved ascent-descent, ascent on the forget set, and fine-tuning on the retain set exhibit large bounds, indicating poor unlearning. Our auditor provides a practical tool for empirically falsifying unlearning claims through a hypothesis-testing framework, and we validate it on CIFAR-100 and Shakespeare text.
[LG-24] No Subspace to Track: Non-Identifiability and Optimizer State in Low-Rank Training
链接: https://arxiv.org/abs/2607.05872
作者: Noel Thomas
类目: Machine Learning (cs.LG); Optimization and Control (math.OC); Machine Learning (stat.ML)
*备注: 21 pages, 5 figures
Abstract:Memory-efficient optimizers such as GaLore train large language models by projecting gradients onto a rank-r subspace recomputed every T steps, assuming this subspace is a slowly drifting object that can be tracked. We show that beyond a small reproducible core, there is no such object. Two estimates of the top-r subspace computed at the same step from disjoint minibatches disagree as much as estimates computed T steps apart (0.73 vs 0.74 of the maximal chordal distance sqrt(2r), at Pythia-160M with r=128): the apparent rotation at each refresh is dominated by estimator noise. This holds across four model families in three architecture classes from 70M to 6.9B parameters, strengthening with scale, and more weakly in a vision transformer. Only ~39 of 128 directions are reproducible across minibatches, and averaging cannot recover the rest: under N-fold averaging the gradient’s spectral tail shrinks as N^(-1/4) rather than the N^(-1/2) of pure noise, so no averaging budget makes the subspace well defined. What helps instead follows from treating each refresh as a change of coordinates for Adam’s state. Carrying the second moment blindly is provably about (r-k*)/2 worse than the best rotation-blind estimator, while the first moment transports exactly through the rotation, the optimal linear map under isotropic gradients and the rule LDAdam uses. At 1B over 40k steps (3 seeds), full LDAdam reaches 18.7 perplexity at beta2=0.999, beating untransported GaLore after its best beta2 fix (19.3); shortening the second-moment memory to beta2=0.99 helps the refreshing optimizers, though for canonical GaLore the effect is small and a full-rank control reverses it. One measurable fact, subspace non-identifiability, clarifies why GaLore works, which patches work, and what to check before trusting a low-rank assumption: the reproducible rank k*.
[LG-25] Strategic Bargaining in Multi-Buyer Markets: Reinforcement Learning from Verifiable Rewards for LLM Negotiations
链接: https://arxiv.org/abs/2607.05863
作者: Shuze Daniel Liu,Claire Chen,Jiabao Sean Xiao,Xin Chen,David Simchi-Levi
类目: Machine Learning (cs.LG); Computer Science and Game Theory (cs.GT)
*备注:
Abstract:Negotiation is a fundamental strategic interaction in management science, characterized by agents attempting to reach agreements while protecting private information, such as reservation costs and hidden valuations. A prevalent yet complex scenario involves a single seller negotiating concurrently with multiple buyers, each possessing heterogeneous, private budgets. In such settings, constrained by a limited number of communication turns, the seller must balance exploring the broader market to discover the highest valuation with concentrating sufficient turns on a single target buyer to secure the best possible outcome. Our analysis reveals a significant gap in standard Large Language Models (LLMs): while these models are linguistically proficient, they fail to act as effective economic decision-makers. Specifically, they exhibit a failure to explore the buyer pool, often fixating on the current highest bid rather than strategically investigating the market to discover latent high valuations. In this paper, we propose a specialized training recipe using Reinforcement Learning from Verifiable Rewards (RLVR). By anchoring the reward function to objective economic outcomes, the strategic balance between market discovery and surplus extraction emerges natively through the learning process. Our results demonstrate that the trained seller undergoes a multi-stage strategic evolution, learning to leverage price anchoring and strategic probing to identify more profitable counterparties. The agent extracts a substantially higher surplus than frontier models by both improving its persuasive bargaining skills and consistently closing deals with high-value buyers. Finally, we show that our seller strategies generalize robustly to unseen buyer negotiation styles and budget distributions. Subjects: Machine Learning (cs.LG); Computer Science and Game Theory (cs.GT) MSC classes: 90B50, 90C40, 68T05, 91A80, 91B26 ACMclasses: I.2.6; I.2.7; I.2.11; H.4.2; J.4 Cite as: arXiv:2607.05863 [cs.LG] (or arXiv:2607.05863v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2607.05863 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-26] Level-Crossing Density as a Mesh-Free High-Frequency Auxiliary Loss for Implicit Neural Representations
链接: https://arxiv.org/abs/2607.05815
作者: Gunner Levi Howe
类目: Machine Learning (cs.LG)
*备注: 14 pages, 8 figures, 4 tables. Code, raw experiment logs, and estimator unit tests: this https URL
Abstract:The Minkowski functionals of a field’s excursion sets – area, boundary measure, and Euler characteristic – describe its level-set morphology; the Euler characteristic is the cheapest handle on topology. We derive smooth Monte-Carlo estimators for all three of a continuous neural field, evaluated at scattered points via the co-area formula and Gauss-Bonnet, using only autodiff: no grid, no complex, no persistence. The estimator is accurate to 1-3% against exact topology in 2D and 3D, and costs about 3 ms per iteration where a persistent-homology (PH) loss on a cubical grid costs 650-1000 ms – a 250x gap. We establish four design rules without which these losses silently fail: a dense level ladder (invariants are flat in the parameters away from transitions), a C^2 backbone (ReLU nets hide curvature in kinks), the full Minkowski vector (Euler characteristic alone is an alternating sum, gamed by debris-hole cancellation; pricing perimeter closes the channel), and sampling-scale coverage. In 2D the vector-valued cap is the only method in a controlled comparison that both repairs topology (3/3 seeds) and preserves fidelity – uniform smoothing repairs at 11-17x the fidelity cost, and the Euler term alone repairs nothing. In 3D neural-SDF fitting, however, a failure mode we believe general to any sampled soft topology objective appears: gradient descent adversarially hides topological noise below the sampling density, where the estimator is blind – spurious-feature counts are invariant to 4x more samples, and closing the window needs cubically many points, erasing the cost advantage. A grid-based PH baseline, whose complex is the evaluation resolution, solves the same benchmark ( 4/9 exact; median b_1 error 1 vs. ours above 10^4 ). The 250x cost of persistence is, at present, the price of having no null space. We release estimators, receipts, and benchmarks.
[LG-27] Contextual Procurement Auctions with Bandit Learning
链接: https://arxiv.org/abs/2607.05813
作者: Yiling Chen,Shi Feng,Sadie Zhao
类目: Computer Science and Game Theory (cs.GT); Machine Learning (cs.LG)
*备注:
Abstract:We study repeated contextual procurement auctions in which the platform must learn context-dependent product values from bandit feedback. We give an exactly truthful explore-then-commit mechanism with \widetilde O((ng)^1/3T^2/3) regret. We also give a frozen-payment UCB mechanism with a regret-incentive tradeoff: the near-UCB tuning attains (\widetilde O(\sqrtngT)) welfare regret, while for fixed (n,g) its total incentive error is (\widetilde O(T^3/4)); the balanced tuning gives (\widetilde O(T^2/3)) on both scales. Regret is measured as welfare loss relative to the full-information efficient allocation. We prove a matching lower bound for the frozen-payment regret-incentive tradeoff.
[LG-28] Heckman-Corrected Epistemic Uncertainty: Selection on Unobservables Defeats Importance Weighting
链接: https://arxiv.org/abs/2607.05806
作者: Gunner Levi Howe
类目: Machine Learning (cs.LG)
*备注: 9 pages, 4 figures. Every number machine-generated from committed result JSONs (regenerate-and-diff verified); faithfulness-gated Heckman implementation reproduces published Stata reference output. Code, data, and tests archived at this https URL . Companion paper (survivor bias in learning-curve surrogates) forthcoming
Abstract:Training data for machine learning is routinely collected by a selection process the model never sees: loans are observed only when granted, outcomes only when a test was ordered. The standard fixes – importance weighting, covariate-shift correction, MAR imputation – assume selection is ignorable given observables. Econometrics solved the harder case in 1979: Heckman’s two-equation model jointly fits a probit selection equation and an outcome equation linked through correlated errors, and the inverse-Mills-ratio term corrects for selection on unobservables, where importance weighting is structurally helpless. We instantiate this for deep epistemic uncertainty: a deep outcome network, a linear selection head, and a joint bivariate-normal likelihood over all units, ensembled for predictive variance. In a controlled generator where sampling probability depends on an unobservable correlated (rho up to 0.9) with the outcome noise, deep ensembles, MC dropout, and GP baselines are overconfident exactly where data was avoided: coverage of nominal-90% intervals falls to 64.4% at rho=0.9, and importance weighting with oracle propensities does not fix it (43.1%) – reweighting corrects the covariate distribution, not the conditional bias E[y|x,selected] != E[y|x]. The Heckman correction restores coverage (88.9%) when the selection equation has an instrument – a variable affecting selection but not the outcome – and degrades measurably without one (40.3%); we chart this honesty curve rather than hide it. On real tabular data with induced MNAR selection, the corrected intervals are the best-calibrated (lowest region-ECE) non-oracle method in selected-against regions; baselines matching its raw coverage do so only by over-widening everywhere. Our estimators reproduce classic Stata output to seven digits. We state which identification regime a practitioner is in, and release the code.
[LG-29] wo Sides of the Same Coin: Learning the Backdoor to Remove the Backdoor
链接: https://arxiv.org/abs/2607.05748
作者: Qi Zhao,Christian Wressnegger
类目: Machine Learning (cs.LG)
*备注:
Abstract:The community has recently developed various training-time defenses to counter neural backdoors introduced through data poisoning. In light of the observation that a model learns poisonous samples responsible for the backdoor easier than benign samples, these approaches either use a fixed threshold of the training loss for splitting or iteratively learn a reference model as an oracle for identifying benign samples. In particular, the latter has proven effective for anti-backdoor learning. Our method, HARVEY, leverages a similar yet crucially different technique: learning an oracle for poisonous rather than benign samples. Learning a backdoored reference model is significantly easier than learning a reference model on benign data. Consequently, we can identify poisonous samples much more accurately than related work identifies benign samples. This crucial difference enables near-perfect backdoor removal as we demonstrate in our evaluation. HARVEY substantially outperforms related approaches across attack types, datasets, and architectures, lowering the attack success rate to the very minimum at a negligible loss in natural accuracy. The figure below shows an overview of our methods working principle. Subjects: Machine Learning (cs.LG) Cite as: arXiv:2607.05748 [cs.LG] (or arXiv:2607.05748v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2607.05748 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Journalreference: AAAI 2025 Related DOI: https://doi.org/10.1609/aaai.v39i21.34441 Focus to learn more DOI(s) linking to related resources
[LG-30] Multimodal Molecular Representation Learning with Graph Neural Networks Deep Cross Networks and SMILES Embeddings
链接: https://arxiv.org/abs/2607.05736
作者: Qiwei Han,Chi Zhou,Ruobing Wang,Zheng Ma
类目: Machine Learning (cs.LG)
*备注: 14 pages, 3 figures
Abstract:Molecular property prediction often relies on isolated data modalities, where continuous 3D graph neural networks (GNNs) struggle to efficiently capture long-range topological dependencies and exact macroscopic heuristics. In this work, we introduce a parameter-efficient Tri-Branch Modular Fusion Neural Network that synthesizes three orthogonal modalities: 3D spatial geometry (SchNet), discrete topological grammar (SMILES via ChemBERTa), and explicit macroscopic physicochemical descriptors (Deep Cross Network). By bypassing standard scalar readouts and employing a shared late-fusion architecture, the framework establishes a mathematically rigorous multimodal latent space that effectively resolves the arithmetic and oversmoothing limitations of local message passing. We evaluate the proposed architecture on the QM9 benchmark, targeting the extensive thermodynamic property of atomization energy at 0 K ( U_0^\mathrmatom ). Through systematic combinatorial ablation and latent bottleneck optimization ( d_e=64 ), the tri-modal framework achieves a validation Mean Absolute Error (MAE) of 0.0207 eV. Operating with fewer than one million parameters, this architecture decisively surpasses the sub-chemical accuracy threshold and yields a substantial 20.6% error reduction over a strictly controlled geometric baseline. Ultimately, our findings demonstrate that integrating orthogonal macroscopic and topological data streams provides a synergistic, \mathcalO(1) physical shortcut. This multimodal alignment offers a highly efficient alternative to brute-force parameter scaling, establishing a robust surrogate model for high-throughput virtual screening (HTVS) pipelines.
[LG-31] Low-Overhead Error-Corrected QCNNs Using Bivariate Bicycle Codes
链接: https://arxiv.org/abs/2607.05724
作者: Alejandro Rosales,Animesh Yadav
类目: Machine Learning (cs.LG); Quantum Physics (quant-ph)
*备注: 10 pages, 6 figures, under review
Abstract:Quantum convolutional neural networks (QCNNs) combine the power of quantum computing and classical CNN for computational speedup in classification tasks. However, noise levels on state-of-the-art quantum devices remain too high for practical QCNN execution. In addition, despite the reliable surface code providing a method for error rates below a threshold value, they have a prohibitively large qubit cost. Recently introduced bivariate bicycle (BB) codes are of particular interest for their high error threshold, constant encoding rate, and linear code distance. Through simulation with realistic hardware noise sources, we demonstrate that a 4-qubit unprotected QCNN fails to converge and exhibits a worse learning rate compared to numerical simulations. Addressing both limitations, we propose a distance-4 BB quantum error-correction (QEC) technique for QCNNs. In doing so, we validate that our low-overhead QEC technique for QCNNS represents a step toward practical QCNNs.
[LG-32] Uncertainty-Aware Velocity Correction for Proprioceptive Vehicle Localization using Evidential Mamba
链接: https://arxiv.org/abs/2607.05669
作者: Abinav Kalyanasundaram,Karthikeyan Chandra Sekaran,Wolfgang Utschick,Michael Botsch
类目: Robotics (cs.RO); Machine Learning (cs.LG)
*备注: Accepted at the 2026 International Conference on Indoor Positioning and Indoor Navigation (IPIN 2026), Rome, Italy. 6 pages, 4 figures
Abstract:Reliable localization in GNSS-denied environments remains a fundamental challenge for intelligent vehicles, as inertial navigation systems accumulate unbounded drift without external correction. Existing approaches provide drift correction through dedicated infrastructure, expensive external sensors, or complex multi-sensor fusion, each introducing practical deployment barriers. We propose Evidential Velocity Correction using Mamba (EVC-Mamba), a learning-based architecture that transforms onboard vehicle sensor data into a virtual velocity sensor for IMU drift correction without additional hardware. A Mamba-based selective state space model captures the temporal dynamics of vehicle motion, while evidential deep learning with a Normal-Inverse-Gamma distribution provides principled uncertainty quantification. The resulting uncertainty-aware velocity estimate is incorporated as a virtual correction measurement into an Error-State Extended Kalman Filter to reduce position drift. Evaluation on real-world vehicle data demonstrates that inertial navigation using the proposed velocity correction achieves localization accuracy within 10% of a dedicated external velocity sensor across different outage durations. The proposed architecture supports real-time onboard deployment at 40 Hz on edge hardware, enabling reliable localization during prolonged GNSS outages.
[LG-33] Orthogonal Dendritic Intrinsic Networks: An Architecture for Significance-Ordered Orthogonal Latent Spaces
链接: https://arxiv.org/abs/2607.05653
作者: Jeanie Schreiber,Tyrus Berry,Zeeshan Ahmed
类目: Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注:
Abstract:Principal Component Analysis or PCA-like properties (orthogonality, variance ranking) are seldom realized in deep autoencoder architectures. In this work, we present ODIN (Orthogonal Dendritic Intrinsic Network), a novel autoencoder architecture that recovers PCA-like latent structure in a fully non-linear regime. By incorporating a set of geometric constraints directly into the training objective, ODIN encourages latent dimensions to be mutually orthogonal and ordered by explained variance, mirroring the interpretable decomposition of PCA while retaining the expressive power of deep networks. We provide theoretical grounding for these constraints and demonstrate their compatibility with standard encoder-decoder frameworks. We also establish empirical results for both synthetic and real world datasets, establishing a principled path toward interpretable, structured feature learning and dimensionality reduction.
[LG-34] Domain-Adaptive Climate Downscaling Under Temporal Distribution Shift
链接: https://arxiv.org/abs/2607.05645
作者: Shuochen Wang,Nishant Yadav,Auroop R. Ganguly
类目: Machine Learning (cs.LG); Atmospheric and Oceanic Physics (physics.ao-ph)
*备注:
Abstract:Deep-learning-based climate downscaling aims to learn relationships from historical low-resolution (LR) and high-resolution (HR) climate data to generate HR climate projections. However, this setting faces a temporal out-of-distribution (OOD) challenge: models trained on historical data are commonly applied to future projections whose distributions may differ substantially from the training period. This study investigates temporal OOD shift for daily temperature downscaling over the Continental United States using paired LR-HR model simulations. We propose a temporal domain-adaptive downscaling framework that combines supervised HR reconstruction on historical data with domain alignment between historical and future climate distributions. Experiments across future validation periods show that the proposed domain-adaptive model consistently outperforms statistical and deep-learning-based bias-correction methods, with the largest gains occurring when the temporal distribution shift is strongest. Spatial analyses indicate stronger improvements over high-elevation and topographically complex regions, along with higher spatiotemporal correlation with the HR target. The extreme analysis shows that domain adaptation also reduces upper-tail temperature bias relative to the non-adaptive model. These results demonstrate that temporal domain adaptation can improve the robustness of HR climate projections under non-stationary climate conditions.
[LG-35] Intuitionistic Fuzzy Graph Embedded Random Vector Functional Link with Multiview Learning
链接: https://arxiv.org/abs/2607.05635
作者: Vrushank Ahire,Yogesh Kumar,M.A. Ganaie
类目: Machine Learning (cs.LG)
*备注:
Abstract:Random Vector Functional Link (RVFL) networks are popular due to their fast training and universal approximation capabilities. However, RVFL models face challenges in preserving geometric relationships and utilizing multiple feature views effectively. To address these limitations we propose the Intuitionistic Fuzzy Graph Embedded Random Vector Functional Link with Multiview Learning (IFGRVFL-MV) model. The proposed approach comprises three key components: intuitionistic fuzzy sets for uncertainty handling, graph embedding to capture intrinsic geometric structures, and multiview learning to use complementary information from multiple feature spaces. The model assigns intuitionistic fuzzy membership and non-membership values to data points making it robust to outliers. Also, the graph embedding framework preserves topological structures, increasing the generalization performance. We performed experiments on benchmark datasets from UCI and KEEL repositories which concludes that IFGRVFL-MV outperforms existing models in classification accuracy. Our results establish that IFGRVFL-MV is a promising advancement in the domain of uncertainty and multiview environments.
[LG-36] A Coin Flip Per Token: Bernoulli Sparse Steering of Large Language Models
链接: https://arxiv.org/abs/2607.05615
作者: Nima Eshraghi,Lovedeep Gondara,Yuqing Huang,Sagarika Suresh,Leizer Teran,Jithin Pradeep,Xiaotong Xu,Fanny Chevalier
类目: Machine Learning (cs.LG)
*备注:
Abstract:Activation steering via sparse autoencoders (SAEs) enables behavioral control of large language models without task-specific fine-tuning, but standard methods apply the steering signal at every generated token, incurring constant per-token perturbation that risks degrading fluency. We ask: is dense intervention necessary? We introduce Stochastic Token Steering (STS), which gates each token independently with probability p , and Stochastic Block Steering (SBS), which gates a leading window once per sequence; neither requires a reward model or learned gating policy. Across two model families and two behavioral tasks, steering only 50% of the tokens recovers most of the dense-steering effect while preserving fluency, and steering as few as 30% surpasses prompt-based control. The optimal steering magnitude scales inversely with the intervention ratio, revealing that SAE-mediated control is rate-limited: the behavioral outcome depends on cumulative signal dosage across a sequence.
[LG-37] SafeImpute: Reliable Clinical Data Imputation via Conformal Selection KDD2026
链接: https://arxiv.org/abs/2607.05613
作者: Xinrui He,Mengting Ai,Junting Wang,Curtiss B. Cook,Jingrui He
类目: Machine Learning (cs.LG)
*备注: Accepted at KDD 2026. Author accepted manuscript
Abstract:Clinical care often relies on key laboratory indicators, yet real-world patient visits are sparse and tests are ordered irregularly, leading to pervasive missingness. While many imputation methods improve average accuracy, they provide limited guidance on which imputed values are reliable enough for high-stakes downstream use. In this work, we study reliable clinical imputation, aiming to produce accurate imputations while selectively releasing the reliable results, with statistical control over clinically unacceptable errors. To achieve this goal, we propose SafeImpute, a reliable imputation framework for irregular and sparse clinical longitudinal records. SafeImpute constructs an event graph that captures both intra-patient temporal trajectories and inter-patient clinical similarity, and learns imputations with a two-relation GNN and adaptive fusion, regularized by an auxiliary masked reconstruction objective. For reliability guarantees, SafeImpute converts a proxy risk score into conformal p-values and applies the Benjamini–Hochberg procedure to control the false discovery rate (FDR) of unacceptable errors among released imputations at a user-specified tolerance. Experiments on our Mayo Clinic data, the public MIMIC-III and MIMIC-IV datasets show that SafeImpute achieves strong imputation accuracy while providing reliable error control, outperforming diverse baselines in both standard imputation evaluation and FDR-controlled selective-release evaluation.
[LG-38] EquiFiLM: Charge-Conditioned Equivariant Force Fields via Feature-wise Linear Modulation
链接: https://arxiv.org/abs/2607.05559
作者: Samuel Sahel-Schackis,Ken-ichi Nomura,Aiichiro Nakano,Matthias F. Kling,Thomas Linker
类目: Machine Learning (cs.LG); Materials Science (cond-mat.mtrl-sci); Chemical Physics (physics.chem-ph); Computational Physics (physics.comp-ph)
*备注:
Abstract:Foundation machine learning force fields (MLFFs) such as MACE-MP-0 and UMA cover broad chemical space at near density functional theory (DFT) accuracy. However, they assume equilibrium ground-state physics and do not natively handle externally induced changes to the electronic state, such as charging, applied fields, or electronic excitation, which limits their use for driven processes such as photoexcitation and charge injection. We propose EquiFiLM, a lightweight extension that adds continuous external conditioning to any equivariant foundation MLFF via a per-layer Feature-wise Linear Modulation (FiLM) block, learning externally driven changes to the potential energy surface from minimal training data. The block modulates only scalar channels and preserves E(3)-equivariance exactly. We demonstrate the recipe on charged liquid water with the foundation model MACE-MatPES as the backbone, yielding E-MACE. On the four training charges, E-MACE delivers a 3.1\times reduction in force RMSE ( 21.3 to 6.96 meV/ \mathringA ) and a 61\times reduction in per-atom energy RMSE ( 6.1 to 0.1 meV/atom) over a baseline without EquiFiLM trained on the same data, at indistinguishable inference cost. Across seven held-out interpolation and extrapolation charges, force RMSE stays within 18-61 meV/ \mathringA and energy RMSE within 0.7-5.4 meV/atom. The model runs stable molecular dynamics across the full range tested and predicts the charge-dependent first-shell response of the reduced pair distribution function probed by ultrafast electron diffraction. Adding this conditioning axis to the foundation requires only a few thousand DFT-labeled frames, against the \approx 10^8 structures of a charge-aware foundation trained from scratch. The recipe is backbone- and conditioning-agnostic: it applies without architectural change to any equivariant MLFF with scalar interaction-layer channels.
[LG-39] Federated Physics-Grounded Reinforcement Learning for Distributed Stability Control in Smart Grids
链接: https://arxiv.org/abs/2607.05553
作者: Omar Al-Refai,Ibrahim Shahbaz,Adam Ali Husseinat,Eman Hammad
类目: Machine Learning (cs.LG); Systems and Control (eess.SY)
*备注: Accepted at IEEE SmartGridComm 2026. This is the accepted manuscript version. The final published version will appear in IEEE Xplore
Abstract:Transient stability control in smart grids requires rapid post-fault damping of generator frequency and rotor angle deviations to prevent cascading failures. This paper proposes FedPPO-PG, a Federated Multi-Agent Proximal Policy Optimization framework with Physics-Grounded neighborhoods, which reformulates transient stability control as a cooperative multi-agent reinforcement learning problem optimized directly against closed-loop stability objectives. Each generator hosts an independent local actor augmented with the frequency deviations of its two most strongly coupled electrical neighbors, identified from the post-fault Kron-reduced susceptance matrix. A guided policy initialization phase warm-starts all actors from the classical decentralized controller, while a centralized critic guides advantage estimation under the centralized training–decentralized execution (CTDE) paradigm. Evaluated on a simulation of the IEEE 39-bus benchmark system across five training and three unseen fault contingencies, FedPPO-PG achieves 100% stabilization in all 24 trials, reduces mean stability time by 72.4%, and cuts the control power by 7-14 times compared to the centralized baseline. Each actor executes independently with no central coordinator at deployment, and the per-actor inference latency satisfies the IEEE/IEC 60255-118-1-2018 real-time reporting requirements.
[LG-40] InvWeaver: Deductive Feedback for Invariant Synthesis in Interacting-Loop Programs
链接: https://arxiv.org/abs/2607.05478
作者: Guangyuan Wu,Weining Cao,Zehui Tan,Yuan Yao,Hengfeng Wei,Taolue Chen,Xiaoxing Ma
类目: Machine Learning (cs.LG); Programming Languages (cs.PL)
*备注:
Abstract:Loop invariant inference is a fundamental yet challenging problem in program verification. Recent LLM-aided guess-and-check techniques have shown strong performance on single-loop programs, but they often struggle with programs containing multiple interacting loops. This paper presents InvWeaver, a neuro-symbolic framework for synthesizing invariants for such programs. The key idea is to expose inter-loop dependencies and propagate proof obligations through a combination of loop-level abstraction, obligation-guided inference, and weakest-precondition-based refinement. We evaluate InvWeaver on a comprehensive benchmark suite, including a newly curated dataset derived from classic algorithms. Experimental results show that InvWeaver substantially outperforms existing invariant inference methods, solving 72 out of 82 multi-loop benchmark problems and maintaining strong performance on single-loop tasks.
[LG-41] Parameter-Free Encoders Remain Viable for RDB Foundation Models ICML2026
链接: https://arxiv.org/abs/2607.05476
作者: Linjie Xu,David Wipf
类目: Machine Learning (cs.LG)
*备注: ICML 2026 Workshop on Foundation Models for Structured Data
Abstract:Given a relational database (RDB) storing heterogeneous tabular information, how can we predict missing (or future) values in some target column of interest? As the space of potential targets is vast across enterprise settings, it is preferable to avoid learning a new model from scratch each time there is a new prediction task. Frozen foundation models based on RDB-specific encoders provide a viable solution, but ideal design remains an open question. On the one hand, it has recently been argued that certain parameter-free subgraph encoders combined with single-table foundation models can achieve near SOTA performance, with no RDB-specific pre-training required. Meanwhile, other contemporary studies advocate for parameterized encoders pre-trained to exploit observable labels for learning task-specific representations. To address this ambiguity, we analyze RDB encoder properties specifically when labels are present as inputs, proving limitations on the potential efficacy of trainable encoder parameters. As empirical validation, we demonstrate that considerably simpler parameter-free encoders are still capable of strong performance across many relevant benchmarking tasks.
[LG-42] Exogenous Dropout: A Simple Strong Baseline for Corruption-Robust Time Series Forecasting with Covariates
链接: https://arxiv.org/abs/2607.05452
作者: Hao Hu,Xue-shan Ai
类目: Machine Learning (cs.LG)
*备注: 21 pages, 4 figures, 6 tables
Abstract:Time series forecasters that use exogenous covariates are fragile in deployment: when those covariates are noised, temporally misaligned, or missing, strong exogenous-fusion and exogenous-adapted models can degrade far above the endogenous-only floor. We study whether such robustness requires specialized architectures, or whether it can be obtained through a simple training intervention. We propose exogenous dropout, a model-agnostic method that randomly zeros whole exogenous channels during training. Across electricity-price forecasting, reservoir hydrology, and meteorology, exogenous dropout substantially improves robustness under Gaussian noise, temporal misalignment, and fully missing channels, while preserving clean accuracy. Applied to a dual-correlation network, it yields the most robust model in our experiments, outperforming a deliberately strong bounded architectural foil, BoundEx, which combines a learnable gate, a fallback residual to the endogenous backbone, and per-channel exogenous FiLM modulation. Architecture-by-dropout ablations, gate-behavior diagnostics, and a representation-level bound show that explicit architectural boundedness is not necessary for this robustness: an unbounded model trained with exogenous dropout is more robust than the bounded model in every domain. We release a corruption-robustness benchmark and recommend exogenous dropout as a simple, strong baseline for future work on time series forecasting with covariates.
[LG-43] Design-CP: Context Parallelism for Design of Protein Nanoparticles ICML2026
链接: https://arxiv.org/abs/2607.05439
作者: Lorenzo Tarricone,Helen E. Eisenach,Aiko Muraishi,Charlotte M. Deane
类目: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC); Quantitative Methods (q-bio.QM)
*备注: Accepted at the 2026 Workshop on Generative and Agentic AI for Biology (ICML 2026)
Abstract:Many all-atom generative protein models can in principle design large multimeric complexes by jointly modelling all chains, but their quadratic token- and atom-pair representations quickly exceed single-GPU memory as the number of chains and residues modelled grows. We introduce Design-CP, two context-parallel (CP) inference strategies for RFdiffusion 3 (1D row-sharding and 2D grid sharding with ring attention) that distribute the quadratic activations across a multi-GPU mesh while preserving pretrained weights. We characterise their scaling when sampling icosahedral assemblies, showing that the maximum feasible asymmetric subunit (ASU) size grows with the expected square-root trend in GPU count and that 2D sharding achieves better wall-clock scaling. Moreover, we show how strong point-group symmetry constraints make CP usable out of the box for end-to-end, all-atom design of icosahedral nanoparticles, yielding favourable in silico structural and interface metrics. Finally, we demonstrate octahedral nanoparticle design on a small cluster of workstation-grade 16GB GPUs, illustrating how Design-CP can be a practical path towards democratising large-assembly protein design.
[LG-44] Statistically Meaningful Geometry and Gauge Symmetry Breaking: A Geometric Foundation for Scientific Discovery and Intelligence Emergence
链接: https://arxiv.org/abs/2607.05436
作者: Bing Cheng,Yi-Shuai Niu,Howell Tong,Shing-Tung Yau
类目: Machine Learning (cs.LG)
*备注:
Abstract:The rapid scaling of over-parameterized machine learning architectures, particularly LLMs, raises a profound crisis: do these systems exhibit genuine intelligence, or are they merely sophisticated statistical pattern matchers? Classical flat Euclidean statistics cannot differentiate continuous interpolation from the autonomous discovery of novel causal laws. To resolve this, we introduce Statistically Meaningful Geometry (SMG), a framework modeling over-parameterized learning systems as infinite-dimensional non-parametric Orlicz fiber bundles. We prove that under persistent out-of-distribution (OOD) stimuli governed by unmodeled causal mechanisms, continuous optimization fails. Unmodeled variance is rejected by the visible horizontal base manifold, leaking into the unobservable vertical fiber space and generating an accumulation of Active Acausal Tension. Driven by the statistical manifold’s non-linear curvature, this tension inevitably strikes a conjugate focal boundary ( T_\textcrit = \pi^2 / K_\textmax ), triggering localized volumetric collapse and a catastrophic matrix singularity ( [G_f]^-1 \to \infty ). We demonstrate this geometric breakdown acts as the strict non-equilibrium trigger for a Gauge Symmetry Break (GSB). The system purges hidden tension from unobservable gauge redundancies, spontaneously crystallizing a new, mathematically independent horizontal coordinate axis. This non-parametric phase transition registers as a discrete +1.0 integer step-jump in observable Structural G-Entropy. By decoupling parameter charts and subjecting emergent axes to a Minimal Energy Path Criterion and a Causal Invariance Filter, we distinguish genuine discovery from malignant hallucinations. Ultimately, SMG provides a parameter-free, falsifiable dashboard to mathematically certify true intelligence, transforming AI for Science into an engine of autonomous paradigm shifts.
[LG-45] Life Cycle Assessment of Pre-training the Lucie 7B Open-Source Large Language Model on the Jean Zay Supercomputer
链接: https://arxiv.org/abs/2607.05408
作者: Marc Léobet,Pierre-François Lavallée,Jean-Pierre Lorré
类目: Computers and Society (cs.CY); Machine Learning (cs.LG)
*备注: 12 pages, 2 tables, 1 figure. OpenLLM-France Environmental Report v1, January 2026. CC BY-SA 4.0
Abstract:The environmental impact of training large language models (LLMs) is increasingly scrutinised, yet most published estimates focus on operational energy and disclose little about manufacturing (embodied) emissions, water consumption, or the underlying highperformance computing (HPC) infrastructure. We present a life cycle assessment (LCA) of the pre-training of Lucie 7B, an open-source multilingual Foundation Model developed by the OpenLLM-France consortium and trained on the NVIDIA H100 partition of the Jean Zay supercomputer operated by IDRIS (CNRS). The assessment is framed by the AFNOR SPEC 2314 “Frugal AI” reference and applies the Labos 1point5 methodology for greenhouse gas(GHG) accounting in computing. The study scope extends from data preparation to model validation, and integrates the full life cycle of the hardware infrastructure: manufacturing (including raw-material extraction), use (compute, temporary storage, system administration, cooling), and end-of-life. We report (i) an annual footprint of 417.5 tCO2eq for the Jean Zay H100 partition, split almost equally between manufacturing and operation; (ii) an effective intensity of 36.7 gCO2eq per H100 GPU-hour; (iii) a total training footprint of 21 tCO2eq for Lucie 7B (574 564 H100 GPU-hours), inclusive of amortised hardware manufacturing; (iv) on-site water consumption of approximately 76m3 for the training campaign and an annual Water Usage Effectiveness (WUE) of 0.07 L/kWh for IDRIS; (v) a heat-reuse factor (ERF) of 0.37 thanks to waste-heat recovery into the urban heating network. The study contributes one of the few publicly documented LCAs of an LLM training campaign that explicitly couples operational data with embodied emissions decomposed by subsystem (compute, storage, power chain, cooling), and discusses the implications for the design of frugal-by-construction AI systems in Europe.
[LG-46] Performance Optimization and Comparative Analysis of Generative AI Models on Advanced Accelerators
链接: https://arxiv.org/abs/2607.05400
作者: Amitash Nanda,Javier Hernandez Nicolau,Madhusudan Gujral,Mahidhar Tatineni,Amitava Majumdar,Debashis Sahoo
类目: Performance (cs.PF); Machine Learning (cs.LG)
*备注: 6 pages, 3 figures
Abstract:Generative AI models, such as Large Language Models (LLMs) and diffusion models, have demonstrated impressive performance across a wide range of tasks. Despite these advances, deployment remains challenging due to substantial memory requirements, extended inference latency, significant computational demands, and high hardware costs. These issues are further complicated when evaluating models across heterogeneous platforms, where differences in numerical formats, memory bandwidths, and software stacks interact with model architecture and workload characteristics in complex ways. To address these challenges, we present a systematic study focused on performance optimization and comparative analysis of several Generative AI models across diverse downstream tasks. This work introduces a novel mixed-precision post-training quantization evaluation, examines fine-tuning strategies, and assesses performance across modern high-performance computing (HPC) systems and advanced accelerators.
[LG-47] A Function-Space Dichotomy for Compositional Learning: Exponential Sub-Optimality of the Neural Tangent Kernel
链接: https://arxiv.org/abs/2607.06382
作者: Arkaprabha Ganguli,Emil Constantinescu
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:A persistent empirical observation is that trained neural networks outperform their neural tangent kernel (NTK) limit on tasks with compositional structure, yet a quantitative account of \textbfwhen and \textbfby how much has been lacking. Working on the unit circle, we give such an account through a dichotomy between two complexity measures of the target: its \textbfFourier complexity , which controls NTK kernel regression, and its \textbfarchitectural complexity , which controls learning over depth- L , width- w ReLU networks with the variation norm of the weights bounded by R . We first characterize the minimax rate of the architecture class \mathcalC_L,w,R , pinning it down up to a single factor of L : between \Omega(Lw^2R^2/n) and \tildeO(L^2w^2R^2/n) . We then show the NTK estimator sits \textbfexponentially above this floor whenever the two complexities decouple: for the depth- L iterated sawtooth, NTK regression needs \Omega(4^L) samples while the minimax floor is polynomial in L . Numerical experiments confirm the theoretical claims: on bandlimited smooth targets, the NTK is competitive or better, while on the hypercube sparse-parity model, a standard two-layer network beats the NTK by four to six orders of magnitude in test error. The gap is thus a function-space property, a mismatch between the kernel’s smoothness bias and the target’s compositional structure, rather than a generic kernel-versus-network phenomenon.
[LG-48] A Convex Approximation Framework for Neural Likelihood-Based Bayesian Inverse Problems
链接: https://arxiv.org/abs/2607.06252
作者: Fabian Schneider,Tapio Helin,Leila Taghizadeh
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Probability (math.PR); Methodology (stat.ME)
*备注: 22 pages, 3 figures
Abstract:Many problems in science and engineering are difficult to model accurately, either due to unknown physical mechanisms, poorly quantified measurement uncertainty, or prohibitive computational costs of high-fidelity simulations. These challenges limit the applicability of classical probabilistic inference methods such as Markov chain Monte Carlo, especially in high-dimensional Bayesian inverse problems. As data from scientific experiments become increasingly available, machine learning methods offer a flexible alternative to explicit parametric modelling. We study neural likelihood approximation, where the goal is to learn the likelihood function directly from data without explicit knowledge of the underlying data-generating process. A common approach trains likelihood surrogates by minimizing the Kullback-Leibler divergence between the true posterior and an approximate posterior, which is equivalent to minimizing the expected negative log-likelihood. This work improves the theoretical foundations of neural likelihood approximation by alleviating limitations of restrictive model classes: we show that, by working with un-normalized potentials and folding normalization into the training objective, the resulting learning problem is strictly convex. We show that empirical minimizers of the resulting data-driven objective converge to the true likelihood as the sample size grows. Numerical experiments for the neural likelihood approximation are conducted for a deblurring and a non-linear PDE based imaging problem.
[LG-49] Entanglement as a Structural Complexity Axis: A PAC-Bayesian View of Generalization in Quantum Policies and Value Functions
链接: https://arxiv.org/abs/2607.06230
作者: Jian Xu,Delu Zeng,John Paisley,Qibin Zhao
类目: Quantum Physics (quant-ph); Machine Learning (cs.LG)
*备注:
Abstract:Parameterized quantum circuits (PQCs) are increasingly used as policies and value functions in quantum reinforcement learning, yet it remains unclear when and why quantum policies generalize. We give a PAC-Bayesian account in which generalization is governed not by the raw number of circuit parameters, but by the effective dimension of the Fisher geometry induced by the circuit. This quantity is inflated by entanglement, making entangling connectivity an independent axis of this http URL controlled experiments that fix the number of trainable rotations and vary only entanglement, we find that circuits with larger Fisher effective dimension exhibit larger train-test gaps, while parameter count is a weak predictor. The resulting bound acts primarily as a ranking certificate: it correctly orders circuits with identical parameter count, which parameter-counting bounds cannot do. We validate this mechanism across supervised classification, quantum contextual bandits, and value-function generalization, where entangled circuits consistently generalize worse than non-entangled circuits of equal parameter count, with gaps shrinking as sample size this http URL strongest evidence comes from low-variance decision models, including single-observable classifiers, value heads, and one-step policies. In end-to-end multi-step policy learning, entanglement effects remain statistically significant but high return variance leaves the full ordering only partially resolved. Partial-correlation analysis shows that Fisher effective dimension screens off entangling pattern, and controls for training accuracy, readout, and optimizer rule out major optimization confounders. The effect also persists on an IBM Heron quantum processor under real noise. Overall, our results reframe quantum policy design around an entanglement–generalization trade-off rather than expressivity alone.
[LG-50] Separation Capacity of Scattering Networks on Low-Dimensional Datasets
链接: https://arxiv.org/abs/2607.06048
作者: Konstantin Häberle,Helmut Bölcskei
类目: Machine Learning (stat.ML); Information Theory (cs.IT); Machine Learning (cs.LG); Classical Analysis and ODEs (math.CA)
*备注: 19 pages
Abstract:We aim to identify scattering network architectures that maximize the separation capacity on data with low intrinsic dimension. The networks we consider employ a fixed monomial nonlinearity and no pooling, so that the only design variable is the frame generated by the network filters. For data modeled as rectifiable sets, we first characterize and bound the separation capacity of general feature extractors in terms of the geometry of the dataset. We then particularize to scattering networks and obtain two design criteria: (i) the filters should meet the data on sufficiently many frequencies, and (ii) the matrices coupling the frame to the geometry of the data should be well-conditioned.
[LG-51] On the convergence of graph Laplacians with a symmetric divergence
链接: https://arxiv.org/abs/2607.05892
作者: Liane Xu
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注: 51 pages, 10 figures
Abstract:When analyzing a manifold learning algorithm for data lying on a smooth, compact, connected Riemannian submanifold (\mathcalM, g) of \mathbbR^d , a key estimate for the geodesic distance d_g is that there exists K 0 such that 0 \leq d_g(p, q)^2 - |p-q|^2 \leq K d_g(p, q)^4 for all p, q \in \mathcalM . We observe that more generally, when \mathcalM is equipped with a smooth symmetric divergence D satisfying a non-degeneracy condition and g is given by g_p := \frac12\mathrmHess_p(D(p, \cdot)) for all p \in \mathcalM , there exists K 0 such that \left| D(p, q) - d_g(p, q)^2 \right| \leq K d_g(p, q)^4 for all p, q \in \mathcalM . We demonstrate that this is sufficient for the pointwise convergence of graph Laplacians constructed with D and discuss examples where D is given by the Sinkhorn divergence on a family of probability measures parametrized by a manifold.
[LG-52] On the Condition Number Upper Bound of the L-BFGS Inverse Hessian Approximation Matrix with a Two-Sided Geometric Envelope Safeguarding Mechanism
链接: https://arxiv.org/abs/2607.05836
作者: Don Li
类目: Optimization and Control (math.OC); Machine Learning (cs.LG); Numerical Analysis (math.NA)
*备注: 22 pages, 3 figures
Abstract:The limited-memory BFGS (L-BFGS) algorithm is a cornerstone of large-scale optimization due to its linear memory and computational costs. However, in ill-conditioned or non-convex landscapes, the implicit inverse Hessian approximation can suffer from an exploding condition number, leading to numerical instability and degraded convergence. To address this, we propose Two-Sided L-BFGS, a safeguarded variant that dynamically constrains the condition number of the inverse Hessian operator via a two-sided geometric envelope. Moreover, we show that Two-Sided L-BFGS preserves accumulated curvature information and maintains standard O(mn) memory and per-iteration time complexities. We prove that this geometric envelope yields a uniform bound on the condition number of every inverse Hessian approximation generated by the algorithm. By tracking the algebraic evolution of the extreme eigenvalues through m consecutive quasi-Newton updates starting from a scaled identity matrix, the resulting bound is expressed explicitly as a function of the memory depth, problem dimension, and envelope hyperparameters. Moreover, we show that Two-Sided L-BFGS preserves asymptotic global convergence in non-convex regimes under standard smoothness and strong Wolfe line-search assumptions, matching the theoretical guarantees of L-BFGS variants utilizing the Li-Fukushima cautious update rule. Numerical experiments on high-dimensional optimization problems demonstrate that the proposed method maintains well-conditioned inverse Hessian approximations and improves robustness and convergence behavior on ill-conditioned benchmarks.
[LG-53] Latency-Constrained Hardware-Aware Quantum Error Correction Co-Design with Adaptive Confidence-Gated Neural Decoding for the Rotated Surface Code
链接: https://arxiv.org/abs/2607.05814
作者: Sumit Chongder
类目: Quantum Physics (quant-ph); Emerging Technologies (cs.ET); Machine Learning (cs.LG)
*备注: 29 pages, 18 figures, 12 tables. Source code, trained models, and benchmark data: this https URL
Abstract:Real-time decoding is a major bottleneck in scaling quantum error correction (QEC) from noisy intermediate-scale quantum (NISQ) devices to fault-tolerant quantum computing. We present an adaptive confidence-gated decoding framework for the rotated surface code that treats decoding as a two-stage inference problem. A lightweight feed-forward neural network performs fast-path decoding for the majority of syndrome measurements, while only low-confidence predictions are escalated to a minimum-weight perfect matching (MWPM) refinement stage. We benchmark the framework on rotated surface codes with distances d \in \3,5,7,9,11\ under circuit-level depolarising noise using the Stim stabiliser simulator. The evaluation characterises logical accuracy, confidence-controlled accuracy-latency trade-offs, decoding throughput, per-shot latency, and decoding-graph resource scaling. Routing only 3.3%-6.2% of syndromes to the refinement stage improves logical accuracy from 99.21% for the neural-only baseline to 99.81% at a confidence threshold of 0.95 while incurring only a bounded increase in average decoding cost. Neural-decoder throughput saturates near 4.6 \times 10^5 samples s ^-1 at batch size 512 on commodity CPU hardware, indicating that the neural fast path is not the dominant throughput bottleneck beyond code distance d=7 . We release the complete benchmarking pipeline, trained models, raw benchmark data, and source code, and explicitly distinguish the experimentally validated contributions from the broader hardware-aware QEC co-design roadmap, including hardware-constrained code discovery, GPU-accelerated inference, and multi-noise optimisation, which remain directions for future work.
[LG-54] Boosting with List-Decodable Codes COLT2026
链接: https://arxiv.org/abs/2607.05791
作者: Addison Prairie,Li-Yang Tan
类目: Machine Learning (stat.ML); Computational Complexity (cs.CC); Data Structures and Algorithms (cs.DS); Machine Learning (cs.LG)
*备注: COLT 2026
Abstract:Boosting is a fundamental technique for generically improving the accuracy of learning algorithms (Schapire 1989). Existing boosting algorithms construct a strong learner using O(\log(\frac1\epsilon)/\gamma^2) calls to a \gamma -advantage weak learner, and this round complexity is known to be optimal for generic boosters that succeed on all concept classes (Freund 1995). We show that this lower bound can be circumvented for concept classes that satisfy a mild closure property. Specifically, we present a new boosting algorithm that, for any class \mathcalF closed under O(\log \frac1\gamma) -XOR, strong learns \mathcalF using O(\log \frac1\epsilon) calls to a \gamma -advantage weak learner and a single batch of \tildeO(\log(\frac1\epsilon)/\gamma^2) additional samples. Our algorithm arises from a new and simple connection between boosting and list-decodable codes. Viewing the target function as a message, we run the weak learner on its encoding and view the resulting weak hypothesis as a corrupted codeword. Feeding this corrupted codeword to a list decoder, we obtain a small list of candidate hypotheses, at least one of which is a strong hypothesis for the original function. Using additional samples, we identify and output this strong hypothesis. Comments: COLT 2026 Subjects: Machine Learning (stat.ML); Computational Complexity (cs.CC); Data Structures and Algorithms (cs.DS); Machine Learning (cs.LG) Cite as: arXiv:2607.05791 [stat.ML] (or arXiv:2607.05791v1 [stat.ML] for this version) https://doi.org/10.48550/arXiv.2607.05791 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-55] From Closed-Loop Optimization to Open Decision Making: Coupled Digital Twins for Predictive and Autonomous Microscopy
链接: https://arxiv.org/abs/2607.05758
作者: Yu Liu,Boris Slautin,Ian Mercer,Jon-Paul Maria,Sergei V. Kalinin
类目: Materials Science (cond-mat.mtrl-sci); Machine Learning (cs.LG)
*备注:
Abstract:Automated experimentation is moving from closed-loop optimization toward open decision-making, where human or AI planners must forecast the consequences of candidate actions before executing them. Such forecasts require a model of both sides of the experiment: how the sample is likely to respond and what the instrument is likely to detect. We therefore introduce a coupled digital-twin framework that separates these roles and then links them. In this framework, the sample twin encodes material state inferred from prior knowledge and measurements till the moment. The instrument twin captures signal formation, feedback dynamics, and operating constraints based on prior knowledge. When coupled, the two twins estimate expected outcomes, uncertainty, and risk for candidate microscope operations. For amplitude-modulation scanning probe microscopy, we realize this framework with a physics-informed encoder of force-distance curves, a deterministic scanner model of cantilever and feedback dynamics, and sparse learned residual corrections. The encoder first recovers scanner-driving descriptors with sub-nanometer accuracy. The calibrated scanner then reproduces typical traces within a few nanometers and identifies operating-point noise amplification as the main source of mismatch. Supplementary phase analysis localizes residual error to the phase channel, which clarifies where added physics is needed. Together, these results establish coupled sample and instrument twins as a practical foundation for predictive microscope operation and autonomous experimental planning.
[LG-56] Width-Robust Learnability in Mean-Field Bayesian Neural Networks
链接: https://arxiv.org/abs/2607.05735
作者: Dmitry Vaintrob,Kaarel Hänni
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注: 36 pages (with appendices), 2 figures
Abstract:Infinite-width limits are a standard way to reason about neural networks, but it is not automatic that the limiting learner has the same complexity-theoretic inductive bias as large finite networks. We study this question for Bayesian neural networks at the mean-field, or critical feature-learning, scaling. The central quantity is the \emphreduced entropy [ s_\infty(y,\varepsilon)=\limsup_N -\frac1N\log \pi_N^0(L\le \varepsilon), ] the intensive prior cost of representing a target function y to population mean-squared error \varepsilon . Our main result is a width-robust learnability theorem. At fixed depth, a family of Boolean-cube targets is learnable from polynomially many samples at infinite width if and only if it is learnable at polynomial width, if and only if its reduced entropy is polynomially bounded. Equivalently, up to polynomial slack in accuracy, the Bayesian mean-field learner generalizes exactly on the targets that can be represented by polynomial-size networks. The forward direction is proved by a form of subsampling: from the infinitely many hidden neurons in the mean-field solution, one can select polynomially many representatives and still preserve the learned function on every input simultaneously. At the critical scaling this subsampling has both an active'' component, which keeps the data-dependent low-dimensional statistics, and a lazy’’ component, which resamples the entropy-dominated directions from the prior. Thus the infinite-width mean-field limit gives a clean analytic description of learning without introducing spurious width-dependent generalization power. Comments: 36 pages (with appendices), 2 figures Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG) MSC classes: 68Q32 (Primary) 68T07, 60K35, 62F15 (Secondary) Cite as: arXiv:2607.05735 [stat.ML] (or arXiv:2607.05735v1 [stat.ML] for this version) https://doi.org/10.48550/arXiv.2607.05735 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Dmitry Vaintrob [view email] [v1] Tue, 7 Jul 2026 01:41:42 UTC (45 KB)
[LG-57] Beyond Heuristic Tuning: Power-Calibrated LLM Watermarking ICML2026
链接: https://arxiv.org/abs/2607.05694
作者: Xiaopu Wang,Zelin He,Chengyuan Liu,Runze Li
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注: Accepted ICML 2026
Abstract:Logit-based watermarking is a widely used mechanism for identifying LLM generated content, yet its effectiveness is governed by a fundamental trade-off between detectability and semantic distortion. Existing analyses provide limited guidance for principled hyperparameter selection, leaving practical deployments reliant on heuristic tuning. In this work, we develop a power-calibrated statistical framework that establishes explicit quantitative relationships between watermark hyperparameters, detection power, and distortion. This characterization transforms watermark design into a guided optimization problem. Building on these results, we derive practical parameter selection procedures that achieve optimal tradeoffs under constraints. Extensive experiments across multiple language models and datasets validate the theory and demonstrate that the proposed framework consistently identifies Pareto-optimal points.
[LG-58] Integrating GNSS-Derived Zenith Wet Delay into a Weather Foundation Model Improves Precipitation Forecasting
链接: https://arxiv.org/abs/2607.05658
作者: Leonardo Trentini,Fanny Lehmann,Laura Crocetti,Benedikt Soja
类目: Atmospheric and Oceanic Physics (physics.ao-ph); Machine Learning (cs.LG); Geophysics (physics.geo-ph)
*备注: Submitted to Geophysical Research Letters
Abstract:Global Navigation Satellite Systems (GNSS), best known for positioning, also serve weather science, as atmospheric water vapour delays their signals. This delay, the Zenith Wet Delay (ZWD), is a direct, all-weather measure of column moisture. Although assimilated into numerical weather prediction for decades, ZWD is not yet used by leading machine learning weather models (MLWM), despite addressing a known deficiency: the underestimation of severe precipitation. Here we present the first integration of GNSS-derived ZWD into Aurora, a state-of-the-art weather foundation model. Our extended Aurora learns ZWD with skill comparable to its pretrained variables. More importantly, including ZWD systematically improves forecasts when fine-tuning for six-hour accumulated precipitation. Gains grow with severity, reaching an 8.8% increase in Equitable Threat Score at the 99th percentile, while the precipitation power spectrum becomes more realistic at synoptic and planetary scales. Direct GNSS observations therefore encode information that MLWM can exploit for high-impact precipitation.
[LG-59] DBNN: Neural Spike Classification Using a Deep Binarized Neural Network
链接: https://arxiv.org/abs/2607.05590
作者: Binyi Ren,Luca M. Meyer,Majid Zamani
类目: ignal Processing (eess.SP); Machine Learning (cs.LG)
*备注: Paper under review
Abstract:Implantable brain-computer interfaces require on-node spike sorting to reduce telemetry bandwidth and power while maintaining reliable neural decoding. This paper presents a hardware-oriented deep binarized neural network (DBNN) spike-sorting system with two binarized hidden layers with 256 neurons and a fixed-point output layer to enable multiplier-free inference dominated by sign-controlled accumulation and bit-wise logic. The proposed classifier operates on compact 16-sample spike waveforms to reduce the implementation cost (16-256-256-3) and achieves a median classification accuracy of 98.7% on both synthetic and in-vivo datasets. An FPGA prototype on a Cyclone V device operates at 50 MHz and requires 528 cycles per spike, corresponding to a 0.01 ms compute latency, while consuming 828 ALMs and 1023 registers with zero DSP blocks. For ASIC feasibility, the DBNN is implemented using FreePDK45-based flow; synthesis in Synopsys Design Compiler indicates an estimated silicon area of 0.014 mm2 and an operating power of 122 nW at 20 kHz under a 1.1 V supply. These results demonstrate that the proposed DBNN spike sorter offers a favorable trade-off between accuracy and implementation cost, supporting low-power, implantable neural interfaces. Overall, the proposed DBNN spike sorter achieves high accuracy (98.7%) with extremely low hardware cost (0.014 mm2, 122 nW at 20 kHz) and multiplier-free operation, making it suitable for low-power, implantable neural interfaces. This paper introduces the first DBNN designed for real-time neural spike sorting, striking an excellent balance between input data size and network complexity.
[LG-60] Deep Neural Variation Spaces: A Unifying Perspective on Depth and Complexity
链接: https://arxiv.org/abs/2607.05546
作者: Julia Nakhleh,Robert D. Nowak
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Functional Analysis (math.FA)
*备注:
Abstract:We develop a unified function space theory of deep fully connected neural networks. Functions in our spaces are defined recursively as \ell^1 -bounded linear combinations of activated functions from preceding layers, with a dictionary of affine functions at the first layer. Unlike existing theories that are largely specialized to homogeneous activations such as the ReLU, our framework provides a meaningful notion of functional complexity for deep networks with a broad range of homogeneous and non-homogeneous activation functions commonly used in practice. This simple construction unites several seemingly disparate ideas from the literature, including norm-based complexity bounds and variational characterizations of depth, and facilitates novel analyses of what kinds of functions deep norm-constrained networks can represent. To this end, we prove a novel representer theorem for our spaces and establish novel function-space complexity bounds showing that the associated function classes remain qualitatively small at arbitrary depth. In the univariate ReLU case, we prove a “depth saturation” result: depth in this setting yields only a small constant rescaling of the function class, with no added functional diversity. As a consequence, we show that deep norm-controlled ReLU functions in any dimension cannot exhibit high frequencies along any direction. This finding reveals that some commonly cited expressivity benefits of depth disappear once network complexity is controlled by an appropriate function space norm, rather than parameter count or other representational costs that permit compounded rescaling across layers. Overall, our results illustrate how a function space perspective yields new structural insights into the relationship between depth and complexity.
[LG-61] Higher-Order Certified Robustness for Regression ICML2026
链接: https://arxiv.org/abs/2607.05536
作者: Jie Zhang,Natalie Frank
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注: Accepted at the 43rd International Conference on Machine Learning (ICML 2026)
Abstract:Randomized smoothing has emerged as a scalable technique for certifying the adversarial robustness of classifiers. However, its application to regression remains under-explored and faces unique challenges. Existing regression certificates rely on probabilistic acceptance regions and fail to exploit the local geometry of the function. In this work, we present a novel framework for certified robust regression that addresses these limitations. We derive a prediction-centered certificate that guarantees the stability of the smoothed model’s prediction and ensures practical computability at test time. We investigate several alternatives for constructing these certificates by explicitly incorporating means, variances, and gradients. In particular, we demonstrate on the MNIST rotation task that utilizing gradient information yields significantly tighter robustness certificates compared to the current state-of-the-art, alpha-smoothing.
[LG-62] Black Hole Black Boxes: Numerical Black Hole Metrics via AInstein Neural Networks
链接: https://arxiv.org/abs/2607.05489
作者: Tancredi Schettini Gherardini,Edward Hirst,Alexander George Stapleton
类目: General Relativity and Quantum Cosmology (gr-qc); Machine Learning (cs.LG); High Energy Physics - Theory (hep-th)
*备注: 33 pages; 7 tables; 9 figures
Abstract:The AInstein architecture introduced an unsupervised neural method for solving the Riemannian Einstein equations on arbitrary manifolds. This Physics Informed Neural Network approach (PINN) is extended here to Lorentzian signature, validated by recovering the maximally extended Schwarzschild geometry, and tested as novel search method for arbitrary black hole solutions. The topology is built into the architecture by treating S^2 globally through its standard embedding, such that the network learns an ambient metric on the manifold \mathbbR^2 \times \mathbbR^3 , where Penrose coordinates are chosen for \mathbbR^2 and the metric on S^2 is obtained by pullback. The architecture is first trained with the objective of recovering the Schwarzschild metric via losses encoding the vacuum Einstein equation, a quadratic Weyl scalar constraint, and the SO(3) symmetry of the resultant metric; directly motivated by the Birkhoff–Jebsen theorem. Following this, the objective is generalised to use the Petrov speciality index, a horizon curvature anchor, and a trapped-surface constraint, to allow search for algebraically general Petrov type I solutions, finding potentially novel general-type Lorentzian Einstein metrics with a genuinely trapped interior.
[LG-63] SHARC: SHAP-Based Interpretability in Machine Learning Risk Models for Regulatory Capital under ICAAP and CCAR
链接: https://arxiv.org/abs/2607.05484
作者: Ujjwala Vadrevu
类目: Risk Management (q-fin.RM); Machine Learning (cs.LG)
*备注: Builds on: Paper 1: arXiv:2605.17275 , Paper 2: arXiv:2606.07575
Abstract:The adoption of non-parametric machine learning models for regulatory capital estimation introduces a fundamental governance challenge: the inability to explain model outputs in a manner auditable by supervisory bodies. This ‘black box’ problem remains a major barrier to the adoption of Gaussian Process Regression (GPR) and related ML architectures in ICAAP and CCAR workflows despite their predictive advantages over traditional parametric approaches. This paper addresses this barrier through SHARC (SHAP for Regulatory Capital), an explainability framework for the Hybrid GPR-HS architecture and its stress-testing extension. SHapley Additive exPlanations (SHAP), derived from cooperative game theory and satisfying the properties of Local Accuracy, Missingness, Consistency, and Efficiency, are applied to Stressed Value-at-Risk (SVaR) outputs under three macro scenarios: West Asia War, Climate Risk, and AI Bubble/Regulatory Burden. SHARC decomposes SVaR into baseline, mean-driven, and volatility-driven components, enabling transparent linkage between scenario design and capital outcomes. Two findings emerge. First, SHARC consistently links non-linear SVaR outputs to underlying scenario inputs, confirming framework fidelity and providing auditable traceability of capital drivers. Second, under stress conditions, the mean return component (directional loss magnitude) dominates the variance component (volatility baseline) in determining capital levels, with implications for capital limit-setting, position management, and hedging strategy. The results establish SHARC as a regulator-aligned explainability layer that makes the Hybrid GPR-HS framework fully auditable and consistent with FRTB, ICAAP Pillar 2, and CCAR transparency requirements. Comments: Builds on: Paper 1: arXiv:2605.17275, Paper 2: arXiv:2606.07575 Subjects: Risk Management (q-fin.RM); Machine Learning (cs.LG) Cite as: arXiv:2607.05484 [q-fin.RM] (or arXiv:2607.05484v1 [q-fin.RM] for this version) https://doi.org/10.48550/arXiv.2607.05484 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-64] Broken Ergodicity and the Violation of the Fluctuation-Dissipation Theorem Lead to Generalization Beyond Overfitting in Machine Learning
链接: https://arxiv.org/abs/2607.04135
作者: Chan Li,Nigel Goldenfeld
类目: Disordered Systems and Neural Networks (cond-mat.dis-nn); Statistical Mechanics (cond-mat.stat-mech); Machine Learning (cs.LG)
*备注: 62 pages, 8 figures; includes Supplemental Material
Abstract:The remarkable ability of modern neural networks to generalize improves with increasing network capacity, even when the number of model parameters or effective degrees of freedom exceeds the number of training data points. This phenomenon is all the more surprising given that generalization error diverges when the number of model parameters approaches a critical value from below. Here we use dynamical mean field theory to show that this so-called “double descent” behavior is the outcome of a phase transition in the stochastic field theory describing the training process. We calculate the critical exponents and scaling function of the double descent phase transition, and show that it is marked by a breakdown of the fluctuation-dissipation theorem associated with broken ergodicity. The corresponding response function has the same functional form as the simple London model of the superconducting transition, with the rigidity of the wave function corresponding to the neural network’s ability to generalize accurately.
附件下载
