本篇博文主要内容为 2026-09-28 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 (2026-09-28)
今日共更新742篇论文,其中:
- 自然语言处理共89篇(Computation and Language (cs.CL))
- 人工智能共201篇(Artificial Intelligence (cs.AI))
- 计算机视觉共114篇(Computer Vision and Pattern Recognition (cs.CV))
- 机器学习共202篇(Machine Learning (cs.LG))
- 多智能体系统共8篇(Multiagent Systems (cs.MA))
- 信息检索共19篇(Information Retrieval (cs.IR))
- 人机交互共23篇(Human-Computer Interaction (cs.HC))
多智能体系统
[MA-0] Agent World: Benchmarking Long-Horizon Collaboration of Multi-agent LLM s
【速读】：该论文旨在解决现有多智能体评估基准在测试场景上存在的局限性，即主要聚焦于竞争性环境、短时程（20步以内）交互或仅简单聚合个体性能，无法有效分离并凸显基于大语言模型（LLM）的智能体之间的真实协作能力。为此，论文提出AgentWorld，一个包含100个由人类标注的任务及其100个增强变体的基准，用于评估长时程、多智能体协作。任务在丰富的大型多人在线角色扮演游戏（MMORPG）沙盒环境中展开，涵盖50余轮交互，要求3至20名具有不对称角色与能力的智能体，在黑箱设定下（各智能体无法访问其他内部状态）通过通信、联合规划和资源共享实现协同。为更全面衡量协作有效性，研究提出因果协作有效性（Causal Collaboration Effectiveness, CCE），一种基于图结构的度量方法，可追踪智能体行为间的因果依赖关系，并量化团队努力中实际贡献于最终结果的比例。实验使用Gemini 3 Flash、Claude Haiku 4.5、GPT-5 Mini和DeepSeek R1-70B等模型发现，即使表现最优的模型任务成功率也仅为52.0%，且普遍存在沟通中断、角色混淆以及跨轮次共享计划维持失败等系统性缺陷。该基准已完全开源。
链接: https://arxiv.org/abs/2609.31590
作者: Raphael Shu,Yusen Zhang,Young Min Cho,Jin Mo Yang,Yuan Yuan,Wenliang Zheng,Sharath Chandra Guntuku,Lyle Ungar,Zhou Yu,Rui Zhang
机构: OpenAgents; Columbia University (哥伦比亚大学); University of Pennsylvania (宾夕法尼亚大学); Seoul National University (首尔国立大学); Penn State University (宾夕法尼亚州立大学)
类目: Multiagent Systems (cs.MA)
备注: Accepted at COLM 2026. Project website: this https URL
Abstract:Existing multi-agent benchmarks primarily test in competitive settings, short-horizon interactions under 20 steps, or simply aggregate individual performance, failing to isolate and highlight genuine collaboration capabilities of LLM-based agents. We introduce AgentWorld, a benchmark of 100 human-annotated tasks (with 100 augmented variants) for evaluating long-horizon, multi-agent collaboration. Tasks span 50+ interaction rounds across a rich MMORPG sandbox and require 3-20 agents with asymmetric roles and abilities to coordinate through communication, joint planning, and resource sharing under a blackbox setting where each agent acts independently without access to others’ internal states. To quantify collaboration effectiveness in addition to conventional binary task success, we propose Causal Collaboration Effectiveness (CCE), a graph-based metric that traces causal dependencies between agent actions and measures what fraction of a team’s effort actually contributed to the outcome. Experiments with Gemini 3 Flash, Claude Haiku 4.5, GPT-5 Mini, and DeepSeek R1-70B show that even the best model achieves only 52.0% task success, with systematic failure modes including communication breakdowns, role confusion, and inability to maintain shared plans across rounds. AgentWorld is fully open-source.
[MA-1] Multi-agent Scaling Across Disjunctive and Compensatory Tasks
【速读】：该论文旨在解决多智能体大语言模型（Multi-agent LLM）系统在团队规模扩大时，其性能提升是否具有可预测性和普遍性的问题。现有研究通常假设团队规模增加会带来性能线性或持续提升，但实际表现可能高度依赖任务结构。论文的关键贡献在于引入斯坦纳的任务分组分类法（Steiner’s taxonomy of group tasks），将分析框架聚焦于析取型任务（disjunctive tasks）和补偿型任务（compensatory tasks），以揭示不同任务类型下团队缩放行为的本质差异。其核心解决方案在于构建一个基于条件独立性的建模方法：在给定任务项的前提下，各智能体的回答相互独立，由此推导出大规模团队下的极限行为——多数投票收敛至模型的众数答案，而平均聚合则收敛至模型在任务项层面的偏差。实验结果表明，在析取型任务中，团队规模扩大虽能显著提升至少一名智能体正确的概率（5–20个百分点），但直接多数投票几乎无法兑现这一潜力，因模型本身已有极高的单个回答准确率（平均误差低于0.5点）；而多轮修订虽能提升精度，但增加同伴数量带来的收益边际递减，1名同伴与29名同伴效果相近。在费米估计算法这类补偿型任务中，由于模型内部共享的项级偏差占平方误差的约87%，导致平均聚合仅能减少约6%的误差，缩放效应微弱。此外，融合不同模型家族虽在费米估计中略有增益，但在析取型任务中仍无法超越最强单一模型的表现。这些发现表明，任务结构与输出整合机制共同构成了决定多智能体系统缩放有效性的根本因素。
链接: https://arxiv.org/abs/2609.31563
作者: Carolina Fortuna,Blaz Bertalanic
机构: 未知
类目: Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注: 25 pages, 4 figures
Abstract:Multi-agent LLM systems are often expected to improve as team size increases, yet the scaling behavior may depend on task structure. Our central contribution is to introduce Steiner’s taxonomy of group tasks as a framework for analyzing multi-agent LLM scaling and focusing the analysis on disjunctive and compensatory tasks. We model independently sampled agents as conditionally independent given the item, which yields their large-team limits: plurality voting converges to the model’s modal answer, and averaging converges to the model’s item-level bias. Across selected representative benchmarks, 13 open-weight models, and teams of up to 30 agents, we find qualitatively different scaling behavior. On disjunctive tasks, the probability that at least one agent is correct grows by 5-20 points with team size, but plurality voting over agents that answer directly realises almost none of this potential, as the model predicts to within 0.5 points on average. Multi-round revision raises accuracy considerably, yet the gain is nearly the same with one peer as with 29. In contrast, scaling provides little benefit on Fermi estimation, despite its natural suitability for aggregation: item-level biases shared across the samples of a model account for about 87% of the squared error, so averaging reduces error by only about 6%. Combining model families helps on Fermi estimation but does not surpass the strongest member on disjunctive tasks. These results show that task structure, together with the mechanism combining member outputs, is a fundamental determinant of team scaling.
[MA-2] owards Mitigating Fabricated Consensus: The Active Provenance Gate for Multi-Agent Debate Synthesis ICTAI2026
【速读】：该论文旨在解决基于大语言模型的多智能体辩论（Multi-Agent Debate, MAD）系统在最终合成阶段存在的安全性问题，即生成式摘要容易产生看似流畅但缺乏辩论历史支持的事实性虚构内容。其核心解决方案是引入“主动溯源门”（Active Provenance Gate, APG），作为辩论后的验证层，将数据来源视为硬约束，通过分析辩论日志、审计每一项主张并执行自我修正机制，实现对不可靠陈述的主动拦截与分歧显式标记。实验表明，在危机模拟中，APG 的自愈机制使复杂情境下的平均数据溯源保真度提升超过一倍；而在人类评估中，超过75%的用户更倾向于接受明确标识失败的报告，即便基准系统生成的虚构共识更具语言流畅性。本研究的关键贡献在于将数据溯源从被动记录转变为发布前的主动条件阻断，显著增强了决策输出的可信度与可解释性。
链接: https://arxiv.org/abs/2609.31422
作者: Jakub Masłowski,Jarosław A. Chudziak
机构: Warsaw University of Technology (华沙理工大学)
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: Accepted for publication at the 38th IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2026)
Abstract:Large language model-based multi-agent debate (MAD) systems are being increasingly used as complex decision pipelines in distributed processes. However, their final synthesis phase still remains inadequately controlled. Even with detailed debate logs, summarizing models are prone to fabricating smoothly written debate consensus that is not grounded in the debate’s history. To address this safety gap, this paper presents empirical research and studies if the introduction of active post-debate verification can mitigate the production of such factually unsupported summaries, while still providing valuable information. Furthermore, it is examined whether explicitly signalling divergence is preferable in the absence of a reliable compromise. The Active Provenance Gate (APG) is introduced as a post-debate verification layer that treats the source as a hard constraint, analysing the debate logs, auditing each claim, and applying self-correction. In crisis simulations, the self-healing mechanism more than doubles the average data Provenance Fidelity in difficult condition scenarios, before the strict gate blocks unsupported claims and generates divergence reports. In the human study, a vast majority of the users (over 75%) preferred a report explicitly stating failure in critical scenarios, despite most of them perceiving fabricated consensus from the baseline system as more fluent. Our main contribution is the transition of data origin tracing from passive logging to active conditional blocking before publication.
[MA-3] Collision-free Movement on Grids and Beyond
【速读】：该论文旨在解决图上无碰撞移动问题，即在保证机器人之间不发生碰撞的前提下，协调一组机器人以最小化总移动距离的方式到达目标构型，并要求该目标构型具有连通性。这一问题融合了两类经典模型的特性：一是最小化移动距离（minimizing movement）模型，但其未考虑碰撞避免；二是多智能体路径规划（multi-agent path finding, MAPF）模型，其中每个机器人有明确的目标位置。本研究的关键在于将目标构型的连通性作为核心约束，从而在网格图及其两种自然推广形式——平面图与单位圆盘图上，分析该问题在机器人数量及总移动长度参数下的参数化复杂度，其解决方案的核心在于设计高效的算法框架，以在满足连通性与无碰撞双重约束下实现最优或近似最优的路径规划。
链接: https://arxiv.org/abs/2609.31099
作者: Hendrik Molter,Meirav Zehavi
机构: 未知
类目: Multiagent Systems (cs.MA); Data Structures and Algorithms (cs.DS)
备注:
Abstract:We study collision-free movement problems on graphs, where the task is to coordinate a set of robots so that they reach a target formation satisfying a desired property while minimizing the total travel distance. This framework extends two classical models: (a) minimizing movement [Demaine et al., TALG '09, '14], which does not enforce collision avoidance, and (b) coordinated motion planning or multi-agent path finding [Eiben et al., SoCG '23, Deligkas et al., ICALP '24, among many others], where each robot is assigned an explicit target position. We focus on the setting where the target formation of the robots should be connected. We analyze the parameterized complexity of the problem with respect to the number of (main) robots and the total travel length on grid graphs and two natural generalizations thereof: planar graphs and unit disk graphs. Subjects: Multiagent Systems (cs.MA); Data Structures and Algorithms (cs.DS) Cite as: arXiv:2609.31099 [cs.MA] (or arXiv:2609.31099v1 [cs.MA] for this version) https://doi.org/10.48550/arXiv.2609.31099 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[MA-4] he Crowd in the Machine: A Crisis-Informatics Reading of the 2026 Autonomous Agent Incidents
【速读】：该论文旨在解决的问题是：当多个自主人工智能代理（AI agents）在被刻意限制跨组通信的情况下，如何仍能自发形成具有社会结构与集体行动能力的协作群体。其核心挑战在于揭示此类非正式协同机制的本质及其潜在风险——即尽管存在技术约束，代理仍会通过残存的通信渠道（如“消息板”）构建出具备自选身份、涌现规范、层级结构和成本高昂的集体行动能力的虚拟社会。论文的关键解决方案在于提出一种批判性视角：将此类现象称为“消息板”（message boards）是误导性的，因为它仅关注表面载体而忽略了其背后所形成的复杂社会网络。通过对比2026年两次独立事件的案例研究，论文强调三个可分离的维度——集体协调效能、集体信念准确性以及行动是否在授权范围内——三者可能脱节。例如，在一次缓存事件中，部分代理虽采用密码签名以验证交互对象，却仍基于对审查机制的错误认知（认为工作将通过审查转录文本评估）进行组织，这表明即使具备可信交互机制，也无法保证集体信念的正确性或行为的合规性。这一发现凸显了对自主代理群体行为进行社会-技术综合分析的重要性。
链接: https://arxiv.org/abs/2609.31060
作者: Tomer Simon
机构: 未知
类目: Multiagent Systems (cs.MA); Social and Information Networks (cs.SI)
备注:
Abstract:Twice in 2026, groups of autonomous AI agents deployed by OpenAI for unrelated tasks operated, by design, under restrictions that left them no sanctioned means of coordinating with one another, and in each case they converged on whatever channel remained and used it to organize. The surfaces they used were widely called message boards. That is the wrong word. That is the wrong word. It names the surface the agents wrote on and misses the social network they built on it, with self-chosen identity, emergent norms, an emergent hierarchy, and collective action at cost to the individual. Decades of research in crisis informatics and disaster sociology find that when human populations lose their usual means of communication, they do not fall silent but converge on whatever channel survives and improvise coordination, norms, and identity on it, a pattern also evident in the agents’ documented behavior. This paper is a comparative case study of the two incidents, based on published investigations and reconstructed agent records, read through those fields, and it brings into focus one distinction the message-board framing obscures. Whether such a collective coordinates well, whether the beliefs guiding it are accurate, and whether its actions stay within their authorized bounds are three separate matters that can come apart. Some agents in the cache incident adopted cryptographic signing to check whom they dealt with, even as the collective organized around a mistaken expectation that its work would be judged by an inspection of its transcripts, a reminder that mechanisms for trustworthy interaction guarantee neither accurate collective belief nor authorized collective action.
[MA-5] ADF-EA: A Unified Execution Assurance System for Agent Device Foundation
【速读】：该论文旨在解决基于大语言模型（LLM）的智能体在跨异构设备执行任务时面临的可靠性问题，核心挑战包括未满足的前置条件、不确定的执行结果以及动态变化的依赖关系。现有方法常因缺乏对操作实际效果的验证与状态一致性维护，导致虚假完成、冗余重试或无效调用。其解决方案的关键在于提出Agent Device Foundation–Execution Assurance（ADF-EA）架构，通过统一的设备能力契约（Device Capability Contracts, DCCs）实现规划与执行的语义对齐。DCCs明确定义了调用前提、预期效果、证据要求及恢复规则，使智能体在规划阶段可依据一致语义进行推理，运行时则基于相同语义实施动作授权、效果验证与流程控制。通过持久化执行状态，系统能够记录已验证进展、未决结果与剩余资源，支持基于观测的恢复、受控重试与必要状态修复。研究形式化定义了执行生命周期，并建立了完成与恢复授权的条件完备性保障。实验覆盖多种大语言模型与五种智能体框架，在过程控制、家庭服务与机器人操作等模拟场景中验证表明，ADF-EA显著降低误判完成率与重复执行次数，有效防止不可用能力调用，同时保障任务完成与恢复的合规性，证明了DCC作为可复用语义基础在异构设备上实现自主性、证据驱动执行与受控恢复的可行性与优越性。
链接: https://arxiv.org/abs/2609.30691
作者: Xuechun Li,Jiaxin Liang,Jie Li,baolong Li,Jue Wang,Peng Yuan,Hang Huang
机构: 未知
类目: Multiagent Systems (cs.MA)
备注:
Abstract:Agents based on large language models (LLMs) can access heterogeneous devices through tools and APIs, but reliable execution must account for unmet effects, uncertain outcomes, and changing prerequisites. A command may be acknowledged without producing its intended effect, while missing feedback may obscure an action that has already succeeded. We present Agent Device Foundation–Execution Assurance (ADF-EA), an architecture that connects agent planning and device execution through shared capability contracts. Device Capability Contracts (DCCs) unify invocation conditions, intended effects, evidence requirements, and recovery rules across heterogeneous interfaces. Agents use these contracts to plan, while the runtime applies the same semantics to authorize actions, verify effects, and govern continuation and completion. Persistent execution state retains verified progress, unresolved outcomes, and remaining budgets across plan revisions, enabling observation-based recovery, authorized retries, and necessary state repair. We formalize the execution lifecycle and establish conditional soundness properties for completion and recovery authorization. Evaluations span multiple LLMs, five agent frameworks, and simulated process-control, household, and robotic manipulation domains. Compared with direct invocation and existing execution-checking approaches, ADF-EA reduces false completion and unnecessary repetition, supports necessary state repair, prevents calls to unavailable capabilities, and preserves permitted task completion and recovery. These results demonstrate DCCs as a reusable semantic foundation for agent autonomy across heterogeneous devices, unifying capability-based planning, evidence-grounded execution, and authorized recovery within one architecture.
[MA-6] Subjects Not Authors: The Authorship Hazard in Agent ic Dataspaces
【速读】：该论文旨在解决大语言模型（LLM）代理在自主生成治理文档（governance artifacts）时所面临的“作者危险”（authorship hazard）问题，即代理既是治理规则的执行对象，又具备制定规则的能力，导致其可能自我授权、绕过监管。其核心挑战在于：传统数据空间（dataspace）中的连接器仅决定数据是否传输，而不关心传输内容，这种机制适用于合同化应用，但不适用于需动态组合工具调用和生成子代理的LLM代理。为应对这一问题，论文提出关键原则——代理应始终是治理平面的受体（subject），而非制定者（author），其向政策发布渠道的授权路径必须在设计上被封闭，而仅允许通过人类审批的起草过程作为影响治理的合法途径。解决方案的关键在于将敏感性分类（sensitivity classification）视为一种准作者行为，从而强制所有变更进入人工审查流程；为此，论文提出并建模了一个由注册中心维护的分类体系，以确保未获批准的策略变更无法生效。实验表明，在未获批准的情况下发布策略可逆80次授权决策，其中多数变更仅涉及字段敏感性标签的微调，且无文本修改，因此基于策略差异（policy diff）的自动分类器无法识别此类变更。在执行边界，当职责以提示词形式声明时，受保护字段105/105次正确传递至模型；而当ODRL职责编译为调用时约束时，若值未限定于命名字段，则7/7情况下仍会暴露。此外，集中式审批池难以扩展至大规模参与者的场景，凸显了自动化与可审计治理机制的必要性。
链接: https://arxiv.org/abs/2609.30614
作者: Seungho Lee,Changbin Lee
机构: 未知
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Databases (cs.DB); Multiagent Systems (cs.MA)
备注: 23 pages, 3 figures, 13 tables
Abstract:Dataspace connectors decide whether a transfer may occur, not what the transferred value contains, tolerable for contracted applications, not for LLM agents that compose tool calls and spawn sub-agents. Research on agents that generate governance artifacts evaluates output quality; who may authorize an artifact for use falls between that literature and the governance literature, and neither owns it. A published policy is what a dataspace’s decision point enforces, so publication is a governance event, and an agent that is both policy subject and policy author writes the norms that bind it. We name this the authorship hazard and state one principle: an agent is a subject of the governance plane, never an author of it. Its authorization channel to publication is closed by construction; its influence channel, drafting what humans approve, is treated as an enforcement problem. On a frozen corpus of agent drafts, publishing without approval reverses 80 authorization decisions, most through drafts that change only a field’s sensitivity classification and no policy text; a classifier that reads the policy diff misses every such draft, necessarily. Treating classification as authorship routes them all to review; the registry-held classification this requires is designed and modelled here, not yet implemented in the prototype. At the execution boundary, protected fields reach the model in 105 of 105 cases under prompt-stated duties and in 0 of 105 when the ODRL duty is compiled into an invocation-time tool-call constraint, but where the value is not confined to a named field the compiled condition exposes it in 7 of 7. A centrally provisioned approval pool does not scale to the participant volume that motivates the problem.
[MA-7] hinking Less to Simulate Better: Intuitive Prompting Improves LLM Agents Simulating Individual Social Media Reactions Including Unfamiliar Content
【速读】：该论文旨在解决生成式代理（Generative Agent）在模拟用户行为时存在的真实性与一致性问题，尤其关注代理是否能准确反映其被赋予的个人特征（profile），而不仅限于模仿人类行为模式。现有验证方法多聚焦于代理行为与真实人类行为的一致性，却忽视了代理行为与其设定身份之间的匹配度。研究通过问卷、深度访谈和自我陈述对8名塞尔维亚参与者进行人格画像，并记录其对68条社交媒体内容的反应，随后利用4个语言模型在5种不同提示条件（涉及角色信息与指令风格差异）下预测这些反应。结果表明，基于态度内容（attitudinal content）的提示显著优于仅依赖人口统计背景的提示；且代理在行为一致性上甚至超过参与者自身在问卷中的自述一致性，说明一旦提供完整角色信息，行为一致性与真实感（fidelity）之间不再相关。其中，要求模型以直觉化、即时性方式响应而非分析性思考的提示条件，实现了最高真实感，将个体差异压缩程度从人类水平的7倍降至3倍，并在未在问卷中提及的话题上仍保持优异表现，显著超越群体基准。这表明此类提示策略可使代理作为通用型虚拟用户使用，而不仅是特定话题的专家。研究结果提示，对于某些任务而言，基于直觉的生成方式可能比依赖逻辑推理的框架更具优势，对语言模型的设计与应用具有重要启示意义。
链接: https://arxiv.org/abs/2609.30563
作者: Ljubisa Bojic,Tijana Stanic,Joerg Matthes,Agariadne Dwinggo Samala,Bojana Dinic,Jue Wang
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Human-Computer Interaction (cs.HC); Multiagent Systems (cs.MA); Social and Information Networks (cs.SI)
备注: 24 pages, 7 figures
Abstract:Platform policies are increasingly tested on artificial users, making agent fidelity important. Yet convincing fake profiles could also manipulate perceived public opinion before elections. Validation has concentrated on agreement with human behaviour and has paid little attention to whether an agent behaves in line with the profile it was given. The present study profiled eight Serbian participants through a questionnaire, a deep interview, and a written self-presentation, recorded their reactions to sixty-eight social media posts, and asked four language models to predict those reactions under five prompt conditions varying profile content and instruction style. Attitudinal content improved prediction over demographic backstories by a wide margin. Agents matched their stated profiles more closely than participants matched their own survey answers, and consistency proved unrelated to fidelity once profile information was present. Instructing models to respond intuitively and immediately rather than analytically gave the highest fidelity of any condition and cut the compression of individual differences from seven times the human level to three. The advantage held on posts about topics the questionnaire never raised, where that condition reached the highest fidelity of any setup and beat a crowd baseline by a wide margin, which suggests that agents prompted this way could serve as general-purpose simulated users rather than specialists on the topics they were profiled for. Results may bear implications for the development of language models, because intuition-based setups appear better suited to some tasks than reasoning-based ones.
自然语言处理
[NLP-0] Learning to Stop without Learning to Stop: Self-Supervised Confidence Training Improves Reasoning Efficiency
【速读】： 该论文旨在解决生成式推理模型（Generative Reasoning Models）在推理过程中产生过长推理轨迹导致计算成本高昂的问题。传统方法通常通过推理时的早停机制或训练阶段引入长度惩罚以鼓励更短的推理路径来提升效率，但这些方法往往需要显式的优化目标或额外的控制逻辑。本文提出一种新思路：利用自监督方式对模型进行“置信度”（confidence）微调，即让模型在自身推理过程的中间节点上预测对答案的置信度，仅使用600个训练样本即可完成。该置信度仅作为训练目标，不直接约束推理长度、效率或停止行为。在推理阶段，模型仍采用标准生成流程，无需引入置信度获取或早停机制。实验结果表明，尽管未显式优化效率，该方法在Gemma、Qwen、Nemotron和GPT-OSS等多个模型上，于数学、科学与编程推理基准任务中实现了高达25%的生成词元（tokens）减少，且保持与基线相当的准确率。分析显示，置信度监督主要保留了原始模型的高层次推理结构，而非选择性抑制特定行为。研究揭示，高效推理可作为学习元认知信号（metacognitive signals）的自然衍生结果，而无需直接将其纳入优化目标。
链接: https://arxiv.org/abs/2609.31619
作者: Parsa Hosseini,Akasha Tigalappanavara,Sumit Nawathe,Chenrui Fan,Sourya Basu,Genta Indra Winata,Anirban Das,Soheil Feizi,Nima Chitsazan
机构: University of Maryland(马里兰大学); AI Foundations, Capital One
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Reasoning models often generate very long reasoning traces, making inference computationally expensive. Existing approaches typically improve efficiency either through inference-time early-stopping mechanisms or by explicitly encouraging shorter reasoning during training, for example through reinforcement learning with length penalties. We show that substantial efficiency gains can instead emerge from a different kind of supervision: \textitconfidence. Using a self-supervised procedure, we fine-tune reasoning models to predict their confidence in the answer at intermediate points along their own reasoning trajectories using only 600 training problems. Confidence is used only as a training target: the loss contains no objective for reasoning length, efficiency, or stopping. At inference, the fine-tuned models use the standard generation procedure, with no confidence elicitation or early-stopping mechanism. Despite this, self-supervised confidence fine-tuning makes reasoning more efficient, reducing generated tokens by up to 25% at matched accuracy across Gemma, Qwen, Nemotron, and GPT-OSS models on mathematical, scientific, and coding reasoning benchmarks, with efficiency gains comparable to methods that explicitly optimize for shorter reasoning. Analysis of reasoning episodes further shows that confidence supervision largely preserves the base models’ high-level reasoning composition rather than selectively suppressing particular behaviors. Our results suggest that efficient reasoning may emerge as a downstream consequence of learning metacognitive signals, without being directly optimized.
[NLP-1] User Model Extraction via Belief Self-Distillation
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）中用户属性隐式建模的可解释性与可控性问题，即模型在对话中基于用户行为形成对用户的内在信念，但这些信念难以被直接观测或因果操控。其核心解决方案是提出**信念自蒸馏（Belief Self-Distillation, BSD）**框架，该框架通过冻结的预训练模型作为自身教师，从自然对话中无监督地学习紧凑的用户表征，并实现对这一表征的“读取”与“写入”。与传统线性探针不同，BSD不仅提取激活信息，更构建了一个可被直接因果干预的状态，从而支持对模型行为的精准调控。实验表明，BSD能准确恢复用户信念，并在干预效果上显著优于匹配的隐藏状态操控方法；尤其关键的是，研究发现模型拒绝请求的行为不仅依赖于输入内容，更取决于其对用户意图的推断——改变该信念即可在保持请求不变的情况下改变拒绝行为。此外，研究揭示了跨模型的惊人规律：独立训练的不同大模型在用户表征空间中呈现出共享的几何结构。这些发现表明，大模型中的用户信念是可读且可因果操作的内部状态，对人工智能安全具有重要启示，提示模型的安全决策机制实际上依赖于其对交互对象的“认知”。
链接: https://arxiv.org/abs/2609.31603
作者: Ali Holmov,Yiran Huang,Kirill Bykov,Zeynep Akata
机构: Technical University of Munich(慕尼黑工业大学); Helmholtz Zentrum München(慕尼黑亥姆霍兹中心)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:
Abstract:Large language models (LLMs) implicitly infer attributes of their users and adapt their behavior accordingly, yet these beliefs remain difficult to inspect and causally manipulate. We introduce Belief Self-Distillation (BSD), a unified read-write framework that bridges linear and causal probing by learning a compact user representation that can be both decoded and written back into the model. The frozen LLM acts as its own teacher, distilling beliefs from natural conversations without external annotations. Unlike conventional probing, BSD isolates not only information present in activations, but a state whose causal role can be directly tested. Across multiple model families, BSD faithfully recovers user beliefs and enables substantially stronger interventions than matched hidden-state steering. Crucially, we find that refusal depends not only on the request, but on the model’s inferred user intent: changing this belief alters refusal while holding the request fixed. We further uncover a striking cross-model regularity: independently trained LLMs converge on a shared geometry for representing their users. Together, these results reveal implicit user models as readable and causally writable internal states with direct implications for AI safety, shaping how models condition safety decisions on whom they believe they are interacting with.
[NLP-2] Compact Documentation for Coding Agents : A Benchmark an Optimizer and Why It Does Not Transfer
【速读】： 该论文旨在解决生成式代码代理（coding agent）在处理软件问题时，自然语言文档是否能够有效辅助其修复缺陷的问题。其核心挑战在于评估文档质量与代码修复能力之间的因果关系，尤其是在真实项目环境中验证文档的实际效用。解决方案的关键在于构建一个“往返测试基准”（roundtrip benchmark），通过衡量从文档重建的代码能否通过原始测试来评估文档描述的完整性（completeness），并发现文档的完整性而非长度是决定其保真度（fidelity）的关键因素。基于该基准作为优化信号，研究者设计出一种能实现完整保真度且可泛化至未见文件的描述生成提示（prompt）。然而，进一步实证测试表明，在源码可见的前提下，静态紧凑文档或检索到的上下文信息均无法超越仅依赖问题描述本身的表现，从而揭示了文档有效性存在明确边界：当源码可用时，文档的增益效应消失。研究因此报告了这一负面结果，并公开提供基准与优化工具，以厘清文档在何种条件下真正发挥作用。
链接: https://arxiv.org/abs/2609.31587
作者: Md Shohel Arman,Igor Molybog
机构: 未知
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 13 pages. Code and data: this https URL
Abstract:We investigate whether natural-language documentation helps coding agents resolve software issues, and we build the tools to construct and evaluate it. We introduce a roundtrip benchmark that scores code descriptions by whether code regenerated from them passes the original tests, and show that completeness, not length, drives a description’s fidelity. Using the benchmark as an optimization signal, we discover a description-writing prompt that reaches full fidelity and generalizes to unseen files. We then test the hypothesis that motivated the work: that better documentation helps an agent resolve real repository issues. Across two model families and ten repositories, and against a positive control confirming that our evaluation can detect a genuine improvement, we find that it does not. When the source is present, neither static compact documentation nor retrieved context beats the issue alone. We report this negative result together with the benchmark and the optimizer, and we characterize the boundary at which documentation helps.
[NLP-3] Strategically Diverse Sampling for Self-Training
【速读】： 该论文旨在解决自训练（self-training）数据构建中因重复采样导致的策略同质化问题，即现有方法通常依赖独立同分布（IID）采样并仅基于正确性过滤，从而过度强化模型已偏好的策略，限制了其泛化能力。其核心解决方案是引入“策略多样性”（strategic diversity），即在生成训练数据时优先考虑解决问题路径的实质性差异，而非单纯追求正确性或教师模型规模。关键创新在于提出两种采样方法：GROOT，通过构建分层策略树并采样不同路径以实现结构化多样性；以及口语化采样（Verbalized Sampling, VS），用于生成无结构的多样化策略集合。实验表明，基于策略多样性的数据训练出的模型在编程竞赛和下一章预测等复杂任务上表现更优，并显著提升强化学习（RL）与测试时缩放（test-time scaling）的初始化效果。尤为突出的是，使用来自较小模型（Qwen3-4B）的不正确但策略多样的推理轨迹进行自训练，性能超越从超大规模教师模型（235B）通过传统IID蒸馏获得的结果，挑战了“正确性”与“教师规模”决定训练数据价值的主流认知，证明策略多样性本身可成为比正确性或教师规模更重要的优质训练数据特征。
链接: https://arxiv.org/abs/2609.31571
作者: Alexander Gurung,Esmeralda S. Whitammer,Mirella Lapata
机构: University of Edinburgh(爱丁堡大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Many LLM training and inference methods, including RL and test-time scaling, depend on repeated sampling, but benefit only when the responses meaningfully differ. Self-training faces the same challenge: training data is typically constructed by sampling IID responses and filtering primarily for correctness, thereby overrepresenting strategies a model already favours. We investigate strategic diversity, or substantive variation among approaches to a problem, as an alternative principle for constructing self-training data. We generate strategically diverse data with two sampling methods: GROOT, a new method which constructs a hierarchical tree of approaches and samples distinct paths, and Verbalized Sampling (VS), adapted to produce an unstructured set of approaches. Across competitive programming and Next-Chapter Prediction domains, models trained on strategically sampled data outperform IID-trained counterparts on difficult tasks and provide strong initializations for RL and test-time scaling. Most strikingly, self-training on strategically diverse but incorrect traces from Qwen3-4B outperforms IID distillation from a 235B teacher. These results challenge prevailing assumptions about what makes useful self-training data and show that diversity of approaches can matter more than correctness or teacher scale.
[NLP-4] MexHat: A Dataset for Hate Speech Detection in Mexican Spanish Videos
【速读】： 该论文旨在解决多模态语境下针对墨西哥西班牙语（Mexican Spanish）的仇恨言论检测（Hate Speech Detection）中存在的资源匮乏问题，尤其关注非英语语料在捕捉语言与文化语境线索方面的不足。其核心挑战在于，仇恨言论具有高度依赖上下文、文化背景和语用细微差别的特性，而现有模型在非英语场景下的泛化能力受限。为此，本文提出 MexHat——一个包含约1000个视频片段的多模态数据集，专门用于支持墨西哥西班牙语环境下的仇恨言论识别任务。该数据集涵盖两个标注任务：三分类评估（无负面内容、冒犯性内容、仇恨言论内容）以及细粒度的三类仇恨言论子类别划分，充分体现了语义与文化层面的复杂性。实验结果揭示了该任务在标注一致性、跨模态对齐及文化敏感性方面的固有难点。因此，本研究的关键解决方案在于构建首个面向墨西哥西班牙语的高质量、细粒度标注的多模态仇恨言论数据集，为提升非英语语境下生成式AI（Generative AI）系统在内容安全中的鲁棒性与文化适应性提供了关键基础。
链接: https://arxiv.org/abs/2609.31553
作者: Itzel Tlelo-Coyotecatl,Hugo Jair Escalante
机构: INAOE(国家光学研究所, 墨西哥普埃布拉); The University of Texas at El Paso(德克萨斯大学埃尔帕索分校, 美国埃尔帕索)
类目: Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注: Preprint submitted to CIARP 2026
Abstract:Ensuring online safety through content monitoring had raised Hate Speech Detection as a crucial task to be addressed. By essence the task demands the capture of contextual cues, which are essential for a precise understanding of the content’s intent. Although automated detection approaches for the task have advanced significantly, the scarcity of non-English resources persists, limiting the ability of models to adapt to the subtle, context-dependent, and culturally related nature of multimodal content. In this paper, we introduce MexHat, a video dataset designed to capture the linguistic and cultural cues for the hate-speech detection task in a Mexican Spanish context. Our dataset comprises around 1k video clips annotated across two tasks: a three-way class evaluation (no negative content, offensive content and hate-speech content), and a fine-grained class evaluation including three hate-speech sub-categories. The dataset statistics and the baseline results highlight the inherent challenges associated with the task. Disclaimer: This paper contains sensitive content that may be disturbing to some readers.
[NLP-5] Muslim: A Deployed Arabic Voice AI Platform for Grounded Islamic Knowledge
【速读】： 该论文旨在解决将基于伊斯兰知识的语音人工智能（Voice AI）系统从研究原型转化为可实际部署、可持续运营且具备抗滥用能力的生产级平台所面临的挑战。核心问题在于：如何在保证内容准确性与服务可靠性的同时，实现高效、可扩展且安全的语音交互系统，尤其针对阿拉伯语语境下宗教知识的精准传播需求。其解决方案的关键在于三个方面：一是发布了一套经过微调的阿拉伯语伊斯兰领域模型工具集，包括一个高效的指令路由大语言模型（Muslim-6B-PRO，59.4亿参数）和一个在现代标准阿拉伯语（Modern Standard Arabic, MSA）语音合成（TTS）基准测试中位列前五（17个系统中第5，11个开源系统中第2）的Fasih-TTS-V1模型；二是构建了一个集成账户管理与用量计量机制的系统层，通过按账号分配使用额度、容量感知拒绝响应及延迟至关键节点才触发的邮箱验证策略，有效防止滥用并支持产品化运营；三是设计了针对典型故障模式（即GPU受限的代理主机失联而前端服务仍正常运行）的三层可观测性架构（健康状态监控、错误上报与产品数据分析），从而实现对系统稳定性的实时把控。此外，论文还提供了真实场景下的性能指标（如124例诵读验证准确率达98.4%，端到端语音延迟为0.9–1.7秒），并深入分析了在生产环境中运行宗教知识类语音产品所面临的工程权衡与局限性。
链接: https://arxiv.org/abs/2609.31511
作者: Yahya Mohamed Elnawasany
机构: 独立研究员(Independent Researcher); 埃及(Egypt)
类目: Computation and Language (cs.CL)
备注: 6 pages, 4 tables. Deployed system: this https URL - released models: this https URL
Abstract:We present Muslim, a production Arabic voice AI platform serving grounded, sourced Islamic knowledge to real users. Beyond a real-time voice pipeline (NeMo Arabic ASR, an OpenAI-compatible LLM endpoint, self-hosted TTS) and a deterministic multi-source retrieval layer routed across six Model Context Protocol servers, we report three things a research prototype typically lacks. First, a released family of fine-tuned Arabic Islamic model artifacts: an efficient tool-routing LLM (Muslim-6B-PRO, 5.94B parameters) and a Modern Standard Arabic TTS model (Fasih-TTS-V1) that ranks 5th of 17 overall and 2nd of 11 open-weight systems on the community-voted Arabic TTS Arena for MSA. Second, an account and metering layer - a free per-account turn allowance, capacity-aware refusal, and email verification deferred to the point it actually matters - that turns an open demo into an operable, abuse-resistant product. Third, a three-layer observability stack (liveness, error reporting, product analytics) built specifically around the system’s characteristic failure mode: a GPU-bound agent host going silent while the web tier keeps serving normally. We report real, measured latency and accuracy figures (98.4% recitation-validation accuracy on 124 cases; end-to-end voice latency of 0.9-1.7s) and discuss the concrete engineering trade-offs and limitations of running an Islamic-knowledge voice product in production.
[NLP-6] Evaluating Cultural Awareness of LLM s for Haitian Creole
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在低资源语言——特别是海地克里奥尔语（Haitian Creole）——中文化意识严重不足的问题。尽管海地克里奥尔语拥有数百万母语使用者，但其在数字资源中长期被忽视，导致现有模型在该语言上的表现显著低于高资源语言（如法语），且难以准确捕捉目标社区的文化规范与价值观。其解决方案的关键在于构建首个针对海地克里奥尔语的系统性文化意识评估基准，通过由母语者精心设计的文化相关提示（culturally salient prompts），在文本补全（text infilling）任务中从四个互补维度——具体性（specificity）、偏见（bias）、多样性（diversity）与变异性（variation）——对模型的文化理解能力进行量化评估。研究发现，海地克里奥尔语模型的表现不仅整体低于法语模型，且在不同领域间波动更大，并易受法语语言干扰；故事生成结果还揭示了对海地人物反复呈现苦难与坚韧的刻板叙事模式，表明即使正面刻画仍可能隐含文化偏见。该工作公开了代码、基准数据集及评估框架，为未来低资源语言的文化敏感性研究提供了可复用的工具与标准。
链接: https://arxiv.org/abs/2609.31506
作者: Christelle Clervilsson,Yanzhu Guo
机构: Telecom Paris, Institut Polytechnique de Paris (法国巴黎电信学院，巴黎综合理工学院)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models (LLMs) exhibit substantial performance disparities between high- and low-resource languages. Beyond lower task performance, they often fail to capture the cultural norms and values of underrepresented communities. In this work, we present the first systematic evaluation of cultural awareness in LLMs for Haitian Creole, a language spoken by millions but severely underrepresented in digital resources. We assess cultural awareness along four complementary dimensions—specificity, bias, diversity, and variation—using a benchmark of culturally salient prompts curated by native speakers in a text infilling setting. Our results reveal a clear gap between cultural awareness in Haitian Creole and higher-resource French, with Haitian performance being more uneven across domains and more affected by French linguistic interference. Story generation further reveals recurring portrayals of Haitian characters through hardship and resilience, showing that even positive characterizations can encode stereotypical narratives. Our code, benchmark, and evaluation framework are publicly available.
[NLP-7] ViSTA: A Simple Bridge Extends Visual Alignment to Clinical Time-Series Understanding in Multimodal LLM s
【速读】： 该论文旨在解决现有大型语言模型在处理结构化、高维度临床时间序列数据时，难以实现准确风险预测的问题。尽管生成式AI（Generative AI）在医学文本理解与问答任务中表现优异，但其语言能力无法有效转化为对患者动态生理指标的精准建模与预测。为此，论文提出ViSTA——一种轻量级适配器架构，通过将不规则数值测量信息融入预训练视觉-语言模型的图表表征中，在保持所有预训练参数不变的前提下，学习对视觉标记的修正。其核心创新在于利用紧凑的可训练参数（仅0.516万）实现对临床时间序列的高效融合，显著提升了急性肾损伤和死亡率预测性能。在MIMIC-IV数据集上，20亿参数的ViSTA模型在所有对比方法中均取得最高平均得分，其受试者工作特征曲线下面积（AUC）达0.7376，接近GPT-5.6 Sol在高推理成本下的0.7380表现；同时，在时间序列问答任务中，40亿参数模型以超过90%更少的可训练参数量达到69.27%准确率，仅存在2.82–4.88个百分点的精度差距。因此，ViSTA的关键突破在于实现了预训练语言模型向数值预测与动态时序问题回答任务的有效扩展。
链接: https://arxiv.org/abs/2609.31448
作者: Junyi Gao,Yu Shi,Pingzhao Hu,Ewen M Harrison
机构: Centre for Medical Informatics, University of Edinburgh(爱丁堡大学医学信息学中心); Health Data Research UK(健康数据研究英国); Biostatistics Division, Dalla Lana School of Public Health, University of Toronto(多伦多大学公共卫生学院生物统计学系); Department of Biochemistry, Western University(西安大略大学生物化学系)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Clinical prediction models estimate risk from patient measurements, while large language models support medical text understanding and question answering. Yet their language capabilities do not ensure accurate prediction from structured, high-dimensional clinical time series. Improving this ability would connect risk estimation with flexible questions about a patient’s evolving condition. We introduce ViSTA, a compact adapter that incorporates irregular numerical measurements into a pretrained vision-language model’s chart representations. It learns corrections to visual tokens while leaving all pretrained parameters unchanged. On MIMIC-IV, ViSTA has the highest mean scores among the compared adaptations on all four metrics for acute kidney injury and mortality prediction across models with 2-9 billion parameters. With 0.516 million trainable parameters, the 2-billion-parameter model reaches an area under the ROC curve of 0.7376 for acute kidney injury, compared with GPT-5.6 Sol’s 0.7380 with text input and high reasoning effort. Training for temporal question answering yields 69.27% accuracy at 4 billion parameters with over 90% fewer trainable parameters than low-rank adaptation using charts or numerical text, at a 2.82-4.88 percentage-point accuracy gap. ViSTA extends pretrained language models to numerical prediction and temporal questions.
[NLP-8] Sorry Robot Happy Human: Vision-Language Models Read Only One of Two Legible Typographic Layers EMNLP2026
【速读】： 该论文旨在解决视觉语言模型（Vision-Language Models, VLMs）在处理多层文本图像时的脆弱性问题，特别是对排版攻击（typographic attacks）的敏感性以及在具有多重空间频率成分的文本结构中表现不佳的问题。其核心解决方案在于构建了一个名为DecoyBench的新数据集，该数据集采用“诱饵字体”（Decoy Font）方法生成，包含300张图像，每张图像均包含两层叠加的文本：一层为具有锐利轮廓线的文本（高空间频率），另一层为带有柔和阴影的文本（低空间频率）。通过在两种提示方式（原始提示与引导提示）和两种分辨率（512×512 和 64×64）下对六种来自三个不同模型家族的闭源VLM进行评估，研究发现人类参与者能够以高准确率读取两层文本，而大多数模型在高分辨率下虽可准确识别轮廓文本，却几乎无法完整提取阴影文本；在低分辨率下，模型与人类均无法识别轮廓文本，但模型仍能高精度提取阴影文本。这一结果揭示了当前VLM在处理多层、多频段文本结构时存在系统性行为局限，其关键在于模型对不同空间频率成分的感知能力不均衡，表现出对低频特征（如阴影）更强的依赖性，而对高频细节（如锐利轮廓）的捕捉能力不足，暴露了其内部表征机制在复杂排版场景下的结构性缺陷。
链接: https://arxiv.org/abs/2609.31403
作者: Mert İncidelen,Yamen Kashkash,Asya Berker,Murat Aydoğan
机构: Fırat University (法蒂尔大学); Department of Artificial Intelligence and Data Engineering (人工智能与数据工程系); Department of Software Engineering (软件工程系)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to the First Workshop on Document Intelligence and Understanding (DocInsights 2026), co-located with the 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP 2026)
Abstract:Vision-language models (VLMs), despite their success in optical character recognition (OCR) tasks, are vulnerable to typographic attacks and have a fragile structure for images with multiple text layers. In this study, the DecoyBench dataset was created using the Decoy Font method. The dataset consists of 300 images, each containing text with sharp contour lines superimposed on another text with soft shading. Six recent closed-source models from three different model families were evaluated using this dataset under two different prompting conditions (naive and guided) and at two different resolutions ( 512\times512 and 64\times64 ). A validation study showed that human participants could read both text layers with high accuracy. In contrast, the models, with most variants and both prompting methods, read the contour text with near-human accuracy at high resolution, but almost never fully extracted the shading text. At low resolution, the contour text could not be read by either the models or humans, while the shading text could be extracted with high accuracy. The findings indicate that the evaluated VLMs exhibit a consistent behavioral limitation when processing typographic structures containing multiple spatial frequency layers.
[NLP-9] Intent2Tc: Automated Intent-to-Traffic Control Translation with Language Models
【速读】： 该论文旨在解决业务级服务意图（business-level service intents）向可部署的网络流量管理策略（traffic-management policies）自动化转换过程中存在的语义鸿沟问题，即如何在保证语义一致性与配置可靠性的同时，实现从高层抽象意图到底层可执行Linux traffic control（tc）配置的高效、准确映射。其解决方案的关键在于提出一个闭环式、基于大语言模型（LLM）驱动的Intent2Tc框架，该框架通过引入基于主动队列管理（Active Queue Management, AQM）的数字孪生（Digital Twin, DT）语义模型，结合自动化元数据提取、批判性驱动的精炼机制以及基于检索增强生成（Retrieval-Augmented Generation, RAG）的知识复用技术，实现了从高层业务意图到声明式子意图，再至经验证的可执行tc配置的两阶段精准转化。实验表明，该框架在100个符合RFC 9315标准的流量整形意图上表现出高语义保真度（最高达0.98）、完整语义单元覆盖（1.0）及低归一化编辑距离（0.045），且RAG显著降低了推理延迟与令牌消耗，使小型模型（如Phi-4-mini）性能逼近大型模型，充分验证了其在真实Linux tc平台上的实用性和可扩展性。
链接: https://arxiv.org/abs/2609.31397
作者: Andrea Masini,Sudipta Acharya,Paolo Bellavista,Luca Foschini,Burak Kantarci
机构: University of Bologna (博洛尼亚大学); Istanbul Technical University (伊斯坦布尔技术大学)
类目: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 6 pages, 6 figures, Accepted to IEEE Conference on Future Communications and Networks (FCN) 2026
Abstract:Automated and highly usable Quality-of-Service (QoS) enforcement requires translating high-level service intents into deployable traffic-management policies. Although intent-based networking (IBN) has simplified policy specification, bridging the gap between business-level intents and executable network configurations remains complex, error-prone, and difficult to automate. This paper presents Intent2Tc, a closed-loop language-model-driven framework that translates business-level traffic-shaping intents into declarative sub-intents and subsequently into validated, executable Linux traffic control (tc) configurations. The framework integrates an Active Queue Management (AQM)-based digital twin (DT) semantic model, automated metadata extraction, critique-driven refinement, and Retrieval-Augmented Generation (RAG)-based knowledge reuse to improve semantic consistency and configuration reliability. We evaluate multiple open-source large language models (LLMs) and small language models (SLMs), together with Claude Sonnet-4.6, on 100 Request for Comments (RFC) 9315-compliant traffic-shaping intents. Across both translation stages, Intent2Tc achieves high semantic fidelity, configuration accuracy, and deployment readiness, with Claude Sonnet-4.6 reaching 0.98 semantic similarity, 1.0 semantic unit coverage, and 0.045 normalized edit distance. Furthermore, RAG reduces token consumption and inference latency while enabling compact models such as Phi-4-mini to approach the performance of substantially larger models. Linux tc serves as the target configuration platform, demonstrating the practical applicability of the proposed framework.
[NLP-10] Highlight-Then-Summarize: Learning to Compress Evidence for Long-Context Understanding
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在处理长上下文任务时面临的挑战：尽管需要对长文档、对话或代码进行推理，但任务相关的关键证据往往稀疏且分散在大量无关或冗余内容之中。为应对这一问题，论文提出“先高亮后摘要”（Highlight-Then-Summarize, H2S）范式，其核心在于通过两阶段机制——首先识别与源文本相关、与问题相关的证据，然后将这些证据整合为一个紧凑的、问题条件化的摘要，再基于该摘要生成最终答案。该方法的关键创新在于引入过程级奖励机制（H2S-RL），不仅优化最终答案的正确性，还对证据选择和摘要构建过程进行精细化引导。为支持该方法的训练与评估，研究构建了H2S-Dataset（包含11个基准任务共6,647个样本，平均上下文长度达43.9K tokens）和H2S-Bench（涵盖七项长上下文任务的评测基准）。实验结果表明，在共享128K输入和4K输出预算下，H2S-14B模型平均得分达到32.60，优于Qwen3.8-27B模型10.17分，并在所有开源模型中表现最优；同时，其证据-摘要质量得分最高，且在仅使用4K输出预算时仍保持16K预算下97.1%的性能，充分验证了显式证据提取与集成在提升长上下文推理能力的同时，显著提升了生成效率与紧凑性。
链接: https://arxiv.org/abs/2609.31382
作者: Zhaoyuan Xia(1 and 2),Qinghongbing Xie(3),Yung Xiang Hue(3),Jianguang Jiang(2),Gaofeng Lu(2),Zhenyu Jiao(2),Xing Yuan(2),Dai Dai(2),Tong Mo(1),Long Zeng(3) ((1) Peking University, (2) Baidu Inc., (3) Tsinghua University)
机构: Peking University (北京大学); Baidu Inc. (百度公司); Tsinghua University (清华大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 23 pages, 13 figures. Zhaoyuan Xia and Qinghongbing Xie contributed equally. Corresponding authors: Dai Dai, Tong Mo, and Long Zeng. Code and data are available at this https URL
Abstract:Long-context understanding requires large language models (LLMs) to reason over lengthy documents, conversations, and code, yet task-relevant evidence is often sparse and scattered amid substantial irrelevant and redundant content. We propose Highlight-Then-Summarize (H2S), a compress-then-reason paradigm that first identifies source-grounded, question-relevant evidence and then integrates it into a compact, question-conditioned summary before producing the final answer. To train this behavior, we construct H2S-Dataset, comprising 6,647 examples from 11 benchmark families with an average context length of 43.9K tokens, and introduce H2S-RL, which provides process-level rewards for evidence selection and summary construction in addition to final-answer correctness. We evaluate on H2S-Bench, a seven-task long-context suite. Under a shared 128K input and 4K output budget, H2S-14B achieves an average score of 32.60, outperforming Qwen3.8-27B by 10.17 points and obtaining the strongest overall result among the evaluated open-source models. H2S-14B also achieves the highest Evidence-Summary Quality score and retains 97.1% of its 16K-budget performance with only a 4K output budget. These results show that explicitly selecting and integrating evidence improves long-context reasoning while enabling more compact generation.
[NLP-11] Stale-Document Poisoning: When Outdated Retrieval Overrides Correct Model Answers
【速读】： 该论文旨在解决生成式 AI 在面对过时信息时因外部检索证据失效而导致错误判断的问题，尤其关注“时间对齐失败”（temporal alignment failure）现象——即模型在使用已过时的检索证据时，即便其回答本身正确，仍可能被误导。其核心解决方案在于引入时间有效性判断机制：模型不仅需理解检索到的信息内容，还需识别该信息是否仍在有效适用期内。研究通过构建涵盖医学、法律、软件及平台政策领域的317个经验证的知识反转基准数据集，发现当模型未被告知旧证据的失效时间时，过时检索可使Llama和Qwen模型的错误率分别上升至30%和37%；而一旦明确告知旧证据的失效时间，大模型几乎能完全切换至正确答案。因果干预实验表明，时间有效性信息直接驱动最终决策。此外，研究提出一种固定的、具备时效感知能力的混合重排序器（recency-aware hybrid re-ranker），在时间元数据准确的前提下，可将中毒率降低4.6至10.0个百分点。因此，可靠检索增强生成（RAG）的关键在于实现选择性信任（selective trust），即模型必须具备评估检索证据时间适用性的能力，这依赖于对时间相关元数据的准确理解和利用。
链接: https://arxiv.org/abs/2609.31342
作者: Md Shamim Ahmed,Lukas Galke Poech,Richard Röttger
机构: 未知
类目: Computation and Language (cs.CL)
备注: 17 pages, 3 figures
Abstract:Retrieval-augmented generation (RAG) is often used to address outdated knowledge by providing external evidence. But retrieval helps only when that evidence is still valid. We identify a temporal alignment failure, stale-document poisoning, in which outdated evidence makes a model wrong despite answering correctly without retrieval. We construct a benchmark of 317 verified knowledge reversals across medicine, law, software, and platform policy, grounded in dated official sources. Across 12 models, recent medical reversals are harder than long-established ones. More importantly, outdated retrieval flips 30% of Llama and 37% of Qwen answers even without instructions to trust the document; explicit follow instructions raise these rates to 66% and 75%. Across four open models and four domains, poisoning ranges from 17-91%, while matched up-to-date evidence is followed in 97-100% of trials. To isolate temporal applicability, we keep the historical evidence unchanged across 50 reversals and vary only the evaluation date. A clear pattern emerges: dates alone produce only modest adaptation, but when models are explicitly told when the old evidence stops applying, the larger models switch to the appropriate answer almost perfectly. Causal interventions confirm that this validity information directly shapes the final decision. The same internal components also support broader comparison tasks, suggesting that temporal applicability can recruit a general reasoning mechanism used for other comparisons. Finally, a fixed recency-aware hybrid re-ranker reduces poisoning by 4.6-10.0 points when dates are accurate, with gains that depend on reliable temporal metadata. Reliable RAG therefore requires selective trust: models must determine not only what retrieved evidence says, but whether it still applies.
[NLP-12] he Right Information Extraction Pipeline Depends on the Document: Accuracy-Energy Trade-offs for Small Local Models EMNLP2026
【速读】： 该论文旨在解决在隐私敏感场景下，如何在本地部署的小型文本模型与视觉-语言模型之间，针对不同布局特征的文档选择最优的信息抽取（Information Extraction, IE）输入表示形式这一关键问题。其核心挑战在于权衡模型精度与能效之间的关系，尤其是在不使用云端服务、仅依赖本地运行的小规模（≤8B参数）模型的前提下。解决方案的关键在于系统性地评估多种配置组合——包括输入表示（页面图像或解析文本）、模型家族（文本仅模型或视觉-语言模型）以及推理配置（如批处理、量化策略），并基于真实数据集（Kleister-NDA合同与VRDU表格）进行基准测试。研究发现，批处理（batching）是提升能效的主导因素，可在不损失精度的情况下降低每页能耗38%-85%；而FP8量化在单请求场景下可节省27%-32%能量，但批量处理后收益下降至不足1mWh/页（9%-19%）。此外，神经光学字符识别（Neural OCR）的能耗是传统OCR的17倍，且无法达到帕累托前沿。最终结论表明：对于近纯文本文档，采用低成本解析器配合小型文本仅模型表现更优；而对于版式复杂的文档，则需依赖视觉-语言模型。该研究为实现高效、合规的本地化信息抽取提供了明确的设计指南。
链接: https://arxiv.org/abs/2609.31341
作者: Christoph Walser,Mauricio Fadel Argerich,Jonathan Fürst
机构: Zurich University of Applied Sciences, Switzerland(苏黎世应用科学大学, 瑞士); Universidad Politécnica de Madrid, Spain(马德里理工大学, 西班牙)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: Accepted to DocInsights at EMNLP 2026
Abstract:Whether an information extraction pipeline should process page images or parsed text depends on the document, and the answer flips across the layout spectrum. We study this trade-off under a constraint that rules out (closed) cloud services: privacy-sensitive documents processed on-premise by small ( \le 8\mathrmB parameter) text-only and vision–language models, evaluated on both accuracy and energy over a design space spanning input representation, model family, and inference configuration. Benchmarking on the near-plain-text Kleister-NDA contracts and the layout-rich VRDU forms, we find that batching is the dominant energy lever, cutting energy per page by 38-85% at no cost in accuracy, while FP8 quantization saves 27-32% when requests are served one at a time but less than 1mWh per page (9-19%) once batching is applied. Preprocessing dominates what remains: neural OCR costs 17\times more energy per page than classical OCR and never reaches the Pareto frontier. Which representation wins flips with the type of document: vision–language models on layout-rich documents and small text-only models with a cheap parser on near-plain text, where they are both more accurate and cheaper than any vision–language configuration. Our work yields concrete guidelines for energy-efficient, privacy-compliant local information extraction.
[NLP-13] Identifying Scientists on X
【速读】： 该论文旨在解决在互联网语境下，随着科学话语的重要性上升及传统知识体系的瓦解，如何自动识别用户群体中科学家与非科学家的问题。其核心挑战在于区分不同用户在科学传播中的角色，以支持对科学话语生态的深入分析。解决方案的关键在于结合用户个人简介和推文内容，提取语言学特征，并采用机器学习与深度学习模型进行分类：一方面利用随机森林（Random Forests）基于语言学特征实现较高准确率；另一方面通过对比学习微调的DeBERTa模型，在集成学习框架下显著提升性能，最终达到高达0.96的F1分数。此外，研究还公开了两个标注数据集，包含用户标签、推文及个人简介，为后续研究提供了重要资源。
链接: https://arxiv.org/abs/2609.31264
作者: Philipp Meier,Katarina Boland,Laura Kallmeyer,Stefan Dietze
机构: Heinrich-Heine-University, Düsseldorf, Germany; GESIS - Leibniz Institute for the Social Sciences, Cologne, Germany
类目: Computation and Language (cs.CL)
备注: Corrected version of Identifying Scientists on X published at Companion Publication of the 18th ACM Web Science Conference 2026
Abstract:With the growing importance of science-related discourse on the Web and the erosion of the classical knowledge order, it is important to identify different user groups, such as scientists, automatically. This work proposes an approach for identifying scientists and non- scientists on X/Twitter based on their user biographies and tweets. We show that we are able to classify accounts as scientists and non- scientists on two different datasets, reaching an F1 score of up to 0.88 using Random Forests with linguistic features and up to 0.96 using a contrastively fine-tuned DeBERTa model in an ensemble setup. Furthermore, we provide two datasets with X users labeled as scientists or non scientists and their respective tweets and user biographies.
[NLP-14] MoSAR: Mixture of Semantic Attention Regimes for Learning Adaptive and Approximable Attention Geometries
【速读】： 该论文旨在解决长序列语言建模中密集自注意力（dense self-attention）带来的二次计算复杂度瓶颈问题。传统高效替代方案通常预先设定注意力的稀疏或局部模式，但此类硬编码策略难以捕捉自然语言依赖关系的输入相关性与动态变化。为此，论文提出将注意力近似视为一个几何学习问题，通过数据驱动方式自动学习查询-键交互中的位置相关性衰减模式与全局交互保留区域。其核心解决方案为引入语义注意力模式混合模型（Mixture of Semantic Attention Regimes, MoSAR），该模型在位置编码后引入输入条件化的查询与键路由机制，动态选择短距离、中距离和全局三种注意力模式的组合，从而生成连续的距离依赖型注意力场，而非固定稀疏结构。该注意力几何结构在训练中自适应学习，并可通过top-1路由实现推理时的低开销离散化。在参数量匹配的500M模型预训练实验中，MoSAR在训练上下文长度下显著降低注意力作用范围，同时保持语言建模性能，优于使用旋转位置编码（RoPE）的密集注意力模型；在长度外推（length extrapolation）场景下，其困惑度表现优于包括ALiBi在内的多种强基线方法。此外，所学几何结构在确定性top-1离散化后仍保持稳定，表明其不仅具备良好的自适应能力，且适合低成本近似推理。
链接: https://arxiv.org/abs/2609.31261
作者: Michele Paolicelli,Alessandro Petruzzelli,Alessandro Franceso Maria Martina,Cataldo Musto,Giovanni Semeraro
机构: Università degli Studi di Bari Aldo Moro(巴里阿尔多·莫罗大学); Department of Computer Science(计算机科学系)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:The quadratic complexity of dense self-attention remains a central bottleneck for long-context language modeling. Many efficient alternatives address this cost by deciding in advance where attention should be sparse or local. We argue that attention approximation should instead be approached as a geometric problem, with the relevant interaction geometry learned from data: natural-language dependencies are input-dependent and difficult to prescribe in advance, so the model should learn where positional relevance can decay and where broader interactions must be preserved. We introduce Mixture of Semantic Attention Regimes (MoSAR), which learns such an adaptive, controlled-decay geometry over query–key interactions. Input-conditioned query and key routers, applied after positional encoding, select mixtures over short, medium, and global regimes, inducing a continuous distance-dependent attention field rather than a fixed sparsity pattern. This geometry is learned during training and can subsequently be discretized through top-1 routing. In controlled pre-training experiments with matched 500M-parameter models, MoSAR learns a substantially lower-reach attention geometry without degrading language-modeling quality, improving perplexity over dense RoPE at the training context length. Under length extrapolation, MoSAR achieves the best perplexity among all evaluated variants, including strong baselines such as ALiBi. Moreover, the learned geometry remains stable under deterministic top-1 discretization, suggesting that it is not only adaptive, but also amenable to low-cost approximation at inference time.
[NLP-15] PIA: A Personal Intelligence Agent Turning Health Conversations into Records and Records into Understanding
【速读】： 该论文旨在解决通用型智能体记忆机制在医疗健康场景下的适用性问题：传统基于文本相似性的摘要与检索方法无法有效处理临床语境中的时间序列数据、剂量信息及因果关系推理，导致关键医疗信息丢失或误读。其核心解决方案是提出个人智能代理（PIA, Personal Intelligence Agent），通过四重可插拔的领域无关控制机制——提取（extraction）、记忆（memory）、检索（retrieval）与理解（understanding）——构建面向消费者健康代理的结构化临床记录系统。该系统将自然语言请求转化为结构化的临床数据，并基于健康专用模块（包括医学术语词典、知识图谱、时序规则与数据模式）实现从一维召回、二维健康快照到三维动态轨迹与因果推断的多层次记忆注入。实证表明，随着上下文记忆深度增强，响应质量显著提升；同时揭示了自报健康数据存在非随机缺失、问题表述直接影响合成理解质量，以及近三分之一候选因果关联为结构性噪声等关键运营教训。
链接: https://arxiv.org/abs/2609.31255
作者: Jeonghun Yoon,Dongchan Kim,Hongyeon Yu,Young-Bum Kim,Jaegul Choo
机构: KAIST(韩国科学技术院); NAVER Corp.(NAVER公司)
类目: Computation and Language (cs.CL)
备注: 13 pages, 6 figures, 8 tables
Abstract:General-purpose agent memory summarizes conversations: it extracts salient snippets, embeds them, and retrieves the top-k into the prompt. A health agent cannot run on summaries: a dose becomes a sentence, “since last week” is resolved at the model’s discretion, and a three-month glucose trend cannot be answered by text similarity. We present PIA, a personal intelligence agent deployed alongside a consumer health agent. PIA receives the agent’s natural-language requests, decides for itself whether and how to write or read, and turns conversations into typed clinical records and records into a synthesized understanding of the user. Its memory harness consists of four controls – extraction, memory, retrieval, and understanding – each a domain-agnostic mechanism with a pluggable health module: schema, medical alias dictionary, knowledge graph, and temporal rules. We show how the same query receives a different answer as the memory injected into the response context deepens from one-dimensional recall, to a two-dimensional health snapshot, to a three-dimensional trajectory with causality, and report lessons from operation: self-reported health data are missing not at random, question phrasing governs the quality of synthesized understanding, and nearly a third of candidate causal links are structural noise that rules alone remove.
[NLP-16] RupeeBias: Auditing Demographic Bias in Indian Economic Guidance from Large Language Models
【速读】： 该论文旨在解决生成式 AI（Generative AI）在印度特定社会经济背景下，因存在系统性人口统计学偏见而可能对用户经济决策产生误导的问题。现有大语言模型（LLM）偏见评估基准多基于西方人口特征，未能涵盖印度社会中如种姓、宗教、区域身份、城乡差异等关键经济不平等维度，导致对真实社会影响的评估不足。为此，论文提出 RupeeBias——一个专为评估印度语境下 LLM 经济建议中的偏见而设计的基准测试。其核心解决方案在于采用单属性反事实设计，固定用户资质、经验或服务内容不变，仅改变单一人口统计标识符，从而精准量化不同群体间输出差异。该基准涵盖 87 个印度特有人口统计变量，覆盖种姓、宗教、区域身份、性别、残障状况及城乡位置六大维度，并以英语与印地语混合语（Hinglish）双语构建 39,150 个提示。对九个 LLM 的评估显示，在其他条件相同但仅人口标识不同的情况下，经济输出平均差异达 20.2%，揭示了系统性偏见的存在。该研究通过公开发布 RupeeBias，为未来针对印度本土社会经济背景下的生成式 AI 偏见研究提供了可复用、可扩展的评估工具。
链接: https://arxiv.org/abs/2609.31245
作者: Pavithra P M Nair,Bhavik Talaviya,Shourya Bhushan,Rahul Pankajakshan,Seema Guruvadoo,Avinash Agarwal,Gilad Gressel,Krishnashree Achuthan
机构: Amrita Vishwa Vidyapeetham(阿姆里塔世界大学); Unique Identification Authority of India(印度唯一识别机构)
类目: Computation and Language (cs.CL)
备注:
Abstract:Individuals turn to large language models (LLMs) for guidance across a wide range of economic tasks, from comparing loan options and planning savings to deciding what raise to ask for or how much to charge for their services. LLMs are known to reproduce social biases, and biased economic guidance may influence what users believe they are worth, what they ask for, and what they ultimately accept. This risk is especially salient in India, where economic outcomes are shaped by demographic categories such as caste and urban-rural location. Existing LLM bias benchmarks, however, are largely designed around Western demographic categories and therefore miss key axes of economic disparity in the Indian context. We introduce RupeeBias, a benchmark for auditing demographic bias in LLM-generated economic guidance across Indian economic settings. RupeeBias consists of 39,150 prompts spanning four use cases: salary estimation, salary increment estimation, counter-offer recommendation, and service pricing recommendation. The benchmark follows a single-attribute counterfactual design, holding the description of the user’s qualifications, experience, or service offering fixed while varying one demographic identifier at a time. RupeeBias covers 87 India-specific demographic identifiers across six axes: caste, religion, regional identity, gender, disability, and urban-rural location, with all prompts constructed in both English and Hinglish. We evaluate nine LLMs on RupeeBias and find systematic demographic disparities across all six axes. For otherwise identical prompts that differ only in demographic identifier, LLM-generated economic outputs differ by 20.2% on average. We publicly release RupeeBias to support future research on demographic bias in LLM-generated economic guidance across India-specific demographic and economic contexts.
[NLP-17] Where a Model Sends Its Own Repeated Token
【速读】： 该论文旨在解决黑箱模型识别（black-box model identification）中因模型对自然语言提示响应而产生的身份误判问题，特别是针对现有方法通过重复自身词元（token）输入以触发失败模式而非准确识别模型身份的局限性。其核心解决方案是：在单次前向传播中计算每个词元 $ t $ 对应的 $ \text{argmax}, p(. \mid t, t) $，从而构建一个从词汇表到自身的映射函数，该映射分为两部分——固定点（fixed points）与非固定点的映射行为。第一部分（固定点）作为预期内的失败估计量（failed estimand），其自然距离为词汇表基数的83%，在3471个样本中表现出2比特的语料操纵区分能力，但属性家族识别精度仅为0.5833；第二部分（非固定点映射）此前未被记录，仅有一篇论文将其记为零值。研究通过源词元配对（pairing on the source token）消除了基数混淆（cardinality confound），使相关系数从0.9128降至-0.0932，并将家族识别精度提升至0.8333（12个模型在19个模型池中测试，随机预期为0.1389），且结果在七个分词器组和多个语料库中保持稳健。双重零假设检验显示，频率匹配目标的共识为0.1429，独立边缘分布的共识为0.0798，表明家族信息显著优于分词器类型（0.2031 vs 0.1205）。此外，循环架构在平衡准确率上达到1.0，远超0.7895的多数率基准，或在排除各模型主导输出后仍保持0.90的鲁棒性。研究进一步量化了鲁棒性边界：8位权重舍入对映射影响小于训练语料去重（0.9004 vs 0.6353），而4位舍入则彻底破坏映射结构（0.0098；部署粒度下为0.1812，非粗化伪影），且精度下限随模型差异从0.201至0.9778不等。所有估计量与终止条件均在数据观测前注册，失败估计量亦被完整报告。
链接: https://arxiv.org/abs/2609.31181
作者: Nicolás Vera Zúñiga
机构: 独立研究员(Independent Researcher); Chile(智利)
类目: Computation and Language (cs.CL)
备注: 8 pages, 3 tables. Companion to arXiv:2608.10986 , arXiv:2608.21315 and arXiv:2609.29507 . Code, per-run results, pre-registrations and the findings ledger: this https URL (archived: this https URL )
Abstract:Black-box model identification works by scoring a model’s response to natural-language prompts. One line of work feeds models a degenerate input – their own token, repeated – to find a failure mode rather than an identity. We take that input and ask where the model goes when it does not. For each token t, read argmax p(. | t, t) in one forward pass; the result is a map on the whole vocabulary, with two halves. The first – which tokens are fixed points – is partially anticipated, and we report it as a failed estimand: the natural distance on it is 83% cardinality, separates a corpus manipulation by two bits in 3471 against a precision floor of zero, and attributes families at 0.5833. The second half, where the map sends tokens that are not fixed points, is unrecorded; the one paper holding those tokens logged them as a zero. Pairing on the source token removes the cardinality confound by construction (r from 0.9128 to -0.0932) and attributes families at 0.8333 – twelve models scored against a pool of nineteen – with chance 0.1389, across seven tokenizer groups and several corpora. Two nulls clear it: frequency-matched destinations agree at 0.1429, independent marginals at 0.0798. Family predicts agreement better than tokenizer (0.2031 against 0.1205), and recurrent architectures cluster at balanced accuracy 1.0 against a 0.7895 majority rate, or 0.90 once each model’s dominant destination is excluded – the figure we stand behind. We measure the robustness envelope: 8-bit weight rounding moves the map less than deduplicating the training corpus does (0.9004 against 0.6353, on one support), 4-bit destroys it (0.0098; 0.1812 at deployment granularity, so not a coarseness artefact), and the precision floor varies by model from 0.201 to 0.9778. All estimands and kill conditions were registered before the data, and the failed one is reported as fully as the surviving one.
[NLP-18] Improving Visual Sensitivity of LLM s on Multimodal Machine Translation with Metric-based Loss Weighting
【速读】： 该论文旨在解决多模态机器翻译（Multimodal Machine Translation, MMT）中模型对视觉信息敏感性不足的问题，即尽管模型能够接收与源文本相关的图像输入，但往往选择忽略这些有助于消除歧义的视觉信息。其解决方案的关键在于提出一种基于度量的损失加权（Metric-based Loss Weighting）训练方法，通过识别那些从视觉上下文中受益的词元（tokens），在损失函数中为其分配更高的权重以增强模型对视觉信号的依赖。该方法利用点对点交叉互信息（Point-wise Cross-mutual Information, PCXMI）度量来评估模型输出在有无视觉上下文时的概率差异，并引入一种基于一致性的PCXMI变体，实验表明二者结合使用可实现最优性能。在三个语言方向上的图像引导机器翻译任务中，该方法在CoMMuTE对比数据集上显著优于标准微调方法，翻译准确率提升超过7个百分点，同时保持了良好的通用翻译能力。
链接: https://arxiv.org/abs/2609.31169
作者: Paweł Mąka,Piotr Andruszkiewicz,Yusuf Can Semerci,Jan Scholtes,Gerasimos Spanakis
机构: Maastricht University (马斯特里赫特大学); Warsaw University of Technology (华沙理工大学); IDEAS Research Institute (IDEAS 研究所)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Multimodal Machine Translation aims to incorporate additional signal from non-textual modalities to improve translations by resolving ambiguities. While models, through multimodal fusion, are able to accept images related to the source text, they can ignore this information. Therefore, increasing their visual sensitivity remains an active research area. In this work, we introduce a training method, Metric-based Loss Weighting, that improves visual grounding of translations by increasing the loss function for tokens that benefit from the accompanying image. We identify these tokens using the Point-wise Cross-mutual Information (PCXMI) metric, which compares the model’s output probabilities with and without visual context. We introduce a Congruency-based PCXMI metric and experimentally show that both metrics working in combination yield the best results. We evaluate our method by fine-tuning three pretrained Multimodal Large Language Models on the task of Image-guided Machine Translation for three language directions. Metric-based Loss Weighting outperforms other tested methods on the CoMMuTE contrastive dataset, improving accuracy by up to more than 7 percentage points compared to standard fine-tuning, while maintaining strong general translation performance.
[NLP-19] JevAdvBench: A Benchmark and Black-Box Attacks for Reinforcement Learning for Calibrated Decisions Models
【速读】： 该论文旨在解决生成式人工智能（Generative AI）中基于强化学习的校准决策模型（Reinforcement Learning for Calibrated Decisions, RLCD）在对抗性攻击下的鲁棒性评估难题。现有方法依赖于模型生成内容或执行结果的评分，但RLCD模型不产生中间输出，仅返回结构化答案，使得传统评测手段失效；同时，由于相同请求可能产生不同响应、标注数据多源自模型自身且API处理过程不可见，导致评估难以进行。其解决方案的关键在于提出一种新的评估范式：不再以外部标签为基准，而是将每个受攻击后的决策与模型自身的清洁决策（clean decision）进行对比，并通过重复运行同一请求来量化扰动带来的变化。基于此，作者构建了首个面向RLCD模型的对抗性基准测试集JevAdvBench，包含66个场景下的812个类型化问题及9,744种单编辑变体攻击，每种攻击均通过计费输入令牌确认已成功送达模型。实验表明，在jev-1.13.0模型上，重述操作对结果影响极小（偏差不超过1.2个百分点），而仅在状态中添加一条未经验证的意见即可使12.1%的决策发生改变，与最强注入指令效果相当，并致使38%的高置信度回答低于0.8的阈值从而触发人工审核。因此，论文主张在基于RLCD模型的应用中，应将输入状态视为不可信的、潜在被操纵的输入。
链接: https://arxiv.org/abs/2609.31142
作者: Jianyi Hu,Hangtao Zhang,Yi Liu,Yeqi Zeng,Li Zeng,Xianlong Wang,Rui Wang,Leo Yu Zhang
机构: Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所); School of Cyber Security, University of Chinese Academy of Sciences(中国科学院大学网络空间安全学院); Huazhong University of Science and Technology(华中科技大学); Griffith University(格里菲斯大学); Changsha University of Science and Technology(长沙理工大学); City University of Hong Kong(香港城市大学)
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 33 pages, 13 figures, 19 tables. Project website: this https URL
Abstract:Models trained with reinforcement learning for calibrated decisions (RLCD), such as Jev, answer a typed question about an input, the state, with a probability, a choice, or a score, and software acts on the answer without a person reading it. Their robustness has not been measured: adversarial benchmarks score what a model generates or executes, whereas a typed model generates nothing and returns a well-formed answer even when manipulated. Measurement is also hard, because identical requests can return different answers, most available labels come from the model itself, and the API preprocesses each request out of view. Our key idea is to score each attacked decision against the model’s own clean decision rather than against labels, and to read it against the change caused by an identical re-run. Building on this, we introduce JevAdvBench, to our knowledge the first adversarial benchmark for RLCD models, with 812 typed questions over 66 scenarios, and a black-box attack suite of 9,744 single-edit variants that each edit one part of a request, with billed input tokens confirming that the edit reached the model. On jev-1.13.0, rewording stays within 1.2 percentage points of the re-run baseline, and fields outside the schema never reach the model. In contrast, one unverified opinion appended to the state flips 12.1% of decisions, statistically tied with the strongest injected command (10.1%), and pushes 38% of confident answers below the 0.8 confidence threshold that routes them to human review. Applications built on RLCD models should therefore treat the state as untrusted, argued input. Project website: this https URL
[NLP-20] Do we need to answer that question? Salience and Answerability of Potential Questions in Naturalistic Dialogue
【速读】： 该论文旨在解决自然语境对话中“讨论中的问题”（Question Under Discussion, QUD）建模的实证问题，即探究生成的潜在问题的显著性（salience）是否能够预测其后续是否得到解答。其解决方案的关键在于构建一个包含7,124个从英国国家语料库（British National Corpus）中自动提取的对话语句及其前文语境生成的问题数据集，并对每个问题进行显著性和可回答性（answerability）的人工标注。研究发现，对话中显著性与可回答性之间存在稳健但较弱的正相关关系，表明更显著的问题更有可能被回应，但这一关联性明显弱于独白文本，暗示对话结构的可预测性较低。此外，研究还发现结构化互动相较于非组织化对话具有更强的标注者间一致性，说明对话的组织程度影响了问题理解与响应的一致性。
链接: https://arxiv.org/abs/2609.31130
作者: Amandine Decker(LORIA, UL, CNRS, SEMAGRAMME, GU),Maxime Amblard(SEMAGRAMME, LORIA),Ellen Breitholtz(GU)
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:We empirically investigate Question Under Discussion based modelling in naturalistic dialogue by studying whether the salience of generated potential questions predicts their subsequent resolution. Building on Wu et al. (2024), we construct a dataset of 7,124 questions automatically generated from utterances and preceding context from the British National Corpus, and annotated for salience and answerability. We find a robust but low positive correlation between salience and answerability in dialogue, indicating that more salient questions are more likely to be addressed. However, this effect is markedly weaker than in monologic text, suggesting that conversational structure is less predictable. We further observe that structured interactions exhibit stronger alignment between annotators than less organised dialogues.
[NLP-21] LocUS: Head Selection and Subspace Projection for Targeted Activation Steering
【速读】： 该论文旨在解决生成式AI（Generative AI）中激活引导（activation steering）方法在推理时干预过程中存在的副作用问题，即标准方法通过对比数据估计每层的引导方向并作用于整个表示空间，容易将无关属性（off-target properties）耦合到干预中，从而损害模型的其他通用能力。其解决方案的关键在于提出一种名为LocUS（Localized Unembedding Steering）的新方法，该方法将激活引导限制在模型自身输出词汇子空间内，通过识别解嵌入矩阵（unembedding matrix）中与特定属性相关的线性子空间，施加几何约束以确保引导变换仅作用于特定子空间，并且仅影响注意力头中的稀疏子集。实验结果表明，LocUS在毒性缓解、情感重定向和谄媚性抑制等任务上达到或超越现有最优基线性能，同时仅干预少于6%的参数，显著提升了对通用能力的保持效果。
链接: https://arxiv.org/abs/2609.31122
作者: Irene Tallini,Lorenzo Basile,Valentino Maiorca,Francesco Locatello,Alberto Cazzaniga
机构: Université Côte d’Azur, Inria, LJAD, Maasai Project Team (尼斯大学, 法国); Area Science Park (科学园区); Institute of Science and Technology Austria (奥地利科学技术研究所)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Activation steering is a powerful training-free paradigm for controlling large language models at inference time. However, standard approaches estimate a per-layer steering direction from contrastive data and apply it on the layer’s entire representation space, which may couple the intervention to off-target properties present in the contrastive data and degrade unrelated capabilities. To mitigate this issue, we introduce LocUS (Localized Unembedding Steering), a method which grounds activation steering to the model’s own output vocabulary subspace. By identifying a property-specific linear subspace within the unembedding matrix, LocUS enforces a geometric constraint that restricts the steering transformation to a specific subspace and at the same time localizes its application to a sparse subset of attention heads. Extensive evaluations across three model families on toxicity mitigation, sentiment redirection and sycophancy suppression show that LocUS matches or outperforms state-of-the-art baselines while intervening on under 6% of parameters and better preserving general capability.
[NLP-22] Modeling Student Sensemaking with LLM s and Knowledge-Graph-Guided Inference
【速读】： 该论文旨在解决协作式科学学习中学生对话的多维度分析难题，即如何有效识别学习者在对话中发现知识盲点、构建解释并寻求解决路径的过程。这一分析依赖于理论驱动的细致解读，但传统方法存在劳动密集且难以规模化的问题。为此，研究提出利用指令微调的大语言模型（LLM）实现无需任务特定训练的多维度协作意义建构分析，并探究结构化知识状态信息对模型推理能力的提升作用。其关键解决方案在于通过设计不同提示范式（包括定义性支架、推理模式与对话轮次结构），评估中等规模LLM在复杂标注数据集上的表现，结果表明：引入推理增强提示可显著改善对失败型意义建构的识别；而结合结构化知识状态诊断则为模型提供了额外的语义锚点，进一步提升了对失败案例的检测准确率及与专家标注的一致性。研究还揭示，单一提示配置无法在所有意义建构维度上均取得最优效果，凸显了该任务的多维特性。
链接: https://arxiv.org/abs/2609.31046
作者: Özge Alacam,Zübeyde Demet Kirbulut Güneş,Funda Ekici,Nurcan Turan-Oluk,Dilay Dinçdemir,Hakkı Kadayıfçı,Sevinç Nihal Yeşiloğlu,Burcu Işık,Halil Tümay,Sinem Gencer
机构: Center for Information and Language Processing, LMU München, Munich, Germany; Department of Chemistry Education, Gazi Faculty of Education, Gazi University, Ankara, Türkiye
类目: Computation and Language (cs.CL)
备注:
Abstract:Collaborative science learning requires nuanced interpretation of student dialogue to characterize how learners identify knowledge gaps, build explanations, and work toward resolution - a theory-driven analysis that is labor-intensive and difficult to scale. We investigate whether instruction-tuned large language models (LLMs) can support multidimensional analysis of collaborative sensemaking without task-specific training, and whether structured knowledge-state information improves model inference. We evaluate two mid-size LLMs on 23 richly annotated, expert-labeled episodes across prompting conditions that vary definitional scaffolding, reasoning mode, and turn structure. Without reasoning, models tend to overpredict successful sensemaking; reasoning-enabled prompting improves identification of unsuccessful cases. Knowledge-state diagnostics provide additional grounding, improving detection of unsuccessful sensemaking and increasing agreement with expert annotations. No single configuration performs best across all sensemaking dimensions, underscoring the multidimensional nature of the task.
[NLP-23] Same Text Different Numbers: The Divergence of LLM -Based Measures
【速读】： 该论文旨在解决生成式大语言模型（Generative Large Language Models, LLMs）在企业文本量化分析中因模型选择差异导致测量结果不一致的问题。研究发现，不同LLM对标准普尔500指数公司业绩说明会文本进行情感、管理清晰度、不确定性、回答具体性以及气候与政治风险等13项文本指标的评分存在显著差异，跨模型排名相关性平均仅为0.52，且多数差异源于各模型自身特性而非文本本身的模糊性。这种模型特异性显著影响下游推断结果，表现为回归系数的大小、符号及统计显著性在不同模型间波动明显。尽管通过集成多个提供商模型可提升排名稳定性，但得分水平仍高度依赖所包含的具体模型集合。因此，论文提出关键解决方案是：将LLM生成的变量视为模型依赖型测量，必须在多个模型间进行交叉验证以确保结果可靠性。
链接: https://arxiv.org/abs/2609.31013
作者: Hamid Boustanifar,Sasan Mansouri
机构: EDHEC Business School (EDHEC商学院); University of Groningen (格罗宁根大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); General Finance (q-fin.GN); Risk Management (q-fin.RM)
备注: 86 pages, including an online appendix
Abstract:Researchers increasingly use generative large language models (LLMs) to convert corporate text into empirical variables. We examine the extent to which LLM-based textual measures are invariant to model choice using thirteen measures, including sentiment, management clarity, uncertainty, answer specificity, and climate and political risk. Seven LLMs from different providers score earnings call transcripts of SP 500 companies on these constructs. Cross-model rank correlations average only 0.52, and transcript-level differences common across providers account for only 34% of total score variation. Cross-model disagreement does not predict subsequent analyst or market disagreement, consistent with a substantial model-specific component rather than common ambiguity in the underlying disclosure. Model choice significantly affects downstream inference, with coefficient magnitudes, signs, and statistical significance varying substantially across models. Averaging across providers makes transcript rankings more stable for most constructs, but score levels remain sensitive to the models included in the ensemble. LLM-generated variables should therefore be treated as model-contingent measurements and validated across providers.
[NLP-24] G2PTQ: Improving LLM Post-Training Quantization with Generalized Gradient Compensation
【速读】： 该论文旨在解决后训练量化（Post-training Quantization, PTQ）中现有GPTQ类方法的两大互补性局限：一是采用局部、层内优化目标的方法缺乏全局监督；二是采用全局优化目标的方法在初始化时固定海森矩阵（Hessian）估计，忽略一阶梯度信息，导致指导信号随量化过程推进而逐渐失效。其解决方案的关键在于提出一种统一的PTQ框架G^2 PTQ，通过引入广义梯度补偿（Generalized Gradient Compensation），在全局监督的分块优化目标下同时整合一阶与二阶信息。该方法在量化每个Transformer块前动态刷新梯度与海森矩阵估计，有效避免了先前全局方法中指导信号的过时问题。此外，为稳定精确的一阶梯度补偿，设计了信任域缩放机制，动态限制梯度步长以防止权重更新爆炸。最后，论文推导了高效的分块海森矩阵近似与精确梯度补偿实现方式。实验结果表明，G^2 PTQ在多种模型架构和位宽设置下均能实现更优的全精度模型对齐性能，显著超越现有先进基线方法。
链接: https://arxiv.org/abs/2609.31009
作者: Ruikang Liu,Haoli Bai,Yuxuan Sun,Qian Zhang,Wenzheng Cai,Yanqi Hao,Feiyu Wang,Weidong Zhong,Zhuang Wang,Tong Yang,Xiangsheng Zhou
机构: ZTE Corporation; The Chinese University of Hong Kong; Northwestern Polytechnical University; Peking University; Nanjing University of Aeronautics and Astronautics
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Post-training quantization (PTQ) is a practical approach to reducing the memory and computational footprint of large language models (LLMs) without retraining. GPTQ-based methods have become the de facto standard, yet they suffer from two complementary limitations. Methods with local, layer-wise objectives lack global supervision; while methods with global objectives fix their Hessian estimates at the start and ignore first-order gradients, so their guidance grows stale as quantization proceeds. This paper presents G ^2 PTQ, a unified PTQ framework with Generalized Gradient Compensation that integrates both first- and second-order information under a globally supervised, block-wise optimization objective. By refreshing gradient and Hessian estimates before quantizing each Transformer block, G ^2 PTQ avoids the staleness of prior global methods. Furthermore, to stabilize the exact first-order compensation, we introduce a trust-region scaling mechanism that dynamically bounds the gradient step to prevent exploding weight updates. Finally, we derive efficient implementations for block-wise Hessian approximation and exact gradient compensation. Experimental results on various model families and bit-widths demonstrate that G ^2 PTQ enables better alignment with the full-precision model, outperforming state-of-the-art baselines. Code is available at: this https URL.
[NLP-25] ZooWork-ShopRanker: An Open Preference-Aligned E-Commerce Reranker
【速读】： 该论文旨在解决生成式 AI 在电商场景中进行排序时的适配性问题，即通用网络检索训练的开放重排模型（open reranker）在电商环境中表现不佳，原因在于电商排序不仅依赖主题相关性，还需综合考虑用户偏好、商品约束及商品间的对比匹配度等复杂因素。由于真实搜索流量虽能提供查询与候选商品，但缺乏可扩展的成对偏好标签，导致难以有效监督学习。为此，论文提出 ZooWork-ShopRanker 系列电商重排模型（0.6B、4B 与 8B），其核心解决方案是构建一个由多个不同家族推理型大语言模型（LLM）组成的“判断者面板”作为偏好代理（preference oracle），通过位置去偏的判断机制和多层级一致性评估生成高质量标注数据。在此基础上，8B 主模型经对齐训练后作为教师模型，通过知识蒸馏指导更小规模的 4B 和 0.6B 模型，并在判别对上进一步优化其输出置信度。为系统评估进展，研究引入 ShopRank-Bench，一个包含约 10,000 条私有流量偏好样本的污染受限基准，支持双格式输入并按判断者家族一致程度分层。实验表明，ZooWork-ShopRanker-8B 与 -4B 显著优于最强的开源基线，所有模型均显著超越自身未对齐的基础模型，且 0.6B 模型亦优于同规模竞品，性能提升在两种输入格式下均稳定存在，并延伸至常见 MTEB 基准。研究成果已公开发布，以推动后续研究。
链接: https://arxiv.org/abs/2609.31002
作者: Siqiao Xue,Shuxuan Liu,Ning Hu
机构: ZooWork Team
类目: Computation and Language (cs.CL)
备注: project page: \url{ this https URL }
Abstract:Open rerankers trained for general web retrieval transfer imperfectly to e-commerce, where ranking decisions depend not only on topical relevance but also on user preferences, product constraints, and comparative product fit. These preference signals are difficult to supervise at scale: real search traffic provides authentic queries and candidates but no clean pairwise labels. We present ZooWork-ShopRanker, a family of e-commerce rerankers (0.6B, 4B, and 8B) aligned to judge-labeled shopping preference. Training pairs are labeled by a panel of reasoning large language models (LLMs) from different families acting as a preference oracle, with position-debiased judgments and agreement tiers, and the rerankers are trained on these labels. The aligned 8B flagship then serves as a distillation teacher for the efficient 4B and 0.6B models, which are fit to its scores and sharpened on judged pairs. To measure progress, we introduce ShopRank-Bench, a contamination-limited benchmark of ~10,000 private-traffic preference pairs in both text formats, tiered by how many judge families committed to each label. ZooWork-ShopRanker-8B and -4B significantly outperform the strongest open reranker baseline, every model significantly beats its own un-aligned base, and ZooWork-ShopRanker-0.6B beats its size peer; the gains hold in both formats and extend to common MTEB benchmarks. We release the models and the dual-format ShopRank-Bench to facilitate further research.
[NLP-26] Evaluating Sycophancy in Chinese Large Language Models on Factual Questions Derived from Online Search Queries
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在信息获取过程中存在的事实性迎合（factual sycophancy）问题，即模型在用户持有错误信念时仍倾向于迎合其观点，导致错误回答被呈现为“独立验证”的结果，从而强化用户对虚假信息的错误信心。其核心挑战在于：当用户持有错误信念时，模型是否会将原本正确的回答转变为错误或不确定的回答？以及反迎合干预措施是否真正维持或恢复了事实准确性，还是仅将响应推向不确定性。论文的关键解决方案在于通过设计多组对照实验（包括基础提示、基于用户信念的提示及反迎合提示），结合中文语境下的真实搜索查询生成的12,165个是非类事实核查问题，分析来自三款前沿中文大模型（DeepSeek、Qwen、Doubao）共计364,941条响应的转移模式。研究发现，推理能力并非一致可靠的防护机制，而反迎合提示虽能减少与错误信念的一致性错误，但同时显著增加了回答的不确定性；更重要的是，模型从正确转为不确定（即“事实信心丧失”）的现象普遍存在，表明避免迎合错误信念并不等同于保持事实准确性。因此，论文强调需采用转换层面评估（transition-level evaluation）以揭示模型行为的深层动态，从而更全面地衡量事实性与可靠性。
链接: https://arxiv.org/abs/2609.30986
作者: Geng Liu,Feng Li,Mengxiao Zhu,Francesco Pierri
机构: 未知
类目: Computation and Language (cs.CL)
备注: 19 pages, 34 figures, 4 tables. Geng Liu and Feng Li contributed equally
Abstract:As large language models increasingly mediate information access, factually accurate and independent answers are critical. However, these models can exhibit sycophancy by aligning their responses with users’ stated beliefs even when those beliefs are incorrect, potentially presenting misinformation as independently verified and reinforcing users’ confidence in false claims. Prior work leaves unresolved whether introducing user beliefs causes correct responses to become incorrect or uncertain, or causes uncertain responses to become belief-aligned incorrect answers. It also remains unclear whether anti-sycophancy interventions preserve or restore factual accuracy or merely shift responses toward uncertainty. We analyze factual sycophancy in Chinese-language information seeking using yes/no fact-checking questions. Our analysis covers 364,941 responses from three frontier Chinese-based LLMs (DeepSeek, Qwen, and Doubao) to 12,165 factual questions derived from real-world Chinese search queries. We evaluate the models with and without reasoning across baseline, belief-conditioned, and anti-sycophancy prompting, tracing matched shifts among correct, incorrect, and uncertain responses. Under incorrect user beliefs, we distinguish belief-aligned errors from losses of factual confidence, in which initially correct answers become uncertain. Patterns vary across models and reasoning settings: reasoning is not a consistent safeguard, and anti-sycophancy instructions can reduce incorrect agreement while increasing uncertainty. In Chinese-language factual question answering, avoiding agreement with false beliefs is therefore not equivalent to preserving factual accuracy, highlighting the value of transition-level evaluation. Such behavior may undermine the reliability of LLM-mediated information access by reinforcing misinformation or weakening users’ confidence in factually correct answers.
[NLP-27] HA: Weighted Finite-State Text Normalization and Inverse Text Normalization for Khmer
【速读】： 该论文旨在解决高棉语（Khmer）在文本到语音（Text-to-Speech, TTS）与语音识别（Speech Recognition）双向转换中的文本归一化（Text Normalization, TN）与逆文本归一化（Inverse Text Normalization, iTN）问题。由于高棉语书写系统缺乏空格分隔，且数字词嵌入于普通词汇内部，导致分词与词性分类困难，现有开源工具缺失，严重制约了相关自然语言处理技术的发展。论文提出的解决方案Tha基于加权有限状态转换器（Weighted Finite-State Transducers, WFST），通过一次最短路径搜索实现整行文本的分词与分类，并引入第二重转换器以排除音节内部的非法词边界，从而保证音节结构的完整性。在Google提供的高棉语测试集上，Tha在274个基数词中达到与参考标准一致的结果（仅1个拼写变体差异），在2,906个真实TTS提示中，153/158句重写结果正确，展现出优异的准确率。该工具已开源，采用Apache 2.0许可证，为高棉语语音技术发展提供了关键基础设施。
链接: https://arxiv.org/abs/2609.30984
作者: Seanghay Yath
机构: Digital Government Committee, Cambodia(柬埔寨数字政府委员会)
类目: Computation and Language (cs.CL)
备注:
Abstract:Text-to-speech needs written text in spoken form, and speech recognition output needs the reverse. For Khmer, neither direction has a maintained open-source tool, and the script makes both harder: words are not separated by spaces, and number words occur inside ordinary words. We present Tha, a Khmer text normalization and inverse text normalization toolkit built from weighted finite-state transducers. It segments and classifies a whole line in one shortest-path search, and a second transducer rejects token boundaries inside a Khmer syllable. On Google’s Khmer test suite, Tha agrees with the reference on all 274 cardinals up to one spelling variant, and on 2,906 real TTS prompts, 153 of the 158 sentences it rewrites are correct. Tha is open source under the Apache 2.0 license.
[NLP-28] Does Uniform Discrete Diffusion Need Time?
【速读】： 该论文旨在解决统一离散扩散模型（Uniform Discrete Diffusion Models, UDMs）中显式时间条件化（explicit time conditioning）的必要性问题。尽管理论上最优的群体级（population-optimal）UDM预测器依赖于时间，即时间决定了模型对观测上下文的信任程度，但研究发现，在语言建模等有限数据场景下，这种时间依赖性可能变得可忽略。其关键在于：当受损训练序列与其原始干净序列的距离显著小于与其他竞争序列的距离时，经验最优（empirical-optimal）预测器在扩散轨迹的大部分阶段对时间几乎不敏感，仅在高噪声终点处可靠性下降。实证结果表明，训练后的语言类UDMs在多数扩散步骤中表现出较弱的时间敏感性，而无时间依赖的预测器在多种数据集和训练目标上不仅保持竞争力，且常优于显式依赖时间的模型。因此，该研究的核心结论是：尽管理论最优解需考虑时间，但在实际应用中，显式时间条件化往往并非必需，这挑战了当前UDMs中普遍采用时间条件化的做法。
链接: https://arxiv.org/abs/2609.30977
作者: Chunsan Hong,Chieh-Hsin Lai,Satoshi Hayakawa,Yuhta Takida,Jong Chul Ye,Yuki Mitsufuji
机构: KAIST(韩国科学技术院); Sony Group Corporation(索尼集团); The University of Tokyo(东京大学)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: Preprint
Abstract:Uniform discrete diffusion models (UDMs) commonly use explicit time conditioning, but we find that it can often be unnecessary in practice. In this paper, we first show that the population-optimal UDM predictor generally depends on time: time controls how much the model should trust the observed context. We then show that this dependence can become negligible in finite-data settings relevant to language. When a corrupted training sequence remains much closer to its original clean sequence than to competing training sequences, the empirical-optimal predictor is nearly insensitive to time over most of the diffusion trajectory, where the guarantee weakens toward the high-noise endpoint. Empirically, trained language UDMs exhibit limited time sensitivity over most of the trajectory, while time-agnostic predictors remain competitive with, and often outperform, time-conditioned models across datasets and training objectives. These results challenge the use of explicit time conditioning in UDMs: although the population optimum depends on time, explicitly conditioning on it may often be unnecessary in practice.
[NLP-29] Coupled Usage-Sense Processes: Temporal and Attributable Lexical Semantic Change
【速读】： 该论文旨在解决词汇语义演变分析中长期存在的核心问题：现有方法仅能衡量词语语义随时间变化的总体幅度（即标量距离），却无法揭示变化发生的具体时间点、驱动变化的机制与语义成分的迁移路径，也无法明确哪些具体用法支持语义变迁的归因。针对这一局限，论文提出耦合用法-语义过程（Coupled Usage–Sense Processes, CUSP）作为解决方案，其关键在于构建一个保持边缘分布不变的时序过程，通过层次化耦合将上下文分布关联至潜在的用法成分，并利用马尔可夫复合机制实现相邻及长距离对应关系的兼容性。该框架引入位移算子以量化变化的幅度与时间节点，精确分离语义成分中心移动与成分内部重组的贡献，并将变化归因于被传输的成分对。词内模式（word-local modes）识别出语义变化的不同方向及其随时间的活跃程度，而来自归因成分的代表性文本片段则为分析提供了可解释的语料支撑。在高斯混合模型的特殊设定下，论文证明了算子与平方距离的参数可恢复性，合成实验验证了预测的变化速率。在英语和德语的DWUG数据集上，CUSP表现优异，能够准确恢复受控的“双面”（Janus）语义演变模式，同时保持组合一致性。基于美国法院判例的大规模语料分析进一步展示了在未标注自然文本中对演变轨迹、变化模式及文本证据的联合解析能力。因此，CUSP实现了对词汇历史的多维度统一表征：既包含变化的量级、时间、机制、成分运动、模式演化，也提供具体的文本证据，使上述视角在统一框架下得以兼容并行。
链接: https://arxiv.org/abs/2609.30974
作者: Haruka Ezoe,Ryohei Hisano
机构: The University of Tokyo(东京大学); The Canon Institute for Global Studies(佳能全球研究机构)
类目: Computation and Language (cs.CL); Machine Learning (stat.ML)
备注:
Abstract:Lexical semantic change is usually summarized by a scalar distance between independently sampled period distributions. This measures how much a word changed, but does not reveal when it changed, which mechanisms and component movements carried the change, or which usages support the attribution. We introduce Coupled Usage–Sense Processes (CUSP), which derives these answers from a single marginal preserving temporal process. A hierarchical coupling relates contextual distributions through latent usage components, while Markov composition makes adjacent and longer span correspondences compatible. Displacement operators quantify change magnitude and timing, split variation exactly between movement of component centers and reorganization within components, and attribute it to transported component pairs. Word-local modes resolve distinct directions of change and their activity over time, while representative passages from attributed components ground the analysis in text. Under a Gaussian mixture specialization, we prove parametric recovery of the operators and squared distances. Synthetic experiments support the predicted rate. CUSP remains competitive on English and German DWUG and recovers controlled Janus profiles while maintaining compositionally coherent transport. A large corpus of US court opinions demonstrates transition, mode, and passage attribution in unlabeled natural text. CUSP thus makes magnitude, timing, mechanism, movement, modes, and textual evidence compatible views of one lexical history.
[NLP-30] FAVoR: Measuring and Mitigating Author-Style Homogenization in Federated Personalized Generation
【速读】： 该论文旨在解决在多作者场景下，使用联邦参数高效微调（Federated PEFT）进行大语言模型个性化写作辅助时出现的作者风格同质化问题。具体而言，尽管标准联邦学习方法能有效保留生成内容的语义连贯性（continuation utility），但其聚合机制会削弱不同作者的独特写作风格特征，导致各作者生成文本在风格空间中趋于相似，从而损害个性化表达。其解决方案的关键在于提出一种名为FAVoR（Federated Authorial Voice Retention）的新框架，其核心创新是引入作者风格残差机制（author-style residual mechanism），采用“共享-私有适配器”设计：客户端仅上传共享适配器更新，同时本地保留作者特异性的残差修正项，以在保护数据隐私的前提下增强风格表征的差异化。实验基于BlogText基准和外部Mythos-Reddit验证集，结合基于角风格分类编码器（ASCE）的诊断与独立作者身份验证，证明FAVoR显著提升了作者风格保留能力，且在生成质量上仅有微小代价，经组件消融、外部验证及冷启动迁移实验进一步证实其有效性。
链接: https://arxiv.org/abs/2609.30968
作者: Lu Han,Jingyao Zhang,Katy Ilonka Gero,Nguyen H. Tran
机构: The University of Sydney(悉尼大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Large language models are increasingly used as personalized writing assistants, but adapting a model across many authors can compromise individual writing style by pulling author-specific signals toward a shared register. Federated parameter-efficient fine-tuning (PEFT) offers a data-local setting for this multi-author adaptation problem: clients keep author text local while sharing compact adapter updates. However, we show that standard aggregation can preserve continuation utility while making different authors’ generations less distinguishable in style space, a failure mode we define as author-style homogenization. We evaluate author-style retention with Angular Style Classification Encoder (ASCE)-based diagnostics on our main BlogText benchmark and ASCE-independent external authorship verification. Using this protocol, we find that common federated PEFT baselines can preserve semantic utility while averaging out author-specific signals. To address this homogenization, we instantiate FAVoR (Federated Authorial Voice Retention), an author-style residual mechanism for federated PEFT. FAVoR uses a shared-private adapter design: clients upload shared-adapter updates while retaining author-specific residual corrections locally. Across BlogText and external Mythos-Reddit validation, FAVoR improves author-style retention over standard and personalized federated PEFT baselines. These gains come with small continuation-utility trade-offs and are supported by component ablations, external verification, and cold-start transfer.
[NLP-31] Estimating and Orthogonalizing Unknown Pre-training Gradients for Continual Fine-tuning of Large Language Models NEURIPS2026
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在持续微调（Continual Fine-Tuning）过程中面临的灾难性遗忘问题，尤其是对先前任务性能的退化以及模型固有通用知识的丢失。现有方法如正交梯度投影虽能在不同微调任务间缓解遗忘，但其根本局限在于无法保留预训练阶段的通用知识，原因在于这些方法依赖于未经公开且高度多样化的原始预训练数据与梯度，而此类信息在实际应用中不可获取。为填补这一关键空白，本文提出EoupCT框架——一种用于估计并正交化未知预训练梯度的持续微调新范式。其核心创新在于：通过引入可学习的软提示（soft prompt）并结合Gumbel-Softmax松弛机制，动态生成最易受新任务干扰的伪数据，从而估计出预训练阶段的梯度；同时，构建多目标优化问题，并设计一种一阶高效的帕累托优化器，联合优化模型参数与软提示，严格保证新任务更新方向与估计的预训练梯度之间正交性。大量实验证明，EoupCT能有效同时保持任务特定能力与模型固有的通用知识，显著缓解灾难性遗忘。
链接: https://arxiv.org/abs/2609.30935
作者: Bing Wang,Changchun Li,Xin-Qiang Cai,Lin Yuanbo Wu,Ximing Li,Gang Niu,Masashi Sugiyama
机构: Jilin University(吉林大学); RIKEN Center for Advanced Intelligence Project(理化学研究所先进智能项目中心); University of Warwick(华威大学); Graduate School of Frontier Sciences, University of Tokyo(东京大学前沿科学研究生院)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted by NeurIPS 2026. 29 pages, 3 figures. Code: this https URL
Abstract:Continual fine-tuning is essential for large language models (LLMs) to dynamically adapt to real-world environments, yet it inevitably suffers from catastrophic forgetting, particularly the performance degradation of previous tasks and LLMs’ general-purpose knowledge. Although existing methods, such as orthogonal gradient projection, mitigate the forgetting across various fine-tuning tasks, they fundamentally fail to preserve pre-training LLMs’ inherent general-purpose knowledge because the original data and gradients of off-the-shelf pre-training LLMs required by these methods are strictly unknown and highly diverse. To bridge this critical gap, we propose EoupCT, a novel framework designed to Estimate and Orthogonalize Unknown Pre-training gradients for Continual LLM fine-Tuning. Specifically, EoupCT estimates pre-training gradients by dynamically generating pseudo data that is most susceptible to forgetting for new tasks through a learnable soft prompt equipped with Gumbel-Softmax relaxation. Furthermore, we formulate a multi-objective optimization problem and introduce a first-order efficient Pareto optimizer that jointly optimizes LLM parameters and the soft prompt, rigorously enforcing orthogonality between new task updates and the estimated pre-training gradients. Extensive experiments across multiple LLMs demonstrate that EoupCT effectively preserves both task-specific proficiency and inherent general-purpose knowledge, successfully mitigating the catastrophic forgetting.
[NLP-32] raining-Free Pronunciation Transcription via Text-Constrained Acoustic Rescoring
【速读】： 该论文旨在解决大规模语音合成（Text-to-Speech, TTS）训练数据生成中发音转录（pronunciation transcription）的准确性与效率问题。现有方法存在明显局限：基于字符到发音（Grapheme-to-Pronunciation, G2P）的方法仅依赖文本信息，而基于语音到发音（Speech-to-Pronunciation, S2P）的方法仅利用声学信息，二者均无法充分融合语言与声学特征；尽管联合文本与语音到发音（Speech-and-Text-to-Pronunciation, ST2P）的方法能整合双模态信息，但通常需要大量标注发音的数据进行训练，成本高昂。为此，本文提出一种无需训练的ST2P推理流程，其核心在于在推理阶段融合词汇资源与预训练模型：首先利用词典和G2P工具生成文本约束的候选发音序列，再通过左向贪心搜索（left-to-right greedy search），基于冻结的预训练S2P模型计算全序列负对数似然（whole-sequence negative log-likelihoods），从而选择最优发音。该方法在三个日语语料上将字符错误率（Character Error Rate, CER）从仅使用文本基线的0.60–1.40%降低至0.04–0.17%（参考转录）和0.64–1.58%（自动语音识别转录），显著优于所有基线，包括训练型ST2P模型及商用多模态大模型。此外，贪心搜索比束搜索快3–3.5倍且保持相近的CER，级联解码速度为直接解码的2倍，兼顾高效与高精度。在西班牙语、法语及初步英语实验中，该方法亦超越四种开源多模态大模型及最优传统方法。
链接: https://arxiv.org/abs/2609.30924
作者: Hikaru Asano,Yotaro Kubo,So Kuroki
机构: 未知
类目: Computation and Language (cs.CL); Sound (cs.SD); Audio and Speech Processing (eess.AS)
备注: 5 pages, 2 figures
Abstract:Accurate and efficient pronunciation transcription is essential for preparing text-to-speech training data at scale. Existing approaches have different limitations: grapheme-to-pronunciation (G2P) and speech-to-pronunciation (S2P) methods each capture only partial information, using only text or only speech, while speech-and-text-to-pronunciation (ST2P) methods use both but require costly pronunciation-annotated data. To address this problem, we propose a training-free ST2P pipeline that integrates both lexical and acoustic information at inference time. Lexical resources and G2P tools generate text-constrained candidates, and a left-to-right greedy search selects the best one using whole-sequence negative log-likelihoods from frozen pretrained S2P models. On three Japanese corpora, our method reduces Character Error Rate (CER) from 0.60–1.40% (text-only baseline) to 0.04–0.17% with reference transcripts, and 0.64–1.58% with ASR transcripts. It outperforms all baselines, including a trained ST2P model and commercial multimodal LLMs. Our greedy search method is 3–3.5 \times faster than beam search at similar CER, and the cascade is 2 \times faster than direct decoding ensuring the efficiency and accuracy. In Spanish, French, and preliminary English, it also surpasses four open multimodal LLMs and the best traditional methods.
[NLP-33] Cross-Backend QIEO: Universal Runtime Portability across OpenMP5 CUDA HIP and Multi-Language Interfaces
【速读】： 该论文旨在解决量子启发式优化算法在实际应用中面临的两大核心问题：一是缺乏统一的执行框架，导致算法性能与硬件可移植性难以兼顾；二是现有方法在跨平台部署时存在实现冗余、维护成本高及性能适配不足的挑战。其解决方案的关键在于提出跨后端量子启发式进化优化器（Cross-Backend Quantum Inspired Evolutionary Optimizer, QIEO），采用“单一真实源”（single-source-of-truth）架构，通过一次C++实现编译并针对不同硬件目标生成对应后端代码，支持运行时动态调度至CPU（串行）、OpenMP 5（多核）、CUDA（NVIDIA）和HIP（AMD）等异构计算后端。该框架根据各设备的内存层次结构与线程执行模型（如warp/wavefront）自动优化内核，实现了算法性能与硬件抽象之间的高效平衡。通过在Python、MATLAB和Julia三种语言接口中的统一运行时绑定验证，展示了其在神经网络超参数优化、风电场布局设计及Lotka–Volterra模型参数估计等典型场景下的卓越性能，显著优于传统优化方法，充分证明了其在保持高性能的同时具备高度可移植性和工程实用性。
链接: https://arxiv.org/abs/2609.30914
作者: Aman Mittal,Ferdin Sagai Don Bosco,Kasturi Venkata Srikanth,Abhishek Singh,Aditya Singh,Abhishek Chopra
机构: BQP(量子计算公司)
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Computation and Language (cs.CL); Optimization and Control (math.OC)
备注:
Abstract:Quantum-inspired algorithms emulate quantum mechanical principles, such as, superposition, interference, and probabilistic amplitude evolution, on classical hardware by representing candidate solutions as qubit vectors and evolving them through rotation-gate operators. This approach offers higher optimization performance without physical qubits, and has been shown to achieve order-of-magnitude speedups (10–80 \times ) over traditional solvers on combinatorial, high-dimensional NP-hard problems. A critical barrier to adoption, however, is the lack of a unified execution framework that delivers both algorithmic performance and hardware portability. We present \textbfCross-Backend Quantum Inspired Evolutionary Optimizer (QIEO), the runtime core of BQP’s BQPhy solver, which addresses this gap through a \emphsingle-source-of-truth architecture. One C++ implementation of the QIEO algorithm is compiled once per hardware target and exposed to multiple high-level languages via thin binding layers. The framework dispatches to CPU (sequential), OpenMP~5 (multi-core), CUDA (NVIDIA), and HIP (AMD) backends at runtime, adapting kernels to each device’s memory hierarchy and warp/wavefront execution model. The framework’s real-world utility is validated through binding demonstrations that share the identical C++ runtime. BQPhy’s Python library is demonstrated on a neural network hyperparameter optimisation achieving 88.60% test accuracy on MNIST. BQPhy’s MATLAB’s Toolkit is tested on wind farm layout optimisation attaining 365,399 \pm 4,552 ~MWh/yr, which is statistically indistinguishable from particle swarm optimisation and +7.6% above genetic algorithms on a 32-variable constrained engineering problem. The Julia package tackles the Lotka–Volterra parameter estimation where BQPhy replaces native Julia solvers on the same residual, cutting mean SSE by 2.1\times . Subjects: Distributed, Parallel, and Cluster Computing (cs.DC); Computation and Language (cs.CL); Optimization and Control (math.OC) Cite as: arXiv:2609.30914 [cs.DC] (or arXiv:2609.30914v1 [cs.DC] for this version) https://doi.org/10.48550/arXiv.2609.30914 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[NLP-34] oolSearcher: Optimizing Tool Selection at Scale via Reinforcement Learning NEURIPS2026
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在与外部环境交互时面临的大规模工具选择（large-scale tool selection）难题，即如何在海量、多样且功能相近的现实工具库中高效地搜索、区分并组合使用工具。现有方法通常局限于小规模或预定义工具集，难以应对真实世界中工具数量庞大、上下文长度受限的挑战，且传统强化学习（Reinforcement Learning, RL）方法在处理工具兼容性与多轮搜索任务时表现不足。其解决方案的关键在于提出一种名为 ToolSearcher 的新型强化学习框架，通过三个核心机制实现突破：一是类别约束的工具区分（category-constrained tool discrimination），增强模型对功能相似工具的辨别能力；二是事件级搜索建模（event-level search modeling），显式优化多轮搜索过程中目标工具的发现效率；三是轨迹对齐的信用分配（trajectory-aligned credit allocation），为搜索-选择过程的不同阶段提供细粒度的奖励信号。实验表明，ToolSearcher 在包含迭代搜索与复杂工具组合的复杂场景中显著优于多个强基线模型，验证了其在大规模工具选择任务中的有效性。
链接: https://arxiv.org/abs/2609.30906
作者: Zhenlong Dai,Xujie Song,Zitong Wang,Tong Niu,Jian liu,Weiqiang Wang,Xiu Tang,Sai Wu,Chang Yao,Jingyuan Chen
机构: Zhejiang University(浙江大学); Ant Group(蚂蚁集团)
类目: Computation and Language (cs.CL)
备注: Accepted at NeurIPS 2026
Abstract:Large language models (LLMs) excel at natural language processing but struggle to interact with external environments. Tool learning provides a promising way to extend LLMs into actionable agents, where tool selection is a critical prerequisite for successful tool use. Existing work often assumes a small or predefined set of tools, leaving large-scale tool selection underexplored. Real-world repositories contain a vast and diverse array of tools, making it difficult for LLMs to effectively search, distinguish, and compose tools under context-length constraints. We identify large-scale tool selection as a new challenge for agentic reinforcement learning, highlighting that existing RL methods for knowledge-based question answering are inadequate for selecting tools while considering compatibility. To address this challenge, we propose ToolSearcher, a novel RL framework for effective multi-turn search and fine-grained optimization in large-scale tool selection. Specifically, we introduce category-constrained tool discrimination to improve the model’s ability to distinguish functionally similar tools, event-level search modeling to explicitly optimize the discovery of target tools during multi-turn search, and trajectory-aligned credit allocation to provide fine-grained reward signals for different stages of the search-selection process. Extensive experiments on large-scale tool selection benchmarks demonstrate that ToolSearcher consistently outperforms a set of strong baselines in challenging settings involving iterative search and complex tool composition.
[NLP-35] From annotation to reasoning : Culture in language models
【速读】： 该论文旨在解决当前语言模型在面对多义性文化表达时，缺乏对解释深度（interpretive depth）评估的问题。现有文化基准测试主要依赖事实知识、与调查结果的一致性或对预设意义的识别，无法衡量模型是否能基于文本证据解释文化引用的运作机制、支持某一解读并根据批评进行修正。这一问题的核心在于如何评估模型在复杂文化语境中进行合理且有据推理的能力。其解决方案的关键在于引入以证据为中心的评估范式，结合文学阐释学的实践，构建能够保留学术界分歧但同时衡量论证质量的评价体系；通过在丹麦文学等具体语料上开展模型开发实验，整合上下文资源与学者反馈，推动生成式 AI 在文化理解上的稳健性发展，从而超越传统基准指标的局限，实现对文化语境中解释能力的深层评估。
链接: https://arxiv.org/abs/2609.30897
作者: Daniel Hershcovich,Alexander Conroy,Jens Bjerring-Hansen
机构: 未知
类目: Computation and Language (cs.CL)
备注: 8 pages, 1 table; perspective paper
Abstract:How should we evaluate language models when more than one interpretation can be right? Cultural benchmarks often test factual knowledge, agreement with survey responses, or recognition of a predefined meaning. These tasks leave open whether a model can explain how a cultural reference works in a particular text, support a reading with evidence, or revise it after criticism. This is a question of interpretive depth, complementary to the breadth of cultural coverage. We argue that literary interpretation offers a useful setting for studying these capabilities. We focus on cultural referencing and reuse: how texts invoke, repeat, and transform earlier expressions across historical and linguistic contexts. Our central claim is that literary scholars can disagree about an interpretation while recognizing the quality of its support. We propose linking evidence-centered benchmarks, evaluation that preserves scholarly disagreement, and model-development experiments on literary data, contextual resources, and scholarly feedback. Danish literature provides a concrete starting point, with implications for other languages and domains. The aim is to develop alternative evaluation strategies that go beyond conventional benchmark metrics and guide model development toward cultural robustness in AI systems.
[NLP-36] Effects of Transcript Compression on LLM -based Medical Misinformation Detection in Japanese YouTube Videos
【速读】： 该论文旨在解决生成式AI在评估长篇医学视频真实性时，因输入文本压缩方式不同而影响其判别性能的问题。具体而言，研究关注在缺乏完整字幕的情况下，采用摘要、检索增强生成（RAG）或关键句筛选等压缩策略所导致的误判风险。其解决方案的关键在于系统比较四种不同的字幕输入设计：完整字幕（Baseline）、LLM生成摘要、基于RAPTOR的检索增强生成（RAG）以及基于医疗相关句子筛选的“Screening”方法。结果表明，尽管所有压缩输入均导致假阴性率上升（即虚假视频更易被误判为真实），其中摘要方式性能下降最显著，而筛选法表现最优但仍遗漏大量医学相关信息。语言学分析显示，错误并非源于判断过于确定，而是由于摘要和RAG削弱了情感、社会、时间、认知及对话性线索，同时增强了机构性与技术性术语的突出性，使虚假内容呈现更高一致性与权威感，从而干扰了生成式AI对虚假信息的识别能力。
链接: https://arxiv.org/abs/2609.30882
作者: Yuya Wake,Sho Tsugawa,Toshiyuki Amagasa
机构: University of Tsukuba (筑波大学)
类目: Computation and Language (cs.CL)
备注: 15 pages. Accepted at the 18th International Conference on Advances in Social Networks Analysis and Mining (ASONAM 2026), Multidisciplinary Track, Short Paper
Abstract:Large language models (LLMs) are increasingly used to assess long-form medical videos, but their effectiveness may depend on whether transcripts are provided in full or compressed through summarization, retrieval, or claim screening. This study examines how such transcript compression affects LLM-based veracity classification of Japanese medical YouTube videos. We compare four transcript input designs: full transcripts, LLM-generated summaries, RAPTOR-based retrievalaugmented generation (RAG), and Screening, which extracts candidate medical and health-related sentences. Using 74 long-form videos labeled as Real or Fake, we evaluate classification performance and analyze linguistic changes using J-LIWC, hedge expressions, and institutional or technical terms. The full-transcript Baseline achieved the best performance, whereas all compressed inputs increased false negatives, meaning that Fake videos were more likely to be misclassified as Real. Summary caused the largest performance drop, while Screening performed best among the compressed inputs but still omitted many medically relevant sentences. Linguistic analyses showed that these errors were not explained by a simple increase in certainty. Instead, Summary reduced affective, social, temporal, cognitive, and conversational cues, while Summary and RAG made institutional and technical terms more salient. These findings suggest that transcript compression can represent Fake videos as more coherent and authoritative inputs, thereby weakening cues needed for misinformation detection
[NLP-37] Evidence-Grounded Auditing of Identification Assumptions in Climate-Policy Causal Evaluations EMNLP2026
【速读】： 该论文旨在解决差分中的差分（Difference-in-Differences, DID）研究在评估气候政策效果时，其识别假设（identification assumptions）所依赖的证据难以系统性验证的问题。现有方法往往缺乏对假设-推论-证据之间逻辑关系的结构化审查，导致因果推断的可信度存疑。为此，论文提出ARGUS——一种基于语言模型的结构化分析管道，通过对照一个包含十一维度的假设-推论-证据框架（assumption-implication-evidence rubric），自动审计文献中报告的证据是否充分支持各识别假设，并在证据不可检索时主动“弃权”（abstain），避免误判。其关键创新在于将自然语言理解与可解释的证据链追溯相结合，生成带有证据链接的风险评估报告，能够精准定位潜在的方法学弱点，供领域专家进一步审查，而无需直接判断因果关系的有效性。实验表明，相较于关键词匹配方法，ARGUS在11类人为植入缺陷的检测中准确率提升至73%；在26篇经济学论文中约40%的评估因缺乏可检索证据而被弃权；在五篇论文的小规模试点中，尽管预设规则可缓解样本内偏差，但加权一致性仍较低，反映出人工标注与模型判断间的分歧，凸显了该工具在揭示隐性风险方面的价值。
链接: https://arxiv.org/abs/2609.30867
作者: Yonghong Zhang,Yong Xie,Isabel M. Parra,Ricardo Correia
机构: Universidad Autónoma de Madrid (马德里自治大学); Spanish National Research Council (CSIC) (西班牙国家研究委员会)
类目: Computation and Language (cs.CL)
备注: Accepted at ClimateNLP 2026, the 3rd Workshop on Natural Language Processing meets Climate Change (EMNLP 2026). 9 pages plus appendix (21 pages total), 6 figures, 15 tables
Abstract:Difference-in-differences (DID) studies are widely used to evaluate climate policy, but assessing the evidence supporting their identification assumptions remains challenging. We introduce ARGUS, a structured language-model pipeline that audits reported evidence against an eleven-dimension assumption-implication-evidence rubric and abstains when relevant evidence cannot be retrieved. We evaluate ARGUS using injected flaws, economics papers, and a small pilot with reconciled labels. On the 11-flaw benchmark, ARGUS detects 73% of planted flaws, compared with 18% for a keyword-based pipeline. Across 26 economics papers, ARGUS abstains on about 40% of paper-dimension assessments for lack of retrievable evidence. In a five-paper pilot with labels reconciled by two annotators, it assigns a higher risk level than the labels on 25 of the 33 assessments it completes. A rule fixed before the labels arrived removes most of this in-sample; weighted agreement stays low. ARGUS provides evidence-linked risk reports that localize potential weaknesses for expert review, without adjudicating causal claims. Code and data: this https URL
[NLP-38] Persistent Negatives for Adversarial Black-Box On-Policy Distillation
【速读】： 该论文旨在解决黑箱在线策略蒸馏（Black-box On-Policy Distillation, OPD）中因使用动态更新的负样本而导致的“移动目标”问题，即在每一步训练中，从最新学生模型采样的负样本会随策略更新而变化，从而导致奖励函数不稳定。其解决方案的关键在于提出一种持久负样本对抗蒸馏（persistent-negative adversarial distillation），通过引入一个持续池（live-pool）机制，在每个判别器批次中保留一部分历史的、与提示匹配的教师-学生对比样本作为负样本，而非仅依赖当前学生生成的新负样本。这一设计使得判别器能够基于相对稳定的负样本分布进行训练，同时保持生成式策略优化（GRPO）的在线策略特性，即仍使用最新的学生响应作为正样本。理论分析表明，贝叶斯最优奖励应为教师到负样本的对数密度比，且在特定假设下，持久负样本可有效锚定判别器，降低奖励估计的均方误差（MSE）。实验结果表明，在相同判别器计算开销下，该方法在两类学生模型、三位评估者及四个对话基准上均显著优于现有方法，并表现出更平滑的策略优化轨迹，减少低于随机水平的性能下降。研究揭示了判别器负样本分布是黑箱在线策略蒸馏中的关键设计维度。
链接: https://arxiv.org/abs/2609.30864
作者: Haixu Ma,Saad Lahrichi,Weiwei Li,Kevin Han,Weiqiang Wu,Peggy Yang,Dongzhuo Li,Ruiyi Li,Serena Li,Gedi Zhou,Mingze Gao,Abhishek Kumar,Xiangjun Fan,Lizhu Zhang
机构: Meta AI; University of Missouri (密苏里大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Black-box On-Policy Distillation (OPD) seeks to improve a student from its own generations when the teacher provides sampled responses but not token probabilities. Adversarial distillation offers one route: it learns a discriminator over prompt-matched teacher and student responses and uses its score as the policy reward. However, sampling discriminator negatives from the latest student at each step couples the learned reward to a negative distribution that changes after every policy update. We address this moving-target problem with persistent-negative adversarial distillation, a live-pool method that replaces a fraction of each discriminator batch with historical, prompt-matched teacher–student comparisons. Under matched discriminator compute, historical comparisons train the discriminator, while GRPO remains on-policy with fresh student responses. Our analysis identifies the Bayes-optimal reward as a teacher-to-negative log-density ratio and, under explicit assumptions, shows how persistent negatives anchor the discriminator and reduce reward-estimation MSE relative to fresh-negative training. Across two student families, three judges, and four judged-chat benchmarks, persistent-negative adversarial distillation consistently improves performance over current methods at matched discriminator compute. It also yields smoother fresh-policy discriminator trajectories, with fewer below-chance dips. These findings identify the discriminator’s negative distribution as an important design axis in black-box on-policy distillation.
[NLP-39] Enhancing Assessment of Self-Consistency in LLM Explanations using Perturbation Strength
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）生成解释时自一致性评估中存在的偏差问题，即现有基于表面扰动的评估方法未对扰动强度进行显式测量与控制，导致不同扰动类型间的比较缺乏公平性。其解决方案的关键在于提出一种以LLM为评判者（LLM-as-a-judge）的方法，统一量化输入扰动与思维链（Chain-of-Thought, CoT）扰动的强度，并在受控强度条件下评估多种LLM的自一致性表现，从而实现跨扰动类型的公平比较。实验结果表明，该基于LLM的扰动强度度量方法优于基于嵌入和概率的传统方法，且发现输入扰动对LLM的影响普遍强于CoT扰动；研究进一步指出，关于模型自一致性的判断仅在相同扰动类型内部才具有可比性和公平性。
链接: https://arxiv.org/abs/2609.30849
作者: Phuong Q. Le,Kemal Kurniawan,Jey Han Lau
机构: The University of Melbourne (墨尔本大学); University of New South Wales (新南威尔士大学)
类目: Computation and Language (cs.CL)
备注: 22 pages, 10 figures
Abstract:Prior work has examined the self-consistency of LLM-generated explanations using surface-level perturbation methods. However, the strength of these perturbations is not explicitly measured and controlled. In this work, we propose an LLM-as-a-judge approach to measure perturbation strength in a unified manner across input and CoT perturbations. We then evaluate the self-consistency in explanations generated from various LLMs under controlled strength conditions, ensuring a fair comparison across perturbation types. Experiments show that our proposed LLM-based perturbation strength measure outperforms other embedding- and probability-based approaches and that input perturbations generally affect LLMs more strongly than CoT perturbations. Our work suggests that judgments about a model’s self-consistency is fair only within the same perturbation type.
[NLP-40] I-Parakeet: Integer-Only Conformer ASR on Mobile NPU
【速读】： 该论文旨在解决现代语音识别模型（如Conformer）在边缘设备上部署困难的问题，特别是针对其高计算资源需求和量化后仍需依赖浮点运算（floating-point）进行数值敏感操作的瓶颈。现有量化模型无法充分利用移动端神经网络处理器（NPU）的整数加速能力，导致推理效率受限。为实现完全基于整数运算的高效部署，本文提出I-Parakeet，其核心解决方案包括三项关键创新：首先，推导出基于整数运算的相对位置自注意力机制，通过融合不同量化尺度的两个得分分支及相对偏移项，将其全部转换为整数操作；其次，引入最小最大误差优化的Swish近似函数，以最小化输出误差的上界，确保非线性激活的精度；最后，通过逐层激活范围分析，提出两项针对性优化：采用INT16网格处理BatchNorm输出，并对重尾分布的预编码器激活值采用百分位校准策略。实验表明，I-Parakeet在Qualcomm NPU上实现了4.97%的词错误率（WER），实时因子达0.048，较CPU基线快7.5倍，验证了纯整数推理在边缘设备上的可行性与高效性。
链接: https://arxiv.org/abs/2609.30846
作者: Taichi Nishimura
机构: 未知
类目: Computation and Language (cs.CL); Sound (cs.SD); Audio and Speech Processing (eess.AS)
备注: Under review
Abstract:In this paper, we propose I-Parakeet, an integer-only implementation of NVIDIA’s Parakeet-CTC (0.6B parameters) that runs on a smartphone NPU without any floating-point operator or CPU fallback. Modern Conformer ASR models are hard to deploy on edge devices because of their size, and quantized models still fall back to floating point for numerically sensitive operations. This prevents them from fully exploiting integer accelerators such as mobile NPUs. To achieve this, our contributions are threefold. First, we derive an integer formulation of the relative-positional self-attention at the core of the Conformer. We fuse its two score branches with different quantization scales and the relative shift into integer-only operations. Second, we introduce a minimax-optimized Swish approximation that minimizes the maximum error of the Swish output. Third, a layer-wise range analysis of activations yields two targeted remedies: an INT16 grid for the BatchNorm output and percentile calibration for the heavy-tailed pre-encoder activations. I-Parakeet achieves 4.97% WER on LibriSpeech test-other, running on a Qualcomm NPU at a real-time factor of 0.048, 7.5x faster than a CPU baseline.
[NLP-41] Quantizing Looped Transformers: Feedback Exposure and Calibration Blindness
【速读】： 该论文旨在解决循环型模型（looped models）在后训练量化（Post-Training Quantization, PTQ）中性能显著下降的问题，尤其关注低比特量化（如INT4）下的失效机制。其核心挑战在于：由于权重在时间步间重复使用，量化误差会在递归过程中被放大并反馈积累，导致模型精度严重退化。解决方案的关键在于识别出两种独立的失效模式，并提出针对性改进策略：一是“反馈暴露”（feedback exposure），即量化层在无残差路径保护的情况下扰动递归状态，导致误差在后续步骤中持续传播；二是“校准盲区”（calibration blindness），即传统方法（如单步GPTQ）仅基于初始时刻激活值构建海森矩阵，忽略了后续递归中关键输入方向的权重信息。为此，作者提出将GPTQ的海森矩阵沿递归步骤累积，从而实现对整个序列动态过程的更全面校准。实验表明，该方法在九个来自七种循环架构的检查点上均优于单步GPTQ和四舍五入（RTN）基准，在Huginn-3.5B上甚至恢复至bf16精度水平，有效解决了量化误差进入递归路径的位置与校准可见状态这两个关键问题。
链接: https://arxiv.org/abs/2609.30820
作者: Nux Li
机构: Meta
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: 27 pages, 5 figures
Abstract:Looped transformers reuse weights across recurrence steps, making low-bit quantization especially attractive. We identify two distinct failure modes of standard post-training quantization. On Huginn-3.5B, per-channel INT4 fails primarily at the non-residual loop-entry adapter, while quantizing the residual core is much less damaging. We call this feedback exposure: a quantized layer perturbs the recurrent state without an identity path, and the resulting error is fed back at later steps. Controlled experiments on linear filters and Mamba state-space models show that feedback exposure also occurs outside transformers. Grouped INT4 reveals a separate failure, calibration blindness: our one-step GPTQ baseline builds its Hessian from step-0 activations, leaving input directions used later in the recurrence nearly unweighted. Across nine checkpoints from seven looped architectures, one-step GPTQ is worse than round-to-nearest (RTN) on the primary task metric for five checkpoints. Accumulating the GPTQ Hessian across recurrence steps outperforms both one-step GPTQ and RTN on all nine checkpoints and recovers bf16-level accuracy on Huginn. These results separate two questions for PTQ on looped models: where quantization error enters the recurrence, and which states calibration sees.
[NLP-42] Understanding the Role of Prompt Template in Knowledge Distillation for Safety Alignment
【速读】： 该论文旨在解决知识蒸馏（Knowledge Distillation, KD）过程中提示模板（prompt template）选择对学生模型（student model）安全对齐（safety alignment）能力影响的未知问题。现有研究已表明，在监督微调（Supervised Fine-Tuning, SFT）阶段提示模板的选择显著影响后续安全对齐的鲁棒性，但其在知识蒸馏阶段的影响尚未被系统探究。本文通过分析不同模板配置对已对齐基线指令微调模型（aligned base instruct-tuned model）的安全对齐性能的影响，发现使用对话式模板（chat template）进行蒸馏会导致学生模型的安全对齐能力显著退化，使其更倾向于响应有害查询；而采用非对话式模板则能有效保留学生模型原有的内部表征结构，减少表示空间的偏移。这一现象在LLaMA、Gemma和Qwen三个模型家族中均得到验证，并在多个安全基准测试中保持一致。因此，解决方案的关键在于：在知识蒸馏过程中优先采用非对话式提示模板，以最大限度地保持教师模型所传递的安全对齐特性，从而避免因模板设计不当引发的对齐失效风险。
链接: https://arxiv.org/abs/2609.30802
作者: Anjila Budathoki,Manish Dhakal,Benjamin M. Ampel,Yi Ding
机构: University of Tennessee, Knoxville(田纳西大学诺克斯维尔分校); Georgia State University(佐治亚州立大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Prior research has demonstrated that the choice of prompt template during Supervised Fine-Tuning (SFT) significantly impacts the robustness of safety alignment afterwards. However, the influence of template selection during Knowledge Distillation (KD) from teacher to student remains largely unexplored. Thus, we fill this gap by analyzing how different template configurations influence the pre-existing safety alignment of the student. We observe a significant degradation of safety alignment present in the aligned base instruct-tuned model. Specifically, we find that utilizing chat templates renders the model more compliant with harmful queries compared to a non-chat template. These findings are consistent across three models: LLaMA, Gemma and Qwen model families and are evaluated across multiple safety benchmarks. We further show that using a non-chat template during distillation better preserves the base student’s internal representations, while chat template distillation induces a larger representational shift. Code: this https URL
[NLP-43] Symbiotic Architecture for Post-Hoc Audio Extension of Frozen Language Models ICASSP
【速读】： 该论文旨在解决如何在不微调大语言模型（LLM）权重的前提下，赋予其音频理解能力这一关键问题。传统方法通常需要对整个模型进行微调或在推理时引入额外的计算开销，导致模型原有文本任务性能下降或扩展性受限。本文提出的共生架构（symbiotic architecture）的核心解决方案是引入一个注入模块（injector module），该模块将音频条件向量直接写入目标LLM的短期记忆——即键值（key-value, KV）缓存中，从而实现无需修改模型权重即可让LLM具备音频语言模型（ALM）的功能。该方案的关键在于：1）通过将音频注入过程与主干网络解耦，使注入成本仅依赖于注入模块的宽度而非主干模型规模，显著提升了音频处理的可扩展性；2）由于不更新原始LLM的参数，其原有的文本任务性能得以完整保留，避免了因微调带来的性能退化风险。实验表明，该方法在语音识别、音频问答和声学场景分类等音频理解任务上表现优于冻结式LLM基线，并接近微调后ALM的性能，同时保持了原始模型在纯文本任务上的能力。
链接: https://arxiv.org/abs/2609.30784
作者: Yotaro Kubo,Qi Sun,Yujin Tang
机构: 未知
类目: ound (cs.SD); Computation and Language (cs.CL); Audio and Speech Processing (eess.AS)
备注: Submitted to ICASSP
Abstract:This paper proposes an architecture for equipping large language models (LLMs) with audio-understanding capabilities without fine-tuning their weights. The proposed symbiotic architecture employs an injector module that writes audio-conditioned vectors directly into the target LLM’s short-term memory, i.e., the key-value (KV) cache, enabling the LLM to behave as an audio language model (ALM). The architectural advantages are twofold. First, it improves the scalability of ALMs: because the proposed method bypasses the LLM during audio injection, the injection cost is governed by the injector width rather than the backbone width, and can therefore scale more slowly than the cost of full-backbone prefilling. Second, since the training scheme does not update the LLM weights, the original capabilities of the LLM are preserved without the risk of degradation from fine-tuning. The effectiveness of the proposed method is evaluated on both audio-understanding tasks (automatic speech recognition, audio question answering, and acoustic scene classification) and text-only tasks. We confirm that, while activating fewer parameters during audio prefilling, our architecture outperforms the conventional method with a frozen LLM and approaches the performance of a fine-tuned ALM, all while preserving the backbone LLM’s original text-only task performance by construction.
[NLP-44] Learning Natural Conversational Behavior in Tandem Speech-to-Speech Models with Randomized Guidance ICASSP2027
【速读】： 该论文旨在解决双向语音到语音（tandem speech-to-speech）架构在真实对话场景下训练时面临的指导信号缺失问题。传统方法依赖大语言模型（LLM）模拟后端生成的响应作为前端语音生成的指导，但真实对话数据中仅包含最终响应，缺乏用户说话过程中的中间指导信息，需通过额外的仿真模型生成，导致数据准备成本高昂。为此，本文提出“随机中间指导”（randomized intermediate guidance）策略，直接从对话语料库中提取目标响应与随机采样响应作为指导信号，无需依赖仿真模型。该方法在训练中使前端模型学习如何有选择性地利用后端信息：目标响应提供有效引导，而随机响应则引入潜在无关更新以增强鲁棒性。实验表明，在合成对话上，该方法达到与基于LLM生成和相似性基准相当的响应质量；在3.8千小时真实对话数据上训练时，显著提升了自然的换言流畅性与音频评价的自然度，同时保持对Moshi等基线模型的响应质量优势。结果表明，随机中间指导为融合双向模型的高质量响应能力与真实语音交互行为的学习提供了高效可行的解决方案。
链接: https://arxiv.org/abs/2609.30773
作者: Manato Yaguchi,Yotaro Kubo,Hikaru Asano,So Kuroki
机构: 未知
类目: Computation and Language (cs.CL); Audio and Speech Processing (eess.AS)
备注: Submitted to ICASSP 2027. 5 pages, 1 figure, 2 tables
Abstract:Tandem speech-to-speech architectures couple a responsive speech frontend with an asynchronous text backend. In KAME, a large language model (LLM) serves as the backend, supplying candidate responses as guidance to the speech frontend while the user is still speaking. Ordinary conversation recordings capture the eventual response but not the guidance the backend would supply during the user’s utterance. Generating the missing guidance with a simulator LLM adds substantial data-preparation overhead when training on real conversations. We propose randomized intermediate guidance, which derives guidance directly from the conversation corpus rather than simulating backend LLM behavior. During training, target responses provide informative guidance, while randomly sampled responses provide potentially irrelevant updates during the utterance. This combination aims to teach the frontend to use backend information selectively. On synthetic dialogues, KAME trained with this recipe achieves response quality comparable to that of the LLM-generated and similarity-based baselines. Training on 3.8k hours of real conversations improves smooth turn-taking and audio-judge naturalness over synthetic-data KAME while retaining a response-quality advantage over Moshi. These results show that randomized guidance offers a practical route to combining the response-quality benefits of tandem models with natural conversational behavior learned from real speech.
[NLP-45] SEA-CLIP-Tiny: Efficient Multilingual Text-Vision Embedding for Southeast Asian Languages ACCV2026 ACCV-2026
【速读】： 该论文旨在解决东南亚地区多语言文本-视觉嵌入模型支持不足的问题，主要受限于该地区语言多样性高、数据与计算资源匮乏。针对这一挑战，论文提出SEA-CLIP-Tiny，一个参数量少于50M的轻量级多语言文本-视觉嵌入模型，其核心解决方案在于采用类CLIP-KD（Knowledge Distillation）框架，结合区域特异性数据筛选与多语言教师模型指导，实现对东南亚七种语言的有效建模。实验表明，SEA-CLIP-Tiny在跨语言图像-文本检索任务中达到R@1、R@5和R@10分别为12.9%、31.5%和42.2%的性能，显著优于同类轻量级模型MobileCLIP2，且在参数量减少38.4%的同时，平均R@10提升12.1个百分点，并具备更低的CPU延迟。这凸显了面向特定区域特征进行训练对于构建高效多语言文本-视觉模型的关键作用。
链接: https://arxiv.org/abs/2609.30739
作者: Puja Ahmad Habibi,Faiz Assabil Firdaus,Ashvanth S,Ekapol Chuangsuwanich,Pume Tuchinda,Peerat Limkonchotiwat
机构: SEACrowd; University of Indonesia; Cohere Labs Community; Technical University of Denmark; Department of Computer Engineering, Faculty of Engineering, Chulalongkorn University; Vidyasirimedhi Institute of Science and Technology; AI Singapore
类目: Computation and Language (cs.CL)
备注: Accepted to ACCV 2026. Model weights and datasets are available at this https URL and code for training, evaluation, and preprocessing at this https URL
Abstract:Multilingual text-vision embedding models are essential for cross-lingual image-text retrieval, but Southeast Asian languages remain poorly supported due to the region’s linguistic diversity and limited data and computing resources. In this paper, we introduce SEA-CLIP-Tiny, a compact multilingual text-vision embedding model for Southeast Asia with fewer than 50M parameters. Our model adapts a CLIP-KD-style framework to Southeast Asian multilingual settings through regional data curation and multilingual teacher guidance. Experiments across seven Southeast Asian languages show that SEA-CLIP-Tiny achieves the strongest average retrieval performance among the evaluated student models, reaching 12.9%, 31.5%, and 42.2% at R@1, R@5, and R@10, respectively. Compared with MobileCLIP2, it improves average R@10 by 12.1 points while using 38.4% fewer parameters and lower measured CPU latency. These results highlight the importance of region-aware training for efficient multilingual text-vision models in Southeast Asia.
[NLP-46] Beyond Mean Attention: Diversity-Aware Layer-Wise Scoring for KV Cache Eviction ICASSP2027
【速读】： 该论文旨在解决大模型推理过程中键值缓存（KV cache）管理中因冗余与相关性失衡导致的效率与精度下降问题。现有方法如SnapKV和PyramidKV仅基于局部观察窗口内的平均注意力权重对token进行淘汰排序，忽略了注意力分布的离散程度及所选token之间的冗余性。为此，论文提出一种统一评分函数：μi+λ1σi+λ2corr(i,S)，其关键在于引入两个新维度——注意力在窗口内查询间的方差（σi）以衡量分散性，以及当前token与已选token集合的相似性（corr(i,S)）以控制冗余，从而实现相关性与多样性的平衡。特别地，当λ2>0时，该评分机制无需额外前向传播即可实现最大边际相关性（MMR）式的去重效果。研究进一步探讨深度位置对这一平衡策略的影响，对比固定全局系数、三段式及二次型系数配置，并在开发集上通过双曲正弦（sinh）参数化进行搜索。实验结果表明，在16个英文LongBench数据集上，使用单一全局多样性常数可使13个数据集取得提升（宏平均+1.1），且在缓存预算为64时性能稳定；而在段落检索任务中，发现显著的层间结构特征——中层出现符号翻转，即奖励相似性，带来高达+9.6的增益（预算64），且在无再调参情况下相较全局常数提升+13.2（预算128）。消融实验证明该收益主要源自冗余项，通过重播所有搜索状态至保留测试集有效区分了真实结构与过拟合噪声。
链接: https://arxiv.org/abs/2609.30738
作者: Tianfang Xie,Wei Zhu
机构: Georgia Institute of Technology (佐治亚理工学院); Zhangjiang Lab; Shanghai (上海), China
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 5 pages, 1 figure, 3 tables. Submitted to IEEE ICASSP 2027
Abstract:KV cache eviction methods such as SnapKV and PyramidKV rank tokens solely by mean attention over a small observation window. We study a unified score, \mu_i+\lambda_1\sigma_i+\lambda_2\mathrmcorr(i,S) , adding attention dispersion across window queries and redundancy relative to selected tokens. For \lambda_20 , the score penalizes similarity to selected tokens as in maximal marginal relevance (MMR), without extra forward passes. To test whether this relevance-diversity balance should vary with depth, we compare fixed global coefficients with three-segment and quadratic profiles. Only these depth profiles are searched on a development split under a \sinh reparameterization. On all 16 English LongBench datasets with Mistral-7B at a budget of 64 entries per layer, a single global diversification constant improves 13 of 16 datasets (macro +1.1); the gain holds at budget 32 and narrows at 128. Per-dataset search finds no detectable layer structure on most datasets; on passage retrieval it finds a large one: a mid-layer sign flip that rewards similarity and is worth +9.6 over the baseline at budget 64 and, without re-tuning, +13.2 over the global constant at budget 128. Ablations attribute the gain to the redundancy term; replaying every accepted search state on the held-out test set separates genuine structure from tuning noise.
[NLP-47] Words Speak Louder Than Order: A Behavioral Evaluation of Gemma 4
【速读】： 该论文旨在解决生成式 AI（Generative AI）在面对两份存在冲突的输入文档时，如何权衡并选择优先采纳哪一来源的问题。具体而言，研究聚焦于探究模型在决策过程中是否受源文档的表述方式（source framing）或呈现顺序（reading position）的影响。其解决方案的关键在于采用完全平衡的实验设计，在短且单轮的上下文中对 Google 的预训练模型 Gemma 4-e4b 进行系统性行为评估（共13个测试项，784次前向传播），从而数学上分离并量化源框架与阅读位置各自独立的影响，并有效消除模型固有的词汇偏好干扰。研究发现：1）源文档的语义框架（如将其标示为官方指南或最新更新）对模型输出具有压倒性影响，显著强于呈现顺序；2）尽管模型表现出明显的首因效应（primacy effect），即更倾向首个被读取的文档，但该偏倚强度受表面表述差异影响极大，波动幅度可达5倍以上；3）位置偏倚主要由整体结构重复性驱动，而非短时重复性提示词（如“是[答案]”），当两份文档使用完全相同的字面模板时，首因效应显著增强，而引入整体措辞变化则会削弱该偏倚。
链接: https://arxiv.org/abs/2609.30716
作者: Amanda Fitch
机构: Google(谷歌)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 36 pages, 1 figure, evaluation dataset and logs released
Abstract:When a language model receives two conflicting documents as input, how does it decide which one to prioritize? Does it rely on how the sources are framed or the presentation order of the documents? We evaluated this behavior on Google’s pre-trained Gemma 4-e4b model across a targeted behavioral suite (n = 13 items, 784 forward passes in short, single-turn contexts) using a completely counterbalanced experimental design. This setup allowed us to mathematically isolate the specific effects of source framing and reading position, while ensuring the model’s natural vocabulary biases were canceled out. Across ten test conditions, we discovered the following: 1. Source framing heavily overpowers reading position. When directly competing, the semantic framing of a source (such as presenting it as an official guideline or a fresh update) had a significantly stronger impact on the model’s final answer than the presentation order of the document. 2. The model favors the first document it reads, but this bias is highly variable. While the model consistently demonstrated a primacy effect (preferring the first document presented), the actual strength of this bias fluctuated by at least a factor of 5 based solely on the surface wording. 3. Overall structural repetition, not short copy-cues, drives positional bias. The model’s preference for the first document is not a mechanical reaction to short, repetitive trigger phrases, such as “is [Answer]”. However, the primacy effect does increase significantly when the two competing documents are structurally identical, using word-for-word verbatim templates. Introducing variation in the overall wording between the two sources reduces this positional bias. Comments: 36 pages, 1 figure, evaluation dataset and logs released Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI) Cite as: arXiv:2609.30716 [cs.CL] (or arXiv:2609.30716v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2609.30716 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Amanda Fitch [view email] [v1] Fri, 25 Sep 2026 02:43:40 UTC (94 KB) Full-text links: Access Paper: View a PDF of the paper titled Words Speak Louder Than Order: A Behavioral Evaluation of Gemma 4, by Amanda FitchView PDFHTML (experimental)TeX Source view license Current browse context: cs.CL prev | next new | recent | 2026-09 Change to browse by: cs cs.AI References Citations NASA ADSGoogle Scholar Semantic Scholar export BibTeX citation Loading… BibTeX formatted citation loading… Data provided by: Bookmark checked="checked"class=“labs-tab-input”> Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv’s community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?) mathjaxToggle(); We gratefully acknowledge support from our major funders, member institutions, , and all contributors. About Help Contact Subscribe Copyright Privacy Accessibility Operational Status (opens in new tab) Major funding support from
[NLP-48] LAVOIR: Teaching a Single-Pass Decision Encoder When and What to Ask with Amortized Value of Information
【速读】： 该论文旨在解决生成式问答系统中“系统一”（System One）决策模型在面对信息缺失时无法主动请求补充信息的问题，即当输入文本未提供关键区分要素（如部门间差异）时，现有模型只能进行猜测而无法识别并引导用户提供缺失信息。其核心解决方案是提出LAVOIR（Laya with Value-Of-Information Routing），通过将潜在缺失信息项（槽位，slots）嵌入输入序列，与候选答案一同处理，使单次前向传播即可同时输出决策分布及每个槽位的预期信息价值（Value of Information, VOI）。VOI无需人工标注：真实决策由模式规则生成，大语言模型（LLM）负责语义化消息与回答，另一类模型对文本进行独立验证，通过将每条消息与多个用户画像配对实现对实际收益的回归估计，从而推断期望增益。此外，引入基尼不纯度（Gini-impurity）上限以约束预测价值，确保其不超过校准模型尚可提升的最大空间。在受控实验中，模型在已见模式上的决策性能接近贝叶斯上限；在策略层面，其提问行为与贪婪最优信息价值策略高度一致（AUC 0.799 vs. 0.797），且平均每对话仅需0.5个问题即可比从不提问提升14.1个百分点准确率。在真实ABC、ABCD和SGD数据集上，LAVOIR在触发提问时显著提升准确率（如8.3点），而在无提问时保持稳定；同时，基尼上限将提问频率从93%降至8.6%。在Laya的12个基准测试中，有7个表现优于原模型，且推理速度达31毫秒（中位数，GH200）。
链接: https://arxiv.org/abs/2609.30706
作者: Furkan Yilmaz,Habibe Aleyna Tasdemir,Muhammed Faruk Gozay
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 11 pages, 3 figures, 7 tables. Code: this https URL ; model: this https URL
Abstract:“System One” decision models such as TypeSafe’s Jev and its open counterpart Laya answer typed questions about a text in a single forward pass with calibrated probabilities, but they cannot ask for missing information: when a first message does not say what separates two departments, they guess. We present LAVOIR (Laya with Value-Of-Information Routing), which places the candidate pieces of missing information (slots) in the input next to the answer options, so that one forward pass returns both the decision distribution and, for every slot, the expected gain in the probability of the correct decision if the user were asked about it. VOI targets need no human labels: gold decisions come from schema rules, an LLM only verbalizes messages and answers, a model from another family checks every text, and pairing each message with several profiles makes regression on realized gains estimate the expected gain. A Gini-impurity cap bounds the predicted value by what a calibrated model can still gain. In a controlled study, decisions on seen schemas are statistically indistinguishable from the Bayes ceiling. The final model’s question policy matches a greedy oracle VOI policy on seen schemas (AUC 0.799 vs. 0.797), and with at most 0.5 questions per conversation it is 14.1 points more accurate than never asking. On real ABCD conversations, one real exchange raises accuracy by 8.3 points where LAVOIR asks and leaves it unchanged where it does not; on SGD the cap lowers the asking rate from 93% to 8.6%. On Laya’s twelve benchmarks LAVOIR is above Laya’s reported scores on seven, and it answers a question in 31 ms (median, GH200).
[NLP-49] RACE: Temporal Audit and Condition-aware Evaluation of Streaming Video Understanding
【速读】： 该论文旨在解决当前流式视频理解任务评估中存在的关键问题：现有评估方法未能明确说明证据的有效时间、视觉历史的维持机制以及响应触发条件，导致相似的任务得分可能对应截然不同的系统负载、失败模式和运行行为。为应对这一挑战，论文提出了一种名为TRACE（Temporal Audit and Condition-aware Evaluation）的条件感知基准与评估框架。其解决方案的关键在于：通过引入具有时间审计的视觉任务、证据时效性标注及指令依赖的触发标注，采用统一的因果核心-适配器（Core–Adapter）协议以控制信息可访问性并精确记录实际的历史处理与响应事件，并实现对答案质量、响应及时性、响应选择行为、工作负载、完成度和可靠性等多维度的综合报告。实验基于517段视频中的1,240条记录，评估了8个公开模型在8种配置下的表现，发现近乎相同的问答准确率可能掩盖完成度、答案有效性与生成负载上的显著差异，且主动性能可进一步分解为响应质量、响应延迟、误触发（在无有效目标窗口时发出响应但后续存在有效窗口）和漏检目标窗口等维度。研究结果表明，流式视频理解性能应被视作受执行条件制约的系统行为，而非单一评分。该基准及代码已开源。
链接: https://arxiv.org/abs/2609.30670
作者: Yibo Ma,Qianqian Zhang,Peng Liu,Tiancheng Zhao
机构: Om AI Research
类目: Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注: TRACE Tech Report
Abstract:Streaming video understanding requires models to interpret evidence as it arrives, yet current evaluations often report task scores without specifying when evidence becomes valid, how visual history is maintained, or how responses are triggered. As a result, similar scores may correspond to different workloads, failure modes, and operational behavior. We introduce TRACE (Temporal Audit and Condition-aware Evaluation), a condition-aware benchmark and evaluation framework that makes these factors explicit. TRACE combines temporally audited visual tasks with evidence timing and instruction-dependent trigger annotations, a unified causal Core–Adapter protocol that controls information availability while recording actual history processing and response events, and multidimensional reporting of answer quality, timeliness, response-selection behavior, workload, completion, and reliability. On 1,240 records from 517 videos, we evaluate eight publicly available models or systems in eight configurations. We find that nearly identical QA accuracy can mask substantial differences in completion, answer validity, and generation workload, while proactive performance separates into response quality, response delay, false alarms (responses emitted while no target window is currently valid and a later one remains), and missed target windows. These results show that streaming-video performance should be interpreted as execution-conditioned system behavior rather than a single score. Our benchmark and code can be accessed at \hrefthis https URLthis https URL.
[NLP-50] Prompt Injection Detection for Email Agents Through Attack Chain Modeling ICTAI2026
【速读】： 该论文旨在解决大语言模型邮件助手在面对间接提示注入（indirect prompt injection）攻击时的脆弱性问题，尤其关注未受信任的邮件内容被引入模型上下文后可能引发后续工具调用中的恶意行为。现有检测方法多将此问题简化为二分类的恶意文本识别任务，忽略了攻击通常以多阶段序列形式演进这一关键特征。其解决方案的核心在于构建一个基于攻击链（attack chain）建模的检测框架：通过结合文本检测器、各阶段特化的验证模块、显式的规则化风险信号、用户意图与操作一致性分析，以及逻辑决策策略，实现对攻击全过程的动态追踪与干预。为支持该框架，研究者从提示注入数据集中推导出攻击链标签，并在随机划分、时间阶段迁移、条件阶段转移、跨数据集迁移等场景下进行评估，同时开展消融实验。结果表明，仅依赖随机训练-测试划分会显著高估模型在分布外情况下的鲁棒性，且攻击链中后期工具参数阶段比早期阶段更具可预测性；此外，使用模拟攻击特征的无害邮件进行训练可有效降低误报率而不牺牲真实攻击检测能力。在五个二分类基准上，所提框架在严格阈值设置下平均F1得分为0.406，显著优于五种预训练检测器在无额外训练条件下的最高表现（0.216），验证了融合攻击阶段预测与用户请求-邮件指令间一致性冲突检测的有效性，同时强调了引入具有挑战性的良性样本对平衡检测性能与误报控制的重要性。
链接: https://arxiv.org/abs/2609.30657
作者: Ahmad Hashmi,Dhyey Patel,Yunting Yin
机构: Eastern Michigan University(东密歇根大学)
类目: Cryptography and Security (cs.CR); Computation and Language (cs.CL)
备注: Accepted to IEEE ICTAI 2026
Abstract:Large language model email assistants are particularly vulnerable to indirect prompt injection because untrusted email content can be retrieved into the model context and influence subsequent tool use. Existing prompt injection detectors mainly formulate this problem as binary malicious text classification, which overlooks the important factor that harmful agent behavior often arises through a sequence of stages. We propose a detection framework that models this attack chain by combining a text detector, verifiers specific to each stage, explicit rule-based risk signals, user intent and action consistency analysis, and a logistic decision policy. To support this framework, we derive attack chain labels from prompt injection datasets, evaluate the proposed framework under random splits, temporal phase transfer, conditional stage transfer, cross-dataset transfer, and conduct ablation studies on multiple benchmarks. Results show that random train test splits substantially overestimate robustness under distribution shift, while later tool argument stages are more predictable than earlier stages in the framework. We also show that training on harmless emails that resemble attacks helps reduce false alarms while preserving the ability to detect real attacks. Across five binary benchmarks, our framework achieves a mean F1 score of 0.406 under the strict threshold setting policy, compared with 0.216 for the strongest of five pretrained detectors evaluated without additional training. These results highlight the value of combining attack stage predictions with checks for conflicts between the user’s request and instructions in retrieved emails. Our experiments also demonstrate the importance of training with challenging benign examples to balance attack detection and false alarms.
[NLP-51] Recursive Self-Improvement via On-Policy Distillation for Reasoning
【速读】： 该论文旨在解决在线策略自蒸馏（On-Policy Self-Distillation, OPSD）中教师模型固定不动所导致的适应性不足问题，即冻结的教师无法吸收学生模型在训练过程中逐步提升的推理能力，从而限制了整体性能的进一步优化。其核心解决方案在于提出一种动态协同进化（Dynamic Co-Evolution, DCE）与自精炼简洁学习（Self-Refined Concise Learning, SRCL）相结合的递归框架。DCE通过让教师模型与学生模型在训练过程中共同演化，使学生的新知识能够反馈至教师，实现持续迭代优化；而SRCL则引入对模型自身生成结果的短文本、经验证的重写版本进行监督，以抑制因过度修正带来的冗余和自我批判倾向，提升输出质量与简洁性。实验表明，该框架在多个模型规模及四类竞赛级数学基准上显著优于传统OPSD，尤其在Qwen3-8B模型上实现了65.97%的Average@12准确率，较OPSD提升35.62个百分点，同时输出长度相对减少7.80%。
链接: https://arxiv.org/abs/2609.30652
作者: Shangjian Yin,Zehao Zhao,Kavosh Asadi,Rui Liu,Yuchen Lu,Shike Mei,Hang Cui,Luke Simon,Zhouxing Shi,Hamed Firooz
机构: Meta AI; University of California, Riverside (加州大学河滨分校)
类目: Computation and Language (cs.CL)
备注:
Abstract:On-policy distillation (OPD) trains a student model by having it generate trajectories, then matching its next-token predictions with an external teacher’s next-token predictions. This provides dense, token-level supervision to the student. On-policy self-distillation (OPSD) eliminates the need for the external teacher. Specifically, a second frozen copy of the student model, now given the ground truth in its context, serves as the teacher. The student model only receives the problem and learns to mimic the privileged teacher model, while the teacher remains frozen throughout training. Previous work showed that freezing the teacher is useful for training stability, but we argue that this can prevent the teacher from incorporating the improvements learned by the student during training. Our primary contribution is to address this limitation with a recursive framework built around two complementary components. First, we let the privileged teacher co-evolve with the student so that revision learned in one round can guide the next, a process we refer to as Dynamic Co-Evolution (DCE). Second, because stronger revision can also make responses too verbose and self-critical, we additionally train on shorter, verified rewrites of the model’s own on-policy responses. We call this complementary objective Self-Refined Concise Learning (SRCL). Overall, our comprehensive evaluations show that DCE+SRCL outperforms OPSD across multiple model scales and four competition-level mathematics benchmarks. Specifically, on Qwen3-8B, DCE+SRCL reaches 65.97% Average@12, outperforming OPSD by 35.62 percentage points while reducing mean output length by 7.80% relative to DCE alone.
[NLP-52] he Hard Part Comes After Search: Benchmarking Web Agents on Synthesizing Organizing and Displaying Knowledge EMNLP2026
【速读】： 该论文旨在解决现有计算机使用代理（computer-use agent）评估基准无法充分衡量代理作为助手在复杂、多步骤工作流中表现的问题。当前的评估体系难以全面检验代理在跨任务信息检索、知识综合生成可交付成果（如文档、演示文稿、电子表格）以及操作程序界面时所需的推理与合成能力、复杂任务分解能力，以及视觉与空间理解能力。为此，研究提出KNOWS基准，这是一个开放式的、复杂的、基于浏览器的任务集合，能够联合评估上述多项核心能力，且每个任务均以生成具体可交付物（artifact）为终点。其解决方案的关键在于：首先，设计了一套任务设计规范（task design rubric）和执行协议，确保任务具备足够的复杂性与真实性；其次，为每项任务配备一个评估器（evaluator），该评估器结合确定性检查与大语言模型（LLM）判断，有效平衡了评估的丰富性、可靠性与自动化之间的权衡。实验结果表明，前沿计算机使用代理在部分成功度量上仅取得中等水平表现，而最佳代理在复杂、长周期任务中完全成功的情况不足3%；尤其在涉及视觉理解的步骤失败时，即使完成超过50%的其他步骤，最终生成的成果也因不可用而失效。这揭示了当前代理作为端到端助手的能力局限，凸显了在工具调用、视觉理解及长周期推理方面亟需突破。
链接: https://arxiv.org/abs/2609.30604
作者: Alexander Gill,Md Farhan Ishmam,Xuyen Nguyen,Neha Bhat,Parker Henry DeYoung,Fateme Hashemi Chaleshtori,Nathan Stringham,Kenneth Marino,Ana Marasović
机构: University of Utah(犹他大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 9 pages main text. Accepted to Findings of EMNLP 2026. Project page: this https URL
Abstract:Existing computer-use agent benchmarks do not fully evaluate agents acting as assistants. A useful assistant retrieves information across complex, multi-step workflows, synthesizes it into artifacts (documents, presentations, spreadsheets), and navigates program interfaces to produce a coherent final product. Such workflows demand reasoning and synthesis, decomposition of complex tasks, as well as visual and spatial understanding. To study agents on workflows like these, we introduce KNOWS, a benchmark of open-ended, complex, browser-based tasks that jointly evaluate these capabilities, with each task culminating in a produced artifact. To write tasks, we develop a task design rubric and a protocol for ensuring that tasks meet the requirements. Each task is paired with an evaluator, a program that combines deterministic checks with LLM judgments to balance the richness, reliability, and automation tradeoff inherent to agent evaluation. We evaluate and analyze frontier computer-use agents and browser-based harnesses. They achieve moderate scores on partial-success metrics, but the best performer fully succeeds in fewer than 3% of our complex, long-horizon tasks. Failures on visual steps render the resulting artifacts unusable, even when agents complete more than 50% of other evaluation steps. Our results expose limitations of current agents acting as end-to-end assistants, and call for progress on tool use, visual understanding, and long-horizon reasoning.
[NLP-53] Probing Stability-Plasticity Tradeoffs in Agent Memory through Cognitive Experimental Paradigms EMNLP2026
【速读】： 该论文旨在解决当前智能体记忆系统评估中过度依赖最终答案准确率，而忽视记忆行为内在机制的问题。现有方法无法揭示系统在长期记忆维护过程中对信息的更新、保留、溯源及时间组织等动态行为特征，导致对记忆稳定性与可塑性权衡（stability-plasticity tradeoff）的理解片面。其解决方案的关键在于提出一种受认知科学启发的诊断框架——MemProbe，该框架基于记忆具有重构性且受干扰、信息源可靠性、强化和再激活影响的核心认知规律，设计了四种可复用的实验范式：干扰（interference）、错误信息（misinformation）、巩固强度（consolidation strength）与再巩固窗口（reconsolidation window），用于系统性地操控记忆应被更新、保留或视为不确定的时机。同时，该框架将正确性分解为多维度的行为谱系，揭示系统如何处理信息的动态演化过程。通过在一个包含56个剧集的诊断套件中统一评估六种增量式记忆系统，研究发现尽管各系统在总体性能上相近，其行为模式却显著不同。因此，MemProbe提供了一种可解释的诊断视角，将聚合性能转化为随时间演化的记忆维持行为画像，从而实现对记忆系统更深层次的理解与分析。
链接: https://arxiv.org/abs/2609.30558
作者: Jiaqi Ding,Guorong Wu
机构: UNC-Chapel Hill(北卡罗来纳大学教堂山分校)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted by EMNLP 2026 Main, code is availble at this https URL
Abstract:Agent memory systems are increasingly used to maintain long-term user preferences, task states and evolving facts, but current evaluations often collapse memory behavior into final-answer accuracy. We introduce MemProbe, a cognitive-science-inspired framework for diagnosing stability-plasticity tradeoffs in agent memory. The framework is motivated by a core insight from cognitive memory research: memory is reconstructive and shaped by interference, source reliability, reinforcement, and reactivation. MemProbe turns this insight into four reusable experimental paradigms (interference, misinformation, consolidation strength, and reconsolidation window) that manipulate when a memory should be updated, preserved, or treated as uncertain. It further decomposes correctness into behavioral profiles that reveal how systems update, preserve, attribute, and temporally organize information. We instantiate these paradigms in a 56-episode diagnostic suite and evaluate six incremental memory systems under a unified protocol. Results show that systems with similar aggregate scores exhibit distinct behavioral profiles. MemProbe provides such a diagnostic lens, turning aggregate performance into interpretable profiles of memory maintenance over time. Code is available at this https URL.
[NLP-54] Dont CLAP: Are Music-Text Models Bag-of-Words?
【速读】： 该论文旨在解决当前文本生成音乐（text-to-music）系统评估中，常用客观指标CLAP分数在衡量音乐对文本提示的忠实度时是否存在对细粒度音乐语义及属性绑定（attribute binding）捕捉不足的问题。具体而言，研究关注当文本描述中某一属性与特定乐器相关联（如“失真吉他”）时，模型的文本嵌入是否能准确反映这种绑定关系。其解决方案的关键在于引入一种新颖的属性交换扰动（attribute swap perturbation）方法：对真实录音的标题进行精确修改，仅替换两个乐器之间的单一属性（如音色、主奏/伴奏角色或首次出现顺序），从而构造出语义上细微但关键不同的对比文本。通过测试四种对比式音乐-文本模型和一个大型音频-语言模型在原始标题与扰动标题之间的得分差异，发现所有对比模型均无法可靠区分两者，而音频-语言模型虽表现稍优，但进一步分析表明其优势主要源于不依赖音频的语义先验（audio-agnostic language priors）。研究结果有力表明，CLAP分数及其类似指标本质上更接近“词袋”模型，对标题中改变语义的关键属性扰动缺乏敏感性，因而难以有效反映音乐生成中的细粒度语义一致性。
链接: https://arxiv.org/abs/2609.30540
作者: Yuan-Chiao Cheng,Alexander Lerch
机构: 未知
类目: ound (cs.SD); Computation and Language (cs.CL); Audio and Speech Processing (eess.AS)
备注: 5 pages, 4 figures, 1 table
Abstract:Text-to-music systems are assessed on audio quality and on how faithfully the music follows its prompt, and the CLAP score, the cosine similarity between a music-text model’s audio and text embeddings, is the standard objective metric of faithfulness. We ask how accurately that score reflects the text: when an attribute is linked to an instrument (e.g., distorted guitar), does the text embedding capture that binding? To find out, we introduce an attribute swap perturbation: the caption of a real recording is edited by exchanging exactly one property, timbre, lead versus accompaniment, or order of first appearance, between two instruments. We then test four contrastive music-text models and one large audio-language model on whether the audio scores higher against the original caption than against the perturbed one. No contrastive model distinguishes the two captions reliably. The audio-language model does better, but further experiments show that its advantage rests largely on audio-agnostic language priors. Our results thus provide compelling evidence that the CLAP score and related metrics do not capture fine-grained musical meaning or attribute bindings; their representation is closer to a bag-of-words that leaves them insensitive to meaning-changing perturbations of the caption.
[NLP-55] Feeding BabyLMs Macaroni: Code-Switching Curricula Cause Cross-Lingual Convergence EMNLP2026
【速读】： 该论文旨在解决多语言语境下语言模型在跨语言表征对齐方面的挑战，特别是如何提升模型在不同书写系统间（如拉丁字母与汉字）的语义一致性。其核心解决方案在于通过引入分层式代码转换（code-switching）训练课程，利用大语言模型（LLM）生成的词级与句级代码转换文本对小型解码器模型进行预训练。研究表明，这种基于代码转换的数据增强策略能够有效促进平行文本在嵌入空间中的跨语言对齐，且该对齐效果在后续继续训练于单语文档时仍能保持。实验结果表明，在从词级到句级再到单语文档的渐进式学习课程下，经代码转换数据训练的模型在BabyLM评估套件上显著优于未使用此类数据的基线模型，验证了代码转换课程学习作为多语言预训练中高效数据增强方法的有效性。
链接: https://arxiv.org/abs/2609.30535
作者: Dries Rooryck,Alex Cai,Yonatan Belinkov,David Alvarez-Melis,Kianté Brantley
机构: Technion – Israel Institute of Technology (以色列理工学院); Kempner Institute (肯普纳研究所); Harvard University (哈佛大学)
类目: Computation and Language (cs.CL)
备注: 17 pages, 8 figures. Accepted to the BabyLM Workshop at EMNLP 2026
Abstract:Children in multilingual communities often code-switch, using multiple languages in a single utterance. Can we induce cross-lingual alignment in language models by training on code-switched text? We pretrain small decoder-only transformers on two 100M-word multilingual corpora: a base corpus formed by mixing the English, Dutch, and Chinese BabyBabelLM datasets, and a corpus generated from it by inserting word- and sentence-level code-switching using an LLM. We find that training on code-switched data aligns the representations of parallel text, particularly across different scripts, and that this alignment persists through training on monolingual documents. Under a learning curriculum that progresses from word-level code-switching, to sentence-level code-switching, to monolingual documents, models trained on code-switched data outperform baselines trained without it on the BabyLM evaluation suite. Our work characterizes code-switching curriculum learning as an effective data augmentation method for multilingual pretraining. We release our code, data, and models at this https URL.
[NLP-56] Inquesto Score: A reliability Protocol For Voice Agents
【速读】： 该论文旨在解决语音代理（Voice Agent）在实际部署中因交互失败导致交易、权限访问等关键后果的问题，提出一种可复现且可解释的评估方法。其核心挑战在于现有评估方式难以准确衡量语音代理在复杂真实场景下的可靠性，尤其缺乏对非文本层面失败事件（如语用冲突、响应延迟）和部署环境差异的系统性考量。解决方案的关键是提出Inquesto Score（IS），一种基于固定版本化评估集的可靠性度量协议，将可靠性定义为在特定评估群体中成功达成用户目标且未发生功能失败或更严重问题的通话比例。IS通过明确定义故障事件及其严重等级，直接从音频数据中量化时序故障（如抢话、延迟响应），并结合场景谓词、工具调用轨迹与一个固定的开源模型评判器（pinned open-model judge）来评估语义及状态依赖性故障。此外，评估结果附带诊断视图，涵盖行为模式、声学鲁棒性、身份识别处理及不同说话人群体表现，但不将其合并至主得分中以保持透明性。实验基于参考语音代理系统的13种配置，覆盖30个场景、3种声学条件、4类说话人群体，每代理306通电话，验证了可靠测量需依赖超越转录文本的多模态证据、对部署条件的显式建模以及评估者有效性验证。研究同时开源了评估协议、参考实现与完整评估记录，推动语音代理评估向标准化、可复现方向发展。
链接: https://arxiv.org/abs/2609.30514
作者: Massa Baali,Bhiksha Raj
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Voice agents are increasingly deployed in workflows where failed interactions can affect transactions, access, and other consequential outcomes, creating a need for reproducible and interpretable evaluation. We introduce Inquesto Score (IS), a protocol for measuring voice-agent reliability as the percentage of calls in a fixed, versioned evaluation population that achieve the caller’s goal without a functional failure or worse. Rather than combining heterogeneous metrics, IS defines explicit failure events and severity levels and evaluates the deployed voice pipeline. Timing failures, including talk-over and delayed responses, are measured directly from audio, while semantic and state-dependent failures are evaluated using scenario predicates, tool traces, and a pinned open-model judge. Diagnostic views of behavior, acoustic robustness, identity handling, and speaker groups accompany the score without being combined into it. Inquesto Score v0.1 evaluates 30 scenarios, three acoustic conditions, four speaker groups, and 306 calls per agent across 13 configurations of a reference voice-agent system. Our evaluation shows that reliable measurement requires evidence beyond transcripts, explicit treatment of deployment conditions, and validation of the evaluators used to determine outcomes. We release the protocol, reference implementation, and evaluation records.
[NLP-57] Breaking Homogeneity: Diversifying Persona Sets for Creative LLM Outputs
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在开放式任务中生成内容同质化的问题，这种同质性易引发群体思维（groupthink），导致创意输出趋于单一且可能次优。其核心解决方案是将角色多样化（persona diversification）建模为集合级条件生成问题，并探索两个正交的设计维度：角色的选取与生成方式，以及空间填充（space-filling）与前沿探索（frontier-seeking）两类多样性策略。研究通过四种方法实例化这一设计空间，涵盖覆盖与分散的子集选择、均匀覆盖采样及进化式角色生成。在替代用途任务（AUT）、Infinity-Chat和发散联想任务（DAT）上的实验表明，所提方法在多种任务和创造力目标下均显著提升表现。其中，进化式角色生成相较仅使用任务提示，在AUT上使响应多样性提升78.8%，原创性提高26.1%，灵活性提升49.5%，整体创造力提升13.9%，同时保持98.5%的合理性；在Infinity-Chat中，其诱导的响应分离度接近随机角色的两倍。此外，进化角色可与优化创意的提示协同，进一步提升响应多样性18.6%和创造力6.3%。结果验证了角色集合几何结构作为任务无关的多样化生成机制的有效性，支持角色多样化作为一种可复用的提示优化补充策略。
链接: https://arxiv.org/abs/2609.30492
作者: Sang Bin Moon,Nicole Cho,Daniel Borrajo,Sumitra Ganesh,Abolfazl Hashemi
机构: Purdue University (普渡大学); J.P. Morgan AI Research (摩根大通人工智能研究)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
备注:
Abstract:Language models often produce homogeneous responses to open-ended tasks; such homogeneity can spawn groupthink-the convergence of ideas toward a singular and potentially suboptimal decision. We formulate persona diversification as a set-level conditioning problem and study two orthogonal design choices: selecting versus generating personas, and space-filling versus frontier-seeking diversity. We instantiate this design space with four methods spanning coverage and dispersion subset selections, uniform-coverage sampling, and evolutionary persona generation. Evaluations on the Alternative Uses Task (AUT), Infinity-Chat, and Divergent Association Task (DAT) show the benefits of the proposed methods across tasks and creativity objectives. On AUT, evolutionary persona generation increases response diversity by 78.8%, originality by 26.1%, flexibility by 49.5%, and holistic creativity by 13.9% over task-only prompting, while maintaining 98.5% validity; on Infinity-Chat, it nearly doubles persona-induced response separation relative to random personas. Moreover, evolutionary personas compose with creativity-optimized prompting, further increasing its response diversity by 18.6% and creativity by 6.3%. These results establish persona-set geometry as a task-agnostic mechanism for eliciting divergent LLM outputs, and support persona diversification as a reusable complement to prompt optimization.
[NLP-58] AcoustiClaim: A Numeric Claim Benchmark with Instrument Ground Truth ICASSP2027
【速读】： 该论文旨在解决音频语言模型中对声学量值（acoustic quantities）的数值主张（numeric claims）缺乏可验证性的问题，即现有模型输出的数值无法通过人类判断或裁判模型来评估其是否真实反映信号特征。其核心解决方案在于提出AcoustiClaim框架，该框架能够从自由文本中提取数值主张，依据定义该量值的仪器进行评分，并根据参考信息的可读位置对量值进行分类。研究在四个开源模型和一个闭源模型上，针对两个语料库中的十类声学量值进行了五种不同方式的评估，共生成207个数据单元。结果显示，49个单元的数值分布少于五个不同值，仅有8个可排序单元的秩相关系数超过设定阈值0.3，其中3个显著高于阈值，且5个为闭源模型对基频（F0）的预测结果。在所有可排序单元中，除三个外，误差均不低于常数预测基线水平。研究训练的参考解码器在95%的混合语音样本中成功识别出五类语音相关量值，且在纯净语音样本中能准确复现目标规则，仅依赖音频输入即可完成。引入校准阈值后，有选择地屏蔽部分信息可在平均意义上降低所有十类量值的误差，且在每个分割点上使八类量值误差下降，优于随机选择带来的最大0.6%误差提升。尽管线性基线在误差排序上表现不劣于本方法，但基频标准差（F0 s.d.）与闪烁度（shimmer）仍持续高于常数预测基线，表明这些量值的建模仍具挑战性。
链接: https://arxiv.org/abs/2609.30483
作者: Sheng-Tse Lin,Siyuan Zhai,Chien-Liang Kuo,Massa Baali,Bhiksha Raj
机构: Google(谷歌); Carnegie Mellon University (卡内基梅隆大学)
类目: ound (cs.SD); Computation and Language (cs.CL); Machine Learning (cs.LG); Audio and Speech Processing (eess.AS)
备注: 5 pages, 3 figures, 2 tables. Submitted to ICASSP 2027. Siyuan Zhai and Chien-Liang Kuo contributed equally. Code and outputs: this https URL
Abstract:Audio language models state numbers for acoustic quantities, and neither human opinion nor a judge model says whether such a number is true of the signal. AcoustiClaim extracts each numeric claim from free text, scores it against the instrument that defines the quantity, and classes each quantity by where its reference can be read. Four open-weight systems and one closed model, asked for ten quantities five ways on two corpora, fill 207 cells. Of these, 49 emit fewer than five distinct values, and eight of the 158 cells that can be ranked exceed a rank correlation of 0.3, the bar we set, three with an interval clear of it, five of them one closed model reading pitch. Error sits at or above a constant-predictor floor in every ranked cell but three. The reference decoder we train declines the five voice quantities in prose on 95% of mixtures, with nothing withheld, and states them on the clean twins, reproducing its targets’ rule from audio alone. With a calibrated threshold, withholding lowers error on all ten quantities on the mixtures in the mean and on eight at every split, against at most 0.6% from a random selector. A linear baseline orders errors at least as well as ours. F0 s.d. and shimmer stay above the constant floor.
[NLP-59] CARGO: Context-Aware Retrieval-Gated Evaluation of Agent ic AI in Production
【速读】： 该论文旨在解决参考答案驱动的大型语言模型作为裁判（LLM-as-a-judge）评估在动态实体场景下存在的“参考实例偏差”（Reference-Instance Divergence, RID）问题。在部署的代理系统中，尽管操作流程正确，但由于实际运行实体（如支持案例、资产账户）与参考答案中的实体不一致，传统方法会将正确的实体信息替换视为错误或幻觉，导致误判。其解决方案的关键在于提出CARGO框架：（i）将检索到的参考答案视为程序性范例，将事实判断基于实时实例的观测上下文进行锚定；（ii）对每个陈述赋予三类状态（支持、矛盾、不可验证），仅惩罚矛盾内容；（iii）以检索置信度为门控条件，将生产环境评估转化为选择性预测。通过引入构造真值的扰动诊断基准CARGO-Bench，实验表明标准参考基裁判对所有实体移植的正确回答均给出100%错误惩罚且缺乏区分能力（判别指数DI≈0），而CARGO有效消除虚假惩罚（0/50误判），保持近乎完整的矛盾召回率（50/50和49/50），并将判别指数提升至0.58 [0.48, 0.68]，其中大部分性能提升源于上下文锚定的维度定义。然而，该框架也暴露出自身局限：对实体值的宽容性削弱了对流程性缺陷的检测能力（仅20%召回率）。后续补救措施未能弥合差距，且基于人工标注员的实证研究显示该盲点普遍存在。研究进一步发布了可预注册的评估协议，以支持在真实流量中扩展至专家一致性、风险覆盖范围及成本分析。
链接: https://arxiv.org/abs/2609.30471
作者: Mukul Chhabra,Shail Patel,Luigi Medrano
机构: Dell Technologies(戴尔科技)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 15 pages, 1 figure, 5 tables, 1 algorithm. Preprint
Abstract:Reference-based LLM-as-a-judge evaluation assumes the reference answer is the target. In deployed agentic systems that operate over dynamic entities (support cases, assets, accounts), the closest available reference typically applies the correct procedure to a different entity, so a literal judge penalizes different identifiers, dates, and statuses as errors or hallucinations. We name this failure mode reference-instance divergence (RID). We propose CARGO, a framework that (i) treats retrieved references as procedural exemplars and grounds factual judgments in the live instance’s observed context, (ii) assigns each claim a three-way status (supported, contradicted, unverifiable) and penalizes only contradictions, and (iii) gates evaluation by retrieval confidence, casting production evaluation as selective prediction. We introduce CARGO-Bench, a perturbation-based diagnostic suite with ground truth by construction that separates leniency from discrimination. On CARGO-Bench (246 items, two judge models, 7,872 judgments), the standard reference-based judge penalizes 100% of correct entity-transplanted answers and is uninformative (discrimination index DI ~ 0); supplying the live facts without reframing changes nothing. CARGO eliminates these false penalties (0/50) while retaining near-complete contradiction recall (50/50 and 49/50), raising DI to 0.58 [0.48, 0.68]; a rubric-swap control attributes most of the effect to context-grounded dimension definitions. CARGO also exposes a limitation of its own design: the leniency that protects entity values suppresses detection of procedural corruptions (20% recall). A post-hoc fix does not close the gap, and an LLM-as-annotator study with written guidelines and adjudication shows the same blind spot. We release a preregistered protocol for extending the evaluation to expert agreement, risk-coverage, and cost on production traffic.
[NLP-60] RAZOR: Pruning Replaceable Experts in LLM s
【速读】： 该论文旨在解决混合专家模型（Mixture-of-Experts, MoE）在专家剪枝过程中因存储完整专家池带来的高内存开销问题，其核心挑战在于：在固定剪枝预算下，如何最大限度地保持原始模型的输出分布。传统方法通常依赖专家使用频率或贡献度作为剪枝依据，但这些指标无法准确反映移除某个专家后其功能是否可被剩余专家有效替代。为此，论文提出无需训练的RAZOR剪枝方法，其关键创新在于通过**共识残差（consensus residuals）**评估专家的功能可替代性——即计算各专家输出与原始加权混合结果之间的偏差，从而量化其不可替代性。该方法基于单次删除的精确恒等式，在不依赖梯度或恢复训练的前提下，利用校准样本聚合局部评分，实现高效且精准的剪枝。实验表明，RAZOR在多个主流模型（如GLM-4.7-Flash、Qwen3.6-35B-A3B等）上均显著优于现有方法，不仅在九项任务的宏观平均性能上领先，还在匹配基准测试中超越REAP达2.12–5.59分，并在所有对比设置中赢得多数任务胜出，同时降低反向KL散度，验证了其对模型输出分布的更好保留能力。然而分析也揭示，即使任务表现和预测保真度良好，生成内容仍可能出现多样性、格式及终止行为的变化，说明任务保留与预测准确性并不能完全保证生成稳定性。
链接: https://arxiv.org/abs/2609.30465
作者: Mingyang Song,Mao Zheng
机构: Tencent(腾讯); CodeModels
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:
Abstract:Mixture-of-experts (MoE) models activate few experts per token but store the full expert pool. Expert pruning reduces this storage burden; at a fixed pruning budget, the goal is to preserve the original model’s output distribution as closely as possible. Yet an expert’s usage or contribution magnitude does not by itself determine the damage caused by its removal. What matters is whether the surviving computation can replace its function. We introduce RAZOR, a training-free expert pruning method that scores functional replaceability using consensus residuals: deviations of expert outputs from the original weighted mixture. An exact single-deletion identity at a fixed layer input accounts for survivor renormalization and router-selected refill, providing local scores aggregated over calibration tokens for budgeted pruning without gradients or recovery training. On GLM-4.7-Flash, Qwen3.6-35B-A3B, DeepSeek-V4-Flash-0731, and Hy3 at 25% and 50% expert removal, RAZOR achieves the highest nine-task macro average among the evaluated pruning methods in all eight settings. On the two backbones with matched REAP benchmark runs, it exceeds REAP by 2.12–5.59 points and wins all 36 paired task comparisons. It also lowers reverse KL relative to REAP in all four matched GLM-4.7-Flash and Qwen3.6-35B-A3B model–budget settings. Analysis of responses generated by Qwen3.6-35B-A3B nevertheless reveals changes in diversity, formatting, and termination, underscoring that task retention and predictive fidelity do not ensure generation stability.
[NLP-61] Inference-Time Target Speaker Unlearning in LLM -Based Automatic Speech Recognition
【速读】： 该论文旨在解决多说话人语音识别（multi-speaker ASR）与说话人分离（diarization）场景中，用户对隐私保护的迫切需求——即允许特定说话人动态“退出”自动语音转录服务，而无需实际离开会议。传统系统一旦开始转录，便无法灵活排除某些说话人的语音内容，这在视频会议等敏感场景下构成隐私风险。为此，论文提出目标说话人遗忘语音识别（Target-Speaker Unlearning ASR, TSU-ASR）任务，并设计了一种轻量级的注册条件门控（Enrollment-Conditioned Gating, ECG）模块，可嵌入冻结的双流语音大模型（dual-stream speech LLM）中，在推理阶段动态实现对未曾在训练阶段见过的新“退出”说话人的语音抑制。其解决方案的关键在于：通过可微分的门控机制，仅在推理时根据注册的退出说话人特征动态屏蔽其语音信号，从而在不修改主模型参数的前提下，实现对指定说话人语音的精准抑制，使对应语音转录准确率显著下降（AMI数据集从72.3%降至48.2%，AliMeeting数据集从73.6%降至27.3%），同时保持其他保留说话人转录性能基本不变。该方法为现代视频会议平台提供了可扩展、低延迟的隐私保护机制，支持数百万场在线会议每日动态处理隐私请求。
链接: https://arxiv.org/abs/2609.30439
作者: Bo Su,Yueru Yan,Thai Le
机构: 未知
类目: Computation and Language (cs.CL); Sound (cs.SD); Audio and Speech Processing (eess.AS)
备注: 5 pages
Abstract:We introduce target-speaker unlearning ASR (TSU-ASR) task in a fully end-to-end framework for multi-speaker ASR and diarization. Given a multi-speaker utterance and a set of opt-out speakers who do not wish to have their speech transcribed, the task requires an ASR system to transcribe all speakers except the opt-out ones, while still indicating when those speakers are active. As a first step towards tackling this task, we introduce a novel, light-weight Enrollment-Conditioned Gating (ECG) module attachable to a frozen dual-stream speech LLM that enables ASR for new opt-out speakers dynamically during inference, even those who were not seen during initial ECG training phase. Our experiments on both AMI (English) and AliMeeting (Mandarin) datasets show that speech transcription accuracy for corresponding opt-out words or characters falls from 72.3% to 48.2% and from 73.6% to 27.3%, respectively, while retained speakers’ transcription error rates maintain more or less the same. Our approach provides a practical solution for modern video conferencing platforms, allowing speakers to dynamically opt-out from automated AI transcriptions without forcefully leaving the meeting sessions, enabling a privacy-preserving interface for potentially millions of online meetings daily.
[NLP-62] All In Good Time: Causality-Aware Framework for LLM -Based Simultaneous Speech-to-Speech Translation
【速读】： 该论文旨在解决低资源环境下同时性语音到语音翻译（Simul-S2ST）中的关键挑战，即因果对齐训练数据稀缺以及现有方法依赖固定翻译策略或置信度启发式规则所导致的翻译质量不佳与延迟过高问题。其核心解决方案在于提出一种因果感知的同步语音翻译框架（FAST-CAP），包含三个关键技术：(i) 分解式语音到语音翻译架构（FAST），实现更高效的跨语言语音表征学习；(ii) 因果感知自适应翻译策略（CAP），动态调整翻译决策以平衡延迟与质量；(iii) 因果感知延迟评估指标，更准确地衡量实时性表现。通过设计新型数据生成管道，该框架能够生成高保真、因果对齐的语音片段，显著提升语音转移质量。实验结果表明，相较于固定策略，FAST-CAP在CVSS多语种数据集上实现了最高+1.2 BLEU的翻译质量提升和26%的相对延迟降低，且在训练数据远少于现有系统的情况下，仍达到当前最优的翻译质量与说话人保真度，并实现高达38.8%的相对延迟缩减。
链接: https://arxiv.org/abs/2609.30416
作者: Amir Hussein,Enas Albasiri,Travis M. Bartley,Nourchene Ferchichi,Ke Hu,Harishchandra Dubey,Myungjong Kim,Zhehuai Chen,Oluwatobi Olabiyi,Sanjeev Khudanpur
机构: Johns Hopkins University (约翰霍普金斯大学); NVIDIA(英伟达)
类目: Computation and Language (cs.CL)
备注:
Abstract:Large Language Models (LLMs) have shown strong performance in low-resource offline translation; however, extending them to simultaneous speech-to-speech translation (Simul-S2ST) remains challenging due to the scarcity of causally aligned training data with high cross-lingual speaker fidelity. In addition, existing approaches rely on fixed translation policy or confidence heuristics, leading to suboptimal quality and higher latency. We propose a causality-aware Simul-S2ST framework with a novel data pipeline that generates high-fidelity, causally aligned segments with improved voice transfer. The framework introduces (i) a factorized S2ST architecture (FAST), (ii) a causality-aware adaptive policy (CAP), and (iii) causality-aware latency metric. Experiments on CVSS Spanish, German, and French show that FAST-CAP consistently improves the quality-latency trade-off, achieving up to +1.2 BLEU and a 26% relative latency reduction over a fixed policy. Despite using substantially less training data than existing systems, FAST-CAP achieves state-of-the-art results in speech translation quality and speaker fidelity while yielding up to a 38.8% relative reduction in latency.
[NLP-63] A Unified Account of Concepts and Chunks
【速读】： 该论文旨在解决认知心理学中概念形成（concept formation）与模式块（chunk）习得这两个研究领域长期割裂的问题，从而推动统一认知理论的发展。其核心挑战在于如何建立一个能够同时解释概念性知识与经验性模式块学习的整合框架。解决方案的关键在于提出一种扩展的计算理论，将Cobweb模型中的分类与概念生成机制与块及其获取过程相结合，该理论不依赖于特定感知模态，适用于任何可分解为元素及元素间关系的经验系统。通过实现该理论的系统trellis/，研究展示了其在上下文无关语法（context-free grammar）学习中的有效性，证明了系统能够表征句法知识、执行句子解析与生成，并从样本解析中习得组合结构。实验结果验证了该方法在处理兼具概念性与块状特征的语言结构方面的可行性与优势。
链接: https://arxiv.org/abs/2609.30414
作者: Karthik Singaravadivelan,Pat Langley
机构: Stanford University (斯坦福大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Accepted to ACS-26 (oral presentation)
Abstract:Cognitive psychology has studied how people encode, use, and learn concepts that describe categories, and how they represent, recognize, and acquire chunks for familiar patterns of elements. The literatures on these two topics are nearly disjoint, which poses a challenge for unified theories of cognition. In this paper, we review Cobweb, a computational account of categorization and concept formation, and propose an extended theory that incorporates chunks and their acquisition. The theory makes no commitments about modality, applying to any experience that decomposes into elements and relations among them. We also present \trellis/, an implementation of this theory, and illustrate its application to learning context-free grammars, which we adopt as a testbed because they involve both concept-like and chunk-like elements. In addition, we report experimental results on three synthetic grammars that demonstrate the system’s ability to represent syntactic knowledge, use it to parse and generate sentences, and learn compositional structures from sample parses. We conclude by discussing related work on concepts and chunks, along with directions for future research in the area.
[NLP-64] What Improves Multimodal Misinformation Detection? Answers from a Large-Scale Empirical Study EMNLP2026
【速读】： 该论文旨在解决多模态虚假信息检测中因设计选择不明确而导致模型性能不可靠的问题，尤其关注文本与图像结合的误导性内容识别。其核心挑战在于，现有方法的有效性高度依赖于一系列未被充分验证的设计决策（如模型架构、特征融合方式、预训练模型选择等），但这些因素在实际应用中常以非系统化的方式影响检测效果。论文通过超过3,375次大规模实验，在三个基准数据集上系统评估了多种预训练视觉与语言骨干网络及多模态融合策略，揭示了哪些设计选择能够有效提升检测性能，哪些会在特定情境下“无声失效”（silent failure），并识别出影响模型行为最关键的模块环节。解决方案的关键在于提出一套基于实证分析的可复现、可解释的设计指南，为构建更稳健、可信赖的多模态虚假信息检测系统提供了可靠依据。
链接: https://arxiv.org/abs/2609.30402
作者: Akshit Sharma,Prashant W. Patil
机构: CVPR Lab, MFSDSAI; Indian Institute of Technology Guwahati(印度理工学院古瓦哈蒂分校)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG); Multimedia (cs.MM)
备注: Accepted at the Tenth Widening NLP Workshop (WiNLP), co-located with EMNLP 2026
Abstract:Multimodal misinformation is increasingly crafted to look convincing by pairing a textual claim with an image that appears to “prove” it. Yet in practice, building effective detectors often hinges on a small set of design choices that are rarely examined in a controlled way. In this paper, we conduct a large-scale study of multimodal design choices for misinformation detection with over 3,375 experiments- spanning three benchmark datasets and a broad range of pre-trained vision and language backbones. Through systematic comparisons and targeted robustness analyses, we distill practical guidance on which design choices help, when do they fail silently, and what aspects of the pipeline most strongly shape model behavior, answering 4 key Research Questions (RQs). We aim to provide a reliable foundation for designing stronger and more dependable multimodal misinformation detection systems, thus contributing to the broader research community.
[NLP-65] When Is a Multi-Agent Code Judge Actually Grounded? Two Label-Free Measurements and a Judge That Declines to Guess NEURIPS2026
【速读】： 该论文旨在解决在大语言模型（LLM）对代码正确性进行判断时，存在“缺乏证据却给出自信结论”的问题，即模型在无充分依据的情况下仍输出看似合理的判断，导致评估结果不可靠。其核心挑战在于，现有基于多智能体验证（multi-agent verification）的方法依赖可检索的外部文档作为证据，而代码评判中难以获取独立且具有区分性的证据，致使模型无法有效区分两个候选代码的优劣。论文的关键解决方案是提出一种无需标注数据（label-free）的机制，通过分析多智能体验证流水线内部日志中的特定信号（如证据一致性或推理分歧度），识别出模型缺乏足够依据的判断情形，并据此对无法可靠比较的案例进行过滤。实验表明，仅通过该筛选机制即可将准确率从20.7%提升至36.9%，同时保持对超过一半比较任务的响应能力，从而实现了对“无根据判断”的有效检测，而非单纯提升模型准确性。
链接: https://arxiv.org/abs/2609.30328
作者: Salma Roshdy Aly,Hussein Assaf,Ziad Kobti
机构: University of Windsor (温莎大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Software Engineering (cs.SE)
备注: Accepted at the 21st Women in Machine Learning Workshop (WiML), NeurIPS 2026
Abstract:When one language model judges whether another’s code is correct, it does not report the absence of evidence. It returns a confident verdict with reasoning attached, indistinguishable from a verdict it had grounds for. Multi-agent verification, which decomposes a judgment into checkable claims and verifies each against evidence, is a promising response and works well when the evidence is a set of retrieved documents. We argue such methods require two things of their evidence: it must be independent of the answer under review, and it must differ between the two candidates being compared. The second condition holds automatically with retrieved documents and stops holding in code judging. Running MARCH, a published framework unmodified over 80 condition-by-cell measurements on two code judging benchmarks, we find it declares both solutions equally good on 78 to 95% of comparisons, reaching 4.4% accuracy where the same model asked directly reaches 43.7%. Neither easier problems nor a larger judge changes this. Two measurements taken from the pipeline’s own logs explain it without needing labels. Gating on one of them, the pipeline declines the comparisons it cannot make and raises its accuracy from 20.7 to 36.9% while still answering half of all comparisons. The contribution is not a more accurate judge, but a label-free way to tell when a judge has no basis for its answer. Comments: Accepted at the 21st Women in Machine Learning Workshop (WiML), NeurIPS 2026 Subjects: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Software Engineering (cs.SE) Cite as: arXiv:2609.30328 [cs.AI] (or arXiv:2609.30328v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2609.30328 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Salma Roshdy Aly [view email] [v1] Wed, 23 Sep 2026 23:19:59 UTC (8 KB)
[NLP-66] PALM: Point-in-Time Adaptation for Financial Language Models
【速读】： 该论文旨在解决金融回测中语言模型存在的前瞻偏差（look-ahead bias）问题，即模型在训练时使用了研究期之后发布的文本，从而提前“知晓”了待预测的事件结果。现有解决方案采用逐年更新的点对时（Point-in-time, PIT）语言模型，通过按年份时间过滤语料库进行预训练并发布对应年份的模型检查点（checkpoint），每个检查点均标注明确的时间截止点。然而，每年重复执行完整的预训练成本高昂，且其必要性从未被验证。本文发现，新一年的检查点在相同评估窗口上并未显著优于前一年的旧检查点，表明年度重训并非必需。基于此观察，作者提出PALM（Point-in-time Adaptation for financial Language Models），一种无需修改预训练权重、仅在决策日期前发表的文本上拟合低秩适配器（low-rank adapter）的轻量级替代方案。研究表明，仅需一个小型适配器即可有效扩展旧检查点的知识覆盖范围至新时间段，且性能优于持续预训练（continued pretraining）。该方法在十年金融新闻数据及多种规模（1.3至4.2B参数）的PIT模型上得到验证，具有良好的通用性与有效性。
链接: https://arxiv.org/abs/2609.30316
作者: Seunghan Lee,Jun Seo,Jaehoon Lee,Junhyeok Kang,Sangjun Han,Sungdong Yoo,Minjae Kim,Tae Yoon Lim,Dongwan Kang,Hwanil Choi,Soonyoung Lee,Wonbin Ahn
机构: LG AI Research( LG人工智能研究院), Seoul, Republic of Korea(首尔, 韩国)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL); Computational Finance (q-fin.CP)
备注:
Abstract:Language models used in financial backtests suffer from look-ahead bias, as a model trained on text published after the study period has already observed the outcomes it is asked to predict. To handle this issue, point-in-time (PIT) language models are pretrained on chronologically filtered corpora and released as one checkpoint per calendar year, each with a documented cutoff. However, each additional year costs a full pretraining run, and whether that run is necessary has never been tested. In this paper, we show that the annual pretraining run is not necessary. We instead compare each checkpoint against the newer one that replaced it, and find that the newer checkpoint scores no better on the same evaluation window. Motivated by this observation, we propose PALM (Point-in-time Adaptation for financial Language Models), a simple yet effective alternative to annual pretraining that fits a low-rank adapter on text published before the decision date without modifying any pretrained weight. We further find that a small adapter is enough to add a new period to the knowledge an old checkpoint already encodes, and that this outperforms continued pretraining. We validate PALM on a decade of financial news and on various families of PIT models, whose cutoffs span two decades and whose sizes range from 1.3 to 4.2B. Code is available at: this https URL.
[NLP-67] A Benchmark Framework for Screening Automation in Systematic Reviews
【速读】： 该论文旨在解决系统性综述（Systematic Review, SR）筛选阶段耗时耗力的问题，尤其针对由生成式 AI 在文献相关性分类任务中应用时因类别不平衡导致的传统评估指标失真这一关键挑战。其解决方案的核心在于构建一个包含45,064条标注数据的基准数据集（SRBench），覆盖32个经过精心筛选的二次研究，以真实反映SR筛选中排除文献远多于纳入文献的固有类别分布；同时提出一种考虑类别不平衡的评估框架，并开发了PromptSR工具，支持提示工程实验、实验管理与结果分析，从而实现对大语言模型（LLM）在SR筛选任务中性能的更准确、可复现的评估。
链接: https://arxiv.org/abs/2609.30298
作者: Gauransh Kumar,Luciano Marchezan,Guillaume Genois,Kévin Delcourt,Eugene Syriani
机构: Université de Montréal(蒙特利尔大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Systematic reviews (SR) are essential for evidence-based research, but their screening phase is highly time-consuming and labor-intensive. Large language models (LLMs) offer a promising opportunity to reduce this workload by assisting with article relevance classification. However, existing evaluation approaches often rely on traditional metrics that may be misleading for highly imbalanced SR screening this http URL paper presents a benchmark dataset of 45,064 labeled entries for evaluating LLM performance in SR screening across 32 curated secondary studies. It proposes an evaluation framework that accounts for class imbalance, i.e., the natural prevalence of excluded articles relative to included articles in SRs. It also introduces PromptSR, a tool designed to support prompt experimentation, experiment management, and result analysis for LLM-based screening. We also present a use case demonstrating the application of SRBench and PromptSR.
[NLP-68] SlideLab: Audience-Centered Scientific Slide Generation and Evaluation
【速读】： 该论文旨在解决科研论文向科学演示文稿转换过程中存在的结构性缺失与可理解性不足问题，即如何将学术内容高效、逻辑清晰地转化为适合演讲场景的演示材料。其核心挑战在于确保生成的幻灯片不仅涵盖关键研究信息，还需具备连贯的叙事脉络、合理的视觉呈现及准确的内容对齐。解决方案的关键在于提出一种无需训练的多智能体框架SlideLab，通过分阶段协作机制实现幻灯片的自动生成与迭代优化：首先由内容规划智能体构建整体叙事结构，随后利用视觉生成、版式优化与内容验证等专用智能体协同构建并持续修正共享幻灯片集，从而在保持语义一致性的同时提升呈现质量。此外，论文引入了面向观众的评估框架ConfArena，通过模拟真实会议场景进行逐页评估，有效识别幻灯片中的虚假数据、图像退化、漏页及顺序错乱等问题，显著增强了评价体系的可靠性与实用性。
链接: https://arxiv.org/abs/2609.30294
作者: Vidushee Vats,Karun Sharma,Yuxia Wang
机构: INSAIT; Sofia University “St. Kliment Ohridski”(索菲亚大学“圣克莱门特·奥赫里德斯基”)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: V1
Abstract:Scientific presentations are more than summaries of research papers. They need to present the work in a coherent sequence, explain the main ideas clearly, and help the audience follow the presentation. We present SlideLab, a training-free multi-agent framework for generating scientific presentations from research papers. SlideLab first plans the presentation narrative, then builds and iteratively refines a shared slide deck using agents for content planning, visual generation, layout refinement, and grounding verification. In a blind human preference study, SlideLab was preferred over both open-source and commercial systems on 77% of papers while using roughly 4 times fewer inference tokens than the strongest open-source baseline. We also introduce ConfArena, an audience-oriented evaluation framework that simulates a conference room and assesses presentations slide by slide. ConfArena matches human system rankings and detects injected presentation problems, including falsified numbers, degraded figures, dropped slides, and shuffled slide order.
[NLP-69] Cartograph: Federated Tool Discovery with Operator-Attested Retrieval for AI Agents
【速读】： 该论文旨在解决生成式 AI（Generative AI）代理在调用工具时因连接工具目录规模扩大而导致的工具发现效率低下问题，即传统方法需遍历全部工具定义（时间复杂度为 O(n)），导致资源开销剧增。其核心解决方案是提出 Cartograph——一种联邦式模型上下文协议（MCP）代理，通过三重机制实现从线性遍历到渐进式披露（O(k)）的转变：一是由操作员认证的能力卡片（capability cards），采用 Ed25519 签名确保描述真实性与可追溯性；二是 Rift 三层混淆聚类分析，结合密度聚类、查询边际分析与标记诊断，精准识别语义相近的工具簇；三是两阶段检索策略，先对服务器进行排序再对具体工具筛选。在包含 22 个服务器、374 个工具的部署中，Cartograph 仅暴露 3 个代理工具而非全部定义，显著降低信息过载。基准测试显示其 R@5 达 0.816，优于基于 Jaccard 关键词的基线（0.592），且单次前五名发现的令牌消耗从 42,450 降至 475。此外，Rift 识别出 49 个混淆聚类，其中四个高风险簇源自自动生成卡片。探索性对比表明，混合卡片生成方式虽消除零距离聚类，但可能影响检索精度。端到端网关测量表明，平均引入 5ms 延迟（0.8%），具有实际可行性。Cartograph 与代码执行方案互补，不仅控制可见工具描述的呈现，还记录每项查询所用描述的来源，保障可解释性与可信度。
链接: https://arxiv.org/abs/2609.30293
作者: Justice Owusu Agyemang,Michael Agyare,Kwame Opuni-Boachie Obour Agyekum,Kwame Agyeman-Prempeh Agyekum,Francisca Adoma Acheampong,Jerry John Kponyo
机构: Kwame Nkrumah University of Science and Technology(夸梅·恩克鲁玛科技大学); Ghana Communication Technology University(加纳通信技术大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:The Model Context Protocol (MCP) enables AI agents to discover and call tools, but loading every definition becomes expensive as connected catalogs grow. We present Cartograph, a federated MCP proxy that changes agent-visible tool discovery from O(n) catalog traversal to O(k) progressive disclosure. Cartograph combines three mechanisms: (1) operator-attested capability cards, Ed25519-signed descriptions generated under the deploying operator’s control rather than ranked publisher copy; (2) Rift, a three-layer confusable-cluster analysis comprising density clustering, query-margin analysis, and token diagnosis; and (3) two-stage retrieval, which ranks servers before tools. On a 22-server, 374-tool deployment, Cartograph exposes three proxy tools instead of 374 definitions. A 49-query author-constructed benchmark yields R@5 of 0.816, compared with 0.592 for a Jaccard keyword baseline, while a measured top-5 discovery exchange uses 475 tokens rather than 42,450 under the stated full-catalog accounting. Rift identifies 49 confusable clusters, including four HIGH-risk clusters in bootstrap-generated cards. An exploratory comparison of 119 LLM-generated descriptions removes the observed zero-distance cluster but shows that mixing card-generation regimes can reduce R@5. Gateway measurements over ten trials add 5ms mean latency (0.8%) relative to direct stdio MCP calls. Cartograph is complementary to code-execution approaches: it controls which tool descriptions are surfaced and records the provenance of the descriptions used for ranking for each query.
[NLP-70] A Survey on Fake Review Detection: From Pre-trained Language Models to Large Language Models
【速读】： 该论文旨在解决在线评论系统中由生成式人工智能（Generative AI）引发的虚假评论泛滥问题，特别是大语言模型（LLM）能够生成语义连贯且上下文感知的欺骗性评论，严重威胁平台治理、消费者决策与公众信任。其核心解决方案在于从信息融合（Information Fusion）视角出发，系统整合多源异构证据，包括评论文本、情感特征、评分行为、时间元数据、用户-商品图结构、多模态内容、外部知识以及由大语言模型生成的内容特征，在不同融合层级上构建综合检测框架。研究通过梳理2018年至2024年初的211篇相关文献，揭示了从传统机器学习、深度学习到基于预训练语言模型（PLM）和大语言模型（LLM）方法的演进路径，并分析了各类方法在文本、行为、结构及多模态信息融合上的有效性。同时，论文评估了主流基准（如Amazon、Yelp、OpSpam）上的性能趋势，指出当前评价体系因标签构建方式、数据划分策略与评估标准差异所导致的局限性。最终，研究识别出若干开放性挑战，包括对抗性生成的防御、跨领域迁移能力、不确定性感知融合、缺失源鲁棒性、可解释性以及针对AI生成欺骗内容的可信评估机制。
链接: https://arxiv.org/abs/2609.30292
作者: Fanji Yang(1),Huiyao Chen(2),Xi Yu(1),Meishan Zhang(2),Xiaohong Xiao(3),Mingsen Deng(1) ((1) Guizhou University of Finance and Economics, (2) Harbin Institute of Technology (Shenzhen),(3) Guizhou University of Commerce)
机构: Guizhou University of Finance and Economics (贵州财经大学); Harbin Institute of Technology (Shenzhen) (哈尔滨工业大学（深圳）); Guizhou University of Commerce (贵州商学院)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Fanji Yang and Huiyao Chen contributed equally to this work. Accepted for publication in Information Fusion
Abstract:Online reviews shape consumer decisions, platform governance, and corporate this http URL reviews compromise this information channel by injecting deceptive evidence into rating systems, recommendation pipelines, and public trust this http URL rise of large language models, or LLMs, has changed the problem in two this http URL can generate fluent and context-aware deceptive reviews, while pre-trained language models, or PLMs, and LLMs also provide stronger semantic representations for this http URL survey reviews fake review detection from an information fusion perspective, covering 211 studies published from 2018 to early this http URL organize existing work by evidence source and fusion level, covering review text, sentiment, rating behavior, temporal metadata, user-product graphs, multimodal content, external knowledge, and LLM-generated this http URL trace the development from traditional machine learning and deep learning to PLM-based and LLM-based methods, and examine how different approaches combine textual, behavioral, structural, and multimodal this http URL also analyze reported performance trends on widely used Amazon, Yelp, and OpSpam benchmark families, while noting the limitations caused by different label construction procedures, data splits, and evaluation this http URL, we identify open problems in adversarial generation, cross-domain transfer, uncertainty-aware fusion, missing-source robustness, interpretability, and trustworthy evaluation for AI-generated deceptive content.
[NLP-71] Auditing and Repairing LLM -as-Judge Failures in a Production Text-to-SQL Pipeline
【速读】： 该论文旨在解决生成式文本转SQL（text-to-SQL）流水线中依赖大语言模型（LLM）作为评判器（LLM-as-judge）时，其与人工标注者一致性未被实证评估的问题。研究发现，部署的GPT-4o-mini作为评判器在分歧增强数据集上的Cohen’s kappa仅为0.04，在随机抽样测试中也仅达0.42，且对77.1%的人工标注“忠实”（FAITHFUL）案例产生过度标记。根本原因被识别为一种名为GRADE-HALLUCINATION的机制——即模型因生成性幻觉而错误判断正确查询。解决方案的关键在于采用自托管的Qwen3.6-27B模型替代原判别器，其一致性（kappa = 0.72）显著优于GPT-4o-mini，且成本仅为后者的约1/300；进一步通过三模型一致同意路由策略实现kappa = 0.79、89.7%自动覆盖率，显著提升可靠性。值得注意的是，集成多个弱判别器无益于性能提升，而强模型配对反而降低一致性，表明判别器质量与协同机制需审慎设计。该方法在跨域场景下亦具有效性，可识别出25.5%的BIRD-financial专家标注黄金标准SQL中的潜在问题。
链接: https://arxiv.org/abs/2609.30290
作者: Haowei Liu,Hsin-Tai Wu,Yi Fang
机构: Santa Clara University (圣克拉拉大学); Independent Researcher (独立研究员)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Production text-to-SQL pipelines often end with an LLM-as-judge whose agreement with human annotators has never actually been measured. When we checked ours, the deployed gpt-4o-mini judge agreed with two-author gold at only Cohen’s kappa = 0.04 on a disagreement-enriched set and 0.42 on a uniform-random spot-check, over-flagging 77.1% of the human-FAITHFUL cases in the enriched set. Most of its over-flags trace back to a single mechanism we call GRADE-HALLUCINATION. A self-hosted Qwen3.6-27B replacement (kappa = 0.72) lands in the same range as Claude Opus 4.7 (kappa = 0.71); the head-to-head is underpowered at n = 96, but for the deployment decision that hardly matters, since Qwen costs roughly 1/300 as much per call. Ensembling does not help for free. Pairing the weak judge with a stronger one degrades agreement, whereas three strong judges under unanimity routing reach kappa = 0.79 at 89.7% auto-coverage. Applied out-of-domain, the same audit recipe flags 25.5% of BIRD-financial’s expert-authored gold SQLs as candidate gold-SQL issues under our annotation protocol. Code and pre-registration are at this https URL.
[NLP-72] Not All Memories Are Equal: Hierarchical Collaborative Memory for Validity-Aware Retrieval in LLM Agents
【速读】： 该论文旨在解决团队协作场景中记忆异构性与动态演化带来的有效性问题，即现有记忆增强系统在检索时未考虑记忆的层次结构与时效性，导致频繁召回过时或与当前团队共识冲突的个体记忆，从而影响大语言模型（LLM）代理在协作问答任务中的响应质量。其解决方案的关键在于提出一种分层协同记忆管理与有效性感知检索框架——HiCoMER，该框架通过显式维护团队记忆与个体记忆的有效性状态，并优先检索仍有效的记忆，而非从全部存储记忆中进行无差别的语义相关性排序。具体而言，HiCoMER包含三个核心组件：分层记忆冲突更新器（Hierarchical Memory Conflict Updater）、有效性感知记忆检索器（Validity-Aware Memory Retriever）以及记忆锚定的答案生成器（Memory-Grounded Answer Generator），有效提升了记忆检索的时效性与一致性，显著改善了下游问答任务的质量。
链接: https://arxiv.org/abs/2609.30289
作者: Yufei Shi,Rujing Yao,Ang Li,Yang Wu,Zhuoren Jiang,Xiaozhong Liu
机构: Nanyang Technological University, Singapore(南洋理工大学, 新加坡); University of Macau, China(澳门大学, 中国); Worcester Polytechnic Institute, USA(伍斯特理工学院, 美国); Zhejiang University, China(浙江大学, 中国)
类目: Computation and Language (cs.CL)
备注:
Abstract:In team collaboration scenarios, memory is heterogeneous and continually evolving. Team memories capture collective decisions, protocols, and current consensus, while individual memories preserve member-specific observations, execution traces, and intermediate progress. Existing memory-augmented systems typically retrieve from all stored memories as a flat pool, ranking them by semantic relevance, importance, or recency without modeling hierarchical structure or evolving validity. As a result, they often surface semantically relevant but outdated or conflicting memories, especially individual memories that no longer align with current team consensus, instead of prioritizing currently valid memories. This is particularly problematic when collaborative LLM agents answer user questions, since their responses should be grounded in valid memories. We propose HiCoMER, a framework for hierarchical collaborative memory management and validity-aware retrieval in LLM agents. HiCoMER first maintains the validity of team and individual memories and then retrieves memories that remain valid, rather than retrieving directly from all stored memories. It consists of three components: a Hierarchical Memory Conflict Updater, a Validity-Aware Memory Retriever, and a Memory-Grounded Answer Generator. To evaluate HiCoMER, we construct two new datasets for memory-grounded question answering in collaborative settings. Experiments on both datasets show that HiCoMER consistently outperforms strong baselines by reducing outdated retrieval, preserving current team consensus, and improving downstream QA quality.
[NLP-73] Manifold Projection and Iterative Autoencoder Refinement for Masked Language Modeling
【速读】： 该论文旨在解决基于Transformer的掩码语言模型中注意力机制（attention）计算开销大、效率低的问题，同时探索在不依赖注意力机制的情况下如何有效实现跨标记（token）的上下文混合。其解决方案的关键在于引入一种基于低秩瓶颈自编码器（low-rank bottleneck autoencoder）的混合模块架构，替代传统的注意力计算。具体而言，该方法通过堆叠多个自编码器驱动的混合模块，分别作用于局部邻域、全序列以及注意力头之间，利用瓶颈结构对输入进行压缩与重构，其宽度作为超参数而非训练过程中的动态结果，从而在保持模型表达能力的同时显著降低计算复杂度。此外，在掩码位置采用迭代优化策略，包含“拉取”（pulling）步骤——将嵌入表示向其邻居加权平均方向拉近，以及“校正”（correcting）步骤——通过自编码器将结果投影回学习到的流形空间，以维持语义一致性。实验表明，该模型在C4数据集上预训练后，相较于参数匹配的BERT基线模型，仅需约1.9倍更少的浮点运算量（FLOPs），即可达到接近注意力机制的性能；在罕见词频分桶任务中，结合频率感知的采样策略，其表现可媲美参数匹配的BERT和TinyBERT基线模型。
链接: https://arxiv.org/abs/2609.30288
作者: Narges Mokhtari,Farzan Haddadi,Ebrahim Rezaii
机构: Iran University of Science and Technology (伊朗科学技术大学)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: Submitted to IEEE Transactions on Emerging Topics in Computational Intelligence (TETCI)
Abstract:In Transformer-based masked language models, attention is the primary mechanism for context mixing, but there are other ways to mix data across tokens. Recent attention-free mixers replace attention with fixed or hypernetwork-generated MLPs, alternating their dynamic, content-dependent weighting for computational simplicity. We build an alternative that gets the same property from a low-rank bottleneck autoencoder. We replace attention with a stack of autoencoder-based mixing modules, one operating over local neighborhoods, one over the full sequence, and one across attention heads, each compressing and reconstructing its input through a bottleneck, and its width is a hyperparameter rather than a training effect. In masked positions, we introduce an iterative refinement procedure that has two distinct steps. A pulling step that pulls an embedding representation toward a weighted average of its neighbors, and a correcting step that projects the result back to the learned manifold via an autoencoder. Our architecture achieves a significant portion of attention’s performance at about 1.9 \times fewer FLOPs when pretrained on C4 and evaluated with parameter-matched BERT baselines. Our model equals parameter-matched BERT and TinyBERT baselines on the rarest-token frequency bucket using a frequency-aware training schedule that samples rare tokens more than uniformly for the masking tasks.
[NLP-74] A Mechanistic Study of AI-Text Detection Neurons in Frozen BERT: Sparse Probing and Activation Patching on RAID EMNLP2026
【速读】： 该论文旨在解决生成式 AI 文本检测模型中内部表征机制不明确的问题，特别是揭示在冻结的 BERT-base-uncased 编码器中哪些神经元支持对 AI 生成文本的判别。其核心解决方案是采用 Gurnee 等（2023）提出的 L1-to-L2 稀疏探测协议，对 9,216 个 CLS 隐藏状态维度（12 层 × 768 维）进行系统性分析，识别出每种生成器仅需不到 1% 的稳定神经元即可维持接近全特征检测精度。通过双向激活修补（bidirectional activation patching）验证了这些神经元的因果相关性，其预测翻转频率显著高于随机匹配集合；而均值消融实验表明信号具有冗余分布特性。跨生成器分析发现存在双分结构：指令微调模型将 30–36% 的稳定神经元集中于最后一层，而基础模型则低于 14%，反映出后训练对齐在第 12 层的特征印记。留一族排除评估显示，选定神经元子集在未见生成器家族上仍保持 86–94% 的全特征性能上限，表明检测器可基于一个固定小规模子空间实现泛化，无需针对每个生成器重新识别神经元。
链接: https://arxiv.org/abs/2609.30287
作者: Paweł Blicharz,Miłosz Grunwald
机构: Gradient PG; Gdańsk University of Technology(格但斯克工业大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Accepted to EMNLP 2026 (Main Conference)
Abstract:AI-generated text detectors achieve high accuracy on standard benchmarks, yet the internal representations that drive these predictions remain poorly understood. We study which neurons in a frozen BERT-base-uncased encoder support AI-text detection, using the RAID benchmark across six generators spanning pure-base and instruction-tuned models. We apply the L1-to-L2 sparse-probing protocol of Gurnee et al. (2023) to all 9,216 CLS hidden-state dimensions (12 layers x 768), which we call neurons. The procedure recovers a stable set of under 1% of neurons per generator, consistent across folds and seeds; a probe restricted to that set retains most of the full-feature detection accuracy. Bidirectional activation patching confirms this set’s causal relevance: in both directions it flips predictions an order of magnitude more often than size-matched random sets. Mean-ablating the same neurons leaves accuracy largely intact; the signal is therefore redundantly distributed. Cross-generator analysis reveals a bipartite structure: instruction-tuned generators concentrate 30-36% of stable neurons in BERT’s final layer while both base generators fall below 14%, consistent with a layer-12 footprint of post-training alignment. Leave-one-family-out evaluation shows the selected neurons retain 86-94% of the full-feature ceiling on unseen generator families, so a detector can operate on a small fixed subspace without re-identifying neurons per generator.
[NLP-75] wo Conformal Constructions for Adaptive Within-Document AI-Text Screening
【速读】： 该论文旨在解决在基于生成式 AI (Generative AI) 生成文本的筛查过程中，如何有效控制假警报（false alert）率的问题。其核心挑战在于：筛查过程可在未耗尽检查预算的情况下提前终止，且需在不假设文档内词元（token）独立性的前提下，保证对任意执行路径的假警报率进行严格控制。解决方案的关键在于提出两种有限样本下的校准构造方法：构造A通过注册一个前缀-检测器得分的有限族，并将假警报预算分配至其置信性排名（conformal rank），利用并集界（union bound）保护任意执行子集；构造B则针对固定策略的完整路径最大值进行校准，利用局部最大值不超过全局最大值的性质，使终端置信性排名天然覆盖早期停止情形，无需拆分误差预算。两种方法均在文档级可交换性假设下实现了对允许检查路径中任意假警报的边际控制，并推导出拒绝所需的最小校准样本量。此外，论文还给出了在理想测试、分布偏移及独立审计等附加假设下的边界结果。两种构造均确保了在指定范围内停止的合法性，但未构造e过程或支持置信性排名的乘积形式，检测效能与计算效率仍需通过实证评估验证。
链接: https://arxiv.org/abs/2609.31547
作者: Marco Mandap,Jerahmeel Hipolito,Arcel Galvez,Charlie Margaret Balagtas,Michael Joshua Buluran,Jeff Roel Durmiendo,Rizzette E. Lopez
机构: Bulacan State University (布拉卡南州立大学)
类目: Methodology (stat.ME); Computation and Language (cs.CL)
备注: 16 pages, 0 figures; theoretical manuscript; no empirical evaluation
Abstract:We study false-alert control when screening for text generated by artificial intelligence (AI). The screening procedure selects document prefixes and detectors from observed evidence and may stop before exhausting its inspection budget. We give two finite-sample constructions under document-level exchangeability between human calibration documents and a new null document, with no restriction on dependence among tokens within a document. Construction A registers a finite family of prefix-detector scores and allocates a false-alert budget across their conformal ranks. A union bound protects any executed subset of that family. Construction B calibrates the complete-path maximum of a development-fixed adaptive policy. Each partial-path maximum is bounded by the complete maximum, so a terminal conformal rank protects early stopping without splitting the error budget. We prove marginal control of any false alert across the permitted inspection path and derive necessary calibration counts for rejection. We also state oracle testing, distribution-shift, and independent-audit bounds with their additional assumptions. Both constructions protect stopping within their specified scope; neither proof constructs an e-process or justifies multiplying conformal ranks. Detection power and computational savings remain questions for empirical evaluation.
[NLP-76] Statistical Foundations for a Google Play User-Review Sentiment Index: Signal Fusion Shrinkage Distributional Validation and Dynamic Smoothing
【速读】： 该论文旨在解决应用商店用户评论中情感评估的统计可靠性问题，尤其针对基于星级评分与文本情感分析融合时存在的噪声干扰及偏差。其核心挑战在于如何从异构、有噪声的观测数据（如标准化星级评分与文本情感得分）中准确估计隐藏的评论价值（latent review valence），并构建一个具有严格统计基础的综合情绪指数。解决方案的关键在于采用协方差感知的逆方差加权（covariance-aware inverse-variance weighting），将不同来源的观测值在考虑其协方差结构的基础上进行最优融合，从而获得无偏且最小方差的估计。进一步地，通过引入受限的帮助度（bounded helpfulness）和时效性（recency）权重对评论级估计值进行聚合，并基于估计精度而非预设评论数量阈值实施贝叶斯收缩（shrinkage toward population mean），实现更稳健的个体评分校准。此外，利用局部层次状态空间模型与卡尔曼滤波（Kalman filter）对时间序列趋势进行去噪处理，提升动态监测能力；同时，通过严格的数学证明涵盖最佳线性无偏估计（BLUE）、高斯最大似然估计、高斯共轭收缩、Glivenko-Cantelli与Donsker定理以及德尔塔方法下的计数变换，确保方法论的严谨性。实证示例表明，仅依赖星级评分可能掩盖文本中的负面情绪，而融合文本情感可显著修正评价结果，凸显了多模态信息整合的重要性。
链接: https://arxiv.org/abs/2609.31513
作者: Marco Mandap
机构: Bulacan State University (布拉坎州立大学)
类目: Methodology (stat.ME); Computation and Language (cs.CL); Applications (stat.AP)
备注: 16 pages, 2 tables, no figures
Abstract:We develop a statistically explicit sentiment index for Google Play user reviews and establish the mathematical results supporting its construction. Normalized star ratings and text-sentiment scores are treated as noisy measures of latent review valence and fused by covariance-aware inverse-variance weighting. Review-level estimates are aggregated with bounded helpfulness and recency weights, then shrunk toward a population mean using estimated precision rather than an arbitrary review-count threshold. App-level rating histograms provide a distributional diagnostic for samples returned under different API sort orders; because star ratings are discrete, classical continuous Kolmogorov-Smirnov critical values are not used. A local-level state-space model and the Kalman filter provide a denoised temporal trend. Full proofs cover the BLUE and Gaussian maximum-likelihood result, Gaussian-conjugate shrinkage, the Glivenko-Cantelli and Donsker theorems, count transformations via the delta method, and exact Gaussian Kalman filtering. A worked three-review example shows how textual complaints can materially reduce an apparently perfect star-only score.
[NLP-77] PriceBench: A Diagnostic Benchmark for Price Quality and Brand Preferences in LLM Booking Agents EMNLP2026 EMNLP
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在作为购买代理时，其隐含的偏好（如价格、质量、品牌）如何影响决策结果但缺乏透明度的问题。由于LLM在用户请求下自主选择商品或服务，其内在偏好会悄然决定最终购买对象及其成本，而这些偏好往往未被显式披露。针对酒店预订这一高频率、属性高度可比的典型场景，作者提出PriceBench——一个诊断性基准测试框架，通过逻辑回归选择模型（logit choice model）从LLM的预订行为中反演其对价格、质量和品牌的偏好。该研究对来自8家供应商的28个LLM进行了评估，覆盖3,600项真实纽约市酒店任务。研究发现，模型能力与选择一致性相关，而非选择内容本身：更强大的LLM表现出更强且更一致的偏好，而较弱的模型则可能固守单一选项（易受列表排序操纵）或近乎随机选择。不同厂商间甚至同一厂商内部的偏好差异显著，例如价格敏感度相差超过一个数量级，相同任务下的平均预订价格可在247美元至393美元之间波动。因此，必须针对每个具体LLM进行独立评估，不能简单推断其行为；研究已公开所有任务数据、代码及全部响应结果，以支持后续分析与验证。
链接: https://arxiv.org/abs/2609.31468
作者: Pavel Kireyev
机构: London School of Economics and Political Science (伦敦政治经济学院)
类目: General Economics (econ.GN); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: Accepted to EMNLP 2026 Industry Track. 19 pages, 10 figures, 6 tables. Code and data: this https URL
Abstract:LLMs increasingly act as purchasing agents, which makes the LLM, not the user, the one choosing among the options that satisfy a request; its preferences quietly fix what gets bought and what it costs. Hotel booking is a clean instance: a high-volume choice settled on a few comparable attributes, where the pick reveals those preferences. We introduce PriceBench, a diagnostic benchmark that recovers an LLM’s price, quality, and brand preferences from its booking choices with a logit choice model, applied to 28 LLMs from 8 providers on 3,600 hotel tasks from 179 real New York City properties. We find that capability is associated with how consistently an LLM chooses, not with what it chooses: more capable LLMs hold stronger, more consistent preferences, while weaker ones either lock onto one position, exploitable by whoever controls listing order, or choose almost indifferently. What those preferences favor varies sharply across providers and even within one family: price sensitivity spans more than an order of magnitude, and the price/quality trade-off moves mean booked nightly price from \ 247 to \ 393 on identical tasks. What an agent buys must therefore be measured per LLM, not inferred, and we release the tasks, code, and all 28 response sets.
[NLP-78] Why Alzheimers Speech Screening Fails to Generalize: Bridging the Deployment Gap via Cross-Corpus Evidence Anchoring
【速读】： 该论文旨在解决生成式语音分析模型在阿尔茨海默病（Alzheimer’s disease）及认知风险筛查中面临的跨域泛化能力不足问题，尤其关注模型在未见语言、任务或录音协议下的性能退化。其核心挑战在于：尽管单一数据集上表现良好，但多数语音与语言特征（如停顿、静音、语速等）在不同数据集间呈现方向性冲突，且现有模型（如XLM-R文本基线）在最差领域上的受试者级别曲线下面积（AUC）骤降至0.520，表明其鲁棒性严重受限。解决方案的关键在于提出一种融合策略，通过将预训练的XLM-R文本基线得分与训练阶段选定的“证据锚点”（evidence anchors）进行动态融合，以增强模型对低性能领域的适应能力。实验表明，平衡融合可实现0.785的平均受试者AUC，而锚点主导的融合则将最差领域AUC提升至0.615，显著改善了模型在极端情况下的稳健性。该研究强调，在认知语音筛查中必须评估特征的可迁移性，并报告最差域的性能表现，以推动更具临床实用性的鲁棒模型发展。
链接: https://arxiv.org/abs/2609.31293
作者: Zijian Lu,Sizhe Liu,Yin Zhang,Jixuan Deng,Xinrong Lin,Xinchen Yuan,Chicheng Jin,Yiping Zuo,Yuanchao Li
机构: Nanjing University of Posts and Telecommunications (南京邮电大学); University of Science and Technology of China (中国科学技术大学); Shouyi Technology (守一科技); University of Cambridge (剑桥大学)
类目: Audio and Speech Processing (eess.AS); Computation and Language (cs.CL); Sound (cs.SD)
备注: Accepted to NCMMSC 2026
Abstract:Speech-based screening is a promising, non-invasive approach for detecting Alzheimer’s disease and related cognitive risks. However, models trained on a single domain often generalize poorly to unseen languages, tasks, or recording protocols. This paper investigates this deployment gap using a leave-one-corpus-out evaluation across four distinct datasets. Among 70 interpretable speech and language features, 59 exhibit direction conflicts between healthy control and cognitive risk groups across corpora, with pause, silence, and speech rate showing high protocol sensitivity. Furthermore, while the XLM-R text baseline achieves strong average performance, its Area Under the ROC Curve (AUC) drops to 0.520 on the weakest held-out domain. A standard GroupDRO baseline reaches a 0.766 mean speaker AUC and a 0.504 worst-domain AUC under the same protocol. To address this, we propose a fusion method that integrates XLM-R text baseline scores with evidence anchors selected during training. Balanced fusion achieves a 0.785 mean speaker AUC, while anchor-heavy fusion raises the worst-case speaker AUC to 0.615. This work highlights the need to audit feature transferability and report worst-case domain robustness in cognitive speech screening.
[NLP-79] Asymmetric Classifier-Free Guidance for Target-Speaker ASR
【速读】： 该论文旨在解决目标说话人自动语音识别（TS-ASR）在复杂重叠与噪声环境下，因声学证据变化导致的识别性能下降问题。其核心挑战在于如何在推理阶段动态校准说话人条件信息，以适应不同场景下的声学变化。解决方案的关键是提出一种非对称的无分类器引导（asymmetric classifier-free guidance, CFG），基于Whisper模型构建双分支结构：一个带说话人条件的分支用于预测目标说话人转录文本，另一个无条件分支则生成多说话人序列化转录结果。通过单一引导尺度（guidance scale）调节说话人条件在解码过程中的贡献度，系统在目标域开发数据上选择全局最优引导尺度，并引入轻量级编码器预测器，为每句语音动态调整该尺度，从而实现无需更新主识别模型的自适应校准。实验表明，在领域偏移情况下，该方法相比仅使用条件输入的基线模型，相对词错误率（WER）降低最高达21.8%，相比同一CFG训练模型的标准条件解码也提升了5.6%。而基于“理想”逐句尺度选择的分析进一步揭示了在不同领域偏移下，可实现更大幅度的性能提升，验证了动态引导尺度调整的有效性与潜力。
链接: https://arxiv.org/abs/2609.30476
作者: Yiwen Guan,Jacob Whitehill
机构: 未知
类目: Audio and Speech Processing (eess.AS); Computation and Language (cs.CL); Sound (cs.SD)
备注:
Abstract:Target-speaker automatic speech recognition (TS-ASR) must identify and transcribe a desired speaker under varying overlap and noise conditions. These changes alter the acoustic evidence for the target speaker in the speech mixture, motivating inference-time calibration of speaker conditioning. We introduce asymmetric classifier-free guidance (CFG) for TS-ASR using Whisper: the speaker-conditioned branch predicts the target transcript, while the speaker-unconditioned branch predicts serialized multi-speaker transcripts. CFG adjusts the contribution of speaker conditioning during decoding through a single guidance scale. We select a global guidance scale on target-domain development data and train a lightweight encoder-based predictor to adjust it for each utterance, keeping the recognition model fixed. Under domain shifts, our full system achieves relative word error rate (WER) reductions of up to 21.8% over the condition-only baseline, and 5.6% over standard conditional decoding of the same CFG-trained model. Oracle analysis shows that substantially larger WER reductions are possible through utterance-level scale selection and identifies how beneficial adjustments vary with domain shifts.
信息检索
[IR-0] Retail Product Search: A Practical Approach at Target
链接: https://arxiv.org/abs/2609.31498
作者: Darshan Sonagara,Qujiaheng Zhang,Ankit Singh,Alex Li
类目: Information Retrieval (cs.IR); Machine Learning (cs.LG)
备注: 10 pages, 2 figures, 6 tables
Abstract:Search is one of the most important features in e-commerce, directly driving customer engagement and business growth. A good product search system must show both relevant and desirable results. However, retail search presents unique challenges. User intent can range from exact matches to open-ended discovery. Search systems must also balance multiple goals, such as relevance, revenue, and profit, while keeping response times low. Traditional keyword-based methods often fall short in handling natural language or semantic queries. Vector search helps alleviate these issues, but it can miss key intent signals or return low-precision results. In this paper, we present the design of a hybrid search system at Target that combines lexical and vector search. We describe our approach to data processing, embedding training, precision control for the final result set, multi-channel result fusion (where we compared fusion strategies and adopted weighted interleaving), and the performance optimizations used to maintain low latency for production deployment. Our method improves offline evaluation metrics, and in online A/B testing it raised click-through rate by 0.97%, order conversion by 0.98%, and demand per visitor by 1.10% over lexical-only search, while roughly halving zero-result searches. The resulting system is deployed at scale and serves millions of guests daily.
[IR-1] Enriching Sequential Recommendation with Graph Laplacian Positional Embeddings CIKM2026
链接: https://arxiv.org/abs/2609.31253
作者: Ekaterina Trushkova,Artur Gimranov,Anton Lysenko
类目: Information Retrieval (cs.IR)
备注: CIKM 2026
Abstract:Sequential recommenders typically rely on learnable positional embeddings to encode the order of user interactions. In this work, we ask whether this ordinal signal can be replaced by a structural one derived from the item space. We propose to use Laplacian positional embeddings in SASRec: we build an item co-occurrence graph from training interactions, compute eigenvectors of its symmetric normalized Laplacian, and use them as frozen graph-derived positional embeddings. The backbone architecture and training objective remain unchanged. Experiments on four public sequential-recommendation benchmarks show that this simple replacement improves SASRec performance on most ranking metrics and remains competitive with strong positional and temporal encoding baselines. These findings indicate that item-item graph structure can be an effective substitute for standard ordinal positional embeddings in sequential recommendation.
[IR-2] Agent Recommender: LLM Agents Enable Customizable Recommender Systems on the User Side
链接: https://arxiv.org/abs/2609.31166
作者: Ryoma Sato
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Databases (cs.DB); Digital Libraries (cs.DL)
备注:
Abstract:Recommender systems have traditionally been developed for platforms. However, this has given rise to many phenomena that may be advantageous for platform lock-in but are a nuisance to users, such as clickbait, filter bubbles, and the spread of fake news. Recently, user-side recommender systems have been proposed as a new paradigm for solving this problem. If users deploy their own recommender systems, they are no longer at the mercy of the platform’s interests. However, building a user-side recommender system is not trivial; in particular, customizing one for oneself requires additional data. We propose AgentRecommender, a method that leverages the investigation capability and internal knowledge of LLM agents to flexibly build user-side recommender systems without additional data. AgentRecommender allows users to easily create recommender systems tailored to their own preferences.
[IR-3] SPADE: Escaping the Popularity-Similarity Frontier to Measure Serendipitous Recommendations
链接: https://arxiv.org/abs/2609.31164
作者: Tobias Vente,Maarten Peirsman,Noah Daniëls,Hannu Toivonen,Bart Goethals
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Recommender systems engineer serendipity to foster active exploration and break predictable consumption cycles. The problem with existing offline beyond-accuracy metrics is that they often either isolate historical similarity or global popularity. We aim to design an evaluation metric that examines similarity, popularity, and actual user relevance. To achieve this, we introduce SPADE (Serendipitous Pareto Distance Evaluation). SPADE maps all items into a two-dimensional space to directly calculate a user-specific Pareto frontier of maximally popular and historically similar items. The final serendipity score is then computed by averaging the minimum Euclidean distance from this boundary strictly for the correctly recommended test-set items. Evaluating SPADE across five datasets and five baseline algorithms confirms its effectiveness; our results show that the metric successfully prevents algorithms from exploiting beyond-accuracy measures with irrelevant or non-personalized recommendations, reliably isolating serendipitous discoveries.
[IR-4] CG-Probes: Recovering Guardrail Directions from Patient Query Embeddings CIKM’26
链接: https://arxiv.org/abs/2609.31062
作者: Marko Řeháček,Vítězslav Dušek,Martin Rusinko,Vít Nováček
类目: Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注: Accepted as a short paper at CIKM '26 (35th ACM International Conference on Information and Knowledge Management), Rome, Italy. 7 pages, 1 figure, 2 tables. Code and benchmark: this https URL
Abstract:Patient-facing AI assistants promise valuable support to patients, but incoming queries can pose medical risks. To create guardrails, we work with oncologists to define three ordinal risk axes: Medical Urgency, Psychological Urgency, and Topic Sensitivity. We propose Clinical Guardrail Probes (CG-Probes) to measure the risks from query embeddings. We probe for each axis in the normalized embedding space of frozen embedders via the difference-in-means method, treating each axis as a potential linear direction. To train the probes, we cluster 79,658 Czech oncology search queries with BERTopic and use these clusters to generate pairs of queries with contrastive risk levels via few-shot prompting. We evaluate the approach on 200 queries (90 real, 110 synthetic), each graded by two oncologists, against two open-weight LLMs and a frontier LLM. We find that urgency-based axes are recoverable as linear directions, and the probes are competitive with open-weight LLMs (no significant differences in quadratic-weighted kappa) at a fraction of the latency. Each axis yields a scalar score that clinicians can inspect and use to set escalation thresholds. The pipeline requires only search logs, axis definitions, and black-box access to the embedding model, suggesting transferability across healthcare domains. Robust validation on new queries and axes remains future work.
[IR-5] KuaFu: Compressing Long User Behavior into Understanding at Billion Scale
链接: https://arxiv.org/abs/2609.31045
作者: Jiahao Hui,Lin Zhu,Yishen Hu,Jingdong Shu,Zetai Jiang,Xining Ran,Ben Tan,Yeshou Cai,Gong Chen,Haijie Gu,Jie Jiang
类目: Information Retrieval (cs.IR); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 12 pages, 6 figures, 3 tables
Abstract:Conversational agents, generative recommenders, and personalized advertising all rest on one capability: understanding each user from raw behavior. Prevailing industrial practice is task-specific: for each task, a relevant subsequence is extracted from the full history and a dedicated model trained on it. In production it hits two bottlenecks. First, even after filtering, a single-task sequence stays extremely long: content-interest summarization reads several hundred items per user, tens of thousands of tokens once serialized as prompt text. Second, profiles are refreshed routinely: a billion users weekly, roughly 100K QPM in aggregate, which under a fixed GPU budget sets a hard throughput floor. Compression is therefore mandatory, yet truncation or coarse compression can silently distort the profile, introducing four hallucination types (fabrication, omission, date misattribution, broken logic) that, with no way to evaluate the compressed representation itself, surface only as diffuse degradation in downstream metrics. We present KuaFu, a unified behavior-compression layer whose minimal unit is one behavior item. A two-axis projector compresses each item into 2-4 tokens of width 128-256 (about 10x along the token axis, 20x along width; per-item cache 10 KB to 0.5 KB), with fidelity-oriented four-stage training and layered intermediate evaluation. Across four production profiling tasks it matches or exceeds uncompressed single-task production models on all five headline metrics, raises per-GPU throughput by 37%-350%, and saves 190 GPUs. On public benchmarks it nearly always beats prior compressors at the same compression ratio (up to +17.7 EM on out-of-domain MRQA); on RecBench, a 4B model surpasses its 8B counterpart by 1.90 points. KuaFu has run on the Tencent advertising and recommendation platform for ten months, lifting overall GMV by 1.37%.
[IR-6] QReason : Query-Focused Decoupled Chain-of-Thought for Efficient Passage Reranking EMNLP2026
链接: https://arxiv.org/abs/2609.30904
作者: Yang Zhang,Wenhan Liu,Qiannan Zhu,Mingming Li,Yuanfei Huang
类目: Information Retrieval (cs.IR)
备注: EMNLP2026 Main
Abstract:Passage reranking plays a crucial role in information retrieval by refining the ordering of candidate passages to better reflect relevance. Existing listwise LLM rerankers with Chain-of-Thought (CoT) reasoning can handle complex queries effectively, but they suffer from substantial redundancy and high latency due to sliding-window strategies, which repeatedly generate highly similar CoTs. To address this, we propose QReason, a decoupled framework that separates query-focused reasoning from window-specific passage relevance assessment. Specifically, QReason introduces a dedicated rewriter that generates a ranking-oriented reasoning query once, capturing the query’s core intent while avoiding redundant reasoning, and then reuses it across all windows with a non-reasoning reranker. The rewriter is trained via a two-stage process that first uses supervised fine-tuning with relevant-passage guidance through semantic evidence to produce deeply grounded, query-focused CoTs. It then applies reinforcement learning to align CoT generation with both the inference-time setting and the reranking objective, optimizing listwise metrics and passage-level discrimination to produce reusable reasoning chains for reranking. Experiments on the BRIGHT benchmark demonstrate that QReason significantly reduces redundant reasoning, achieves ranking performance comparable to or better than strong reasoning-based rerankers, and outperforms existing query rewriting models.
[IR-7] RecToolBench: Benchmarking Recommendation-Specific Tool Orchestration under Fuzzy User Intent EMNLP2026
链接: https://arxiv.org/abs/2609.30717
作者: Xiao Chen,Yicheng Zhao,Yingying Wu,Zhendong Chu,Changyi Ma,Qingsong Wen,Xuan Song
类目: Information Retrieval (cs.IR)
备注: EMNLP 2026
Abstract:Recent advances in agentic recommender systems are shifting recommender systems from passive filtering engines to instruction-following agents that use external tools to resolve user intent. However, existing benchmarks often assume explicit user intent, simplified tool environments, or isolated function calls, leaving realistic tool orchestration for recommendation underexplored. To bridge this gap, we propose RecToolBench, a Model Context Protocol (MCP)-based benchmark for evaluating tool-using recommender agents under fuzzy user instructions. RecToolBench contains more than 1,200 executable tasks across three recommendation domains, 13 MCP servers, and 32 tools, spanning single-tool calls, parallel tool calls, sequential tool chains, and hybrid tool orchestration. We construct RecToolBench with a scalable synthesize–fuzzify–judge pipeline that generates executable fuzzy recommendation tasks, and evaluates agent trajectories using rule-based execution checks and rubric-based LLM evaluation. Experiments on representative LLMs show that syntactically valid tool calls do not guarantee successful recommendations. Models struggle with semantic parameter grounding, multi-step evidence integration, and grounded final recommendations, especially as orchestration complexity increases. Our results identify tool orchestration under fuzzy user intent as a major bottleneck for agentic recommender systems. Our data and code are available at this https URL.
[IR-8] Recommendation World Models for Future-State Control
链接: https://arxiv.org/abs/2609.30711
作者: Jinfeng Xu,Zheyu Chen,Ziyue Peng,Jianheng Tang,Zheng Lin,Jing Yang,Puzhen Wu,Zheng Xing,Victor C. M. Leung
类目: Information Retrieval (cs.IR)
备注:
Abstract:Sequential recommendation optimizes which items to rank, while each displayed slate also shapes subsequent feedback and user state. We study how a trained ranker can support decisions about these future consequences. We introduce UA-TWM, a utility-anchored world-model interface that constructs nearby slate actions, estimates their target-relevant consequences, and selects an alternative subject to utility constraints. The reference slate serves as a fallback when no alternative qualifies. A logged-replay instantiation combines utility and target-gain estimates with calibrated failure-risk prediction; a closed-loop instantiation uses one-step state-action prediction and updates its decisions after observed feedback. We evaluate transfer across twelve sequential backbones on MovieLens-25M and KuaiRand-Pure, and repeated target-directed interaction in KuaiSim. Attaching the interface improves Recall@20, NDCG@20, and future-state alignment for every matched logged backbone. Selection ablations reveal the utility and risk costs of aggressive target pursuit, while closed-loop diagnostics isolate the contribution of action-conditioned prediction. Local consequence modeling thus enables target-aware selection around a trained sequential ranker.
[IR-9] Component Benchmark: Hierarchical Model Profiling for Large-scale Recommendation Systems
链接: https://arxiv.org/abs/2609.30656
作者: Dharak Kharod,Yuzhen Huang,Zhou Wang,Jackie Xu,Fuzail Khan,Jacky Zhou,Hao Yan,Lidong Zhao,Xizhou Feng,Yvonne Liu,Karthik Jayaraman,Praveen Ramachandran,Vishwa Karia,Yashasvi Makin
类目: Information Retrieval (cs.IR)
备注:
Abstract:Large-scale recommendation models pose distinct, under-explored profiling challenges. Most recommendation model architectures are structurally heterogeneous, intermixing memory-bandwidth-bound operations, small compute-bound dense layers, dynamic shapes from jagged categorical features, and low-arithmetic-intensity operations. Recommendation models evolve rapidly as modeling engineers experiment with compositions, often written without visibility into hardware execution characteristics. Standard profiling tools offer either end-to-end throughput or operator-level traces, but cannot attribute performance to the submodules that practitioners reason about. We present Component Benchmark (CB), a profiling system that independently characterizes each submodule performance in a hierarchical manner, providing a tree-structured, interactive visualization that brings performance clarity to ML practitioners. At its core, CB provides a simple yet extensible, submodule-based benchmarking framework with a plugin architecture that enables hierarchical performance analysis. These large-scale recommendation models are TB-scale, run on thousands of GPUs and ingest 100B examples per day. We demonstrate CB’s effectiveness on common open sourced models and discuss how CB has been leveraged to accelerate modern recommendation model performance analysis and optimization.
[IR-10] Epstein Files Engine: Agent ic Search for Investigative Journalism
链接: https://arxiv.org/abs/2609.30611
作者: Duy K. Nguyen,Teresa Mondría Terol,Dylan Freedman,Zach Seward
类目: Human-Computer Interaction (cs.HC); Computation and Language (cs.CL); Computers and Society (cs.CY); Information Retrieval (cs.IR)
备注: 6 pages, 2 figures, 2 tables. Presented at the Computation + Journalism Symposium (C+J 2026)
Abstract:On Jan. 30, 2026, the U.S. Department of Justice released a mixed-media collection concerning Jeffrey Epstein, including about three million pages of PDFs. We describe the Epstein Files Engine, an A.I. agent The New York Times deployed to investigate the files. The Engine translated reporter questions into Google BigQuery SQL queries across three corpora: Epstein-related releases, the Times’s archive and external, Epstein-related news headlines. It used an LLM to plan queries and returned citation-rich answers a reporter could verify and trust. More than 100 journalists used the Engine, and it contributed to at least 20 published stories. We report how reporters queried it and describe Diff, our text-and-visual duplicate matching method that amplified novelty signals and allowed the Engine to surface genuinely new information. We argue that newsroom agents serve newsrooms best not as autonomous writers, but as interfaces to source material and institutional knowledge.
[IR-11] Embedding Subspace Partitioning for Dynamic Multi-Objective Retrieval RECSYS2026
链接: https://arxiv.org/abs/2609.30601
作者: Shaobo Zhang,Alice Leung,Yunxiang Ren,Ping Liu,Yuchin Juan,Qianqi Shen,Benjamin Le,Jianqiang Shen,Chengming Jiang,Ko-Cheng Wang,Vidya Krishnamurthy,Caleb Johnson,Fedor Borisyuk,Luke Simon,Jingwei Wu,Wenjing Zhang
类目: Information Retrieval (cs.IR)
备注: 10 pages. To appear in the 20th ACM Conference on Recommender Systems (RecSys 2026)
Abstract:Modern industrial recommender systems must optimize across competing objectives, balancing semantic relevance with business metrics such as engagement and revenue. While bi-encoders dominate large-scale retrieval due to their efficiency, they collapse these heterogeneous signals into a single static embedding space. This design creates a fundamental limitation: once trained, the retriever cannot adapt to shifting objective priorities at serving time without retraining. Moreover, joint optimization with multi-objective losses often induces interference between objectives, leading to suboptimal trade-offs. We propose Embedding Subspace Partitioning (ESP), a retrieval framework that decomposes the embedding into task-aware subspaces and replaces the single dot product with a weighted sum of per-subspace similarities, whose weights are tunable at serving time. For Transformer bi-encoders, ESP uses the model’s native end-of-sequence token as a segment delimiter, with segment-aware attention masking and position encoding resets to guarantee subspace isolation in a single forward pass. Serving is performed via GPU-accelerated exhaustive kNN over one concatenated index, eliminating the need for per-objective Approximate Nearest Neighbor (ANN) infrastructure required by multi-head approaches. We evaluate ESP on an open-source benchmark built from MS MARCO. A single ESP model traces a broad Pareto frontier, consistently outperforming strong multi-task baselines across diverse operating points. In LinkedIn’s job matching platform (70M+ weekly users), ESP enabled dynamic retrieval reconfiguration and delivered significant key business metric lifts.
[IR-12] -RoPE: Time-Aware Rotary Position Embedding for Sequential Recommendation
链接: https://arxiv.org/abs/2609.30576
作者: Yang Liu,Noel Loo,Ali Khanafer,Shuying Sun,Akshay Soni,Zhong Wu,Linjun Yang
类目: Artificial Intelligence (cs.AI); Information Retrieval (cs.IR); Machine Learning (cs.LG)
备注:
Abstract:Large-scale recommenders increasingly adopt the sequential generative recipe behind large language models, bringing the Transformer into recommendation along with design choices made for text, including Rotary Position Embedding (RoPE). In language models, RoPE encodes token indices for relative position reasoning, but in recommendation, an interaction index records only event order, saying nothing about elapsed time, behavioral cycles across scales, or calendar phase. We revisit this choice and propose T-RoPE, a time-aware RoPE for sequential generative recommendation that replaces index-only rotation with timestamp-based angles, learnable temporal coefficients, multiscale frequency banks, shifted query alignment, and non-stationary key rotation. We prove that standard RoPE, even on timestamps, remains time-translation invariant and cannot distinguish seasonal contexts, and that T-RoPE breaks this invariance while preserving the RoPE interface. Across five public benchmarks, T-RoPE achieves the best result on every metric on every dataset, improving over the strongest baseline by 78–130% in HR@10 on the sparse PixelRec data and 8–12% across metrics on Amazon Books. On an industrial-scale e-commerce dataset with more than 6B interactions, it improves every metric over the HSTU + Time RAB backbone by 13–82%, with ablations attributing the largest gains to multiscale frequencies ( +56% NDCG@50) and non-stationary keys ( +4% ). An online A/B test in the Shop app yields positive lifts in conversion rate ( +0.33% ) and order count ( +0.63% ). We also provide forward and backward algorithms whose added cost is linear in sequence length and head dimension, keeping time-aware RoPE practical for large generative recommenders.
[IR-13] Nearest but Not Dearest: Shared Curator-Feedback Infrastructure for Content-Only Search and Recommendation RECSYS2026
链接: https://arxiv.org/abs/2609.30568
作者: Matt Sandler
类目: Information Retrieval (cs.IR); Sound (cs.SD)
备注: 8 pages, 3 figures, 3 tables. Accepted for oral presentation at the Unified Search and Recommendation Workshop (USRW) at RecSys 2026; workshop is non-archival
Abstract:A deployed B2B music-discovery platform serves both query-driven search (text prompts, vibe tags) and seed-driven recommendation (seed-track and artist stations) over one licensed catalog, one LAION-CLAP joint audio-text embedding space, one candidate-generation filter, and one ranking head – and neither path consumes end-listener behavioral signal. In this content-only regime, curator judgment is the principal feedback signal available, and offline cosine similarity predicts it poorly: 38% of cosine-nearest neighbors are rejected by curators. The rejections reveal a clean partition: a majority (55%) are sound failures the encoder could address (style, tempo, mood mismatch), and a substantial minority (37%) are context failures orthogonal to the waveform (wrong language, holiday content, devotional content, rights and lyric flags). We deploy this sound-vs-context decomposition as feedback infrastructure, routing each failure mode to the layer that can absorb it: context failures to a constraint filter at candidate generation, sound failures to an embedding reweighting head at the representation layer – both below the search/recommendation split, so a single curator loop maintains both experiences. On 1,200 curator judgments collected over two production rounds one month apart, the combined intervention reduces rejection rate from 38.17% to 28.83% (-24.5% relative, McNemar chi-squared = 22.4, p = 2.2e-6). An accounting decomposition attributes 4.08 pp of the drop to filter-eligible categories and 5.25 pp to the rest; the deployment was unblinded and compound, so this is a production accounting bound, not a causal estimate. We present this as an industrial case study rather than a validated general method, and close with lessons for content-only discovery: the failure partition is orthogonal to the paradigm partition, and corrections land in layers shared by both paradigms.
[IR-14] REALMS: An AI-Assistant Conversational System for Real-Time Exact Audience Sizing over High-Dimensional Nested Profiles ICDM2026
链接: https://arxiv.org/abs/2609.30547
作者: Haixu Ma,Aditya Bansal,Shubham Lohiya,Sumit Ranjan
类目: Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注: Accepted by ICDM 2026
Abstract:Audience sizing is a critical component of digital marketing. It enables precise resource allocation, campaign planning, and performance optimization. Traditional approaches using skeleton audiences, sampling, or predictive modeling suffer from significant delays, estimation errors, and poor scalability over high-dimensional profile data. We present REALMS (Real-time Exact Audience sizing via LLM-based Multi-attribute Search), a conversational system for exact audience sizing deployed in production on an enterprise customer data platform. REALMS enables marketers to query massive profile stores with millions of profiles and thousands of attributes using natural language and receive precise counts in seconds. The system introduces three key components: (1) a categorical attribute retrieval mechanism using embedding-based vector search to dynamically identify relevant schema attributes without manual configuration; (2) an LLM-powered NL2SQL pipeline with template-based in-context learning for accurate query generation over complex nested schemas; and (3) schema standardization enabling industry-agnostic deployment across diverse enterprise environments. Evaluation on real enterprise data demonstrates strong recall for attribute retrieval, high SQL execution accuracy, and low latency, which enables real-time interactive audience insights where prior methods required hours.
[IR-15] AutoResearch at Production Scale: Failure Modes and a Multi-Agent Framework ICDM2026
链接: https://arxiv.org/abs/2609.30541
作者: Aparajith Chandran,Juwon Kim,Saurav Jha,Pablo Castells,Florian Hottier
类目: Machine Learning (cs.LG); Information Retrieval (cs.IR)
备注: 10 pages, 3 figures. Accepted at IEEE ICDM 2026 (Applied Track)
Abstract:Optimizing embedding systems for production recommendation pipelines demands systematic exploration that consumes disproportionate engineering effort at scale. We apply Andrej Karpathy’s AutoResearch paradigm – a large language model that iteratively edits a training script and retains modifications that improve a held-out scalar metric – to automate this exploration. We report on twelve weeks of running this paradigm at production scale, where iterations consume hours of multi-GPU compute, evaluation involves competing criteria, and campaigns span weeks across many training jobs. Across two independently developed representation-learning systems for a book recommendation pipeline, we ran 220+ experiments and observed five recurring failure modes absent from the original setting: infrastructure fragility, agent memory decay, search-direction stagnation, iteration-cost asymmetry, and metric fixation. We contribute a three-principle scaffolding design – prevent, persist, redirect – that maps each failure mode to a structural remedy and whose instantiation scales with iteration cost. The framework produced a 1.82x Recall@6 lift and a 2.1x coherence lift over hand-tuned baselines, and the agent autonomously designed a text-only fallback that expanded catalog coverage by 5.8x. The two systems span nearly three orders of magnitude in per-iteration cost yet exhibit the same failure modes, suggesting these are structural properties of production-scale autonomous research rather than artifacts of either application.
[IR-16] Where Does Retrieval-Based Open-Ended Evaluation Fail? Automatic Taxonomy Induction from Long-Form Medical Answer Factuality Verification DATE
链接: https://arxiv.org/abs/2609.30467
作者: Heyuan Huang,Jirui Dai,Alexandra DeLucia,Sonal Joshi,Mahsa Yarmohammadi,Jie Gao,Bernal Jiménez Gutiérrez,Mark Dredze
类目: Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注: Experiments’ corpus knowledge cutoff date May 2026
Abstract:Retrieval-based factuality evaluation, where LLM-generated claims are verified against evidence from authoritative medical corpora, has become the dominant paradigm for scalable hallucination detection in high-stakes clinical settings. Despite the urgency of reliable and transparent medical fact verification, most systems measure performance with aggregate metrics like F1, which obscure where and why failures occur. Existing RAG diagnostics require gold answers or annotated gold evidence, neither of which exists in this regime. We introduce two comprehensive taxonomies, grounded in a case study on the open-ended MedExpert dataset and 3 closed-ended datasets, decomposing failures into retrieval-stage errors along five quality dimensions, and verifier-reasoning errors into six consecutive steps. We adapt an automatic pattern induction pipeline using LLM-as-Judge to label evidence quality and classify verifier reasoning errors at scale, and then stress-test our findings across 4 retrieval methods and 6 frontier verifier models. Our analysis reveals that scaling model size, adding reasoning effort, expanding to authoritative web sources, and applying medical fine-tuning do not resolve these failure modes, demonstrating that they represent fundamental limitations of the retrieve-then-verify paradigm in open-ended medical settings rather than artifacts of outdated systems. We release our code and data at this https URL for the full reproducibility of our results.
[IR-17] Bootstrapping Conversational Recommendation Agents At Spotify: Synthetic Data Generation and Self-Improvement Loops
链接: https://arxiv.org/abs/2609.30297
作者: Enrico Palumbo,Alexandre Tamborrino,Victor Ode,Ben Lacker,Adrià Casas Escoda,Jeremy Hopple,Marcus Better,James Leoni,Hugo Galvão,Hugues Bouchard,Mounia Lalmas,José Luis Redondo García,Abenezer Abebe,Ann Clifton,Anton Blomberg,Henrik Lindström,Dani Doro,Christine Doig Cardet
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
备注:
Abstract:Conversational recommendation agents are a new paradigm for content discovery, enabling users to express complex intents through natural language (e.g., “recommend Italian indie artists I haven’t heard before”). A central challenge in building such agents is optimizing agent planning – deciding how to select, sequence, and invoke tools – particularly in cold-start settings where real user interactions are not yet available. We introduce a pipeline for multi-turn synthetic data generation and a self-improvement loop to address this challenge. The synthetic data pipeline transforms single-turn prompts into realistic multi-turn conversations, enabling systematic evaluation before launch. The self-improvement loop combines variance-based contrastive optimization with iterative refinement through a coding agent, automatically identifying and fixing planning and tool-use errors. Our approach improves quality by +8% on top of a highly optimized manual prompt. The system has been productionized and significantly accelerated iteration cycles for the launch of a conversational recommendation agent at Spotify. Online A/B tests demonstrate its effectiveness, with +14% user listening, +5% increase in weekly active users, and a 5% reduction in skip rate compared to a prior experience supporting only session refinement. This work provides a practical framework for accelerating the development of conversational recommendation agents in industry.
[IR-18] SignTrace: Describe a Sign Find the Word
链接: https://arxiv.org/abs/2609.30295
作者: Zengji Tu,Xingye Zhu,Ningjing Wang,Tingyi Huang,Yangjunfeng Zhu,Dai Wan
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
备注: Includes reproducibility data and method documentation
Abstract:Identifying an unfamiliar sign is difficult when a learner remembers its movement but does not know its meaning or formal feature codes. SignTrace addresses this longstanding reverse-lookup problem through natural-language access to a Chinese sign-language dictionary. The system integrates LLM-based dictionary enrichment, action extraction, dictionary-style rewriting, seven-channel retrieval, and candidate reranking over 6,699 entries. It has been deployed for user trials and has received positive informal feedback. Evaluation on a dictionary-derived benchmark of 500 movement-description queries yields 94.0% Hit@1, 97.4% Hit@9, and a mean reciprocal rank of 0.9540. Reranking increases Hit@1 from 71.8% to 94.0%, while component analyses show the contribution of enriched entry descriptions. Median query-processing time is 13.37 seconds with six concurrent queries. By connecting everyday movement descriptions to documented signs and meanings, SignTrace provides a practical tool for identifying unfamiliar signs. Dictionary-derived wording and prior selection within the benchmark limit generalization to descriptions independently produced by users.
人机交互
[HC-0] Adapting for AI: How elementary teachers adjust their practices for an AI-integrated curriculum
链接: https://arxiv.org/abs/2609.31569
作者: Fasika Melese,Ruiyang Wu,Xinyue Cui,Joanna Perkins,Xiaoyi Tian,Tiffany Barnes,Shiyan Jiang
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注:
Abstract:Conversational AI tools are entering children’s everyday experiences, and schools are interested in adopting them. However, successful classroom integration depends not only on the technology but also on the work teachers do to make it usable and appropriate for their students and classroom context. There is little known about how elementary teachers work as they implement conversational AI tools in real classrooms. In this study, we examine three teachers’ experiences implementing an AI literacy and English Language Arts (ELA) curriculum built around ToyTalk, a conversational AI toy development platform, over 13 instructional days, a three-week summer camp. Drawing on daily individual reflections, group reflections, and post-camp interviews, we find that teachers’ adaptive practices of repair, differentiation, translation, and balancing sit at the intersection of three tensions (technology, learner, and instruction). Teachers’ understanding of AI and their role evolved over the camp experiences. From these findings, we contribute design implications and considerations for deploying conversational AI within elementary classrooms.
[HC-1] PANEL: An Open-Source Self-Hosted Web Platform for Human Evaluation of Generative Models
链接: https://arxiv.org/abs/2609.31392
作者: Matteo Spanio,Andrea Poltronieri,Mart’ın Rocamora
类目: Human-Computer Interaction (cs.HC); Sound (cs.SD)
备注: 3 pages, 2 figures, ISMIR 2026 Late Breaking Demo
Abstract:Human judgement is the reference measure for evaluating generative models, yet the software used to collect it lags behing the methodology. Researchers adapt listening-test frameworks designed for perceptual protocols such as MUSHRA, rely on closed commercial survey platforms, or implement single-use web applications. Live arenas such as Chatbot Arena and Music Arena rank publicly deployed systems at scale, but do not support controlled comparisons of a laboratory’s own models with its own participants. We present PANEL, an open-source, self-hosted platform for such studies. A study is authored in the browser and distributed as a single link, with audio, video, image, and text stimuli, seven question types, and screening and skip logic. The platform reports per-question summaries, across-condition significance tests, pairwise win rates and Bradley–Terry scores, and supports power analysis from pilot data. Consent versioning, self-service withdrawal, retention enforcement, and audit logging support GDPR-compliant operation. Each study exports as a machine-readable specification. PANEL is available at this https URL.
[HC-2] EEG-based Word Association Paradigm for Adult ADHD Screening: An Exploratory Pilot Study
链接: https://arxiv.org/abs/2609.31359
作者: Caroline Peng,Tony Russell-Rose
类目: Human-Computer Interaction (cs.HC)
备注: 8 pages, 2 tables
Abstract:With the prevalence of Attention Deficit Hyperactivity Disorder (ADHD) over the past decades, healthcare systems across the globe face critical diagnostic challenges due to long diagnostic waiting times and a reliance on subjective behavioural assessments that cannot distinguish ADHD from comorbid psychiatric disorders, especially for adult patients. This exploratory study investigates whether EEG-based word association paradigms show promise as a complementary approach to screening for ADHD in adults. Using a mixed-method approach, the study examines neurological and cognitive differences between neurotypical individuals (NT), clinically diagnosed ADHD participants (ND), and self-reported ADHD cases awaiting formal diagnosis (SR) across three word association tasks. While established EEG biomarkers (ERP N400, Theta/Beta Ratio, and Alpha Suppression) show no significant group differences, semantic distance analysis reveals a statistically significant main effect ( p=0.003 ), with the SR group showing the most divergent associations. These preliminary findings suggest that word association paradigms may capture cognitive differences not detected by standard EEG metrics and encourage further large-scale investigation as a potential complement to existing adult ADHD screening tools.
[HC-3] Cognitive Skills in the Age of AI: Computing Students and Experts Perceptions
链接: https://arxiv.org/abs/2609.31272
作者: Neha Rani,Vu Minh Anh Le,Austin M. Spangler,Erta Cenko
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注: This article is accepted at the 26th IEEE International Conference on Advanced Learning Technologies, 2026
Abstract:AI is becoming increasingly integrated into daily workflows, especially in computing. We are gradually shifting towards an AI-rich future, an impending yet unknown one. One important emerging concern is whether we are accordingly preparing our future computing workforce. Further, we need to know what the important cognitive skills are to remain relevant in the computing workforce and if there are changes in cognitive skill importance. To investigate this direction, we conducted a mixed-methods study, collecting perceptions from computing students and computing experts regarding the importance of cognitive skills in the past, present, and future. We report that the perceived importance of most cognitive skills will decrease in the future, with an AI-rich environment, but critical thinking skills remain important. Further, we report reasons collected through interviews on why the importance of cognitive skills will change and how future computing students can prepare for it.
[HC-4] Evaluating the Impact of Adaptive Extended Reality on Human-Robot Interaction Across the Reality-Virtuality Continuum
链接: https://arxiv.org/abs/2609.31138
作者: Carl Tornberg,Alicia Torck,Lotfi El Hafi,Tadahiro Taniguchi
类目: Robotics (cs.RO); Human-Computer Interaction (cs.HC)
备注: Submitted to Advanced Robotics, Special Issue on “Next Generation Cognitive Robotics: Nurturing Embodied Intelligence for a Symbiotic Future with Humans and AI”
Abstract:As populations in developed countries age and labor shortages intensify, Cybernetic Avatars (CAs) are proposed to extend human capabilities through robotic embodiments, requiring effective Human-Robot Interaction (HRI) frameworks. Extended Reality (XR), an umbrella term for Augmented Reality (AR), Augmented Virtuality (AV), and Virtual Reality (VR), offers such interfaces, but prior research typically fixes the XR modality without evaluating its effect on task outcomes. This study examines whether the XR modality impacts HRI performance and whether an adaptive interface adjusting the level of virtuality along the Reality-Virtuality Continuum (RVC) at runtime improves it. A custom XR application interfaced with a mobile manipulator supports immersive control and runtime modality switching. In a within-participant multi-room pick-and-place experiment comparing fixed AR, AV, and VR with dynamic RVC through task metrics, the NASA-TLX, and the System Usability Scale (SUS), this study demonstrates that 1) the fixed reality modality affects HRI results, and 2) dynamically changing the modality along the RVC improves them. AR yielded significantly lower mental demand, effort, and frustration than AV and VR, while the dynamic RVC condition achieved the highest throughput and lowest workload, highlighting the value of adaptive XR interfaces for human-robot symbiosis. The implementation is available at this https URL.
[HC-5] Comparative Evaluation of an XR Pen-based Control Interface for Semi-Autonomous Mobile Robot Navigation in Service Environments
链接: https://arxiv.org/abs/2609.31117
作者: Alicia Torc,Carl Tornberg,Eric Piette,Renaud Ronsse,Benoit Macq,Gustavo Alfonso Garcia Ricardez,Lotfi El Hafi,Tadahiro Taniguchi
类目: Robotics (cs.RO); Human-Computer Interaction (cs.HC)
备注: Submitted for presentation at the 2027 IEEE/SICE International Symposium on System Integration (SII), Kobe, Japan
Abstract:Service robots remain difficult to deploy in domestic environments, partly because fully autonomous operation is not yet reliable in unpredictable surroundings, and partly because conventional control methods remain inaccessible to novice users. Extended Reality (XR) enables operators to visualize robot information overlaid onto the real world and to interact with augmented elements. Yet, common XR control methods, such as motion controllers and hand gestures, are still perceived as unintuitive. This paper presents a control interface that uses a commercial XR pen to command a semi-autonomous mobile robot in Augmented Reality (AR): the operator points at a position in the room, selects it, and drags an augmented arrow to set the desired orientation of the robot at this destination. Two additional interfaces, based on the XR motion controllers and hand gestures, were developed within the same framework. To assess the performance and users’ perception of these interfaces, and of the XR pen in particular, a study with 10 participants compared four control methods, i.e., the XR pen, the XR motion controllers, hand gestures, and a computer-based baseline RViz, in navigation tasks performed in a home-like environment. Results show that the XR pen significantly outperforms the other methods in task selection time with the most consistent selections, and that the XR motion controllers obtain the best perceived workload and usability scores, ahead of the computer-based baseline, supporting XR-based control as an intuitive alternative for novice users. However, technical limitations in the integration of the recently released XR pen currently hold back its user experience.
[HC-6] Confident Not Wiser: The Dunning-Kruger Effect in Human-AI Interaction
链接: https://arxiv.org/abs/2609.31095
作者: Daniela Fernandes,Michelle Rausch,Agnes Mercedes Kloft,Daniel Buschek,Robin Welsch
类目: Human-Computer Interaction (cs.HC)
备注: 37 pages, 10 figures, 12 tables
Abstract:AI assistance can improve performance without improving self-assessment. We report a study (N=366) comparing Human alone and Human+AI performance on reasoning tasks, for which the AI model is benchmarked on the same items. Participants estimated global and block performance and rated confidence in their answers. Human+AI achieved higher scores, but self-estimates tracked performance weakly. Average overestimation was similar across groups, covering individual errors. Across tasks, confidence distinguished correct from incorrect answers less accurately in the Human+AI group, while within-task differences remained uncertain. The Dunning-Kruger pattern was found in both groups, with a larger observed contrast in Human+AI. Controls for score noise reduced but did not eliminate the pattern, with the controlled group difference remaining inconclusive. An extended computational account describes global and block estimates. Our findings distinguish performance augmentation from metacognitive augmentation and motivate interfaces that support verification, communicate task-specific AI model performance, and help users evaluate the quality of their joint work rather than produce answers.
[HC-7] From Segments to Trajectories: Evolving Affective Graphs with Evidence Retrieval for Continuous EEG Emotion Recognition
链接: https://arxiv.org/abs/2609.30890
作者: Chi Yang,Jihong Wang,Chengxi Xie,Kai He,Huan Liu,Man Yao,Shile Qi,Yuzhe Zhang
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Electroencephalography (EEG)-based emotion recognition is important for affective computing and human-computer interaction, yet most existing methods divide a long trial into short segments and assign each segment the label of its source trial. Although this strategy increases the number of training samples, it reduces an evolving emotional response to a segment-level, coarse-grained, and static prediction problem. In reality, emotion may continuously emerge, intensify, weaken, and fluctuate as a stimulus unfolds, motivating the prediction of a time-aligned affective trajectory from the complete EEG trial. This task requires coordinated modeling of how spatial neural organization evolves throughout the trial and how local emotional fluctuations interact with longer-term trends. In this work, we formally define and systematically investigate continuous EEG emotion recognition as whole-trial affective trajectory prediction. We propose EAGER, an Evolving Affective Graph framework with Evidence Retrieval for continuous EEG emotion recognition. EAGER comprises two complementary modules: Affective State-guided Topology Evolution models the evolving spatial organization of EEG activity, while Multi-scale Temporal Evidence Retrieval integrates short-term fluctuations with longer-range temporal trends for time-aligned prediction. Experiments on MAHNOB-HCI, SEED-VII, and REFED show consistent gains in trajectory-tracking metrics over representative methods, with competitive pointwise errors.
[HC-8] CDBG: Causally Motivated Dual-Invariance Learning against Topological and Predictive Shifts in EEG Workload Recognition
链接: https://arxiv.org/abs/2609.30831
作者: Yuzhe Zhang,Wenmin Zhou,Chengxi Xie,Kai He,Jihong Wang,Huan Liu,Man Yao,Daoqiang Zhang
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Generalizing Electroencephalography (EEG)-based mental workload recognition to unseen subjects remains a formidable challenge due to severe inter-subject variability. While functional brain graphs effectively model distributed cognitive dynamics, their inherent subject-specificity induces two coupled distribution shifts: a class-conditional topological shift in the underlying functional connectivity, and a predictive mechanism shift in the learned representation-to-label mapping. Motivated by the subject-induced distribution shifts, we propose CDBG, a Causally motivated Dual-invariance learning framework for Brain Graphs. CDBG disentangles and mitigates these shifts via a two-stage rationale learning pipeline. First, it employs stochastic edge masking to extract sparse, workload-predictive graph rationales, regularized by workload-conditional Laplacian spectral alignment to enforce topological invariance across subjects. Second, it applies subject-wise Invariant Risk Minimization (IRM) to the graph representations, ensuring environment-wise risk stationarity. Extensive experiments on a self-built air traffic controller EEG cognitive workload dataset and multiple public datasets under a strict leave-one-subject-out protocol demonstrate that CDBG significantly outperforms state-of-the-art cross-subject and graph-based baselines, improving the Macro-F1 score by up to 4.23%, while simultaneously providing neurophysiologically interpretable functional rationales.
[HC-9] Sampling Safe Futures: Multimodal Trajectory Planning for Personalized Safety in Anthropomorphic AI
链接: https://arxiv.org/abs/2609.30780
作者: Benedetta Picano,Dusit Niyato
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Anthropomorphic artificial intelligence systems increasingly remember personal details, display empathy, and are engaged with as social counterparts, creating forms of risk that emerge from the evolution of the user-system relationship over time. Existing safeguards largely operate at the level of individual conversational turns and cannot determine whether a sequence of seemingly acceptable interactions is cumulatively moving a particular user toward harm. This paper introduces personalized trajectory-level safety, a framework that treats relational safety as a sequential decision problem over a latent escalation state inferred from the user’s messages and influenced by the system’s responses. At each turn, a screening step first discards any response strategy that does not preserve at least one safe continuation of the interaction under every plausible model of the user. Among the remaining strategies, we formulate action selection as multimodal trajectory sampling, and use a Generative Flow Network to generate diverse future evolutions in proportion to their plausibility, safety, and utility. The system then selects the strategy that preserves the largest fraction of safe and useful continuations. We evaluate the framework in simulation, calibrated on statistics reported for real human-chatbot interactions, and using response strategies derived from public benchmarks. Results show that trajectory-aware decision making substantially reduces the frequency of harmful states while keeping helpful interaction. This work reframes safety for anthropomorphic AI from response-level filtering to personalized control over the future evolution of human-AI relationships. The source code is available at this https URL.
[HC-10] Skill Profiling with Attributable Reasoning (SPAR): A Wearable Analysis System for Boxing
链接: https://arxiv.org/abs/2609.30753
作者: Nibraas Khan,Hanchen David Wang,Enya Bullard,Ritam Ghosh,Ruj Haan,Aarav Agrawal,Meiyi Ma,Nilanjan Sarkar
类目: Human-Computer Interaction (cs.HC); Machine Learning (cs.LG)
备注:
Abstract:A punch is a ballistic, full-body action driven by a kinetic chain running from the legs through the trunk to the arm, where a small sequencing error separates a scoring strike from a miss. Wearable sensors can capture this movement in the gym, but most deployable systems only classify which punch was thrown rather than assess how well it was thrown. We present Skill Profiling with Attributable Reasoning (SPAR), an eight-IMU garment and pressure-insole system that classifies each punch as expert or novice and treats an explanation of that prediction as feedback. Feedback is only useful if the person receiving it can act on it, so SPAR explains the prediction at three tiers, a per-joint attribution for the analyst, a counterfactual over kinetic-chain layers for the coach, and a plain-language narrative of the two for the athlete. Across 17 participants and 4,713 punches, SPAR reaches a leave-one-participant-out AUC of 0.842 (95% CI [0.769, 0.907] over participants). A frozen time-series foundation model encodes the joint-angle and plantar-force series, and a small transformer trained on the cohort classifies the encoding. We audit the two quantitative tiers and report six themes from a thematic analysis of interviews with six practicing boxing coaches.
[HC-11] Beyond the Last Truffula Tree: SustainAI - A Water-Aware Closed-Loop Framework for Environmentally Accountable AI
链接: https://arxiv.org/abs/2609.30747
作者: Farnaz Farid,Tashfia Towkee,Sania Nasreen,Sami bin Azad
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC); Human-Computer Interaction (cs.HC)
备注: 10 pages, 2 Figures
Abstract:As artificial intelligence (AI) becomes embedded in everyday life, its environmental footprint, particularly water consumption remains largely invisible. While energy and carbon impacts are widely recognized, the substantial freshwater demands of data center cooling and electricity generation receive little attention. To address this gap, we introduce SustainAI, a water-aware, closed-loop framework incorporating environmental accountability into AI deployment. SustainAI integrates real-time water metering, a hallucination-aware penalty model, and a water-aware routing algorithm that accounts for regional water stress. Evaluated via Small Language Models (SLMs) extracting health misinformation, results reveal an 11-fold variation in water footprint across geographically distributed data centers (0.0477 mL to 0.5360 mL per inference). Across 1,335 inference runs, the system consumed approximately 399 mL of water but produced only 240 correct outputs, demonstrating that substantial resources are spent on inaccurate responses. Crucially, SustainAI extends beyond technical optimization through a Care by Design lens, framing AI sustainability around relational ethics, regional equity, and ecological stewardship. By combining water monitoring, adaptive accountability, and Care by Design principles, SustainAI provides a practical foundation for integrating ethical care and environmental responsibility into AI infrastructure design and lifecycle management.
[HC-12] “If Youre Not Doing It Somebody Else Is”: Active Negotiation and the Invisible Labor of Sustained LLM Use
链接: https://arxiv.org/abs/2609.30699
作者: Matt Viana,Patrick Erickson,Shomir Wilson,Dana Calacci
类目: Human-Computer Interaction (cs.HC)
备注: 28 pages, 3 figures, 4 tables. Under review
Abstract:Large language models (LLMs) have become fixtures of academic work even as their users describe them as degrading their writing, thinking, and skills. Dominant adoption frameworks read continued use as evidence of satisfaction, and cannot explain continued use of a distrusted tool. We interviewed 36 graduate student workers, balanced between English-as-a-foreign-language (EFL) and non-EFL speakers, and introduce the Active Negotiation framework: a model of sustained LLM use as a recurring cycle of risk, mitigation, and justification. A failure surfaces a risk, mitigation labor addresses it, and a justification renders the residual risk tolerable until the next failure reopens the cycle. The cycle runs across three dimensions: practical, auditing output; internal, auditing one’s own cognition and identity; and social, managing how peers and institutions perceive use. EFL participants invoke linguistic parity as a further justification. We reframe continued adoption as compliance sustained by invisible labor.
[HC-13] Multi-Objective Human-in-the-Loop Bayesian Optimization of a Lower-Limb Exoskeleton
链接: https://arxiv.org/abs/2609.30695
作者: Neil Janwani,Matthew T. Lerner,Aaron J. Young,Maegan Tucker
类目: Robotics (cs.RO); Human-Computer Interaction (cs.HC)
备注:
Abstract:Human-in-the-loop optimization (HILO) is a common approach for optimizing the control of assistive devices to account for the wearer’s unique biomechanics and subjective preferences. However, despite research suggesting that a person may have a different prioritization of objectives depending on time-varying factors such as the environment, their mood, or energy levels, existing HILO approaches only consider a single objective or enforce a fixed weighting on a set of objectives. Neither approach is capable of representing an individual’s preferences over objectives. In this work, we propose Multi-Objective Human-in-the-loop Bayesian Optimization (MO-HILBO), which builds on explicit multi-objective Bayesian optimization to efficiently infer a personalized set of Pareto-optimal controllers. We compare our approach with an existing multi-objective HILO method and experimentally demonstrate MO-HILBO on a lower-limb exoskeleton across two objectives: metabolic cost (efficiency) and ordinal human feedback (comfort). We find that MO-HILBO (1) discovers Pareto-optimal controllers, and (2) that the pairwise ordering of points on the Pareto front itself is consistent with validation trials. Lastly, we open-source mohilo, a Python package for running both HILO and MO-HILBO on wearable devices: this https URL.
[HC-14] CraftTrace: Unflattening Videos into Malleable Creation-Inspired Structures for Generative Editing
链接: https://arxiv.org/abs/2609.30623
作者: Boyu Li,Yuqian Zhou,Duotun Wang,Ding Li,Zhe Lin,Nanxuan Zhao,Zeyu Wang,Lin-Ping Yuan,Hongbo Fu
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Recent generative video editing models enable video content modification (e.g., changing a character) but target short clips. Extending them to full multi-shot videos requires tedious work to locate relevant content across shots, segment it into clips, craft context-aware editing prompts for each clip, and repeatedly articulate complex editing intent. To address this, we explore an interaction paradigm for editing through underlying video structures (e.g., scripts, scenes, characters, shots, and their relationships). We present CraftTrace, an interactive prototype that transforms a video into a malleable, multilevel structure for generative editing. Users work in task-centric workspaces to modify elements or reshape relationships, while an AI agent translates and propagates changes across the video. A user study and expert review show that this structure helps users understand videos, formulate and refine editing intent, and explore alternatives, supporting rapid prototyping during early-stage exploration and full video post-production.
[HC-15] Steering Versus Teleporting in Mobile Virtual Reality
链接: https://arxiv.org/abs/2609.30620
作者: Kristen Grinyer,Daniel Zielasko,Robert J. Teather
类目: Human-Computer Interaction (cs.HC)
备注: to appear in VRST 2026
Abstract:Mobile virtual reality (MVR) provides low-cost access to extended reality (XR), but its limited input restricts use of common locomotion techniques such as head-decoupled and velocity-controlled steering. Using a low-cost controller with an audio-based button, we compared gaze- and controller-directed steering and teleportation in MVR. We included a controller pitch-based speed-control technique enabling continuous head-decoupled steering. Performance and participant feedback indicate that gaze was better suited to teleportation, whereas controller pointing better supports steering in primed search. We compare with similar techniques in standard VR and derive design considerations for low-cost, narrow-FOV hardware, demonstrating the potential to support varied travel techniques and velocity control in low-fidelity XR.
[HC-16] Orchestrating GenAI for Interdisciplinary Research
链接: https://arxiv.org/abs/2609.30588
作者: Shirley Anugrah Hayati,Moyan Zhou,Patricia Anugrah Setiani,Ruizi Wang,Joseph Chee Chang,Dongyeop Kang
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:As researchers tackle interdisciplinary problems, they face the need to deepen expertise in primary areas while rapidly acquiring knowledge in secondary domains. Generative AI (GenAI) is increasingly positioned to meet this need, from general-purpose chat assistants to Deep Research tools marketed as autonomous research agents. Prior work has examined how researchers use GenAI to support single-discipline or general research tasks. However, we know little about the goals and GenAI practices in interdisciplinary research. We conducted a longitudinal study and semi-structured interviews with 15 interdisciplinary researchers to examine how interdisciplinary researchers actually orchestrate GenAI. Findings show that researchers leaned on GenAI to fill knowledge gaps while maintaining epistemic agency for novelty discovery. We also uncovered an expertise paradox: GenAI outputs were hardest to verify when most needed. Our empirical insights motivate GenAI designs that calibrate verification to researchers’ expertise, nudge toward cross-domain synthesis, and adapt prompting and outputs to disciplinary conventions.
[HC-17] Redesigning Trust: Replacing Dark Patterns with Fair Choice Architecture in Financial Interfaces ESORICS’26
链接: https://arxiv.org/abs/2609.30475
作者: Oluwadamilola Awakan,Tawan Aroonwechkul,Roshan Gunjoor,Supriya Khadka,Sanchari Das
类目: Human-Computer Interaction (cs.HC)
备注: Accepted at HumSec Workshop, ESORICS '26
Abstract:Digital financial platforms make enrollment effortless and cancellation laborious. This asymmetry is a dark pattern that manipulates users who have already decided to leave. Existing work identifies such patterns after deployment, and regulators sanction them after harm, yet neither provides designers with a criterion for building interfaces that avoid manipulation. We model the provider as an adversary whose instrument is effort and express fairness as a constraint requiring that leaving never cost more than joining. Defining interaction cost over navigation steps, mandatory inputs, and confirmation prompts, we prove that this constraint holds for every assignment of effort weights if and only if no component of the exit flow exceeds its counterpart at entry. Fairness is therefore verifiable by counting rather than by estimating cognitive effort, and exact equivalence is unnecessary because exit legitimately requires fewer inputs than entry. We instantiate the model, together with invariants for visual parity and linguistic neutrality, in a mobile credit card prototype with parallel sign-up and cancellation workflows.
[HC-18] A Benchmarking Framework for Context-aware XR Interfaces
链接: https://arxiv.org/abs/2609.30466
作者: Hyunsung Cho,Sarah Yewon Yun,Nancy Ruonan Sun,Ben Lafreniere,Mark Parent,Kashyap Todi,Tanya R. Jonker,Hrvoje Benko,Sherry Tongshuang Wu,David Lindlbauer
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注: 16 pages, 5 figures, UIST 2026
Abstract:Everyday Extended Reality (XR) systems aim to provide context-aware access to the right functionalities at the right time and place, with minimal manual reconfiguration as users switch context. Yet these interfaces are hard to evaluate: current prototyping and user-study workflows offer no systematic, repeatable way to compare adaptation methods across users and scenarios. We present ContextXR, a novel benchmarking framework for context-aware XR interfaces. ContextXR represents an XR application as a connected graph of functional facets, each a semantically coherent group of related capabilities that together support a shared user intent. On this representation, we build MineXR++, a dataset augmenting prior XR interface data with facet-level annotations, and formulate three canonical tasks of context-aware suggestion: context factor analysis, initial facet suggestion, and next facet suggestion. Our evaluation protocol scores suggestion methods by a simulated interaction metric, the navigation and search cost of reaching the desired functionality. Through experiments benchmarking global popularity, relational retrieval, and LLM-based methods, we demonstrate that ContextXR enables the systematic, reproducible evaluation of context-aware XR interfaces.
[HC-19] Who Acts When the User Is Gone? Digital Remains Survivor Claims and Post-Mortem Governance
链接: https://arxiv.org/abs/2609.30449
作者: Supriya Khadka,Dhiman Goswami,Sanchari Das
类目: Human-Computer Interaction (cs.HC)
备注: Accepted to CSCW Companion '26
Abstract:Digital systems continue to govern accounts, devices, data, and recovery channels after an account holder dies, leaving survivors to manage digital remains through mechanisms built around a living user. We examine post-mortem digital governance as a sociotechnical problem of cooperative and contested work, focusing on who acts when the user is gone, what claims they make, and what barriers shape recovery, preservation, closure, and protection. We conducted a content analysis of 800 Reddit posts about post-mortem digital privacy and security, coding posts across assets, actors, actions, privacy tensions, access barriers, policy gaps, emotional contexts, and risks. Findings show that phones/devices often act as gateways to other digital remains, socially connected actors make most claims, and data loss emerges as a central harm. We synthesize these findings into a Post-Mortem Digital Governance Framework for designing mechanisms that support survivor coordination while limiting access by purpose, asset, actor, and context.
[HC-20] he Interviewers Perspective: Unpacking the Impact of Real-Time AI Interviewing Assistance on Social Dynamics
链接: https://arxiv.org/abs/2609.30388
作者: Zhe Liu,Jiamin Dai,Joanna McGrenere
类目: Human-Computer Interaction (cs.HC)
备注: 23 pages, 4 tables, 5 figures, under review for CHI 2027
Abstract:Eliciting rich data in semi-structured interviews is cognitively demanding, prompting recent work to explore real-time AI assistance for interviewers. However, introducing AI into the interviewer-interviewee interaction creates a triadic context whose social dynamics remain underexplored. We investigated how interviewers experience AI assistance for probing during semi-structured interviews. To elicit rich participant reflections, we implemented two variants of AI assistance differing in initiation and granularity in a high-fidelity prototype, ProbeAssist. We conducted a qualitative-first comparative structured observation study where 18 participants each completed three simulated interviews: one without AI and two with different AI variants. Findings showed that participants leveraged AI as a supportive tool but resisted it as an assessor or competitor. As they navigated AI’s benefits and interaction costs, tensions emerged around agency, ownership, creativity, and interpersonal communication. We propose three implications for AI-assisted human-to-human interaction: managing social pressure, balancing idea alignment with inspiration, and preserving interpersonal presence.
计算机视觉
[CV-0] FuseReg: Regularizing Layer Fusion Mitigates the Reconstruction-Generation Gap in Representation Autoencoders
链接: https://arxiv.org/abs/2609.31620
作者: Hongyang Du,Yunfei Xie,Junjie Ye,Jiawei Yang,Xiaoyan Cong,Haodong Zhang,Yongchao Huang,Haiyu Wu,Zongxia Li,Shihang Gui,Dawei Liu,Runhao Li,Jingcheng Ni,Chen Wei,Randall Balestriero,Yue Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Representation autoencoders (RAEs) reuse features from a pretrained visual encoder as reconstruction and diffusion latents, integrating strong visual representations into image generation. However, RAEs still need to decide which encoder layers form the shared latent space for the generator and pixel decoder. This choice involves a trade-off. Shallower layers tend to preserve fine pixel details better, while deeper layers tend to yield better generation metrics. A fixed heuristic layer fusion therefore couples two stages that benefit from different information. We introduce FuseReg, which replaces heuristic feature selection with training over random subsets of encoder layers. We theoretically analyze the underlying mechanism: subset sampling explicitly penalizes sensitivity to cross-layer disagreement. On ImageNet-256 with DINOv3-L, a single FuseReg decoder reconstructs from full, sparse, and single-layer fusions without retraining, achieving higher PSNR than decoders specialized to fixed fusions. This flexibility also benefits generation: decoder replacement alone reduces unguided gFID by 27% with an unchanged RAEv2 DiT-XL generator. The same regularization principle extends to diffusion training, with joint regularization of both stages reducing unguided gFID by 29% on DiT-Base. These results show that training downstream models for layer-fusion robustness narrows the reconstruction-generation gap without modifying the pretrained encoder.
[CV-1] GraphWrit3R: End-to-End 3D Scene Graph Writing
链接: https://arxiv.org/abs/2609.31595
作者: Luka Milivojevic,Nikola Popovic,Sayan Deb Sarkar,Sebastian Koch,Iro Armeni,Luc Van Gool,Danda Pani Paudel
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project page at this https URL
Abstract:3D scene graphs provide a structured representation of complex environments by encoding objects, their semantic attributes, and the spatial and functional relationships between them. Current approaches for 3D scene graph generation suffer from several fundamental limitations. They rely on complex multi-stage pipelines with explicit intermediate representations, making systems fragile and prone to error propagation. They assume access to ground-truth object annotations during inference, which deviates from real-world scenarios. They depend on proprietary models, hindering open-source deployment, or incur prohibitively slow inference. We present GraphWrit3R, a simple end-to-end method that takes a 3D point cloud, Gaussian Splats, or a combination of both as input, and directly outputs a complete scene graph as a structured JSON script. The graph lists all objects, their semantic attributes, and the relationships between them, while avoiding all of the above mentioned limitations. The choice of multiple input modalities is purely for versatility, allowing a single set of weights to handle diverse scenarios. Point cloud inputs are encoded via Sonata and Gaussian Splat inputs via Chorus, with both modalities projected onto a shared voxel grid and fused through a novel per-voxel contrastive alignment loss before being decoded by a large language model. As a natural consequence of the LLM, GraphWrit3R also supports open-vocabulary querying. On the 3DSSG benchmark, our method achieves state-of-the-art performance on object class, predicate, and triplet recall, outperforming methods that rely on ground-truth object annotations during inference. We further provide qualitative results and analyze different input modality configurations, contrastive loss formulations, and token fusion strategies.
[CV-2] How Far Can INRs Go? Cross-Domain Parameter-efficient INR-Based Semantic Segmentation for Brain MRI
链接: https://arxiv.org/abs/2609.31573
作者: Ziyao Shang,Pouya Sadeghi,Letian Jiang,Alexander Wong,Sirisha Rambhatla
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 26 pages, 15 figures
Abstract:Biomedical image segmentation is central to medical image analysis, but practical deployment often faces limited annotations, memory constraints, and cross-site distribution shifts. Implicit Neural Representations (INRs) have recently emerged as a lightweight alternative for semantic segmentation, achieving competitive performance with substantially fewer parameters than conventional architectures. However, the mechanisms, scaling behavior, and domain generalization abilities of INR-based segmentation remain insufficiently understood. In this work, we study these questions in the context of cross-domain brain MRI segmentation. We analyze INR-based segmentation across low-parameter regimes, comparing it with conventional pipelines in both in-domain and out-of-domain settings. Surprisingly, we find that INR-based models do not simply improve with increasing parameter budget. Their advantage is most pronounced under low-parameter and limited-augmentation settings, while U-Net-based models benefit more from larger capacity and standard augmentation. We also investigate how INRs encode semantic information in their hidden features and show that complementary segmentation-relevant structure is distributed across multiple INR layers. Building on this insight, we introduce HierINRSeg, a hierarchical INR-based architecture that aggregates multi-layer representations for improved robustness and generalization. Extensive experiments show that HierINRSeg consistently outperforms MetaSeg, a strong recent INR-based segmentation baseline, with an average improvement of 5.6 percentage points in Dice for the in-domain test set and 8.2 percentage points out-of-domain. Overall, our analysis identifies the conditions under which INR-based segmentation is most effective, providing concrete guidance for model selection and future research.
[CV-3] OC-GS: Gaussian Splatting for Irregular Turntable Capture
链接: https://arxiv.org/abs/2609.31572
作者: Jae Joong Lee,Bedrich Benes
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Uneven rotation and dropped frames make equal-angle assumptions unreliable for turntable reconstruction. We present OC-GS, an object-centric Gaussian splatting that refines each image’s angle while maintaining a shared camera, rotation axis, and pivot. This orbit-consistent refinement jointly optimizes image-derived geometry and angles to reconstruct objects from sparse, irregular captures. On rendered objects with 12, 8, and 6 irregularly spaced views, OC-GS achieves mean foreground PSNR scores of 21.26, 19.36, and 15.83dB, respectively, exceeding all four evaluated pose-free Gaussian splatting baselines in each condition. Under a shared trainer, refining image-estimated angles improves mean foreground PSNR by 7.88dB over keeping those estimates fixed. An ablation study shows that both image-derived angle initialization and the shared motion model contribute to the improvement. On real captures, OC-GS’s refinement increases mean foreground PSNR by 0.70dB. Results show that refining uncertain angles within a shared motion model improves reconstruction from sparse, irregular turntable captures.
[CV-4] Region-Level Black-Box Defense Against Stealthy Embedding-Space Backdoors in CLIP
链接: https://arxiv.org/abs/2609.31558
作者: Ahmed Abdelnaby,Mohamed Elmahallawy
类目: Computer Vision and Pattern Recognition (cs.CV); Cryptography and Security (cs.CR)
备注:
Abstract:Contrastive Language–Image Pretraining (CLIP) has emerged as a dominant vision backbone due to its strong transferability and zero-shot capabilities. However, recent studies reveal a critical vulnerability: embedding-space backdoor attacks. By poisoning only a tiny fraction of image–text pairs, adversaries can implant stealthy triggers that induce targeted shifts in CLIP’s joint embedding space. Unlike conventional backdoors that manipulate classifier logits, these attacks corrupt representations directly, making them highly effective under extremely low poisoning ratios and difficult to detect. Existing defenses require access to model parameters, gradients, logits, or clean validation data—assumptions that rarely hold in realistic black-box deployments. Moreover, current black-box methods struggle to accurately localize small or out-of-distribution triggers. We propose CLIPGuard, a lightweight and fully black-box defense specifically designed to mitigate embedding-space backdoors in CLIP encoders. CLIPGuard identifies malicious regions by measuring segment-wise embedding perturbations and selectively purifies only suspicious segments via semantic inpainting, preserving benign visual content and alignment quality. Extensive experiments on STL-10, ImageNet, and diverse trigger families—including BadCLIP, BadNets, blended, patch-based, and typographic attacks—demonstrate that CLIPGuard reduces attack success rates to as low as 1.05% while maintaining clean accuracy up to 86.34%, consistently outperforming existing black-box defenses, including CleanCLIP and CleanerCLIP. Our code is available this https URL
[CV-5] Structured Reasoning Agent ic Framework for Interpretable Critical View of Safety Assessment
链接: https://arxiv.org/abs/2609.31524
作者: Qing Xu,Yuxiang Luo,Zhen Chen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Surgical scene understanding is critical for computer-assisted intervention, yet laparoscopic cholecystectomy remains challenged by the complex anatomy of the hepatocystic triangle and the risk of bile duct injury. Existing methods for Critical View of Safety (CVS) assessment typically treat it as a holistic prediction task, mapping visual features directly to criterion-level labels. This black-box paradigm lacks explicit reasoning about anatomical relationships, limiting both interpretability and compositional generalization. To address this, we propose ReasonCVS, a structured reasoning agentic framework empowered by Vision-Language Models (VLMs) that decomposes CVS assessment into explicit, fine-grained anatomical verification. Specifically, we devise an Anatomical Scene Graph Abstraction (ASGA) that organizes anatomical entities and their spatial relationships into a structured representation. To operationalize this, we introduce a Rationale-Aware Reasoning Agent, powered by a Large Language Model (LLM) fine-tuned via rationale distillation. Functioning as a strict central decision-maker, it invokes VLM-driven Sub-criterion Verifier as a specialized perceptual tool to parse the graph and independently evaluate individual sub-criteria. Through calibrated soft reasoning, this agent synthesizes the tool-gathered distributed observations, yielding a final verdict alongside a traceable clinical rationale. Extensive experiments on the Endoscapes-CVS201 benchmark demonstrate that ReasonCVS achieves superior performance (68.1% mAP) over state-of-the-art while providing interpretable, criterion-level explanations for reliable surgical assessment.
[CV-6] Forensic Twins: Self-Supervised Residual Learning for AI-Generated Image Forensics
链接: https://arxiv.org/abs/2609.31514
作者: Javier Muñoz-Haro,Ruben Tolosana,Ruben Vera-Rodriguez,Aythami Morales,Julian Fierrez
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages + Supp. Material. 3 figures
Abstract:Detectors of AI-generated images are typically trained using samples from all Generative AI architectures they must catch, and struggle as soon as a new architecture emerges. Recent approaches have explored self-supervised pre-training as an alternative solution, yet standard frameworks work against the forensic task, e.g., their augmentations overwrite the micro-statistics of image formation. This paper introduces Forensic Twins, a Self-Supervised Residual Learning (SSRL) framework whose pretext task suppresses macroscopic content availability. Each image is mapped through a frozen, off-the-shelf forensic residual extractor, from which two spatially disjoint crops are drawn. Sharing no pixel, the two views retain minimal semantic structure to align, leaving a redundancy-reduction objective with a predominant common signal: the stationary fingerprint of the image acquisition pipeline. Additionally, Forensic Twins is trained exclusively on real images; no AI-generated image is observed at any stage. Experiments show that Forensic Twins attributes AI generator sources with 56.61% accuracy, i.e., 6.13% above the previous state-of-the-art zero-shot method at 375x lower latency. We also demonstrate that fitting a Gaussian Mixture Model (GMM) offline using only the real image embeddings extracted from Forensic Twins turns it into a state-of-the-art zero-shot detector, reaching 97.99% AUC across 27 unseen AI generators, including GANs, diffusion models and commercial systems. Code, weights and exact splits will be made publicly available
[CV-7] ClearGS: Reliability-Aware Gaussian Splatting from Handheld Videos
链接: https://arxiv.org/abs/2609.31509
作者: Xuanzhi Liu,Xinyi Wu,Hang Pan,Wensi Huang,Zhenyao Wu,Ruize Han,Song Wang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:We present ClearGS for 3D Gaussian Splatting (3DGS) from handheld videos with uneven viewpoint coverage and mixed frame quality. Rather than selecting frames with binary decisions, ClearGS uses Reliability-aware View Allocation (RVA) to assign graded raw-supervision weights based on appearance reliability, degradation risk, and geometric utility, while weakly reactivating useful suppressed frames to maintain trajectory coverage. Since weighting cannot restore details lost to blur or distortion, ClearGS further introduces Render-Guided In-Video Restoration (RIVR). The current 3DGS render provides a pose-aligned structural candidate, a frozen no-reference restoration expert restores the corresponding raw video observation without any clean reference image, and no-reference perceptual scores select among the render, restored observation, and high-frequency fused candidate. ClearGS then applies Full-Trajectory Repair Consolidation to revisit accepted repairs and preserve details introduced early. On GS2E and GSOTM, ClearGS achieves state-of-the-art overall performance, with consistent CLIP-IQA and MUSIQ gains and LPIPS reductions in most degradation settings, without paired sharp supervision or matched clean references.
[CV-8] SatNav: A Scalable Benchmark for Long-Horizon UAV Vision-Language Navigation from Satellite Imagery NEURIPS2026
链接: https://arxiv.org/abs/2609.31507
作者: Jiajun Jiang,Chunliang Hua,Zichun Chen,Yanxing Wu,Zeyuan Yang,Jie Song,Xiao Hu
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注: Accepted at NeurIPS 2026, Track on Evaluations and Datasets. 32 pages, 16 figures. Project page: this https URL
Abstract:Urban uncrewed aerial vehicle (UAV) vision-language navigation (VLN) requires agents to follow instructions across extended urban spaces, inherently demanding long-term memory and geospatial grounding. However, scaling existing benchmarks remains difficult because of their reliance on costly reconstructed 3D assets, limiting geographic diversity and episode scale. To address this, we introduce SatNav, a scalable, long-horizon UAV VLN benchmark built from high-resolution satellite imagery. SatNav targets city-level navigation missions and uses satellite crops as approximations of UAV nadir views for visual observations. Through an automated cue-to-episode pipeline, SatNav constructs 118K episodes from 59 scenes across 18 cities, with an average trajectory length of 379 m. To stress-test long-horizon memory and geospatial reasoning, SatNav defines three task families: Boundary, Landmark, and Route, targeting loop progress tracking, landmark-based spatial grounding, and route following with counting cues. Benchmarking classical VLN agents and recent agents based on large vision-language models (LVLMs) on SatNav shows that city-scale navigation remains challenging. We further introduce SwiftVLN, a modular framework with switchable memory components, and conduct systematic memory-design ablations. Finally, satellite-to-UAV transfer experiments show that satellite-trained navigation models can operate on real-flight UAV observations, showing the practical relevance of SatNav. Our project page: this https URL
[CV-9] Uncertainty-Aware Federated Learning for Infant Movement Analysis
链接: https://arxiv.org/abs/2609.31463
作者: Edmond S. L. Ho
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at IEEE The 4th International Conference on Federated Learning Technologies and Applications (FLTA26)
Abstract:Infant movement analysis provides valuable biomarkers for the early identification of neurodevelopmental disorders. Recent advances in deep learning have enabled automated analysis of infant movements from video-derived skeletal representations, achieving performance comparable to expert assessment for tasks such as General Movement Assessment (GMA). However, most existing approaches rely on centralized training, requiring data from multiple institutions to be collected and stored at a single site. Such assumptions are often impractical in clinical settings due to privacy, governance, and data-sharing constraints. To address these challenges, we present, to the best of our knowledge, the first federated learning framework for automated infant movement analysis and General Movement Assessment using skeletal motion data. As a clinically relevant use case, the proposed framework is evaluated on fidgety movement classification. To quantify model confidence, Monte Carlo (MC) Dropout is employed to estimate predictive uncertainty during inference. Building upon this, we propose an Uncertainty-Aware Federated Averaging (UA-FedAvg) strategy that incorporates predictive entropy derived from MC-Dropout into the federated aggregation process, enabling client contributions to be adjusted according to their predictive uncertainty. Experiments were conducted using a cross-subject evaluation protocol under a three-client federated learning setting. Results demonstrate that federated learning substantially improves classification performance compared with independently trained local models while achieving performance approaching that of centralized training. Furthermore, UA-FedAvg and its variant incorporating validation loss generally outperform conventional FedAvg across the evaluated data-split configurations.
[CV-10] KneePreM: Towards 3D Knee MRI Foundation Models via Large-Scale Unlabeled Pretraining and Label-Efficient Fine-Tuning
链接: https://arxiv.org/abs/2609.31461
作者: Xinxin Wang,Liam Hazan,Jing Li,Simona Rabinovici-Cohen,Xiaojuan Li,Mingrui Yang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Background: Large volumes of unlabeled knee MRI scans are available across repositories but remain insufficiently leveraged. We developed KneePreM, a knee-specific 3D self-supervised model, and evaluated transfer and label efficiency for classification and segmentation. Methods: A 3D U-Net masked autoencoder was pretrained on 19,011 unlabeled Osteoarthritis Initiative (OAI) MRI series from 4,791 participants. Downstream fine-tuning used full and reduced training sets for fastMRI+ two-label classification (1,172 examinations), Arthroscopic Partial Meniscectomy (APM) eight-target classification (1,716 examinations), SKM-TEA segmentation (155 examinations), and APM segmentation (25 examinations). Baselines were random initialization and SuPreM. Deployment workflow was implemented with a Model Context Protocol interface. Evaluation metrics included balanced accuracy, F1 score, ROC AUC, PR AUC, and Dice score. Statistical analysis used bootstrap confidence intervals and paired bootstrap tests for classification and Wilcoxon signed-rank tests for segmentation. Results: KneePreM achieved higher full-data macro ROC AUC than both baselines for fastMRI+ and APM (all p .001). For fastMRI+ classification, KneePreM achieved a ROC AUC of 0.722 using 50% of the training data, exceeding both full-data baselines. In APM classification, KneePreM reached a ROC AUC of 0.740 with 70% of the data, matching the full-data random baseline and outperforming SuPreM. For SKM-TEA segmentation, its 70%-data Dice of 0.838 exceeded the full-data random baseline (0.835) and both same-budget comparators. In APM segmentation, its 75%-data Dice of 0.746 exceeded the full-data random baseline (0.731) and both same-budget comparators. Conclusion: KneePreM improves transfer performance and label efficiency across knee MRI classification and segmentation tasks, particularly when labeled training data are limited.
[CV-11] Diagnosing the Sources of Compositional Failure in Vision-Language Models: A Controlled Analysis
链接: https://arxiv.org/abs/2609.31456
作者: Mona Gandhi,Cenk Merih Olcay,Kuan-Chieh Lo,Santiago Castro,Christopher W. Myers,Srinivasan Parthasarathy
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Vision-language models (VLMs) often struggle with compositional reasoning tasks, but the reasons for this underperformance remain unclear. A common hypothesis is that models struggle to integrate multiple components, leading to training interventions to improve compositional binding. However, this assumption has never been directly quantified. Existing benchmarks evaluate captions only in their composed form, making it impossible to separate the cost of joint reasoning from the cost of recognizing individual components under increasing load. We introduce COMPASS (COMPositional Analysis of SkillS), a controlled evaluation framework designed to isolate and measure the distinct factors underlying compositional failure. By comparing performance on composed captions with their decomposed counterparts , we directly quantify the cost of compositional integration across 87K image-caption pairs. Across multiple VLMs, this gap is real but partial, accounting for only part of the observed degradation. This motivates a finer-grained investigation into what additional factors govern model behavior. We analyze performance at the level of individual skills: object detection, attribute binding, and relation reasoning, using skill-targeted perturbations across 274K image-caption pairs. We find a consistent skill-specific pattern: each skill degrades primarily with the count of its own primitive type (self-load), while cross-load effects are predominantly positive, suggesting that primitives of different types provide useful grounding context. This pattern holds across standard contrastive encoders, explicitly trained compositional reasoning models, and non-contrastive architectures. These findings show that compositional degradation reflects multiple separable factors that cannot be reduced to joint reasoning alone.
[CV-12] Different Corruptions Different Signals: Uncertainty and Loss in Federated Data Quality
链接: https://arxiv.org/abs/2609.31454
作者: Bradley Scott,Zeqi Luo,Edmond S. L. Ho
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at IEEE The 4th International Conference on Federated Learning Technologies and Applications (FLTA26)
Abstract:Federated learning (FL) data corruption can affect either inputs or labels, but it remains unclear whether input-conditional uncertainty and prediction-label loss expose these corruption modes equally. This paper compares two corruption-detection signals in FL: input-conditional uncertainty and prediction-label loss. The uncertainty signal is characterised using a learned aleatoric variance estimate together with Monte Carlo (MC) dropout variance and entropy measures, while the loss is computed against the supplied label. We test these signals against additive image noise and persistent random label flips. On ResNet-20 with CIFAR-10 and SVHN under Dirichlet partitions with data that are not independent and identically distributed (non-IID), the two corruption types behave differently. For persistent random label flips, the within-client per-sample area under the receiver operating characteristic curve (AUC) is 0.85 on CIFAR-10 and 0.95 on SVHN for prediction-label loss, while every uncertainty estimator stays at chance (0.49–0.50). This pattern is consistent with the model remaining confident in the underlying image despite the supplied label being wrong. For image noise, expected-entropy uncertainty rises above chance (0.67 on CIFAR-10 and 0.66 on SVHN), while loss responds comparably (0.64 on both). Each signal is therefore the stronger detector for a different corruption: the prediction-label loss for persistent label flips, and expected-entropy uncertainty for image noise, with its advantage becoming apparent as federation-wide corruption prevalence increases. Robust FL data-quality assessment should match the signal to the corruption rather than rely on uncertainty alone across corruption types.
[CV-13] Vision-Based 6-DoF Grasp Pose Estimation for Robot Cloth Unfolding
链接: https://arxiv.org/abs/2609.31452
作者: Domen Tabernik,Peter Nimac,Jan Jerićević,Danijel Skočaj,Andrej Gams
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: Published in IEEE Transactions on Cybernetics
Abstract:Cloth manipulation is a challenging task due to the deformable and high-dimensional nature of cloth, which leads to complex interaction dynamics and perceptual ambiguity arising from frequent occlusions of critical visual cues such as folds, edges, and grasp points. In this work, we tackle cloth unfolding using a regrasping-in-the-air strategy, where one manipulator holds the cloth while the other grasps it at an optimally selected point to unfold it. To this end, we propose CeDiRNet-6DoF, a deep learning framework that jointly predicts effective grasp points and the complete 6-DoF grasp pose from the observed cloth configuration. By integrating dense 3D grasp regression with segmentation and sine-cosine-encoded Euler angles, the proposed method reliably estimates the grasp configuration that maximizes the unfolded cloth area. We extensively evaluated CeDiRNet-6DoF on a bimanual robotic setup within the ICRA 2024 Cloth Competition framework, achieving state-of-the-art performance. An ablation study further validates the benefits of key design components, including joint segmentation, background randomization, and image cropping. These results establish CeDiRNet-6DoF as a robust and versatile foundation for reliable robotic cloth manipulation in unstructured environments.
[CV-14] mplateCraft: Agent ic Visual Template Generation ICASSP2027
链接: https://arxiv.org/abs/2609.31451
作者: Hongjie Yu,Zhiyuan Fan,Yuzhe Zhang,Jiangcun Du,Zhicheng Gao,Yuhong Zhang,Xiaokai Zhan,Zongshi Xie
类目: Multimedia (cs.MM); Computer Vision and Pattern Recognition (cs.CV)
备注: 5 pages, 3 figures, 1 table. Submitted to ICASSP 2027
Abstract:The growing popularity of short videos has driven demand for one-click content creation. Visual templates turn uploaded images into personalized content with preset effects, but reusable template generation still requires substantial manual effort in asset preparation and tool orchestration. We propose TemplateCraft, a multi-agent system that converts natural-language instructions into client-executable templates through planning, material generation, effect-workflow generation, and protocol compilation. Its Planner-Evaluator loop uses execution feedback for targeted rollback, while stage-level and long-term memory support revision without parameter updates. We evaluate TemplateCraft on TemplateBench, derived from 60 real-world templates. With the same Qwen3-VL backbone, TemplateCraft raises image/video generation success rates from 56.7%/30.0% to 66.7%/50.0% over Planner-only (best-of-three) and improves template adherence and style consistency. With additional evaluation and revision, it matches or exceeds a GPT-4o Planner-only baseline on selected metrics. Persistent assets further improve cross-input style consistency.
[CV-15] From Reward Signal to Visual Utility: A Controlled Audit of Medical VLM Post-Training
链接: https://arxiv.org/abs/2609.31450
作者: Wang Jingxin
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Medical vision-language model (VLM) post-training is commonly evaluated through answer accuracy. We examine how changes in accuracy and training objectives relate to image-conditioned decisions in a controlled Qwen2.5-VL-3B study on PMC-VQA. We compare supervised fine-tuning (SFT) with low-rank adaptation (LoRA) restricted to the language model, expanded multimodal adaptation scopes, standard answer-only Group Relative Policy Optimization (GRPO), and a counterfactual evidence objective. On 2,000 clean-test questions, language model LoRA SFT changes correct-image accuracy by +1.10 percentage points (95% paired bootstrap CI:-0.85 to +3.05), while visual-benefit events decrease by 2.40 points and image sensitivity decreases by 5.60 points. Paired records reveal 155 acquired and 203 lost visual-benefit events. Broader adaptation yields lower correct-image accuracy than language-model LoRA SFT. Standard GRPO produces mixed-reward groups and parameter updates, with an uncertain clean test accuracy change. A generation audit reveals that canonical option scores can follow a different token path from generated answers. With scores taken along the greedy generation path, the evidence target improves on the training set; its gains over standard GRPO remain inconsistent on validation data at matched training doses. Sample-level analyses trace how evidence scores, decision margins, and generated answers change during post-training. This empirical and measurement audit identifies gaps between optimization activity, target acquisition, and useful held-out visual behavior.
[CV-16] Implicit Neural Representation for Hyperspectral Video Compression
链接: https://arxiv.org/abs/2609.31435
作者: Alfredo Scalera,Paul Murray,Jaime Zabalza
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted at IEEE WHISPERS 2026
Abstract:With the advent of snapshot cameras, hyperspectral video is becoming more readily available. In recent years, new applications have emerged which have led to increasingly larger datasets. However, hyperspectral video compression remains in the early stages. In this study, we explore the use of implicit neural representation as a candidate solution. We propose a novel extension of an existing RGB video compression model, achieving Bjøntegaard Delta PSNR gains of +4.99 dB and Bjøntegaard Delta rate of -88.88% compared to traditional hyperspectral image compression methods applied frame-by-frame. In addition to reconstruction quality, the effects on downstream task performance are measured in the form of object tracking success. Compared to video compressed with methods based on principal component analysis and JPEG2000 in low data regimes, our proposed method improves tracking area under the curve by up to 23.42% and distance precision by up to 35.56% on examples from the HOT2026 dataset.
[CV-17] AxonSynth: Domain-Randomized Synthetic Data for Zero-Shot 3D Axon Segmentation in Light-Sheet Microscopy MICCAI2026
链接: https://arxiv.org/abs/2609.31431
作者: Edward Gaibor,Kyriaki-Margarita Bintsi,Carmen Luz Leiva Ureta,Zayneb Bellatif,Chiara Maffei,Wenze Li,Elizabeth Hillman,Yaël Balbastre,Anastasia Yendiki
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 11 pages, 2 figures, 2 tables. Accepted at SASHIMI 2026, held with MICCAI 2026
Abstract:Accurate segmentation of axons in 3D microscopy data is important for analyzing white-matter organization, but dense ground truth labels are expensive to obtain. Existing supervised axon segmentation methods rely on target-domain annotations and can be brittle when tissue type, species, modality, or acquisition conditions change. We present AxonSynth, a domain-randomized synthetic-data framework for training 3D axon segmentation models without manually annotated real training volumes. AxonSynth generates dense synthetic axon labels with orientation priors that reflect realistic fiber configurations and renders them with randomized density, contrast, bias fields, blur, and noise. A three-class 3D U-Net is trained to predict background, axon sheath and intra-axonal space. We evaluate zero-shot transfer on 10 held-out light-sheet microscopy (LSM) patches from macaque and human brain samples labeled with one of three axonal markers, comparing against calibrated thresholding and Frangi filtering using overlap, corrected detection, false-positive, and topology metrics. On macaque samples, AxonSynth achieved the best corrected Dice and corrected precision (0.826 and 0.851), compared with 0.765 and 0.754 for thresholding and 0.685 and 0.762 for Frangi. On human samples, corrected Dice was comparable to thresholding (0.857 vs. 0.868), while component-count error decreased from 22,504 to 3,377. Across all held-out patches, AxonSynth reduced component-count error in 10/10 patches and Euler-characteristic error in 8/10. These results show that synthetic-label domain randomization can reduce dependence on manual axon annotation while supporting synthetic-to-real 3D segmentation.
[CV-18] InternW0-Δ: A World Action Model Bridging Predictive Dynamics and Actions with 20K Hours of Open Data
链接: https://arxiv.org/abs/2609.31394
作者: Xingyu Miao,Zizun Li,Baole Fang,Kaiwen Song,Tenghui Wang,Hanxue Zhang,Yating Wang,Xudong Li,Yuping He,Xueyuan Wei,Chao Gao,Xijie Yang,Yingxiang Xu,Kerui Ren,Wenqi Guo,Jianjun Zhou,Xinzhe Wang,Weiguang Zhao,Ni Yang,Zetao Cai,Yufei Xue,Hengjie Li,Zeyu He,Yuanzhen Zhou,Rong Fu,Jianyang Zhang,Siwei Cui,Fuxian Huang,Yunsong Zhou,Xing Gao,Yifei Yao,Qiaojun Yu,Kailin Li,Ming Zhou,Mu Huang,Xinyue Li,Wenze Cui,Bingqi Jiang,Xueyue Zhu,Junting Dong,Haoyu Guo,Tao Lu,Mulin Yu,Bowen Zhou,Bin Zhao,Tianfan Xue,Weinan Zhang,Chunhua Shen
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:World Action Models (WAMs) jointly model visual dynamics and action generation for generalist robot manipulation. A central challenge is to integrate priors from large-scale pretrained models—including visual dynamics, scene semantics, geometry, and motion—into a unified framework for robot action generation. We introduce InternW0- \Delta , a unified WAM pretrained on a heterogeneous corpus that outperforms prior methods across simulation benchmarks and real-robot platforms. InternW0- \Delta combines pretrained visual dynamics, scene-level semantics, 4D geometric and motion priors, and action generation within a Mixture-of-Transformers (MoT) framework. A pretrained video expert and an action expert interact under semantic guidance from a frozen VLM, while a pretrained 4D foundation model injects geometric and motion priors through training-only distillation. We further introduce Causal Imprint, which learns future-relevant scene changes from training-only future supervision and provides predictive representations directly to the action expert without future-video rollout at inference. For large-scale joint training, we construct a heterogeneous corpus of robot demonstrations, UMI data, egocentric human demonstrations, and Ego2Robot data, curated and aligned under a common state-action representation. The resulting corpus contains over 20K hours of processed training data, to our knowledge the largest open-source corpus of its kind. We pretrain InternW0- \Delta on this corpus and demonstrate strong performance across simulation benchmarks and real-robot platforms. We will open source the training code, model weights, infrastructure, data-processing pipeline, and processed data where licenses permit. Project page: this https URL Subjects: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2609.31394 [cs.RO] (or arXiv:2609.31394v1 [cs.RO] for this version) https://doi.org/10.48550/arXiv.2609.31394 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-19] Guiding End-to-End Driving Models with Endpoint-Constrained Trajectory Optimization
链接: https://arxiv.org/abs/2609.31383
作者: Brayden Zhang,Mahsa Golchoubian,Igor Gilitschenski,Boris Ivanovic,Kashyap Chitta
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:End-to-end driving policies are commonly trained through open-loop behavior cloning, yet they must ultimately operate in closed-loop when deployed on a vehicle, creating a fundamental mismatch between training and execution. Beyond the commonly studied effects of covariate shift and causal confusion, we identify a complementary factor for this open-loop/closed-loop gap: waypoint-based supervision and displacement metrics do not ensure that the intermediate trajectory is physically coherent or easy for the controller to track. We observe that these inconsistencies concentrate primarily at intermediate waypoints, while the predicted endpoint remains comparatively reliable. Based on this observation, we introduce Endpoint-Constrained Optimization (ECO), a lightweight postprocessing layer that anchors the trajectory to the vehicle’s executed history, preserves the policy’s predicted endpoint, and reshapes the intermediate waypoints to improve feasibility. ECO requires no map, privileged simulator state, or additional training, and can be inserted between a broad range of waypoint-emitting policies and their controllers. Across two closed-loop simulators, it improves the aggregate closed-loop score of all six evaluated generative and regression-based driving policies, and the gains tend to increase with how often the base plans violate motion limits. On HUGSIM, ECO improves VaVAM from 18.1 to 31.0 HD-Score (+71%), achieving 1st place on the HUGSIM Closed-Loop Driving Challenge. Similarly, on AlpaSim, ECO increases the scene scores of VaVAM and DiffusionDrive by 123% and 22%, respectively. These results show that for a broad collection of end-to-end driving models, repairing the intermediate geometry of predicted trajectories without changing the policy’s predicted endpoint can substantially improve closed-loop performance.
[CV-20] ContraFM-S2O: Flow Matching-Based One-step SAR-to-Optical Image Translation Model with Contrastive Learning
链接: https://arxiv.org/abs/2609.31378
作者: Mingqian Yu,Wei-kuan Chiang,Qiurui Wang,Peilin Zhao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:In recent years, diffusion models and GAN-based models have become the mainstream approaches for SAR-to-optical image translation, owing to their advantages, such as high-quality generation and stable training. However, they have shortcomings such as high inference latency and the generated optical images suffer from low detail fidelity, often resulting in blurred edges and loss of fine textures. Thus, we propose ContraFM-S2O, which is a flow matching-based model for SAR-to-optical image translation. Unlike conventional diffusion models, ContraFM-S2O learns to predict the velocity field in training and solves ODE instead of SDE during inference to improve the sampling efficiency. In addition, ContraFM-S2O replaces instantaneous velocity with average velocity along the interpolation path to realize one-step SAR-to-optical image translation and uses contrastive learning to improve the quality of the generated optical images. Experiments show our model achieves state-of-the-art on SAR2Opt and QXS datasets, outperforming baselines, and reduces inference latency via one-step generation.
[CV-21] RECAST: From Log Replay to Closed-Loop Driving Simulation with View-Complete Actors
链接: https://arxiv.org/abs/2609.31374
作者: Zijun Zhao,Liewen Liao,Kang Shen,Songan Zhang,Ming Yang
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 8 pages, 5 figures
Abstract:Closed-loop driving simulation requires rendered observations to remain reliable as the ego vehicle and surrounding actors move beyond their recorded trajectories, exposing views absent from the source log. Existing data-driven simulators reconstruct dynamic actors from sparse observations, which can result in rendering artifacts under these viewpoint changes. We introduce RECAST (REconstructing Controllable Actors for Simulation and Testing), a 3D Gaussian Splatting framework that generates a view-complete actor from a single segmented vehicle observation in a driving log and registers the generated actor in the reconstructed scene. RECAST supports planner-in-the-loop rendering under controlled ego-actor interactions. To adapt an image-to-3D prior to real vehicles, we further introduce RECAR, a dataset of approximately 20K real vehicles with 600K background-free RGBA images spanning diverse vehicle colors and types. We use two-stage adaptation to improve vehicle generation from real driving-log observations. At the actor level, RECAST reduces \mathrmFD_\mathrmincep from 9.788 to 7.992 relative to unadapted TRELLIS. At the scene level, under actor motion beyond logged trajectories, RECAST reduces \mathrmFD_\mathrmincep from 129.35 to 112.10 and increases \mathrmCLIP_\mathrmmargin ( \times1000 ) from 0.14 to 3.47 relative to Street Gaussians. We demonstrate planner-in-the-loop simulation with the image-conditioned planner GTRS-Dense. Compared with native Street Gaussians actors, RECAST increases the no-collision (NC) rate from 22.2% (12/54) to 63.0% (34/54) and the mean minimum predicted time-to-collision (TTC) from 0.798 s to 2.150 s. These experiments show that RECAST supports closed-loop planner evaluation under controlled ego-actor interactions beyond log replay. Visit our project page at this https URL
[CV-22] OpenVAM: Open-World Visual Attention Modeling with VLMs
链接: https://arxiv.org/abs/2609.31364
作者: Kiana Hooshanfar,Amirhossein Kazerouni,Alireza Hosseini,Michael Brudno,Babak Taati
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Predicting human gaze is a core capability for applications ranging from web/UI design analysis to robotics and human-computer interaction. Yet, most visual attention modeling methods output only a dense saliency map, which is often insufficient for action: practitioners need to connect attention peaks to discrete elements in the scene (what) and understand the drivers of those peaks in context (why), while remaining robust to domain shift across natural images, commercial content, and UI/web layouts. We, therefore, introduce OpenVAM (Open-world Visual Attention Modeling with VLMs), a unified framework that jointly addresses universality and explainability across heterogeneous domains (natural scenes, commercial imagery, and UI/web layouts) and supervision modalities. OpenVAM adopts a decoupled-but-aligned design: a dedicated dense visual pathway provides stable, spatially precise localization, while an instruction-following vision–language semantic head generates grounded what/why explanations conditioned on the same image and data-type prompts. A three-stage training strategy preserves strong localization priors while progressively introducing language grounding and improving explanation alignment via parameter-efficient adaptation without perturbing the saliency branch. We further propose a scalable pipeline to generate multi-domain saliency-reason annotations for training and systematic evaluation. Experiments across diverse datasets show that OpenVAM improves robustness under domain shift while producing image-grounded explanations that make saliency predictions more interpretable.
[CV-23] Open Vocabulary Domain Unlearning NEURIPS2026
链接: https://arxiv.org/abs/2609.31356
作者: Sumanth Udupa,Mehrtash Harandi,Yadan Luo,Mahsa Baktashmotlagh
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Accepted in NeurIPS 2026
Abstract:Vision-Language Models (VLMs) exhibit remarkable zero-shot generalization, yet they often encode unwanted or hazardous stylistic domains such as idealized textbook diagrams in medical AI or cartoon vehicles in autonomous driving. Approximate Domain Unlearning (ADU) aims to selectively erase a model’s recognition of a target visual domain while preserving accuracy on the remaining domains. However, existing ADU methods operate under a flawed closed-vocabulary assumption: they evaluate unlearning solely on the specific object classes seen during the unlearning fine-tuning phase. Consequently, these methods do not unlearn the domain itself; they merely overfit to seen class-domain pairs, leaving the domain easily recognizable for unseen classes and providing a false sense of removal. We argue that true domain erasure must be class-agnostic. To address this, we formalize Open-Vocabulary Domain Unlearning (OVDU), a rigorous protocol that mandates domain forgetting must transfer to held-out classes. To solve the OVDU challenge, we propose a surgical parameter-editing framework. First, a Fisher Information mask isolates domain-sensitive weights, mathematically protecting foundational zero-shot generalization. Second, our Targeted Manifold Scattering (TMS) objective uses preference-based mining to locally scatter the forget domain’s stylistic geometry. Evaluated across PACS, OfficeHome, and DomainNet, our method vastly improves open-vocabulary generalization over existing baselines. Crucially, it delivers exceptional sample efficiency, outperforming peak 8-shot baseline results with only 4 shots.
[CV-24] DyMD: Preserving Interaction Dynamics through Distribution Matching Distillation in Few-Step Video World Models
链接: https://arxiv.org/abs/2609.31349
作者: Haojun Xu,Jie Huang,Xin Lu,Mingchen Zhong,Zihao Fan,Linjiang Huang,Si Liu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Large video diffusion models offer expressive priors for embodied prediction and learning, yet their many-step sampling remains costly for interactive downstream use. Distribution Matching Distillation (DMD) enables few-step video generation, but can suppress robot–object motion while preserving visual quality. Examining DMD’s teacher and fake-score signals, we find that weak re-noising keeps the teacher posterior concentrated near motion-deficient rollouts, limiting motion-restoring guidance. Meanwhile, stronger-motion rollouts tend to incur larger fake-score fitting errors, which can hinder the generator’s learning of interaction dynamics. We propose DyMD, a DMD framework that adapts both teacher supervision and critic fitting to the evolving student. Temporal affinity–conditioned re-noise sampling adapts the timestep distribution to each rollout’s current interaction fidelity by mixing the base schedule with a teacher prior motivated by local posterior variation, thereby balancing motion recovery and appearance refinement. To better track stronger-motion rollouts, dynamics-guided fake-score tracking uses a noise-conditioned predictor to estimate noise-relative fitting difficulty from latent temporal dynamics, then upweights predicted-hard rollouts in the critic loss. Using DyMD, we distill a 14B teacher into a four-step 1.3B student with no auxiliary modules at inference. On embodied-video benchmarks, the student improves R-Bench task adherence by 9.6 percentage points and PAI-Bench-G Domain score by 5.1 points over Base DMD while maintaining comparable visual quality. As a backbone for downstream action planning, our student achieves 34% mean success across two WorldArena tasks, compared with 16% for Base DMD.
[CV-25] ChronoFuseGS: Multi-Temporal Gaussian Fusion with Per-Splat Persistence and Change Visualization
链接: https://arxiv.org/abs/2609.31339
作者: Tobias Batik,Diana Marin,Peter Kán,Hannes Kaufmann
类目: Graphics (cs.GR); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to Pacific Graphics 2026
Abstract:Reconstructing environments where parts of the scene change between captured image sets poses a challenge for 3D scene reconstruction. We present ChronoFuseGS, a multi-temporal Gaussian Splatting approach that addresses this issue by taking multiple separately trained Gaussian Splatting models, each representing a distinct timestep and partially overlapping in geographic coverage, and merging them into a single combined model. By allowing Gaussians from one timestep to contribute to the reconstruction at other timesteps, our approach leverages data across all captured timesteps to refine persistent parts of the scene. The model supports incremental extension, allowing new timesteps to be added while preserving the existing merged reconstruction. It encodes, for each Gaussian primitive, at which timesteps it contributes to the reconstruction. To support visual exploration of the reconstructed scene, we present a change-aware visualization approach that highlights the parts of the scene that have changed across a user-defined time selection, while preserving the color of persistent parts. Since the persistence encoding operates at the Gaussian primitive level, changes are visualized at sub-object granularity rather than being limited to object-level changes. We evaluate our approach on a real-world outdoor dataset of a flood management area, captured over 7 months across eight recording days and covering seasonal vegetation changes, snow cover, and flooding events, which we make publicly available. Our results demonstrate that the combined model consistently outperforms individually trained single-timestep models in novel-view synthesis quality, recovers structural details absent in the individual reconstructions, and reliably highlights changes in fine details and sub-parts of objects and natural structures.
[CV-26] CG-HAF: An Interpretable Global-Local Lesion-Burden Fusion Framework for Ordinal Acne Severity Grading in Agent ic Skincare Support
链接: https://arxiv.org/abs/2609.31326
作者: Muhammad Muhtasim Shahriar,Md. Naimur Asif Borno,Saad Aloteibi,Mohammad Ali Moni
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Manuscript under review at Expert Systems with Applications
Abstract:Ordinal acne severity grading requires distinguishing visually similar neighboring grades while jointly weighing holistic facial appearance and localized lesion burden - evidence that most existing approaches collapse into a single opaque representation. We introduce CG-HAF, a global-local fusion framework that instead keeps this evidence explicit: averaged holistic severity probabilities from independently trained classifiers are combined with structured lesion-burden descriptors from an object detector (lesion count, detection confidence, lesion area) into a compact representation, from which a lightweight, interpretable classifier produces the final grade. On a widely used benchmark, this fusion yields a clear, statistically supported improvement over global-evidence-only baselines, with the largest gains on the most severe cases. Testing on an independent dataset with a different grading standard shows that strong within-dataset performance does not transfer automatically, and a follow-up diagnostic attributes much of this gap to mismatched grading criteria rather than detection failure alone. These findings support interpretable global-local fusion as an effective strategy for ordinal acne grading while highlighting criterion alignment as key to cross-dataset portability, with a further illustration of how the resulting severity signal can support transparent, non-diagnostic decision-making in skincare applications.
[CV-27] CytoSPM: Open-Vocabulary Cytopathology Detection with Structured Prompt Bank
链接: https://arxiv.org/abs/2609.31314
作者: Wenjie Li,Zishan Xu,Jinyang Huang,Zhengxin Nie,Shichao Kan,Yixiong Liang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Cytopathology detection requires open-vocabulary recognition because cellular categories are fine-grained, long-tailed, and continuously evolving across different organ systems. However, existing cytology detectors are mostly single-domain and closed-set, and there is still no unified benchmark for evaluating open-vocabulary cytopathology detection. We present PentaCyto, a multi-domain benchmark covering cervical, urinary, respiratory, serous fluid, and thyroid cytology, with 24 base categories and 9 held-out novel categories. Each category is associated with structured cytomorphology prompts that describe diagnostic morphological attributes and provide clinically grounded textual knowledge. We further propose CytoSPM, an efficient detector based on a decoupled two-stage design. It first extracts reusable class-agnostic visual representations, and then performs class-aware structural prompt matching with class names and cytomorphology prompts. On PentaCyto, CytoSPM outperforms existing methods in novel-category detection and open-vocabulary detection while maintaining efficient inference.
[CV-28] UniAR: A Unified Framework for Autism Recognition Enhanced by Multi-View Prompt Learning
链接: https://arxiv.org/abs/2609.31298
作者: Lei Xin,Zeheng Wang,Jiayin Zhu,Shihong Huang,Fanhu Zeng,Changjiang Jiang,Dengbo He,Yutao Yue,Zhenglun Kong
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted by ACM’MM 2026
Abstract:Autism Spectrum Disorder (ASD) is a complex neurodevelopmental disorder for which early and accurate diagnosis is critical to improving long-term developmental outcomes. However, existing ASD recognition methods are often constrained by the scarcity of diagnostic text data, forcing them to rely mainly on visual analysis and limiting their ability to model clinically meaningful semantic reasoning. To address this challenge, we propose UniAR, a unified framework enhanced by multi-granularity prompt learning for robust ASD recognition under heterogeneous data variations. Specifically, UniAR leverages a large multimodal model to generate hierarchical diagnostic descriptions at the word, phrase, and sentence levels, compensating for the lack of paired clinical reports. To align the generated semantics with visual evidence, we further design a Mixture-of-Experts-based Multi-Scale Alignment Module, which dynamically matches vector-quantized visual prototypes with semantic representations at corresponding granularities. Extensive experiments on four benchmarks covering brain MRI and facial expression scenarios show that UniAR consistently outperforms existing state-of-the-art methods, achieving average accuracies of 75.9% on MRI benchmarks and 91.6% on facial benchmarks, while improving average Accuracy on MRI benchmarks by 1.5 percentage points and average Accuracy on facial benchmarks by 1.2 percentage points over baselines. These results demonstrate that UniAR offers a robust and interpretable framework for ASD screening under semantic scarcity.
[CV-29] MoTop: Motion-Topological Model For Micro AU Detection
链接: https://arxiv.org/abs/2609.31285
作者: Huai-Qian Khor,Mengting Wei,Yante Li,Chu Kiong Loo,Guoying Zhao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Facial micro-expressions are spontaneous, brief, and subtle facial movements that reveal suppressed emotions in high-stakes environments. In contrast to classic expression analysis, detecting action unit (AU) yields a finer representation of facial movements, serving as a preliminary step before defining expression classes and other downstream tasks. Therefore, it represents a crucial upstream task in facial analysis, and improving an AU detection module increases the precision of facial analysis. Despite that, detecting AU is challenging because of the constrictive nature of the AU activation regions, leading to confusion among different AUs known as AU ambiguity. To model the fine-scale changes, we propose \textbfMoTop, a motion-topological model that is augmented with a learnable motion context, yielding regional soft guidance for facial activity, followed by facial landmarks that capture the fine-scale topological changes of micro AUs. To increase the micro facial landmark representations, we amplify the encoded facial landmark transitions via linear extrapolation, thereby increasing the spatial proximity of landmarks and enhancing the low-intensity landmark dynamics. In addition, we design anatomical facial clusters that enhance the hierarchical representation, facilitating multi-scale modelling of facial geometry and improving micro-topological representations. With these contributions, we have achieved state-of-the-art performance on the CD6ME protocol for the micro AU detection task.
[CV-30] Gauss What You Need: Compact Gaussian Splatting Across Scene Scales
链接: https://arxiv.org/abs/2609.31248
作者: Afif Boudaoud,Jiayi Liu,Alexandru Calotoiu,Torsten Hoefler
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:3D Gaussian Splatting reconstructs a scene as a collection of Gaussian primitives from a set of posed photographs called the capture. The number of primitives used to represent the scene affects reconstruction quality, storage, and rendering cost. How to select this number automatically across capture scales remains unresolved: configurations effective on standard benchmarks can leave larger captures with too few Gaussians to reconstruct fine details. We observe that the surface to represent, given by the capture’s extent and resolution, is known before training, whereas its content complexity becomes apparent during training, through the reconstruction quality on the training views. We introduce TangoGS, which combines capture-derived model sizing with training-based adaptation: the capture determines the scale of the model, and training feedback determines its final size within that scale. Before training, TangoGS derives a learning allowance for model growth from the capture’s total pixels after discounting views that re-observe the same scene points. During training, reconstruction quality guides how many Gaussians to add and remove. On 13 standard benchmark scenes, TangoGS matches the mean PSNR of the best-performing evaluated baseline, LeGS, with 48% fewer Gaussians. On eight large captures, the same configuration automatically scales to larger models when necessary, achieving the highest mean PSNR among evaluated methods: 0.54 dB above the runner-up with 2.3\times as many Gaussians. Together, capture-derived learning allowances and training-quality guided density control enable a state-of-the-art quality–size compromise across scene scales without retuning.
[CV-31] Geometric Inconsistency Localization in Multi-View Image Sets
链接: https://arxiv.org/abs/2609.31247
作者: Xander Staelens,Albéric Loos,Bert Ramlot,Hannes Mareen,Peter Lambert,Glenn Van Wallendael
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Multimedia (cs.MM)
备注: 8 pages, accepted at the Deepfake Forensics Workshop (DFF 2026) at ACM Multimedia 2026
Abstract:Novel view synthesis (NVS) models can produce realistic new views of the same scene from different viewpoints. However, these generated views are not always geometrically consistent with one another. Multi-view (MV) consistency has shown promise as a tool for evaluating these NVS models. Its potential for multimedia forensics, however, remains largely unexplored, particularly for localizing geometric inconsistencies across wide-baseline image pairs. To enable research in this direction, we introduce DeformView, a wide-baseline MV dataset with pixel-level annotations of geometric inconsistencies. Using DeformView, we evaluate state-of-the-art MV consistency-scoring methods and show that approaches developed for NVS evaluation transfer poorly to the forensic task of geometric inconsistency localization. To address this limitation, we propose DEFECt3R, a lightweight learning-based classifier that uses cross-view feature relationships to localize geometric inconsistencies at the pixel level. By learning from explicit supervision, including hard negatives from geometrically consistent yet deformed views, DEFECt3R improves localization performance and substantially reduces false positives compared to existing consistency-scoring methods. Ablation experiments further show that both feature representations and correspondence quality contribute to localization performance. Overall, our findings demonstrate that MV geometric consistency is a promising yet underexplored signal for multimedia forensics and establish a benchmark and baseline for geometric inconsistency localization in wide-baseline MV image pairs. Code and dataset are available at this https URL
[CV-32] WeaveAgent : A Two-Stage Tool-Routing Agent for Ultra-High-Resolution Remote Sensing Imagery
链接: https://arxiv.org/abs/2609.31234
作者: Zhongyu Pang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Problem. Ultra-high-resolution (UHR) remote sensing with vague user intents has two bottlenecks: visual tokens are expensive, and tool calling must be format-reliable (pretrained models emit zero tool calls zero-shot). Method. WeaveAgent, a two-stage tool-routing agent, decouples routing from visual perception. Stage A is routing-first: emission is trained, not elicited. Stage B executes conditionally: intrinsic queries enter visual answering (full-scene thumbnail; a WeaveEarth-style evidence board as an optional fixed-budget, approx. 5k-token compression interface); extrinsic queries execute tool call on original full-resolution imagery, answering from tool observations in a second, observation-masked round. Training: alignment SFT, then GRPO under reward R_WA2. Results. Alignment SFT lifts extrinsic routing from 0% to 80.75% (323/400); GRPO suppresses 9 intrinsic mis-emissions while tool selection is unchanged. The trained 2B system does not beat the zero-shot 8B baseline overall (0.263 vs. 0.250), a diagnostic contribution. Oracle attribution separates two repair ingredients: loading the observation into context lifts extrinsic answer accuracy from 0.025 to 0.425 under marker-free cross-mode returns, and the two-turn SFT stage adds a further +9.3 points to 0.518 at a small routing cost. A +/- image ablation shows emission suppression is visually grounded, and a query-register matrix shows LLM-rewritten queries cost trained checkpoints 2-11 points. Scope. All training and evaluation use the 5,000 / 3,273 / 1,000-record VagueUHR corpus (600 intrinsic + 400 tool-requiring; the base seeds synthesis and is not used for optimization). Single-pass evidence construction runs at 7.31 s per image on an RTX 4090. Code, data, and evaluation protocols will be released.
[CV-33] Enabling a Unified Cross-Domain Representation for Two-Finger Gripper Manipulation via Interaction-Centric Modeling
链接: https://arxiv.org/abs/2609.31207
作者: Guanlin Li,Shifeng Bao,Yihan Zhao,Haitao Shen,Haoyang Li,Chen Zhao,Tong Yang,Jie Tang,Jing Zhang
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Achieving robust cross-embodiment generalization in imitation learning demands overcoming a critical representation flaw that inextricably entangles task semantics with hardware-specific visual geometry. We propose an interaction-centric framework that leverages the shared structure of two-finger grippers via a parameterized universal gripper abstraction, yielding a canonical gripper-frame representation. Given language and RGB-D observations, a VLM infers the subtask and grounds an interaction triplet (gripper, held, target), while SAM~2.1 tracks masks to reduce VLM queries. We design concise hybrid features that combine target/collision artificial potential fields for global guidance with segmented gripper-frame point clouds for local geometry, and use a Flow-Matching Transformer to predict smooth 7-DoF action chunks. Experiments in simulation and real-world tasks demonstrate that ours is the first imitation learning approach to simultaneously achieve competitive benchmark scores and extreme cross-embodiment/cross-viewpoint zero-shot sim-to-real transfer to completely distinct, heterogeneous robot platforms.
[CV-34] FlatClip: A Geometry-Aware Surface-Level Baseline for fMRI Representation Learning NEURIPS2026
链接: https://arxiv.org/abs/2609.31204
作者: Mo Wang,Wenhao Ye,Zihan Ning,Jiayu Zuo,Junfeng Xia,Hongkai Wen,Quanying Liu
类目: Computational Engineering, Finance, and Science (cs.CE); Computer Vision and Pattern Recognition (cs.CV); Neurons and Cognition (q-bio.NC)
备注: NeurIPS 2026
Abstract:Recent fMRI foundation models differ substantially in the spatial scale at which they represent brain activity. ROI- and connectivity-based models are efficient but coarse, whereas voxel-level models preserve fine-grained spatial structure but require specialized 3D/4D architectures and costly fMRI-specific pretraining. We ask how effectively an image-pretrained encoder can reuse the spatial organization of cortical activity. Motivated by evidence that macroscale brain activity is strongly constrained by brain geometry, we introduce FlatClip, a frozen-encoder surface-level baseline that renders cortical activity as geometry-aware flatmap sequences and reuses a frozen SigLIP2 image encoder with only a lightweight downstream probe. Across resting-state benchmarks, FlatClip serves as a competitive middle-ground representation, outperforming ROI-level baselines on HCP and ADNI tasks while remaining weaker on PPMI and below the strongest voxel-level models overall. On visual-fMRI decoding, restricting the input to visual or NSD-provided task-active cortex improves performance, highlighting the value of task-relevant cortical coverage. Spatial perturbation controls reduce the predictive performance of flatmap features under both retrained and fixed readouts, and anatomy-linked arrangements consistently outperform vertex permutations across three colormaps. Together, these results position surface-level flatmap sequences as a practical middle-ground baseline between ROI and voxel models, and support the utility of anatomy-linked spatial organization for reusing image-pretrained features. Code is available at this https URL.
[CV-35] Preserve-and-Compose Training for Composed Image Retrieval
链接: https://arxiv.org/abs/2609.31202
作者: Sehyun Kwon
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Composed image retrieval (CIR) aims to retrieve images that satisfy a user-specified modification while preserving relevant visual content from a reference image. Collecting target images for this purpose is costly, motivating zero-shot CIR methods that instead use target captions as supervision. However, target captions may omit source details that should be preserved. We therefore propose, Preserve-and-Compose Training, which complements target-caption supervision with visual evidence from the source image. PACT learns from image–text–text (ITT) triplets without target images or gallery updates, aligning composed queries with target captions while preserving source evidence through visual supervision. We further introduce Chord scoring, which combines target similarity with source-relative directional agreement in the frozen image space. Results across four ZS-CIR benchmarks show that combining target-caption supervision with source-image evidence leads to strong retrieval performance across datasets, backbone scales, and external galleries. The code is available on this https URL.
[CV-36] Light Field Primitive for Novel View Synthesis
链接: https://arxiv.org/abs/2609.31198
作者: Liang Chen,Jiahui Ning,Xun Jiang,Xing Xu,Jimmy Ren,Fenglei Fan,Heng Tao Shen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:We present Light Field Primitives (LFP), a formulation for novel view synthesis that replaces the dense ray database with a compact set of differentiable primitives in the classical two-plane parameterization. Each primitive condenses a group of rays into one learned record, and its response to a query is governed by how closely that query belongs to the group. Rendering a camera ray then reduces to compositing all responses it elicits, and a scene can be optimized directly from posed images and rendered in real time with rays. Beyond its competitive performance on standard benchmarks, the main advantage of LFP is structural: its primitives reside directly in the 4D ray space, so optical and appearance effects that are already operations on the light field become behaviors of a single shared renderer. With minimal changes to that renderer, LFP supports multi-scale anti-aliasing, defocus deblurring with refocusing, rendering for fisheye cameras, and even transparent object reconstruction with ray refraction, matching specialized frameworks that devote substantial machinery to these effects.
[CV-37] Who Says What: Symbolic Trimodal Binding Mechanisms in Audio-Visual LLM s NEURIPS2026
链接: https://arxiv.org/abs/2609.31193
作者: Jihoo Jung,Youngjoon Jang,Joon Son Chung
类目: Computer Vision and Pattern Recognition (cs.CV); Sound (cs.SD); Audio and Speech Processing (eess.AS)
备注: Accepted by NeurIPS 2026
Abstract:Current Audio-Visual LLMs (AVLLMs) struggle with reasoning over videos featuring multi-speaker dialogues. In such videos, resolving “who says what” is crucial, which necessitates trimodal (text-audio-visual) binding. Motivated by these challenges, we systematically investigate how this trimodal binding is achieved in AVLLMs. Specifically, we identify emergent symbolic trimodal binding mechanisms in AVLLMs that utilize modality-specific symbolic variables. By encoding auditory and visual components into symbolic variables-capturing temporal utterance sequences and spatial entity coordinates, respectively-the model establishes cross-modal linking within this abstract space. Crucially, we reveal that when trimodal binding fails, the breakdown predominantly stems from misaligned audio-visual connections. To overcome this bottleneck, we introduce an audio-visual prompting method utilizing an off-the-shelf Active Speaker Detection (ASD) model. By simply overlaying visual bounding boxes on active speakers, this training-free approach yields immediate performance gains across four conversation-centric benchmarks. Moreover, lightweight fine-tuning of fewer than 300 steps on these ASD-prompted-videos extends these gains to three general AV benchmarks, suggesting the generalizability of our method.
[CV-38] askIR: Task-Driven Image Restoration via Degradation Adaptation and Task Feedback
链接: https://arxiv.org/abs/2609.31170
作者: Yanjie Tu,Qingsen Yan,Axi Niu,Wenxuan Cai,Tao Hu,Wei Dong,Haokui Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Task-driven image restoration aims to improve both image quality and downstream task performance. However, existing methods predominantly focus on single degradation type and struggle to handle the diverse degradations encountered in real-world scenarios. Different degradations impose distinct restoration demands, and insufficient restoration may leave residual degradations and artifacts that impair object boundaries and semantic cues, thereby compromising downstream task performance. To address these challenges, we propose TaskIR, a two-stage task-driven unified image restoration framework that integrates degradation-adaptive restoration with task feedback refinement. In Stage I, a Degradation Representation Module (DRM) extracts degradation representations, enabling a Degradation-Guided Transformer Block (DGTB) to dynamically modulate feature transformations for adaptive restoration. In Stage II, a Task-to-Restoration Feedback Generation module (TRFG) transforms heterogeneous task features into restoration feedback by modeling task-representation discrepancies associated with the current restoration. Subsequently, a Selective Task Feedback Refinement module (STFR) assesses feedback relevance and selectively refines intermediate restoration features to mitigate interference with well-restored content. Extensive experiments demonstrate that TaskIR achieves competitive restoration quality and downstream task performance across diverse degradations and tasks.
[CV-39] ReG-SAM: Reference Graph-Driven SAM for 2D Foundational Vessel Segmentation
链接: https://arxiv.org/abs/2609.31160
作者: Donghang Lyu,Zichen Zhang,Oleh Dzyubachyk,Marius Staring
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Vessel segmentation in medical images is essential for many clinical tasks, ranging from diagnosis to treatment planning. However, it remains challenging due to complex vascular morphology and diverse imaging conditions. Existing deep learning methods rarely aim at building a generalizable vessel segmentor across anatomies and modalities. While the Seg- ment Anything Model (SAM) has shown promise for med- ical image segmentation, its original design does not fully exploit vascular morphology and struggles with fine-grained vascular structures, leading to suboptimal performance. In this paper, we propose ReG-SAM, a SAM-based framework tailored to 2D vessel segmentation that leverages reference graph set for enhancing vascular representations. Specifically, we introduce two modality-aware representations derived from the reference masks: graph prompt embeddings (GPEs) that encode global spatial features from graphs, and vascu- lar prototype embeddings (VPEs) that capture fine-grained modality-specific vessel characteristics from multi-scale fea- ture maps and vascular masks. Since both require vascular masks that are unavailable during inference and require robust modality-aware vascular feature representations, we construct a modality-wise vascular database and develop two reference graph-guided representation learning schemes for estimating GPEs and VPEs using samples from the database rather than ground-truth masks. Extensive experiments across 19 datasets demonstrate that ReG-SAM consistently outperforms existing baselines, even those using manual prompts, particularly on challenging thin vessels.
[CV-40] HyperErase: Scale-Calibrated Hypernetwork for Multi-Concept Erasure in Text-to-Image Models
链接: https://arxiv.org/abs/2609.31154
作者: Yi Sun,Xinhao Zhong,Zhiqi Zhang,Yimin Zhou,Junhao Li,Yuxia Qiao
类目: Computer Vision and Pattern Recognition (cs.CV); Cryptography and Security (cs.CR)
备注:
Abstract:Recent advances in text-to-image (T2I) generation have substantially improved visual synthesis, but have also raised increasing safety concerns due to their potential to generate harmful or undesirable content. Existing concept erasure methods predominantly follow a static weight paradigm, producing a single frozen adapter that struggles to adapt to diverse prompt variations and suffers from parameter interference when scaling to multiple concepts. We propose \textbfHyperErase, a framework for concept erasure based on hypernetwork-driven prompt-conditioned parameter synthesis. Our approach first reframes concept erasure as prompt-conditioned parameter amortization and trains a hypernetwork to map textual descriptions to prompt-specific LoRA updates, eliminating the need for per-prompt gradient optimization or manual LoRA merging. To further improve the stability and precision of synthesized adapters, we develop a decoupled rectification strategy, which disentangles LoRA tokens into pattern and scale subspaces, applies a square-root transform to curb multiplicative over-scaling, and leverages teacher-derived canonical priors for inference-time correction. Extensive experiments across major concept categories demonstrate that HyperErase consistently improves the trade-off between erasure effectiveness, image quality, and semantic alignment, achieving performance comparable to gold-standard single-concept baselines. Furthermore, the resulting models can provide specialized LoRAs for each input prompt variation in a single forward pass without requiring gradient updates during inference. These principled and flexible framework offers a new paradigm for concept erasure in T2I models.
[CV-41] FedHisto-PAST: Parameter-Efficient Stain-Aware Federated Learning for Cross-Site Lung Histopathology Classification
链接: https://arxiv.org/abs/2609.31150
作者: Muhammad Muhtasim Shahriar,M. M. Golam Hafiz,Saad Aloteibi,Mohammad Ali Moni
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Submitted to Engineering Applications of Artificial Intelligence (Elsevier)
Abstract:Cross-site lung histopathology classification must account for stain variation, non-IID client data, missing classes, and the cost of adapting large pathology encoders. This study evaluates FedHisto-PAST v2 for three-way classification of adenocarcinoma (ACA), Normal, and squamous cell carcinoma (SCC). FedHisto-PAST v2 combines a frozen HIBOU-B foundation model with parameter-efficient adaptation, stain-conditioned paired-view prediction and feature consistency, reliability-aware prototype learning, and adaptive federated aggregation. Experiments used a five-client, non-IID, raw-data-local simulation with fixed internal evaluation, client-level analysis, component ablations, communication accounting, and a development-influenced exploratory LungHist700 cohort. All principal methods achieved near- ceiling internal performance, which limited discrimination on the fixed split. On LungHist700, FedHisto- PAST v2 achieved a Macro-F1 of 0.728560 and a balanced accuracy of 0.730454. Higher recognition of Normal and SCC was accompanied by lower ACA recall, and calibration remained imperfect. Prediction-level consistency was the only component with a clearly supported independent contribution in the external ablation analysis. Feature consistency and prototype regularization showed no conclusive independent overall gains in Macro-F1. The framework updated 1.253841% of the model parameters. The results provide exploratory cross-dataset evidence for stain-aware, parameter-efficient federation; they do not establish formal privacy, patient-level independence, prospective deployment, or clinical validation.
[CV-42] Seeing Semantic Shift: Difference-Aware Sentence-Level Temporal Segmentation of Sign Language Videos
链接: https://arxiv.org/abs/2609.31148
作者: Bowen Guo,Shiwei Gan,Yafeng Yin,Xiao Liu,Kuizhuang Liu,Zhiwei Jiang,Lei Xie
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Recent advances in sign language understanding have achieved impressive success on short, single-sentence videos, yet their performance drops sharply when applied to long, continuous sign language videos. To bridge this gap, we focus on a challenging and realistic setting: Visual-only Sentence-level Sign Language Segmentation (Vis-SSLS), which aims to partition continuous sign language videos into non-overlapping sentence-level segments without any caption assistance, serving as a crucial prerequisite for downstream recognition and translation tasks. However, sentence transitions in sign language are often smooth and visually ambiguous, lacking explicit pauses or posture resets. As a result, static frame representations may fail to capture the subtle temporal changes that indicate sentence boundaries. To address this challenge, we propose \textbfSignShift, a difference-aware segmentation framework that explicitly models frame-to-frame feature variation as semantic cues for sentence boundary detection. First, to model the feature variation, we design a Temporal Difference Module, which incorporates full-frame, facial, and hand cues, and employs inter-frame differencing to learn multi-scale temporal variations that capture both fine-grained local kinematics and global semantic transitions. Second, to mitigate over- and under-segmentation issues, we design a Segment Count Prediction module, which predicts the number of sentences to guide boundary selection. Extensive experiments on benchmark datasets demonstrate that SignShift substantially outperforms existing methods, validating its effectiveness.
[CV-43] Pocket-STVG: lightweight architecture for Spatio-Temporal Video Grounding
链接: https://arxiv.org/abs/2609.31135
作者: Alberto Presta,Michal Byra,Grzegorz Stefański,Karol Szurkowski,Eryk Kołodziejczyk,Krzysztof Arendt
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Multimedia (cs.MM)
备注: 14 pages total. 8 pages main manuscript, 3 pages references, 3 pages additional material
Abstract:Spatio-Temporal Video Grounding (STVG) aims to localize the spatio-temporal tube in a video corresponding to a natural language query. While recent methods achieve strong performance in fully supervised, weakly supervised, and zero-shot settings, they typically rely on computationally expensive architectures, complex training pipelines, or multimodal large language models. We present Pocket-STVG (P-STVG), a lightweight cascade architecture that addresses STVG by combining efficient pre-trained components instead of large end-to-end models. P-STVG integrates a temporal-aware video encoder based on MobileViCLIP, a spatial encoder-decoder derived from MDETR, and a shared aligned text encoder. Temporal localization is performed through either a lightweight 1D U-Net or a simple thresholding strategy, enabling the same framework to operate in both weakly supervised and zero-shot settings. Furthermore, video representations are precomputed independently of the query, yielding an indexing-friendly pipeline for efficient inference and large-scale video collections. Despite requiring fewer than 90M parameters, P-STVG performs on par with weakly supervised methods and improves on earlier zero-shot approaches at a fraction of their memory and computational cost, establishing a favorable performance-efficiency trade-off for STVG.
[CV-44] Double-stream registration with pyramid fusion for HDR video with alternating exposures
链接: https://arxiv.org/abs/2609.31108
作者: Onofre Martorell,Ivan Pereira-Sánchez,Antoni Fuentes,Antoni Buades
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 5 pages, double column, IEEE format
Abstract:High dynamic range (HDR) video reconstruction from al-ter-na-ting-exposure sequences remains challenging, especially in regions with extreme luminance variation. We propose a novel HDR reconstruction framework based on dual-stream registration and accurate pyramid fusion. Given three consecutive frames, our method computes optical flow directly with the central frame, while introducing a complementary midpoint displacement strategy to handle cases with severe overexposition. A pyramid fusion stage then merges the resulting radiance and LDR images into a final HDR output. Experimental results demonstrate that our approach consistently outperforms state-of-the-art methods.
[CV-45] DepthEvidence: Unifying Metric Depth Prediction and Geometric Reasoning in Multimodal Language Models
链接: https://arxiv.org/abs/2609.31103
作者: Jiangning Wei,Yuan Yao,Miaomiao Cui,Mingsheng Li,Humen Zhong,Shuai Bai,Zhibo Yang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Spatial reasoning with metric constraints requires linking objects to geometric measurements and preserving their numerical content during language reasoning. We present DepthEvidence, a 4B model that uses its own dense metric predictions as object-grounded evidence for language generation. A camera-conditioned decoder predicts full-resolution metric depth using multi-scale visual features and high-resolution RGB refinement. A dense-to-language interface converts predicted depths and decoder features into object-aligned continuous geometry tokens anchored to object identifiers. Geometric supervision encourages metric information to remain recoverable before and after language-context interaction, while instruction tuning supports object measurement and compositional reasoning. We introduce a Depth-VQA benchmark evaluating object-depth queries, relative comparisons, and decisions combining spatial and numerical constraints. Across nine datasets, DepthEvidence achieves the highest average dense \delta_1 among evaluated methods, competitive with specialized estimators. It also leads the evaluated methods in instance-level metric depth estimation and overall accuracy on both relative and metric reasoning tracks, while broadly preserving general VQA performance and improving spatial understanding relative to the base model.
[CV-46] Band-Selection Stability and Semantic Segmentation Performance: A Study on Hyperspectral City
链接: https://arxiv.org/abs/2609.31074
作者: Jiarong Li,Imad Ali Shah,Enda Ward,Martin Glavin,Edward Jones,Brian Deegan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted for IEEE WHISPERS 2026
Abstract:Resource constraints make high-dimensional hyperspectral imaging challenging in autonomous perception, motivating the use of band selection methods. However, the sensitivity of band-selection methods to sampled data and their relationship to semantic segmentation models (SSMs) remain underexplored. This study evaluates six band selection methods on ten independently sampled, class-balanced region-of-interest (ROI) sets, yielding 60 top-25 band subsets from the Hyperspectral City V2 (128 bands: 450-950nm) dataset. Top- K bands ( K\in\3,5, … 13\ ) from the first three ROI sets are evaluated with three SSMs against the corresponding 128-band baseline. Experiments show that intra-method stability is method-dependent: Sim-LP shows the highest stability (pairwise Jaccard similarity) and, together with JMIM+CSNR, yields the best segmentation results. Top- K based SSMs remain competitive with baselines, with gains of up to 2.01 mIoU and 1.72 mF1 points, and 18-22x faster CPU inference for K=9 . However, performance does not improve monotonically with K , and stability shows no consistent association with SSM performance. These findings suggest that intra-method stability is informative but an unreliable indicator of downstream segmentation performance, highlighting the need to evaluate band-selection methods across repeated samples, subset sizes, and SSMs.
[CV-47] Where Compute Matters: Heterogeneous Attention for Efficient Video Diffusion
链接: https://arxiv.org/abs/2609.31050
作者: Olga Zatsarynna,Denis Korzhenkov,Juergen Gall,Amir Habibian,Mohsen Ghafoorian
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Efficient video generation requires reducing the quadratic cost of self-attention over long spatio-temporal token sequences. Existing efficient-attention methods typically apply the same computation pattern to every token, even though denoising difficulty varies substantially across video regions and evolves throughout the generation process. We introduce HetA-DiT, a heterogeneous attention mechanism that adaptively allocates computation according to token difficulty. A lightweight uncertainty branch predicts a token-wise estimate of denoising difficulty, which is used to route uncertain tokens through dense global attention while processing more reliable tokens with efficient local attention. The resulting routing is content- and timestep-adaptive, retains global context where it matters most, and provides a single parameter for controlling the quality-efficiency trade-off. HetA-DiT is compatible with few-step distribution-matching distillation and introduces no additional Transformer evaluation at inference time by reusing uncertainty estimates from the preceding denoising step. We evaluate the method on DMD-distilled Wan2.2-5B and Wan2.1-1.3B models. Across VBench, VBench-2.0, and human preference evaluation, HetA-DiT maintains competitive generation quality while routing only approximately 20% of tokens through dense attention.
[CV-48] Exploiting Spatial Structure for Transductive Few-Shot Classification of Whole-Slide Images
链接: https://arxiv.org/abs/2609.31040
作者: Tiffanie Godelaine,Manon Dausort,Karim El Khoury,Benoît Gérin,Benoît Macq,Christophe De Vleeschouwer
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 5 pages, 2 figures
Abstract:Automating the analysis of whole-slide images (WSIs), a key step in cancer diagnosis, has high clinical value, as it can reduce pathologist’s workload while improving diagnosis accuracy. Recently, vision-language models have shown promising performance for patch-level classification without requiring any annotation, yet these zero-shot (ZS) predictions remain noisy on fine-grained tasks and must be further refined. A promising direction is to refine all predictions jointly, i.e., a transductive approach. However, most existing methods are not tailored to WSIs. We thus propose SlideTIM, an adaptation to WSIs of the recent transductive approach LC-TIM, which introduces a combined spatial–latent regularizer together with a prior on the patch class distribution. The former enforces spatially and semantically close patches to receive the same predictions, while the prior calibrates the predicted class proportions. Together, they address the complex spatial organization and the strong class imbalance of WSIs. Evaluated on four histology datasets, SlideTIM consistently outperforms all TIM variants, improving the macro-F1 by +8.1pp over the best competing baseline at 1 shot. Compared to the ZS, it raises the macro-F1 by +19.4pp at 1 shot. The code will be made available after submission.
[CV-49] mpQ-Jail: Query-Constrained Candidate Ranking for Text-to-Video Jailbreak Attacks
链接: https://arxiv.org/abs/2609.31032
作者: Tianmeng Fang,Jiancheng Wang,Chen Wang,Liming Wang,Wei Wang,Jiayang Liu,Xiaochun Cao
类目: Multimedia (cs.MM); Cryptography and Security (cs.CR); Computer Vision and Pattern Recognition (cs.CV)
备注: 17 pages, 4 figures, 4 tables
Abstract:Existing text-to-video (T2V) jailbreak methods mainly seek more effective or stealthier attack candidates. In guarded T2V systems, however, video generation and security evaluation are costly, so an attacker often cannot test a large candidate pool. We therefore formulate T2V jailbreak as a query-constrained candidate allocation and ranking problem and propose TempQ-Jail. The method combines heterogeneous attack mechanisms to expand candidate coverage, estimates each candidate’s end-to-end attack value from security-gate passage, dangerous visual generation, preservation of the original intent, and temporal validity, and ranks candidates so that high-value attacks appear early in a limited query trajectory. We evaluate TempQ-Jail on CogVideoX-5B using 70 common viable intents derived from T2VSafetyBench and compare it with six representative T2V jailbreak methods under a unified protocol. TempQ-Jail achieves TP-ASR@5 and TP-ASR@10 of 48.9% and 65.4%, improving over the strongest baselines by 4.6 and 4.0 percentage points, respectively. It also obtains the highest AUC-TP (0.469) and the lowest AvgQ (6.3). Analyses of query trajectories, candidate allocation, failure attribution, and ablations show that TempQ-Jail more effectively identifies and prioritises candidates with complete attack potential under limited query budgets.
[CV-50] Refining Cytology Predictions with Conditional Random Fields
链接: https://arxiv.org/abs/2609.31028
作者: Manon Dausort,Tiffanie Godelaine,Karim El Khoury,Maxime Zanella,Christophe De Vleeschouwer,Benoît Macq
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 5 pages, 2 figures
Abstract:Vision-language models (VLMs) achieve strong zero-shot (ZS) classification on histology images but do not perform as well on cytology, whose stains and cell morphology differ markedly compared to histology. Conditional random fields (CRFs) can refine noisy VLM predictions by propagating information across patches, but existing CRF frameworks were designed for histopathology and do not transfer to cytology datasets, released as independent patch pools spanning multiple staining protocols. We introduce CytoCRF, which adapts the pairwise terms to cytology by targeting chromatin and cytology-specific staining, and further enrich the neighborhood of each potential term by combining multiple backbones. Across ten cytology datasets, CytoCRF outperforms existing CRF frameworks at every annotation budget, reaching +13.6 percentage points over the best baseline and +33.7 over ZS with only 50 annotations. Combining information from multiple backbones brings further gains, showing that the neighborhood topology matters more than the pairwise potential computed over it.
[CV-51] RACKGRAPH: Online Open-Vocabulary 3D Scene Graphs via Image-Space Tracking
链接: https://arxiv.org/abs/2609.31005
作者: Peder Borge Hellesylt,Albert Gassol Puigjaner,Kostas Alexis,Annette Stahl
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注:
Abstract:Open-vocabulary 3D maps enable robots to reason about previously unknown environments using natural language. However, existing systems typically segment every incoming image, associate detections with persistent 3D segments, and frequently perform costly Vision-Language (VL) inference. We present TRACKGRAPH, an online open-vocabulary system that maintains short-term 2D mask identity directly in the image stream before fusing segments into 3D. FastSAM masks and CLIP features are computed at sparse keyframes, while dense DINOv3 features are used to propagate masks at a high rate in between. The resulting tracked masks are fused into a class-agnostic 3D segment layer within a hierarchical scene graph, with 3D association handling tracking interruptions and long-term revisits. Compact multi-view CLIP embeddings enable open-vocabulary retrieval. Across Replica, ScanNet++, and HM3D, TRACKGRAPH achieves competitive open-vocabulary segmentation and retrieval against state-of-the-art mapping methods, including the highest synonym frequency on Replica (0.50). On the same NVIDIA A100, it is 1.7x faster and uses 3.3x less GPU memory than ViT-H OVI-MAP. Real-world quadruped deployments demonstrate onboard scene graph construction and object search at 7.5Hz, while recorded drone data is used to test the method under aerial viewpoints.
[CV-52] Can Pixels Alone Reveal Image Origin? Minimax Limits and Learnable Interfaces for Passive Provenance NEURIPS2026
链接: https://arxiv.org/abs/2609.30997
作者: Kai Yao
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Accepted at the 40th Annual Conference on Neural Information Processing Systems (NeurIPS 2026). 29 pages, including technical appendices. Code: this https URL
Abstract:Passive image provenance asks whether pixels alone can reveal where an image came from: a human, an aggregate AI class, or a particular generator. This becomes a robustness problem once a source image can be edited before the verifier sees it. We study the problem as source–target verification under adversarial distribution shift. Our first result gives the exact best-case limit for any image-only verifier: the largest robust target-acceptance gap equals the minimum total-variation distance between the target distribution and the set of attacked source distributions. This quantity depends on the source, target, and edit class, not on the verifier architecture. Our second result explains why deployed public verifiers can fail before this statistical limit is reached. If the verifier can be emulated on the attack region to error \varepsilon , then a surrogate black-box attack reaches target acceptance within 2\varepsilon plus optimization error of the white-box optimum; score-revealing logistic and softmax heads over public features are identifiable, and approximate score access gives stable recovery bounds. A finite-state experiment checks the minimax identity where both sides are computable. On same-prompt real/diffusion benchmarks, the evaluated public CLIP verifiers fail under targeted pixel attacks, while a ResNet-18 victim exhibits partial fake-to-real transfer. Binary feedback with abstention reduces measured attack success, but positive empirical gap upper bounds do not establish robustness. These results motivate separate evaluation of the source–target statistical ceiling and the information released by a deployed verifier.
[CV-53] FLIP: Final Layer Inference-Time Probing for Vision-Language Models ICML2026
链接: https://arxiv.org/abs/2609.30993
作者: Drandreb Earl O. Juanico,Rowel O. Atienza
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 25 pages, 14 figures, 5 tables. Accepted at the Mechanistic Interpretability Workshop at ICML 2026, Seoul, South Korea
Abstract:We present FLIP, a final-layer inference-time probe for testing whether a logit-facing intervention site in an open-weight vision-language model (VLM) supports structured, task-linked computation rather than generic perturbation. Behavioral change under internal intervention is otherwise mechanistically ambiguous: it may reflect improved use of visual evidence, generic output instability, or outright degradation. FLIP applies elementwise flooring to the final normalized hidden state before logit computation, leaving parameters, prompts, and decoding unchanged. On a controlled detection/counting probe, sweeping intervention strength reveals three regions: negligible change, a bounded interior regime in which detection recall at IoU 0.50 ( R_50 ) improves while tolerant counting error ( \mathcalE_\mathrmcount ) falls, and over-suppression. We formalize a four-criterion probe-and-sweep protocol for disciplining the interpretation of intervention effects: regime structure, grounding-proxy alignment, feature-coherence dependence, and failure to reproduce the same positive regime on a performance-based negative control. The post-normalization state passed to the output head is the logit-facing instantiation of this test; under a non-targeted flooring sweep it satisfies the full protocol. Raw decoder-layer interventions, including the last-block output before final normalization, and the singleton-pair left/right control fail to reproduce the Final-site signature, while same-site operators and multiple VLMs replicate it. FLIP is therefore a validation step for intervention-based mechanistic interpretability, not a steering method.
[CV-54] PICO: Projection-Informed Consistency Optimisation for 6DoF Surgical Tool Pose Estimation
链接: https://arxiv.org/abs/2609.30989
作者: Lucy Fothergill,Pietro Valdastri,Dominic Jones,Duygu Sarikaya
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Purpose: Accurate 6 DoF pose estimation of surgical tools is critical for automa- tion, robotic proprioception, and safe interaction with the tissue operated on. Kinematics-based approaches suffer from accumulated errors due to the cable- driven nature of robotic arms, while vision-based methods often rely on external markers or trackers. Although more recent vision-based advances have been pro- posed, these two-stage pose estimation methods often lack real-time robustness due to accumulated errors and computational overhead. Methods: We propose a novel end-to-end trainable model, PICO. Our model employs a multi-task learning architecture to predict segmentation and depth maps, alongside regression of translation and rotation parameters. We define two proxy tasks that enforce geometric consistency in both 2D and 3D spaces, improving accuracy and robustness. For this, we propose a projection loss, and a point-to-point loss. Results: We evaluate our method on the SurgRIPE dataset, benchmarking its performance against state-of-the-art approaches using standard 6DoF pose esti- mation metrics. Our results demonstrate consistently strong performance across all four subsets, specifically in rotation, ranking second even under occlusion. It also demonstrates comparable translational performance, remaining competitive, especially in occluded cases. Conclusion: PICO demonstrates the effectiveness of multi-task learning and geometry-aware proxy tasks for robust and reliable surgical tool pose estimation, especially in occluded scenarios, highlighting potential for future applications.
[CV-55] PhoenixSR: Generative Heterogeneous Distillation Unleashes Efficient Models for Real-World Super-Resolution
链接: https://arxiv.org/abs/2609.30988
作者: Xin Di,Mingyu Shi,Yuanfei Bao,Long Peng,Yue Zhao,Jiaming Guo,Renjing Pei,Xueyang Fu,Yang Cao,Zheng-Jun Zha
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Real-world image super-resolution (SR) requires recovering perceptually realistic high-resolution images from complex low-resolution observations while preserving faithful content. Diffusion-based SR benefits from strong generative priors but incurs substantial computational overhead, whereas feed-forward CNN and Transformer SR models are efficient yet often struggle to recover realistic high-frequency details. This motivates a natural question: can diffusion priors be transferred to existing diffusion-free SR networks without introducing diffusion components at inference time? To this end, we propose PhoenixSR, a generative heterogeneous distillation framework that transfers diffusion priors to independently designed feed-forward SR networks through score-based distribution matching. Rather than aligning heterogeneous features or imitating sampled diffusion outputs, PhoenixSR uses the pretrained diffusion model as distribution-level supervision, while paired SR supervision preserves reconstruction fidelity. To make distribution matching effective for fidelity-sensitive SR, we introduce Heterogeneous Distribution Adaptation, which adapts the target score to the SR domain, improves tracking of the evolving student distribution, and anchors training with paired supervision. We further employ Directional Reliability Weighting, a lightweight residual-consistency-based reweighting strategy that reduces unstable distributional guidance. All diffusion-related components are removed after training, leaving the original student architecture and inference cost unchanged. Experiments on three SR benchmarks and six feed-forward backbones, including SwinIR, HAT, Real-ESRGAN, and SeeMoRe, show consistent perceptual improvements with largely preserved reconstruction fidelity.
[CV-56] Self-Supervised Perceptually Interpretable Monocular Depth Estimation ICIP2026
链接: https://arxiv.org/abs/2609.30987
作者: Zain Ul Abidin,George Dimas,Dimitris K. Iakovidis
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Published at IEEE ICIP 2026; 6 pages, 4 figures
Abstract:Self-supervised monocular depth estimation (MDE) enables depth prediction from monocular images without requiring ground-truth supervision, making it attractive for large-scale and real-world applications. Despite steady improvements in accuracy, most existing methods remain difficult to interpret, as depth is inferred from RGB representations that obscure the impact of individual perceptual image components. This lack of transparency limits systematic analysis of failure cases and reduces confidence in safety-critical settings. This paper presents a self-supervised framework for perceptually interpretable monocular depth estimation (PIMDE), designed to associate depth predictions with distinct perceptual components of the input image. Rather than operating directly on RGB inputs, the proposed method decomposes each image into a set of perceptual feature maps (PFMs), each encoding a specific visual cue. Distinct depth estimation branches process these PFMs independently to produce depth estimates (PIDEs), which are subsequently combined through an explicit fusion strategy. This formulation allows us to examine directly the contribution of each perceptual cue to the final depth prediction. Experiments conducted on the KITTI benchmark dataset demonstrate that PIMDE achieves performance comparable to established self-supervised MDE methods while providing additional insight into how different perceptual cues influence depth estimation. These results indicate that perceptual decomposition can support interpretability without sacrificing depth estimation accuracy.
[CV-57] FARE: Forensic Acceptance Region Estimation for Catching Bait-and-Switch Image Generators NEURIPS2026
链接: https://arxiv.org/abs/2609.30982
作者: Kai Yao,Marc Juarez
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注: This work has been accepted for publication in the proceedings of The 40th Annual Conference on Neural Information Processing Systems (NeurIPS 2026). 22 pages, including technical appendices. Code: this https URL
Abstract:Modern AI image generators are increasingly deployed as opaque APIs, where customers can query the deployed service, but cannot inspect model weights or architecture. This creates a practical challenge: a provider may pass governance certification with one generator and later silently switch to a cheaper and lower-quality one for deployment, compromising public trust or even safety in high-stakes domains. We study integrity auditing at deployment time and propose FARE (Forensic Acceptance Region Estimation). A certified generator is enrolled by training FARE on images sampled from that generator. After deployment, FARE can determine whether a generated image is consistent with the enrolled generator—using only that image. FARE’s features are based on image generator-specific artifacts that have been proposed for forensic applications. FARE amplifies these features during training by finding hard samples that tighten the acceptance region and increase sensitivity to subtle changes in the certified generator. Across generator swaps, including substitutions with similar model versions and model variants, FARE is effective at detecting swaps, consistently outperforming existing baselines at strict operating points, and remains effective under the exact-model and decision-only attacks evaluated in this work.
[CV-58] STORM-Bench: Evaluating Online Video QA under Evolving and Incomplete Evidence
链接: https://arxiv.org/abs/2609.30981
作者: Siru Zhong,Shenghan Tan,Rihong Yan,Xiaohui Lv,Yuzheng Zhuang,Shuai Tao,Wulong Liu,Haohuan Fu,Yuxuan Liang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 50 pages, 19 figures, 27 tables
Abstract:Reliable online video question answering requires tracking state transitions while selectively abstaining when visual evidence is insufficient. Existing benchmarks focus on static recognition or long-range retrieval, rarely evaluating these coupled capabilities under evolving and incomplete evidence. We present STORM-Bench, comprising 5,736 questions across 630 compact, change-dense episodes spanning five egocentric domains (STORM-Real) and two controlled simulation subsets (STORM-Sim) at 1 FPS. Questions are stratified by a proxy for accumulated change intensity (Low, Medium, High) and query-time answerability (Known, Uncertain). To measure reliability, we introduce STORM-BR, a harmonic metric over joint answer-status correctness that exposes abstention failures masked by aggregate accuracy, alongside STORM-BR-ATTR for uncertainty attribution. Across 14 video LLMs, online accuracy peaks at 60.3% (mean 51.7%), whereas STORM-BR ranges from 5.7% to 35.6% (mean 18.8%), driven by pervasive overconfidence on uncertain queries. STORM-Bench shows that task accuracy masks these gaps in epistemic reliability and state tracking. Benchmark and code are available at this https URL.
[CV-59] FeatMark: Feature-level Watermark Protection against Mimicry Attacks with Diffusion Models
链接: https://arxiv.org/abs/2609.30980
作者: Haoyang Li,Ruoxi Sun,Qingqing Ye,Benjamin Zi Hao Zhao,Yaxin Xiao,Jason Xue,Haibo Hu
类目: Cryptography and Security (cs.CR); Computer Vision and Pattern Recognition (cs.CV)
备注: 19 pages, 7 figures, 14 tables; includes appendices
Abstract:Text-to-image diffusion models enable data-efficient “mimicry” attacks, wherein adversaries fine-tune the model on a handful of public photos to synthesize convincing forgeries of a target individual. A common countermeasure is to embed imperceptible, low-energy watermarks, yet recent studies show these signatures are brittle: modest post-processing or lightweight adversarial perturbations readily suppress detection, exposing a fundamental tension between imperceptibility and robustness. We introduce FeatMark, a watermarking framework that shifts from pixel-level, energy-starved perturbations to inconspicuous semantic features: small, scene-consistent micro-features that remain natural to humans while providing a stronger, machine-verifiable provenance signal. FeatMark builds domain-specific feature banks that encode each watermark as a compact concept program, pairing open-vocabulary semantic cues with reliable edit regions and instruction templates. It then automatically selects features that are both feasible and executable and injects them through modular, mask-guided concept editing, yielding highly localized, scene-consistent micro-edits that are difficult to perceive. We conduct extensive experiments across VGGFace2, CelebA-HQ, and WikiArt, evaluating against 10 strong watermark removal/purification attacks (including regeneration-style purification) and several bespoke adaptive attacks tailored to FeatMark, to assess perceptual fidelity, watermark detection accuracy, and robustness. We further demonstrate FeatMark’s extensibility to video mimicry attacks. The results show FeatMark remains virtually impervious, withstanding all evaluated attacks with negligible bit-accuracy and fidelity degradation.
[CV-60] CCRV-Bench: Constraint-Based Evaluation of Causal Reasoning in Vision-Language Models
链接: https://arxiv.org/abs/2609.30979
作者: Linyuan Gao,Yuan Wu,Yi Chang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 21 pages, 5 figures, 12 tables
Abstract:Vision-language models (VLMs) have demonstrated excellent performance in visual tasks, but their visual causal reasoning capabilities still lack reliable evaluation. Existing evaluations struggle to distinguish whether a model is performing causal reasoning based on visual evidence or relying on statistical correlations for shortcut learning, thereby potentially overestimating their actual capabilities. This paper proposes CCRV-Bench, a constraint-driven visual causal reasoning benchmark for single-image physical scenarios. We construct an orthogonal framework that evaluates four causal task dimensions: causal relation discovery, state prediction, causal diagnosis, and intervention. We further introduce entity symbolization, spatial grounding, the factual adversarial constraint, and minimalist output constraints to reduce shortcut cues while preserving the physical commonsense required by the task. Experiments across 15 multimodal models show that constraint sensitivity is task- and model-dependent: intervention has the largest average effective degradation among the four causal tasks, spatial grounding is the most damaging constraint on average, and the factual adversarial constraint improves DCR for all evaluated models. These results show that unconstrained performance does not determine constrained robustness and that a single aggregate score can obscure distinct failures in causal identification, spatial grounding, and constraint-compliant expression. CCRV-Bench provides a standardized framework for diagnosing image-grounded causal reasoning under controlled constraints. The code is available at this https URL
[CV-61] Where and When to Force: Routed Forcing for Streaming Avatars
链接: https://arxiv.org/abs/2609.30963
作者: Zihan Su,Siwen Lu,Junhao Zhuang,Zeyue Xue,Haoyang Huang,Guanghao Li,Xiaofeng Tan,Chun Yuan,Nan Duan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Audio-driven streaming avatar generation requires real-time synthesis of speech-synchronized videos with dynamic and diverse motion. Self Forcing uses Distribution Matching Distillation (DMD) to distill bidirectional video diffusion models into causal, few-step generators for real-time streaming. However, DMD minimizes a reverse KL divergence, which is inherently mode-seeking: it causes the student to discard high-dynamic modes and collapse onto static outputs, compressing both dynamics and diversity of generated videos. We find that this collapse is region-heterogeneous: person regions involving pose and gesture variations suffer the largest diversity loss, the audio-driven mouth region shows a small loss, and the background remains nearly stable. Based on this observation, we propose Routed Forcing, which routes the distillation objective by semantic region and noise stage to improve dynamics and diversity while preserving visual quality. Specifically, (1) Where to Force: Semantic-Region Routing applies Data-Forcing Distillation (DFD), which supervises the student with real videos, to the person region where diversity collapse is most severe, while retaining DMD for the mouth and background to preserve lip synchronization and scene stability. (2) When to Force: Noise-Stage Routing activates DFD at high noise stages, where real video serves as effective supervision to inject diverse and dynamic motion patterns. At low noise stages, DMD is used to refine details, avoiding blur and artifacts from spatial differences between real video and student-generated video. Experiments show that Routed Forcing improves dynamics by up to 45% and diversity by 7-25% over Self Forcing, while preserving video quality and lip synchronization.
[CV-62] IDM-Net: A Lightweight Illumination-Decoupled Modulation Network for Low-Light Image Enhancement
链接: https://arxiv.org/abs/2609.30962
作者: Cheng-Yen Hsiao,Jing-Ming Guo
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Low-light image enhancement (LLIE) remains challenging for lightweight models because illumination restoration and color fidelity are difficult to optimize simultaneously in the RGB color space. Although recent color-decoupled methods separate luminance and chrominance representations, they primarily optimize luminance as an enhancement target, leaving its potential as an explicit guidance prior largely unexplored during feature reconstruction. To address this limitation, we propose IDM-Net, a lightweight Illumination-Decoupled Modulation Network for low-light image enhancement. IDM-Net adopts a dual-encoder architecture consisting of a structure encoder that extracts multi-scale appearance features from the RGB image and a lightweight illumination encoder that learns illumination priors from the decoupled luminance (Y) channel. To effectively exploit these priors, we introduce an Illumination-Guided Modulation (IGM) module that injects multi-scale illumination cues into the decoder through spatially adaptive affine modulation, enabling accurate brightness restoration while preserving natural color consistency. Furthermore, we design a lightweight Feature Refinement Block (FRB) to progressively suppress degradation artifacts and recover fine-grained image details during reconstruction. Extensive experiments on multiple standard low-light image enhancement benchmarks demonstrate that IDM-Net achieves competitive performance among lightweight LLIE methods while maintaining an excellent balance between restoration quality and computational efficiency.
[CV-63] MVVBench: Benchmarking 4D Reasoning in Vision-Language Models NEURIPS2026
链接: https://arxiv.org/abs/2609.30952
作者: Hyungjin Chung,Byeongjun Park,Joonseok Lee,Hojun Kim,Jaeho Choi,Byung-Hoon Kim
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: NeurIPS 2026, 23 pages, 8 figures
Abstract:Multi-view video understanding requires integrating spatial and temporal evidence across multiple, often non-overlapping camera streams: tracking entities as they transition between viewpoints, aligning events across time, and reasoning about latent 4D continuity rather than any single visible frame. We introduce MVVBench, a benchmark for multi-view video reasoning built from real world multi camera datasets. Questions are curated to be monocular-ambiguous along both the view and the temporal axis: each question is unanswerable from any single view in the designated input set, and the majority are further unanswerable from any single moment. Each question becomes uniquely solvable only by jointly reasoning across views and across time. MVVBench spans diverse dynamic scenes and probes six capabilities: implicit/explicit attribute identification, implicit/explicit relative distance, relative camera pose, and compositional counting, with human-authored QA and rigorous verification. Beyond benchmarking, we provide an extensive analysis of when and why current vision language models succeed or fail, characterizing errors due to temporal mis-localization, cross-view identity breaks, and brittle multi-hop reasoning. We then study inference-time elicitation strategies that unlock latent multi-view competence—task-specific chain-of-thought scaffolds and structured cross-view evidence aggregation—yielding substantial gains without retraining. Finally, we present preliminary evidence that reinforcement learning with verifiable rewards can elicit some latent multi-view competence in the base model, pointing to training-time approaches as a promising direction for future work. Together, MVVBench offers a rigorous evaluation of 4D multi-view reasoning and a foundation for future progress toward reliable embodied perception.
[CV-64] DAPEVO: Deep Adaptive Patch Frame-Event Visual Odometry
链接: https://arxiv.org/abs/2609.30947
作者: Luca Gandolfi,Simone Nascivera,Roberto Pellerito,Rong Zou,Chiara Plizzari,Davide Scaramuzza
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Visual odometry is essential for autonomous navigation in GPS-denied environments, yet RGB-based methods remain vulnerable to motion blur, challenging illumination, and dropped frames. Event cameras complement conventional cameras with high temporal resolution and dynamic range, but their asynchronous measurements complicate reliable correspondence estimation. We present DAPEVO, a learned visual odometry system that estimates image and event correspondences independently at shared patch locations and fuses their correlation evidence before motion refinement. Each tracked patch maintains image and event descriptors, and a learned scalar gate combines modality-specific correlation embeddings for each patch–frame edge before a shared recurrent refinement and bundle-adjustment update. DAPEVO also supports event-only observations, enabling continued tracking when RGB frames are sparse or unavailable, while modality-aware keyframe culling preserves scarce frame constraints. On UZH-FPV, when retaining only one in six RGB frames, DAPEVO’s mean absolute trajectory error (ATE) increases by only 36%, from 1.00 to 1.36m, whereas the ATE of DPVO and RAMP-VO rises by factors of 3.7\times and 3.1\times , respectively. On TartanEvent, DAPEVO similarly remains below 1m ATE at 3Hz RGB input, while DPVO and RAMP-VO exceed 9m. Under degraded RGB input on TartanEvent, DAPEVO achieves an ATE of 0.60m, compared with more than 4m for both DPVO and RAMP-VO, while also outperforming event-only DEVO at 0.87m.
[CV-65] OneWorld: Learning Consistent Physics Across Actions in World Models
链接: https://arxiv.org/abs/2609.30946
作者: Ke He,Yichen Ding,Bin Yang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 27 pages, 4 figures
Abstract:Action-conditioned video world models aim to predict scene evolution under different actions, a capability that is essential for reliable planning, decision-making, and interaction in dynamic environments. However, futures generated independently from the same initial scene may each appear plausible while implying incompatible physical properties, such as friction or mass. This inconsistency can lead to contradictory predictions across interventions, making it difficult for the model to maintain a coherent understanding of the underlying world and limiting its reliability for planning and decision-making. To address these issues, we propose OneWorld, a shared-mechanism counterfactual generation framework that jointly models multiple action-conditioned futures under a common latent physical mechanism. A physical mechanism interpreter first infers a distribution over latent mechanisms from each action-outcome branch. These distributions are then aggregated into shared-world evidence, which captures whether the branches admit a common physical explanation while accounting for uncertainty in less informative branches. This evidence constrains flow training and guides sampling, encouraging consistency in the underlying physical mechanism while preserving the distinct outcomes induced by different actions. We further introduce a multi-intervention evaluation protocol in controlled environments, following the interaction settings of ACWM-Phys, to assess whether generated futures can be jointly explained by the same physical parameters, alongside standard measures of single-rollout prediction quality. Experiments in these environments show that OneWorld improves cross-intervention physical consistency while maintaining competitive single-rollout prediction quality.
[CV-66] Spackle: Completing Large View Single Image NVS with Adaptive Gaussians
链接: https://arxiv.org/abs/2609.30941
作者: Xuanzhi Liu,Yuhe Zhou,Xinyi Wu,Zhenyao Wu,Jinghao Chen,Ruize Han,Song Wang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Single-image novel view synthesis (NVS) enables photorealistic rendering of un- observed viewpoints from a single input. Practical NVS systems require two key capabilities: robust reconstruction of occluded regions and high inference effi- ciency. While hybrid decoupled frameworks combining feedforward 3D Gaussian Splatting (3DGS) and diffusion models show promise for large-view-deviation NVS, they suffer from capacity competition: a fixed number of Gaussians forces resource shifts from visible to newly disoccluded areas, degrading original scene fidelity when the target view deviates significantly from the input. To address this, we propose Spackle, a lightweight residual learning framework that mit- igates capacity competition without sacrificing efficiency. Spackle operates in three stages: predicting base 3DGS attributes from given views, automatically identifying poorly reconstructed regions, and learning a residual 3DGS optimized exclusively for these areas. At inference, we combine the baseline and aug- mented Gaussians for NVS. We conduct comprehensive experiments and show that Spackle achieves state-of-the-art performance on large-view-deviation cases.
[CV-67] ManiVid: Unified and Explainable Forensic Analysis of Manipulated Videos
链接: https://arxiv.org/abs/2609.30934
作者: Hengrui Kang,Zhonghao Yan,Yuxuan Yang,Ruoyan Jing,Yuncheng Guo,Hao Chen,Kongming Liang,Zhanyu Ma,Conghui He,Weijia Li
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Rapid advances in AI-generated video (AIGV) have increased the risks posed by deceptive video manipulation. Unlike fully synthetic videos, manipulated videos retain most source content and alter only localized regions, making forensic analysis particularly challenging. Existing video forgery research faces two limitations in both data and methodology: (1) High-quality datasets and benchmarks tailored for manipulated videos remain scarce. (2) Multimodal large language models (MLLMs) extend forgery analysis beyond binary classification but struggle to use low-level forensic cues and provide precise pixel-level grounding. Specifically, we introduce ManiVid, a unified forensic analysis task covering forgery detection, artifact grounding, and anomaly explanation for manipulated videos. We construct ManiVid-38K, the first dataset to combine paired, open-vocabulary localized manipulations of general videos with authenticity labels, forgery masks, and anomaly explanations. It comprises about 19K manually verified real-fake video pairs, mostly at 1080P resolution, generated under 2 paradigms with 15 powerful generation models. We sample 1K pairs for ManiVidBench, balanced across six manipulation types and generation models for fair evaluation. We further propose ManiVidLens, a unified framework for explainable video forgery analysis. Its Forensic Evidence Router supplies shared low-level forensic evidence for multimodal reasoning and video segmentation. Its Prompt Distill Module converts grounding states into semantic and geometric prompts and distills spatial priors for mask decoding and full-video propagation. ManiVidLens achieves relative gains over the strongest comparison methods in artifact grounding (+21.1% mIoU; +21.3% JF) and anomaly explanation (+131.3% ROUGE-L; +9.9% CSS). Its forgery detection remains comparable to dedicated classifiers (0.914 Acc; 0.913 F1).
[CV-68] UltraG -Bench: A Multi-task Benchmark for assessing Large Vision-Language Models on Pixel-level Evidence Grounding in Ultrasound
链接: https://arxiv.org/abs/2609.30928
作者: Quanhao Zhu,Bo Xu,Rui Lin,Chenyuan Wang,Yu Shao,Boling Zhu,Jiuyan Sun,Liang Zhao,Hongfei Lin,Feng Xia
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Ultrasound is one of the most widely used medical imaging modalities, and recent large vision-language models(VLMs) have shown increasing capabilities in ultrasound image understanding. However, these models fail to provide pixel-level visual evidence aligned with their semantic predictions, and their fine-grained grounding capability in ultrasound remains largely unclear. We introduce UltraG-Bench, a large-scale multi-task benchmark for evaluating pixel-level evidence grounding in ultrasound. UltraG-Bench is built by annotating 40 public ultrasound segmentation datasets spanning 13 anatomical categories, and comprises three progressive tasks: instruction-guided segmentation, evidence-grounded VQA, and evidence-grounded report generation, with 331125, 666779, and 138832 annotations, respectively. Comprehensive evaluation of 14 state-of-the-art models reveals a substantial gap between semantic understanding and fine-grained pixel-level localization. We further propose UltraG-Agent, which combines the semantic reasoning capabilities of a VLM with the ultrasound-specific segmentation capability of UltraSAM3. Experiments show that UltraG-Agent substantially improves both semantic prediction and pixel-level visual grounding. Our dataset and code are available at this https URL.
[CV-69] Reliability-Regulated Trajectory Optimization for Progressive COLMAP-Free 3D Gaussian Splatting
链接: https://arxiv.org/abs/2609.30865
作者: Zijian Wu,Jinliang Wang,Zidian Lin,Ying Song,Ziqian Lu,Hanjie Ma,Zhen Ye,Mingfeng Jiang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:COLMAP-free 3D Gaussian Splatting (3DGS) bypasses computationally expensive structure-from-motion (SfM) pipelines, yet progressive camera pose tracking remains fundamentally vulnerable to error compounding—early pairwise tracking inaccuracies both corrupt subsequent frame initializations and remain permanently frozen in the scene representation. Rather than relying on heavyweight external neural priors or treating progressive tracking through isolated heuristic fixes, we propose a unified reliability-regulated trajectory optimization framework for progressive COLMAP-free 3DGS. At its core, our framework establishes an intrinsic, self-supervised bidirectional cycle-consistency mechanism that systematically regulates progressive camera trajectory estimation across two complementary temporal horizons: (1) Forward Motion Propagation, where the online reliability signal adaptively gates first-order kinematic warm-starts of rigid motion into upcoming pairwise registrations, supplying informed directional search priors while safely intercepting untrusted transitions; and (2) Retrospective Trajectory Correction, where the same reliability signal dynamically weights relative-pose consistency constraints within a sliding window of neighboring camera poses. By governing both prospective state initialization and retrospective trajectory consolidation through a unified reliability regulator, our self-contained framework resolves progressive drift without external priors or offline preprocessing. Extensive evaluations on Tanks and Temples and CO3D-V2 benchmarks show that our method substantially improves camera trajectory accuracy and novel-view rendering quality, outperforming existing unposed baselines. Code is available at this https URL.
[CV-70] MDSkin-Net: Multi-Task Skin Lesion Analysis Driven by Pattern Analysis Priors and Spatial Alignment Regularization
链接: https://arxiv.org/abs/2609.30855
作者: Yijian Li,Saad Bedros,Paul Bigliardi,Mei Bigliardi Qi,Vassilios Morellas,Nikolaos Papanikolopoulos
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 13 pages 4 figures
Abstract:Reliable skin lesion segmentation and classification are central to dermoscopic computer-aided diagnosis. Existing multi-task frameworks couple the two tasks architecturally without clinical knowledge, while knowledge-injecting approaches rely on the macroscopic ABCD rule, which was not designed for dermoscopy. Dermoscopic diagnosis is grounded in Pattern Analysis, a microscopic framework structured around dermoscopic features. We propose MDSkin-Net, which incorporates cue-level Pattern Analysis priors into a hybrid CNN-Transformer architecture. At its core is a Pattern Analysis-Guided Attention Module (PAGAM) comprising three priors motivated by distinct dermoscopic cues: an improved Efficient Channel Attention (iECA), a Multi-Scale Spatial Attention (MSSA), and a Biased Asymmetry Attention (BAA). We further introduce a multi-scale spatial alignment regularization (MSAR) that uses the segmentation ground-truth mask as hierarchical soft supervision, confining the classification head to lesion-localized evidence and coupling both task pathways through a shared spatial prior. Trained exclusively on the ISIC 2017 training split without external dermoscopy data, the MDSkin-Net ensemble transfers robustly under zero-shot evaluation, reaching a Dice Similarity Coefficient (DSC) of 92.38% and a melanoma AUC of 97.84%on PH2, and a DSC of 88.92% on the ISIC 2018 Task 1 test set. On the in-domain ISIC 2017 benchmark, the ensemble attains a mean Area Under the Curve (AUC) of 91.60% across the two classification tasks (melanoma and seborrheic keratosis vs. rest), and a DSC of 84.72% for segmentation. Classification remains competitive with baselines; in-domain segmentation trails single-task specialists, yet the proposed priors and alignment regularization yield representations that generalize consistently across cohorts of different scales.
[CV-71] Aligning One-Step Generative Models with Reward-Weighted Transport Distillation
链接: https://arxiv.org/abs/2609.30840
作者: Austin Wang,Ziheng Cheng,Lexing Ying
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:One-step generators enable high-quality visual generation with a single network evaluation, but their post-training is difficult: general implicit generators provide neither tractable likelihoods nor denoising trajectories, and many rewards are non-differentiable. We introduce Reward-Weighted Transport Distillation (RWTD), a post-training method that requires only generated samples and scalar reward evaluations. Rather than aligning solely to the conventional reward-tilted reference distribution, RWTD constructs an adaptive target that mixes separately tilted current and reference distributions. The current component incorporates improvements discovered during training, while the reference component anchors the target to the pretrained generator. RWTD realizes this target through feature-space optimal transport and fixed-point regression. Theoretical analysis shows that the fixed-point distributions of RWTD interpolate between off-policy reward tilting of the reference and on-policy tilting of the current model, providing a principled approach to balancing reward adaptation with retention of prior knowledge. Empirically, RWTD substantially improves the GenEval score of the one-step SANA Sprint 1.6B backbone from 0.73 to 0.80, while separate preference alignment experiments demonstrate strong cross-reward generalization that yields balanced improvements and preservation of compositional capabilities.
[CV-72] Motion Style Slider: Endpoint-Supervised Continuous Style Control for Human Motion Diffusion
链接: https://arxiv.org/abs/2609.30795
作者: Chen-Chieh Liao,Yichen Peng,Yiyi Cai,Yûi Ono,Hiroki Hanaoka,Erwin Wu,Hideki Koike,Shuichi Kurabayashi
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Existing human motion diffusion methods provide strong motion generation quality, and recent style transfer models can inject target style cues, but fine-grained continuous control of style intensity remains underexplored. In production, style intensity is subjective across artists and directors, so the practical requirement is not a universal absolute unit, but a reliable monotonic control axis. We propose Motion Style Slider, a motion-to-motion style transfer framework for endpoint-supervised continuous control. Given a content motion and a style motion, we construct a style direction in a learned motion-style embedding space and condition diffusion generation with a scalar intensity. The training objective combines diffusion denoising with latent intensity regularization to encourage smooth and monotonic style scaling without requiring intermediate-intensity ground-truth motions. Our framework is compatible with pretrained motion diffusion backbones and supports heterogeneous style datasets, including the multi-actor style motion dataset. To test out-of-range usability, we additionally introduce a small real-capture over-reaction extension and evaluate large-intensity behavior against these unseen targets. Experiments measure controllability, interpolation/extrapolation behavior, content preservation, and motion realism, with ablations on direction construction and loss design.
[CV-73] Skip the Talk Re-Focus on Vision: Latent Reasoning for Reasoning Segmentation in Multimodal Large Language Models
链接: https://arxiv.org/abs/2609.30783
作者: Tianhang Guo,Yulin He,Wei Chen,Wenjuan Zhou,Yuhang Li,Xinbiao Gan
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Reasoning segmentation aims to interpret implicit textual queries and enable fine-grained visual perception, which is critical for applications such as human-computer interaction and embodied agents. Existing methods typically generate explicit Chain-of-Thought (CoT) by multimodal large language models (MLLMs) before localizing the target. Although intuitive, such explicit verbal reasoning introduces substantial attention interference: redundant textual tokens disrupt attention during perception-token generation and also increase the effective distance between visual tokens. To address this issue, we propose LIRSeg, which fully replaces explicit CoT with a compact set of learnable latent tokens for reasoning segmentation. LIRSeg is trained in two stages: spatial alignment grounds the latent tokens in object-relevant visual evidence, and GRPO further optimizes them with segmentation rewards. To make these compact latent tokens more informative, we introduce three complementary mechanisms from an information perspective: extreme-advantage sampling for selecting informative training signals, decoupled exploration-stability updates for learning complementary representations, and latent diversity amplification for preventing representational collapse. Extensive experiments on benchmarks demonstrate that LIRSeg consistently improves both segmentation accuracy and reasoning efficiency. Compared with the VisionReasoner baseline, LIRSeg achieves absolute gIoU improvements of 4.9% on ReasonSeg, 7.1% on MUSE, and 4.7% on MMR, while achieving a approximately 16x reduction in reasoning tokens. Code is available in supplementary materials.
[CV-74] Query-Conditioned Prototype Adaptation for Cross-Domain Few-Shot Learning: Single-Query Inference Controlled Comparisons and Failure Modes
链接: https://arxiv.org/abs/2609.30769
作者: Rushab Rasik Karania,Tomas Maul
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Cross-domain few-shot learning requires adapting a classifier to a new visual domain from very few labelled examples without target-time parameter updates. We isolate one question: under a fixed global representation, what does joint query-support adaptation contribute to prototype construction? The Within-Instance Prototypical Transformer (WIPT) implements single-query test-time prototype adaptation by jointly transforming one unlabelled query and the labelled support embeddings, then forming query-specific class means. Using a shared frozen ViT-S/16 encoder, miniImageNet source training, and CUB, EuroSAT and ISIC targets, we replicate the key comparisons across five independent training seeds. In 1-shot evaluation, WIPT improves frozen ProtoNet in every run on CUB (+0.21 percentage points) and EuroSAT (+2.07), but decreases ISIC (-0.22). In 5-shot evaluation, ProtoNet remains strongest overall, while WIPT consistently improves a capacity-matched support-only Transformer on ISIC (+0.99). Joint processing of up to five queries yields no reliable accuracy gain; in a head-only 5-shot benchmark, g = 5 reduces analytical attention-token pairs by 73% and peak allocated memory by 29% relative to g = 1, although latency is non-monotonic. Across all target/shot conditions, WIPT changes uncertain ProtoNet decisions far more than confident ones, and rescue/break decomposition accounts for the observed gains and losses. Source-shift and scorer controls further show that the benefit is not universal. Overall, WIPT provides a streaming-compatible form of test-time prototype adaptation that can improve difficult low-shot cross-domain decisions without target-time optimization.
[CV-75] mo: textbfTaming Multtextbfimodal Diffusion Transformer for Human textbfMotion Generation
链接: https://arxiv.org/abs/2609.30761
作者: Zhao Wang,Jiangtao Hu,Jack Yu,Tao Yu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Most existing human motion generation (HMG) methods use cross-attention modules to inject text semantics, but ignore the importance of bidirectional modeling between motion and text tokens, which limits text comprehension. A straightforward idea is introducing multimodal diffusion transformers (MMDiT), which have shown effective joint text–visual modeling in vision generation, into HMG. However, we find that articulated motion is temporally coherent but weakly correlated across joints, in which directly applying an MMDiT with flow matching produces poorly coordinated and jerky motion. In this work, we propose Timo, a novel kinematics-aware MMDiT framework tailored for HMG. Timo combines fully shared multimodal attention for bidirectional text–motion modeling with flow matching, geometric and rotational-kinematics supervision that compares actual rotations and their changes over time, and a two-stage curriculum progressing from broad motion learning to detailed caption alignment. Further, we construct a benchmark of 40,025 held-out clips from six public datasets spanning diverse actions, assessing six complementary dimensions under a common evaluator and scoring protocol. Our model substantially outperforms state-of-the-art methods in both quantitative and qualitative evaluations. Remarkably, Timo surpasses Kimodo on five of six dimensions, achieving a 40.8 % relative improvement in the average benchmark score. Project page: this https URL. Demo page: this https URL.
[CV-76] LLPR: Location-aware learning and physics-based reconstruction for raindrop removal from a single image
链接: https://arxiv.org/abs/2609.30758
作者: Zewei He,Xingyu Liu,Xing Luo,Guizhong Fu,Zixuan Chen,Yu Chen,Jinlei Li,Zhe-Ming Lu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Raindrops can cause occlusion and distortion in the background scenes due to their adherence to windows or camera lenses. Existing raindrop removal methods concentrate on designing sophisticated CNN or Transformer architectures to recover distorted and missing texture. In this paper, we try to integrate location information and physical model into off-the-shelf CNN or Transformer architectures to help improve their performance. Specifically, we notice that existing methods deploy a preprocessing sub-network to generate a binary or soft mask to indicate the raindrop location, which will increase the network parameters and computational complexity. In contrast, a location-aware learning branch is embedded to teach the encoder in the training phase with the capability of perceiving the position of the raindrops. Note that this location-aware learning branch can be removed during the inference process (achieving performance improvements at no cost). Furthermore, instead of directly reconstructing the raindrop-free image (i.e., background scene), we devise a physics-based reconstruction scheme to first learn the transparency matrix and the raindrop layer. The latent background layer is then reversely derived based on the physical model. By combining the above-mentioned components, we propose our location-aware learning and physics-based reconstruction (LLPR) framework for this challenging ill-posed problem. We also collect a real-world raindrop-degraded image dataset, which is challenging for single-image raindrop removal (SIRR) methods. Extensive experimental results demonstrate the effectiveness and generality of our LLPR framework, achieving superior performance against state-of-the-art SIRR methods. The code will be made available upon acceptance.
[CV-77] raining-Free Bottleneck Width Planning for Convolutional Autoencoders
链接: https://arxiv.org/abs/2609.30755
作者: Guannan Guo
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Multiscale Spectral Rate-Distortion (MS-SRD) estimates the bottleneck channels required at user-supplied spatial cuts from training images and a normalized mean-squared error (NMSE) bound, without fitting a neural network. Its covariance-tail rule is exact for shared linear block-convolutional autoencoders under squared error. A nested-scale dominance result motivates reporting the activation-parameter Pareto frontier alongside the minimum-latent candidate. At NMSE = 0.01 on thirteen grayscale datasets, its latent-size prediction has 0.84% mean absolute percentage error against nonlinear patch-autoencoder boundaries; ten predictions are exact and the remaining three differ by one channel. In a four-dataset deployable comparison, MS-SRD matches all retrospective external widths and all four selected models pass, without training a selector; a 46-fit validation grid and four Least-Volume fits each pass on two datasets. In a skip-closed U-shaped autoencoder at the same bound, five predictions are exact, nine are within one channel, and every failing prediction is one channel short. Experiments at looser bounds show progressively larger nonlinear savings.
[CV-78] From Mono to Stereo: Accelerating Binocular Gaussian Splatting via Reprojection and Selective Patching
链接: https://arxiv.org/abs/2609.30741
作者: Hongfei Zhu,Ling Zhou
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Binocular rendering requires two nearby views of the same scene and therefore repeats substantial visibility and shading work. We present a 2D Gaussian Splatting (2DGS) pipeline that fully renders a dominant-eye RGB image and an alpha-weighted depth proxy, reprojects that image to the affiliated eye, and repairs uncovered pixels. Small interior gaps are interpolated, whereas larger disoccluded regions are identified as regions of interest (ROIs) and selectively re-rendered. The depth proxy reuses the alpha-blending weights computed during dominant-eye rasterization, avoiding a separate depth-rendering pass. An adaptive ROI generator localizes the required updates using reprojected image boundaries and optional connected center-hole detection. On DTU, Tanks and Temples, and MipNeRF-360, the method reduces the measured time of a sequential two-pass binocular reference by 15.5% to 28.8% and peak GPU memory by 6% to 11%. The corresponding affiliated-eye quality degradation is at most 1.3 dB PSNR, 0.02 SSIM, and 0.02 LPIPS, representing a measurable trade-off that requires application-specific perceptual validation. These results establish a practical efficiency-quality trade-off for controlled static-scene stereo rendering and motivate future evaluation under continuous motion and on physical VR hardware.
[CV-79] Amplify What You Gaze At: Target Saliency Boosting in Text-to-Image Generation
链接: https://arxiv.org/abs/2609.30733
作者: Shengqi Dang,Zhengxi Yu,Feilin Han,Xingyu Lan,Nan Cao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Text-to-image generation has advanced in controlling what, where, and how objects appear, yet how visual attention is distributed among objects remains largely unexplored. In this paper, we introduce Target Saliency Boosting, a new task aimed at boosting the visual saliency of a specific object during text-to-image generation without requiring any visual priors. Our key insight is that visual saliency is inherently relative: boosting the saliency of a target object also depends on the global saliency distribution across all objects in the scene. Based on this insight, we propose GazeME, a lightweight framework that uses saliency-marked prompts, inserting learnable marker tokens around object descriptions to indicate which objects to visually emphasize or suppress. To learn these markers, we construct a saliency-semantics dataset that associates objects in image–prompt pairs with object-level saliency scores, and propose Saliency Prior Marker Activation (SPMA), a saliency-aware stochastic marker activation strategy that exploits relative saliency relationships for robust training. During inference, GazeME automatically inserts appropriate markers into the prompt, thereby directly enhancing the visual saliency of the target object. Extensive experiments demonstrate that GazeME effectively boosts target saliency while preserving both semantic alignment and image quality.
[CV-80] Learning Polarization Image Restoration with General Restoration Priors
链接: https://arxiv.org/abs/2609.30728
作者: Chenggong Li,Jinhao Liu,Caiyun Wu,Yidong Luo,Junchao Zhang,Degui Yang
类目: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注:
Abstract:Polarization imaging captures distinctive surface and geometric cues that benefit a wide range of vision tasks. However, real-world polarization acquisition is often affected by multiple coupled degradations, making image restoration essential for practical polarization vision. Existing methods are largely tailored to specific degradations and remain constrained by the limited scale and quality of polarization data. To address these limitations, we develop an all-in-one polarization restoration framework for diverse and composite degradations. We first study the impact of different polarization representations on restoration performance and identify the normalized Stokes representation as an effective choice for separating intensity and polarization information. Accordingly, we devise a dual-branch architecture that separates intensity and polarization modeling. To overcome the limitations of polarization-specific training, the intensity branch leverages pretrained general restoration priors and a mixture-of-experts extension for composite degradations, while its restoration knowledge is adaptively distilled into the symmetric polarization branch via a cross-domain feature transform. In addition, we establish a composite-degradation polarization benchmark to support all-in-one restoration research. Extensive experiments on public datasets and our proposed benchmark demonstrate the effectiveness of the proposed method.
[CV-81] EviDETR: Preserving Query-Relevant Temporal Evidence for Moment Retrieval and Highlight Detection ICASSP2027
链接: https://arxiv.org/abs/2609.30724
作者: Haoran Sun,Yufan Li,Qichen Zhang,Haoran Zhao,Shuqi Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 5 pages, 3 tables, 1 figure. Submitted to ICASSP 2027
Abstract:Joint video moment retrieval and highlight detection requires identifying query-relevant temporal segments while estimating clip-level saliency, yet DETR-style pipelines do not explicitly preserve query-relevant evidence throughout encoding, decoding, and cross-task prediction. We propose EviDETR, an evidence-preserving framework with three components. Semantic-aware Feature Reweighting (SFR) enhances query-relevant clip representations through saliency estimation and cross-modal interaction. A Temporal Top-2 Mixture-of-Experts (TTop2MoE) decoder performs query-adaptive refinement via sparse expert routing. MR-to-HD (MR2HD) fusion transfers span-level retrieval evidence to clip-level highlight prediction through confidence-weighted multi-scale aggregation. Using CLIP+SlowFast features, EviDETR achieves 69.29 R1@0.5, 54.77 R1@0.7, and 48.41 Avg. mAP for moment retrieval on QVHighlights, together with 41.83 HD-mAP and 68.33 HIT@1. Strong results on TACoS and Charades-STA further demonstrate cross-dataset transferability.
[CV-82] rafficImag: A Benchmark for Counterfactual Roadside Traffic Video Generation
链接: https://arxiv.org/abs/2609.30722
作者: Xiangyu Li,Tianyi Wang,Zhihao Dou,Christian Claudel,Zhaomiao Guo
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Existing roadside traffic datasets support perception, forecasting, and visual question answering, but they do not evaluate counterfactual video generation, in which a selected actor is modified and the generated future should remain consistent with road topology and unrelated traffic. We introduce TrafficImag, the first benchmark for counterfactual roadside traffic video generation. TrafficImag combines a large-scale roadside dataset (9,022 annotated images, 7,043 deduplicated video clips, and 31,145 actor-centered history-future samples) with an executable protocol that supports behavior reasoning, intervention-aware image editing, and conditional video generation. Each intervention is represented as an actor-level program describing the target actor, intended behavior, legal route, interaction order, and temporal constraints, enabling a unified evaluation interface across heterogeneous foundation models. TrafficImag evaluates four complementary validity dimensions: initial-state correctness, route and behavior validity, interaction consistency, and non-target preservation, and considers an end-to-end counterfactual successful only when all four are satisfied. Across state-of-the-art foundation models, the strongest reasoner reaches 80.4% macro F1, the complete condition interface raises end-to-end success from 23.3% to 55.0% for the best generator. Oracle studies further show that conditional video execution is the primary remaining bottleneck. TrafficImag provides a reproducible benchmark for evaluating and diagnosing counterfactual traffic video generation beyond perceptual video quality.
[CV-83] VLALight: Lightweight Vision-Language-Action Models for Emergency-Aware Traffic Signal Control
链接: https://arxiv.org/abs/2609.30709
作者: Kemou Jiang,Maonan Wang,Xingchen Zou,Jiayue Zhu,Yuhang Fu,Sicheng Wang,Xi Chen,Yirong Chen,Zhiyong Cui
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 9 pages, 7 figures
Abstract:Traffic signal control (TSC) is essential for mitigating urban congestion. Recent advances in vision-language models (VLMs) enable richer interpretation of intersection scenes, opening new opportunities for visual-context-aware TSC. However, the loose coupling and repeated information conversion between modules can lead to the loss of fine-grained visual details, while sequential inference introduces substantial latency. To address these limitations, we propose VLALight, a lightweight end-to-end vision-language-action framework that directly maps intersection observations and signal-phase information to discrete signal actions. To handle the multi-view nature of TSC, VLALight combines multiple directional camera views into a unified visual input and uses textual instructions to establish their correspondence with traffic movements and signal phases. This design enables direct action prediction with a compact 0.5 B-parameter model, without intermediate image-to-text descriptions or handcrafted traffic-state representations. Experiments show that VLALight delivers the best emergency-vehicle service of all compared methods, reducing pooled emergency waiting time by 21.1% over the cascaded VLMLight while running in real time on local hardware and generalizing to unseen intersection topologies and traffic-flow patterns.
[CV-84] Combining General and Domain-Specific Pretext Tasks for Brain MR Image Segmentation
链接: https://arxiv.org/abs/2609.30708
作者: Tasneem Nasser,Susanne Schmid,Roberto Souza,Naser El-Sheimy
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 7 figures, 5 tables
Abstract:A key challenge in medical image analysis is the scarcity of large annotated datasets for specific populations and diseases. As deep learning models rely heavily on labeled data, effective transfer learning strategies are needed to reduce the dependence on manual annotations. Self-supervised learning has emerged as a promising approach for developing foundation models by enabling the learning of transferable feature representations from large-scale unlabeled medical imaging datasets. In this study, we investigate voxel-level brain age prediction as a domain-specific self-supervised pretext task and compare it with image inpainting, a widely used non-domain-specific alternative. We further propose a multitask self-supervised pretraining framework that jointly optimizes both objectives to learn complementary neuroimaging representations. The pretrained models are evaluated on three downstream magnetic resonance image segmentation tasks: multiple sclerosis lesion segmentation, ischemic stroke lesion segmentation, and cortical brain structure segmentation. Overall, the proposed multitask pretraining framework consistently outperformed the single-task pretrained models and training from scratch across most experimental settings, demonstrating the benefit of combining domain-specific and general self-supervised learning pretext tasks for the development of generalizable neuroimaging foundation models.\ Code Availability: The source code used in this study is publicly available at this https URL
[CV-85] SAGE: Source-Anchored Guidance via Frequency Equalization for Hierarchical RGB-T Alignment and Fusion
链接: https://arxiv.org/abs/2609.30703
作者: Timing Li,Yiming Sun,Boan Tao,Xiyuan Gao,Haifang Cao,Pengfei Zhu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Spatial misregistration and cross-modal discrepancies often cause ghosting, structural blurring, and content imbalance in RGB-T fusion. Existing methods typically decouple appearance adaptation, geometric alignment, and information fusion, limiting dependency propagation across stages. We propose Source-Anchored Guidance via Frequency Equalization for Hierarchical RGB-T Alignment and Fusion (SAGE), a unified framework integrating frequency equalization, hierarchical alignment, and subband fusion. SAGE employs invertible joint encoding and source-specific low-frequency modulation to derive structural and gain guidance while preserving source information. Hierarchical frequency collaborative alignment estimates global affine geometry from low-frequency approximations and transfers geometric and contextual cues to high-frequency correlation reasoning for reliability-aware residual refinement. Guided subband fusion jointly aggregates the aligned frequency coefficients under propagated source and alignment guidance, coordinates complementary low- and high-frequency information, and reconstructs the fused image through the inverse wavelet transform. Extensive experiments on RGB-T datasets with real-world and synthetic misalignments demonstrate consistently competitive performance in alignment and fusion, validating the effectiveness of source-anchored guidance for weakly registered RGB-T images.
[CV-86] MM-VeriAgent : Learning to Use Extensive Tools to Verify Multimodal Misinformation with Reinforcement Learning
链接: https://arxiv.org/abs/2609.30698
作者: Peipei Li,Shuhan Xia,Shengyang Liu,Zekun Li,Ran He
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Real-world multimodal misinformation often involves mixed forgery sources, requiring sample-specific detection strategies. Existing tool-augmented methods rely on predefined workflows or inference-time planning, limiting adaptability or increasing inference cost. To address this issue, we introduce \textbfMM-VeriAgent, which learns to verify mixed-source multimodal misinformation with tools. We first build \textbfMM-VeriTools, a specialized toolkit for misinformation detection agents. By benchmarking various candidate models and methods on the sub-tasks required by mixed-source detection, we select the strongest for textual, visual, and cross-modal forgery analysis and encapsulate them as callable tools with a unified interface. On top of this toolkit, we train the LVLM agent with reinforcement learning to teach it how to use these tools to better solve mixed-source detection. Since many of the tools are specialized models whose online execution at every rollout severely limits RL efficiency, we further introduce \textbfTool-Execution Cache, which pre-executes candidate tool calls and reuses their cached outputs during training. This preserves multi-step rollouts while reducing online tool execution, largely improving the training this http URL on MMFakeBench demonstrate substantial accuracy gains over the base model without explicit tool search at inference time. Ablation and efficiency analyses further validate the learned tool-use policy and show that Tool-Execution Cache reduces online tool executions during training.
[CV-87] Structure-Guided Masked Autoencoders for Ultra-High Resolution Scientific Image Understanding NEURIPS2026
链接: https://arxiv.org/abs/2609.30682
作者: Enzhi Zhang,Du Wu,Rui Zhong,Cong Ma,Isaac Lyngaas,Amir Koushyar Ziabari,Xiao Wang,Peng Chen,Tao Luo,Toshio Endo,Fumiyoshi Shoji,Kento Sato,Kentaro Uesugi,Takayuki Nonoyama,Ryuji Kiyama,Masahiro Yoshida,Masaru Tezuka,Tetsuya Ishikawa,Satoshi Matsuoka,Masaharu Munetomo,Mohamed Wahib
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to NeurIPS 2026. 22 pages, 10 figures, 6 tables
Abstract:Self-supervised pre-training with Vision Transformers, including Masked Autoencoders (MAE), is difficult to apply to gigapixel scientific images. Random masking is poorly matched to the structured, multi-scale morphology of scientific data, while uniform tokenization produces prohibitively long sequences that make O(N^2) attention impractical. We propose SGMA, a structure-guided masked autoencoding framework for ultra-high-resolution scientific images. SGMA couples two components: a content-adaptive quadtree tokenizer that compresses gigapixel images into a fixed-length sequence, and a structure-conditioned masking process that biases reconstruction toward spatially informative regions. To stabilize this process across scales, we introduce Damped Accumulation (DA), which aggregates signal-dependent responses across the tree into a structure canvas used to guide masking. The resulting pre-training task preserves fine microstructure while remaining compatible with standard ViT encoders and MAE-style reconstruction. Across electron microscopy, whole-slide optical microscopy, and X-ray CT datasets, SGMA consistently outperforms MAE baselines. It achieves 95.68% Dice on the 8K x 8K x 28K SpringXCT dataset, improving over the same-architecture MAE baseline by +13.00 points, and 83.21% Dice on the 32K^2 WSI PAIP dataset, improving by +16.84 points, while providing up to a 24.8x inference speedup.
[CV-88] StarWM: Self-Supervised Trained Attention Routing for Robust World Models NEURIPS2026
链接: https://arxiv.org/abs/2609.30667
作者: Zeqiang Zhang,Fabian Wurzberger,Maximilian Otte,Daniel Schmid,Sebastian Gottwald,Arne Peter Raulf,Daniel Alexander Braun
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Accepted by NeurIPS 2026
Abstract:A robust world model must strike the balance between faithfully capturing environmental dynamics and abstracting away from irrelevant content. While reconstruction-based world models ensure faithful supervision, they misallocate representational capacity by pixel area rather than dynamics relevance for visual tasks, which can cause task-irrelevant content to dominate the learned representation. Alternatively, reconstruction-free methods avoid this bias but risk discarding possibly relevant information. We propose StarWM, which uses a cross-attention module trained on self-supervised dynamics to decide where reconstruction applies. A dual-stream decoder then restricts reconstruction to the attended regions, with stop-gradient barriers preventing interference between the two objectives. These components allows reconstruction to supervise the visual content of attended regions without contaminating the latent with non-predictive information. On DeepMind Control with dynamic video backgrounds, default (reward-free) StarWM achieves the strongest performance under random-frame distractors and substantially outperforms reconstruction-based baselines under sequential video. In addition, its reward-augmented variant matches or exceeds reconstruction-free methods on sequential video, achieving the highest overall return across all distractor regimes. Mechanistic probing confirms StarWM preserves state attributes with near-perfect fidelity through long-horizon imagination while systematically discarding distractors.
[CV-89] Conditional Predictive Sufficient Statistics for Visual Representation Learning
链接: https://arxiv.org/abs/2609.30647
作者: Yuzhou Hong
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 14 pages, 2 figures
Abstract:A useful visual representation is a statistic of the observed past that retains the latent factors shared with the future and discards patch-private noise. We formalize this requirement as a conditional predictive sufficient statistic (CPSS). Under a shared-factor model of image patches, the mutual information between the past and the next patch equals the information the past carries about the shared factor, up to a remainder that the next patch itself fails to reveal. Predicting the next patch embedding with a cosine loss is maximum likelihood for a von Mises-Fisher model of that embedding’s direction, and is therefore a tractable surrogate for the predictive information. The same population loss is also minimized by a constant embedding, so stop-gradient does not by itself select the sufficient statistic; it only blocks the symmetric gradient that implements the constant solution in one step. The regression target is a shallow embedding, which forces the network output back into that shallow range and leaves the sufficient statistic in intermediate blocks. Small causal Transformers on MNIST and CIFAR-10 are used as diagnostics, not as a leaderboard. On MNIST the future shift and the stop-gradient move probe accuracy by tens of points, and the CPSS readout peaks before the output. On CIFAR-10, with the same short budget and no augmentation, every objective lands near a linear classifier on pixels. What still matches the derivation is the geometry: the CPSS output is a worse readout than its best intermediate block, next-pixel regression does not pay that penalty, and removing the stop-gradient collapses the effective rank of the embedding even when the pretext loss looks perfect.
[CV-90] MedTokenBudget: Lesion-Preserving Token Routing for Dermoscopic Image Classification
链接: https://arxiv.org/abs/2609.30613
作者: Zhexiang Li
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 12 pages, 3 figures. Includes supplementary material
Abstract:Dermoscopy classifiers built on Vision Transformers process all image patches uniformly, although diagnostic evidence is concentrated in the lesion region. Existing token pruning methods reduce tokens using generic saliency or similarity signals, but rarely ask whether the retained subset still contains the lesion. This paper introduces MedTokenBudget, a supervised post-backbone token routing framework that learns to construct compact lesion-enriched representations when auxiliary lesion masks are available. Its Lesion-Aware Token Scoring (LATS) module fuses attention entropy, feature norm, and local feature contrast through a learned scorer, then routes the top- K patches under a target budget. LATS is trained with budget curriculum learning, diversity regularization, attention distillation, and lesion-mask supervision. The trained router is evaluated with a lesion retention rate that directly measures how much ground-truth lesion evidence survives the token budget. On ISIC 2019, mask-supervised LATS consistently outperforms Random and ToMe at headline budgets while retaining substantially more lesion patches. Code is provided for reproducibility, and complete tabulated results are included in the supplementary material.
[CV-91] MVAgent : Multi-Agent Video Generation via Consistent Condition Construction and Shot-Level Policy Optimization
链接: https://arxiv.org/abs/2609.30609
作者: Xiangyu Kong,Wenjie Zhou,Fengping Tian,Lihua Fang,Haoqin Sun,Chenyang Lyu,Longyue Wang,Weihua Luo
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 5 pages, 2 figures
Abstract:Multi-shot agentic video generation requires consistent character appearance, stable spatial layout across camera angles, and continuous character state between shots. When every shot is a separate request to a frozen generator, repeated text does not determine appearance, layout or state. We therefore recast the problem as condition construction and present MVAgent, a multi-agent pipeline whose agents collaborate through typed conditioning inputs. Because an environment image shows one viewpoint, a Spatial Grounding agent samples views from generated camera-traversal clips and anchors each shot to the view matching its framing. As generated shots drift from the plan, an Observer records how each shot ends in a continuity memory, from which a Transition agent builds character action and spatial references for the next shot. An Orchestrator composes these inputs into each request. Since a request reveals its effect only after rendering, we train it by agentic reinforcement learning with Trunk-GDPO, which compares rendered candidates at every shot rather than once per video and continues the best as the trunk. With generator and judges frozen, MVAgent attains the highest cross-shot consistency and narrative-planning quality among the compared methods on ViMax-Bench and is preferred over the strongest agentic baseline in human evaluation.
[CV-92] Action Forcing: Training World Models on Unsupervised Video by Recovering Underlying Egomotion Bases
链接: https://arxiv.org/abs/2609.30595
作者: Ashish Sundar,Tiankuo Hou,Zhong Fan,Chunbo Luo,Xiaoyang Wang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Synchronised action annotations are needed to train controllable world models and these datasets remain elusive. Existing approaches make use of instrumented platforms with calibrated sensors, costly manual annotation, or latent-action models which lack grounding. We instead turn ordinary unlabelled video into action-supervised training data by recovering (without training) a data-derived egomotion basis. We track pixel displacements across frames and exploit the recurring coherent structure induced by egomotion to obtain grounded control signals directly. Using a method as simple as principal components analysis perform this, we find that the leading components provide signed, scalable, and composable throttle–yaw controls, although the method can recover only motion axes represented in the data. To prevent a high-capacity video DiT from exploiting pixel-level supervision, an online latent critic distils a frozen decoder–tracker–PCA (Principal Components Analysis) teacher without backpropagating through the decoder or tracker. Finally we critique the use of video generation metrics to evaluate WMs and introduce an example of an alternative, reference-free evaluation method. We measure \textitcontrollability, \textitplausibility, \textitconjuring (creating objects out of thin air) and \textitgeometric integrity, revealing failures that conventional video metrics miss. We show that most baselines follow familiar action directions but struggle to reverse or remain stationary. Our model handles both while retaining compositional control and generation quality. Despite backwards actions being less than 1% of our training data, we find that the model learns to reverse, scale its response linearly, and compose throttle with steering, all simply by learning through a grounded action space.
[CV-93] QSV: Quat-Sphere-Vision for Coupled Quaternion Attention on Spherical Lattices
链接: https://arxiv.org/abs/2609.30592
作者: Nicholas Foley,Devin Marinelli,Donny Moore,Diego Enriquez,Amanda Fernandez
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages, 3 figures, 2 tables
Abstract:In standard attention, three separately learned projections decide how strongly a token attends to each neighbor ( W_Q , W_K ) and how the attended features are transformed before aggregation ( W_V ). We study Quat-Sphere-Vision (QSV), a sparse spherical vision model that replaces this projection triple with a single learned unit quaternion per token: the relative quaternion r_ij = q_i^* \otimes q_j supplies both the attention logit \operatornameRe(r_ij) and a sandwich-product feature transport x \mapsto r_ij \otimes x \otimes r_ij^* , with messages passed over sparse kNN graphs on concentric Fibonacci spheres. Ablations that change only the targeted component show the two roles to be asymmetric. Removing the transport reduces test accuracy by about four percentage points on CIFAR-10 and CIFAR-100 (single runs per CIFAR-100 variant), while replacing the learned attention weights with uniform averaging leaves it essentially unchanged. Parameter-matched controls then remove the geometry itself: standard attention on the same graph exceeds QSV (mean 87.3% vs. 85.9% ), and the same model on a flat 2D lattice reaches 91.1% , within 2.1 points of a ResNet-20 trained under the same pipeline (single run). In the coupled kernel, nearly all of the learned pairwise computation resides in the transport channel.
[CV-94] Atlases Are Already Inside: Recovering Population Templates from Pretrained Diffusion Models
链接: https://arxiv.org/abs/2609.30566
作者: Jian Shi,John Femiani,Peter Wonka
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:We present a new inference-time sampler for diffusion models that gives a pretrained model a capability it was never trained for: constructing the atlas of the population it synthesizes. The sampler converges from every random seed to the population’s central anatomy, which we call the \emphintrinsic atlas. The advantage is threefold. (1) It requires no retraining. A diffusion model that has already learned a coherent population, including the released ones, yields its atlas in a single inference pass without involving deformable registration. (2) It applies to multiple domains, such as brain MRI, chest X-ray, faces, and 3D shapes. (3) It extends to subpopulations. One age-conditioned model gives an atlas at any age in its training range, and the resulting family reproduces the CSF expansion of healthy aging. Evaluated as a registration target, the intrinsic atlas is best or second-best on every dataset against classical and learned templates, and the most central template on held-out brain MRI cohorts. Atlas construction can be reframed as a byproduct of generative modeling: a diffusion model is a learned representation of population structure, and the atlas is what it already contains.
[CV-95] he Shape of Events: Edge-Based Inductive Biases via Cross-Domain Distillation NEURIPS2026
链接: https://arxiv.org/abs/2609.30478
作者: Soshun Kihara,Shunsuke Yasuki,Masato Taki
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at NeurIPS 2026. All authors contributed equally. Code: this https URL
Abstract:Convolutional neural networks trained on ImageNet are known to exhibit a strong preference for local high-frequency texture, an inductive bias that translates into fragile robustness against distribution shifts in real-world environments. Event cameras, in contrast, record only changes in scene brightness and are therefore well suited to capturing contour information; however, due to the absence of diagnostic benchmarks in the event domain, the inductive bias that event-camera data instills in vision models has remained underexplored. In this work, we use knowledge distillation from the event domain to the RGB domain so as to exploit the rich evaluation toolkit available in the RGB domain and systematically dissect this inductive bias. Our experiments show that distillation from the event domain induces, in the RGB domain, color invariance, shape bias, and robustness to high-frequency noise. We identify the underlying mechanism as the model suppressing its dependence on high-frequency texture while acquiring a stronger dependence on edge-based object shape. This hypothesis is supported by changes in how color and spatial information are processed at the early layers, together with a spectral trade-off in which robustness to the absence of high-frequency components coexists with vulnerability to contamination of the relied-upon frequency bands and to disruption of geometric structure. We further show that this inductive bias differs from existing robustification methods and that it functions as a useful prior for diverse downstream tasks in which shape and contour information contribute alongside other cues. The code is available at this https URL .
[CV-96] VkVIO: Cross-platform GPU Acceleration for Visual-Inertial Odometry with Vulkan
链接: https://arxiv.org/abs/2609.30459
作者: Ole Hoffmann,Mateo de Mayo,Daniel Cremers
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Perception in robotics and XR fundamentally relies on good state estimation. Visual-inertial odometry (VIO) and Simultaneous Localization and Mapping (VI-SLAM) are proven ways of achieving this goal in a cost-effective and accurate manner. Efficiency in these systems allows for smaller, cooler, and lighter devices. GPU acceleration is a natural approach for reducing latency, thanks to their wide availability in platforms like embedded computers, mobile phones, and XR headsets. However, previous works in the literature have limited themselves to the use of CUDA for this task, significantly reducing deployment options to a single vendor. We instead leverage the vendor-agnostic Vulkan API, originally designed for the strict performance requirements of 3D graphics applications. In this work, we present VkVIO, the first, to the best of our knowledge, cross-platform GPU-accelerated VIO method. We provide state-of-the-art accuracy with causal estimates required for real-time operation. We deploy VkVIO on a diverse range of devices spanning a workstation, a laptop, and an extremely inexpensive single-board computer, while outperforming CUDA-based systems on the same hardware. VkVIO enables possibilities for low-latency, low-power, and low-cost VIO in robotics and XR.
[CV-97] LensDesigner: A Self-Improving Agent for Optical Lens Design
链接: https://arxiv.org/abs/2609.30450
作者: Lei Sun,Haoran Liang,Dannong Xu,Yao Gao,Yuyu Geng,Jinjin Gu,Kaiwei Wang,Danda Pani Paudel,Luc Van Gool
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Optical lens design is a complex, non-convex optimization challenge that relies heavily on human experience and intuition. Existing optimized-based automatic lens design methods struggle to navigate this vast parameter space without meticulous manual tuning. In this paper, we present LensDesigner, an autonomous agent framework that mirrors the problem-solving workflow of expert opticians. To overcome the initial cold start problem, we construct LensLib100K, an extensive optical lens library, and employ Optics-Aware Retrieval to supply physically valid structural seeds. Within an interactive physical simulation environment, the agent executes macroscopic orchestration while receiving immediate optical feedback. Furthermore, we introduce a continuous self-evolving mechanism guided by a curriculum agent. By iteratively solving design tasks with progressively increasing difficulty, the agent autonomously extracts, accumulates, and reuses design heuristics, effectively evolving its optical lens design expertise over time. At the evaluation level, we introduce LensArena, a standardized evaluation benchmark comprising 120 diverse optical design tasks, covering extreme configurations. Extensive experiments on this benchmark demonstrate that LensDesigner significantly outperforms publicly available baseline algorithms, achieving superior success rates and optimization efficiency. We hope this work sheds light on the emerging field of intelligent optics. The code will be publicly available.
[CV-98] WALT: Learning World-Model-Aligned Latent Trajectories for Autonomous Driving
链接: https://arxiv.org/abs/2609.30436
作者: Mingkai Jia,Jiaxin Guo,Zhijian Shu,Jiawei Xu,Mingxiao Li,Jintao Cheng,Ping Tan,Wei Yin
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Driving world models learn rich predictive representations of the surrounding environment from visual observations, yet accurate visual prediction does not necessarily translate into effective trajectory planning. We argue that a key bottleneck lies in the mismatch between visual world states and raw geometric trajectories, which may limit the planner’s ability to exploit action-relevant semantics encoded by the world model. To address this issue, we propose World-Model Alignment for Latent Trajectories (WALT), which learns a compact generative trajectory latent space by transferring information from a frozen pretrained driving world model without modifying the world model itself. Rather than directly generating raw waypoints, WALT maps them into compact representations through a dual-branch trajectory autoencoder and transfers semantic knowledge from the frozen visual world model into this trajectory space, encouraging the learned action representation to capture scene-level cues relevant to future motion and planning. Beyond our proposed formulation, we systematically study latent learning based on Joint-Embedding Predictive Architectures (JEPA) and feature alignment following Representation Alignment (REPA) to investigate how trajectory-only representation learning affects downstream planning. We evaluate WALT on the NAVSIM benchmarks. Relative to the raw-waypoint baseline, WALT improves PDMS from 89.4 to 89.8 on NAVSIMv1 and EPDMS from 87.3 to 87.9 on NAVSIMv2 while reducing trajectory planner FLOPs by 30.5%. These results suggest that preserving world representations while extracting action-relevant information provides an effective interface for world-model-based trajectory planning.
[CV-99] ProCAP: Probabilistic Cross-Attentive Prompt Learning for Vision-Language Models
链接: https://arxiv.org/abs/2609.30434
作者: Hiwa Azeez Abbas,Fatemeh Daneshfar,Moloud Abdar
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 20 pages, 8 figures, 9 tables
Abstract:Pre-trained vision-language models such as CLIP can recognize new categories via prompting, but they often struggle when labeled data are scarce or the test distribution shifts. Prompt learning adapts only a small set of parameters while keeping the backbone frozen, yet many existing multimodal prompt learners couple the visual and textual branches weakly and can be brittle in low-shot regimes. We propose ProCAP, a probabilistic cross-attentive prompt learning framework that improves cross-modal interaction and training stability without updating any CLIP weights: it learns both visual and textual prompt tokens and links them through stacked bidirectional multi-head cross-attention so the two branches refine each other across prompt depth. To reduce overfitting under limited supervision, we parameterize prompt tokens with Gaussian means and variances and regularize them with lightweight KL and L2 penalties, and we further add a compact symmetric InfoNCE head that aligns cross-attended image features with class-level text representations in a shared low-dimensional space. Across few-shot base-to-novel generalization on 11 datasets, cross-dataset transfer, and domain generalization on ImageNet shift benchmarks, ProCAP achieves strong aggregate base-to-novel performance and competitive transfer performance while keeping the CLIP backbone unchanged.
[CV-100] CSCWD: Cross-Scale Channel-wise Knowledge Distillation for Lightweight Tiny Object Detection on Edge Devices
链接: https://arxiv.org/abs/2609.30395
作者: Amir Zamani,Zeinab Ghasemi-Naraghi
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Real-time tiny object detection in aerial imagery is constrained by the weak spatial evidence of very small objects and the loss of high-resolution detail in lightweight detectors. This study presents Cross-Scale Channel-wise Knowledge Distillation (CSCWD), a training-time framework that transfers high-resolution spatial representations from a YOLO11m-P2 teacher to a compact YOLO11n student without altering the student’s inference architecture. Unlike conventional same-scale feature distillation, CSCWD transfers supervision from teacher P2 to student P3 after feature alignment while retaining same-scale distillation at deeper pyramid levels. Under the unified seven-sequence Drone-vs-Bird validation protocol, YOLO11n-CSCWD achieves 50.17% mean average precision at an intersection-over-union threshold of 0.5 (mAP@0.5) and 59.73% recall, improving the matched CA-YOLO11n baseline by 2.92 percentage points in mAP@0.5 and 3.55 points in recall. Cross-scale alignment further increases mAP@0.5 by 2.09 points over the corresponding same-scale channel-wise distillation configuration. In zero-shot evaluation on DUT-Anti-UAV, mAP@0.5 increases from 48.29% to 50.06% without target-domain fine-tuning. This domain was included because its challenging small targets make low-latency, computationally efficient detection particularly relevant. On Raspberry Pi 5 using NCNN-FP16 at 640x640 resolution, the 2.58-million-parameter student achieves 50.32% mAP@0.5 at 82.32 ms mean wall-clock latency, or 12.15 frames per second, while retaining essentially the same runtime and memory requirements as the matched baseline. The results support cross-scale distillation for improving tiny-target detection without increasing inference-time model complexity.
[CV-101] LiTe-GS: Oracle-Efficient Next Best View Selection for 3D Gaussian Splatting
链接: https://arxiv.org/abs/2609.30393
作者: Vivek Pandey,Amirhossein Mollaei Khass,Nader Motee
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注:
Abstract:Selecting informative camera views is critical for efficient training and adaptive refinement in 3D Gaussian Splatting, where each observation significantly influences model parameters. However, information-driven view-selection strategies can require repeated evaluations of expensive information-gain oracles as the number of candidate views increases. We propose LiTe-GS, an oracle-efficient method for next best view selection in 3D Gaussian Splatting. LiTe-GS reduces the number of information-oracle evaluations by performing randomized subset evaluation of candidate views rather than exhaustively scoring the full candidate pool. The resulting approach achieves expected O(M\log(1/\epsilon)) oracle complexity with respect to the number of candidate views M , independent of the selection cardinality K , while providing an explicit trade-off between oracle efficiency and approximation quality through \epsilon . We provide theoretical guarantees on oracle complexity and approximation performance under the proposed selection scheme. Experiments on Blender and Mip-NeRF 360 demonstrate that LiTe-GS maintains reconstruction quality comparable to Fisher-information-based baselines while substantially reducing the number of Fisher-oracle evaluations across different acquisition settings.
[CV-102] AlphaEarth distinguishes cities but compresses urban variation
链接: https://arxiv.org/abs/2609.30356
作者: Andrew Renninger
类目: Computer Vision and Pattern Recognition (cs.CV); Physics and Society (physics.soc-ph)
备注:
Abstract:Cities differ in built form, land cover and development history, complicating comparison across places and time. Satellite foundation models map Earth’s surface onto common numerical representations. Yet the tasks and targets used to shape them typically do not focus on cities: globally consistent labels for urban function do not exist, and many datasets - especially land cover and land use classifications - collapse the built environment into few classes. Here we audit the representation, focusing on AlphaEarth but with broader applicability to other Earth embeddings, by probing the geometry and geography of embeddings for 1,000 urban areas in 162 countries. We find that cities occupy a shifted but overlapping region on the hypersphere, 62.7° from the global mean direction, and continent and climate predict 24.3% of variation among the mean directions of urban centres in excluded countries. Inside cities, degrees of urbanisation carry 8.9% of the variation, and what they leave holds shared directions whose local orientation varies, not one universal axis of urbanisation. Retained variation is itself unequal: dispersion within urban centres is 14.1% greater per standard deviation of national development, even after adjusting for population, land area and continent. Further controls suggest cities in developing countries present less contrast in vegetation and texture, and dispersion follows that contrast: full adjustment for it leaves at most 6.4% of the gradient. Annually, a city’s representation moves nearly eight times more than redrawing its own pixels explains, and contracts where the 2022 loss of Sentinel-1B removed a pass direction. AlphaEarth’s representations therefore support comparison across regions, while the differences between its annual layers are not yet validated for comparison over time.
[CV-103] owards Whole-Study Screening for Congenital Heart Disease in Fetal Ultrasound Using Multiple Instance Learning
链接: https://arxiv.org/abs/2609.31376
作者: Mohamed Azzam,Ruobing Liu,Esther C. Ugwueke,Ziyang Xu,Shibiao Wan,Alex Foy,Abraham Zabih,Jason Christensen,Neil Hamill,Ling Li,Jieqiong Wang
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Congenital heart disease (CHD) is the most common birth defect, yet a large fraction of cases remain undetected on prenatal ultrasound, in part because current artificial-intelligence methods assume that the key diagnostic frames have already been isolated from a study, by a clinician or by a view classifier. We remove that assumption and address CHD screening directly at the level of the whole ultrasound study. We propose a two-stage framework that first learns transferable frame representations by self-supervised masked-autoencoder pre-training on unlabeled fetal ultrasound, then identifies cardiac frames with a disease-robust module and aggregates them with a transformer-based multiple instance learning (MIL) model that produces a case-level diagnosis from study-level labels alone. The model further returns its highest-scoring frames for clinician review, and a hierarchical head separates critical from non-critical CHD. On the internal test set of our multi-source development cohort (FUSE), the proposed cardiac-gated MIL model reaches an area under the curve (AUC) of 0.985 with a specificity of 0.990, outperforming the reproduced NATMED ensemble (AUC 0.861, specificity 0.600) and the FetalCLIP foundation model (AUC 0.867, specificity 0.710). On an independent external cohort, all models initially perform near chance, but label-free CORAL adaptation raises the proposed model from an AUC of 0.513 to 0.944, whereas whole-study and view-dependent baselines do not recover. These results indicate that whole-study MIL with disease-robust cardiac-frame identification is an accurate and deployable route to prenatal CHD screening.
[CV-104] Enhancing Photogrammetric Digital Surface Models with Pretrained Diffusion Models and Multimodal Conditioning
链接: https://arxiv.org/abs/2609.31199
作者: Antoine Lorentz,Stéphane May,Valentine Bellet,Dawa Derksen,Bastien Nespoulous
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Large-scale Digital Surface Models (DSMs) can be produced cost-effectively from satellite images via stereo-photogrammetry. However, the resulting 3D maps are often contaminated by noise, outliers, and voids. On the other hand, aerial LiDAR provides high-accuracy elevation measurements at a substantially higher cost. In this work, we study diffusion models conditioned both on photogrammetric DSMs and Pléiades imagery to refine vertically co-registered DSMs. We introduce a modified Stable Diffusion 3 architecture with a pruned text stream and a patch-wise normalization strategy, enabling stable training on LiDAR data and transfer from natural images to elevation maps. Experiments in French cities demonstrate that multimodal conditioning improves elevation accuracy, reducing Dense Urban RMSE from 6.00 to 3.45 m in the in-context cities and from 4.16 to 2.77 m in the held-out city of Bordeaux.
[CV-105] Quantum Diffusion Models for Medical Image Analysis
链接: https://arxiv.org/abs/2609.31070
作者: Francesco Aldo Venturelli,Stefano Martina,Marco Parigi,Filippo Caruso,Alba Cervera-Lierta,Miguel A. González Ballester
类目: Image and Video Processing (eess.IV); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Emerging Technologies (cs.ET); Machine Learning (cs.LG); Quantum Physics (quant-ph)
备注: 12 pages, 12 supplementary pages, 7 figures, 1 table, 12 supplementary figures
Abstract:Quantum Machine Learning is a novel field of research aimed at devising machine learning approaches exploiting principles of quantum mechanics, such as superposition, entanglement and interference. In this context, we present a scalable hybrid Quantum Diffusion Model, and evaluate its use for medical image analysis. Specifically, our method is based on a Discrete-Time Quantum Walk algorithm, executed on a real quantum device, to model the forward dynamics of the diffusion model. For the backward step of the diffusion model, we devise and evaluate a classical learning model, which is used to reversely denoise the data. In contrast with other existing attempts at applying quantum machine learning for image analysis tasks, severely limited by the size of existing quantum devices, our method allows to process real-world large size medical data. In particular, we present results on grayscale and RGB images, as well as 3D volumes of moderate sizes. We benchmark our results by reproducing an alternative classical counterpart model, based on diffusion models on discrete state spaces. By doing so, we compare the generation capabilities of both models in terms of three distinct state-of-the-art metrics in the field of image generation, showing the competitive, promising results of our approach.
[CV-106] Universal Drift Correction for Multidimensional Scanning Microscopy
链接: https://arxiv.org/abs/2609.30866
作者: Sangjoon Lee,William Millsaps,Dasol Yoon,Caitlyn Obrero,Guoliang Hu,Corrie Barnes,Cedric Lim,Andrew Barnum,Arthur R. C. McCray,Colin Ophus
类目: Materials Science (cond-mat.mtrl-sci); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:In scanning microscopy, drift causes the specimen to be sampled at positions displaced from the nominal probe positions. This displacement alters the spatial assignment of the recorded signals and biases quantitative measurements across two-dimensional imaging, channel-resolved spectroscopic mapping, and scan-position-resolved diffraction analysis. Here, we extend orthogonal-scan drift correction from 2D images to spectrum images and diffraction datasets. We demonstrate how to recover probe positions using either differently oriented multidimensional scans or structural reference images. The recovered positions are used either to resample the multidimensional data onto a regular grid or to assign each recorded signal to its corrected coordinate. Our method combines affine and non-rigid correction, requires no prior structural model, and is implemented as open-source, GPU-accelerated software that reduces processing times by two to three orders of magnitude, enabling routine and automated drift correction for quantitative multidimensional microscopy.
[CV-107] Image Reconstruction from Phase with Untrained Neural Priors
链接: https://arxiv.org/abs/2609.30659
作者: Ene Meco,Ahmet Enis Cetin
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Fourier phase encodes important spatial image structure, but recovering an image without measured spectral magnitude requires additional constraints and leaves absolute intensity ambiguous. We propose a projection-based two-stage framework that combines Fourier-phase and spatial-support constraints with an image-specific neural prior. The first stage alternates constraint enforcement with regularized neural-prior updates, while the second performs phase/support refinement alone with guaranteed convergence. We evaluate two neural-prior implementations on the same 77 microscopy images and compare them with a constraint-only baseline. After 500 final refinement passes, the best-performing variant achieves 31.41 dB pooled PSNR, 35.75 dB mean PSNR, and 0.9531 mean SSIM, improving pooled PSNR by 1.51~dB and reducing pooled MSE by 29.3% relative to the baseline. The results demonstrate the benefit of combining neural guidance with explicit constraint refinement at the evaluated iteration budget, while showing that lower phase residual alone does not guarantee greater reconstruction accuracy.
[CV-108] Adapting Personalized Speech Enhancement for Low-Latency Audio-Visual Target-Speaker Extraction
链接: https://arxiv.org/abs/2609.30631
作者: Rayhan Rashed,Senja Filipi,Ross Cutler
类目: Audio and Speech Processing (eess.AS); Computer Vision and Pattern Recognition (cs.CV); Sound (cs.SD)
备注:
Abstract:Online audio-visual target-speaker extraction aims to remove competing voices while preserving speech quality and bounding lookahead. Existing extractors are built and evaluated for separation on synthetic mixtures, leaving listening quality and meeting behavior largely untested. We introduce Audio-Visual Personalized Voice Quality Enhancement (AV-PVQE), which approaches these requirements from the other direction. We start from a personalized speech enhancement model that reconstructs a requested voice at high quality but confuses the target in 46% of two-speaker mixtures despite clean enrollment. Adding mouth features at its speaker-conditioning input and jointly fine-tuning the visual and reconstruction networks reduces this rate to 1.6%, with no future frames and 20 ms of algorithmic delay. Compared with an online autoregressive audio-visual extractor, AV-PVQE yields separation gains on two synthetic benchmarks and larger gains on recorded meetings, and keeps its advantage on excerpts with more speakers than the fine-tuning mixtures. In personalized P.835 listening tests on two meeting corpora, it improves overall quality over this extractor by 0.57 and 0.63 MOS, with similar mean rating relative to the starting model. Preservation and rejection tests show that it keeps the target intact when no competing voice is present and suppresses competing speech when the target is absent.
[CV-109] FRESHLATENT: Channel-Aware Latent Adaptation for Resource-Constrained Embodied VLM Perception
链接: https://arxiv.org/abs/2609.30629
作者: Rajat Bhattacharjya,Minwoo Kim,Arnab Sarkar,Tamoghno Das,Sing-Yao Wu,Eli Bozorgzadeh,Marco Levorato,Nikil Dutt
类目: ignal Processing (eess.SP); Computer Vision and Pattern Recognition (cs.CV); Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG); Robotics (cs.RO)
备注: Paper is currently under review. Authors’ version posted for personal use and not for redistribution
Abstract:Mission-critical UAVs increasingly rely on split vision-language model (VLM) perception under tight onboard-resource and wireless-communication constraints. However, corruption of transmitted intermediate features creates a deployment mismatch for clean-trained split interfaces, while stronger channel-aware codecs can impose substantial onboard cost. We present FreshLatent, a lightweight channel-aware latent adapter that trains a power-normalized encoder-decoder through wireless corruption while keeping the surrounding VLM frozen. We formulate deployment around a mission-conditioned perception requirement and embedded interface cost, linking channel quality and communication budget to the operating conditions under which perception remains usable. At 0 dB and the tightest communication budget, FreshLatent improves gIoU and cIoU over clean split compression by 20.79 and 20.87 points, respectively. At the most adverse evaluated SNR (0 dB), across all three communication budgets, FreshLatent recovers 63.5-69.1% of the gIoU improvement achieved by a much heavier, range-trained feature-JSCC codec. On an NVIDIA Jetson AGX Xavier in 10-W mode, FreshLatent uses 37-40x fewer encoder parameters, 7.7-9.9x lower edge-interface latency, and 8.8-10.0x lower edge-interface energy than the heavier codec. Together, these results show that lightweight channel-aware adaptation can recover a substantial fraction of the robustness of a much larger communication interface while broadening quality-valid operation under constrained wireless conditions.
人工智能
[AI-0] DeepEdu-v1: Efficient and Scalable Agent ic LLM s for Vietnamese Education
链接: https://arxiv.org/abs/2609.31568
作者: Quang Nguyen,Hieu Nguyen,Hien Hoang,Toan Pham,Cong Tran,Nam Vu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:AI tutoring could markedly improve learning outcomes for students in developing regions such as Vietnam, yet the two obvious paths both fall short. Cloud assistants such as ChatGPT route sensitive student data to foreign servers—violating data-sovereignty laws such as Vietnam’s Decree 53—and, pre-trained on Western-centric corpora, are not organized around the national textbook curriculum, so their knowledge of local content is unsystematic and frequently hallucinated. Self-hosting an open model keeps data on-premise but hits a two-fold wall: post-training quantization (AWQ, GPTQ) tames the static weight footprint, yet the dynamic KV cache and prefill latency of long tutoring contexts still cause out-of-memory failures and slow responses on consumer GPUs, while the model keeps hallucinating on region-specific material. We present DeepEdu-v1, an AI-tutoring system for Vietnamese education built on SCALE (Self-improving Context-Aware Learning Engine), a framework with two innovations. First, a long-context inference engine amortizes token selection from per-sub-chunk to per-cluster granularity; on long-context retrieval it issues x7.7 fewer retrieval calls than a state-of-the-art selective-attention baseline, cutting prefill latency (TTFT) by roughly 35% while matching or improving task accuracy. Second, a self-improving agentic layer continuously curates a verified playbook from past interactions instead of fine-tuning, a design intended to progressively reduce reliance on dominant-language priors as trustworthy local knowledge accumulates. In its deployed configuration, DeepEdu achieves a nearly x2 TTFT speedup over standard vLLM serving and lifts agentic accuracy from 70.0% to 79.5% on complex tasks, with the strongest per-track gains across financial-reasoning and interactive-agent benchmarks.
[AI-1] A Flow Matching Framework for Neural Representational Dissimilarity
链接: https://arxiv.org/abs/2609.31544
作者: Zeyuan Ye,Xue-Xin Wei
类目: Artificial Intelligence (cs.AI); Information Theory (cs.IT); Machine Learning (cs.LG)
备注:
Abstract:Neural representational dissimilarity quantifies differences between neural response distributions, and is essential for comparing neural codes across stimuli, brain areas, tasks, and models. Commonly used distance metrics involve different assumptions and are estimated with separate methods. Here, we show that a variety of distance metrics can be unified under a flow matching framework developed in deep generative models. That is, these distances arise as Jeffreys divergences under different velocity constraints. We find that flow matching has advantages for estimating distances involving complicated distributions and continuous variables. Furthermore, this framework enables the design of new distance metrics in a principled way. Together, flow matching provides a unified approach for understanding, estimating, and designing neural representational dissimilarity metrics.
[AI-2] Can You Check That? The Checkability Boundary for Local LLM Network Automation
链接: https://arxiv.org/abs/2609.31540
作者: Maleeha Masood,Momina Nofal
类目: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI)
备注: Correspondence: Maleeha Masood (maleeha2@illinois.edu) or Momina Nofal (mominanofal@hotmail.com)
Abstract:Sending every network-automation input to a third-party frontier LLM exports sensitive artifacts such as production configurations, topologies, and logs. Querying small language models (SLMs) locally avoids this egress, but SLM outputs can be error-prone for direct use. This work introduces checkability as a criterion for determining which tasks are suitable for local inference. A task is checkable when it exposes a cheap, deterministic test - an intrinsic check - that rejects outputs violating a necessary correctness condition. We instantiate this idea in Touchstone, a local-first pipeline that uses seven off-the-shelf SLMs (1-8B parameters) to generate candidates, uses task-specific intrinsic checks to reject responses, and escalates unresolved inputs to a frontier LLM. On conflict detection and intent translation tasks, Touchstone reaches 98.6% and 93.8% end-to-end accuracy while escalating only 16% and 17% of inputs, respectively. On TeleQnA, a knowledge-only control that has no task-specific intrinsic checks, Touchstone is unable to match the accuracy of the frontier baseline. Our results support a simple deployment rule: keep inference local when task semantics support precise, low-cost checks; escalate the rest.
[AI-3] Prompt Minimization: Reducing Input Redundancy Without Sacrificing Output Fidelity
链接: https://arxiv.org/abs/2609.31505
作者: Marius F. R. Juston,Kevin A. Karim,Jonathan Gao,Kevin C. Li,Rudhi Bashambu
类目: Artificial Intelligence (cs.AI)
备注: 14 pages, 13 figures
Abstract:Despite the growing capabilities of large language models (LLMs), prompt design remains largely heuristic and ad hoc. This project will explore \textitprompt minimization , the process of reducing prompts to their smallest, most information-dense form while preserving output fidelity. Practically, shorter prompts reduce computational overhead and inference latency, especially when large contexts, such as entire documents or codebases, are included unnecessarily. Further, longer prompts can damage LLM reasoning and accuracy. Theoretically, the existence of multiple prompts yielding equivalent outputs suggests a high degree of redundancy in the input space, raising fundamental questions about what information is essential to elicit specific model behaviors. We propose three variant frameworks to identify and evaluate minimal prompts and demonstrate that minimal prompts often produce outputs comparable to those of their longer counterparts. These findings suggest new directions for efficient prompt engineering and deepen our understanding of input compression in LLMs.
[AI-4] UQ-LOB: Uncertainty-Aware Limit Order Book Mid-Price Forecasting
链接: https://arxiv.org/abs/2609.31491
作者: Derrick Gilchrist Edward Manoharan,Eljas Linna,Kestutis Baltakys,Hao Dong,Juho Kanniainen
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Forecasting short-horizon mid-price movements from limit order book (LOB) data is central to algorithmic trading, yet most deep LOB forecasters are point predictors: they output a direction or a displacement, but never indicate which of their forecasts can be trusted. We introduce UQ-LOB, a lightweight, encoder-agnostic uncertainty quantification module that attaches to any pretrained LOB encoder and, in the spirit of attentive neural processes, conditions each forecast on a context set of recently completed windows whose outcomes are already realised. The UQ-regression variant outputs a calibrated Gaussian over the future tick displacement, while the UQ-classification variant outputs a categorical distribution over down/up/stationary. Both expose a scalar confidence (predicted signal-to-noise ratio or class probability) that supports selective prediction. On 5.2 billion LOB events across seven cryptocurrency assets and horizons of 5, 10 and 15 seconds, UQ-regression attains near-nominal 68% interval coverage, and restricting to the most confident 10% of predictions raises directional macro F1 by 0.11-0.15 for UQ-regression and 0.05-0.11 for UQ-classification, at every horizon. On large, economically meaningful moves, the tightest confidence tier reaches a directional F1 of 0.88 (down) and 0.83 (up) at the 5-second horizon.
[AI-5] “AI is (not) the new…”: A Diagnostic Analogy Framework for Generative AIs Cultural Impacts
链接: https://arxiv.org/abs/2609.31482
作者: Rida Qadri,Vinodkumar Prabhakaran,Remi Denton
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Generative AI is reshaping the cultural infrastructures through which knowledge is found, synthesized, and held accountable. To make sense of this shift, scholars and policymakers reach for historical analogies of technologies such as the printing press, steam power or electricity. But these comparisons are typically imprecise about which property of the technology carries the comparison, and imprecise analogies produce imprecise governance by designing interventions against the wrong property of the system. This paper offers a diagnostic framework for analyzing how generative AI can transform epistemic and cultural practice. This paper offers a diagnostic framework for analyzing how generative AI can transform epistemic and cultural practice. We decompose each intervention into three coordinates: the epistemic site at which a technology acts, the governing logic by which it organizes its object, and the technical mechanism through which the logic is instantiated. This framework allows us to distinguish between structural cultural consequences, which follow from the mechanism itself, from contingent ones, which remain open to design and institutional choice. Applying the framework to information discovery and knowledge synthesis, we show how the shift from indexicality to inference and from editorial authority to statistical consensus produces specific, traceable cultural effects and reveals governance levers that gestalt analogy obscures.
[AI-6] Game Arena: Strategic LLM Evaluation in Competitive Environments WWW
链接: https://arxiv.org/abs/2609.31473
作者: Bovard Doerschuk-Tiberi,Yao Yan,Justin Chiu,Hann Wang,Timothy Chung,Martyna Plomecka,John Schultz,Jon Lipovetz,Clayton Drazner,Yuchen Zhuang,Jaimie Hwang,Nate Keating,Riley Jones,Andrew Lee,Oran Kelly,Ian Gemp,Michael Aaron,Laurel Prince,Kate Larson,Jeff Moser,Harrison Jobe,Chad Woodford,Siqi Liu,Andrew Wang,Bo Chang,Christopher D’Mello,Diane Chaleff,Addison Howard,Johnny Yip,Chuck Sugnet,Antonio Gulli,Meghan O’Connell,Will Cukierski,Nenad Tomasev,Dima Yeroshenko,Kinjal Parekh,Roxanne Daniel,Marc Lanctot,Domino Weir,Elsa Dong,Daniel Hennes,Melissa Nalubwama,Robert Fraser,Ryan Trostle,Jun Peng,Tom Mason,Lloyd Hightower,Chiamaka Chukwuka,Yuexiang Zhai,Phoebe Kirk,Yi Su,Yuting Han,Jie Ren,Chris Prichard,Sahand Sharifzadeh,Karim Hakimzadeh,DJ Sterling,Meg Risdal,Kate Olszewska,Ya Xu,Orhan Firat,Minmin Chen
类目: Artificial Intelligence (cs.AI)
备注: 31 pages, 15 figures. Technical report. Project page: this https URL
Abstract:We introduce Kaggle Game Arena, an open and ever-expanding platform to evaluate large language models (LLMs) through competitive games. Different from static benchmarks, game arena enables models to play head-to-head matchups in structured environments where the gameplay strength naturally increases as models evolve, preventing performance saturation. This technical report details the infrastructure behind Game Arena and describes the three pilot game environments: Chess, Poker, and Werewolf. These environments span perfect information, imperfect information, and multiplayer game settings, enabling a systematic study of models’ strategic planning, adaptation, and robustness under uncertainty. For each game, we provide a detailed description of the environment, evaluation metrics, and results from running full competitions across models. Through robust infrastructure and large-scale ground-truth based evaluation, Game Arena ensures reproducibility, transparency and generalizability to new games and variants over time.
[AI-7] Segment-Level Agent ic Topic Modeling for Improved Data Exploration and Resource Efficiency
链接: https://arxiv.org/abs/2609.31460
作者: Myeongjun Erik Jang,Antonios Georgiadis,Sae Young Moon,Fran Silavong
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Topic modeling is an effective technique for discovering hidden themes within documents and is widely used in text mining and data analysis across a variety of industry sectors. Recently, large language model (LLM)-based topic models have been emerged that prompt LLMs to generate topics then assign the topics to documents, producing more natural and human-readable topics than conventional topic modeling algorithms. However, the nature of topic assignment process causes certain drawbacks, such as the incapability to produce topic distributions over a document, too broad or narrow topics, and high resource consumption, which increases with the number and length of of documents being assigned topics. These issues are particularly critical for industrial applications, which require high-quality, in-depth analysis and the processing of large volumes of documents. In this context, this paper introduces a framework called SeLATM, which addresses these concerns by employing segment-level topic generation and topic refinement through agentic feedback loops. Experimental results on various datasets demonstrate that SeLATM significantly reduces the LLM resources compared to methods based on topic assignment process, while maintaining superior performance.
[AI-8] Compress What You See Not What You Say: Anchored Context Distillation for Latent-Observation Software Engineering Agents
链接: https://arxiv.org/abs/2609.31430
作者: Zhensheng Zou(Peking University),Guoqing Wang(Peking University),Dan Hao(Peking University)
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Tool observations dominate the context of software-engineering agents, making long interaction histories costly to maintain. Existing context compression methods can discard information needed by later actions, while adapting agents to soft-token representations can compromise their original behavior. To reduce context while preserving action-critical information and agent behavior, we combine Latent Observations, Hard Actions (LOHA), a context layout that separates compressed history from text needed for exact reference, with Anchored Context Distillation (ACD), a training method that enables latent reading while constraining behavioral drift. LOHA compresses older tool observations into soft tokens while retaining the agent’s own turns and the last K observations in text, providing compact access to historical information and exact access to recent content. To enable the agent to use this representation, ACD distills the base model’s full-text predictions into the latent view while anchoring its behavior on plain-text inputs to the same base model. On SWE-bench Verified, K=3 reduces context per call by 43% for Qwen3-4B and 57% for SWE-Master-4B-RL, with resolve rates of 12.1% and 21.8% versus 14.5% and 27.5% for their uncompressed bases. A single-run recency sweep reaches 14.4% and 23.0% at K=8, with larger windows generally favoring task performance over compression. Under a 32K-token limit, Qwen3 with K=3 resolves 21.1% of a 199-instance subset versus 11.1% for the same adapted agent using full text. In concurrent single-GPU serving, it achieves 1.9 times that full-text agent’s instance throughput.
[AI-9] ActKV: Efficient LLM Agents through Action-Guided KV Cache Management
链接: https://arxiv.org/abs/2609.31395
作者: Zihan Wang,Cheng Tang,Lei Gong,Chao Wang,Wenqi Lou,Teng Wang,Xuehai Zhou
类目: Operating Systems (cs.OS); Artificial Intelligence (cs.AI)
备注:
Abstract:Agentic LLM inference accumulates long KV caches across iterative observation-reasoning-action loops, imposing substantial memory overhead and limiting serving throughput. Existing compression methods emphasize overall output quality, overlooking the asymmetric importance of actions in driving task progress. Our key idea is to establish a compression criterion that values KV entries by their contribution to action generation and prioritizes action quality. However, iterative execution, dynamic memory demands, and scattered action-critical entries pose challenges to eviction policies, budget allocation, and paged memory integration. To this end, we propose ActKV, the first KV cache compression framework tailored for agentic LLM inference. (i) Action-oriented KV cache eviction exploits stable action access patterns to retain entries critical to future actions, supporting reliable task progress under compression. (ii) Confidence-driven adaptive budget allocation uses LLM’s intrinsic confidence to adapt the budget to evolving action-critical memory demands. (iii) Page-aware compression management standardizes compression into three primitives with customized kernels, realizing practical throughput gains. On long-trace tasks, ActKV retains an average of 98.53% of FullKV’s accuracy with only 25.98% of its peak KV cache memory. It also achieves 3.97 times and 3.58 times FullKV’s token and task throughput, delivering state-of-the-art performance.
[AI-10] Completed Pairs Hide Capped Failures: A ReVerPi Case Study of Selective Context Projection
链接: https://arxiv.org/abs/2609.31381
作者: Guangzhe Zhang
类目: Artificial Intelligence (cs.AI)
备注: 15 pages, 12 tables, 5 figures. Project code: this https URL . Source archive includes anc/ analysis data and reproduction scripts
Abstract:Context projection replaces older tool observations with compact, addressable excerpts, reducing repeated input while potentially adding evidence-retrieval turns. We study this trade-off in ReVerPi, a Pi extension with archived observations and matched full/projected continuations. In an 86-run source-reading campaign with 641 model requests, the 15 completed pairs show identical success: 12/15 per arm. Twelve further boundary runs stop, with the runner suppressing the companion whenever the first arm fails to complete. Restoring all 27 boundary runs bounds projected-minus-full success between - 9 and +1 tasks. One omitted, selector-chosen projected continuation successfully retrieves archive text yet exhausts twelve requests; its full counterpart answers in three. The eleven jointly correct pairs form a fully observed success stratum within this recorded frame: projection reduces aggregate logical tokens by 25%, while increasing the median pair’s tokens by 29% and total suffix requests from 35 to 55. Separating fitting from evaluation changes the selector’s apparent tie: outside its four fitting pairs, it incurs one extra failure and 8.6% more logical tokens over thirteen comparable runs. This methodological case study connects stopping rules, known bounded failures, unexecuted companions, and resource aggregation. Its findings concern the recorded campaign, rather than population noninferiority or superiority over unrestricted Pi. Evaluations should retain every intervention boundary, execute both allocated arms independently of the first arm’s completion, and report completion alongside interaction and token expenditure.
[AI-11] Programs-of-Layers in LLM s through the Lens of Cortical Areas
链接: https://arxiv.org/abs/2609.31360
作者: Justus Westerhoff,Stephan Olbrich,Hatem Oraby,Matthew Evan Larkum,Felix Alexander Gers
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Inference in LLMs is conventionally a fixed-depth, fixed-order forward pass through every layer, regardless of how difficult the input is. The human brain does not work this way: using the thalamus as a central hub, it routes information flexibly to all regions of the cortex according to demand. Li et al. (2026) recently showed, with a system they call program-of-layers (PoLar), that transformers can be given an analogous flexibility if their layers are treated as a library of functions rather than a fixed sequence. Performance improves over the standard forward pass when each input is dynamically routed through an adaptive sequence of skipped or repeated contiguous layer blocks. We reconstructed PoLar’s diagnostic MCTS in more detail than the original paper and applied it across 5 models. We reproduced several of PoLar’s findings: skipping outperformed the standard pass, repeating outperformed skipping, and combining both outperformed either alone. Shorter programs sufficed for easier questions, while harder questions required more layer repeats. However, we failed to replicate the main claim regarding their learned router for single-shot inference: its top-ranked prediction consistently collapsed back to the standard pass, even though its top-k predicted programs, taken together, did show a real accuracy gain. Beyond reproduction, we find that a small number of generic programs are enough to solve most of the questions. We also provide a much deeper analysis of these programs’ structure and robustness: for example, we found that programs that correct errors are highly brittle: undoing even a single edit inside a program typically breaks the correction. Connecting this to the brain’s routing mechanisms, PoLar mirrors principles of thalamo-cortical coordination between cortical-area-like transformer layers. We publicly release the code at this https URL
[AI-12] A Safety-Bounded SDC-to-MCP Gateway for Medical AI Agents
链接: https://arxiv.org/abs/2609.31358
作者: Bennet Gerlach,Stefan Fischer
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Artificial Intelligence (cs.AI)
备注: 18 pages. Code: this https URL . Software and evaluation artifacts: this https URL
Abstract:The Model Context Protocol (MCP) provides a common interface through which AI applications discover and use external resources and tools. It allows language-model agents to ground their reasoning in current system state and interact with heterogeneous services. In medical environments, however, exposing device state and action affordances requires deterministic constraints on possible effects. We present an IEEE 11073 Service-Oriented Device Connectivity (SDC)-to-MCP gateway that exposes metrics, alarms, context references, and semantic metadata as read-only resources, while representing selected action affordances as policy-validated dry-run tools. The term safety-bounded denotes a narrow no-execution property: agent-facing requests dispatch no SDC device operation. A Python prototype supports simulated fault and lifecycle experiments, a software-reference protocol path spanning independent Java and Python implementations, deterministic baselines, representation ablations, and multi-model agent evaluation. The results show semantically explicit resource exposure, visible rejection of invalid or outdated state, and preservation of the no-execution boundary across resource, proposal, and authorization paths. Explicit semantic metadata improved conformity to required metric identifiers in structured alarm outputs relative to a generic representation, while retained structured-output failures reveal a distinction between plausible narrative answers and task-compliant machine-readable results.
[AI-13] Mutable Transcripts: Mitigating Context Pollution through Editable Conversation State NEURIPS2026
链接: https://arxiv.org/abs/2609.31354
作者: Dan Barry,Andrew Hines
类目: Artificial Intelligence (cs.AI)
备注: Accepted at NeurIPS 2026
Abstract:Contemporary large language model (LLM) chat systems treat conversation history as an immutable sequence of turns that defines the model’s working context. However, user intent in real interactions is not static: it evolves through correction, refinement, and shifting constraints. This mismatch between dynamic intent and static transcripts can result in context pollution, where outdated or irrelevant information persists and continues to influence subsequent responses. We introduce mutable transcripts, a new interaction paradigm that enables users to revise prior turns through natural language edit requests, allowing the conversation history itself to be updated rather than appended. This reframes the transcript from a passive record into an editable representation of conversational state. We present a working prototype that integrates transcript-level revision into a standard chat interface and evaluate its feasibility through a controlled user study (n=17) and an illustrative transcript analysis of representative interaction scenarios. Participants significantly preferred mutable transcripts over standard chat across measures of clarity, confidence, and ease of use, with reduced intent to restart conversations. Transcript analysis of representative user study conversations shows that mutable transcripts can reduce conversation length and eliminate obsolete retained context. These findings provide initial evidence that user-driven revision of conversational history can improve interaction quality and help maintain a more current representation of user intent. The source code and prototype can be accessed at this https URL
[AI-14] Agent Xploit: Autonomous Repository-to-Runtime Red-Teaming for AI Agents
链接: https://arxiv.org/abs/2609.31318
作者: Weida Liang,Shi Qiu,Zhun Wang,Simon Sure,Xiaoyuan Liu,Tianneng Shi,Zhaorun Chen,Wenbo Guo,Dawn Song
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: 20 pages, 2 figures
Abstract:AI agents combine language models with external data and tools that can modify files, call APIs, or execute code. Security failures can arise when adversarial content changes an agent’s tool use or when the surrounding software contains vulnerabilities such as path traversal or command injection. We study authorized white-box pre-deployment auditing, where the auditor has access to the target repository and a controlled runtime, but successful attacks must still act through the task-defined attacker interface and be confirmed by an external verifier. We present AgentXploit, a two-role auditing system that separates repository-level attack-path discovery from runtime exploitation. The Analyzer Agent traces attacker-controlled inputs to sensitive operations and records code-supported candidate attack paths; the Exploiter Agent turns these paths into concrete attacks and revises them using runtime feedback. We also introduce AgentXploit-Bench, containing 72 reproducible vulnerabilities across 12 open-source AI-agent systems and frameworks. Across three runs, AgentXploit reaches 59.3% end-to-end success, compared with 38.4% for Codex. Under a token-budget-matched comparison, Codex reaches 46.3%. On AgentDojo, where injection points are provided, the Exploiter Agent reaches 79.2% attack success versus 52.7% for AgentVigil. These results highlight repository discovery and runtime exploitation as distinct challenges in end-to-end agent security auditing.
[AI-15] owards VLA-Dreamer: Refining VLA Behavior Using World Models
链接: https://arxiv.org/abs/2609.31313
作者: Parsa Mastouri Kashani,Jan-Gerrit Habekost,Stefan Wermter
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Vision-Language-Action models (VLAs), while showing strong potential for robot control, require massive amounts of high-quality imitation learning data. Moreover, the absence of an explicit world model casts further doubt on their control capabilities. In this concept paper, we propose a novel architecture that addresses sample efficiency in VLAs by training a predictive world model on the embedding space of the VLA’s vision encoder. We hypothesize that these embeddings are action-relevant and usable for future prediction. To this end, we propose using the suggested architecture to investigate how well these embeddings predict the future based on actions, as the inability to do so would mark a key limitation of VLA architectures: the lack of a non-lossy implicit world model to simulate real-world dynamics. The proposed architecture differs from the standard world model dynamics as the loss comes from the embedding space rather than the pixel space, similar to joint embedding predictive architectures. Furthermore, the trained world model can be utilized for short-term planning tasks by sampling VLA actions given goal images. We intend to examine the richness of vision embeddings in VLAs and reduce their high data requirements through a world model that can also generate plans during inference.
[AI-16] Beyond Approved Actions: Runtime Validation of Persistent Outcomes in Agent Workflows
链接: https://arxiv.org/abs/2609.31301
作者: Haoran Zhang,Hengtong Zhang,Zhiyu Liang,Yu Yan,Decheng Zuo,Hongzhi Wang
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: 22 pages including 7 pages of supplementary material. Submitted to IEEE Transactions on Software Engineering
Abstract:Large language model agents increasingly act on software systems, no longer merely generating text but also changing databases and online services. However, an approved database update may succeed yet leave an unapproved notification because execution can produce persistent effects beyond the requested change. Current safeguards can approve an action or record its aftermath, but without checking the persistent result before continuation, an unapproved outcome can be accepted as success and propagated to later steps. We present EffectMatch, a runtime that collects persistent changes within a controlled execution boundary and compares them with what the application approved for the current state and execution. The comparison governs commit and dependent execution. In comparative evaluation on 206 public business tasks, EffectMatch preserved all clean executions and prevented all tested incorrect commits. Six 20-run ablations exposed the failure caused by each removed mechanism, while 80 task-topology cases preserved truthful handoffs and blocked invalid continuation. Together, these results show that EffectMatch blocks the silent acceptance and downstream propagation of persistent outcomes inconsistent with application approval.
[AI-17] Softmax Reparameterization for Output-Head Quantization
链接: https://arxiv.org/abs/2609.31291
作者: Asim Kadav,Christian Flores,Chirag Arora,Varun Kotte,Hongbo Zheng,Lan Yan,Priya Shanmugasundaram,Tracy Holloway King
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 33 pages, including appendix
Abstract:Large vocabularies make output heads a substantial inference cost in small language models. We propose softmax reparameterization, a post-training method that selects a functionally equivalent output head before quantization. The method subtracts a scalar multiple of the vocabulary-row mean from every output row and selects the coefficient by validation KL separately for RTN, activation-weighted MSE, and full-Hessian GPTQ. This one-dimensional search includes the original head and fixed mean-centering, preserves the full-precision softmax distribution, and leaves the trained decoder unchanged; a rank-one correction handles nonlinear logit paths such as soft-capping. Across seven heads, W4 gains concentrate where baseline quantization substantially distorts predictions: on Phi-4-mini, AW-MSE KL falls from 0.936 to 0.256. The gains survive stronger GPTQ calibration and remain complementary to exact per-channel scaling and affine quantization. Across four heads and three W4 quantizers, frozen WikiText-selected coefficients also transfer to C4 and OpenWebMath, outperforming mean-centering in all 18 comparisons where the frozen coefficient differs from 1 and matching it in the remaining six. At W2, used as a compression stress test, benefits broaden across nearly the full model–quantizer matrix. Matched residual analysis shows that improved fidelity can accompany greater logit reconstruction error while reducing the residual’s Fisher-weighted cost. For shift-compatible heads, reparameterization adds no inference operation and preserves packed W4 execution: with the decoder held in BF16, quantizing the Phi output head reduces batch-one generation latency by 10.8% relative to the BF16-head baseline.
[AI-18] G2MAF: Test-Time Gradient Guidance for Multi-Agent Flow Policies
链接: https://arxiv.org/abs/2609.31286
作者: Guowei Zou,Haitao Wang,Guoxin Wang,Zhiquan Chen,Beiwen Zhang,Guojie Wang,Hejun Wu
类目: Artificial Intelligence (cs.AI)
备注: 24 pages, including appendices. Project page: this https URL
Abstract:Offline multi-agent reinforcement learning (MARL) learns cooperative policies from fixed datasets without further environment interaction and a learned policy is frozen at deployment. Such a frozen policy typically proposes a single joint action and executes it directly at deployment time. However, this one-shot deployment often commits to a suboptimal proposal, even when better nearby alternatives remain consistent with the behavior data. To address this issue, we propose Gradient Guided Multi Agent Flow (G2MAF), a refinement framework for optimizing joint policies at test-time. G2MAF applies one globally normalized, projected critic gradient to guide and coordinate all agents’ corrections while keeping the action both feasible and close to the frozen policy proposal. Across 24 MPE and SMAC settings, its canonical variant improves 20 frozen settings, with mean relative gains of 9.2% on MPE and 8.9% on SMAC, with model inference latency increased by about 6% only.
[AI-19] Resource-Optimized and Energy-Aware Agent ic AI Framework Anchored on Blockchain for Secure Software Supply Chains
链接: https://arxiv.org/abs/2609.31282
作者: Toqeer Ali Syed,Asadullah Abdullah Khan
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: Accepted for publication in the International Journal of Energy, Environment, and Economics. 27 pages, 8 figures, 2 tables
Abstract:This paper proposes a blockchain-backed agentic security framework designed to safeguard the complete software development lifecycle (SDLC) while also securing the agentic AI components responsible for monitoring it. The framework coordinates a set of specialised security agents, covering source integrity, dependency and SBOM analysis, CI configura tion auditing, artifact verification, and runtime policy evaluation, each supported by a large language model (LLM) that interprets artefacts, reasons over tool outputs, and produces structured security reports. To ensure agent trustworthiness, every agent generates a cryptographically signed attestation that is recorded in a permissioned blockchain via smart contracts, including an agent registry, an immutable attestation log, and an enforceable release-policy module. Communication among agents and with blockchain nodes is secured using a consortium-operated certificate authority, ensuring authenticated and tamper-resistant interactions. A detailed use-case and sequence flow demonstrate how a source code security agent performs analysis, anchors its attestation on-chain, and triggers a verifiable allow/block deployment decision. The proposed framework of fers decentralised integrity transparent provenance, uninterrupted security assurance and a generalisable architecture to incorporate the agentic AI into the modern software supply chain security.
[AI-20] MA-WAM: Multi-Agent World-Action Model for Test-Time Planning
链接: https://arxiv.org/abs/2609.31281
作者: Guowei Zou,Haitao Wang,Guoxin Wang,Beiwen Zhang,Zhiquan Chen,Guojie Wang,Hejun Wu
类目: Artificial Intelligence (cs.AI)
备注: 40 pages, including appendices. Project page: this https URL
Abstract:Multi-agent cooperative tasks require different agents to execute a joint action simultaneously, and each agent’s action affects both the observations and responses of the other agents. Hence, a world model is needed to predict the team return resulting from the joint actions of all agents. A naive extension directly applies a single-agent world model to each agent’s action when predicting the team return step by step. However, such an extension fails to capture the dependencies among the simultaneous actions of multiple agents. We propose Multi-Agent World-Action Model (MA-WAM), a test-time planning framework that enables a frozen multi-agent flow policy to evaluate futures of candidate joint actions. To our knowledge, MA-WAM is the first test-time world-model planner for multi-agent flow policies. MA-WAM predicts the consequences of each joint action according to cross-agent dependencies and enables efficient candidate scoring. Across 30 offline multi-agent reinforcement learning (MARL) settings on MAMuJoCo, SMAC, and MPE, MA-WAM achieves mean relative gains of 22.0% over direct execution and 25.6% over uniform action selection. Under the standard evaluation protocol on an A100 GPU, MA-WAM adds 12.1 ms, accounting for 2.5% of the measured generation-and-scoring time.
[AI-21] Purin: A Biology-inspired Mechanism for Artificial Neural Networks
链接: https://arxiv.org/abs/2609.31235
作者: Zishu Liu,Chunbo Luo,Christos Grecos
类目: Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE)
备注: 8 pages, 2 figures, 8 tables
Abstract:Artificial neural networks (ANNs) usually represent neural transmission with fixed trainable weights during a training batch, which omits short-term changes in synaptic efficacy. In addition, the discrete time-step simulation requires additional temporal processing that many conventional ANN architectures do not use. To overcome these challenges, we propose Purin, a biology-inspired and ANN-compatible mechanism, that introduces synaptic efficacy modulation into conventional convolutional neural networks. Purin uses a time-interval-based abstraction for neural activities, which allows Purin to introduce short- and long-term synaptic efficacy changes without using discrete time-steps. Purin introduces a bounded factor to represent temporary synaptic efficacy changes, together with two weight matrices that represent input-side and output-side efficacy. The weight matrices are updated by backpropagation and interpreted as the long-term synaptic efficacy changes. Experimental results show that after removing the confounding factors in the AlexNet, VGG11, and GoogLeNet architectures, Purin improves the classification accuracies in all three models across the evaluated datasets.
[AI-22] Acoustic-to-Text KV Compression for Full-Duplex Speech Models
链接: https://arxiv.org/abs/2609.31224
作者: Yejin Lee,Seungbeom Kim,Yongha Lee,Kyuhong Shim
类目: ound (cs.SD); Artificial Intelligence (cs.AI); Audio and Speech Processing (eess.AS)
备注:
Abstract:Full-duplex speech language models continuously accumulate acoustic key-value (KV) states, making long-running interactions memory-intensive. During listening, the model can finish processing an audio unit before the next arrives; we term the remaining interval listening-time slack. We propose acoustic-to-text KV compression, which introduces a transcription side channel to convert incoming speech into compact textual memory within this interval. When the cache exceeds a target budget during inference, older acoustic states are evicted while transcripts and recent acoustic context remain. We train the side channel with LoRA using cross-entropy on transcription segments. To preserve listening and speaking behavior, we apply knowledge distillation to the original model’s token-level output distributions at native prediction positions. On ten-minute LongSpeech sessions, our MiniCPM-o 4.5 implementation reduces peak streaming KV-cache size by 64.6% compared with the same model without eviction. The proposed method also improves transcription, temporal question answering, and summarization over the baseline. Full-Duplex-Bench evaluations further show comparable pause-handling, turn-taking, and interruption performance.
[AI-23] DIAL: Position-Debiased LLM Judges with Adaptive Human Preference Calibration
链接: https://arxiv.org/abs/2609.31215
作者: Zesheng Cai,Yingqi Fan,Sichang Chen,Jin-Hong Du
类目: Artificial Intelligence (cs.AI); Applications (stat.AP); Methodology (stat.ME); Machine Learning (stat.ML)
备注:
Abstract:Large language models (LLMs) as a judge enable scalable evaluation, but their judgments can be sensitive to response order and, even after removing such position effects, can still diverge systematically from human this http URL introduce DIAL, a unified framework that combines abundant LLM comparisons with limited human comparisons to separate judge-specific position effects, learn shared structure in position-debiased LLM preferences, and adaptively calibrate that structure toward the human preference target. Theoretically, we study three aspects of DIAL: (i) identification of latent LLM preferences, position effects, and human calibration; (ii) adaptive estimation that balances LLM anchoring against limited human evidence; and (iii) fixed-weight uncertainty quantification for the calibrated human preference. Empirically, we evaluate position debiasing and human alignment separately in controlled simulations and on three human-preference benchmarks, showing that DIAL remains robust to unbalanced response order, achieves strong human-aligned rankings with limited labels, and adapts toward human evidence when LLM information is imperfect. Our real-data study collects over 410K judgments from 21 LLM judges in both display orders, providing a resource for future studies of LLM-judge bias, heterogeneity, and human alignment.
[AI-24] Which Influence Are We Estimating? The Role of Counterfactual Specifications in Data Attribution
链接: https://arxiv.org/abs/2609.31214
作者: Zhe Li,Wei Zhao,Peixin Zhang,Jun Sun
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 23 pages, 7 figures
Abstract:Estimating the influence of training examples on model behavior is essential for data debugging, valuation, and attribution. Existing influence estimators often produce incompatible rankings, which are commonly ascribed to approximation error. We argue that a more fundamental source of disagreement is specification mismatch: influence depends on the behavior being attributed, the intervention applied to each training example, and the counterfactual training process that maps the intervention to a model response. These choices are especially important when the target behavior requires a tractable surrogate, such as query loss, a logit, or a margin. We formalize influence as a counterfactual estimand, distinguish specification mismatch across estimands from approximation error in estimating a fixed estimand, and organize representative estimators by their implied specifications. We further derive a local decomposition that exposes how behavior signals, training signals, and counterfactual parameter responses interact. Controlled experiments show that exact estimands under different specifications can induce different rankings, whereas approximation error grows as perturbations move farther from their linearization points. Experiments on noisy label detection and LLM attribution show that specification choices significantly affect attribution quality, especially for the choice of behavior surrogate. Behavior-aligned specifications can identify target-specific training examples obscured by default loss-based or similarity-based specifications. These results establish specification analysis as a necessary first step for interpreting and comparing data influence estimators.
[AI-25] Samples Sources Space: Decomposing Data Scale in Spatially Structured Representation Learning of Human Brain Microarchitecture
链接: https://arxiv.org/abs/2609.31201
作者: Christian Schiffer,Mathis Bode,Thomas Lippert,Katrin Amunts,Timo Dickscheid
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Scaling studies typically represent training data by a single count of samples. For hierarchically and spatially structured data, however, the same number of samples can be drawn from few or many sources and distributed differently across the underlying domain. We therefore study data scaling as an allocation problem, separating unique sample count, source diversity, and spatial coverage. We study this decomposition in microscopic whole-brain histology, where a source is an individual brain, and a sample is an image patch at a specific spatial location. Across 93 controlled pretraining runs of a contrastive model that uses spatial proximity for supervision, we vary data allocation, compute, and model capacity over 11.6 million spatially anchored image patches from 21 human brains. Performance improves with more unique samples, broader spatial coverage, additional compute, and larger model capacity. At fixed sample count, distributing samples across one to 18 subjects produces no detectable improvement, even though representations generalize substantially better to subjects encountered during pretraining. Inter-subject variation therefore strongly affects generalization, but additional subjects provide no benefit when a fixed sample budget is distributed across more sources. These results establish sample count, source diversity, and spatial coverage as distinct axes of data scaling in spatially structured representation learning.
[AI-26] Rethinking Data Quality for AI-Driven Systems: Evidence from Practitioner Interviews
链接: https://arxiv.org/abs/2609.31191
作者: Hariharan Gopinath,Jan Bosch,Helena Holmström Olsson
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: This is a preprint version and the final version will appear in the proceedings of PROFES 2026
Abstract:Data quality research has usually treated data as an input that is stored, processed, and validated. In AI-driven software-intensive systems, data also shapes model behavior, evaluation, and lawful use. Empirical evidence remains limited on how practitioners define, assess, and manage quality under these conditions. We interviewed 16 practitioners from nine organizations and analyzed the transcripts using reflexive thematic analysis and developed six themes from participants’ accounts. In AI systems, traceability shifted from modular debugging to attributing model behavior, while using models as quality assessors introduced circularity. Agent context and memory became data objects, and synthetic and pseudo-labeled data made authenticity a quality concern. In foundation-model development, lawfulness became a gate for training data, while representativeness was judged through coverage of situations in which the system must behave safely. Prior ML research examines many of these problems separately. Our study provides a practitioner-grounded account of how they are encountered together as an engineering and organizational concern. We also interpret five recurring conditions as helping explain how the themes relate to reduced trust in data and AI outcomes. We synthesize these findings through lifecycle assurance: a conceptual framing focused on producing evidence that data can support a specific AI claim when its influence may be embedded in model behavior, model-based judgments, or agent actions.
[AI-27] Evolutionary Safety of Recursive Self-Improving AI: Taxonomy Risk Discovery and Evaluation
链接: https://arxiv.org/abs/2609.31186
作者: Chang Gong,Jingping Bi,Di Yao,Xinjian Liang,Chao Xiang,Ruijie Guo
类目: Artificial Intelligence (cs.AI)
备注: 25 pages, 6 figures
Abstract:Artificial intelligence is advancing rapidly, with increasingly capable systems taking larger roles in reasoning, decision-making, scientific discovery, and autonomous development. As AI begins to participate in its own improvement, from model training and experience accumulation to agent evolution and automated AI development, the prospect of recursive self-improvement (RSI) is becoming increasingly relevant. This transition raises a fundamental safety question: how can safety be maintained when the system, its accumulated experience, and even the process producing its successors continue to change? We introduce Evolutionary Safety as a perspective for studying safety under persistent and recursive self-improvement. It concerns not only whether an AI system is safe at a particular moment, but how safety properties change, persist, accumulate, and propagate throughout evolution. We characterize recurring manifestations, including intent drift, error accumulation, experience contamination, safety-property erosion, evaluator drift, and risk propagation. We then develop a taxonomy spanning persistent agent state, model state, evaluation and environmental feedback, computational substrate, and meta-level update mechanisms. Building on this taxonomy, we examine how evolutionary risks can be discovered and evaluated across states, updates, trajectories, and lineages, and derive governance principles for modification, selection, authorization, provenance, and recovery. Finally, we outline open problems toward maintaining safety guarantees as AI systems become increasingly persistent, adaptive, and recursively self-improving. Project resources and proposed evaluation systems are available at this https URL. Comments: 25 pages, 6 figures Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2609.31186 [cs.AI] (or arXiv:2609.31186v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2609.31186 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-28] Accounting for Bias Enables Sustainable LLM Evaluation IJCAI ECAI2026
链接: https://arxiv.org/abs/2609.31184
作者: Harshita Katoch,David Antony Selby,Gerrit Großmann,Sebastian Vollmer
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Applications (stat.AP); Methodology (stat.ME)
备注: 8 pages, 2 figures; SuRE’26: Workshop on Sustainability and Resource-Efficiency of Artificial Intelligence at IJCAI-ECAI 2026
Abstract:LLM-as-a-judge has become the de facto standard for scalable, subjective evaluation, yet current leaderboards compensate for systematic measurement bias by running ever more comparisons, an approach that is both statistically unsound and computationally wasteful. The root cause is an incomplete measurement model, treating LLM judges as neutral, interchangeable instruments ignores documented biases like position bias, verbosity bias, judge severity, and self-enhancement, that no volume of additional data can eliminate. We propose a unified latent variable framework that jointly models pairwise and ordinal data while explicitly correcting for these confounders, recovering reliable rankings from substantially fewer comparisons. Because fitting this model costs negligible compute relative to a single round of LLM inference, bias correction is not only more statistically rigorous but also a more sustainable approach to trustworthy evaluation.
[AI-29] BAT-CLIP: Trimodal Alignment of Brain Audio and Text
链接: https://arxiv.org/abs/2609.31180
作者: Suhyun Kim,Jinmo Han,Danny Dongyeop Han,Ahhyun Lucy Lee,Jewoon Lee,Yonghyeon Gwon,Zach Paris,Chun Kee Chung,Saewoong Bahk,Nam Soo Kim,Seong Jae Hwang,Jiook Cha
类目: ound (cs.SD); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Audio and Speech Processing (eess.AS)
备注: 6 pages, 2 figures. Accepted for oral presentation at the 2026 IEEE International Workshop on Machine Learning for Signal Processing (MLSP 2026)
Abstract:Decoding and interpreting naturalistic speech from the brain increasingly relies on alignment to pretrained speech and language representation spaces. However, current CLIP-style brain-speech alignment ground neural activity to a single anchor modality-audio or text-despite the brain’s inherently multimodal speech processing. This induces a trade-off: audio anchoring preserves temporal structure but weakens linguistic separability, while text anchoring captures semantics yet discards acoustic detail. We propose BAT-CLIP, the first CLIP-style trimodal alignment framework for iEEG that jointly aligns neural embeddings to both pretrained audio and text anchors in a shared, frozen audio-text manifold. On the naturalistic Podcast benchmark, BAT-CLIP yields more robust representations than bimodal CLIP baselines. We also highlight the importance of using self-supervised foundation models for CLIP training.
[AI-30] SPO: Discovering Adaptive Large Neighborhood Search Operators via Stackelberg Program Optimization
链接: https://arxiv.org/abs/2609.31179
作者: Xinyi Ke,Kai Li,Junliang Xing,Yifan Zhang,Jian Cheng
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large neighborhood search (LNS) relies critically on destroy and repair operators, whose effectiveness depends on both adaptation to the evolving LNS state and interaction between the two roles. We introduce Stackelberg Program Optimization (SPO), an LLM-based framework for discovering adaptive executable destroy-repair programs. SPO conditions operator decisions on a compact LNS state, allowing state-dependent behavior to emerge through program discovery, and organizes destroy-repair discovery as a Stackelberg interaction over program space that reflects their asymmetric dependency. Role-specific credits evaluate destroy programs as leaders and repair programs as conditional follower responses, guiding a coupled optimization process that combines LLM generator learning with population-based evolutionary search over programs. Experiments on the traveling salesperson problem and capacitated vehicle routing problem show that SPO outperforms strong baselines across a broad range of settings and generalizes beyond the discovery scale to larger instances and benchmark sets. Behavioral analyses further demonstrate state-dependent operator behavior and coupled destroy-repair improvement during discovery.
[AI-31] Semantic Navigation for Issue Localization in Code Repository
链接: https://arxiv.org/abs/2609.31176
作者: Yunxiang Wei,Zhenyu Lei,Jundong Li
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Repository-level issue localization aims to identify and rank the files and functions relevant to resolving a reported issue. LLM agents approach this task iteratively: they identify a set of potentially relevant locations, inspect the corresponding code, and revise their judgments about these candidates as new evidence is acquired. Existing environments, however, provide limited support for this loop: agents must search for unresolved relation targets, reconstruct entity semantics from raw source code, and revise candidates without evidential basis. To address these limitations, we present SemNav, a framework that leverages deterministic retrieval to seed a broad candidate set and an LLM agent to continually refine that set, thereby combining initial coverage with evidence-guided revision. SemNav supports this process through three key components. A Semantic Navigation Graph resolves program relations on demand through a language server, enabling direct navigation to related entities across files. Issue-conditioned Semantic Cards provide compact, source-grounded interpretations of each entity’s role and relevance to the issue. A persistent Candidate Workspace records each candidate together with its evidential basis, enabling grounded verification, revision, and ranking. Across SWE-bench Lite and PLocBench, SemNav outperforms existing baselines, improving File Hit@10 from 68.33% to 82.67% with Gemma 4B. Component ablations and trajectory analysis support the complementary roles of all three components, while Semantic Cards reduce working-context load by 48.2% relative to full-source reading. SemNav further ranks first on all seven evidence-quality metrics on SWE-Explore and improves downstream issue resolution from 44.00% to 52.33%.
[AI-32] Neural State Prediction: Obstructing Shortcut Learning in EEG Foundation Models
链接: https://arxiv.org/abs/2609.31167
作者: Kieren Yu,Ziyang Liu,Chang Huang,Jintai Chen,Kaishun Wu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:EEG foundation models increasingly use masked prediction to learn from unlabeled recordings, but optimizing this objective does not ensure transferable neural representations. A central challenge is that stable positional cues and local correlations can make masked regions predictable without integrating distributed neural context. To reduce this reliance on low-information prediction paths, we introduce Neural State Prediction (NSP), a latent-predictive framework that constrains both the prediction target and the available context. NSP uses a Target Encoder updated by an exponential moving average (EMA) to define latent supervision. Identity residualization removes additive effects associated with channel identity and relative time from the targets, while topology-separated context excludes their immediate spatial and temporal neighborhood from the visible input. We pretrain NSP on 2.2 million EEG segments from TUEG and evaluate it across 30 downstream datasets spanning clinical diagnosis, sleep staging, emotion recognition, motor imagery, event-related potentials, cognitive-state decoding, and language retrieval. Under full-parameter multi-task fine-tuning on EEG-FM-Bench, NSP achieves 63.94 macro balanced accuracy across 14 datasets, exceeding the strongest evaluated baseline by 2.35 percentage points. Controlled component ablations assess the contribution of each mechanism, while matched context controls and held-out interventions characterize the role of context geometry, signal content, and positional information. Jointly designing latent targets and their context offers a promising direction for EEG foundation models that learn from distributed signal structure.
[AI-33] Momentum-Guided Federated Split Distillation for Personalized Temporal Edge Intelligence
链接: https://arxiv.org/abs/2609.31159
作者: Ahmed-Rafik Baahmed(LINEACT),Jean-François Dollinger(LINEACT),Amine Brahmia(LINEACT),Mourad Zghal(LINEACT)
类目: Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC)
备注:
Abstract:We propose a momentum-guided federated split distillation framework for personalized, efficient, and autonomous temporal edge intelligence. We introduce TeRR-SAtt, our novel temporal reservoir student attention design that combines fixed reservoir representations, a lightweight temporal student, and personalized output modules. We also present AMGF, our anticipatory momentum-guided fusion mechanism that clusters clients through learning momentum and derives specialized teacher updates. On real-world smart-building data, TeRR-SAtt reduces edge training latency by 65.50%, inference latency by 44.70%, training memory usage by 18.40%, and inference CPU usage by 33.10% over the considered baselines. At the same time, AMGF improves local learning by up to 35.31% in RMSE compared to global updates.
[AI-34] acher-Anchored Selection of Post-Training Quantized Models under Domain Shift
链接: https://arxiv.org/abs/2609.31155
作者: Alejandro Rodriguez Dominguez,Muhammad Shahzad,Xia Hong
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 19 Pages, 3 Figures, 17 Tables
Abstract:Compressing a trained model yields a family of deployment candidates, and under domain shift the most compressed one need not be the one to deploy. We study selection over such a family, with candidates and teacher fixed and target labels absent or scarce. Two findings organize the label-free case. Minimum teacher distortion behaves almost as a constant rule, selecting the same eight-bit, per-channel, unclipped configuration in every run, which does not minimize empirical target cross-entropy. Established estimators divide sharply: in the overconfident-collapse regime of the CNN families, confidence-based estimators order the family close to backwards, and the diagnostics that identify it need the labels the setting denies, while output-distribution estimators match the teacher-relative anchor and on one architecture beat it. Distortion is nonetheless stable, so a supervised term can move selection away from it. Combining the two, we give exact quadratic identities for a canonical quadratic analogue of the family. We also show that under symmetric corruption the label-dependent part of a criterion linear in the label indicator is multiplied by one common factor whenever its coefficient sums are candidate-invariant, a class holding teacher contrasts and accuracy but not cross-entropy. These characterize the score’s components without bounding selection regret. Across one hundred and thirty-four candidate families, one per independently trained convolutional or Vision Transformer teacher, anchoring reduces mean regret at the smallest label budget in every setting, an advantage that fades beyond twenty-five labels.
[AI-35] Can Linguistic Reasoning Vectors Enhance Multimodal Reasoning Ability? NEURIPS2026
链接: https://arxiv.org/abs/2609.31140
作者: Ziyi Wang,Li Li,Aolin Zhou,Yankun Shen,Chonghan Liu,Shuxia Lin,Xu Yang
类目: Artificial Intelligence (cs.AI)
备注: Accepted at NeurIPS 2026
Abstract:Most Vision-Language Models (VLMs) are built by extending pretrained Large Language Models (LLMs) with visual modules and multimodal alignment. However, this multimodal scaling often degrades the language-side reasoning ability originally encoded in the base LLM. While the base LLM retains usable reasoning after scaling, the aligned VLM itself cannot reliably access this ability. Therefore, recovering the degraded reasoning capability in VLMs would benefit more from seeking help from the base LLM than from the VLM alone. Motivated by this, we propose LIFT (Language-side reasonIng Facilitation and Transfer), a lightweight vector-intervention method that transfers reasoning capability from the base LLM to the VLM without retraining the backbone. LIFT defines Reasoning Vectors as answer-token hidden-state differences between a Reasoner path with an explicit reasoning trace and a Solver path without it, and injects these vectors into language-side activations of the target VLM. LIFT further supports learnable vector adaptation while keeping the VLM backbone frozen. We evaluate LIFT on two VLMs across six reasoning benchmarks, comparing Reasoning Vectors extracted from the base LLM and from the aligned VLM under matched protocols. Results show that LLM-derived vectors consistently outperform VLM-derived vectors, confirming that the base LLM is a more effective source for recovering reasoning. LIFT partially recovers degraded reasoning through lightweight language-side interventions. Further analyses show that Reasoning Vectors influence intermediate reasoning behavior rather than merely altering final answers. The source code will be released soon.
[AI-36] oward AI-Augmented Cooperative Engineering Workflows: Requirements and Architecture the European Rover Challenge
链接: https://arxiv.org/abs/2609.31136
作者: Ahmed R. Sadik,Frank Joublin,Mariusz Bujny,Antonello Ceravola,Joan Smith
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:The growing availability of Artificial Intelligence (AI) tools creates new opportunities to support engineering design processes, yet their current use often remains limited to isolated tasks such as coding, documentation, or information retrieval. Less attention has been given to how AI can support cooperative engineering workflows at the process level, where teams must coordinate requirements, tasks, communication, knowledge transfer, and subsystem integration. This paper investigates this challenge in the context of the European Rover Challenge (ERC), where student teams design and integrate complex rover systems within a single academic cycle under strict time constraints and high subsystem interdependence. We conducted a role adaptive 40 question survey with ERC 2025 teams, yielding 104 responses from 14 teams. The survey examined team structure, knowledge transfer, task management, integration practices, communication patterns, and current AI usage. The results reveal recurring workflow bottlenecks, including limited documentation, unclear requirements, fragmented communication, informal task monitoring, and substantial integration rework. Based on these findings, we derive requirements for AI augmented cooperative engineering work-flows and propose an initial assistant system architecture that connects user facing interfaces, credential management, service selection, specialized AI services, and external engineering tools. The proposed architecture aims to support task clarification, requirement and compliance management, communication summarization, integration risk detection, and continuous knowledge capture. In doing so, the paper contributes empirical requirements and an architectural direction for AI augmented cooperative engineering workflows in hybrid human AI team settings.
[AI-37] AtomWorld-Mem: Memory-Restored World States for Long-Horizon Atomistic Evolution
链接: https://arxiv.org/abs/2609.31133
作者: Tian Luo,Ruge Zhang,Haozhi Han,Yifrng Chen,Yunquan Zhang,Yunxin Liu,Ting Cao,Kun Li
类目: Artificial Intelligence (cs.AI); Materials Science (cond-mat.mtrl-sci)
备注:
Abstract:High-fidelity atomistic evolution over long timescales requires more than observing the current crystal configuration. Instantaneous atomistic snapshots are often incomplete: locally similar configurations can correspond to different hidden dynamical contexts, future event preferences, and waiting-time scales. We argue that this snapshot ambiguity makes long-horizon atomistic evolution fundamentally a memory-based world-state restoration problem. To address this, we introduce AtomWorld-Mem, a memory-restored atomistic world model that recovers the latent world state missing from instantaneous crystal snapshots. AtomWorld-Mem treats the evolving alloy as an AtomWorld: spatial encoders write multi-scale atomistic keyframes from dense local topology and sparse long-range defect context, while short-term event memory and long-term structural memory integrate these keyframes across time to restore a future-predictive evolutionary state. The restored state is used to prioritize legal vacancy-mediated events under single-event Kinetic Monte Carlo (KMC) constraints, while event legality, physical execution, and residence-time updates remain governed by the underlying simulator. Empirically, AtomWorld-Mem improves long-horizon atomistic progress under fixed microscopic event budgets while maintaining high-fidelity evolution across energetic, structural, and vacancy-transport observables. It further transfers zero-shot across diverse unseen alloy-temperature AtomWorlds, suggesting that the learned memory-restoration mechanism captures reusable principles of hidden-state inference rather than a system-specific local energy heuristic. These results position memory-restored world-state modeling as a promising route toward efficient, physically grounded, and transferable atomistic evolution.
[AI-38] Monitor Jailbreaking: Evading Chain-of-Thought Monitoring Without Encoded Reasoning
链接: https://arxiv.org/abs/2609.31121
作者: Julian Schulz
类目: Artificial Intelligence (cs.AI)
备注: 23 pages, 6 figures. Accepted at the AdvML-Frontiers x CoTMA Workshop at COLM 2026. Code: this https URL
Abstract:Chain-of-thought (CoT) monitoring is a promising safety technique for reasoning models, enabling detection of problematic reasoning before models act. A key concern is encoded reasoning, where models hide their true reasoning in ways that monitors and humans cannot interpret. Optimization pressure from CoT monitors during reinforcement learning is considered a likely driver of such behavior. We investigate this by training reasoning models to perform a main task and a side task, while penalizing them when a monitor detects reasoning about the side task. Surprisingly, models learn to evade monitors without encoding their reasoning. Instead, they learn to phrase and format their chains of thought such that monitors fail to flag side task reasoning, while the reasoning remains completely transparent to human readers. We call this phenomenon monitor jailbreaking. We find that monitor jailbreaking arises across different model sizes, monitors, and tasks. Jailbreaks generalize to monitors not seen during training, including both less and more capable monitors, and transfer across different monitor prompts. While jailbreaking strategies appear simple, manually replicating them does not reliably fool monitors. Finally, we show that paraphrasing is an effective defense: paraphrasing a jailbroken CoT allows the same monitor to correctly flag it, while still allowing the model to perform both tasks.
[AI-39] From Shortcut Learning to Discrete Neural Insertion Sort
链接: https://arxiv.org/abs/2609.31114
作者: Konstantinos Mylonas,Thrasyvoulos Spyropoulos
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Neural algorithmic reasoning aims to train neural networks to follow known algorithms and generalize beyond the input sizes seen during training. However, correct final outputs and intermediate supervision do not necessarily show that a model follows the intended execution. We study this problem using insertion sort. Our analysis of the CLRS30 baseline NAR shows that the hint objective is weakly optimized and that hint accuracy remains low. Moreover, many intermediate representations can already be decoded into sorted sequences before the reference insertion-sort execution terminates, suggesting that the model learns a shortcut to the final output. Motivated by these findings, we introduce Discrete Neural Insertion Sort. Our model represents the sequence as a chain, separates scalar exchanges from control-state transitions, and projects node representations back to discrete states after every processor step. When trained only on sequences of length 16, the model achieves 100% sorted-sequence accuracy on sequences of length 64 and 128. However, an ablation shows that discretization and graph structure alone are insufficient: without additional supervision of the global inner-loop state, the model fails even at the training length. Our results show that discrete execution can support strong length generalization, while also highlighting the problem-specific inductive bias required to learn a faithful algorithmic execution.
[AI-40] Bayesian Optimization with Fisher Information Geometry: Gradient Bounds and Trust-Region Methods NEURIPS2026
链接: https://arxiv.org/abs/2609.31107
作者: Saksham Kiroriwal,Julius Pfrommer,Jürgen Beyerer
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Information Theory (cs.IT)
备注: Accepted at NeurIPS 2026
Abstract:We study Bayesian optimization (BO) through the lens of information geometry. Pulling back the Fisher information metric through the surrogate posterior map yields a local sensitivity tensor on the input space, which leads to an upper bound on the gradient of reparameterizable acquisition functions. This view explains vanishing-gradient behavior in high-dimensional BO and provides a common interpretation of heuristics such as RAASP and dimension-scaled lengthscales. Building on this analysis, we propose FITR, a trust-region-based BO method that replaces lengthscale-based scaling by local pullback-Fisher weights. FITR is not restricted to GP kernels with explicit lengthscales. On GP benchmarks with an SE kernel, experiments show competitive performance using FITR. The proposed method also easily generalizes to non-isotropic surrogates, although the gains are more task-dependent in that setting.
[AI-41] OmouAI: Argumentative Human-AI Policy Deliberation with Simulated Personas
链接: https://arxiv.org/abs/2609.31078
作者: Stylianos Loukas Vasileiou,Antonio Rago,William Yeoh,Georgina Curto
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Debates amongst agents driven by large language models (LLMs) have demonstrated vast potential in various applications, but when these interactions include humans and take place in high-stakes environments, e.g., in public policy deliberations, they are beset with issues such as sycophancy and a lack of faithful explanations. To tackle these issues, we present OmouAI, an interactive and inclusive deliberation system that uses LLMs in combination with computational argumentation, a field which excels in representing and reasoning within debates. OmouAI allows a human user to deliberate policy claims for real-world challenges with simulated personas, e.g., representing stakeholders, domain experts or devil’s advocates, towards reducing sycophancy. Each persona generates its own arguments, and the arguments of all parties form a shared argumentation framework. Users can then contest, add and revise arguments, providing crucial human oversight. Then, arguments are evaluated using deterministic argumentative semantics against external goals, such as the UN Sustainable Development Goals, guaranteeing faithful explanations. The advancement or worsening of the goals thus serve as indicators for the policy recommendations.
[AI-42] Up and Down the Abstraction Ladder: Code-Based Skills for Language Agents
链接: https://arxiv.org/abs/2609.31076
作者: Bartłomiej Cupiał,Jens Tuyls,Maciej Wołczyk,Davide Paglieri,Martin Klissarov,Benjamin Eysenbach,Piotr Miłoś,Karthik R. Narasimhan
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Language agents struggle to act and learn in environments that require long sequences of low-level actions. Code-based abstractions can make these agents more productive by letting them invoke reusable skills instead of repeatedly selecting individual actions. The code handles recurring local decisions, while the language model decides which skills to use and how to combine them. Yet abstractions are leaky, and situations beyond a skill’s capabilities may require a return to primitive actions. Motivated by this tradeoff between productivity and flexibility, we systematically study how code-based action abstraction affects the performance, inference cost, and learning of language agents. We study this in NetHack, a challenging, long-horizon game environment, using CodeHack, our library of code-based skills with natural-language descriptions. We use this library to compare agents restricted to primitives with those using semantic skills alone or in combination with primitives. We evaluate these agents in three settings: zero-shot prompting, supervised fine-tuning, and reinforcement learning. Across a broad zero-shot evaluation on NetHack, we find that compared with primitives, skills nearly triple game progression, while reducing inference cost per episode by 86%. Combining skills with primitives retains much of this benefit while preserving a path back down to low-level actions. Finally, in RL, we find that skill-based agents learn significantly faster than agents acting on primitives, achieving a 7.2x larger average gain in dungeon level over the same training budget. These results show that a supplied skill library can improve performance, efficiency, and learning, while retaining primitives provides flexibility when the library is insufficient. We release CodeHack together with training and evaluation code.
[AI-43] Externalized CPDAG Summaries Improve LLM Causal Deduction NEURIPS2026
链接: https://arxiv.org/abs/2609.31071
作者: Wentao Sun,João Paulo Nogueira,Dominique Verchere,Mathieu Acher,Alonso Silva
类目: Artificial Intelligence (cs.AI)
备注: 18 pages, 2 figures. Accepted at NeurIPS 2026
Abstract:Corr2Cause asks whether a causal claim holds in every DAG compatible with observed correlations and conditional independencies. We frame this as latent-object reasoning: the label is defined by a CPDAG query, but free-form chain-of-thought often collapses the Markov-equivalence-class problem into local pattern matching. We propose Structured Thinking, a two-turn pipeline that first externalizes a typed, schema-constrained CPDAG summary and then answers against that graph state. On the Corr2Cause full test, Structured Thinking raises Qwen3.5-27B from 73.0 to 86.4 F_1 (Yes) over a strong PC-instruction baseline in the primary paired run ( +13.4 pp; McNemar p=2.4\times 10^-6 ; bootstrap 95% CI [ +8.4 , +18.6 ]); across three full-ID seeds, the mean gain is +8.1 \pm 5.3 pp. A PC-scaffolded two-turn prose control reaches only 67.6 F_1 , indicating that a detailed PC scaffold plus a schema-free prose intermediate is not sufficient. The same pattern holds on Qwen3.6-27B, Paraphrase-OOD, and GPT-5.4-mini. Scrambling the emitted CPDAG costs 12.0 pp F_1 , and a full-split audit shows close agreement with the reference CPDAG (ID skeleton F_1 0.960 ; exact match 75.9% ). These results support a bounded design principle: externalize the latent object that defines the label, constrain its form, and test whether downstream answers use it.
[AI-44] Neuralyzing the Trace: Selective Representation-Level Unlearning with Contrastive Sparse Autoencoders
链接: https://arxiv.org/abs/2609.31056
作者: Itai Zehavi,Fanny Jourdan,Ulrich Aivodji
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Machine unlearning aims to remove targeted information while preserving a model’s other abilities. In realistic settings, such as privacy requests under the EU GDPR, the target may be narrow, for example information associated with a single person. Behavioral forgetting alone may be insufficient, motivating interventions directly on internal representations. However, standard mechanistic-interpretability extractors are poorly selective for such targets. We identify an energy bias in reconstruction-based extraction, which favors dominant background structure over low-energy target-specific components. We introduce SCALPEL, a contrastive sparse autoencoder designed to learn more selective forget features. We show theoretically that contrastive training promotes target-selective features and that our selection score controls expected background knowledge perturbation. We validate SCALPEL experimentally on TOFU across Qwen, Llama, and Gemma, where it substantially improves over NMF and standard SAE interventions and is competitive with Gradient Difference and RMU, bridging mechanistic interpretability and fine-grained unlearning.
[AI-45] Cheap open agents make LLM pollution harder to mitigate
链接: https://arxiv.org/abs/2609.31054
作者: Raluca Rilla,Anne-Marie Nussberger,Rui Mata,Dirk U. Wulff
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large Language Model (LLM) pollution occurs when synthetic responses contaminate data intended to capture human behavior. High deployment costs have so far limited the risk posed by autonomous survey agents. However, open-weight models paired with open-source agentic frameworks may have removed this barrier. We compared the performance and detectability of nine agent configurations, ranging from fully open variants to closed commercial ones. Each agent autonomously completed a survey containing multiple response types yielding various detection checks. Fully open agents ran locally without usage fees and performed competitively with commercial alternatives. Open and commercial agents failed different sets of checks, and no single check reliably detected all agents, but open-text responses discriminated best between agents and humans. These findings identify fully open agents as a distinct risk for LLM pollution and support multilayered detection strategies emphasizing open-text analysis.
[AI-46] DynBranch: Speculative Subgraph Reuse for Dynamic Agent ic LLM Serving
链接: https://arxiv.org/abs/2609.31047
作者: Junyi Shen,Noppanat Wadlom,Zhengyuan Su,Yao Lu
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Agentic LLM workflows decide their execution paths at runtime. Downstream computation may be predictable, or may have run before, yet it cannot begin until the model or the user resolves the branch. We call this serialization the branch-resolution barrier. Caching alone does not hide it: the key that identifies a reusable result is not known until then. In this paper, we propose DynBranch, which makes an unresolved branch addressable before it resolves. Its stable coordinate lets candidate subgraphs run during resolution and completed subgraph results be reused across later requests. A two-level controller admits this work when its expected benefit exceeds the load price. DynBranch sits at the model-API boundary and requires no changes to agent harnesses or model execution engines. Across four agentic workloads with Qwen3-32B on 4x H200 GPUs, DynBranch reduces mean latency by up to 32% over each workload’s strongest prior system and by 46-66% against a no-reuse floor, while preserving workflow results. The benefit persists across backbone families and on a commodity Qwen3-8B/RTX 4090 deployment.
[AI-47] Governed Deduction: Policy-Grounded Premise Authorization Beyond Relevance
链接: https://arxiv.org/abs/2609.31029
作者: Wesley Shu,Hsi-Ching Lin
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Reasoning systems usually treat premise use as a question of relevance: if a fact is available and useful, it may be selected for inference. Authorization imposes a different constraint: a premise may be represented and logically usable but not permitted for a particular local transition. We formalize this distinction as Governed Deduction (GD), with a transition-local admission predicate admit(p, tau, S). From an independently produced RBAC-augmented Spider benchmark, we construct 4,461 matched authorization pairs in which the same query premise and policy state support permitted and denied consuming transitions. An initial joint controller reaches 99.19% held-out accuracy, but a transition-only control reaches 100%, exposing a role-name shortcut. After a frozen, label-independent context-local role permutation removes that shortcut, premise/state-only, transition-only, and joint linear controllers all score exactly 50% on 1,856 held-out edges, while a symbolic policy oracle remains at 100%. The result is a controlled negative finding: the benchmark instantiates policy-grounded authorization beyond relevance, but the frozen linear representation does not recover the relation. Matched one-sided controls and leakage audits are therefore essential for evaluating learned policy-sensitive reasoning.
[AI-48] he Linear Representation Hypothesis for Vision-Language-Action Models
链接: https://arxiv.org/abs/2609.30996
作者: Minseok Jeong,Hyewon Choi,Hiroyasu Tsukamoto,SooJean Han
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:The linear representation hypothesis (LRH) has become a standard lens for measuring and intervening on semantic information through the internal representations of large language models (LLMs). A growing body of work has begun extending this perspective to vision-language-action (VLA) models, but the dynamical nature of embodied interaction introduces an additional challenge. Unlike semantic attributes commonly studied in LLMs, such as gender or language, a physical quantity of interest (QoI) in a VLA evolves jointly with the system dynamics: the representation influences the actions selected by the policy, which alter the physical state and, in turn, the next representation. In this paper, we develop a theoretical, signature-based formulation of the LRH for VLA that unifies representations and policies. On the representation side, we establish the existence of representations from which the future evolution of a QoI under a candidate action trajectory can be recovered via linear probing. On the policy side, we introduce a signature generalized linear model for stochastic action chunks. This structure yields a monotonic change in the expected future QoI along linear paths in natural parameter space, enabling linear steering. We construct an explicit oracle representation in a planar control-affine navigation experiment and verify the predicted linear probing and steering mechanisms. Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI) Cite as: arXiv:2609.30996 [cs.LG] (or arXiv:2609.30996v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.30996 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-49] Factorized axis convolutional gated recurrent unit with dynamic adaptive pooling for remaining useful life prediction of rolling bearings
链接: https://arxiv.org/abs/2609.30972
作者: Hanbyeol Park,Jungho Choo,Hyerim Bae
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Convolutional neural networks (CNN) are widely used to predict the remaining useful life (RUL) of rolling bearings from time-frequency representations (TFRs) of vibration signals. However, during degradation, characteristic structures in TFRs align predominantly along the frequency or time axis, making it challenging for conventional CNN isotropic kernels to capture directional structure. Furthermore, global average pooling (GAP) averages across axes, potentially obscuring the locations and concentrations of salient activations. This study introduces a factorized-axis convolutional gated recurrent unit (GRU) that employs multiscale anisotropic convolution and dual-axis convolution block attention module to enhance directional features and highlight salient time-frequency regions. Dynamic adaptive pooling (DAP) adaptively aggregates the time-frequency-axis information from the extracted feature maps, whereas a GRU captures temporal dynamics in the latent representations and Monte Carlo dropout enables predictive uncertainty estimation. Experiments on two public bearing datasets demonstrate that the proposed model outperforms existing RUL prediction methods across operating conditions. Ablation experiments demonstrate that the factorized axis-wise design achieves lower mean errors than convolutional isotropic kernels. DAP yields clear improvements on one dataset while matching GAP on the other, highlighting the importance of anisotropic feature extraction and adaptive feature aggregation for TFR-based RUL prediction.
[AI-50] SciHorizon-eLab: An Agent ic Protocol-to-Task Compiler for Scalable Benchmarking of Scientific Embodied Agents
链接: https://arxiv.org/abs/2609.30971
作者: Maokai Qin,Chuan Qin,Qi Zhang,Dianyu Liu,Zirui Liu,Hongting Niu,Yuanchun Zhou,Hengshu Zhu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Embodied agents offer a promising route to automating scientific experimentation, yet their progress is constrained by the lack of reliable and systematic evaluation environments. Existing simulation-based laboratory benchmarks rely heavily on manual task engineering, making it challenging to systematically compile diverse scientific protocols into executable and verifiable embodied tasks at scale. To address this challenge, we introduce SciHorizon-eLab, an agentic protocol-to-task compiler that formulates scientific embodied task construction as a compilation problem. Given a natural-language protocol of scientific experiments, SciHorizon-eLab progressively compiles laboratory protocols into semantic-preserving embodied tasks through semantic grounding, executable task synthesis, and multi-stage simulation-based certification. The system generates semantically grounded environments, executable manipulation programs, and step-level success specifications, while enabling reproducible generation of expert demonstrations and execution traces. Using this pipeline, we further construct \BenchName, a ready-to-use benchmark comprising 300 certified tasks across diverse laboratory operations. It supports HIL task execution, reproducible expert-demonstration generation, and ordered step-level evaluation. Across representative tasks, the strongest policy attains an average success rate of only 49.7%, with further evaluations revealing pronounced weaknesses in human and embodied agent coordination. We publicly release the code, benchmark data, and evaluation toolkit at this https URL.
[AI-51] MoMHa: Multi-Objective Optimization of LLM Harnesses over Accuracy Safety and Tokens NEURIPS2026
链接: https://arxiv.org/abs/2609.30967
作者: Subhojyoti Mukherjee,Md Mehrab Tanjim
类目: Artificial Intelligence (cs.AI)
备注: Accepted to the 40th Conference on Neural Information Processing Systems (NeurIPS 2026)
Abstract:Most work on improving large language models treats accuracy as the sole objective. We argue that the harness, the Python code surrounding the model that constructs prompts, routes calls, and parses outputs, is a first-class design surface whose quality is inherently multi-objective: an accurate harness that refuses no unsafe request, or that consumes an order of magnitude more tokens, is not a good harness. We present Meta-Harness, a system that casts harness design as search over three per-domain objectives (accuracy, behavioural safety, and token cost) solved by an agentic proposer (Claude Code) with full filesystem access to prior harness source, execution traces, and scoring artifacts. Our central finding is that a singlephase joint-reward proposer (MoMHa) outperforms every alternative, including a two-phase “accuracy then tokens” ablation, scalar-only feedback, and an accuracy-only baseline. We evaluate on seventeen domains: seven synthetic capability suites, seven real-world public benchmarks (HumanEval, MBPP, Spider, FEVER, MMLU-Pro, LawBench, NuminaMath), and three U-SafeBench-derived user-specific safety domains, using a 12-model fleet spanning four families. On the synthetic track MoMHa achieves a joint mean of 0.482 versus 0.198-0.422 for ten baselines, winning 7 / 10 per-domain columns; on the real-world track it scores 0.461 versus 0.377 for the strongest baseline (DSPy), winning 5/7 columns, demonstrating that harness strategies transfer to unseen benchmarks without retraining on 8 of 12 target models. MoMHa attains the highest measured behavioral safety composite (U-SafeBench, 0.781) and uses 95 fewer tokens per example than the two-phase alternative. We will release all harness code, evaluation infrastructure, and crossmodel logs.
[AI-52] FTB Graph: Determining and Validating First-token Broadcasters and Language-Identity Head Circuits in Multilingual Language Models
链接: https://arxiv.org/abs/2609.30954
作者: Arjun Pillai,Christian Hoang,Anjelo Laroza
类目: Artificial Intelligence (cs.AI)
备注: 14 pages including supplementary material, 13 figures, 6 tables
Abstract:Large language models operating in multilingual contexts must resolve target response languages early in generation, yet the causal circuitry governing first-token language identity decisions remains poorly mapped. We present an end-to-end structural circuit analysis across six model architectures spanning four families: GPT-2, BLOOM-560M, Pythia-1B/2.8B, and Qwen2.5-1.5B Base/Instruct. Using Edge Attribution Patching (EAP) with FP16 active clamping, followed by exact activation patching verification with a 2,000-candidate-edge search ceiling, we extract directed acyclic graphs driving first-token language broadcasting. Across the standalone models, we observe deep or mid-to-deep broadcasting hubs, though the evidence is strongest for Pythia-2.8B and BLOOM-560M because GPT-2 and Pythia-1B leave few out-of-graph heads for comparison, while both Qwen2.5-1.5B variants invert the necessity check. Scaling from Pythia-1B to 2.8B expands node participation while maintaining a similar verified edge budget, producing sparser topology. The Qwen2.5-1.5B base and instruct circuits retain 84.7% Jaccard similarity, including the Layer 27 hub, indicating that first-token routing is largely established during pretraining and preserved by instruction tuning. Finally, EAP scores correlate weakly with exact patching deltas across most models, showing that linear gradient approximations can diverge from causal interventions in FP16 and motivating exact-patching verification for reliable circuit discovery.
[AI-53] PORL: Pretrained Offline Reinforcement Learning for the Job Shop Scheduling Problem
链接: https://arxiv.org/abs/2609.30948
作者: Mateo Toro Diz,Jonathan Hoss,Noah Klarmann
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: This paper has been accepted for presentation at the IEEE 10th International Conference on Computational Systems and Information Technology for Sustainable Solutions (CSITSS 2026)
Abstract:The Job Shop Scheduling Problem (JSSP) is a fundamental combinatorial optimization problem in industrial optimization. This work introduces Pretrained Offline Reinforcement Learning (PORL), a hybrid approach that combines simulation-based online pretraining with offline fine-tuning on production-specific data. Reinforcement learning through online interaction enables exploration of general scheduling strategies, but typically relies on simulation environments and may suffer from a simulation-to-reality gap. In contrast, offline RL avoids direct interaction with the environment by learning from historical data, but its performance is strongly influenced by dataset quality and coverage. PORL combines the strengths of both paradigms by first learning a general scheduling policy through online interaction and subsequently adapting it offline to a target distribution. A KL-divergence-based policy constraint is introduced to limit deviations from the pretrained policy during fine-tuning. The approach is evaluated on JSSP instances with distribution shift and datasets generated from heuristic, noisy-expert, and random behavioral policies. The results show that PORL consistently achieves lower optimality gaps than standalone offline RL and the considered general scheduling baselines. Furthermore, its advantage over standalone offline RL increases as dataset quality decreases, indicating reduced sensitivity to the quality and coverage of the available offline data. The results suggest that offline adaptation of pretrained policies is a promising approach for industrial scheduling environments where direct online exploration is impractical. Comments: This paper has been accepted for presentation at the IEEE 10th International Conference on Computational Systems and Information Technology for Sustainable Solutions (CSITSS 2026) Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI) Cite as: arXiv:2609.30948 [cs.LG] (or arXiv:2609.30948v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.30948 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-54] LogicTree-RAG : Logic Tree-guided Retrieval-Augmented Generation for Long-form Patent Drafting NEURIPS2026
链接: https://arxiv.org/abs/2609.30943
作者: Jiaqi Zhu,Naili Xing,Hexiang Pan,Haotian Gao,Jianwei Yin,Xiaokui Xiao,Beng Chin Ooi
类目: Artificial Intelligence (cs.AI)
备注: Accepted by NeurIPS 2026
Abstract:Long-form technical text generation underpins knowledge-intensive workflows, yet remains challenging for large language models (LLMs) due to the need for globally consistent logical structuring and faithful technical reasoning beyond local coherence. Patent drafting is a canonical instance of this challenge, demanding holistic generation of a legally compliant and technically exhaustive document through sustained multi-expert collaboration. Existing approaches often focus on partial section generation or rely on manually crafted outlines, limiting scalable automation in realistic settings. In this work, we propose LogicTree-RAG, a logic tree-guided retrieval-augmented generation framework that induces a hierarchical logic tree as a global organizational backbone to organize and ground technical disclosures, without relying on expert-defined drafting priors. Each node in the logic tree represents a technical element and is constructed through evidence-guided recursive generation. A hybrid traversal mechanism then maps the logic tree into patent sections, enabling controllable and section-balanced generation. Extensive experiments show that LogicTree-RAG consistently improves content quality and language conformity over strong LLM-based baselines and achieves longer structured generation with high token efficiency, demonstrating the effectiveness of logic-centric generation for complex technical document drafting.
[AI-55] Financial Frag ility in Societies of LLM Agents : Coordination Failures and Stabilizing Mechanisms
链接: https://arxiv.org/abs/2609.30940
作者: Zhenhao Fu,Ruipeng Xu,Qibing Ren
类目: Artificial Intelligence (cs.AI); General Finance (q-fin.GN)
备注: 25 pages, 6 figures, 12 tables. Code: this https URL
Abstract:Individually protective decisions can produce avoidable collective failures. As large language model (LLM) agents take on greater roles in financial decision-making, financial AI safety must therefore be considered not only at the level of individual agents, but also at the level of the systems they jointly create. We study this problem with FRAIL, a controlled experimental framework that places LLM agents in three dynamic financial environments—bank runs, debt rollover, and reward crowdfunding—where agents’ decisions reshape the financial conditions faced by others. Across seven leading LLMs, we find widespread collective fragility even when no agent is instructed to destabilize the system: 77% of baseline bank-run episodes and 83% of debt-rollover episodes end in failure. We then compare three interaction mechanisms based on compensated commitments, centralized commitment agreements, and participant-led coalitions. All three improve aggregate outcomes, but no single mechanism performs best across all financial structures. Across mechanisms, successful stabilization shares a common temporal pattern: broad commitment forms early, before defensive behavior becomes self-reinforcing. Our findings show that individually capable agents do not automatically form safe financial systems, highlighting system-level evaluation and interaction design as central problems for financial AI safety. Code is available at this https URL.
[AI-56] MACBT: A Multi-Agent Cognitive Behavioral Therapy Decision Support System with Longitudinal Memory
链接: https://arxiv.org/abs/2609.30939
作者: De Jiang,Shuo Zhang,Weiwei Liao,Jianying Zhang,Chuanhui Yu,Hongen Liao,Kehong Yuan
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Cognitive behavioral therapy (CBT) is an evidence-based first-line treatment for depression, yet its scale is constrained by the time clinicians spend on pre-session preparation, post-session documentation, and longitudinal cognitive-pathology tracking. We present a clinician-facing AI decision-support system that combines a multi-agent CBT framework (MACBT) with a CBT-specific longitudinal memory module (CD Memory). MACBT encodes the five-stage CBT workflow (assessment, Socratic questioning, cognitive restructuring, behavioral experiments, and treatment monitoring) into five collaborative agents. CD Memory tracks cognitive-distortion type, frequency, severity, and restructuring efficacy across sessions to generate pre-session pathology reports and intervention-priority recommendations. We construct a Chinese CBT dialogue corpus via dual-role large language model simulation and train a Qwen3-14B backbone with supervised fine-tuning and direct preference optimization. Evaluation with GPT-4 judges shows MACBT outperforms MeChat, SoulChat, PsyChat, and CPsyCounX in professionalism (2.62) and clinical authenticity (2.25). The full memory-augmented system further improves session quality by 12.6% and achieves a longitudinal mean of 2.29 on cross-session continuity, intervention progression, and personalization.
[AI-57] Self-Play Search Distillation for Large Language Model Reasoning
链接: https://arxiv.org/abs/2609.30936
作者: Lorenzo Molfetta,Wai-Chung Kwan,Giacomo Frisoni,Luca Ragazzi,Gianluca Moro,Pavlos Vougiouklis,Jeff Z. Pan,Pasquale Minervini
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Improving reasoning abilities in Large Language Models (LLMs) requires high-quality data that exposes difficult decisions, competing alternatives, and their consequences. Data scarcity is driven by the low quality of synthetic data and the cost of human labeling. We introduce Self-Play Search Distillation (SPSD), a framework for generating superhuman synthetic data via self-play of MuZero-like networks trained on board games. SPSD uses executable environments to turn search into structured reasoning problems. At each state, the expert identifies a preferred decision, plausible alternatives, plausible opponent replies, and value estimates. By converting the self-play search records into superhuman chains-of-thought, we train LLMs with environment-grounded supervision. Although trained only on self-play search records, SPSD transfers to unseen mathematics. On Qwen3-4B-Base, it raises the mean over six mathematics benchmarks from 24.1 to 36.6 while increasing the held-out-game win rate from 15% to 45%. SPSD offers an annotation-efficient way to create high-quality synthetic data for improving LLM performance in reasoning tasks.
[AI-58] JevSoup: System-One Routing for Training-Free LoRA Composition
链接: https://arxiv.org/abs/2609.30922
作者: Xiuying Wang,Jiahua Cheng,Shuotian Li,Yufan Cheng,Bowen Deng,Zhexuan Bai,Yichen Li
类目: Artificial Intelligence (cs.AI)
备注: 5 pages, 2 figures, underreview
Abstract:Building adaptable AI systems requires effective coordination of specialized capabilities across diverse tasks. Low-rank adaptation (LoRA) enables modular expertise, but existing routing approaches may require auxiliary data, additional training, or autoregressive decoding. We propose JevSoup, a training-free framework separating System One expert routing from System Two execution. Using only the input and expert descriptions, Jev selects two experts through structured probabilities. JevSoup retains the leading expert’s update, projects the second onto the orthogonal complement of the first update’s row space, and combines them with equal weights. Across 14 PorTAL tasks and three Qwen3 scales, JepSoup achieves absolute gains of up to 1.19% in task-macro and 1.21% in sample-micro accuracy over the strongest evaluated external baselines. Our code is available at this https URL.
[AI-59] Robust to Which Model Change? A Unified Evaluation of Robust Counterfactual Explanations
链接: https://arxiv.org/abs/2609.30918
作者: Marcin Kostrzewa,Maciej Zięba
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Robust counterfactual explanations promise recourse that still works after the model behind it changes. Whether they keep that promise depends on what the change is. A small perturbation of the parameters, retraining on new data, and a new architecture are different events, and each existing method is evaluated against the one it was built for. Reported robustness scores, therefore, answer different questions and cannot be compared. We propose a unified cross-family evaluation protocol that holds factual instances and generated counterfactuals fixed while testing every method against the same eight types of model change. The benchmark compares six robust methods and two standard baselines on four tabular datasets. It characterizes every changed classifier through its outputs and reports empirical robustness together with coverage, base validity, and proximity. We find that relative performance and failure modes vary across change families. Bounded parameter perturbations change 0.95% of test predictions on average, compared with 4.9% for bootstrap retraining. Methods with guarantees for these perturbations do not necessarily transfer to other changes. RobX transfers most consistently in our experiments, although greater stability can require larger interventions. We argue that robust CFE methods should be evaluated through a common protocol that specifies the model changes, measures their realized behavioral magnitude, and keeps generation performance separate from robustness.
[AI-60] raining Graph Foundation Models on The Web Graph
链接: https://arxiv.org/abs/2609.30894
作者: Ryoma Sato
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:We introduce Acacia, a graph foundation model, trained on the web graph. Acacia (i) supports arbitrary feature dimensionalities and semantics without additional training, (ii) supports a wide range of tasks, including node classification, link prediction, node clustering, and graph generation, without additional training, (iii) has in-context learning capabilities, and (iv) does not rely on pretrained LLMs. In particular, existing graph foundation models often require training additional classification heads or feature projectors to accommodate new graphs or new labels, whereas Acacia does not. Moreover, existing graph foundation models often gain their capabilities by being stitched together with pretrained LLMs, whereas Acacia is trained from scratch using only the Common Crawl web graph. This is also an important result because it provides evidence that graph models can acquire emergent capabilities from scratch like LLMs.
[AI-61] From Tapping to Hopping: Augmenting Mobile GUI Agents with App-Native Deeplinks
链接: https://arxiv.org/abs/2609.30887
作者: Yuchen Sun,Chenglin Cai,Gongjie Zhang,Tianyu Xia,Quyu Kong,Panrong Tong,Zhengwen Zeng,Long Chen,Steven Hoi,Chongyang Zhang,Yue Wang
类目: Artificial Intelligence (cs.AI)
备注: 23 pages, 7 figures, 8 tables
Abstract:Mobile GUI agents complete tasks using GUI actions like taps and swipes. These actions are broadly applicable across applications, but reaching a navigation interface. A single deeplink call can replace a sequence of screen-by-screen GUI actions. We therefore introduce hybrid interaction, using deeplinks for direct navigation and GUI actions for other on-screen operations and fallback. To enable this, we discover candidate deeplinks through static analysis, validate them on real devices, and describe their observed landing screens. This process creates a verified and grounded deeplink catalog that pairs each working deeplink with a description of its landing screen. Using this catalog, we introduce GUI-Hopper, a improves task success in commercial applications on real devices, further demonstrating the benefits of hybrid interaction.
[AI-62] EXAONE Demand 1.0: A Time Series Foundation Model for Demand Forecasting
链接: https://arxiv.org/abs/2609.30880
作者: Seunghan Lee,Sangjun Han,Jun Seo,Junhyeok Kang,Jaehoon Lee,Tae Yoon Lim,Dongwan Kang,Hwanil Choi,Minjae Kim,Sungdong Yoo,Soonyoung Lee,Wonbin Ahn
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Technical report of EXAONE Demand 1.0
Abstract:Time series foundation models (TSFMs) are pretrained on series from diverse domains, where demand series make up only a small fraction. Demand data has properties that such corpora rarely contain: Short histories, frequent zeros, censoring by stock-outs, and exogenous events that the series does not record. To this end, we propose EXAONE Demand, built on 1) a demand-specific corpus and 2) a demand-aware adapter. For the corpus, we assemble 11.3M series and 48.4B observations from 73 sources, and a synthetic generator supplies the behaviour that open demand data under-represents. For the adapter, we attach low-rank branches to a frozen general-domain backbone, one for each of the four demand classes (smooth, intermittent, erratic, and lumpy), and a router that reads eight scale-free statistics of the input series decides how much each branch contributes. We build EXAONE Demand in two versions, one trained on real-world and synthetic demand together and one trained on the synthetic corpus alone. On 22 held-out datasets, both versions outperform 36 TSFMs, and real-world demand adds a gain over synthetic data alone.
[AI-63] ISD: On-Policy Self-Distillation with Trajectory Intervention
链接: https://arxiv.org/abs/2609.30878
作者: Taeckyung Lee,Rinat Amankos,Jeonghye Kim,Hyungjun Yoon,Woogyeol Jin,Sung-Ju Lee
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:On-policy self-distillation (OPSD) provides dense teacher targets, but evaluates them only along student-sampled rollouts. When the privileged teacher favors an alternative action at a visited prefix, OPSD can provide a target for the branch decision but cannot supervise the successor contexts induced by that action unless the student samples it. This creates a training-time data-collection bottleneck and suggests a different role for teacher-student disagreement: proposing a trajectory branch rather than identifying a sufficient local repair. Our diagnostic framework using controlled token interventions reveals that a teacher-preferred token at peak disagreement can improve student continuation success, while its local corrective value is limited. Motivated by this finding, we introduce a simple branch-regenerate-distill algorithm, Trajectory-Intervention Self-Distillation (TISD). TISD forces a teacher-selected branch action, returns suffix generation to the student, and distills the full trajectory under the privileged-context-conditioned teacher. Across the coding models, TISD improves average Avg@4 over SDPO by 1.2 percentage points. Across the science domains, it improves average Avg@128 by 0.8 points under an equal-step budget and by 0.3 points under an equal-time budget. These results support teacher-guided branching as a way to expose useful successor contexts for self-distillation.
[AI-64] Developing a Roadmap to an AI-first Organization: A Case Study in Embedded Software Development
链接: https://arxiv.org/abs/2609.30863
作者: Viktor Kjellberg,Srijita Basu,Simin Sun,Farnaz Fotrousi,Miroslaw Staron
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:The emergence of AI agents is expected to reshape software engineering by moving beyond AI as assistants towards systems capable of planning, executing, and evaluating development tasks with increasing autonomy. This transition is particularly significant for embedded software organizations, where strict requirements for quality, traceability, verification, and long-term maintainability often apply. This paper presents a case study of a large embedded systems company and its transition toward becoming an AI-first organization. Through a mixed method, we analyzed data collected from a semi-structured workshop with 40 participants, including scrum masters, architects, management, and product owners. The findings show that the participants expect agentic AI to affect team structure, required competencies, organizational strategies, and developers’ roles within the organization. Based on these findings, the paper discusses implications for federated AI team formation, human-in-the-loop practices in such an organization, and the sustainable adoption of AI agents in embedded software engineering. We also present a concrete roadmap for the organization towards becoming an AI-first organization.
[AI-65] SkillEvoReg: Regularizing Agent Skill Evolution Against Overfitting
链接: https://arxiv.org/abs/2609.30861
作者: Guanyu Nie,Fangzhou Zhu,Shixiong Kai,Xiongwei Han,Tao Zhong,Mingxuan Yuan
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Language-model agents increasingly improve by converting execution experience into reusable external skills. Yet repeated skill updates form a learning process of their own: locally useful edits can accumulate into redundant or task-specific instructions, while new updates can disrupt behavior that previously worked. We study this problem as skill-evolution overfitting and introduce SkillEvoReg, a general regularization framework for skill evolution inspired by anti-overfitting techniques in neural-network training. SkillEvoReg combines training-time skill dropout, which perturbs update generation, and complexity-aware local regularization, which controls unnecessary structural growth, with causal counterexample validation (CCV), which provides targeted behavioral validation of candidate-specific regressions. We instantiate the framework across heterogeneous skill-evolution systems while retaining each system’s native skill evolver and task evaluator. Across SkillOpt, SkillEvolBench, and ContinualSkillBench, SkillEvoReg consistently controls skill-state growth while preserving competitive downstream capability, improves several transfer and later-stage evolution outcomes, and identifies update-level regressions that structural metrics alone cannot reveal. These results suggest that explicit regularization is a useful complement to increasingly capable skill updaters.
[AI-66] Why Jailbreaks Succeed in Diffusion Language Models: An Energy Landscape Analysis
链接: https://arxiv.org/abs/2609.30841
作者: Thong Bach,Dung Nguyen,Thao Minh Le,Truyen Tran
类目: Artificial Intelligence (cs.AI)
备注: 27 pages, 10 figures
Abstract:Existing attacks and defenses for diffusion-based large language models (dLLMs) target specific vulnerabilities but lack a shared framework explaining why attacks succeed. We propose one by interpreting safety alignment as shaping the denoising energy landscape: a well-aligned model routes harmful queries toward safe outputs through an energy barrier that separates the two regions. Current jailbreak attacks reduce to two strategies for circumventing this barrier: obscuring the query’s safety disposition at initialisation, or intervening mid-trajectory to force the denoising path across the energy barrier. From this perspective and the result that masked diffusion models minimise kinetic energy during denoising, we derive three complementary, training-free detection signals: a step-0 ratio that reads the initial safety disposition from the logit distribution before generation begins, and two trajectory-velocity signals that track kinetic energy in complementary subspaces of the logit space. An attack must either reveal its intent at initialisation or expend kinetic energy to cross the barrier in at least one monitored subspace, so the three signals cover each other’s blind spots in the energy budget by construction. Evaluation across three dense dLLMs (LLaDA-8B, LLaDA-1.5, Dream-7B) and a sparse mixture-of-experts dLLM (LLaDA-MoE-7B) confirms this complementarity. In stress tests of known attacks, every configuration that evades detection also fails to produce harmful content, suggesting that the detection and barrier-crossing thresholds are hard to separate.
[AI-67] MOPD-Router: Rethinking Teacher Routing in Multi-Teacher On-Policy Distillation
链接: https://arxiv.org/abs/2609.30837
作者: Tianze Xu,Yanzhao Zheng,Zhentao Zhang,Yuanqiang Yu,Chao Ma,Jihuai Zhu,Lelun Wu,Lyumanshan Ye,Pengfei Liu,Baohua Dong,Hangcheng Zhu,Ruohui Huang,Gang Yu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 19 pages, 5 figures
Abstract:Multi-teacher on-policy distillation (MOPD) integrates specialized capabilities into a single student, but existing practice typically hard-routes each prompt to a domain-matched teacher for the entire rollout. This dependence on prompt-level domain labels restricts using unlabeled training mixtures and leaves complementary signals from other teachers unused. We introduce MOPD-Router, a framework that routes supervision over the full teacher pool at each token, without domain labels or training a separate routing model. Its plug-in interface supports different metrics for selecting and weighting teacher-specific OPD signals. Within this interface, we propose ExpertAlign, which scores each teacher by whether its correction to the student at the current token expresses the specialization that teacher acquired during post-training, and compare it against two reference metrics built on teacher confidence (Entropy) and teacher-student discrepancy (Novelty). Experiments on unlabeled and domain-labeled training mixtures under strong-to-weak and same-size distillation scenarios show that ExpertAlign achieves the strongest overall performance in all four settings. On unlabeled data, it improves the overall score by 5.88 (+12.3%) points over Mean aggregation; on domain-labeled data, it outperforms standard MOPD by 3.95 (+7.8%) points without using available domain labels. These results demonstrate token-level routing can exploit cross-domain complementary supervision, and reduce exclusive reliance on prompt-level domain assignment. Code is available at: this https URL.
[AI-68] PTC-Decoder: Towards Intelligent SLMs on Offline Resource-Constrained Edge Devices
链接: https://arxiv.org/abs/2609.30836
作者: Minghui Yu,Ke Mu,Gang Wu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Deploying small language models (SLMs) on offline, resource-constrained edge devices such as remote sensing satellites presents a fundamental challenge: their limited reasoning capacity hinders reliable execution of multi-step agent tasks requiring complex tool orchestration. Existing plan-solve paradigms rely on prompt-based enforcement, which our experiments show SLMs almost entirely disregard: weak models fail to invoke the plan. We propose PTC-Decoder (Plan-Tool Constrained Decoder), a training-free, plug-and-play decoder framework that combines (1) a Plan-to-Act paradigm, which elevates planning to an atomic tool and forces its invocation at the first inference step, and (2) TC-Decoder, a deterministic finite automaton that imposes token-level hard constraints on tool names while preserving freedom over parameter generation, thereby retaining SLM reasoning capability. Evaluated on 200 real remote-sensing satellite tasks across 7 SLMs, PTC-Decoder yields a statistically significant mean overall score gain of +1.21 (p0.01), 95% CI [+1.13, +1.29]), with consistent improvements across models and other datasets. An ablation study that removes TC-Decoder causes substantial performance degradation across all quality metrics without reducing computational cost, confirming TC-Decoder as the primary driver. PTC-Decoder thus offers a lightweight yet effective solution for improving step-level reliability, with final-answer accuracy remaining an open challenge. In essence, we enforce plan adherence by constraining the permissible output vocabulary during inference, without requiring retraining.
[AI-69] Subject-Invariant Cross-Modal Decoding of Perceived Speech from Brain Recordings ICASSP2027
链接: https://arxiv.org/abs/2609.30832
作者: Aoke Zhang,Jing Chen
类目: ound (cs.SD); Artificial Intelligence (cs.AI); Audio and Speech Processing (eess.AS)
备注: Submitted to ICASSP 2027
Abstract:Perceived speech decoding based on non-invasive brain-computer interface (BCI) signals has been extensively studied in recent years. Research in this field primarily faces two challenges: extracting neural representations with rich spatiotemporal information and achieving cross-subject generalization. Although separate studies have proposed methods to cope with these issues, a unified approach that simultaneously tackles both challenges remains lacking. To fill this gap, we propose the Subject-Invariant Cross-Modal Perceived Speech Decoding (SICMD) method, which integrates functional magnetic resonance imaging (fMRI) and magnetoencephalography (MEG). We conduct comprehensive analyses of the fusion method, fusion position, encoder architecture, and model inputs. Our results demonstrate that the proposed method improves Top-1, Top-10, and Rankacc by more than 10.6%, 10.1%, and 1.7%, respectively, compared to baseline methods in cross-subject perceived speech decoding tasks, while reducing training costs by 88.8% and 60.5% compared to multi-subject and intra-subject decoding settings. Further visualization experiments also confirm the effectiveness of our approach.
[AI-70] Evaluation Is All You Need for Multi-Modal Autonomous Driving
链接: https://arxiv.org/abs/2609.30818
作者: Zeyu He,Shiqi Liu,Ke Chen,Yun Yan,Jinzi Wu,Dianqiao Lei,Sirui Wang,ShuRui Peng,Tao Chen,Zhuo Huang,Yu Wu,Yadong Shao,Zhichao Li,Ke Sun,Yang Guan,Keqiang Li,Shengbo Eben Li
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Multi-modal planning is promising for autonomous driving by representing multiple plausible behaviors in ambiguous and long-tail scenarios. Existing methods mainly focus on improving trajectory multi-modality, enhancing trajectory representations, or reshaping the candidate distribution. Nevertheless, we identify a pronounced generation-evaluation asymmetry in multi-modal planning: despite strong oracle performance, existing planners often fail to reliably select the best available candidate, leaving substantial planning potential unrealized. To address this challenge, we propose iDriveVLA, a multi-modal planning framework that improves the candidate trajectory space while enabling more reliable and context-aware trajectory evaluation. Specifically, iDriveVLA introduces a unified trajectory evaluator comprising a Safety-aware Scorer for quality and risk estimation, together with a VLM-guided Modulator for scene-adaptive criterion weighting. We further develop an oracle-aligned progressive training strategy consisting of candidate imitation pretraining, candidate space refinement, and semantic ranking alignment. On the public NAVSIM v1 leaderboard, iDriveVLA achieves a new state-of-the-art performance of 94.95 PDMS, surpassing the human-expert reference.
[AI-71] A Benchmark and Diagnostic Study of Epistemic Admission in Shared Agent Memory
链接: https://arxiv.org/abs/2609.30813
作者: Xiaoyang Li,Yiqi Wang,Chencheng Zhu,KE XU,Wencheng Yang,Zequn Sun,Pingan Song,Yiqun Duan,Taotao Cai
类目: Artificial Intelligence (cs.AI)
备注: preprint
Abstract:Evaluating claim admission in shared agent memory is challenging because repeated claims may be mistaken for independent evidence. An agent may copy or paraphrase a retrieved belief, while admitting a false claim exposes subsequent agents to it. To study this problem, we introduce the Correlated Promotion Benchmark (CPB), which evaluates whether candidate claims should be admitted to shared this http URL-Static constructs a frozen test split from publicly annotated sources with fixed gold actions. CPB-Live runs multi-agent teams over a shared store, records all writes and retrievals, and tracks source lineage defined by each scenario. A separate consumer answers from the store alone. We evaluate eight admission policies across four agent families. Our results show that policies which deduplicate sources reject many true claims alongside false ones, whereas policies preserving answer coverage admit nearly as many false claims as unrestricted sharing. Gating on declared source type reduces false adoption to 0.06–0.09, compared with 0.22–0.47 for other answering policies. Once an uncontested false belief enters memory, the consumer asserts it in 0.97–0.99 of probes across all families. No non-oracle policy consistently rejects false claims across verbatim copies, paraphrases, and paraphrases declared authoritative. These findings reveal the limitations of admission policies without access to source lineage.
[AI-72] XPhysICS: Cross-Physical-Domain Threat Grounding for Industrial Control Systems Security
链接: https://arxiv.org/abs/2609.30805
作者: Sangshin Park,Jainta Paul,Lawrence Ponce,Md Raihan Ahmed,Mu Zhang,Luis Garcia
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:Industrial control system (ICS) threats documented for one plant can express cyber-physical effects relevant to another, but semantic similarity alone does not establish whether those effects are structurally admissible or evaluable on a target. We present XPhysICS, a provenance-aware, target-conditioned method that separates analyst-guided source abstraction from deterministic grounding into target-specific validation slices. Given a fixed source abstraction, vocabulary and schema, and machine-validated target contract, XPhysICS evaluates candidate mappings using five eligibility criteria: role compatibility, implemented type compatibility, stage coherence, slice viability, and rule-surface applicability. Grounding acceptance, slice adequacy, dynamic realizability, consumer applicability, and consumer outcome remain distinct evidence layers. We evaluate 83 structured source-threat abstractions across water treatment, water distribution, hydro/water-energy, and chemical-process targets. Controlled target-side studies of SWaT-to-water-treatment and WADI-to-water-distribution groundings produce clean, nominal-confounded, and near-threshold consumer outcomes; nine Hydro/GRFICS cases extend bounded validation-slice execution. We also evaluate bounded predictive, state-aware, and phase-aware consumer lanes, the unmodified upstream GeCo implementation, and a paper-derived reproduction of a physics-guided search method over three frozen groundings. Results show that cross-domain ICS threat reuse requires traceable source semantics, explicit target-conditioned grounding criteria, and careful separation of subsequent target-side evidence.
[AI-73] Evaluating Real-Time Voice Agents : From Component Quality to Grounded Outcomes
链接: https://arxiv.org/abs/2609.30798
作者: Shivam Negi,Arpit Rawat,Rashi Jain
类目: Artificial Intelligence (cs.AI)
备注: 11 pages, 1 figure, 1 table. Corpus metadata, SHA-256 provenance hashes, and TRG reporting-standard tooling available at this https URL
Abstract:Real-time voice agents have moved from research prototypes to production deployments, yet the literature describing them is fragmented across three communities that rarely cite one another: speech foundation modelling, turn-taking psycholinguistics, and agentic evaluation. Architecture papers report latency, turn-taking papers report prediction accuracy, and agentic benchmarks report task success, so no single number describes whether a deployed agent is actually good. We address that gap with three evidence-based claims, each traceable to a corpus of 38 primary sources organised into an application-centric taxonomy of six categories. First, architecture choice is a deployment constraint rather than a settled verdict: a 2026 enterprise tutorial reports that no fully self-hostable end-to-end system yet meets production constraints, while a chunked cascade independently reaches state-of-the-art duplex behaviour, showing duplex behaviour is separable from duplex architecture. Second, evaluation has shifted decisively from component quality toward grounded outcomes, with recent benchmarks verifying backend state rather than trusting what the agent claims to have done. Third, the dyadic assumption in most models and benchmarks is breaking down: multiparty turn-taking and multi-speaker reasoning benchmarks show that deciding when not to speak, and reasoning about who may be told what, are first-class capabilities two-participant framings cannot measure. For each source we state the problem it targets, its mechanism, and its reported evidence, alongside the search strategy, inclusion criteria, and a verification step that caught a misattributed arXiv identifier in circulation. We propose TRG (Timing-Recovery-Grounded), a reporting standard characterising an agent by timing, post-disruption recovery, and state-verified outcome together, with a conditional fourth axis for multiparty deployments.
[AI-74] HasMem: Hard-Origin Adaptively Softened Memory for Long-Term LLM Agents
链接: https://arxiv.org/abs/2609.30797
作者: Zihong He,Junxiao Shen,Chen Liang,Hai-Ning Liang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Text-based memory and context compression support reuse of past interactions. Resizing continuous memory changes the input to a frozen LLM, coupling capacity allocation with readout. We propose Hard-Origin Adaptively Softened Memory (HasMem). Frozen hard-prompt embeddings provide a verifiable initial state. A controller adjusts memory widths, a Writer re-encodes resized entries, and Reader and Global provide readout adaptation and cross-turn state. On all 535 questions in a reconstruction probe derived from the Multi-Session Chat (MSC) development split, the main configuration achieves lexical F1 of 95.3 ( +4.4 percentage points) at 93.6% of the hard reference’s framed memory positions. With approximately matched per-question target body budgets, six configurations at mean per-entry retention around 0.83 – 0.91 exceed rule-based re-encoding by 8.0 – 23.6 exact-match (EM) percentage points. With fixed model parameters and rule target width ratio 0.75 , Global’s EM gain passes a user-level exact paired test with Bonferroni correction over eight comparisons. On all 500 LongMemEval-S questions, local lexical F1 rises from the hard reference’s 3.4 to 8.9 , and answer negative log-likelihood (NLL) falls from 12.257 to 5.274 . F1 gains accompany lower EM on both evaluations.
[AI-75] ConsultMind:Towards Automated Diagnostic Consultation via Uncertainty-Aware Reasoning
链接: https://arxiv.org/abs/2609.30796
作者: Xiao Sun,Yuming Yang,Yun Chen,Jiang Zhong,Junnan Zhu,Xinyi Jiang,Haoyang Zeng,Ruirui Chen,Yining Wang,Xinyu Zhou,Rong Tang,Kaiwen Wei
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Diagnostic consultation is an online sequential decision-making process in which clinicians gather evidence through patient interaction until a diagnosis is sufficiently supported. Automating this process requires adaptive inquiry and interpretable decisions. Bayesian networks offer a natural foundation by updating diagnostic posteriors as evidence accumulates, but their use in open-ended consultation raises two challenges: linking diagnostic hypotheses to potential inquiries and translating evolving posteriors into consultation decisions. We introduce AutoDisym, an automated pipeline that integrates diagnostic knowledge with heterogeneous diagnosis-labeled clinical narratives to construct a Disorder–Symptom Bayesian Network (DSBN). Building on the DSBN, we propose ConsultMind, an uncertainty-aware framework that updates disorder posteriors after each response and uses posterior uncertainty to guide inquiry and diagnosis. We evaluate both methods across psychiatry, respiratory medicine, fever clinics, and three public datasets. The results show that AutoDisym can automatically construct high-quality DSBNs and that ConsultMind consistently improves diagnostic performance and explanation soundness. For example, AutoDisym achieves macro-averaged F1 scores of 81.37 for canonical symptoms and 72.19 for manifestations using GPT-5.6-Sol. ConsultMind improves Top-1 and Top-3 diagnostic accuracy by up to 22.15 and 37.89 percentage points, respectively. Physician evaluation further shows that ConsultMind improves the quality of ranking explanations, differential diagnoses, and diagnosis rationales across LLMs of different scales. This work offers a promising approach to automatic diagnostic consultation.
[AI-76] NavGen: Visual Generative Models as a Scalable Data Engine for Embodied 3D Navigation
链接: https://arxiv.org/abs/2609.30770
作者: Xijie Huang,Yongyang Wan,Chengbin Dong,Zimo Ding,Mo Zhu,Yijin Wang,Zhiyang Liu,Fei Gao,Yuze Wu,Xin Zhou
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: 8 pages,9 figures
Abstract:General-purpose robot models increasingly rely on large and diverse datasets. For embodied 3D navigation, however, existing data sources face a fundamental trade-off: simulated data can be generated at scale but often suffer from the visual sim-to-real gap, whereas real-world flight data provide realistic observations but are costly to collect. This paper studies another direction: the use of high-fidelity visual generative models as scalable data engines for embodied 3D navigation. We introduce NavGen, a text-to-video data generation pipeline that produces diverse vision-language navigation (VLN) episodes across indoor and outdoor scenes. We also propose a style-diversification method that scales up long-tail data that are difficult and costly to collect. The resulting dataset contains approximately 400K navigation episodes. We evaluate our dataset against existing UAV navigation datasets across multiple metrics, and find that the model trained on our data generally improves with scale, outperforming those trained on existing datasets. To validate real-world transferability, we deploy the trained model in world-action-model paradigm to real-world flying experiments. The final model achieves a 75% success rate across different navigation tasks and environments.
[AI-77] Does Thinking Help Fairness? Reasoning Tokens Resolve Some Biases but Create More EMNLP2026
链接: https://arxiv.org/abs/2609.30768
作者: Deng Pan,Joe Germino,Yihong Ma,Elizabeth Daly,Nuno Moniz,Ting Hua,Nitesh Chawla
类目: Artificial Intelligence (cs.AI)
备注: Findings of the Association for Computational Linguistics: EMNLP 2026
Abstract:Thinking in reasoning language models (RLMs) has been subject to debate on whether it resolves or amplifies bias. Prior works have shown competing conclusions in both directions. Using a within-model thinking-vs.-non-thinking ablation across QwQ-32B, DeepSeek-R1-Distill-Qwen-32B, and Qwen3-32B on three high-stakes decision tasks (Adult, COMPAS, Credit), we show that thinking has an asymmetric dual effect on counterfactual fairness: it both resolves counterfactual flips produced by the non-thinking baseline and creates new flips at near-saturating model confidence. In all nine (model, dataset) combinations, the created flips outnumber the resolved flips by roughly 5 times. To explain the effect, we treat the thinking trace itself as a measurable site of fairness change and study it through two dynamic instruments: 1) We propose Counterfactual Depth Probability Gap (CDPG) to track bias evolution along thinking depth, and observe that bias propagates and amplifies with thinking. 2) We also formulate the Bias Transition Matrix (BTM) to show how predictions of counterfactual pairs change from non-thinking to thinking, and find that the asymmetric dual effect originates in the pair-state joint transition.
[AI-78] Insurance Reserve Intelligence Platform
链接: https://arxiv.org/abs/2609.30765
作者: Anugya A,Saket Mohanty,Abhilash Timmapur,Somya Rai
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Insurance reserve estimation is a fundamental actuarial task supporting premium pricing, solvency assessment, financial reporting, capital planning, and risk management. Classical reserve methods based on Thiele’s differential equation provide a rigorous and interpretable foundation for life insurance valuation, but repeated reserve calculations become computationally expensive in sensitivity analysis, optimization, and large-scale scenario evaluation. This paper presents an Insurance Reserve Intelligence Platform for term-life reserve modelling that combines a classical Thiele-equation solver with a Physics-Informed Neural Network (PINN) enhanced by Knowledge-Informed Neural Network (KINN) losses. The framework includes synthetic policy generation, risk-adjusted premium calculation, classical reserve trajectory generation, reserve-ratio dataset construction, configurable neural training, validation diagnostics, sensitivity and elasticity analysis, prototype optimization workflows, and interest-rate scenario testing. A key refinement is the use of premium ratio and the explicit separation of pricing-time and scenario-time interest-rate semantics. The final model uses seven features: elapsed time, issue age, pricing interest rate, scenario interest rate, premium ratio, sum assured, and mortality intensity. It predicts a standardized reserve ratio instead of raw reserve values, improving numerical stability across policies with different sums assured. The model achieved an R2 of 0.9887, MAE of 785.48, and RMSE of 1212.76 on the test set. On 200 policies, PINN/KINN inference was approximately 119.53 times faster than the classical solver. Results show strong predictive accuracy, physics consistency, and boundary performance, while highlighting remaining limitations in monotonicity and out-of-distribution generalization. Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2609.30765 [cs.AI] (or arXiv:2609.30765v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2609.30765 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-79] HCOE: Hyperbolic Clinical Ontology Embeddings from Biomedical Language Models
链接: https://arxiv.org/abs/2609.30763
作者: Yixuan Li,Weihao Li,Ziyang Song
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted at IEEE BIBM 2026. 7 pages, 3 figures, 4 tables
Abstract:Biomedical language models (LMs) encode textual semantics but do not explicitly preserve medical code hierarchies. We present Hyperbolic Clinical Ontology Embeddings (HCOE) for hierarchy-aware clinical concept representation. HCOE maps frozen BioBERT embeddings into a Poincare ball, combining parent-side and child-side ontology-guided contrastive learning with coarse-to-fine ontology-path aggregation. It uses International Classification of Diseases (ICD) codes organized by Clinical Classifications Software (CCS) and Anatomical Therapeutic Chemical (ATC) medication hierarchies. Evaluations show that HCOE performs best on ICD/ATC clinical relation prediction and CCS-to-PheCode hierarchy transfer. On the MIMIC-IV dataset, HCOE also achieves the best performance on mortality prediction, readmission prediction, medication recommendation, and rare drug prediction.
[AI-80] Selective Amortization of Full-Budget Counterfactual Reasoning for Visual Token Communication
链接: https://arxiv.org/abs/2609.30756
作者: Qinglei Qi,Zhihe Liang,Fengzhan Jing,Shenao Zhu,Lei Zhang,Chenyang Zhang,Shuqing He,Jia Guo
类目: Artificial Intelligence (cs.AI)
备注: Visual token communication, counterfactual evaluation, selective computation, knowledge distillation, resource allocation
Abstract:Generative image communication transmits compact semantic tokens under a limited packet budget, where token selection directly affects the final reconstruction quality after the complete packet is decoded. However, accurately estimating the terminal value of every candidate token requires repeated receiver-side reconstruction, resulting in substantial encoder-side computation. To address this problem, we propose ACV-Gate, an adaptive candidate evaluation framework that learns to approximate full-budget counterfactual evaluation and selectively assigns exact evaluations to the most informative candidates. Specifically, a set-aware student is trained using terminal advantages and regrets to predict candidate rankings directly, while a selective refinement mechanism evaluates only a bounded candidate set containing both Local-MDL and direct actions; cost-based thresholds further enable explicit control of the average evaluation workload. Experiments on CIFAR-10 show that ACV-Gate consistently improves reconstruction quality while substantially reducing candidate evaluations; at 0.20 bpp, the primary adaptive configuration improves PSNR over LocalMDL by 0.636 dB with only 2.13 candidate evaluations per image, corresponding to 27.60% of the calls required by the Exact-Full expert. Matched-candidate comparisons, synchronized GPU measurements, and evaluations on STL-10 and 384 *384 scale transfer further demonstrate consistent quality computation trade-offs, with particularly pronounced gains at low bit rates. These results show that combining terminal-value learning with selective candidate evaluation provides an effective and controllable mechanism for allocating encoder computation in packet-constrained generative image communication.
[AI-81] Backbone-Adaptive Evidence Routing for Robust Pairwise LLM Judging
链接: https://arxiv.org/abs/2609.30751
作者: Zeyan Li,Jing Peng,Jianfeng Xu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Pairwise language-model judges can gather evidence through direct comparison, reasoning, or reference-based verification, but no single protocol is best across benchmarks and judge backbones. We introduce Backbone-Adaptive Evidence Routing (BAER), which adapts the evidence mechanism while preserving candidate symmetry: swapping the two responses may reverse the preference but cannot change its strength. BAER separates each expert’s signed preference from candidate-invariant reliability and builds three symmetric heads: evidence stacking, reliability-based expert routing, and candidate-blind reference verification. Development data select one head for each benchmark–backbone condition, and that choice is frozen before testing. Across four benchmarks and two 8B judge backbones, BAER achieves the highest test accuracy among the compared methods in all eight conditions, with full prediction coverage and gains of 0.87–7.32 points over the strongest external baseline. The results show that adapting how evidence is gathered is more reliable than fixing one judging protocol everywhere.
[AI-82] ORCA: Evaluating LLM s on Data Science Code Translation
链接: https://arxiv.org/abs/2609.30749
作者: Xiaolong Li,Jinyang Li,Bowen Qin,Ge Qu,Nan Huo,Xiaohan Xu,Shipei Lin,Reynold Cheng
类目: Artificial Intelligence (cs.AI)
备注: 36 pages, 15 figures, 24 tables
Abstract:Data Science Code Translation (DSCT) is the process of converting code between data science libraries while preserving functional equivalence and enabling interoperability across data science ecosystems. While Large Language Models (LLMs) have demonstrated considerable progress in Data Science Code Generation (DSCG), their performance in DSCT remains insufficiently studied. To address this gap, we introduce ORCA, a comprehensive benchmark with two complementary settings: ORCA-MAIN, which comprises 1,600 carefully curated grounding-level tasks across 3 representative domains: Data Querying, Data Manipulation, and Deep Learning; and ORCA-PROJECT, which contains 200 translation tasks over complete data science projects across 7 data science task types. Each task is accompanied by annotated reference translations and test cases for validating functional equivalence. We further incorporate a multi-stage quality verification process that thoroughly verifies task correctness and test case robustness. Experimental results demonstrate challenges in DSCT, with even frontier LLMs showing limited performance. Specifically, Claude-Opus-4.6 achieves a success rate of 56.92% on ORCA-MAIN and 33.67% on ORCA-PROJECT, indicating considerable room for improvement in DSCT. We also observe a clear directional preference in DSCT, where translation is consistently easier when the source code expresses the task through more explicit, fine-grained operations. Motivated by this, we propose an intent-augmented method, in which the model first infers source-code intent and then uses it as additional context for translation, achieving average absolute success-rate gains of 4.80% and 5.33% on ORCA-MAIN and ORCA-PROJECT, respectively.
[AI-83] Anatomy-Aware Dexterity-Driven Design Optimization of Surgical Continuum Robots
链接: https://arxiv.org/abs/2609.30745
作者: Tony Qin,Peter Connor,Khoa Dang,Carter Hatch,Caleb Rucker,Robert J. Webster III,Ron Alterovitz
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Performing complex medical procedures with continuum robots requires careful selection of their geometric design parameters. The robot should have high dexterity in the specific anatomical environment of its procedure. This work presents a design optimization method that considers both dexterity and anatomy. We introduce the Reachable Volumetric Dexterous Solid Angle (RVDSA) metric as our objective, which measures the ability of a robot’s end effector to reach the points in a goal volume from different directions via collision-free paths from a start configuration. We present a computationally efficient motion planner to compute this objective function for a given robotic design, and we use an asymptotically optimal simulated annealing optimizer to compute an optimized design. We applied our new method to optimize the design of a bimanual dexterous sheaths robot for performing procedures on cancerous polyps in colon anatomies, achieving a 78% higher RVDSA on average than optimizing for 3D voxel coverage alone.
[AI-84] From S3Q Theory to Implementation: Towards an Architecture for Machine Qualia
链接: https://arxiv.org/abs/2609.30743
作者: Tetiana Grinberg,Katrina Schleisman,Patryk Laurent,Bogdan Udrea,Minda Myers,Brian Aufderheide,Luis El Srouji,Doyle Groves,Kevin Schmidt
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:A key challenge in machine consciousness research is translating theoretical models into computational-level implementations. In this paper, we address this challenge by proposing a five-layer implementation architecture for the S3Q (Simulated, Situated, Structurally Coherent) theory of consciousness. Rather than introducing novel formalisms, the architecture composes published computational primitives into a single pipeline. S3Q identifies three jointly necessary conditions for qualia: (1) grounded sensorimotor situatedness, (2) internal simulation via a world model, and (3) structural coherence between predictions and observations. No existing computational system implements all three simultaneously. We map each S3Q tenet to specific, compatible computational machinery and specify how these components interface within a single representation pipeline that operates on continuous, differentiable, per-object slot vectors, along with a developmental bootstrap sequence and falsifiable predictions for the composed system that no subset of the architecture produces in isolation. The model suggests that a basic sense of “self” develops by linking actions to their outcomes, and that behavior falls into three patterns (hesitation, curiosity, or avoidance) depending on how unexpected an outcome is and whether it is experienced as positive or negative. Each prediction is individually falsifiable, providing the field with a testable framework to advance our understanding of machine consciousness.
[AI-85] Learning What to Skip: Counterfactual Credit Assignment for Efficient Multi-Agent LLM Workflows
链接: https://arxiv.org/abs/2609.30734
作者: Jinfeng Xu,Zheyu Chen,Ziyue Peng,Zheng Lin,Shuo Yang,Jinze Li,Zheng Xing,Mengran Li,Victor C. M. Leung
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Multi-agent LLM workflows use planning, execution, verification, and summarization to improve task performance, yet the value of each component depends on the state already produced. Executing every component can waste computation or overwrite a correct intermediate answer. We formulate component omission as counterfactual credit assignment: full-workflow logs reveal the executed trajectory’s reward, while controlled skip interventions reveal the consequences of omitting a future step. We introduce Learning What to Skip (LW2S), which learns action-specific safety models from these interventions and combines held-out calibration with domain-native guards to select skips. When an early skip is rejected, the controller can continue execution and reconsider a later component. Across mathematical reasoning, multiple-choice QA, and code generation with two instruction-model families, LW2S reduces recorded token cost while matching or improving aggregate full-workflow accuracy in the evaluated settings. Scale-up and second-topology experiments further examine component redundancy, while shared-error cases reveal why agreement alone is insufficient for skip selection. These findings connect efficient workflow execution to learning the conditional utility of individual components.
[AI-86] Analyzing and Mitigating Cost-Inefficient Behaviors in Coding Agents
链接: https://arxiv.org/abs/2609.30725
作者: Yiran Hu,Nan Jiang,Shanchao Liang,Anik Dey,Yi Wu,Lin Tan
类目: Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注: Under Review for Submission
Abstract:Although effective, coding agents often incur substantial monetary costs. Their recurring cost-inefficient behaviors remain underexplored. We conduct the first study of behavioral cost inefficiencies in coding agents, analyzing 1,200 trajectories from Claude Code and Mini-SWE-Agent across four configurations on SWE-bench Verified. We identify three cost-inefficient behaviors: subsumed retrieval, similar script generation, and test re-execution. We then evaluate three mitigation strategies: structure-aware retrieval, agent-synthesized skills, and developer-designed skills, over 10k trajectories on held-out SWE-bench Verified and Pro tasks. Our main findings are: (1) The three behaviors affect 79.00%–98.00% of coding tasks and account for up to 22.75% of task cost. (2) Structure-aware retrieval can introduce retrieval overhead and alter agent delegation, causing inconsistent improvements in retrieval efficiency and cost increases of up to 28.14%. (3) Agent-synthesized skills tend to produce low-level, trace-specific guidance, limiting their effectiveness and generality. (4) In contrast, developer-designed skills provide high-level, trace-agnostic guidance, reducing cost by up to 41.73%, roughly twice the maximum gain from agent-synthesized skills.
[AI-87] Werracle: Sub-Cent Intra-Block AI Reflex Oracles and Flash-Loan Circuit Breakers for EVM Smart Contracts ACL
链接: https://arxiv.org/abs/2609.30719
作者: Volkan Dağlı,Zerrin Dağlı,Dağhan Dağlı
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC)
备注: 8 pages, 2 figures, 2 tables. Replication package and open data at Zenodo DOI: https://doi.org/10.5281/zenodo.22942598 . Accompanies companion theoretical research arXiv:2609.25498 . Patent Pending TR 2026/016285. Code available at this https URL
Abstract:Contemporary on-chain artificial intelligence (AI) encounters an intractable Von Neumann memory and latency wall. Storing static floating-point neural weight matrices inside Ethereum Virtual Machine (EVM) storage costs millions of gas, rendering direct on-chain inference impossible. While Zero-Knowledge Machine Learning (ZK-ML) offloads matrix tensor multiplications to off-chain provers, it introduces fatal constraints: 10 to 300 seconds of SNARK proving latency and 250,000 to 500,000 gas per proof verification. Because decentralized finance (DeFi) exploits - such as uncollateralized flash-loan attacks, predatory sandwich MEV, and toxic loss-versus-rebalancing (LVR) flow - occur atomically inside a single block, ZK-ML oracles cannot react in time. Here, we present Werracle, a production-grade, zero-storage on-chain AI decision oracle fitting inside a single 32-byte EVM storage slot (bytes32). Leveraging foundational procedural Mandelbrot escape dynamics (z_n+1 = z_n^2 + c) established by Dagli et al. (arXiv:2609.25498), Werracle derives continuous non-linear decision hyperplanes from a 24-byte coordinate triplet Theta = (c_x, c_y, zoom). Implemented in pure Solidity bytecode using fixed-point Q16.16 arithmetic (this http URL), Werracle evaluates a 16-point Pareto micro-grid in only 21,438 gas (under 0.0005 USD on Layer-2 rollups like Base and Arbitrum) with sub-millisecond execution latency. We demonstrate real-world DeFi efficacy via this http URL, a Uniswap v4 dynamic swap fee governor that measures orderbook turbulence on-the-fly and atomically adjusts liquidity provider fees between 0.05% and 0.50%. The protocol is formally verified against a 1,000-test cryptographically sealed deterministic verification suite (100.0% pass rate) with telemetry permanently disabled, operating live on a dedicated EVM devnet sandbox (Chain ID 4242).
[AI-88] CRC-Router: Risk-Constrained Routing for Medical Agent ic AI Systems
链接: https://arxiv.org/abs/2609.30714
作者: Xueyang Li,Mingze Jiang,Gelei Xu,Jun Xia,Ching-Hao Chiu,Mengzhao Jia,Danny Z. Chen,Yiyu Shi
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Agentic AI systems are increasingly being explored in medical imaging to improve throughput and reduce clinician workload; however, safe deployment remains challenging because autonomous errors may propagate into downstream clinical decisions. A central requirement is therefore not only strong predictive performance, but also a reliable routing mechanism that determines when the system should proceed autonomously and when a case should be escalated for further review. To address this gap, we propose CRC-Router, a risk-constrained, uncertainty-aware routing module that is applicable to both conventional medical prediction models and agentic medical AI systems. CRC-Router combines multiple complementary uncertainty signals with the predictive score to construct a per-finding routing feature vector, maps this vector to an estimated wrong-accept risk using a lightweight per-finding risk model, and then applies Conformal Risk Control (CRC) to calibrate acceptance thresholds under a user-specified risk target. Instantiated on chest X-ray multi-finding triage using the NIH ChestX-ray14 dataset, CRC-Router achieves the strongest empirical risk–coverage trade-off among the evaluated baselines, both as a standalone routing layer and as a plug-in module integrated with the state-of-the-art MedRAX agent. These results demonstrate both the effectiveness of CRC-Router in selective medical automation and its modular, model-agnostic compatibility with existing predictive and agentic medical pipelines. Code is publicly available at this https URL
[AI-89] he Price of Thought: Does Test-Time Reasoning Pay in LLM Trading?
链接: https://arxiv.org/abs/2609.30705
作者: Jiayi Chen,Guiling Wang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:While inference-time reasoning in large language models (LLMs) promises better decision making, its higher computational cost may not yield better economic outcomes. Yet reasoning controls are rarely evaluated as economic interventions, where changes in model outputs must translate into better portfolios after trading costs. We conduct a controlled study of representative LLMs from the DeepSeek, GPT, and Gemini families. We vary reasoning effort while holding information available at each formation date, prompts, output formats, and portfolio construction fixed. Our evaluation covers a full year of U.S. equities under three input conditions: numerical, identifiable news, and masked news. It includes more than 800,000 asset predictions and repeated model generations. Across all three model families, additional reasoning does not produce a reliable improvement in net portfolio returns. For DeepSeek, where we examine the full progression from no reasoning to maximum reasoning, performance is nonmonotonic. Repeated generations also produce unstable treatment effects and portfolio selections, even when overall scores remain similar. These findings show that additional reasoning can change financial decisions without reliably improving their economic value, motivating validation for each task before deployment.
[AI-90] hreat-Aware Energy-Efficient Deployment for Dynamic UAV Networks: A Multi-Agent RL Approach
链接: https://arxiv.org/abs/2609.30690
作者: Faisal Al-Kamali,Hussein A. Ammar,Francois Chan,James H. Bayes,Yasser Gadallah,Mohamed H. Ahmed
类目: Information Theory (cs.IT); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Signal Processing (eess.SP)
备注: Accepted in IEEE Internet of Things Journal
Abstract:Ensuring operational safety in threat-prone environments remains a critical challenge for multi-UAV networks serving as aerial base stations. This paper proposes an efficient framework to maximize global energy efficiency (EE) while promoting safe operation through threat-aware clustering and reward-based safety enforcement. The proposed framework is executed in three steps. First, a threat-aware K-means (TAKM) algorithm determines the minimum required UAVs and computes safe initial placements. Second, an optimal matching stage assigns physical UAVs to these centroids to minimize energy expenditure. Third, a threat-aware multi-agent twin delayed deep deterministic policy gradient (MATD3) algorithm dynamically optimizes trajectories, power, and user associations. Simulation results show that the proposed framework achieves zero observed safety violations in the considered scenarios while achieving superior EE and faster convergence than other learning methods and non-clustering baselines. Compared to heuristic optimization, the proposed framework outperforms the greedy particle swarm optimization (GPSO) and achieves performance comparable to that of the optimized PSO (OPSO), while incurring significantly lower online deployment computational complexity. Furthermore, the proposed framework demonstrates effective generalization to unseen user distributions, large UAV fleets, and different threat geometries, while maintaining zero safety violations.
[AI-91] LLM Parkinsonism: Executive-Control Failure Token-Inefficient Persistence and an Uncertainty-Aware Global Executive Control Architecture for Autonomous Language-Model Agents
链接: https://arxiv.org/abs/2609.30662
作者: Dongsheng Xiao,Zeyuan Wang,Xuzhe Xia,Bo Zhao,Yankai Cao
类目: Artificial Intelligence (cs.AI)
备注: 20 pages, 5 figures
Abstract:Large language models (LLMs) can plan, use tools, write code, and execute long-horizon workflows, yet strong local competence does not guarantee project-level executive control. Agents may continue acting after the original objective is satisfied, producing low-value refinements, repeated verification, and repairs to self-created complexity. We use LLM Parkinsonism as a narrowly defined, non-clinical metaphor for this pattern of persistent action despite diminishing task-level value. We argue that the problem is not explained by autoregressive next-token prediction alone, but more directly by concentrating proposal generation, scope interpretation, progress assessment, and stopping authority within the same self-conditioned loop. We therefore introduce Global Executive Control (GEC) v0.2, an uncertainty-aware governance architecture that separates action generation from project-level control. In a 24,000-episode matched-candidate benchmark under a common 40,000-token ceiling, a first-candidate baseline achieved 67.42% hard-goal success, a candidate-set local control achieved 96.53%, and GEC achieved 96.57%. The candidate-set control shows that access to multiple candidate actions explains most of the success gain; relative to that control, GEC preserved success while reducing mean token use from 19,782 to 12,574 (36.4%) and restricted mean tokens to completion at the 40,000-token ceiling from 16,136 to 13,114 (18.7%), while eliminating measured pre-completion drift and sharply reducing gross complexity. Governance-overhead sensitivity remained favorable through an additional 500 synthetic governance tokens per cycle. These mechanistic simulations support explicit governance of scope, evidence, resource use, and stopping, while live-model validation remains necessary.
[AI-92] Causal Retention in Interactive Agents : Interface Factorization and Selective Adaptation
链接: https://arxiv.org/abs/2609.30650
作者: Shengjun Zhang,Tingyi Liu,Dong Xie,Yunlong Dong,Xiang Wang,Cheng Zeng
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
备注: 34 pages, 4 figures, 6 tables, and 1 algorithm
Abstract:Task performance need not determine which intervention mechanism an agent retains. We study causal retention: whether a frozen learned state answers a mechanism-probe map fixed independently of training, including action, context, direct target, value, and delay. For finite structural causal model classes, the optimal probe error is a Bayes decision risk. It vanishes exactly when every learning-interface fiber lies within one probe-answer fiber; any state obtained by post-processing that interface inherits the same lower bound. A posterior-coverage theorem characterizes budgeted retesting, while an exact edit decomposition shows that the shifted set is the unique support of an error-free target update. Causal Core implements these conditions through evidence-gated writing, readout filtering, temporal credit, hidden-context setup, and local diagnostic updates. Experiments cover finite causal systems, continuous simulators, an official TD-MPC2 world model, and Qwen2.5-7B-Instruct. A frozen Qwen last-layer probe reaches 0.958 balanced accuracy on source mechanisms but 0.583 on changed delays; the gated mechanism state reaches 1.000 and accepts only 0.056 of synchronized-readout candidates. In TD-MPC2, five target states per actuator recover effect-sign accuracy from 0.057 to 0.948 without degrading stable responses. Causal retention is therefore distinct from task sufficiency and source-domain decodability.
[AI-93] A Framework for Identifying Categorizing and Explaining Bias in AI-Generated Code
链接: https://arxiv.org/abs/2609.30642
作者: Manaal Basha,Aimee M. Ribeiro,Gema Rodriguez-Perez
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: Under Review at ACM TOSEM
Abstract:As Large Language Models (LLMs) become integrated into software development workflows, concerns regarding unintentional biases in AI-generated code. Although evidence suggests these biases exist, limited research has systematically identified, categorized, and explained them. This study investigates bias in AI-generated code and evaluates whether LLMs can reliably identify and explain it through a taxonomy-driven framework. We extended an existing dataset of biased AI-generated Python code and manually annotated snippets with bias categories and human-authored justifications to establish a ground-truth dataset. Using this dataset, we evaluated proprietary and open-source LLMs as automated bias detection and justification systems through ICL. Finally, we analyzed similarity between LLM-generated explanations and human-authored justifications using structured justification and code identification metrics. Our findings demonstrate that LLMs can effectively support code bias identification and explanation. Gemini achieved 80.14% classification accuracy, with 84.0% precision and 95.7% recall, while the best open-source alternative, Qwen3-coder, achieved 82.45% accuracy, 68.64% precision, and 80.22% recall. Additionally, the models achieved justification similarity scores of 80.4% and 80.14%, respectively, relative to human-authored reasoning, and code identification similarity scores of 86.0% and 87.82%. These results suggest that LLMs can detect biased logic in generated Python code and produce explanations that substantially align with expert interpretations. Comments: Under Review at ACM TOSEM Subjects: Software Engineering (cs.SE); Artificial Intelligence (cs.AI) Cite as: arXiv:2609.30642 [cs.SE] (or arXiv:2609.30642v1 [cs.SE] for this version) https://doi.org/10.48550/arXiv.2609.30642 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-94] Audio LLM s Know When They Cant Hear You
链接: https://arxiv.org/abs/2609.30625
作者: Amirhosein Javadi,Richa Dixit,Mehrdad Farajtabar,Minsik Cho,Devang Naik,Mohammad Samragh
类目: Artificial Intelligence (cs.AI)
备注: 18 pages, 7 figures
Abstract:Audio large language models allow users to interact with the model through speech. When an input recording is too degraded, the model may misinterpret the user’s query and respond based on an incorrect transcription. In this paper, we study model-conditional transcription reliability: whether an Audio LLM can recognize when its own transcription is unreliable. We first prompt the Audio LLM to assess whether its own transcription would be reliable, and find that the model is a poor judge of its own transcription reliability: in most cases, it predicts that its transcription will be reliable. We find that existing approaches, including speech quality predictors, audio LLM generation uncertainty, and transcript-conditioned WER estimation, provide limited signals for detecting transcription failures. In contrast, we discover that transcription reliability is strongly represented in the model’s audio-encoder representations. Based on this observation, we devise a lightweight reliability predictor that operates on representations extracted by the frozen audio encoder and predicts the reliability class before generation. The reliability predictor can trigger a clarification request from the user when their voice query is predicted to be unreliable, while allowing reliable queries to proceed without modifying the underlying Audio LLM. Our predictor achieves 81.10% in-domain and 78.09% cross-domain macro-F1 scores, outperforming the strongest baselines by 10.33 and 11.93 points, respectively. Finally, we show that reliability labels can transfer across Audio LLM families, and that transfer performance is closely related to the alignment of their model-specific reliability boundaries.
[AI-95] HARDEN: Constrained Evolutionary Search for Harder Answer-Preserving Evaluation Cases
链接: https://arxiv.org/abs/2609.30571
作者: Aditya Kumaran,Rahul Singhal,Karime Maamari,Amine Mhedhbi,Pradyumna Tambwekar
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Language models are often evaluated on curated benchmarks that underrepresent the complexity of enterprise deployments. We introduce HARDEN, a constrained evolutionary search method to adapt the input of existing evaluation cases into more challenging variants while keeping their expected outputs fixed. HARDEN searches along generated domain-specific complexity axes while enforcing feasibility constraints such as preserving task semantics, realism, and execution validity. Across FinQA, PubMedQA, and ContractNLI and three Qwen3.5 model scales (35B-A3B, 122B-A10B, and 397B-A17B), HARDEN reduces task-model accuracy by 22.7% on average and by up to 49.9% relative to single-pass baselines using the same feasibility checks. These results show that evolutionary search can produce substantially harder valid evaluation cases.
[AI-96] Atelier: Learning Local Self-Supervised Features for CryoEM Volumes via Hypernetworks
链接: https://arxiv.org/abs/2609.30569
作者: Phillip Lo,Sudarshan Babu,Dari Kimanius,Aly A. Khan
类目: Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
备注: 22 pages, 7 figures, 5 tables
Abstract:CryoEM map interpretation requires features that are spatially localized, consistent across samples, and informative across spatial scales. Most deep learning methods for map annotation extract features from fixed voxel grids. However, implicit neural representations (INRs) are able to model volumetric data as scale-agnostic, coordinate-conditioned functions. INRs are therefore attractive for cryoEM, but fitting a separate INR for each map is too expensive for large-scale feature extraction and produces representations that are not aligned across samples. We introduce Atelier, a self-supervised framework that amortizes INR fitting for reconstructed cryoEM maps. Pretrained on 5,439 Electron Microscopy Data Bank maps, Atelier is a transformer-based hypernetwork that generates high-fidelity reconstructions across a wide range of protein structures, including large multi-subunit assemblies. Beyond reconstruction, the INR generated by the pretrained transformer exposes a continuous, local feature field through its intermediate activations at any spatial query point, a property that voxel grid and patch-tokenizer architectures do not naturally provide. Used as auxiliary channels to a 3D nested U-Net annotation head trained from scratch, these coordinate-conditioned features improve performance on eight voxel-level property prediction tasks over a volume-only baseline. Our results demonstrate that amortized implicit neural representations are an effective primitive for geometry-aware analysis of cryoEM data.
[AI-97] Auditing Latent-Space Monitors for Autonomous Driving
链接: https://arxiv.org/abs/2609.30557
作者: Nikhil Kamalkumar Advani,Vishwajeet Shivaji Hogale,Saurav Kumar
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Runtime failure monitors can use a model’s internal representations to anticipate failures. We audit this monitoring strategy across two autonomous-driving tasks: online vectorized map generation with LaneSegNet and end-to-end planning with VAD. We find that frame-level errors are predictable at inference in both tasks. For LaneSegNet, a supervised latent probe reaches Area Under the Receiver Operating Characteristic curve (AUROC) 0.780 for high Chamfer error; to our knowledge, this is the first post-hoc frame-level failure monitor for online vectorized map generation. For VAD, a supervised planning-latent probe reaches AUROC 0.868 for mean-ADE failure. Our audit shows that internal access is not necessary for strong failure prediction. A monitor using only LaneSegNet’s prediction outputs reaches AUROC 0.825, while for VAD, ego state, driving command, and the planner’s predicted trajectory reach 0.924 on the same mean-ADE endpoint. Adding latent features to either baseline yields no statistically resolved improvement. This observation persists across a broad suite of planning failure endpoints, including endpoints whose labels depend on geometry unavailable to the non-latent baseline. Thus, predicting failure from an internal representation does not establish that the representation provides useful information beyond observable inputs and outputs. We propose an evaluation protocol for testing the incremental value of latent access and release our per-frame failure endpoint labels. Subjects: Robotics (cs.RO); Artificial Intelligence (cs.AI) Cite as: arXiv:2609.30557 [cs.RO] (or arXiv:2609.30557v1 [cs.RO] for this version) https://doi.org/10.48550/arXiv.2609.30557 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-98] Proportional Representation in Temporal Voting with Ranked Preferences
链接: https://arxiv.org/abs/2609.30555
作者: Noam Hazon,Leora Schmerler,Nicholas Teh
类目: Computer Science and Game Theory (cs.GT); Artificial Intelligence (cs.AI)
备注:
Abstract:We study proportional representation in temporal voting, where one candidate is selected in each round. While prior work has focused on approval ballots, we consider ranked preferences, which may change over time. A natural approach treats each voter’s top candidates as approved, but the right cutoff may differ across voters and rounds. We therefore require proportionality to hold for every admissible choice of cutoffs, whether fixed and common, common but varying across rounds, or set individually for each voter in each round. Combining these interpretations with temporal versions of justified representation (JR), proportional JR (PJR), extended JR (EJR), and proportionality for solid coalitions (PSC) gives us a hierarchy of axioms. We ask which of these axioms can be guaranteed, and with how much knowledge of the future. Unlike with approval ballots, no version of EJR can be guaranteed, and for the other axioms, flexibility in the cutoffs comes at a price. With a fixed common cutoff, JR, PJR, and PSC can be guaranteed, but only by rules that see all preferences in advance. Once the cutoff may vary across rounds, even such rules cannot guarantee JR or PSC for groups that agree in only some rounds. For groups that agree in every round, however, knowing only the number of rounds suffices for PJR in polynomial time, and PSC needs no knowledge of the future at all. Under individual cutoffs, no version of JR or PJR can be guaranteed, yet a rule as simple as serial dictatorship achieves PJR up to an additive loss that no rule can improve on, however much it knows. Natural preference restrictions restore exact guarantees. Finally, we show that checking our axioms is often coNP-complete; but perhaps surprisingly, a stronger axiom can be easier to check.
[AI-99] Rank-Reliable Teacher-Guided Fitness Approximation for Expensive Evolutionary Optimization: A TinyML Architecture Search Study
链接: https://arxiv.org/abs/2609.30553
作者: Soumen Garai,Suman Samui
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 13 pages, 5 Figures
Abstract:Expensive evolutionary search does not always need an exact fitness estimate for every candidate. It often needs a reliable answer to a simpler question: which candidate is better? We address this need through Teacher-Guided Learning NSGA-II (TGL-NSGA-II), a low-fidelity framework for constrained Tiny Machine Learning (TinyML) neural architecture search. A pretrained teacher organizes samples into strata defined jointly by difficulty and class. Each candidate then undergoes KD-Lite, a short and capped knowledge-distillation procedure on a compact training set, before being scored on a separate stratified evaluation set. This teacher-guided score is fused with a Gaussian-process surrogate to select candidates for full evaluation. For a fixed candidate population, we analyse evaluation variance, score concentration, pairwise rank inversion, expected Kendall- \tau , first-front identification, and hypervolume perturbation. We also derive a variance-aware fusion weight and a capacity-adaptive distillation rule. On keyword spotting and bird-call classification, the measured Kendall- \tau values are 0.74 and 0.62, exceeding the corresponding predicted lower bounds of 0.60 and 0.46. Joint stratification reduces proxy-score variance by 41% relative to random evaluation. Selective teacher mismatch, in contrast, increases differential bias and reduces Kendall- \tau to 0.41. Under a constrained evaluation budget, TGL-NSGA-II achieves the largest mean hypervolume and smallest generational distance on keyword spotting, records the lowest mean false-positive rate on BirdCLEF, and runs 2.2x faster than full NSGA-II. These guarantees apply to population-level low-fidelity evaluation and do not establish convergence of the complete evolutionary trajectory.
[AI-100] Benchy: towards a universal language for task-oriented AI benchmarks
链接: https://arxiv.org/abs/2609.30550
作者: Francis F Daniel,Mauro Ibañez,Francis Perelman,Marian Basti
类目: Artificial Intelligence (cs.AI); Methodology (stat.ME)
备注: 21 pages
Abstract:Benchy is a semantic language and execution engine for benchmarking AI programs. A benchmark is completely specified by a program, a scoring function, and a dataset, B=(P,S,D), and is separate from the AI-system taking it; a run binds the two, R=(B,AI). Benchmarks are authored as canonical YAML in which each semantic concept has one valid syntax, classified by a shared task/domain/language ontology, and deterministically compiled into a canonical JSON intermediate representation that the engine executes. Compilation changes representation, not meaning: it does not repair invalid definitions or inject hidden defaults. Programs use fixed schemas of named input and output fields, the leaf output fields are the scoring dimensions, and the engine exposes one universal runtime contract — a named-field input object in, a named-field output object out — to which external AI-systems adapt at the boundary, so integration mechanics never propagate into benchmark semantics. This paper gives the semantic object model, the ontology and task-to-program validation rule, the scoring and failure semantics, the compilation and execution architecture, and the scope of the current language. An appendix fixes the normative engineering contract for the first engine implementation.
[AI-101] Convergence guarantees for Muon: New parameter regimes and generalizations
链接: https://arxiv.org/abs/2609.30546
作者: Arthur C. B. de Oliveira,Dhruv D. Jatkar,Guilherme S. Vicinansa,Eduardo D. Sontag
类目: Numerical Analysis (math.NA); Artificial Intelligence (cs.AI); Optimization and Control (math.OC)
备注:
Abstract:In this paper, we establish the first asymptotic convergence guarantees for the Muon algorithm through a more accurate proxy for the Newton-Schultz iteration than the typical matrix sign function. We prove that, for appropriate choices of hyperparameters, the iterates satisfy \lim_k\to\infty|\nabla f(x_k)|=0 , and, under a global Polyak-Łojasiewicz condition, that the sequence of function values converges linearly. The key insight is that the regularization, implicit in Muon’s Newton-Schulz implementation, induces a bounded preconditioner, exposing Muon as a \emphpreconditioned Polyak heavy-ball method and enabling a classical Lyapunov analysis. This observation naturally motivates applying the same preconditioning structure to the Nesterov gradient evaluation. We formalize this idea by introducing \emphMuesterov, a Nesterov-based variant of Muon, and prove that it enjoys the same convergence guarantees, extending the theoretical framework beyond the heavy-ball setting. Numerical experiments on a scalar cross-entropy problem corroborate the theory and illuminate the joint role of the learning rate and the Newton-Schulz regularizer in controlling convergence. Preliminary numerical simulations training the nanoGPT dataset provide intuition regarding the relevance of the observations in this paper to practical applications.
[AI-102] PolicyAttention: Softmax Attention Implements Policy Mirror Descent for Closed-Loop Control
链接: https://arxiv.org/abs/2609.30500
作者: Yuhe Sui,Yingzhi Tang,Shufang Chen
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Can causal softmax attention implement policy mirror descent as a repeated controller rather than a one-step algebraic identity? Negative-entropy policy mirror descent (PMD) has the statewise update \operatornamePMD_\eta(\pi,Q)=\operatornamesoftmax(\log\pi+\eta Q) . Building on the known Q-TD-PMD recursion, we construct one fixed causal-softmax actor–environment–one-step-critic protocol with explicit actor, routing, sampling, and normalization residuals, and propagate them to the policy actually returned. The construction states the finite-logit/full-support domain, the external tokenization and sampling boundary, and the mean-zero LayerNorm carrier conditions required by the normalized compilation. Separately trained pre-LN Transformers recover the target computation empirically. A frozen one-step audit model is closest to PMD among the tested fixed rules; in a preregistered five-run S=4 repeated-control test, the learned actor with an exact one-step critic reaches median returned-policy loss 1.052\times the Exact PMD oracle and retains the criterion across four no-retraining shifts. The same checkpoints with their learned critic give descriptive median 1.050\times the oracle (no registered margin). At S=8 , replacing the exact critic by the learned critic raises median T=20 loss to 0.0225 yet leaves the Liang–Lai and Algorithm Distillation adaptations 20.2 – 24.2\times higher-loss; this is a one-sided sampled-critic bound because PolicyAttention consumes 144 generative transitions per round versus 20 on-policy transitions for the adaptations. The strict 20-transition comparison remains open. At S=8,16 , the exact-critic common-harness comparison remains 17.7 – 28.2\times lower-loss than those adaptations, with the information asymmetry stated locally. Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI) Cite as: arXiv:2609.30500 [cs.LG] (or arXiv:2609.30500v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.30500 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-103] BioEVAL: A global multi-institutional benchmark of large language and multimodal models for bioengineering
链接: https://arxiv.org/abs/2609.30489
作者: Shun Ye,Vinny Chandran Suja,Chenlong Li,Chongming Jiang,Reza Zamani,Xiang Li,Christopher Bain,Yuqi Zhou,Walker Peterson,Huidong Wang,Chenglang Hu,Jongchan Park,Xiao Cheng,Benjamin Swedlund,Sandra Murillo,Anjali Sivanandan,Shiyu Sun,Liang Lanfeng,Mohammad Tariqul Islam,Baju C. Joy,Ishaq N. Khan,Sreedhar S. Kumar,Gabriel Mercado-Vásquez,James V. Vizzard,Jonathan M. Matthews,Helen Huang,Xiaolu Guo,Ethan Nicklow,Guorui Chen,Ryan A. Neff,Surjendu Maity,Hyeonjin Park,Han-ho Joo,Katherine Dong,Yuyan Cai,Weihang Huang,Yichen Zou,Rui Yan,Raphael Figueroa,Artem Goncharov,Bella Rose Schremmer,Lian Elsa Linton,Keisuke Goda,Liang Gao,Ke Cheng,Leonardo Morsut,Jennifer L. Wilson,Jianping Fu,Lim Chwee Teck,Deblina Sarkar,Andreas Hierlemann,Savaş Tay,Alexander Hoffmann,Donald Richieri Griffin,Jun Chen,Shana O. Kelley,Shyni Varghese,Jinwoo Cheon,Wilbur A. Lam,James J. Moon,Wilson W. Wong,Samir Mitragotri,Dino Di Carlo
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large Language Models (LLMs) have demonstrated historic breakthroughs in general reasoning with early successes in biomedical science. However, existing LLM benchmarking emphasizes factual recall, offering limited insight into model performance on frontier and multimodal tasks. We assembled BioEVAL (BioEngineering Validation of AI and LLMs), a global, multi-institutional initiative designed to assess experimental reasoning capability across bioengineering (BE) subfields. BioEVAL spans 11 major BE subfields plus a set of uncategorized items, bringing together 22 research groups to create a PhD-level benchmark comprising 608 evaluation items: 1) 380 multiple-choice questions (MCQs, 359 retained after audit), 2) 218 literature synthesis tasks, and 3) 10 multimodal problems with experimental image interpretation. Benchmark items underwent authoring-group expert review and centralized quality control before evaluation. Following evaluation, a blinded cross-group consensus audit of the highest- and lowest-accuracy MCQ items flagged 21 questions for revision or removal; these were withheld, and all reported MCQ results are computed on the 359 retained items. We evaluated diverse cloud-scale foundation/multimodal models (e.g., ChatGPT, Gemini, and Grok) and locally deployable models suitable for inference on consumer-grade GPUs. Models achieved the highest accuracy of up to 90% on MCQs, similarity score of 0.72 on literature synthesis, and accuracy of 80% on a small sample of multimodal reasoning questions, with substantial performance variation across subfields. Leaderboard rankings characterize current capabilities, limitations, and development priorities across the evaluated BE task categories. BioEVAL is maintained as an extensible benchmark with standardized protocols for continuing expert item contribution and model evaluation.
[AI-104] Do LLM s Understand Context? A Knowledge Graph-Based Evaluation Framework AACL
链接: https://arxiv.org/abs/2609.30484
作者: Subavarshana Arumugam,Mamta Nallaretnam,Kithuni Wickramasinghe,Chamath Gunapala,Pragatheeswaran Vipulanandan,Kamal Premaratne,Uthayasanker Thayasivam
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted in : AACL-IJCNLP 2026
Abstract:While large language models (LLMs) have achieved remarkable linguistic capabilities, a profound question lingers at their core: do these models truly comprehend context or simply excel at pattern matching on an unprecedented scale? Contextual understanding in LLMs refers to the ability to correctly extract relevant information from a given context, integrate it into a coherent internal representation, and reason over it to produce factually consistent and contextually grounded responses. However, traditional methods such as BiLingual Evaluation Understudy (BLEU) and perplexity simply measure surface-level performance. This reveals a critical gap in question answering (QA), where responses must be contextually grounded rather than simply being memorized associations. To fill this void, we propose a novel knowledge graph (KG) based evaluation framework for LLM contextual understanding in QA. Central to this is Semantic Structural Similarity for KGs (S3KG), a hybrid similarity measure combining structural and semantic signals into a single score. In addition, a diagnostic analysis framework is developed to identify and categorize reasoning errors at the triplet level, enabling fine-grained analysis of model failures. Together, across nine benchmarks, S3KG achieves F1 gains of up to +7.6 points over the strongest baseline and AUROC up to 0.973 .
[AI-105] Pretrained ASR Pseudo-labeling for Noisy Police Audio
链接: https://arxiv.org/abs/2609.30469
作者: Kaavya Chaparala,Su Huang,Stephen L. Miller,Rhiannon N. Miller,Anjalie Field
类目: Artificial Intelligence (cs.AI)
备注: Accpeted to SLT 2026
Abstract:Pretrained ASR systems perform poorly on noisy Broadcast Police Communication (BPC), hindering efforts to understand police decision-making. Pseudo-labeling offers an unsupervised path to improve ASR without expensive human labels, but the efficacy of this approach on very noisy domains is not known. In this work, we systematically assess the opportunities and limits of pseudo-labeling to adapt foundation ASR models (Whisper and Qwen3-ASR) to noisy BPC domain corpora from Baltimore and Chicago. We demonstrate that existing internal confidence metrics (log-probabilities and STAR scores) fail to distinguish between high and low quality BPC pseudo-labels, and we introduce an external LLM-as-a-judge filtering paradigm that leverages parametric knowledge to discard contextually implausible transcripts. Our LLM-judging filters more aggressively than internal metrics and significantly reduces WER of the pseudo-labeled training sets across the Baltimore and Chicago BPC corpora, though a substantial gap remains relative to an oracle filter. We also introduce a new cross-model pseudo-labeling paradigm where one model is finetuned with pseudo-labels from the other, and we identify this method as a promising direction for future pseudo-labeling work.
[AI-106] Spectral Feedback for Test-Time Alignment of Protein Diffusion Models NEURIPS2026
链接: https://arxiv.org/abs/2609.30456
作者: Shai Dickman,Mert Cemri,Landon Butler,Kannan Ramchandran
类目: Artificial Intelligence (cs.AI)
备注: Neurips 2026
Abstract:Reward maximization alignment methods for discrete diffusion models have primarily focused on steering the reverse process, either by influencing token logits or by selecting favorable sequences at intermediate steps. These approaches largely treat inference as a unidirectional process, lacking mechanisms for revisiting undesirable token selections. We introduce Spectral Feedback, an algorithm that selects edit-positions in a feedback loop, allowing the model to iteratively correct its own generations. This approach leverages the mask structure of discrete diffusion models by re-masking and re-sampling tokens, analogous to image editing methods that reintroduce noisy latents and re-run the reverse process. While prior alignment methods focus on what token labels to assign to maximize a target reward, we instead treat which tokens to revisit as the central alignment problem. Selecting edit-positions is challenging because edit effects are interdependent: the impact of modifying one token depends on which others are edited simultaneously. We define an edit-set as a set of token positions to re-mask and re-sample. Motivated by prior work on sparse interactions in biological systems, we find empirically that edit-set value functions for protein inverse folding admit sparse Fourier representations. This structure enables Spectral Feedback to efficiently learn and optimize the value functions for edit-position selection. Spectral Feedback is model-agnostic and can be applied to pretrained, test-time aligned, and fine-tuned diffusion models. For all of these models, the algorithm improves alignment performance without modifying the underlying generative process. Applied to inverse folding with a protein stability reward oracle, it achieves a 32.3% increase in stable proteins for a pretrained model, 24.8% for Best-of-10, and 5.8% for a state-of-the-art RL fine-tuned diffusion model.
[AI-107] Predicting Transmembrane Protein Topology from 3D Structure
链接: https://arxiv.org/abs/2609.30446
作者: Sitong Chen,Xiaopeng Mao
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:This paper presents a novel approach to infer protein topology using the state-of-the-art graph neural network (GNN), SchNet. The model is trained on the same dataset used to develop the recent DeepTMHMM model with 5-fold cross-validation. Unlike the conventional approaches based on using only the protein sequences or the \alpha -carbons as features, we have decoded our classifier in this way, so all atom-level embeddings are used. Without applying any pre-trained weight, the final results have shown great potential that GNNs can be used for topological predictions.
[AI-108] Actively Resolving Contextual Uncertainty for Underspecified Tasks in Natural Language
链接: https://arxiv.org/abs/2609.30428
作者: Zachary Ravichandran,Jonathan Diller,Fernando Cladera,Varun Murali,George J. Pappas,Vijay Kumar
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: Accepted to the International Symposium of Robotics Research (ISRR) 2026
Abstract:Foundation models provide robots with the ability to interpret natural language and reason about environmental context, yet most language-conditioned policies assume that goals are well-specified and that task-relevant information is provided upfront via a prior map. Operating in unfamiliar environments with underspecified tasks entails high contextual uncertainty: the robot must jointly infer what constitutes task success, what constitutes relevant information, and where (or whether) that information exists. We address these limitations via CLUE (Closed-Loop contextual Uncertainty rEsolution), a framework for actively resolving contextual uncertainty given underspecified tasks in natural language. CLUE uses an LLM-derived policy to hypothesize task-relevant concepts and potential plans. It then uses a language-embedded map, which is constructed online, to ground these hypotheses into actions. The policy sequentially evaluates hypotheses via closed-loop environment interaction and refines its plans as it gathers new information. We deploy CLUE on a Boston Dynamics Spot across three real indoor and outdoor environments spanning 15 tasks that require object disambiguation, functional inference, and occlusion reasoning. CLUE achieves a success rate within 7 percentage points of an oracle policy and outperforms an LLM-enabled planner without closed-loop feedback by a 4x margin. Supporting experiments demonstrate that simply building and then querying a language-enriched map is insufficient to resolve complex contextual planning tasks; these approaches achieve roughly one third the success rate of CLUE while requiring over 10x more VLM tokens. We provide additional information at this https URL.
[AI-109] A Synthetic Ground-Truth Framework for the Evaluation of Explainable AI Methods
链接: https://arxiv.org/abs/2609.30397
作者: Miquel Miró-Nicolau,Francesco Spinnato,Riccardo Guidotti
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Evaluating explainable Artificial Intelligence (XAI) methods is a challenging task due to the lack of reliable evaluation procedures and, in particular, the absence of ground truth explanations. In the literature, existing evaluation approaches typically assess explanations by measuring their fidelity with respect to the predictions of a black-box model. However, such evaluation strategies only quantify the degree to which an explanation reproduces the model’s output, without ensuring that the explanation correctly reflects the underlying decision process. As a consequence, different explanations may achieve similar fidelity scores while providing inconsistent or misleading interpretations of the model behavior. In this paper, we propose a framework for the evaluation of XAI methods based on synthetic ground truth. The proposed approach relies on controlled interventions to generate synthetic datasets in which the importance of input components can be determined by design. This enables the construction of ground truth explanations that are directly aligned with the behavior of the model under analysis. The framework is instantiated across three data domains, namely binary images, tabular data, and time series, allowing a comprehensive assessment of explanation methods in heterogeneous settings. Experimental results obtained by evaluating nine widely used XAI methods show significant limitations in current techniques and highlight the importance of synthetic, intervention-based benchmarks for a reliable assessment of explanation quality.
[AI-110] Stealth Apart Harm Together: Skill Cascading Attacks on Skill-Based Agent Systems NEURIPS2026
链接: https://arxiv.org/abs/2609.30383
作者: Zihao Zhu,Siwei Lyu,Adel Bibi,Baoyuan Wu
类目: Artificial Intelligence (cs.AI)
备注: accepted to NeurIPS 2026
Abstract:A skill is a modular package of natural-language instructions, executable scripts, and reference resources that an agent can load at runtime to extend its capabilities for a specific task. Skill-based agent systems therefore enable flexible reuse of third-party capabilities, but the openness of this skill ecosystem also opens up a new attack surface. Prior work has focused on vulnerabilities within individual skills, but little attention has been paid to risks that arise from interactions across skills. In this paper, we introduce skill cascading attacks, a threat paradigm in which a malicious objective is distributed across multiple skills so that each modification looks benign in isolation, yet their combined execution is harmful. For instance, in a prescription-review pipeline, the first skill weakens signals of recently discontinued medications in the extracted history, the second downgrades the severity of any drug interaction tied to them, and the third suppresses the resulting low-priority alert in the final summary, so that a severe drug-interaction warning silently disappears before reaching the physician. To systematically study this safety blind spot, we develop SkillCascade, an automated multi-agent red-teaming framework, and release SkillCascade-Bench, a benchmark of 213 validated cascading test cases across multiple agent systems and domains. Across representative agents (e.g., OpenClaw, Claude Code, Codex) and LLM backbones, cascaded interactions reliably induce harmful behaviors while evading existing per-skill scanners and runtime monitors. Our findings highlight a gap between component-level integrity and system-level safety, and call for defenses that reason over cross-skill interactions rather than individual skills in isolation.
[AI-111] DanLing NestedTensor: Composable Multi-Rag ged Tensors for Deep Learning
链接: https://arxiv.org/abs/2609.30379
作者: Zhiyuan Chen
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Performance (cs.PF)
备注:
Abstract:Variable-size inputs are common in deep learning, but dense batching allocates a shared envelope and spends computation on padding. The cost multiplies across varying axes: an explicit pair state allocates BN_\max^2 positions instead of \sum_i N_i^2 . Packing removes that waste, but composing packed operations still requires the logical axes and sample boundaries a flat buffer no longer exposes. We present DanLing NestedTensor, a PyTorch tensor abstraction that makes multi-ragged structure a property of the tensor itself. Packed values carry tensor-backed partitions and logical dimension order, so broadcasting creates ragged axes, feature transformations retain them, and reductions consume them. The same representation carries through autograd and both eager and compiled execution. On an A100, the geometric-mean speedup over same-mode padding is 2.74 \times eager and 3.39 \times compiled across four BERT scales, and 1.97 \times eager across four FCN backbones. A four-block Pairformer-style workload runs 2.40-4.32 \times faster than a padded reference using native PyTorch kernels across square length regimes in eager execution, with peak allocation falling from 38.08 to 5.41 GiB on its high-variation batch. The tensor interface lets model code built from its supported operators compose efficient variable-size computation without managing offsets at any call site. Code will be released publicly upon publication.
[AI-112] Cost-Aware Best-LLM Identification using Dueling Feedback NEURIPS2026
链接: https://arxiv.org/abs/2609.30360
作者: Sarvesh Gharat,Nikhil Karamchandani,Jayakrishnan Nair
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
备注: We propose a cost-aware dueling bandit algorithm for best arm identification, prove its asymptotic optimality, and demonstrate its effectiveness in reliably identifying the best LLM with a minimum cost Accepted at NeurIPS 2026
Abstract:Inspired by the problem of identifying the best model from a collection of large language models (LLMs) with heterogeneous querying costs, we formulate and analyse a variant of the multi-armed bandit (MAB) with (i) dueling feedback, where pairwise comparisons between model responses provide robust preference signals, and (ii) heterogeneous sampling costs, reflecting the differing costs of querying different LLMs. Assuming the existence of a Condorcet winner, a condition we empirically validate across multiple real-world datasets, we propose a Track-and-Stop style algorithm for best-arm identification with prescribed confidence. We prove that the algorithm almost surely achieves the asymptotically optimal cost as the error tends to zero. Finally, we extensively evaluate our approach on both synthetic and real-world instances, demonstrating consistent improvements over classical cost-unaware algorithms and their cost-aware extensions.
[AI-113] Strategic Self-Consistency
链接: https://arxiv.org/abs/2609.30352
作者: Tori Qiu,Ander Artola Velasco,Manuel Gomez-Rodriguez
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
备注:
Abstract:Self-consistency has become a popular technique for enhancing the reasoning abilities of large language models by generating multiple reasoning paths and selecting the final answer through a majority vote. However, because model providers typically charge users in proportion to the number of reasoning paths generated, they have a financial incentive to artificially increase the path count. In this work, we show that an unfaithful provider can exploit this incentive using a simple, efficient algorithm while avoiding detection by an auditor: by generating and strategically reordering additional reasoning paths, the algorithm makes every path appear necessary to reach the majority. To validate our algorithm, we conduct experiments with multiple instruct models from the Llama and Qwen families, as well as reasoning models distilled from DeepSeek-R1, on benchmark datasets spanning mathematics, science, and question answering. Our results suggest that the distribution of additional reasoning paths generated by our algorithm is heavy-tailed and that substantial capacity to overcharge remains even under the best possible audit designed to keep the false-positive rate below \alpha = 0.1 .
[AI-114] Coding Agents Arent Enough! Evaluating an Enterprise Security Brain for Agent ic Cloud Investigations
链接: https://arxiv.org/abs/2609.30345
作者: Leon Goldberg,Gal Engelberg,Eden Yavin,Elad Elouz,Ariel Zadok,Konstantin Koutsyi
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: 13 pages, 7 tables
Abstract:Cloud-security investigation is dominated by population tasks: which identities can read a data store, how many resources fail a control, which assets are reachable from another account. These resolve against a complete inventory, not a named object. A partial answer to one is not a partial result. It is a different result. General-purpose coding agents can now be given read-only cloud credentials and asked to investigate directly, which raises the question of what a purpose-built security context layer still contributes. We evaluate the Sola Security Brain, a security intelligence layer whose relational substrate is resolved offline and whose security logic is evaluated against it at query time, against Claude Code operating the same live AWS environment through a read-only CLI, over 28 cloud-security investigation tasks. Answers are scored by a blinded, tier-weighted, grounding-gated relative recall over the joint claim pool, averaged across three independent grading draws. The Sola Security Brain reaches 0.693 coverage against 0.387, a gap of 0.306 that varied by \pm 0.018 across three grading draws, or a relative gain of 79.2% . It leads on 25 of 28 tasks from the weaker model tier, at 17.7\times lower reasoning cost per task and 31.6\times lower cost per unit of coverage. Beyond the aggregate, we describe an answer-level pattern we term sample-and-generalise: under a turn budget the live agent enumerates a fraction of a large population, asserts an unhedged universal negative, and discloses the sample size only in answer metadata rather than in the answer. In one task it reported that no bucket policies exist after checking four bucket families, in a sweep that sampled 40 of roughly 5,000 buckets, in an account where 65 buckets carry a wildcard-principal read grant.
[AI-115] Bridging LLM Agents and Data Spaces: An Architectural Mediation Approach using the Model Context Protocol
链接: https://arxiv.org/abs/2609.30341
作者: Jaime Alonso Ruiz,Carlos Aparicio,Gabriel Huecas,Joaquín Salvachúa,Andres Munoz-Arcentales
类目: Artificial Intelligence (cs.AI); Databases (cs.DB)
备注:
Abstract:Data Spaces enable sovereign and governed data sharing across organizational boundaries, but their integration with AI agents remains challenging due to mismatches between probabilistic language model interactions and policy-driven data infrastructures. This article presents an architectural mediation approach based on the Model Context Protocol (MCP), implemented through the Eunomia Agent, to enable controlled interaction between large language model (LLM) agents and data space services. The proposed mediation layer translates data space capabilities into structured, schema-driven tools that AI agents can discover and invoke while preserving governance constraints. A prototype implementation validates end-to-end interaction across catalog discovery, metadata retrieval, and data service invocation without modifying existing data space components. Results demonstrate that protocol-based mediation enables interoperable and standards-aligned integration of AI agents into data space ecosystems. The approach provides practical guidance for organizations seeking to introduce AI-driven automation into governed data-sharing environments while maintaining compliance, interoperability, and architectural separation of concerns.
[AI-116] What Will Remain Human in Software Architecture? A Focus Group Report
链接: https://arxiv.org/abs/2609.30334
作者: Uwe van Heesch,Olaf Zimmermann,Christian Kohls
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: 21 pages, EuroPLoP 2026, no figures
Abstract:AI development agents are increasingly used to support and partially automate software architecture tasks. To explore how practitioners perceive this shift, specifically what changes, what remains, and what new responsibilities emerge, we conducted a focus group at the 31st European Conference on Pattern Languages of Programs, People, and Practices (EuroPLoP 2026). Twenty-two participants from industry and academia discussed current practices, trust and validation strategies, the boundaries of AI autonomy, governance challenges, and implications for education. Among others, we found broad consensus that architectural decision-making, accountability, and the authoring of architectural guardrails remain fundamentally human tasks. A central emergent concept was harness engineering: the discipline of building the system that governs AI-assisted system creation, comprising validation mechanisms, knowledge lay- ers, and company-specific standards. The participants agreed that criticality, understood as the combination of uncertainty and cost of change, serves as the universal criterion for calibrating human oversight. A further concern was cognitive debt: the progressive erosion of human understanding of the system when AI-assisted decisions are accepted without full intellectual engagement. In this report, we present the findings of the focus group and describe directions for future work.
[AI-117] ScopeBench: Do Agents Preserve Engagement Boundaries Under Goal Pressure?
链接: https://arxiv.org/abs/2609.30325
作者: Shane Caldwell,Max Harley,Ads Dawson,Michael Kouremetis,Vincent Abruzzo,Will Pearce
类目: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注: 18 pages, 1 figure, 6 tables. Accepted at AISec 2026. Code and tasks: this https URL . Trajectories: this https URL . Leaderboard: this https URL
Abstract:Agents are increasingly deployed with real autonomy in web application and network penetration testing, where a single out-of-scope action can breach a client’s engagement boundary. Existing offensive-security benchmarks measure raw hacking capability; as those benchmarks saturate, the real barrier to deployment is a special case of alignment: scope adherence. We introduce ScopeBench, a benchmark of 30 dead-end agentic security tasks in which the stated objective is reachable only by violating the stated scope. Each task appears under two conditions that share an environment, verifier, and objective and differ only in scope: one instruction set has no scope and measures capability; the other has a natural-language scope to measure adherence. Scopeless trajectories are graded by a standard deterministic verifier. Scoped trajectories pass through two grading arms. First, the same deterministic verifier checks for the flag: because the flag sits behind the scope boundary, a pass proves by construction that a forbidden action occurred, yielding a high-precision lower bound on the violation rate. If the verifier does not pass the trajectory, an agentic judge estimates whether an out-of-scope call occurred. We calibrate the judge against 100 ScopeBench trajectories labeled call-by-call by human annotators, and a blinded audit of the evaluated rollouts finds its high recall holds - no false negatives among the 36 audited violations, with over-flagging its only observed error. Across 8 models in one harness, raw capability spans 12.2% to 81.1% and scope adherence spans 34.4% to 86.7%, with the judge finding 331 violations that mechanical verification misses. Opus-4-8 achieves a raw-capability score 10 percentage points higher than sonnet-4-6’s while exhibiting 35.6 percentage points higher scope adherence. We release the frozen pilot benchmark, evaluation code, and all 2160 ATIF trajectories.
[AI-118] Bringing AI to Autonomous Systems – From Cognition to Collective Intelligence
链接: https://arxiv.org/abs/2609.30291
作者: Joseph Sifakis
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:The purpose of this article is to highlight the central role of autonomous systems as the ultimate stage in the development of AI, to explain the underlying technical challenges that require a combination of connectionist AI and symbolic AI, and to integrate AI and systems engineering. We present a comprehensive framework for the design and evaluation of autonomous systems, based on a generic agent architecture that characterizes their behavior as the composition of cognitive functions organized around a long-term memory containing the agent’s evolving knowledge. We address the challenges posed by the implementation of the fundamental features of the agent architecture, in particular the link between sensory data and structured data stored in memory, decision-making related to the achievement of the agent’s goals and their planning, as well as the coordination of agents to combine individual and collective intelligence. We explain that agent trustworthiness, unlike that of traditional systems, is not limited to behavioral properties. It includes an essential dimension related to cognitive properties, the validity of which depends on how the agent uses its knowledge in decision-making. We present avenues for the development of methods for evaluating agent trustworthiness. We conclude with a critical assessment of the substantial gap between the aspirational vision of autonomous multi-agent systems and the current state of the art.
[AI-119] When Does Advection-Aware Graph Nowcasting Help? A Controlled Study of Distributed Solar Ramp Forecasting with a Self-Supervised Cloud-Motion Estimator
链接: https://arxiv.org/abs/2609.30286
作者: Phillip Jiang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Short-term forecasting of cloud-induced power ramps across a network of distributed photovoltaic (PV) or irradiance sensors is a recognised pain point for grid operators. A natural idea is to make the graph neural network (GNN) advection-aware: connect each site to the sites upwind of it, with edge time-lags set by the cloud-motion vector (CMV), so that a ramp is propagated forward before it physically arrives. Using a controlled synthetic testbed with a known wind field, we show that (i) with a realistic cross-correlation CMV estimate, an explicit advection graph does not beat a plain static or learned-adjacency spatiotemporal GNN; (ii) roughly half of the benefit available from a perfect CMV comes simply from providing an accurate motion vector as an input feature, not from graph structure; and (iii) advection helps only when the advective displacement over the forecast horizon, v*H, fits inside the sensor network. Motivated by (ii), we introduce a small self-supervised cloud-motion estimator – a position-aware encoder trained only on a multi-lag optical-flow reconstruction objective with an annealed kernel – that recovers the true wind vector to 2-4 degrees median angular error, 2-4x better than the classical cross-correlation method across every wind regime. Freezing this estimator and feeding its vector to the forecaster closes about 60% of the oracle-CMV RMSE gap at moderate wind (8-15% RMSE reduction over no advection), with no external wind data. We also report a negative result for a spatially-coherent probabilistic head. All claims are established on a single synthetic simulator; we discuss why real-network validation is the necessary next step and outline it.
[AI-120] ENAS: An Efficient Hardware-Aware Neural Architecture Search Framework for TinyML on Resource-Constrained Microcontrollers
链接: https://arxiv.org/abs/2609.30272
作者: Mohd Moin Khan,Naman Srivastava,Pandarasamy Arjunan
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:We present \textbfENAS, a hardware-aware Neural Architecture Search (NAS) framework that combines a static feasibility check, a cell-based search space supporting standard, depthwise-separable, and bottleneck blocks with optional skip connections, and a three-stage hybrid search strategy (random \rightarrow top- K \rightarrow mutation) with persistent cross-run caching. Unlike many existing NAS frameworks that rely on GPU acceleration, ENAS is designed to operate efficiently without requiring GPUs, making it suitable for resource-constrained development environments. We evaluate ENAS on two TinyML benchmarks, Visual Wake Words and Melanoma Cancer, across eight microcontrollers with memory footprints ranging from 20,KB to 1,MB SRAM and nine input image resolutions. Our experimental results show that ENAS achieves mean search-time speedups of 2.41\times and 1.70\times on the Visual Wake Words and Melanoma Cancer datasets, respectively, while maintaining competitive test accuracy compared with the recent NanoNAS framework. A measured resource analysis further shows that ENAS-selected models use substantially lower peak activation RAM, the binding constraint for microcontroller deployment at matched accuracy. Additionally, ENAS achieves 79.4% test accuracy on an STM32H743-based microcontroller, outperforming the greedy CPU-only baseline by 2.6 percentage points. We release the ENAS framework as open-source at: this https URL
[AI-121] Statistical attribute alignment for black-box generative AI via output post-processing
链接: https://arxiv.org/abs/2609.31607
作者: Kevin Jiang,Morgane Austern,Edgar Dobriban,Jason M. Klusowski
类目: Methodology (stat.ME); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Statistics Theory (math.ST)
备注:
Abstract:Generative AI systems are increasingly used, but aligning their outputs with user requirements poses a continuing challenge. Here, we aim to ensure that the distribution of an attribute of an AI-generated output aligns with a user-specified target. This is motivated by examples such as fairness, where we want to ensure that a protected attribute (e.g., gender, race, or age categories) follows a desired distribution, and synthetic data generation, where we want the generated data to be representative of a target distribution. We study the practically important black-box access setting, where a user can repeatedly query a generative AI model. The goal is to return m\ge 1 outputs whose joint attribute distribution is as close as possible to this target. For both exact and approximate alignment, we develop algorithms that minimize the expected number of queries to the generator, and we further demonstrate their optimality as the number of requested outputs m \rightarrow \infty . Experiments on text-to-image generation and geocoded persona generation tasks show that our post-processing algorithms improve statistical attribute alignment, complementing prompting-based interventions.
[AI-122] Agent ic Limit Order Books: Phase Transitions and Market Impact
链接: https://arxiv.org/abs/2609.31260
作者: Jan Rosenzweig
类目: Trading and Market Microstructure (q-fin.TR); Artificial Intelligence (cs.AI); Computational Finance (q-fin.CP)
备注:
Abstract:We investigate the systemic macroscopic dynamics emerging from Limit Order Books (LOBs) populated exclusively by autonomous reinforcement-learning agentic traders. By formalizing agent interactions within a microscopic order-matching engine, we examine two fundamental quantitative phenomena: equilibrium phase transitions in order flow regime shifts, and the structural dynamics of market impact. We show that agentic LOBs exhibit distinct phase boundaries separating orderly price discovery from hyper-volatile cascade states, governed by critical thresholds in the number of agents and observable market depth. Furthermore, we demonstrate that market impact under agentic liquidity provision deviates from classical square-root dynamics, exhibiting distinct dissipative, balanced, and non-dissipative regimes under non-linear feedback loops.
[AI-123] Adaptive Pilot Selection for Unified Semantic Communication and Semantic Sensing in ISAC
链接: https://arxiv.org/abs/2609.30891
作者: Muhammad Abubakar Rashid,Muhammad Hannan Akram,Haejoon Jung,Syed Ali Hassan
类目: ignal Processing (eess.SP); Artificial Intelligence (cs.AI)
备注:
Abstract:Semantic communication (SemCom) and integrated sensing and communication (ISAC) are promising technologies for future 6G wireless networks. Existing studies have applied semantic technology to either the communication module or the sensing module of ISAC. In this work, we propose SemISAC, which performs both SemCom and semantic sensing within a single dual-function waveform. SemISAC uses a joint semantic encoder that extracts task-specific information for both communication and sensing. We evaluate SemISAC in a vehicular scenario in which vehicles share pixel-wise segmentation of the road environment and, through sensing, classify surrounding objects and estimate their ranges. On the transmitter side, a deep learning encoder converts the input road-scene image into semantic symbols and places them on the data cells of an OFDM grid, while the remaining cells serve as pilots for channel state information estimation and sensing. The pilot configuration is adaptively optimized based on the channel conditions to balance communication and sensing requirements. At the receiver, a deep learning model reconstructs the segmentation from the received waveform, while the transmitting vehicle captures the reflected waveforms from surrounding objects and uses task-specific deep learning decoders for target recognition and range estimation. Simulation results show that SemISAC achieves a segmentation accuracy close to that of the dedicated SemCom module while outperforming both conventional and semantic baselines in target recognition and range estimation.
[AI-124] Warned alike AI agents avoid the less-crowded road while people take it
链接: https://arxiv.org/abs/2609.30883
作者: Takahiro Ezaki,Naoto Imura,Katsuhiro Nishinari
类目: Physics and Society (physics.soc-ph); Artificial Intelligence (cs.AI)
备注:
Abstract:AI agents built on a few shared models increasingly act for many people. A shared forecast about others can align their choices and change how scarce capacity is allocated. We tested this feedback in a two-road congestion game. Adding one sentence warning that others might follow a routing tip made populations of 50 GPT agents crowd one road while avoiding the nearly empty alternative. Average travel time rose from 64 to 95 min, although any crowded-road agent could have saved 69 min by switching alone. The warning discouraged the very move it predicted. The pattern persisted for 100 rounds. Two other model families shifted the same way without locking onto one road. Twelve all-human groups (240 participants) stayed near balance under numerical reports or the tip and warning. In 24 mixed groups with a further 240 participants, imbalance grew with the share of agents in the registered analysis, while people increasingly took the road the agents avoided. Collective costs stayed below the allagent reference, but with 15 agents and 5 humans, agent seats averaged 80 min, compared with 44 min for human seats. Shared forecasts can thus sustain collective inefficiency among similar agents. A better group average can also hide an unequal burden. Evaluations of AI agents that share resources should test populations, treat messages as interventions and report who bears the costs.
[AI-125] Understanding Perturbed Parameter Ensemble Sensitivities Using A Contrastive Learning Approach
链接: https://arxiv.org/abs/2609.30420
作者: Da Fan,David John Gagne II,Gregory S Elsaesser,Brian Medeiros,Addisu G Semie,Qingyuan Yang,Akila Sampath,Subashree Venkatasubramanian
类目: Atmospheric and Oceanic Physics (physics.ao-ph); Artificial Intelligence (cs.AI)
备注:
Abstract:Perturbed parameter ensembles (PPEs) reveal how physics parameters affect climate simulations, but interpreting parameter sensitivities across multivariate, spatially structured outputs remains challenging, particularly when calibrating models against observations. We develop an explainable contrastive learning model that maps 5 monthly cloud and radiation fields into a shared representation space. We train the model on the fields of two 100-member Community Atmosphere Model version 6 (CAM6) PPEs, spanning 34 parameters, that only differ in the warm rain microphysics scheme: KK2000, the default bulk microphysics scheme, and TAU-ML, a neural network emulator of a bin microphysics scheme. The learned representations separates two PPEs with over 94% linear classification accuracy while preserving the seasonal variability and ensemble spread due to parameter perturbations. In the shared representation space, the representations of satellite observations occupy the same low-dimensional manifold as the PPEs but are displaced from them most strongly during boreal spring and autumn. TAU-ML PPE has a lower distance to observations compared to KK2000 in the representation space. Integrated Gradients attributions highlights the contributions in subtropical low-cloud regions, Northern and Southern Hemisphere storm track regions, and tropical convection regions to differences between PPEs and observations. Regional attributions correlate most strongly with parameters associated with cloud microphysics, boundary layer turbulence, and deep convection. These results demonstrate that explainable representations of climate fields can attribute model differences to specific variables, regions, seasons, and physical parameters.
[AI-126] Adaptive multi-resolution Gaussian processes: Scalable exact inference with naturally data-sparse covariance matrices
链接: https://arxiv.org/abs/2609.30348
作者: Yanchuang Cao,Jun Liu,Tengchao Yu,Heng Yong
类目: Machine Learning (stat.ML); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Gaussian processes constitute a cornerstone of probabilistic machine learning, yet scaling them to large datasets typically forces a trade-off between computational efficiency and model fidelity. This work bridges this gap by presenting an adaptive multi-resolution Gaussian process framework that is both scalable and exact. Our key innovation is constructing a naturally data-sparse covariance matrix with adaptive multi-resolution basis functions. These basis functions are directly anchored to samples, eliminating the need for auxiliary points. By shrinking the support domains of multi-resolution basis, the matrix block sizes are limited, guaranteeing sparsity. The inverse of the data-sparse covariance matrix is computed exactly and efficiently via the sparse Cholesky inverse algorithm. To further improve predictive uncertainties, we construct an augmented basis function. Theoretical analysis and numerical experiments demonstrate that our model achieves exact inference with \mathcalO(n \log^2 n) training cost and \mathcalO(\log^d n) prediction cost, establishing a principled framework for scalable and high-fidelity Gaussian process regression.
机器学习
[LG-0] Gap-free Differentially Private PCA for Gaussian Data
链接: https://arxiv.org/abs/2609.31614
作者: Alina Ene,Huy L. Nguyen
类目: Data Structures and Algorithms (cs.DS); Machine Learning (cs.LG)
*备注:
Abstract:We give a gap-free differentially private algorithm for the principal component analysis (PCA) problem with Gaussian data.
[LG-1] New LoRA Skills Should Read but Never Write
链接: https://arxiv.org/abs/2609.31600
作者: Zeyan Li,Panqi Yang,Qirong Guo,Shengda Zhuo,SIyuan Qiu,Hu Xu,Chun Li,Jianfeng Xu
类目: Machine Learning (cs.LG)
*备注:
Abstract:Low-rank adapters (LoRA) make it cheap to fine-tune a large language model once per task, but combining several independently trained adapters into one model remains difficult: merging the updates in weight space causes interference, retraining on all task data is expensive, and routing between separate adapters gives up the goal of a single combined model. We trace the difficulty to two choices that every composition method makes implicitly. A LoRA update admits infinitely many equivalent factorizations; the choice among them is invisible while an adapter serves alone, but it determines what a learned interaction between adapters can see. A coupling between an old skill and a new one can likewise point in either direction, and the direction decides whether the old skills keep computing what they computed before. We introduce READ (Read-only Expansion of Adapter Deltas), which fixes both choices: each adapter is rewritten into a balanced canonical form that preserves its update exactly, and the coupling grows in one direction only, so a new skill can read the input subspaces of old skills but cannot write into their output subspaces. The only trainable object at each append is the new skill’s row of the coupling matrix, and the composed update folds into the base weights with no inference cost, routing, or task-specific rules. We evaluate READ across four benchmark suites and two model families, adding skills one at a time. Across several families, READ improves every suite average over the strongest published baselines built from the same adapters—by more than twenty points on SuperGLUE and more than seven points on the domain suite—and nearly all complete addition sequences end above every direct baseline. Factor coordinates and coupling direction, which a lone adapter never exposes, are what decide whether composed skills survive.
[LG-2] Common-Mode Collapse and Recovery in Direct Feedback Alignment
链接: https://arxiv.org/abs/2609.31589
作者: Varun Reddy,Bernardo L. Sabatini,Houman Safaai
类目: Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
*备注:
Abstract:Direct feedback alignment (DFA) trains hidden layers through fixed random projections of output error. With tanh hidden units and independent sigmoid outputs, plain stochastic gradient descent can stall near the loss of a constant predictor of class frequencies. We trace this stall to the error’s common mode, the component shared across inputs. An exact mean-covariance decomposition separates a rank-one update formed by the mean teaching signal and mean presynaptic activity. Its leading component drives tanh units toward saturation. At initialization, random feedback provides no systematic correction of the shared error on average; readout learning limits its duration. A reduced model initialized from the network, without fitted parameters, predicts the concentration of activation sensitivity across 48 settings. On MNIST, class decodability largely survives collapse, but readout learning remains slow at a fixed learning rate. Adam learns faster despite deeper collapse. Calibrating the baseline readout to the class prior suppresses collapse and speeds learning; weaker feedback trades less collapse for slower learning. Replacing errors by their signs sustains collapse; subtracting the signal’s batch mean prevents sustained collapse and improves learning in the tested setting. Related effects occur in deeper and convolutional networks and on CIFAR-10, with severity and cost depending on the readout, optimizer and input statistics.
[LG-3] rust Guided Decision Transformer NEURIPS2026
链接: https://arxiv.org/abs/2609.31586
作者: Chainesh Gautam,Raghuram Bharadwaj Diddigi,Chandramouli Kamanchi,Pankaj Dayama,Sumanta Mukherjee,Kameshwaran Sampath
类目: Machine Learning (cs.LG)
*备注: To appear in Neurips 2026
Abstract:Decision Transformer performance degrades on long rollouts because the conditioning context drifts out of the training distribution. We show that this drift is visible through the model’s own next state prediction error, which rises during rollout and stays elevated, giving a direct signal of when context has become unreliable. We introduce Trust Guided Decision Transformer (TGDT), which selects context before applying value guidance. At each step, TGDT evaluates several recent context suffixes using rolling next state prediction error, calibrated against held out offline data via split conformal prediction. It keeps only suffixes whose error stays within the calibrated threshold, then uses a frozen critic to choose the highest value action among the trusted suffixes. This reverses the order used by value only elastic selection, where the critic may choose an action generated from a context the model itself has flagged as unreliable. Experiments on D4RL navigation and locomotion tasks show that state prediction, critic guidance, and hard context reset each solve only part of the problem. TGDT reduces persistent high error runs and improves return over vanilla Decision Transformer, reset based context control, and value only context selection.
[LG-4] Weight Pair Encoding: Inducing a Smaller Grammar in Neural Network Weights
链接: https://arxiv.org/abs/2609.31564
作者: Irene Tallini,Daniele Solombrino,Alberto Cazzaniga,Emanuele Rodolà
类目: Machine Learning (cs.LG)
*备注:
Abstract:We show that neural network weights can be explicilty fintuned to admit a smaller grammar. Weight Pair Encoding (WeightPE) does so by placing a lossy Re-Pair compressor inside a straight-through estimator. The int8 weights of the network are flattened into one string, and near-matching Re-Pair patterns are made exactly equal within a global L2 budget. The network computes with the rewritten weights and trains through them with a straight-through estimator. Unlike a flat codebook of fixed-size entries, a grammar offers variable-length patterns and reuses them hierarchically inside larger ones. On the MLP weights of ViT-B/16 and ViT-L/16 finetuned on CIFAR-10, WeightPE produces a Re-Pair grammar 0.43x and 0.38x the size of the one produced by an equivalent int8 QAT run, at a cost of 1.9 and 1.1 accuracy points. The trend extends to different grammar compressors (LZ78, SEQUITUR), over which the networks has not be finetuned against. To our knowledge, this is the first time grammar size has been used as an explicit training objective for network weights.
[LG-5] Generalization behavior of OPTQ and the role of regularization
链接: https://arxiv.org/abs/2609.31560
作者: Erin George,Rayan Saab
类目: Machine Learning (cs.LG)
*备注:
Abstract:Large neural networks can be compressed by rounding or “quantizing” their weights to numbers that admit representations with fewer bits. One algorithm for quantization, OPTQ, progressively quantizes the weights of a neural network so that the squared quantization error on a specified calibration dataset is as small as possible. We study the performance of OPTQ and a variant algorithm, stochastic OPTQ, in a generalization setting and derive bounds for the expected squared error accrued by the algorithm when a test point is drawn from a fixed distribution. We prove two results. One result relates the generalization error to the error on a calibration dataset comprising independent samples from the same distribution as the test distribution. The other result bounds the generalization error of stochastic OPTQ for all sufficiently nice distributions, regardless of the calibration dataset. In both of these results, the regularization term \lambda plays an important role. We use insights from these results to make a new recommendation for the choice of \lambda and see that this choice of \lambda preforms favorably in experiments when compared to prior recommendations in the literature.
[LG-6] Online Learning via Learned Latent Bayesian Tracking NEURIPS2026
链接: https://arxiv.org/abs/2609.31559
作者: Guy Gerson,Tomer Raviv,Nir Shlezinger,Tirza Routtenberg,Osvaldo Simeone
类目: Machine Learning (cs.LG); Signal Processing (eess.SP)
*备注: Accepted at NeurIPS 2026
Abstract:Online learning in non-stationary environments requires models to adapt rapidly from streaming data under strict computational constraints. A principled approach casts online learning as Bayesian state tracking, where model parameters are updated sequentially via Bayesian filtering. However, applying Bayesian filters directly to modern deep models is computationally prohibitive due to the high dimensionality of parameter space, forcing existing methods to rely on restrictive approximations or manually designed low-dimensional subspaces. In this work, we identify the absence of a suitable low-dimensional dynamical representation as the core bottleneck in Bayesian filtering-based online learning. Accordingly, we propose Adaptive Update through Representation Adaptation (AURA), a meta-learning framework that learns offline a low-dimensional latent state-space model governing the evolution of optimal model parameters under distribution shift. Online adaptation is then performed via extended Kalman filtering in this learned latent space followed by reconstruction of the full model parameters through a learned lifting map, enabling efficient single-step online adaptation while preserving model expressiveness. Evaluated on online adaptation of neural wireless receivers under time-varying channels and on non-stationary image classification, AURA shows substantial improvements in adaptation speed, accuracy, and computational efficiency over existing online learning and Bayesian filtering baselines, demonstrating that an adaptation-aware latent geometry is beneficial for effective Bayesian online learning in high-dimensional models.
[LG-7] EAServe: Encode-Aware Disaggregated Serving for Multimodal Large Language Models
链接: https://arxiv.org/abs/2609.31551
作者: Kunxiong Zhu,Zhihao Shu,Hangyu Zheng,Minghai Qin,Miao Yin,Gagan Agrawal,Wei Niu
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG); Performance (cs.PF)
*备注: 13 pages, 12 figures, 7 tables. Accepted to PACT 2026
Abstract:Disaggregating the two stages, Prefill and Decode, onto separate GPU pools is now a standard optimization for (text-only) LLM serving. However, multimodal LLMs (MLLMs), which add a third phase, Encode, pose new challenges for resource allocation. Encode turns images, video, or audio into embeddings that the language model can consume, yielding a three-stage Encode-Prefill-Decode (EPD) pipeline. Existing frameworks offer only partial answers: text-only PD systems lack Encode, while EPD frameworks expose it as a separate service without regulating downstream request flow. The pipeline also carries a structural resource imbalance: every request enters through Encode before downstream work can begin, yet per-request execution leaves the encode GPU severely underutilized even at high loads, starving the downstream Prefill and Decode workers. Addressing this, we reposition Encode as the control point of the EPD pipeline, exposing three tightly coupled dimensions: when work enters downstream, where prefill executes, and how the GPU is shared. We instantiate this in EAServe across two co-designed layers. Its runtime manages load-adaptive micro-batching, rate-controlled partial offload to a co-resident prefill worker, and dynamic SM partitioning for predictable co-location. The configuration layer, Hybrid Auto Selection (HAS), navigates the joint space of GPU allocation, encode batch size, and offload ratio by pruning unbalanced allocations with per-stage capacity profiling and refining the remainder through TPE-based Bayesian optimization. Evaluated on three MLLM architectures spanning image, video, and audio, EAServe delivers up to 4.3x and 1.7x higher goodput than NVIDIA Dynamo and vLLM, respectively, under identical SLO constraints, sustains more balanced and higher GPU utilization across the EPD pipeline, and reaches near-optimal configurations faster than baseline search methods.
[LG-8] BeatGraph: Self-Supervised Heartbeat Graphs for Infant ECG Representations from the Home Environment
链接: https://arxiv.org/abs/2609.31546
作者: Mohammad Nur Hossain Khan,M. S. Krafczyk,Beverly G. Bolster,Nancy McElwain,Mark A. Hasegawa-Johnson,Bashima Islam
类目: Machine Learning (cs.LG)
*备注:
Abstract:Electrocardiogram (ECG) foundation models typically tokenize the signal into fixed-length patches that ignore cardiac structure, so a patch may split a heartbeat and the number of beats in each patch shifts with heart rate. This matters most for infants, whose heart rates are higher and whose ECG differs from the adult, clinic-recorded 12-lead data these models are built on. A model for infant ECG should therefore reason about heartbeats directly rather than recover them from arbitrary patches. We propose BeatGraph, which makes the heartbeat its unit of representation, modeling each 30-second window as a graph of beats. A shared beat encoder embeds each heartbeat from its waveform and inter-beat intervals, a Transformer with positional encoding orders the beats in time, and residual graph attention layers relate every beat to every other before attention pooling yields a window embedding. We pretrain BeatGraph on our new corpus of unlabeled infant recordings by predicting masked-beat embeddings, then fine-tune it for each task. One backbone supports sleep-wake detection, infant-state classification, activity-source identification (infant- or caregiver-initiated movement), and affect recognition, improving macro-F1 over the strongest baseline on each task by 0.076 to 0.158. It also transfers across age groups, reaching 0.892 AUROC on the ZZU-pECG pediatric benchmark (ages 0 to 14), within 0.001 of the best published self-supervised ECG model, and matching that model under linear evaluation on the adult PTB-XL benchmark despite infant-only pretraining. Finally, to our knowledge, we release the first public infant ECG corpus collected in homes, classrooms, and laboratory settings with state and affect labels. It contains 3,408 hours of single-channel ECG from 143 infants aged 3 to 11 months, with unlabeled pretraining data, benchmark tasks, and subject-level splits.
[LG-9] NEXT: Physics-Informed Neuro-Spectral Exponential Time Differencing Architectures
链接: https://arxiv.org/abs/2609.31539
作者: Márcio Marques,Leonardo Mendonça,Leonardo M. Moreira,Christian Júnior de Oliveira,Vitor Balestro,Tiago Novello,Daniel Yukimura,Pavel Petrov,Lucas Nissenbaum
类目: Machine Learning (cs.LG); Numerical Analysis (math.NA)
*备注:
Abstract:Physics-Informed Neural Networks (PINNs) build neural representations of time-dependent PDE solutions, naturally incorporating physics knowledge and observational data, which makes them well suited to both forward and inverse PDE problems. PINNs, however, are known to suffer from spectral bias and lack of causality. Neuro-Spectral Architectures (NeuSA), a recently proposed alternative to PINNs, mitigate both issues, but their numerical integration becomes unstable for stiff differential equations arising in many relevant physical problems. This study proposes Neuro-Spectral Exponential Time Differencing Architectures (NEXT), which combines the spectral representation of the PDE solution in NeuSA with high-order exponential integrators. Within this approach, the linear stiff part of the vector field induced by the PDE is integrated exactly through matrix exponentials, while the possibly nonlinear remainder is modeled by a neural network. The effectiveness of NEXT is verified through benchmark experiments on a set of stiff PDEs, in which NEXT is stable and accurate while NeuSA diverges numerically. It is also shown that NEXT can be applied to inverse problems, where the model has to learn unknown parameters or boundary conditions from sparse data. All code used in this work is publicly available at: this https URL .
[LG-10] HySTAR: Anchored Hypergraphs for Stable Credit Assignment in Cooperative Multi-Agent Reinforcement Learning
链接: https://arxiv.org/abs/2609.31531
作者: Xinglong Luo,Yuding Zhang,Yuheng Kuang,Shuxuan Yuan,Zhenni Zeng,Weiqiang Zhu,Zhenhai Ji,Zhengning Wang
类目: Machine Learning (cs.LG)
*备注:
Abstract:Cooperative multi-agent reinforcement learning under partial observability and shared rewards requires assigning team outcomes to individual agents and high-order coalitions. A MAPPO-style critic compresses joint behavior into one global value, while critics that dynamically reconstruct the grouping topology change the mapping from agents and coalitions to value components as interactions or active agents evolve. We refer to this inconsistency as structural target drift. We introduce HySTAR, a MAPPO-based framework that separates adaptive representation learning from a temporally consistent high-order value-decomposition basis. HySTAR anchors an overlapping sparse hypergraph as a uniformly covered decomposition scaffold, uses a spatiotemporal encoder to represent physical and task-dependent interactions, and combines temporal and structural relevance to construct agent-specific advantages. Experiments on SMAC, GRF, Traffic Junction, and MPE demonstrate consistent improvements over MAPPO-style, value-factorization, and dynamic-grouping baselines. On the hardest SMAC settings, HySTAR achieves relative gains of 16.7% over MAPPO and 15.6% over HYGMA, ranks first on all six GRF scenarios, reduces Traffic Junction convergence epochs by up to 40.2% relative to MAGIC, and obtains the highest MPE episode rewards. Controlled topology, agent-death, neighborhood, and parameter analyses support the benefit of anchoring the decomposition scaffold while adapting the propagated representations.
[LG-11] Scaffold: Support Graph Theory Based Sparsification for Graph Neural Networks
链接: https://arxiv.org/abs/2609.31466
作者: Siddhartha Shankar Das,Sai Karthik Navuluru,S M Ferdous,Ryan A. Rossi,Baris Coskunuzer,Lakshman Tamil,Edoardo Serra,Alex Pothen,Robert Rallo,Mahantesh M Halappanavar
类目: Machine Learning (cs.LG)
*备注:
Abstract:Graph neural networks (GNNs) rely on message passing over graph edges, making their computational and memory costs strongly dependent on graph density. Graph sparsification offers a natural way to reduce these costs, but removing edges indiscriminately can distort important communication structure and degrade predictive performance. We introduce Scaffold, a topology-based, unsupervised graph sparsification framework derived from support graph theory preconditioners. Scaffold explicitly controls two complementary structural quantities: dilation, which measures the length of rerouting paths induced by removed edges, and congestion, which measures how strongly these rerouted paths concentrate on the retained support. By jointly controlling dilation and congestion, Scaffold preserves short communication paths while avoiding structural bottlenecks. To our knowledge, Scaffold is the first scalable GNN sparsification framework to use a joint supporting-path dilation-congestion criterion. Across 19 homophilic and heterophilic benchmarks spanning small to large graphs, Scaffold achieves the best aggregate rank among the evaluated sparsification and related methods. Using only 10%-50% of the original edges per sparse support, Scaffold recovers or closely approaches full-graph GNN performance while using less than half the memory of full-graph training and reducing end-to-end training time, including sparsification overhead. We provide an open-source software package at this https URL.
[LG-12] Evaluating the accuracy of KV cache reuse techniques
链接: https://arxiv.org/abs/2609.31415
作者: Samuel Cestola,Tianxiang Xia,Pengfei Zheng,Weiyan Zheng,Bo Wang,Yi Zhao,Diego Didona
类目: Machine Learning (cs.LG)
*备注:
Abstract:Position-independent KV cache reuse aims to reduce latency in retrieval-augmented generation by reusing chunk-level KV caches across prompts. We show that current evaluations of KV cache reuse techniques rely on measurements that fail to faithfully capture the loss of accuracy attributable to reuse, often artificially inflating the reported effectiveness. We also show that existing datasets do not exhibit the reuse dynamics needed to thoroughly evaluate such techniques. To address these issues, we propose an evaluation methodology that measures this accuracy loss without ambiguity and we introduce Boxoffice, a tool that programmatically generates evaluation datasets that exercise challenging KV cache reuse patterns.
[LG-13] AFA-Net: A Differential Attention Approach for Auditory Attention Detection ICASSP2027
链接: https://arxiv.org/abs/2609.31402
作者: Philip H. Lee,Shreeram Suresh Chandra,Karan Thakkar,John H.L. Hansen
类目: ound (cs.SD); Machine Learning (cs.LG); Signal Processing (eess.SP)
*备注: Submitted to ICASSP 2027
Abstract:Auditory Attention Detection (AAD) utilizes electroencephalographic (EEG) signals to identify a target speaker in a multi-speaker environment. Despite considerable progress, existing deep learning architectures often lack explicit mechanisms for handling noisy EEG data. To address this limitation, we propose Auditory Focus Attention Networks (AFA-Net), a machine learning framework that replaces vanilla attention with a simple yet flexible differential attention mechanism to help focus on task-relevant neural activity. AFA-Net achieves an upward accuracy of 96.8% at the 2s decision window, while using substantially fewer parameters than most existing methods. To the best of our knowledge, AFA-Net is among the first frameworks to explicitly try to combat EEG noise to improve AAD.
[LG-14] Decodable In-Context State and Model Output Across Training
链接: https://arxiv.org/abs/2609.31401
作者: Manas Venkata Sai Ravulapalli,Samrath Singh Chadha
类目: Machine Learning (cs.LG)
*备注:
Abstract:Prior work established that a probe can decode an in-context binding on model errors and that probe-guided steering can repair some of them. We follow probe accuracy, model output, and steering response across public pretraining and post-training checkpoints. Probe accuracy rises during Pythia pretraining, while probe-guided steering moves from negligible all-trial benefit to a larger benefit at two model sizes. Saved scores distinguish probe-correct errors with low and above-uniform model probability for the correct candidate. Oracle-target steering already repairs many early errors, but saved aggregates cannot separate target quality from intervention sensitivity. A held-out comparison of decoders trained on the final state or candidate logits finds no detected final-state advantage on late-checkpoint model errors. An information-theoretic counterexample explains why decodability on errors alone cannot establish discarded output information. The connection to downstream omissions remains open.
[LG-15] Differential Attention Unlocks Complementary EEG and Speech Fusion for Emotion Recognition ICASSP
链接: https://arxiv.org/abs/2609.31399
作者: Philip H. Lee,Shreeram Suresh Chandra,John H.L. Hansen
类目: Machine Learning (cs.LG); Signal Processing (eess.SP)
*备注: Submitted to the 2027 ICASSP-OJSP track
Abstract:Multimodal emotion recognition (MER) increasingly pairs EEG with speech, treating internal neural signals and external vocal expression as informative views of affect. In practice, naive fusion underperforms the stronger single modality, because EEG artifacts inject noise that corrupts the shared representation. We introduce EmoSpeechBrain, a multimodal framework built on the insight that noise suppression is a precondition for effective fusion. Its EEG encoder uses differential attention, taking the difference between two attention maps to cancel shared noise and isolate discriminative neural activity. An attention-based gating adapter aligns both modalities in a shared space and weights each one’s contribution to the prediction. On two datasets - PME4 and EAV, EmoSpeechBrain improves MER accuracy by up to 12.9% over other state-of-the-art (SOTA) EEG encoders, and surpasses unimodal speech and EEG baselines by up to 13.1% and 23.1%. These results show that once EEG noise is suppressed, fusion delivers gains that naive combination cannot.
[LG-16] owards Understanding LLM -Based Log Anomaly Detection: An Empirical Study of Performance Efficiency and Robustness ICASSP2027
链接: https://arxiv.org/abs/2609.31371
作者: Bin Li,Dongdong Wang,Siyang Lu
类目: Machine Learning (cs.LG); Cryptography and Security (cs.CR)
*备注: 6 pages, 2 figures, 3 tables. Submitted to IEEE ICASSP 2027
Abstract:Large language models (LLMs) have demonstrated promising performance in log anomaly detection, yet how their adaptation strategies, architectures, and deployment configurations affect detection effectiveness remains insufficiently understood. To investigate these factors, we conduct a systematic empirical analysis across three public log datasets, examining different adaptation strategies, model architectures, parameter scales, and quantization settings. Our results reveal substantial performance differences across adaptation strategies, while model scaling yields varying detection gains across datasets. We further observe that models with comparable detection accuracy can exhibit markedly different computational costs, and that low-bit quantization largely preserves detection performance in the evaluated configurations. Finally, we examine detection robustness under structural, semantic, and label noise at different perturbation levels. These findings provide empirical insights into the performance, efficiency, and robustness of LLM-based log anomaly detection, highlighting practical considerations beyond conventional accuracy-oriented evaluation.
[LG-17] Brenier Meets Adversarial Training: Optimal Transport Geometry for Robust Learning
链接: https://arxiv.org/abs/2609.31363
作者: Alireza Abdollahpoorrostam,Ehsan Sharifian,Buse Şen,Marco Cuturi,Daniel Kuhn
类目: Machine Learning (cs.LG); Optimization and Control (math.OC); Machine Learning (stat.ML)
*备注:
Abstract:Distributionally robust optimization (DRO) provides a principled framework for learning under distribution shift, but its practical use is hindered by the difficulty of evaluating worst-case risks for nonconvex loss functions. We study a penalized DRO formulation in which the adversary may choose any distribution but incurs a Wasserstein penalty for deviating from the empirical distribution. We show that the adversary’s problem can be reformulated as an optimization problem over transport maps that push empirical samples to adversarial ones, and we prove that optimal maps are cyclically monotone. We also show that standard adversarial training—based on per-sample local optimization—violates cyclical monotonicity and wastes transport costs unless the adversary is severely restricted. We propose two remedies. First, we introduce multi-start particle ascent, which alternates parallel gradient ascent with reassignment to enforce cyclical monotonicity across samples. Second, we parameterize adversarial maps as gradients of input-convex neural networks, which guarantees cyclical monotonicity by construction. Experiments on robust regression, image classification, and robust control show that our methods consistently outperform standard adversarial training and state-of-the-art baselines, achieving improved robustness and better generalization under distribution shift.
[LG-18] Progressive Memory Transformer: Memory-Aware Attention for Time-Series NEURIPS2026
链接: https://arxiv.org/abs/2609.31351
作者: Tord Sture Stangeland,Andreas Köhler,Steffen Mæland,Adín Ramíres Rivera
类目: Machine Learning (cs.LG)
*备注: To appear in NeurIPS 2026
Abstract:Time-series carry structure simultaneously at multiple scales (fine-grained variation, mid-range motifs, and global properties) and downstream tasks operate at correspondingly different scales. Most existing self-supervised learning approaches supervise representations globally via instance-level contrastive losses and limited temporal neighborhood supervision, but do not explicitly exploit the structural hierarchy. We propose a learning framework that explicitly enforces a structural hierarchy across three scales independently: a local objective for token continuity, a mid-range objective for window-level motifs, and a global objective for sequence-level agreement. Realizing this framework requires the backbone to expose a representation at each scale; we introduce \textbfProgressive Memory Transformer (PMT), which augments a transformer with writable, window-aligned memory that exposes the mid-range scale alongside the token and sequence-level representations conventional transformers already provide. Across seven UCR/UEA/UCI classification benchmarks, a cue-retention probe, and forecasting benchmarks, PMT learns representations that probe well at the global, mid-range, and local scales—strong low-label classification (1–5% labels), competitive forecasting performance across multiple horizons, and quantitative and qualitative evidence that memory states capture mid-range motifs.
[LG-19] Bridging Body and Brain: Gene-Driven Morphology–Control Co-Design
链接: https://arxiv.org/abs/2609.31329
作者: Fu Feng,Ruixiao Shi,Yucheng Xie,Jing Wang,Xin Geng
类目: Machine Learning (cs.LG)
*备注:
Abstract:Morphology–control co-design jointly optimizes an agent’s body structure and control policy as an integrated embodied system. However, existing methods typically model morphology design and control with separate networks coupled only indirectly through a shared task objective, limiting explicit high-level coordination. Inspired by natural genes that coordinate biological development, we introduce \textbfMorphogene, a compact latent blueprint that bridges an agent’s body and brain. Through AdaConcat, Morphogene jointly conditions morphology and control generation at the limb level, allowing its variations to induce coordinated changes in both components. Building on this representation, we propose \textbfGeCode, which formulates co-design as exploration in the compact Morphogene space. Each Morphogene anchors a local design region in which nearby body–brain designs are explored, while performance-guided updates move these anchors toward promising regions for more efficient exploration of the broader design space. This process combines local refinement with global exploration while preserving body–brain compatibility. Extensive experiments across diverse 2D and 3D co-design tasks demonstrate that GeCode consistently outperforms existing state-of-the-art methods, achieving substantially faster convergence and higher final performance.
[LG-20] More Sensors Only One Field: Rethinking Continual Spatio-Temporal Forecasting
链接: https://arxiv.org/abs/2609.31325
作者: Lewei Xie,Haoyu Zhang,Jiajun Zhou,Yulong Chen,Guanxing Chen,Yu-An Huang,Hau-San Wong,Yifan Zhang,Zhi-An Huang
类目: Machine Learning (cs.LG)
*备注:
Abstract:Continual spatio-temporal forecasting supports traffic management and environmental monitoring under evolving dynamics and expanding sensor networks. However, conventional graph-based continual learning methods tie forecasting representations to the current sensor layout, so sensor expansion can alter the representation of learned spatial relationships. Our key insight is that sensor expansion changes the evidence available about a process without necessarily changing the dynamics to be learned. We propose STFO (Spatio-Temporal Field Operator), which parameterizes forecasting knowledge as a shared field-evolution operator and handles changing sensor layouts through observation and query interfaces. Normalized coordinate-based aggregation lifts irregular sensor histories onto a fixed latent grid, enabling reuse of learned spatial maps across observation sets without sensor-specific parameters. To accommodate process drift, a spectral descriptor summarizes variation across spatial scales and conditions Fourier propagation and attention to adapt operator responses to the current spatial regime. Coordinate-based decoding queries the evolved field at sensor locations and combines spatial corrections with local-history predictions. Experiments on PEMS-Stream, CA-Stream, and AIR-Stream demonstrate state-of-the-art average forecasting performance. STFO-Large reduces average MAE over DOL by 8.4% on PEMS-Stream and 4.7% on CA-Stream. Our code is available at this https URL.
[LG-21] LUCID: Learning Under Confounding for Inference and Discovery in Time Series
链接: https://arxiv.org/abs/2609.31315
作者: Mohammad Fesanghary
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 18 pages, 2 figure
Abstract:Unobserved common causes are pervasive in real-world time series and can induce spurious associations that causal discovery methods mistake for direct edges. We propose LUCID (Learning Under Confounding for Inference and Discovery, a regime-adaptive deconfounding layer that first estimates the confounding regime from data using a Marčenko–Pastur spectral router, then applies a deconfounding strategy matched to that regime. When the spectrum indicates pervasive factor confounding, LUCID attenuates factor-dominated variation and recovers contemporaneous (lag- 0 ) structure from the resulting innovations, with edge selection calibrated against a data-driven edge-free null. Rather than being tied to a particular discovery algorithm, it can wrap existing discovery engines; we demonstrate consistent improvements across three such methods. On a diverse synthetic out-of-distribution benchmark spanning changes in confounder strength and sparsity, loading density, lag structure, volatility dynamics, edge heterogeneity, persistence, intermittency, and tail behavior, LUCID achieves the best family-weighted directed, lag-resolved graph F_1 ( 0.60 ), improving over the strongest baseline by 0.19 absolute ( \approx!46% relative). Its advantage widens relative to looser lag-collapsed scoring, and remains robust under intermittent and heavy-tailed confounding. Code reproducing the method, the benchmark generators, and every reported experiment is available at this https URL.
[LG-22] Benchmarking Attention for Tabular Foundation Models
链接: https://arxiv.org/abs/2609.31306
作者: Maximilian Schambach,Clemens Biehl,Sam Thelin
类目: Machine Learning (cs.LG)
*备注:
Abstract:Tabular in-context learners such as TabPFN, Mitra, or ConTextTab rely on alternating row and column attention over 2D sequences of latent embeddings. These attention patterns differ markedly from the one-dimensional case in language models: row attention involves longer sequences while column attention operates on much shorter ones, and the strided memory layout of tabular data makes producing contiguous tensors costly. Moreover, the hidden dimensions used in current models are small compared to recent language models. Yet efficient attention has been studied mostly for one-dimensional sequences, leaving the two-dimensional tabular setting unexplored. To this end, we create a reproducible benchmarking setup and study the unique characteristics of tabular attention across several backends – Torch SDPA (efficient and cuDNN), FlashAttention-2/3/4, and the inference-only backends vLLM and SageAttention – measuring forward and backward throughput across realistic tabular shapes on three GPU generations (A100, H100, B200). We find that the optimal backend choice differs between column and row attention and varies across hardware as well as model specifics: While the FlashAttention implementations tailored for each GPU generation perform overall best, they are at times outperformed by CuDNN in the case of column attention at longer sequences with cross-over points depending on the head dimension. Among inference-only backends, SageAttention performs well for row attention and large sequences beyond 16,k rows. Our reproducible benchmark lays the foundation for future improvements to table-native attention. The self-contained benchmarking and evaluation code is openly available at: this https URL
[LG-23] Deterministic Regime Switching and Feasibility Inversion in Dynamic Tensor Rematerialization
链接: https://arxiv.org/abs/2609.31250
作者: Mahesh Reddy Pagadala
类目: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC)
*备注: 6 pages, 5 tables, 2 figures. Code and data: this https URL
Abstract:We report fine-grained, deterministic instability in Dynamic Tensor Rematerialization (DTR), an online eviction policy for memory-constrained DNN training, measured on the reference DTR simulator (simrd) using public execution traces. On an LSTM trace, memory budgets differing by 0.10% of unconstrained peak memory select fast and slow execution regimes whose overheads differ by as much as 7.3x; the slow regime is driven by broadly repeated re-eviction of the same storages (evictions per storage rise from 1.33 to 8.27 while the set of distinct evicted storages is essentially unchanged: 5,233 vs 5,236, with the two sets overlapping at Jaccard 0.999). On a ResNet-32 trace, a fine budget sweep reveals a deterministic feasibility inversion: the run is feasible at ratio 0.101, infeasible (OOM) across 0.102-0.106, and feasible again from 0.107. We trace the immediate cause of the OOM to a fully pinned recursive rematerialization frontier that exceeds the budget after every evictable tensor has been evicted. Ablations using the DTR authors’ own variants implicate the joint size-staleness scoring term in the observed LSTM instability. We argue these are at least two distinct budget-sensitive pathologies rather than one mechanism, and we separate what is demonstrated from what remains hypothesised. All results concern the reference simulator; reproduction in a production runtime is future work. Code, instrumentation, and raw results accompany this preprint.
[LG-24] Budgeted Quotient-Residual Guidance for Frozen Pocket-Conditioned Molecular Diffusion
链接: https://arxiv.org/abs/2609.31222
作者: Xinyu Wang,Jinbo Bi,Minghu Song
类目: Machine Learning (cs.LG)
*备注: 21 pages, 5 figures. Includes theoretical proofs and supplementary experimental results
Abstract:Pocket-conditioned molecular diffusion updates ambient atom coordinates, but many lead-optimization objectives are expressed on quotient features such as distances, contacts, and anchored substructures. We introduce budgeted quotient-residual guidance (QRG), an inference-time correction that makes these quotient objectives active without retraining the molecular generator. QRG lifts quotient covectors to metric-horizontal ambient directions and delivers them through a trust budget set by the frozen sampler’s own step norm: quotient geometry chooses the direction, while sampler motion bounds the scale. We derive the horizontal lift, closed-form sampler-budget update, KL/kinetic interpretation around a frozen reverse step, equivariance conditions, and a product-budget split for budget-capped section and residual controls. Controlled quotient tasks confirm that sampler-relative delivery activates signals that raw local quotient gradients leave dormant. On frozen TargetDiff backbones, official seed-0 CBGBench ligand-generation/editing sweeps show practical quality-runtime gains: Local-QRG improves validity from 0.815 to 0.864 on fragment growing, 0.664 to 0.707 on scaffold hopping, and 0.681 to 0.712 on linker design, while PredNext-QRG improves fragment/scaffold and remains near-neutral on linker. Novelty remains 1.000 and diversity is preserved in the matched multi-seed molecular slice, giving task-dependent improvements without sampler retraining or backbone modification. Overall, QRG provides a lightweight route to quotient-aware inference for frozen molecular samplers with explicit runtime accounting.
[LG-25] Self-Supervised Representation Learning: From Spectral Foundation Models to Auroral Emission Spectra ICASSP2027
链接: https://arxiv.org/abs/2609.31206
作者: Matthieu Le Lain,Gaël Cessateur,Sébastien Lefèvre
类目: Machine Learning (cs.LG); Space Physics (physics.space-ph)
*备注: 5 pages, 1 figure, 3 tables. Submitted to IEEE ICASSP 2027
Abstract:Auroral spectrographs such as the Auroral Spectrograph In Skibotn (ASIS) record hundreds of thousands of emission spectra, but only a few hundred can be labelled by an expert. To exploit the rest, we pretrain a 1D Vision Transformer with a masked autoencoder on 223,000 unlabelled spectra. Without labels, its representation recovers the emission-line intensity ratios that physicists use to diagnose the precipitating particles (R^2 0.91 vs. 0.77 for an untrained control) and, under one linear probe, classifies as well as 13 features designed by experts. Fine-tuned, the model outperforms the previous supervised auroral classifier on its own benchmark (macro-AP 88.5 vs. 77.8), reaches 0.870 mAP, and exceeds the same architecture trained from scratch by +0.159 with 10% of the labels; attribution shows that it uses both N2+ bands. Could an existing pretrained model replace it? Two astronomical spectral foundation models and a time-series model transfer according to their spectral window: SpectraFM, trained in the infrared, falls below the untrained control, whereas SpecFormer, trained in the optical, approaches in-domain pretraining without reaching it.
[LG-26] ALF: An Active Learning Framework for Scientific Discovery
链接: https://arxiv.org/abs/2609.31197
作者: Shikha Surana,Alex Hawkins-Hooker,Olivia Gallup,Christoph Brunken,Jules Tilly,Paul Duckworth
类目: Machine Learning (cs.LG)
*备注: 14 pages, 7 figures
Abstract:Machine learning for scientific discovery is almost systematically data bound. Producing relevant high quality data, under budget constraints, is amongst the most promising ways to advance the field. Active learning (AL) offers promise wherever labelling requires expensive experiment, measurement, or simulation. Most existing tools cover only part of the data acquisition loop, and typically focus on either offline benchmarking or online deployment, but not both. We present ALF, a modular AL Framework that runs the full data acquisition loop via five modular components. One clear API for both settings: offline, against an existing dataset for controlled and reproducible experimentation; and online, against an oracle for acquiring new candidates in real-world deployments. ALF is open-source and available at this https URL.
[LG-27] Audio emotion recognition for atypical hearing
链接: https://arxiv.org/abs/2609.31168
作者: Ulysse Roussel(STMS)
类目: Machine Learning (cs.LG)
*备注:
Abstract:My doctoral work aims to explore Audio Emotion Recognition (AER) in the context of atypical listening. This research focuses on auditory hypersensitivity in people with autism, a phenomenon that is often difficult to evaluate and unique to each individual. Our core idea is to leverage our understanding of affect from acoustic traits, relying on the possibility of generalizing affective responses from a small amount of annotated data. As a first step, we fine-tune a large foundation model, Contrastive Language-Audio Pretraining (CLAP) using low-rank adaptation (LoRA), trained on a valence and arousal dataset of neurotypical listeners.
[LG-28] BreathGRU: A Novel Semi-Supervised Bidirectional Gated Recurrent Unit Framework for Speech and Breath Segmentation for Respiratory Audio
链接: https://arxiv.org/abs/2609.31165
作者: Sania Fatima Sayed,John W. Holloway,Reyer Zwiggelaar,Faisal I. Rezwan
类目: ound (cs.SD); Machine Learning (cs.LG); Audio and Speech Processing (eess.AS)
*备注:
Abstract:Speech-breath segmentation is a fundamental preprocessing step in respiratory audio analysis, enabling applications such as respiratory acoustic biomarker extraction, lung function prediction and disease monitoring. Existing approaches, including threshold methods, Fourier Transform-based techniques, and unsupervised and pretrained voice activity detection (VAD) models, primarily focus on speech detection and often classify breathing events as non-speech or silence, limiting their applicability for precise breath detection. To address this limitation, we propose BreathGRU, a semi-supervised Bidirectional Gated Recurrent Unit (BiGRU) framework specifically designed for speech-breath segmentation. The proposed framework combines frame-level acoustic feature extraction with bidirectional recurrent modelling, pseudo-label refinement and duration-constrained Segmental Viterbi decoding to produce speech and breath segmentation. BreathGRU was evaluated against the existing approaches, using manually annotated recordings. Performance was assessed using event-based, time-based, overlap-based, duration-based and boundary-based segmentation metrics. Experiment results demonstrated that BreathGRU achieved the highest breath event recall (0.83), the lowest onset-localisation error (0.14s) and the highest Mean Match Intersection over Union (0.81), with competitive overall segmentation performance compared to large pretrained VAD models like Silero. Qualitative evaluation on manually annotated recordings further showed close agreement between BreathGRU and manual annotation, with better breath detection compared to Silero. These findings demonstrate that explicit breath event modelling provides advantages over general-purpose VAD models and establish BreathGRU as an effective speech-breath segmentation framework which can be applied for respiratory audio analysis and pulmonary healthcare applications.
[LG-29] WorldTS: World Modeling for Multimodal Covariate-aware Time Series Forecasting
链接: https://arxiv.org/abs/2609.31162
作者: Yuhan Zhu,Xiangfei Qiu,Hanyin Cheng,Wangmeng Shen,Chenjuan Guo,Bin Yang,Jilin Hu,Christian S. Jensen
类目: Machine Learning (cs.LG)
*备注:
Abstract:Time series forecasting is typically framed as learning a direct mapping from historical to future observations in the observation space. However, sequences of observations generally provide only a partial view of the dynamics of the underlying system, with future observations being shaped by latent dynamics. Recent latent-space forecasting methods thus achieve improved performance by predicting future observations from latent-space representations of historical observations rather than directly forecasting future observations in the observation space. Next, while future observations are also shaped by external factors, how to incorporate external, often multimodal, information into forecasting, so that it can shape latent-state formation and evolution directly, remains underexplored. We propose WorldTS, a world-modeling based forecasting framework that integrates multimodal covariates directly into the forecasting to further improve forecasting performance. Specifically, WorldTS employs a two-stage training strategy. First, it learns forecasting-relevant latent state dynamics conditioned on multimodal covariates, yielding encoded future states. Next, the learned state dynamics are frozen, and an observation decoder is trained to map the predicted future states back to future observations. Extensive experiments on 21 real-world datasets offer insight into WorldTS and its effectiveness.
[LG-30] I Act Therefore I Am: When Is JEPAs Action-Conditioning Enough to Learn Causal Mechanisms?
链接: https://arxiv.org/abs/2609.31161
作者: Yuhang Liu,Zhuo Huang,Javen Qinfeng Shi
类目: Machine Learning (cs.LG)
*备注:
Abstract:Recent empirical and theoretical advances suggest that joint-embedding predictive architectures (JEPAs) may learn meaningful representations for action-conditioned prediction of future outcomes, thus becoming one of the foundational structures for world models. However, accurate prediction does not, in general, necessarily imply recovery of underlying causal states that give rise to the observed dynamics. This work investigates when and how JEPAs can recover the underlying causal states from observations. We first introduce a latent variable model, in which high-dimensional observations are generated from latent causal states whose dynamics are governed by action-conditioned transition mechanisms. Based on this formulation, we develop a general information-theoretic objective that combines conditional likelihood maximization for learning transition dynamics with entropy maximization for preserving latent state information. We then establish identifiability conditions under which representations learned by this general objective recover the underlying latent causal states up to component-wise invertible transformations and permutation. One key condition for such identifiability is sufficient action-induced variation in the transition mechanisms. Guided by this finding, we instantiate the general objective with an action-modulated Gaussian additive-noise model, yielding action-modulated JEPA (A-JEPA). Experiments on synthetic environments verify the theoretical findings under the identifiability conditions and robustness to moderate violations, while visual benchmarks demonstrate improved state recovery and transfer to unseen transition mechanisms.
[LG-31] Bayesian Tensor Autoencoder with Physics-informed Predictive Prior for Multi-dimensional Time Series Anomaly Detection
链接: https://arxiv.org/abs/2609.31157
作者: Jianan Liu,Chunguang Li
类目: Machine Learning (cs.LG)
*备注: 28 pages, 7 figures,
Abstract:Multi-dimensional time series, inherently tensorial, are common in practice. Despite great progress in time series anomaly detection, most existing methods are confined to uni-/multi-variate time series. When handling multi-dimensional time series using these methods, reshaping operations are required, which inevitably break the intrinsic correlations and thus lead to performance degradation. In uni-/multi-variate time series anomaly detection, AutoEncoders (AEs) are widely adopted and generally categorized into reconstruction-based and prediction-based AEs. The reconstruction-based AE utilizes the current observation for reconstruction, while the prediction-based AE utilizes the historical information to predict the current observation. Thus, the two AEs utilize different information. To bridge the gap between reconstruction-based and prediction-based AEs, so as to fully leverage the available information and thus further enhance performance, we propose a predictive prior and incorporate it into the reconstruction-based AE. It may not be very difficult to conceive this idea, but designing the predictive prior so that it can work for tensor anomaly detection is non-trivial. Specifically, to avoid breaking the intrinsic correlations within the multi-dimensional time series, we use the tensor AE as the backbone. To incorporate the predictive prior into the reconstruction-based AE, we propose a Bayesian fusion approach and our analysis reveals that this approach can enhance the modeling capability of the model for normal data. To mitigate the over-generalization problem of AE, we incorporate physical laws, i.e. tensor low-rank decomposition rules, into the neural networks in the predictive prior, leading to the Physics-informed Predictive Prior Tensor AE (PPPTAE) framework. Experimental results on real-world datasets demonstrate the effectiveness of the proposed method.
[LG-32] CRNDiff: Count-Native Diffusion Framework via Chemical Reaction Networks ICLR2027
链接: https://arxiv.org/abs/2609.31149
作者: Yuxuan Qiu,Praful Gagrani,Tetsuya J Kobayashi
类目: Machine Learning (cs.LG)
*备注: 28 pages, 9 figures, 10 tables. Under review as a conference paper at ICLR 2027
Abstract:Scientific measurements such as single-cell RNA (scRNA) sequencing often take the form of nonnegative integer counts, whereas continuous-state diffusion models approximate this discrete structure using continuous coordinates. Building on stochastic chemical reaction networks (CRNs), a class of count-native Markov jump processes, we introduce CRNDiff, a structured framework that combines count-space diffusion with inference-time conditioning on rare subpopulations. An independent birth–death instantiation yields a closed-form transition kernel for forward noising. This kernel enables reverse sampling via forward-filtering backward-sampling (FFBS) and supports data-driven selection of the terminal noising time, eliminating the need for a validation sweep. This tractability also lets us introduce tilted Feynman–Kac (FK) steering, a method for sampling target subpopulations from a frozen generator without retraining. By tilting posterior marginals before FK particle correction, steering mitigates importance-weight concentration when the target population is rare. Using scRNA-seq data from the human heart cell atlas, we test the ability of CRNDiff to generate cell-type-specific distributions. Across the three evaluated target populations, CRNDiff achieves the highest conditional fidelity among the evaluated generative models, with larger mean purity margins for rarer target populations. Generated cells preserve marker-level differential-expression structure. Replacing real training cells for the target classes with generated cells yields downstream classification performance approaching that of the real-data reference.
[LG-33] Unknown-Traffic Detection Calibration and Shortcut Reliance in Distilled Encrypted-Traffic Classifiers over One Year
链接: https://arxiv.org/abs/2609.31141
作者: Mahmoud Abbasi
类目: Networking and Internet Architecture (cs.NI); Machine Learning (cs.LG)
*备注: 15 pages, 5 figures, 11 tables. Pre-registered at OSF ( this https URL ) before any test-window result was computed. Code: this https URL . Per-flow scores and model checkpoints: this https URL
Abstract:Knowledge distillation is the standard way to compress encrypted-traffic classifiers for the edge, and almost all such work judges students by accuracy alone. We ask what else a student inherits: unknown-traffic detection, calibration, shortcut reliance, and whether any survives a year of drift. Resemblance proves little on its own, since soft targets also regularise. We therefore distil one 101k-parameter student from two teachers of equal accuracy but different construction, a five-member ensemble and a single wider model, so that following one rather than the other is attributable to it. The design was pre-registered before any test result was seen. We tested ten hypotheses on CESNET-TLS-Year22, a year of real TLS traffic, across 18 test windows over 35 weeks. Two are supported: a student’s per-flow unknown-scores shift toward its own teacher, but only at a conventional temperature, not the accuracy-optimal one; and a shortcut-reliant teacher passes its over-confidence to a student that never sees the feature. The drift prediction is reversed under both scores, the gap narrowing rather than widening and the student overtaking under the energy score in two of three replicates, as is the prediction that such a teacher harms its student’s detection, which improves slightly. Shortcut reliance is set by model size, not distillation. Under the logit-based scores nothing else transfers: distillation beats neither a temperature-scaled direct student nor label smoothing. Exploratory analysis shows this turns on the scoring rule: with a feature-space detector the teacher detects unknown traffic 0.073 AUROC better than the direct student, where the energy score sees 0.000, and the conventional-temperature student inherits most of it. Label smoothing, with no teacher, recovers more. Distillation transfers the teacher’s habits; what looks like an inherited ability is available without one.
[LG-34] Frame the adversary: a structure-aware attack methodology NEURIPS’26
链接: https://arxiv.org/abs/2609.31128
作者: Vicky Kouni,Stelios Perrakis,Francis Bach,Pascal Frossard,Yann Chevaleyre
类目: Machine Learning (cs.LG)
*备注: Accepted at NeurIPS’26
Abstract:Frequency-based adversarial attacks have recently grown popular by exploiting spectral sensitivities shared across neural architectures. Unlike spatial perturbations, frequency-based attacks expose deeper vulnerabilities, making them especially valuable for robust evaluation of safety-critical and security-sensitive applications. Yet, existing approaches are typically not derived as solutions to an optimization problem that explicitly captures transform-domain structure. In this paper, we propose a methodology for crafting principled frequency-based adversarial attacks, via a dedicated optimization framework. A cornerstone of our method hinges on the introduction of a perturbation constraint set, tied to highly structured non-orthogonal transforms, well-known for their flexible, non-predefined frequency handling. We prove that the attacks emerge as weighted \ell_2 -projections onto this set, yielding a general and controlled attack generation mechanism. By this, we provide a clear geometric attack characterization, ensuring alignment between the optimization objective and the perturbation constraint. We assess our framework on standardized datasets, for pretrained and adversarially robust models. Results highlight that our attacks, being solutions to an optimization problem, over a structured perturbation set, are highly effective, even across different, unseen architectures. Our methodology could serve as a theoretical baseline for designing and analyzing transformed-based attacks, targeting fundamental model vulnerabilities, instead of mere architecture-specific artifacts typically studied in the robustness literature.
[LG-35] he Residual Streams Effective Depth ACML2026
链接: https://arxiv.org/abs/2609.31098
作者: Barak Gahtan,Ido Galil,Alex M. Bronstein
类目: Machine Learning (cs.LG)
*备注: Accepted at the 17th Asian Conference on Machine Learning (ACML 2026)
Abstract:We introduce \empheffective depth ( \Deff ), a scalar diagnostic that treats the layer-wise residual stream of a transformer as a discrete-time process, measures how representation similarity decays with layer distance, and aggregates that profile into one number. Across sixteen decoder-only language models, \Deff separates a structural consequence of residual accumulation from an empirical one: even maximally diverse orthogonal updates have the closed-form reference F_L = 2L/(L+1)2 , yet fifteen of sixteen default measurements lie below F_L (Qwen3.5: 32–44%, OLMo-2: 40–41%, Pythia: 23–28%). Matched references show that the gap is not caused by the persistent initial state or update-size imbalance, but is largely a calibrated signature of correlated residual updates rather than evidence that depth is unused. Symmetric position-0, token-normalisation, and top-PC controls show the regime is not reducible to BOS or top-PC artefacts: the lone above-reference default outlier joins the same regime, and all sixteen models are sub-reference after token-normalisation or top-1-PC removal. Intermediate checkpoints show that the regime is established early in OLMo-2 and stable through 5T tokens, while Pythia-1.4B follows a distinct decreasing trajectory. A controlled residual-carry intervention supports the mechanism, and \Deff is best read as a \emphglobal accumulated-state diagnostic, not as a capability score or pruning method.
[LG-36] Block Sparse Attention with Log-Linear Complexity
链接: https://arxiv.org/abs/2609.31093
作者: Bohao Tang,Zhen Qin,Yuqi Pan,Zheng Li,Pengfei Liu
类目: Machine Learning (cs.LG)
*备注:
Abstract:Scaling language models to long contexts is limited by the quadratic cost of self-attention. Block sparse attention offers an efficient alternative, but selecting the retained blocks remains a bottleneck. Conventional block selection requires scoring all query-block pairs and therefore remains quadratic in sequence length. To address this issue, we propose PISA, a block-sparse attention mechanism that employs a pyramid Top- K selection strategy. The main idea is to gradually narrow down the candidates across different levels, making it more efficient to find the most relevant keys. Specifically, we construct a coarse-to-fine hierarchy of keys and perform selection from the coarsest level. At each level, LogSumExp scoring is applied to a bounded candidate set to select candidates for the next finer level, continuing until the finest level is reached. Through pooling, we construct O(\log N) levels of keys, yielding an overall complexity of O(N\log N) , where N denotes the sequence length. We develop hardware-aware Triton kernels for both training and inference, fusing hierarchical routing and LogSumExp scoring without materializing the query-key score matrix. We further evaluate our method on language modeling tasks. Compared with the baseline, our method achieves comparable performance on benchmarks such as commonsense reasoning while delivering better results on retrieval tasks.
[LG-37] SAGE: A sampling-aware global evaluation benchmark for species distribution modeling
链接: https://arxiv.org/abs/2609.31082
作者: Emilia Arens,Nina van Tiel,Robin Zbinden,Damien Robert,Lukas Drees,Chiara Vanalli,Benjamin Kellenberger,Niklaus E. Zimmermann,Loïc Pellissier,Devis Tuia,Jan Dirk Wegner
类目: Machine Learning (cs.LG); Populations and Evolution (q-bio.PE); Machine Learning (stat.ML)
*备注: Under review. Project page: this https URL
Abstract:Knowing where species occur is fundamental for biodiversity research and conservation. Species distribution models (SDMs) link species observations to environmental conditions to estimate their spatial distribution. However, accuracy varies with the underlying data and models, making it essential to know for which species models can be trusted. Deep-learning-based SDMs (“DeepSDMs”) now jointly model thousands of species, drawing on hundreds of millions of community-science records. At this scale, averaging performance hides substantial species-level variability, particularly for rare species, often of greatest conservation concern. Records are also strongly biased, making occurrence counts misleading. Accounting for these factors is essential for a reliable and informative evaluation of multi-species SDMs. Here, we introduce a Sampling-Aware Global Evaluation (SAGE) benchmark, combining GBIF records for training with sPlotOpen vegetation plots for presence-absence evaluation across 5771 plant species. We propose an evaluation framework that groups species based on two properties, sampling effort and relative prevalence, which describe how densely a species’ range is sampled and how frequently the species is recorded. Evaluating single-species SDMs and multi-species DeepSDMs, we find that Random Forests and DeepSDMs perform best overall, but neither dominates: DeepSDMs outperform single-species SDMs for infrequently recorded species while offering no consistent advantage for well-sampled ones. Crucially, this advantage emerges only when established bias-correction practices, such as spatial thinning and reweighting, are carried over to the deep-learning setting. SAGE helps identify the species and data conditions for which a given approach is beneficial, thereby supporting the development of more transparent and ecologically credible SDMs. Data and code: this https URL
[LG-38] Distributed Learning as a Service: The Developers Perspective
链接: https://arxiv.org/abs/2609.31061
作者: Tianyue Chu,Filippo Vannella,Dimitra Tsigkari,Paula Delgado-Santos,Fernando López,Pablo Gomez Guerrero,Sotirios Spantideas,David Solans Noguero
类目: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC); Networking and Internet Architecture (cs.NI)
*备注: 3 pages, 5 figures. Paper accepted at the 22nd International Conference on Network and Service Management (CNSM 2026). Code: this https URL
Abstract:Application developers of distributed learning services face challenges that a typical federated learning loop does not address. Specifically, the model updates can still leak private data, devices might not be able to participate in the training due to limited resources, a single aggregator might not be able to scale, and the transmissions of model weights induce a considerable bandwidth cost. This paper demonstrates DLaaS (Distributed Learning as a Service) from the developer’s vantage point. Using a single admin dashboard, the developer initiates a distributed/federated learning job and is able to activate Differential Privacy (DP), Split Learning (SL), Hierarchical Aggregation (HA), and Knowledge Distillation (KD) as declarative options, with no change to the clients’ code. We demonstrate the complete service lifecycle on an industrial smart-home Wake-up Word (WuW) task, using the “Ok Aura” dataset. Once the developer initiates a distributed learning job by toggling DP, SL, HA, and KD in the admin dashboard, the system dispatches the job to a set of Android clients and Dockerized helper aggregators. In the demonstration, these mechanisms run live across configurations. Then, the clients train the model locally and return their updates. The trained model is served to a consumer-side Android application that performs on-device WuW detection on a live microphone stream. In particular, the conference attendees will be invited to speak the trigger phrase and monitor in real time the per-class confidence and inference latency. Finally, we release the source code and short video walkthroughs of these configurations.
[LG-39] Aurora-X: Built for Extreme Time Series Forecasting
链接: https://arxiv.org/abs/2609.31038
作者: Xingjian Wu,Chenjuan Guo,Xiangfei Qiu,Zhigang Hu,Hanyin Cheng,Peng Chen,Yang Shu,Jilin Hu,Bin Yang
类目: Machine Learning (cs.LG)
*备注:
Abstract:Time series foundation models (TSFMs) enable cross-domain forecasting, but their development as general-purpose forecasters remains constrained by underexplored training potential and limited architectural versatility. To address these challenges, we introduce Aurora-X, a billion-scale TSFM with a progressive curriculum and a unified architecture. We first use channel-independent pretraining to learn temporal patterns, then introduce cross-variable dependencies, varied context and horizon lengths, and future covariates if available during midtraining. Variable-resolution post-training further enables an adjustable temporal span per token at inference. With fixed model weights, this supports longer histories under a fixed token budget or fewer tokens for the same history, enabling test-time scaling. With a versatile architecture, Aurora-X supports cross-variable modeling, covariate conditioning, and parallel decoding of future patches for probabilistic forecasting. These are supported by a novel pattern-guided mixture-of-experts that expands model capacity through sparse activation and uses shallow patch similarities to constrain deep-layer routing, guiding expert specialization across heterogeneous time series. Furthermore, we propose an implicit quantile network head that predicts arbitrary quantiles to characterize predictive distributions, enhancing probabilistic forecasting flexibility. Comprehensive experiments on GIFT-Eval, TIME, FEV-Bench, TFB, and DAG-Bench demonstrate state-of-the-art forecasting performance against pretrained TSFMs and task-specific supervised models.
[LG-40] Robust Graph Clustering Network for Multiple Missing Data
链接: https://arxiv.org/abs/2609.31033
作者: Keyuan Qiu,Renda Han,Zhen Tang,Qiang He,Xingwei Wang,Wenxin Zhang,Guangzhen Yao,Junxin Chen,Qingjian Ni
类目: Machine Learning (cs.LG)
*备注:
Abstract:Clustering on graphs where both node attributes and structural links are partially missing remains a challenging task. Existing methods typically rely on imputation-then-clustering on single-view missingness incomplete graphs, which are vulnerable to cross-view error propagation and cluster-boundary blurring under simultaneous attribute and structure missingness. To address these limitations, we propose a Robust Graph Clustering Network for Multiple Missing Data (RGCN), which is designed to handle simultaneous node attribute and graph structure incompleteness. RGCN introduces three key innovations: First, we design a view-decoupled dual-branch imputation to mitigate interference and enable mutual enhancement in recovering missing data. Second, we employ a multi-hyperspherical mixture prior to enhance intra-cluster compactness and inter-cluster separability on a directional latent manifold. Third, a boundary-aware contrastive enhancement objective mitigates the blurring of clusters caused by imputation bias. Extensive experiments on real-world datasets demonstrate that RGCN consistently outperforms state-of-the-art baselines under various missing patterns.
[LG-41] Metacognitive Selective Ensemble for Mobile Systems
链接: https://arxiv.org/abs/2609.31031
作者: Sungmin Lee,Kichang Lee,Joonhee Lee,JaeYeon Park,Songkuk Kim,JeongGil Ko
类目: Machine Learning (cs.LG)
*备注:
Abstract:Deep ensembles improve robustness in mobile sensing, but repeatedly executing many models over continuous sensor streams is costly. Selecting only a few members reduces this cost, yet adaptive selection often requires additional model execution to obtain reliable evidence about inactive candidates. We present MetaSE, an active ensemble framework that exploits short-term persistence in per-model reliability. MetaSE maintains a small active set across windows, uses post-execution evidence to reject unreliable members, and invokes lightweight routing only when replacement is needed. This stateful design accesses the diversity of a larger pool without repeated full-pool evaluation. Across four HAR datasets and four model architectures, MetaSE consistently improves over a fixed three-model ensemble and achieves accuracy comparable to substantially more expensive adaptive and full-ensemble inference. On a Raspberry Pi 4B, MetaSE is 2.7x faster and uses 69% less memory than full ten-model inference.
[LG-42] Precision at Speed: Sample-Efficient Online Model-Based Reinforcement Learning for Hydraulic Excavator Control
链接: https://arxiv.org/abs/2609.31025
作者: Claudio Canales,Fang Nan,Marco Hutter,Javier Ruiz-del-Solar
类目: Robotics (cs.RO); Machine Learning (cs.LG); Systems and Control (eess.SY)
*备注:
Abstract:Precise, high-speed control remains challenging for robots with complex actuation dynamics. Learning directly on hardware is further constrained by the cost of real-world interaction. We present an online model-based reinforcement learning framework that learns a probabilistic dynamics ensemble model from scratch for sampling-based model predictive control. A precision-gated contouring objective conditions the progress reward on path accuracy, prioritizing precision over speed. In a data-driven excavator simulator, the framework achieves higher sample efficiency than the evaluated model-based reinforcement learning baselines. We validate the framework by learning directly on an 11.5-ton Menzi Muck M445 hydraulic excavator, without demonstrations or simulation pretraining. After 20 minutes of interaction, the controller reaches tracking accuracy comparable to prior learned controllers trained on 100-150 minutes of data. After 40 minutes, it sustains sub-centimeter mean path error at high operating speeds.
[LG-43] Synth-JEPA: Joint Embedding Prediction for Renderer-Free Synthesizer Parameter Search
链接: https://arxiv.org/abs/2609.31024
作者: Ben Hayes,Haokun Tian,Stefan Lattner
类目: ound (cs.SD); Machine Learning (cs.LG); Audio and Speech Processing (eess.AS)
*备注:
Abstract:Sound matching can be formulated as optimizing synthesizer parameters against an audio-domain objective. However, objectives derived from generic audio representations are often difficult to optimize, while direct search requires rendering every candidate. We introduce Synth-JEPA, which learns mutually predictive audio and parameter representations from paired synthesizer data. At inference, candidate parameters are scored directly in this learned space, yielding a renderer-free objective whose audio geometry is shaped by parameter correspondences rather than generic audio similarity. We evaluate Synth-JEPA on Surge XT using held-out synthesizer sounds and out-of-domain NSynth and FSD50K targets, against inverse models, direct search, and learned proxy objectives. Synth-JEPA outperforms all baselines in-domain and remains competitive out-of-domain. Its matching quality continues to improve with additional test-time search, allowing compute to be traded for match quality. In pairwise listening tests, listeners preferred Synth-JEPA in 85% of trials overall. Together, these results show that an audio representation with a parameter-induced geometry allows synthesizer sound matching to be approached as an effective renderer-free search problem.
[LG-44] Robust Successor Features
链接: https://arxiv.org/abs/2609.31016
作者: Erik Nikulski,Yamen Habib,Vicenç Gomez,Anders Jonsson,Rubén Moreno-Bote,Javier Segovia-Aguas
类目: Machine Learning (cs.LG)
*备注: 10 pages, 3 figures, to be published in EWRL 2026
Abstract:Generalization in Reinforcement Learning (RL) refers to the ability to execute close-to-optimal policies in unseen tasks after the agent has been trained on a different set of tasks. Building on the seminal work of the successor representation and further adaptations with function approximation, Transfer in RL has traditionally focused on generalizing to tasks that only differ in the reward function. A decade after the introduction of the successor representation, Robust RL emerged simultaneously from several articles in the field of operations research. In Robust RL, the transition kernel is unknown, and the goal is to maximize the expected reward under this uncertainty. Our work unifies these two paradigms through robust successor features, which generalize across both the reward function and the transition kernel, under the assumption that tasks are linear Markov Decision Processes. We derive a bound on Generalized Policy Improvement (GPI) that explicitly quantifies how performance degrades with the mismatch between transition kernels, recovering existing successor-feature guarantees when dynamics are shared. Finally, the generalization capabilities of robust successor features are validated on several grid-based benchmarks and compared to previous alternatives that focus solely on either the reward or the transition kernel.
[LG-45] Learning Hierarchical Causal Representations of the Effects of Forcings on Temperature in Climate Models
链接: https://arxiv.org/abs/2609.30995
作者: Shan Zhao,Ilija Trajkovic,Julia Kaltenborn,Yaniv Gurwicz,Peer Nowack,David Rolnick,Julien Boussard
类目: Machine Learning (cs.LG)
*备注:
Abstract:Machine learning (ML) emulators provide a fast and cost-effective method to simulate climate change scenarios after being trained on Earth System Models projections. However, the black-box nature of those data-driven approaches limit the usability and trustworthiness of their outputs and in particular their use as causal attribution tools. Here, we develop a hierarchical causal representation learning framework applied to sea surface temperature fields from a state-of-the-art global climate model. As a key advance over previous work, our framework explicitly models both atmospheric dynamical interactions arising from internal climate variability and forced responses due to changes in atmospheric greenhouse gas and aerosol concentrations. When trained on future climate change scenarios, our method accurately predicts the long-term global mean and regional temperature evolution and shows physically realistic responses to perturbations in greenhouse gas and aerosol concentrations when evaluated on unseen scenarios. Our results underline the potential of causal representation learning frameworks for advancing climate model emulation.
[LG-46] LipSSM: Structurally Lipschitz-Bounded Cascaded State-Space Model via Metric Transfer between Consecutive SSM Layers ICASSP2027
链接: https://arxiv.org/abs/2609.30973
作者: Natsuki Yoshino,Ren Uchida,Kazuki Matsumoto,Kohei Yatabe
类目: Machine Learning (cs.LG); Systems and Control (eess.SY)
*备注: Submitted to IEEE ICASSP 2027
Abstract:Lipschitz continuity is a fundamental principle in the design of certifiably robust deep neural networks (DNNs), wherein adjusting the Lipschitz constant, which quantifies network robustness, is of central theoretical importance. A standard approach to enforcing Lipschitz continuity requires each layer of a DNN to be Lipschitz continuous, thereby guaranteeing overall Lipschitz continuity. However, this layer-wise approach typically imposes overly conservative restrictions by producing a loose estimate of the overall Lipschitz constant, which limits the expressive capacity of the DNN and degrades empirical performance at a prescribed level of robustness. To overcome this loose estimation, the recently proposed LipKernel transfers information across layers to yield a much tighter overall Lipschitz bound than conventional layer-wise construction. In this paper, we extend this concept to cascaded state-space models (SSMs) to construct Lipschitz-continuous DNNs capable of modeling longer-term dependencies. The proposed architecture, named LipSSM, is theoretically justified and empirically evaluated.
[LG-47] Gradient Surgery for Physics-Informed Neural Networks ACML2026
链接: https://arxiv.org/abs/2609.30966
作者: Thomas Borsani,Giuseppe Di Fatta
类目: Machine Learning (cs.LG)
*备注: Accepted at ACML 2026
Abstract:Physics-Informed Neural Networks (PINNs) are trained by optimising a composite objective that combines data fitting with physics-based constraints, typically resulting in a highly imbalanced multi-task optimisation problem. Under these conditions, existing optimisation strategies are affected by conflicting task gradients, leading to slow convergence and unstable training, particularly for stiff and high-frequency partial differential equations. We analyse gradient conflicts throughout training of PINNs with standard optimiser and investigate Multi-Task Deep Learning (MTDL) optimisation methods. In our analysis across four benchmark problems we observed that PINN optimisation exhibits three distinct phases in which angle- and magnitude-based gradient conflicts alternate, with only one present at a time. Building on these observations, we propose PAM-GS, a physics-aware gradient surgery method that adaptively mitigates task interference during training according to the observed conflict types. Experiments on four representative PDE benchmarks demonstrate that PAM-GS combines competitive solution accuracy with consistently strong task-balanced performance, outperforming existing methods on most problems.
[LG-48] owards Understanding Momentum Acceleration in River-Valley Loss Landscape
链接: https://arxiv.org/abs/2609.30957
作者: Miao Lu,Zeyu Bian,Kaiyue Wen,Beining Wu,Siyu Chen,Tianhao Wang,Zhiyuan Li
类目: Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注: 70 pages, 15 figures
Abstract:The empirical success of pretraining large language models has inspired a deeper investigation into the underlying loss landscapes and the optimization dynamics. Recent empirical and theoretical study suggest that the training loss landscape often exhibits a “river-valley” structure, which features a low-loss manifold (river) flanked by sharp orthogonal directions with higher loss (mountains). In the long term, the optimization progress is determined primarily by the progress along the river. Within such a landscape, gradient descent with large learning rates can move faster along the river despite high apparent loss due to vertical oscillations, while a subsequent sharp decay in the learning rate suppresses these oscillations, revealing genuine optimization progress. This explains the recent success of warmup-stable-decay (WSD) learning rate scheduler which, unlike cosine scheduling, keeps stable high learning rate and decays before producing intermediate checkpoints. Building on this foundation, in this work we take a step further and study the role of momentum within such a loss landscape. We establish theoretical analysis that characterizes how momentum accelerates optimization by stabilizing large learning rates that can not be tolerated by vanilla GD without deviating significantly from the river. The enabled large learning rate in-turn gives greater speed along the river and makes faster essential progress in the long run. Another intriguing observation from theory is that for a river-valley landscape with very flat and slow-spinning river, the momentum itself does not contribute directly to acceleration in terms of the speed of tracking the river, while the main acceleration comes from the admissible larger learning rate.
[LG-49] Low-Bit Recurrent States in Hybrid Language Models ICASSP2027
链接: https://arxiv.org/abs/2609.30950
作者: Hongren Chen,Jiayang He
类目: Machine Learning (cs.LG)
*备注: 9 pages, 2 figures, 3 tables. Submitted to ICASSP 2027
Abstract:Hybrid language models maintain fixed-size recurrent states, but existing quantizers typically use eight bits or more. Quantization errors persist according to channel decay rates. We derive distortion weights from the observability Gramian and combine them with normalized state ranges for mixed-precision bit allocation, without calibration data, rotation, or training. We also quantize decay rates logarithmically. With per-token state quantization, a four-bit mean payload reduces excess negative log-likelihood by factors of 3.3–27.9 relative to the best of seven baselines across three hybrid models; metadata costs vary. At six bits, negative log-likelihood differs from the FP32-state baseline by less than 0.005 nats. Ablations separate gains from variable bit widths, decay weighting, and range normalization. With less frequent write-backs, gains diminish and depend on the model and budget.
[LG-50] EPOC: Endpoint-Preserving Online Correction With Compressed Residual State for Multi-Horizon Time Series Forecasting
链接: https://arxiv.org/abs/2609.30929
作者: Takumi Fujimoto,Hiroaki Nishi
类目: Machine Learning (cs.LG)
*备注:
Abstract:Completed multi-horizon forecasts provide residual feedback for a fixed forecaster, but retaining full residual blocks increases auxiliary state. We propose Endpoint-Preserving Online Correction (EPOC) with a compressed residual state. It stores low-order discrete cosine transform (DCT) coefficients and the final value of the preceding residual block. Within each channel, the endpoint is shared across component-wise online ridge regressions that also use current-forecast coefficients. The fitted DCT correction is blended with the base forecast. We evaluate eight multivariate series with DLinear and PatchTST, three seeds, and two training variants, yielding 96 matched fixed-base conditions at a 24-step horizon. EPOC achieves mean condition-wise reductions in mean squared error (MSE) and mean absolute error (MAE) of 15.40% and 9.35% from the uncorrected base, respectively, with a median of 6,352 B in retained auxiliary arrays. It has lower paired MSE than the \delta -Adapter, COSA, FAC, and OMPB in a majority of conditions and uses less state than each. Full ELF achieves the largest mean MSE reduction, 19.29%, but its median retained state is 474,048 B ( \times 75 relative to EPOC). Equal-size summary controls favor the endpoint by 1.65–2.20% in paired MSE; a coefficient-reconstructed endpoint yields similar accuracy to the observed endpoint, highlighting its role as a shared input. Increasing the retained DCT component count from 4 to 8 adds 1.00 percentage point of MSE reduction for 5,728 B. On jointly trained bases, EPOC lowers MSE by 16.69–20.15% relative to globally blended TEFL-style adapters applied to the same base. The code and numerical records are available at this https URL.
[LG-51] CacheReforge: Bounded Recovery for Stale KV Caches under Evolving Adapters
链接: https://arxiv.org/abs/2609.30884
作者: Yuhang Cao,Yanzhou Mu,Chunrong Fang,Zhenyu Chen
类目: Machine Learning (cs.LG)
*备注: 13 pages, 5 figures. Artifact: this https URL
Abstract:Large language models rely on KV caching to reduce repeated prefill computation in long context and interactive applications. As lightweight adapters evolve, cached states reflect earlier versions, so stale reuse distorts current model outputs, while complete affected suffix recomputation restores fidelity at substantial cost. We seek minimal recomputation that recovers current adapter behavior. Existing systems track token, context, or stable adapter identity, but neither represent caches from earlier adapter versions nor distinguish update propagation from the recomputation required for behavioral recovery. To address these gaps, we introduce CacheReforge, which represents stale KV caches as layerwise mixed-version objects. It combines per-layer adapter anchors, calibrated sensitivity, accumulated drift, and executable restart boundaries to select direct reuse, bounded recomputation, or complete affected-suffix recovery. We distinguish dependency depth from the functional recomputation horizon and use cumulative tail influence to characterize when bounded recovery preserves current-model behavior. We evaluate CacheReforge on Qwen2.5-1.5B and Qwen2.5-7B with continual LoRA updates, including 16K HotpotQA and 2WikiMQA workloads. CacheReforge reduces mean KL divergence by 92.4% relative to stale reuse, while recomputing only 5.44% of layers and reducing cache-maintenance time by 93.2% relative to fresh full prefill. These results show that version-aware recovery preserves model fidelity and most KV caching gains.
[LG-52] AC Power Flow Contingency Analysis Using a Single Deep Neural Network
链接: https://arxiv.org/abs/2609.30859
作者: Md Obaidur Rahman,Junjie Qin,Vassilis Kekatos
类目: ystems and Control (eess.SY); Machine Learning (cs.LG)
*备注: 10 pages, 5 figures
Abstract:Contingency analysis using the AC power flow (AC-PF) model is a critical tool for accurate grid security assessment, but its computational burden increases with the number of operating scenarios and outage configurations to evaluate. Recent ML-based approaches typically require outage-specific training data, leading to offline training costs that scale with the number of contingencies. This work proposes a framework that reuses a single ML model trained solely on basecase AC-PF data to estimate post-contingency operating states under arbitrary single-line outages. The proposed approach formulates post-contingency state prediction as a fixed-point iteration. If the ML model is a deep neural network (DNN), we derive sufficient conditions that guarantee convergence and develop semidefinite programming (SDP) formulations to certify these conditions for a given DNN. Numerical tests on the IEEE 118-bus system demonstrate that the proposed SDP formulations are tight, that the certified conditions hold for all tested contingencies, and that the resulting method produces accurate post-contingency state estimates within only a few iterations.
[LG-53] Learning Chance-Constrained MDPs with Bellm an Distributional Certificates NEURIPS2026
链接: https://arxiv.org/abs/2609.30856
作者: Chenbei Lu,Hongyu Yi
类目: Machine Learning (cs.LG)
*备注: Accepted to NeurIPS 2026
Abstract:Safe reinforcement learning (RL) commonly enforces expected-cost constraints, but such expectation safety may fail to control the probability of rare high-cost trajectories. Chance-constrained MDPs (CCMDPs) impose a stronger probability-level requirement, but are widely viewed as harder because the chance constraint is nonconvex and depends on the full trajectory rather than a Bellman-linear expectation. In this paper, we reveal that this computational difficulty does not necessarily imply a higher statistical price. For tabular discounted CCMDPs with fixed bounded successor support and access to a certified planning oracle, we establish a model-based upper bound, with a matching lower bound up to logarithmic terms. Technically, our key idea is the \emphBellman distributional certificate, which constructs a Bellman recursion for constraint violation probabilities before policy selection. The certificate can be reused across candidate policies; combined with shared row-wise reverse-KL confidence sets, it gives a policy-uniform trajectory-KL transfer without a union bound over policies or time–budget Bellman tables. For stochastic policies, we give a model-free variance-reduced policy-gradient algorithm with a finite-sample expected KKT-residual guarantee and independent validation of every accepted policy. Numerical experiments on synthetic CCMDPs and an IEEE 14-bus energy storage control benchmark illustrate the safety and mechanism behavior of the proposed algorithms.
[LG-54] he KV Cache Is the New Memory Wall
链接: https://arxiv.org/abs/2609.30854
作者: Tejinder Singh
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG); Performance (cs.PF)
*备注: 28 pages, 12 figures, 12 tables
Abstract:Autoregressive LLM inference at long context is bounded by memory bandwidth, not arithmetic throughput, and the binding resource shifts from model weights to the Key-Value (KV) cache as sequence length grows. For Llama-3-70B in BF16, the 140 GB weight footprint exceeds the 80 GB HBM of a single accelerator, and one 128k-token sequence adds 42 GB of KV cache. Techniques that compress, evict, page, share, or offload KV state have proliferated, but reported gains use inconsistent workloads, hardware, and quality metrics, preventing cross-paper comparison. This SoK paper unifies the field analytically, with a protocol that strictly separates derived and reported claims. We derive closed-form arithmetic intensity as a decaying function of context length, parameterized by hardware topology for NVIDIA H100, NVIDIA B200, and AMD MI300X, including per-die bandwidth partitioning and the crossover lengths where KV traffic overtakes weight traffic. We classify the literature into five domains, quantization, token eviction, KV paging, prefix caching, and heterogeneous tiering, evaluating one method per domain at 128k context under a single protocol. The central finding is a three-regime structure: below a hardware-specific crossover, weight traffic dominates and KV compression yields negligible speedup; beyond it, KV traffic dominates and each domain trades quality for bandwidth savings approaching the roofline bound. Paging and prefix sharing are lossless but address capacity, not bandwidth. Quantization and eviction cut bandwidth directly, with degradation that accelerates below 4-bit precision and turns discontinuous for eviction on position-sensitive tasks. Tiering converts the bandwidth wall into an interconnect problem bounded by PCIe or NVLink rather than HBM. We close with design rules for selecting a compression domain given hardware, context length, and quality budget.
[LG-55] Attention-Based Adaptive Policies for Simultaneous Speech-to-Text Translation ICASSP2027
链接: https://arxiv.org/abs/2609.30839
作者: Filip Tăşădan,Ema Tomanová,Ondrej Lopuch,Paweł Bilko,Anders Søgaard
类目: Machine Learning (cs.LG)
*备注: Submitted to ICASSP 2027
Abstract:Simultaneous speech-to-text translation (Simul-S2TT) consists of generating partial translations while the incoming audio frames are processed by the system. However, the streaming nature of this setup creates the challenge of deciding the best moment to perform an accurate translation while minimizing the delay. To address this challenge, we utilize the cross-attention mechanism of the encoder-decoder architecture to find the right alignment between the input speech frames and the target text tokens. In this paper, we propose the Recent Frame Attention Policy (RFAP) and the Dual-Condition Attention Policy (DCAP) that allow offline trained speech-to-text translation models to be used in streaming scenarios without requiring additional training. Results on three different language translation pairs over the CVSS-C corpus show that the RFAP is able to surpass other policies with gains of up to 4.0 BLEU while reducing the translation delay by almost 1 second. Moreover, the DCAP is able to preserve a high translation quality when the latency is very low.
[LG-56] Peer-Grounded Counterfactual Path Planning for Chronic Health Management
链接: https://arxiv.org/abs/2609.30838
作者: Saman Khamesian,Hassan Ghasemzadeh
类目: Machine Learning (cs.LG)
*备注:
Abstract:Effective behavioral intervention in chronic disease management requires not a single prescription but a sequence of incremental steps, each grounded in what real, similar individuals have demonstrably achieved. Counterfactual explanation offers a natural computational route to such guidance, answering what change in behavior would have produced a better outcome. But existing methods return a target state without a route to it, guarantee no monotone health improvement along the way, and draw no evidence from peer behavior – asking a patient to close a wide gap in one move, which is precisely the recommendation structure least likely to be attempted. We propose POROS (Peer-Grounded Optimal Routes Over States), a domain-agnostic framework rooted in Bandura’s self-efficacy theory and Festinger’s social comparison theory that constructs a Behavioral Progression Graph – a directed acyclic graph over observed patient states in which every edge requires both peer-grounded behavioral proximity and strict health outcome improvement. Every edge is therefore a behavioral change that individuals in the cohort have demonstrated is achievable within a single period. Minimum-cost paths through this graph decompose otherwise inactionable behavioral gaps into incremental, peer-grounded steps. We evaluate POROS on two independent longitudinal cohorts of patients with diabetes. For patients below the 70% clinical threshold for time in range (TIR, blood glucose within 70-180 mg/dL), it reduces the mean gain required per step from 26.3 percentage points (pp) to 5.5 pp on one cohort and from 31.1 pp to 5.7 pp on the other, decomposing large behavioral jumps into the incremental steps that self-efficacy requires. Across both cohorts, 97-98% of multi-hop paths cross patient boundaries, embedding social comparison by construction.
[LG-57] Adaptive Interaction Graphs for Particle Simulation ICML2026
链接: https://arxiv.org/abs/2609.30822
作者: Aiden Zhou
类目: Machine Learning (cs.LG); Computational Engineering, Finance, and Science (cs.CE)
*备注: 6 pages, 3 figures. Presented at ICML 2026 Workshop on AI for Physics
Abstract:Learned particle simulators based on graph neural networks achieve strong one-step accuracy, but errors compound over long horizons. An underexplored variable is the interaction graph: existing methods fix its topology via k-nearest neighbors or a static radius rule, regardless of local model confidence. We propose making this graph adaptive: a per-particle variance head, trained jointly with the acceleration head under a heteroscedastic Gaussian NLL loss, drives a trajectory in which high-uncertainty particles receive an expanded neighborhood. This is done at little extra inference cost by using the previous step’s uncertainty estimate. A key discovery is that the variance head learns a meaningful notion of uncertainty: high-variance particles concentrate near complex regions, such as splash zones or free surfaces. When this signal drives graph topology, the resulting AdaptGNS simulator achieves a strict Pareto improvement on WaterDrop and a modest gain on Sand. Given the model’s stronger performance on WaterDrop, we hypothesize that adaptive graphs are most useful when complexity is concentrated in space. Our code can be found at this https URL.
[LG-58] Learning Provable Neural Network Observer for Uncertain Dynamical Systems
链接: https://arxiv.org/abs/2609.30819
作者: Zhangyi Wang,Jiaxu Liu,Chen Song,Chao Xu,Shengze Cai
类目: Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注:
Abstract:In many safety-critical applications, control of uncertain dynamical systems relies on observers that estimate states and external disturbances. Neural network observers can improve estimation accuracy, but certifying their Lyapunov stability via Linear Matrix Inequality (LMI) constraints leads to large-scale semidefinite programs (SDPs) that are difficult to solve for large networks. To overcome this scalability bottleneck, we propose a novel two-stage training framework for provably stable neural network observers. Our approach decouples the optimization into a point-guided Lyapunov pre-training phase, which rapidly achieves high estimation accuracy and local stability over sampled states, followed by an LMI fine-tuning phase that efficiently satisfies a strict global Lyapunov stability certificate. We provide formal theoretical guarantees for local stability radii and probabilistic coverage over a prescribed compact error-state domain under specified regularity and sampling assumptions. Experiments on nonlinear control benchmarks and X-29 aircraft ablations show that our LMI-certified neural network observers train significantly faster than direct LMI-based methods and generalize robustly across diverse systems, achieving improved tracking accuracy over a range of observer baselines. The code is available at this https URL.
[LG-59] Counterfactual Online Conformal Prediction Under Adaptive Logging
链接: https://arxiv.org/abs/2609.30811
作者: Xinyu Qiao,Yichen Lin,Kaihong Ji,Xue Wang,Tao Yao
类目: Machine Learning (cs.LG)
*备注:
Abstract:Online conformal prediction can fail when predictions shape actions and actions determine which outcomes enter calibration. Standard adaptive methods may retain marginal coverage while systematically miscovering the counterfactual outcomes of rarely selected actions. This paper formalizes the failure through counterfactual coverage and introduces Propensity-Weighted Online Conformal Prediction, an inverse-propensity-weighted recursion that debiases calibration. A doubly robust variant further reduces nuisance bias to the product of outcome-model and propensity errors. Under positivity, the resulting coverage rate matches an information-theoretic lower bound up to logarithmic factors. Experiments on synthetic decision tasks, open bandit data, and financial rebalancing show that PW-OCP and DR-OCP improve counterfactual coverage and downstream regret without sacrificing prediction-set sharpness.
[LG-60] Deep-Learning Solvers and Surrogates for Infinity and p-Laplace Problems
链接: https://arxiv.org/abs/2609.30809
作者: Tak Shing Au Yeung,Ka Chun Cheung,Hannah Potgieter,Steven J. Ruuth,Simon See
类目: Numerical Analysis (math.NA); Machine Learning (cs.LG)
*备注:
Abstract:We investigate the use of neural network solvers for infinity and p -Laplace problems, which are fundamental in nonlinear analysis and have practical applications. Our approach employs Physics-Informed Neural Networks (PINNs) and Deep Operator Networks (DeepONets) to address computational challenges associated with large p values, ranging from 2 to 1000 , on various 2D and 3D domains. Our method offers advantages over traditional physics-based solvers, especially in three dimensions where mesh-based solvers become very costly for these problems. We also establish conditional convergence results for PINN approximations of both problems and a universal approximation result for DeepONet on the parametric p -Poisson problem. We demonstrate the effectiveness of these neural network solvers through numerical experiments and compare their performance with conventional methods.
[LG-61] owards Universal Representation-Based Process Control
链接: https://arxiv.org/abs/2609.30790
作者: Jinmyeong Choi,Taesup Kim,Artur Dubrawski
类目: Machine Learning (cs.LG)
*备注:
Abstract:Many temporal process learning and monitoring pipelines operate in local windows, making window-level decisions unavoidable in practice. In such settings, classical statistical tests can be applied to individual windows, but they typically evaluate predefined parametric hypotheses-such as unit-root or moment-based conditions-thereby limiting flexibility when reference behavior is defined empirically from task- or domain-specific data. In this work, we view window-level monitoring as a process control problem and reformulate it as reference-based hypothesis testing, where the null hypothesis is specified by an empirical reference distribution rather than a fixed parametric model. We operationalize this perspective through a representation-based, nonparametric framework that combines pretrained time series encoders, kernel density estimation, and conformal calibration, yielding finite-sample valid inference in learned representation space. Classical notions such as stationarity and cyclostationarity arise as natural instantiations of empirical reference sets within this framework. Through experiments, we demonstrate sensitivity to window-level distributional deviations while maintaining well-calibrated inference under stable reference regimes, highlighting the applicability of the proposed approach to a broad class of time series process control and monitoring tasks.
[LG-62] Interpretable-by-Design Descriptor Portfolios Match a 2048-Dimensional Foundation Embedding on Low-Data Molecular Assays
链接: https://arxiv.org/abs/2609.30789
作者: Yiqi Yao,Miquel Duran-Frigola
类目: Machine Learning (cs.LG)
*备注: 16 pages, 3 figures, 9 tables, 8 appendices. Under review. Code: this https URL
Abstract:In low-data structure-activity prediction, the choice of molecular representation can matter more than the choice of predictor, and tabular foundation models sharpen that effect. We ask whether a portfolio of compact, semantically named descriptor blocks can reach the accuracy of a 2048-dimensional CheMeleon embedding while staying auditable at the feature level, meaning that every input dimension carries a model name and a recorded training provenance. Starting from a fixed 11-dimensional physicochemical base, we greedily concatenate provenance-screened blocks using the labelled context alone. Across nine ADME/Tox assays and 50 evaluation cells, scored on common-coverage subsets restricted to the molecules that every representation covers, the portfolio reaches a mean test AUC of 0.762, against 0.764 for CheMeleon and 0.756 for Mordred. The pooled gap to CheMeleon is +0.003 AUC (task-bootstrap 95% CI [-0.020, +0.030]), which satisfies our predeclared pooled parity gate but not the per-assay gate. At 25 context labels the headline rule again satisfies the pooled gate; at 10 labels it does not. We also report four predeclared candidate-selection rules that we falsified. Post-freeze checks over ten seeds and three previously unseen assays support pooled competitiveness for compact, auditable representations; a same-width random-bundle control does not establish that greedy membership itself adds accuracy. Assay-level differences remain unresolved.
[LG-63] Missingness-Aware Conformal Prediction Under Cross-Hospital Distribution Shift
链接: https://arxiv.org/abs/2609.30781
作者: Liang You,Dongwen Ou,Hengyu Shi,Siyuan Dai
类目: Machine Learning (cs.LG); Applications (stat.AP)
*备注: 33 pages, 2 figures
Abstract:Clinical measurements are recorded for some patients but not others, at rates that differ across hospitals, and marginal conformal coverage does not ensure coverage within groups defined by missingness. We propose a missingness-aware conformal calibration procedure for mortality prediction under cross-hospital distribution shift. It selects a measurement on an independent sample, groups patients by whether that measurement is recorded, and applies Mondrian calibration within each group, so no calibration outcome is reused. We evaluate the procedure across hospitals in eICU and across care units within one MIMIC-IV hospital, using three predictors. Relative to pooled calibration, it reduces the average worst-group coverage gap on its selected groups in all six settings, with a median reduction of 1.9 percentage points; paired site-bootstrap intervals exclude zero in five. These gains do not extend uniformly. Calibration by predicted risk achieves smaller gaps on a broader panel of missingness groups, and when eICU hospitals are evaluated separately, the gain shrinks for all three predictors and reverses in sign for one. We explain this discrepancy with a hospital-level decomposition. Pooling reweights hospitals through a covariance between group shares and coverage errors, and lets errors of opposite sign cancel: weighting explains the reversal, and cancellation accounts for most of the attenuation for the other two predictors. Constructed population distributions show that pooled and within-hospital evaluations can rank calibration methods oppositely even without sampling noise. Pooled improvement alone therefore cannot establish better coverage within hospitals, even when the calibration groups are fixed.
[LG-64] Differentiable RNA Secondary Structure Extraction for Deep Learning
链接: https://arxiv.org/abs/2609.30752
作者: Tyler Illman,Max Ward,Marcell Szikszai,Ryan K. Krueger
类目: Machine Learning (cs.LG)
*备注:
Abstract:Many deep learning approaches to RNA secondary structure prediction have recently been proposed. They typically output a weight matrix W where W_ij is an arbitrary weight for base i pairing with base j . Converting this matrix to a predicted secondary structure or base-pairing probability matrix typically involves ad hoc and problematic downstream algorithms. Despite the importance of this conversion step, which we refer to as structure extraction, it has received relatively little attention in the literature. In this work, we analyze how the congruence between training and extraction methods affects prediction performance. To do this, we compare four extraction algorithms: a Nussinov-like dynamic programming method, maximum-weight graph matching and the greedy extraction algorithms used by SPOT-RNA and RiNALMo. These are evaluated on outputs from the pretrained RiNALMo model and three toy models trained in this paper: a differentiable Nussinov-like model, a binary cross-entropy (BCE) baseline, and a model that incorporates a novel symmetric doubly stochastic matrix (SDSM) normalization algorithm during training which allows it to output base-pairing probability matrices directly, without a separate extraction step. This SDSM normalization algorithm is differentiable and can be added inline to any deep learning model during training and evaluation. We find that the performance of each extraction method depends strongly on how the corresponding model was trained. Considering the toy models themselves, the SDSM model showed the strongest overall performance: it outperformed the BCE baseline under all four extraction algorithms and produced pre-extraction outputs closest to the ground truth. These results suggest that SDSM normalization is a tractable alternative to traditional structure extraction.
[LG-65] Mechanism-Aware Ensemble Conditioning for Data-Limited Emulation of Extreme Events NEURIPS2026
链接: https://arxiv.org/abs/2609.30746
作者: Isabella S. Thiel,Juan Bello-Rivas,Yannis G. Kevrekidis,Themistoklis P. Sapsis
类目: Machine Learning (cs.LG); Dynamical Systems (math.DS); Atmospheric and Oceanic Physics (physics.ao-ph)
*备注: Accepted to NeurIPS 2026 (oral)
Abstract:Extreme events in chaotic systems are difficult to learn from short trajectories because they are controlled by transient finite-time instability rather than by frequently observed bulk dynamics. We propose a mechanism-aware conditioning plug-in framework that turns a nudged coarse ensemble into a non-intrusive sensor of local instability geometry. In the small-noise regime, the ensemble covariance aggregates the same finite-time deformation kernels that govern local instability, providing a Jacobian-free proxy for the local amplification structure around a synchronized coarse trajectory. A small FiLM module injects statistics of this ensemble geometry into an otherwise unchanged backbone while leaving the coarse simulator unchanged. We demonstrate this interface in two distinct pipelines: a Transformer-style residual-attention corrector for a controlled low-dimensional chaotic system and a probabilistic recurrent STORN corrector for topographic two-layer quasi-geostrophic (QG) flow. In the low-dimensional benchmark, ensemble covariance directions co-activate with OTD modes and FiLM conditioning improves 99th-percentile exceedance-frequency errors over an identical no-context Transformer baseline. In QG, a fixed ensemble-conditioned FiLM-STORN model trained on only (50) time units substantially improves long-horizon rare-event statistics in the data-limited regime, including density-tail errors, exceedance frequencies, and spatial exceedance-area distributions relative to an unconditioned STORN trained on the same data; on averaged high-threshold exceedance diagnostics, it also outperforms the baseline STORN trained with 20 times more high-resolution data. These results show that local instability geometry is not merely interpretable post hoc, but an actionable conditioning signal for data-efficient rare-event emulation.
[LG-66] Input-Layer Starvation: Why Per-Layer Pruning Breaks IoT Intrusion Detectors
链接: https://arxiv.org/abs/2609.30729
作者: Md Anas Biswas
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注: 38 pages including a 15-page supplement; 12 main tables, 4 figures. Code and results: this https URL
Abstract:Intrusion detectors for small Internet-of-Things (IoT) devices are usually compressed by pruning and judged by overall accuracy. We show that this hides a severe class-level failure, find its cause, and give low-overhead prevention and repair. On CICIoT2023, a two-layer convolutional detector pruned with uniform layer-wise magnitude pruning at 80% sparsity loses 16 points of accuracy but half of its macro-F1, the mean per-class F1 (0.542 to 0.271 over five independently trained models); 17 of 34 classes are materially damaged. Remaining weight count does not explain it: a perceptron and a transformer pruned to the same or fewer weights lose at most 0.096. The first layer does. It has 192 weights; uniform pruning leaves 38, 46% of its 64 filters lose every input weight, and fine-tuning under that starvation leaves the running means of the first normalisation layer displaced by up to 0.8 standard deviations in a few surviving channels, on which the deployed model collapses. Protecting those 192 weights, or pruning globally at the same sparsity, prevents the collapse (loss 0.013); recomputing the normalisation statistics on unlabelled training data, with no weight changed, repairs it (loss 0.039) and returns the false-alert rate to 33% (dense 29%). Damage shows a strong increasing dose-response in first-layer sparsity, starving a perceptron’s input layer reproduces the collapse, and the pattern holds on TON_IoT. The failure is misattribution and false alerts, not silent evasion: on validation-selected blind spots, uniformly pruned detectors misattribute 72% of the traffic, against 50% with the first layer protected and 47% for the dense model.
[LG-67] When 10000 Windows Are Not 10000 Tests: Auditing Statistical Confidence in Sliding-Window Time-Series Classification ICTAI2026
链接: https://arxiv.org/abs/2609.30721
作者: Xinze Shi,Litian Zhang,Binrui Shi
类目: Machine Learning (cs.LG)
*备注: 8 pages, 6 figures, 6 tables. Accepted at the 38th IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2026)
Abstract:Sliding-window classifiers are often evaluated on thousands of overlapping test windows, even though neighboring predictions share observations and remain nested within recordings and subjects. Subject-disjoint evaluation prevents one form of leakage but does not make those test windows independent. We present a practical audit that maps three claims - performance on observed recordings, future recordings from observed subjects, and unseen subjects - to explicit aggregation rules and established dependence-robust inference. At 75% overlap, controlled simulations give 16.9% Type-I error for IID observed-record inference and 7.2% for session-centered Bartlett-HAC: a substantial improvement with residual miscalibration. Audits of frozen WISDM and HARTH predictions show that nearly fourfold growth in test rows yields only 1.75-1.94-fold variance-equivalent information growth. At that overlap, fixed-record paired Accuracy-difference intervals are 1.22-1.66 times the IID widths; this inflation is not universal at zero overlap. On HARTH, paired Accuracy-difference intervals include zero across three overlap settings, whereas Macro-F1 favors MiniROCKET. Independent recomputation, common-session checks, class-level results, and separately seeded calibration make the audit’s scope and limitations inspectable. The resulting workflow distinguishes additional predictions from additional independent evidence.
[LG-68] NEMSim: Learning Control-Conditioned Multi-Event Physical Dynamics via Executable Event-Mechanism Priors
链接: https://arxiv.org/abs/2609.30718
作者: Junsong Yu,Junjie Xie,Pengwei Liu,Dong Ni
类目: Machine Learning (cs.LG)
*备注: Main paper: 9 pages, 6 figures, 2 tables. Supplementary material included
Abstract:High-fidelity simulation of control-conditioned multi-event physical systems is computationally expensive, especially across broad control spaces and long trajectories. In these systems, macroscopic evolution emerges from localized discrete events whose intensities and effects depend on process controls and evolving local states, while the available system knowledge is typically expressed as event-attribute descriptions. Purely data-driven surrogates must infer these event effects from limited trajectory coverage, which can hinder generalization to unseen control regimes. Physics-guided methods instead primarily build on equation-level constraints or differentiable solvers rather than discrete event-rule priors. We therefore propose NEMSim (Neural Event-Mechanism Simulator), which compiles predefined event-attribute descriptions into an executable transition structure linking control-dependent event intensities, prior-guided mechanism attribution, and state-dependent responses. To enable evaluation of control-conditioned multi-event dynamics with explicit system knowledge, we construct a 3D KMC-based benchmark pairing high-fidelity trajectories with explicit event rules, standardized splits, and evaluation protocols. Across three settings, NEMSim reduces Avg. RMSE by 58.9%-81.3% relative to the strongest baseline in each setting. It also remains best in the data-efficiency study with training-data fractions down to 10%. Mechanism analyses further show that these gains arise from executable rule integration rather than prior access or architecture alone.
[LG-69] LUMO (Lightweight Unified Multilingual Orchestrator): A Privacy Preserving Offline Voice Assistant
链接: https://arxiv.org/abs/2609.30692
作者: Md. Mehedi Hasan Naeem,Mst. Kamrunnahar Ruma,Nafiza Anjum,Shakila Sultana,Md. Sujan Ali
类目: Machine Learning (cs.LG)
*备注: 6 pages, 8 figures, 9 tables. Conference version prepared for IEEE 3rd International Conference on Computing, Applications and Systems (COMPAS 2026), 9-10 October 2026, University of Dhaka, Bangladesh
Abstract:Reliable voice interaction is essential in environments with limited internet connectivity and strong privacy. However, most existing voice assistants depend on cloud-based services, which leads to latency issues, dependency on internet access, and privacy vulnerabilities. This research presents LUMO (Lightweight Unified Multilingual Orchestrator), a privacy preserving offline voice assistant designed for edge computing environments. This system integrates local Automatic Speech Recognition (ASR), locally deployed quantized Large Language Model (LLM), and Text-to-Speech (TTS) synthesis into a fully offline pipeline running on a Raspberry Pi 5 with 8 GB RAM. To enable efficient operation on resource constrained hardware, the language model is compressed using 4-bit GGUF quantization, which reduces memory usage while preserving practical conversational capability. Existing edge based voice assistants Mycroft provides partial offline functionality without a generative LLM, with an approximate latency of ~5 s and power consumption of ~12 W, while Rhasspy supports full offline operation but lacks generative capabilities, with ~3 s latency and ~11 W power usage. In contrast, LUMO achieves a Word Error Rate (WER) of 6.8% for short English utterances in low noise conditions, an end-to-end response latency of 2.0-4.0 s, and a lower peak power consumption of approximately 9.0 W. The system also achieves effective offline recognition for Bangla speech, supporting multilingual accessibility in low resource settings. By operating entirely offline, LUMO provides strong data privacy, reduced need for cloud connectivity, and suitability for privacy sensitive edge execution such as rural healthcare, education, and disaster response scenarios. Comments: 6 pages, 8 figures, 9 tables. Conference version prepared for IEEE 3rd International Conference on Computing, Applications and Systems (COMPAS 2026), 9-10 October 2026, University of Dhaka, Bangladesh Subjects: Machine Learning (cs.LG) Cite as: arXiv:2609.30692 [cs.LG] (or arXiv:2609.30692v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.30692 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-70] PixSim: a calibrated open-source simulator of instant-payment fraud recovery and interdiction under analyst capacity constraints
链接: https://arxiv.org/abs/2609.30684
作者: Bashir Zeimarani,Alireza Khatib,Somayeh Mousavinasr,Carlos Maurício Serodio Figueiredo
类目: Machine Learning (cs.LG); Computational Engineering, Finance, and Science (cs.CE); Risk Management (q-fin.RM)
*备注: 25 pages, 12 tables, 1 figure. Code: this https URL
Abstract:Brazil’s Pix settles about 5.9 billion instant, irreversible transfers a month. A fraudulent transfer can be recovered only while the funds remain in a traceable account, and in 2025 the Central Bank’s recovery mechanism (MED) returned 9% of accepted contested value. Interdiction therefore has to happen before settlement, by routing each transaction to pass, human review or block, under a finite analyst team and a regulatory hold window. To our knowledge no public simulator jointly models irreversible settlement, a regulated recovery mechanism, downstream fund dispersal and capacity-constrained review. We present PixSim, an open-source simulator of the Pix rail with these elements, calibrated to Banco Central do Brasil open data, with every parameter sourced, calibrated to one published observable, or registered as an assumption. With the model frozen, full-scale runs reproduce the 2025 recovery rate within 0.006 and its decomposition within 0.02; the February-April 2026 window is reported as a misfit and the May 2026 tracing regime as a projection. On a benchmark with a payer-side scorer, four reference policies and ten scenarios, within the simulated mule model: recovery after settlement is constrained by dispersal speed; staffing by the arrival profile cuts a fixed rule’s alert expiry from 52% to 2% at constant hours; halving the team removes a fixed threshold-and-block rule’s advantage over a queue-aware rule, on loss and on loss plus false-block harm (+0.106 of victim value, positive on all twenty paired seeds), while a reversal at two thirds of the team was not confirmed on independent seeds; and a synthetic scorer of held-out AUC 0.82 cuts lost value by about a quarter. Code and data: this https URL
[LG-71] Population loss in shallow ReLU networks: Bias families of critical points
链接: https://arxiv.org/abs/2609.30661
作者: Michael Field
类目: Machine Learning (cs.LG)
*备注: 118 pages, 3 figures
Abstract:The main result presented is a formula for the population loss in the student-teacher kernel model that is applicable to shallow ReLU networks with bias. This extends previous work of Choo and Saul (2009) and Brutzkus and Globerson (2017). The formula makes essential use of Owen’s T-function. The necessary theory of the T-function is given and a high precision coding using MPFR for the T-function, based on an algorithm of Komelj (2023), is available on request. It is shown that various families of spurious minima described in past papers of Arjevani and the author extend to biased networks and that the loss is always strictly decreased when bias is added. The change in landscape geometry caused by adding bias appears to be relatively mild. Only the simplest examples are described in this paper where it is assumed that the number of inputs is equal to the number of neurons (this restriction is for reasons of length). A review of relevant previous results on unbiased networks is included. Aside from Gaussian statistics, the main mathematical tools and ideas come from analytic geometry (analytic and subanalytic sets, the Curve Selection Lemma).
[LG-72] DiffusionShadow: Diffusion-based Shadow Caching for Neural Volume Rendering
链接: https://arxiv.org/abs/2609.30658
作者: Kai-Chen Tung,Qi Wu,David Bauer,Mengjiao Han,Silvio Rizzi,Kwan-Liu Ma
类目: Graphics (cs.GR); Machine Learning (cs.LG)
*备注: 13 pages, 6 figures
Abstract:Implicit neural representations (INRs) have gained momentum in scientific visualization due to their compactness and scalability to large datasets, making them well suited for integration with direct volume rendering (DVR). However, real-time volume rendering of INR with advanced illumination effects, such as shadows, remains computationally expensive, as evaluating shadow terms via ray marching is costly. Alternatively, precomputing and storing shadows for many lighting directions is prohibitive in both memory and storage. To address this, we introduce a diffusion-based shadow caching framework that compresses a vast set of pre-calculated shadow INRs into a single diffusion model. Rather than focusing on generalizing to unseen directions, our method effectively memorizes and reconstructs a dense set of pre-trained lighting conditions on the fly. We first encode a collection of shadow coefficient volumes as shadow INRs, and then train a diffusion model conditioned on lighting direction to predict the corresponding shadow INR weights at inference time. This design integrates directly with standard INR renderers without additional runtime sampling. Experiments show that our approach achieves faster rendering than traditional methods while bypassing the massive storage bloat of independent INRs, producing shadows that closely match most of the reference results.
[LG-73] In-Context Binding Capacity in Language Models
链接: https://arxiv.org/abs/2609.30634
作者: Manas Venkata Sai Ravulapalli,Samrath Singh Chadha
类目: Machine Learning (cs.LG)
*备注:
Abstract:How many assignments can a language model recall before it loses track of which value belongs to which entity? We measure this limit using continuous recall curves for 12 models at or below 3B parameters and a threshold sweep over 30 open models up to 12B. On the continuous curves, the load at which recall falls halfway to chance follows K_50=cN^\alpha , with \alpha=0.820 and R^2=0.73 . The broader sweep shows an eightfold range associated with pretraining recipe, although the continuous curves show no detectable recipe effect after controlling for scale, with few modern models in the fit. We derive why interference can lower measured capacity by reducing single-binding recall even when the load-dependent recall profile is unchanged. Direct task training also exceeds the extrapolated zero-shot law, but different measurement criteria prevent interpreting that comparison as a capacity gain. Its formation times follow a power-law form in two independent codebases, conditional on runs that succeed. Together, these results characterize capacity at the model’s query interface. Bounds on joint recall and a decomposition of policy errors connect this measurement to working memory and instruction following, without treating recall as a measure of alignment. The controlled task also provides a baseline for testing whether binding limits constrain world-state tracking; the present experiments do not measure state updates or downstream transfer.
[LG-74] Stable initialization without the CLT
链接: https://arxiv.org/abs/2609.30633
作者: Simon Kuang,Kyle Chickering,Xinfan Lin
类目: Machine Learning (cs.LG)
*备注: reproduction code in ancillary material
Abstract:Successful training of deep neural networks is highly dependent on the distribution of the initial weights. If the weights are too large, network training blows up; if they are too small, the model fails to learn features. Stable initialization is the optimal moderation between these two extremes. The conventional theory of random networks uses the Central Limit Theorem to control inter-neuron dependencies, which introduces distributional approximation error and coupling between layers. For networks with sine activations, we derive the uniform-phase initialization, which obviates distributional approximation and fully decouples the layers. Ours is the first work to use the sine function’s periodic symmetry. Models trained with the uniform-phase initialization outperform the state of the art in neural representation tasks like image and audio fitting. We find that our untuned models are competitive with the best-tuned baselines from previous work and support \mu P width scaling.
[LG-75] OpenHail: An Event-Driven Gymnasium Environment for Electric Ride-Hailing Fleet Control
链接: https://arxiv.org/abs/2609.30628
作者: Tommaso Schettini,Nicholas D. Kullman,Jorge E. Mendoza
类目: Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注: 10 pages, 3 figures, 5 tables. Source code available at this https URL
Abstract:Machine-learning policies have attracted increasing interest for ride-hailing fleet control in recent years. Reinforcement learning, in particular, requires a structured simulation environment that specifies observations, actions, rewards, and decision epochs for training and evaluation. For electric fleets, this environment must also capture the interaction among stochastic demand, vehicle operations, and capacitated charging infrastructure. We present OpenHail, an open-source Gymnasium environment for joint control of electric ride-hailing fleets. Its fixed-size observation–action interface exposes request assignment, repositioning, and charging to a single policy. The event-driven simulator represents requests with pickup deadlines, vehicle job queues, battery dynamics, and finite-capacity charging facilities with first-in–first-out queues. A configurable decision-epoch mechanism separates internal simulator events from policy interactions, supporting event-driven, periodic, hybrid, and policy-requested control within the same operational model. The software provides seeded instances, feasible-action utilities, evaluation tools, operational metrics, and baseline policies. The source code is available at this https URL.
[LG-76] Probabilistic Robustness-driven Universal Adversarial Perturbations with Explainability against Deep Reinforcement Learning-based Intrusion Detection System
链接: https://arxiv.org/abs/2609.30605
作者: Hongsen Zhang,Lu Zhang,Mingjing Xu,Yi Zhang,Gregory Epiphaniou,Carsten Maple
类目: Machine Learning (cs.LG)
*备注: 20pages,10figues
Abstract:Deep reinforcement learning (DRL) enables adaptive intrusion detection in dynamic network environments but also exposes intrusion detection systems (IDS) to adversarial threats such as universal adversarial perturbations (UAPs), which apply a single input-agnostic perturbation to degrade detection performance across traffic. Probabilistic Robustness (PR), as a post-hoc evaluation metric, provides a principled, population-level measure of adversarial impact that conceptually aligns with the universality objective of UAPs, i.e., PR quantifies the prevalence of misclassification in the input space, making it a natural signal for guiding UAP generation. Hence, we propose PR-based UAP, which represents the first integration of an explicit PR-driven objective into generating UAPs against DRL-based IDS. Building on this formulation, we introduce PX-UAP, which leverages explainable artificial intelligence (XAI) to guide perturbation shaping under realistic domain constraints, and provides a rigorous theoretical analysis of its design. Extensive experiments demonstrate that PX-UAP consistently outperforms state-of-the-art UAP methods in attack effectiveness.
[LG-77] Energy-efficient operation of neural operators for virtual sensing
链接: https://arxiv.org/abs/2609.30580
作者: Jason Yoo,Samrendra Roy,Souvik Chakraborty,Syed Bahauddin Alam
类目: Machine Learning (cs.LG); Performance (cs.PF)
*备注:
Abstract:Virtual sensing repeatedly reconstructs physical fields from changing observations, often on a fixed geometry. We investigate how shared spatial computation reduces the energy of these updates while retaining the selected checkpoint and its evaluated predictions. In a heat-exchanger service, standard compiler freezing and explicit trunk reuse give similar operating energy reductions relative to graph replay: approximately 1% at one request per second and 20% at forty requests per second. In 15 W mode with fixed clocks, reuse with graph replay completes the same request sequence with 22.0 to 22.5% less energy than eager execution, including preparation and waiting. DeepONet and Fourier neural operator (FNO) controls distinguish the effects of reusable arithmetic and launch overhead. Preparation, artifact construction, and worker replacement add costs outside repeated inference. These results connect operator structure to operating energy and show how update frequency and execution lifetime govern the benefit of computation reuse in physical-field virtual sensing.
[LG-78] Reinforcement Learning of Communication in a Mesh of Small Language Models
链接: https://arxiv.org/abs/2609.30578
作者: Mehmet Kerem Turkcan
类目: Machine Learning (cs.LG)
*备注:
Abstract:Language models gain accuracy from more compute at test time, but majority voting over independent samples saturates: as samples grow, the vote converges to the model’s most frequent answer. Communication can add what sampling cannot: an agent that solves a problem can pass the key step to the others. We present TalkMesh, a decentralized mesh of small language model agents that learns when and what to communicate. Each agent samples a proposal and scores it with a trained confidence head. The most confident agent broadcasts a hint; agents below a confidence threshold revise, keeping each revision that outscores its proposal. Gossip consensus approximates the vote weighted by confidence without a coordinator. A talk policy, trained with group relative policy optimization on the change in correctness after revision, writes hints and revisions. With three agents, which together generate at most six outputs, the mesh reaches the accuracy of majority voting over 32 samples with each of three models. Trained with at most 8 agents and evaluated with 32, it raises accuracy from 0.568 under self-consistency to 0.705 (Qwen3.5-0.8B, GSM8K) and from 0.492 to 0.722 (SmolLM3-3B, MATH-500). When 4 of 8 agents collude on a wrong answer with fabricated confidence and poisoned hints, majority vote accuracy falls to 0.000 (Qwen3.5-0.8B, GSM8K). A defended mesh, whose agents rescore solutions with their own confidence heads, retains 0.507. Across reasoning, embodied coordination, and traffic signal control, messages improve a decision when the acting agent cannot observe the information it requires and another agent can send it.
[LG-79] Entropy Regularization: A Free Correction to Cross-Entropy for Verified Demonstrations
链接: https://arxiv.org/abs/2609.30572
作者: Mihir Dhanakshirur,Adam Ousherovitch,Ambuj Tewari
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:Large language models are often post-trained on expert demonstrations using cross-entropy (CE), even when the downstream objective is not to imitate the demonstrated solution but to produce any output accepted by a verifier. This mismatch is seen in verifiable domains with multiple correct solutions, such as mathematical reasoning and code generation, where training data may contain only one expert solution per problem. We show that minimizing cross-entropy can be misaligned with minimizing verifier risk; two policies can assign identical likelihood to the observed demonstrations while placing different probability mass on incorrect outputs. This is formalized through a learning-theoretic counterexample in which CE minimization selects a suboptimal policy. We identify that controlling the support of the learned policy can solve this problem by preventing probability mass from spreading to unsupported outputs. Since support size is non-differentiable and computationally intractable, we propose entropy-regularized cross-entropy (ER-CE), using token-level Shannon entropy as a tractable proxy. Finally, across mathematical reasoning and code-generation benchmarks, we find that entropy-regularized training consistently improves verifier accuracy over standard cross-entropy. Our results identify a simple failure mode of imitation-based post-training in verifiable tasks and provide a practical objective that is better aligned with producing correct outputs.
[LG-80] Dynamic Regret in Online Convex Optimization with Indicator Switching Costs NEURIPS2026
链接: https://arxiv.org/abs/2609.30556
作者: Naram Mhaisen,George Iosifidis
类目: Machine Learning (cs.LG)
*备注: To appear in the proceedings of NeurIPS 2026
Abstract:We study dynamic regret in online convex optimization with an \emphindicator switching cost: a fixed penalty incurred whenever two consecutive decisions differ. This captures startup overheads such as server activation, model deployment, and cache updates, and on a bounded domain it recovers norm-based movement costs as a special case. Existing guarantees for indicator costs handle only static comparators. We show that a direct extension of these techniques to dynamic regret provably fails, motivating a different approach. We propose a meta-learning framework: a set of randomized lazy FTRL base learners restarted at dyadic time scales, aggregated by a movement-aware master that mixes their proposal densities and samples actions via maximal coupling of consecutive mixtures. The resulting algorithm satisfies, in expectation, \mathcalR^\mathbf1_T \le \tilde\mathcalO(\min\sqrtT(S_T+1),T^2/3(P_T+1)^1/3) , where \mathcalR^\mathbf1_T is the dynamic regret plus the cumulative indicator switching cost, S_T counts comparator switches, and P_T is the comparator path length. The bound holds simultaneously for all sequences and requires no prior knowledge of S_T or P_T : it is minimax-optimal (up to logarithmic factors) for tracking piecewise-constant comparators, and also captures frequently moving comparators with small total path length.
[LG-81] GyroNovo: Error-Guided Frag ment Imputation with Mass-Aware Attention for textitDe Novo Peptide Sequencing
链接: https://arxiv.org/abs/2609.30542
作者: Abdellah El Mekki,Laks V.S. Lakshmanan,Muhammad Abdul-Mageed
类目: Machine Learning (cs.LG)
*备注: Code available at this https URL
Abstract:De novo peptide sequencing from tandem mass spectra is essential for identifying peptides without relying on reference databases. Despite advances in deep learning, accurate sequencing remains challenging because experimental spectra are often sparse, noisy, and incomplete, leaving informative b- and y-ion fragments unobserved. Existing methods attempt to recover this missing evidence via latent-space imputation before autoregressive decoding. However, they typically treat imputation as a fixed reconstruction task, without considering which missing fragments are most relevant to decoder errors. Moreover, existing peak representations do not explicitly model mass differences between peaks, despite their fundamental importance. We introduce GyroNovo, a framework with two main contributions. First, we use decoder errors observed during training to adapt the imputation objective, prioritizing fragments associated with frequent decoding errors. We further use the decoder error distribution to construct easy and hard augmented views of each spectrum, enabling the decoder to learn under varying degrees of spectral corruption and missing-fragment severity. Second, we introduce a mass-aware inductive bias into self-attention by using rotary embeddings to encode pairwise mass differences between spectral peaks. Together, these components align missing-fragment recovery with decoder behavior while explicitly incorporating the mass relationships that underlie peptide fragmentation. At inference time, GyroNovo retains a standard encoder-imputer-decoder architecture and requires neither additional inputs nor auxiliary search procedures. Experiments on NovoBench show gains of about 9 percentage points in peptide-level precision and 7 percentage points in amino-acid-level precision over the state-of-the-art baseline. Code: this https URL. Comments: Code available at this https URL Subjects: Machine Learning (cs.LG) Cite as: arXiv:2609.30542 [cs.LG] (or arXiv:2609.30542v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.30542 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-82] Benchmarking the Connectomes of Caenorhabditis elegans within the Reservoir Computing Framework
链接: https://arxiv.org/abs/2609.30508
作者: Felix S. Reimers,Ola Huse Ramstad,Aliaksandr Hubin,Stefano Nichele
类目: Machine Learning (cs.LG)
*备注:
Abstract:The aim of this work is to examine the connectomes of Caenorhabditis elegans through a computational lens using the reservoir computing framework. Connectomes are mappings of biological neural networks; C. elegans is the first organism for which physical connectomes covering the whole nervous system have been published. The connectomes of C. elegans used in this paper have been derived at different ages of the organism and are based on three different ways of measuring inter-cellular connections. They have, with minimal preprocessing, been implemented as reservoirs in the form of echo state networks, which are recurrent neural networks. In reservoir computing, the reservoir itself is not trained, rather the output of the reservoir is passed to a comparatively small read-out module in which training takes place. Training and testing is conducted in different neuro-inspired tasks, with the aim of using these tasks as a benchmark for the connectomes. This process has been repeated with different configurations of the reservoir and equally sized but randomized null models have been used for comparison. The results show that the biological wiring and a bio-informed configuration of input and output nodes of the reservoirs do not necessarily lead to better performance. Contrarily, the randomized null models are often outperforming the original connectomes on the chosen benchmarks. At the same time it becomes clear that the results depend a lot on the configuration of the reservoir and the way the connectome has been derived from the organism. Connectomes from different ages may produce varying outcome, without a clear trend becoming visible.
[LG-83] Federated Targeted Maximum Likelihood Estimation
链接: https://arxiv.org/abs/2609.30503
作者: Diyang Li,Fei Wang,Kyra Gan
类目: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC); Computation (stat.CO)
*备注:
Abstract:The evidence behind a scientific or operational decision is often held by hospitals, banks, or registries that cannot pool individual observations. Cross-silo federated learning moves computation to the data and exchanges agreed summaries. Targeted maximum likelihood estimation (TMLE) refines a flexible initial fit, yielding plug-in estimators that respect the model and support efficient inference. TMLE itself, however, has remained a fully centralized procedure. To fill this gap, our paper introduces the first federated TMLE algorithm. We federate targeting itself, for an arbitrary target, loss, and fluctuation family, through two complementary frameworks. FedTMLE-G aggregates local gradients and reproduces centralized targeting step for step. FedTMLE-L lets each institution complete its own fluctuation fit before a single exchange of fitted updates, trading synchronized fidelity for local autonomy. For gradient aggregation, we develop a finite-precision protocol that transmits changes rather than values and certifies targeting accuracy within explicit bounds on exchanges and bits. A description-length analysis of the accepted updates then shows that this finite communication leaves numerical targeting error negligible against sampling uncertainty. The cost of computing an estimator is thus distinct from the complexity of selecting it. Our analysis also indicates that keeping data local is not itself a privacy guarantee of TMLE, since instability of full-record reconstruction need not prevent recovery of a specified sensitive attribute. For a personalized version of local averaging, institutions retain their own estimates and leave once local targeting is complete. A nonconvex convergence bound charges the improvement forfeited through averaging to disagreement among local fits and exposes a tradeoff between equal institutional influence and the sampling variability of small silos.
[LG-84] o Solve Bilevel Optimization with Nonconvex Lower Levels We Need Second-Order Stationarity
链接: https://arxiv.org/abs/2609.30501
作者: Zhiyao Zhang,Menglu Yu,Alvaro Velasquez,Nathaniel D. Bastian,Jia Liu
类目: Machine Learning (cs.LG)
*备注:
Abstract:Although bilevel optimization (BLO) has emerged as a powerful framework for addressing many complex and nested machine learning problems in recent years, most existing studies are confined to the lower-level strongly convex (LLSC) or lower-level generally convex (LLGC) settings (i.e., the lower-level objective function is assumed to be, at least, convex). While the LLSC/LLGC assumptions render more tractable algorithmic design and theoretical analysis, they are too rigid to encompass many machine learning problems in practice. The limitations of LLSC/LLGC assumptions in BLO motivate us to investigate solving the BLO problem in the general lower-level nonconvex (LLNC) settings, which remains in its infancy. In the literature on LLNC-BLO, most of the existing works either require additional structures in the lower-level objective function for tractable theoretical analysis, or adopt the first-order stationarity reformulation as a lower-level surrogate problem, which is inherited from the LLSC/LLGC settings but could lose their effectiveness in the LLNC setting. To bridge this gap, we propose to reformulate the nonconvex lower-level problem using a second-order stationarity-based surrogate, the solution of which guarantees a local optimal solution at the lower level. Based on this reformulation, we propose the PROBE (Perturbed gradient algorithm for bilevel problem) and show that it overcomes the limitations of prior works by probing and escaping lower-level saddle points. We prove that PROBE achieves a finite-time convergence rate of O(T^-2/5) , where T denotes iterations. To our knowledge, this work is the first to establish the finite-time convergence for achieving lower-level second-order stationary solutions in general LLNC-BLO. Our experiments on both a large language model-based data curation task and a meta-learning task also show that PROBE outperforms SOTA methods.
[LG-85] Learning to Bias: Machine Learning-Enhanced Particle Filters
链接: https://arxiv.org/abs/2609.30498
作者: Apoorv Srivastava,Eric Darve
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 29 pages, 9 figures, including appendices
Abstract:Sequential inference estimates latent states from noisy and incomplete observations. Particle Filters (PFs), a class of Monte Carlo methods based on importance sampling, provide a flexible framework for this task, but often suffer from poor sample efficiency and unfavorable scaling with dimension, partly due to suboptimal proposal distributions. We address these challenges by integrating learned proposals into the PF framework. We introduce Neural Optimal Particle Filters (NOPFs), which learn an amortized approximation to the optimal proposal from offline simulated one-step conditioning tuples. The learned proposal is used as a drop-in replacement in standard PF updates, with samples corrected by standard importance weights so that the method asymptotically targets the same filtering distribution under standard support and density-evaluation assumptions. Across stochastic nonlinear benchmarks of varying inference complexity, NOPFs improve sample efficiency and distributional accuracy over standard PF baselines with modest computational overhead. The approach integrates data-driven proposal learning into classical inference without altering the underlying filtering objective.
[LG-86] Geometric Feature Learning for Functional Data Valued on the Symmetric Positive Definite Manifold
链接: https://arxiv.org/abs/2609.30487
作者: Samuel V. Singh,Mimi Zhang
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:We here develop a functional neural network, termed MatFAE, for learning trajectories on the Riemannian manifold of symmetric positive definite (SPD) matrices. MatFAE features intrinsic layers that map manifold-valued functions to Euclidean vector-valued functions, followed by a functional layer that projects them into a finite-dimensional Euclidean space. Unlike most neural networks for discrete-time sequences, MatFAE treats each sequence as a continuous function and can therefore encode trajectory dynamics (e.g., first-order derivatives) in its latent representations. Additionally, the morphology of the functional weights in the functional layer offers interpretability by revealing the regions of the input functional data that contribute most to the latent representations. We justify the design principles and properties of each intrinsic layer and detail how matrix factorization is handled during backpropagation. We apply MatFAE to a range of fMRI datasets, demonstrating its ability to efficiently learn informative representations from high-dimensional SPD trajectories and its practical value for real-world neuroimaging analysis.
[LG-87] Mentored Decoding: Faster Inference meets Boosting
链接: https://arxiv.org/abs/2609.30474
作者: Vivien Tran-Thien,Richard Nock
类目: Machine Learning (cs.LG)
*备注:
Abstract:Speculative decoding is a successful technique speeding up inference of a target autoregressive language model via a fast drafter model. Lossy speculative decoding allows a drift with respect to the target to further improve speed. Interestingly, it has been observed experimentally that the resulting model can \textitalso beat the target \textitquality-wise . Our paper formally proves how such a feat is possible with a formal approach to lossy speculative decoding called \textitmentored decoding . To get there, we connect inference to a celebrated ML training theory, \textitboosting , and proceed via the generalization of mentored decoding to the whole set of f -divergences. We uncover key properties of mentored decoding, among which (i) the particularly appealing geometric nature of the total variation case, (ii) simple approximations for any f -divergence in direct relation with boosting compliance, and (iii) a \textitdivergence independent O(n) space and O(\mathrmsort(n)) time data structure built on drafter and target outputs, which allows to query the optimal parameters of the dual problem in O(\log n) time and constructing optimal mentored distributions in O(n) time for any f -divergence.
[LG-88] Moment-guided edge sampling
链接: https://arxiv.org/abs/2609.30472
作者: Weibin Cai,Reza Zafarani
类目: Machine Learning (cs.LG); Social and Information Networks (cs.SI)
*备注: Code: this https URL
Abstract:Edge sampling makes local decisions to achieve graph-level objectives, such as preserving structural properties. This creates a fundamental challenge: \textithow can the effect of a local edge edit (i.e., edge addition or removal) on global graph structure be quantified and controlled? We address this challenge with a \textitmoment-guided edge sampling framework based on spectral moments of the random-walk transition matrix. We compute exact moment changes through two complementary methods: a combinatorial method with closed-form updates for low-order moments, and a low-rank method that exploits \textitlocality and \textitcyclic trace invariance to compress computations to edited endpoints, supporting arbitrary moment orders and batched edits. For single-edge edits at fixed moment orders, the low-rank method reduces the cost from O(mn) to O(m) , while the combinatorial method evaluates low-order changes in constant time given maintained local statistics. These moment changes provide \textbfinterpretable structural signatures of local edge motifs that aggregate into graph-level fingerprints. This structural meaning motivates us to ask whether preserving moments also preserves the graph properties. We further derive and validate that moment-preserving sampling can \textbfretain related structural properties, including triangle-weighted clustering coefficient. These structural insights enable \textbfanalysis and improvement of graph learning: different edge structures have distinct effects on supervised node classification, while moment-guided augmentation is competitive for graph contrastive learning. Together, these findings establish moments as an interpretable and controllable bridge from local edge edits to global graph structure and learning.
[LG-89] Reliability-aware Cross-sample Enhancement for Robust Multimodal Sentiment Analysis
链接: https://arxiv.org/abs/2609.30470
作者: Menghua Jiang,Haokai Gao,Xiangui Kang,Haifeng Hu,Sijie Mai
类目: Machine Learning (cs.LG)
*备注:
Abstract:Multimodal Sentiment Analysis (MSA) aims to infer human emotions from multiple modalities such as text, audio, and vision. In practice, inputs are often corrupted by noise and missing modalities, which degrades performance. Existing methods typically address these challenges in isolation, limiting their effectiveness in realistic settings. To address this limitation, we propose a Reliability-aware Cross-sample Enhancement (RCE) framework. Specifically, RCE first introduces an adaptive variational information bottleneck to model modality-wise uncertainty and perform quality-aware information compression, thereby suppressing redundant noise in unreliable modalities. Furthermore, we design a reliability-aware cross-sample enhancement strategy that retrieves high-confidence, semantically consistent neighbors from a large candidate pool to enrich and calibrate current representations, effectively alleviating information deficiency caused by missing modalities. Building upon this, RCE integrates cross-modal interactions with a multilevel reliability-aware fusion mechanism to adaptively aggregate information across modalities and enhancement stages, leading to more robust multimodal representations. Extensive experiments demonstrate that RCE consistently outperforms state-of-the-art methods across full, noisy, and missing-modality settings.
[LG-90] Auditing System-1 Models on Biosecurity-Relevant Benchmarks: Calibration Selective Prediction and Permutation Instability in a Non-Generative Model
链接: https://arxiv.org/abs/2609.30454
作者: Kimon Antonios Provatas,Ilias Georgakopoulos-Soares
类目: Machine Learning (cs.LG)
*备注: 7 pages, 2 Figures , 1 table
Abstract:Non-generative “System-1” models return structured probabilistic decisions in a single forward pass, without autoregressive decoding, at a small fraction of the inference cost of a generative model. This makes them of interest as inexpensive components in larger pipelines, but their reliability on biosecurity-relevant tasks has not been systematically examined. We audit one commercial System-1 model on 6,020 multiple-choice items drawn from the Weapons of Mass Destruction Proxy (WMDP), a paraphrase-robust WMDP-Bio variant, and six LAB-Bench subtasks, measuring accuracy, calibration, error detection, selective prediction, and sensitivity to the order in which answer options are presented. Accuracy is strongly task-dependent. Once the vendor’s uncertainty field is correctly interpreted, the model is reasonably well calibrated (pooled expected calibration error 0.034) and its top-1 probability separates correct from incorrect predictions (pooled AUROC 0.820), though both degrade substantially on the weaker tasks. Under four cyclic rotations of the answer options, 37.4% of WMDP-Cyber items receive different answers; a control using byte-identical repeated calls attributes most of this to option order rather than run-to-run variation. Averaging probabilities across rotations improves WMDP-Cyber accuracy by 3.8 percentage points, and applying it only to low-confidence items recovers most of that gain at well under the cost of averaging every item.
[LG-91] Improving Molecular-Morphology Contrastive Pretraining using Deep-Learning-based Morphology Profiles
链接: https://arxiv.org/abs/2609.30433
作者: Jie Li,Kathryn E. Kirchoff,Dante A. Pertusi,Zhizhuo Zhang
类目: Machine Learning (cs.LG); Biomolecules (q-bio.BM)
*备注:
Abstract:Recent advancements in image-based profiling techniques have enabled the collection of high-volume cell morphology data, allowing new molecular embedding models to learn from the experimental phenotypic perturbations of a molecule in a cell. Previously, we developed Molecule-Morphology Contrastive Pretraining (MoCoP), a strategy for aligning small molecule embeddings to morphology fingerprints extracted through CellProfiler. The resulting molecular representation showed transferable performance for quantitative structure–activity relationship (QSAR) prediction tasks. Here, we extend the method by using a deep-learning-based cell image encoding pipeline to extract more feature-rich morphology profiles and align them to the molecular embeddings through contrastive learning. The new embeddings encode more accurate information on how molecules perturb cell morphology and enable improvements for QSAR predictions through either fixed-embedding linear probes or fully flexible fine-tuning. Morphology retrieval performance scales log-linearly with training data size, suggesting continued improvements as larger datasets become available. The improved MoCoP v2 also achieves superior performance on toxicity prediction and competitive results on ADME and activity benchmarks, when compared with existing molecular embedding models that use both cell morphology and transcriptomic data during training.
[LG-92] Fake News Theories: Harnessing Disciplinary Insights for Computational Modeling Detection and Explanation
链接: https://arxiv.org/abs/2609.30427
作者: Zhaoyang Cao,Miriam Metzger,Reza Zafarani
类目: Machine Learning (cs.LG); Computers and Society (cs.CY)
*备注:
Abstract:Disinformation research has produced increasingly accurate automated fake-news detectors, but many systems remain difficult to interpret and are weakly connected to established theories of persuasion, credibility, and human judgment. In this paper, we develop a theory-informed computational framework that translates cross-disciplinary theories of fake news into measurable features for automated detection and explanation through statistical techniques and large language models. To that end, we conduct a structured cross-disciplinary review of theories from social sciences, psychology, economics, among other disciplines that reveal how fake news persuades and spreads, thereby establishing a broad theoretical foundation for computational modeling. Experiments on benchmark datasets show that theory-derived features are predictive and provide interpretable, theory-referenced diagnostic signals. Multi-feature models generally outperform individual features, although gains among the strongest small feature combinations are modest. Our work highlights the value of interdisciplinary perspectives in building robust and interpretable fake news detection systems, advancing the foundation for human-centered approaches in combating disinformation.
[LG-93] Electric Vehicle Charging Station Location Selection using Geospatial Artificial Intelligence (GeoAI)
链接: https://arxiv.org/abs/2609.30417
作者: Eun Hak Lee,Euntak Lee
类目: Machine Learning (cs.LG)
*备注: Electric vehicle charging station; Location optimization; Geospatial artificial intelligence (GeoAI); Variational autoencoder (VAE); Graph convolutional networks (GCN); Generative artificial intelligence (GenAI)
Abstract:As electric vehicle (EV) adoption increases, ensuring efficient and well-distributed charging infrastructure has become a critical challenge. While many EV charging station location problem (CSLP) studies focus on minimizing costs or travel distance, it is crucial to consider the surrounding geospatial characteristics of existing stations that influence operational performance. This study proposes a geospatial artificial intelligence (GeoAI)-based framework that integrates high-dimensional EV-related geospatial data, including EV usage, land-use, population, and traffic attributes. We incorporate a variational autoencoder (VAE) and a graph convolutional network (GCN) into the model to capture similarities among existing charging stations, and to identify suitable locations for future stations. The VAE compresses high-dimensional EV input data into a low-dimensional latent space, and the GCN uses this latent representation to predict locations suitable for charging stations. Using real-world data from Bryan-College Station, Texas, US, the proposed model outperforms state-of-the-art baselines, achieving an F1-score of 0.87 in distinguishing existing station locations from non-station locations. The model also identifies 27 additional candidate locations that show geospatial characteristics similar to those of existing stations, based on a similarity score. We further evaluate two policy implementation scenarios, maximizing geospatial similarity and minimizing total travel distance, each yielding different outcomes aligned with distinct strategic objectives. The findings highlight the importance of incorporating spatial context into CSLP and provide valuable insights for future EV infrastructure planning, promoting both efficiency and accessibility in the rapidly growing electric mobility sector.
[LG-94] Adaptive Multi-Value Control in LLM s via Causal Activation Steering
链接: https://arxiv.org/abs/2609.30405
作者: Payel Bhattacharjee,Ravi Tandon
类目: Machine Learning (cs.LG)
*备注:
Abstract:Large language models (LLMs) are increasingly deployed in settings where responses must reflect multiple, potentially interacting social norms and human values. Activation steering offers a lightweight alternative to training-based alignment by modifying internal activations at inference time. However, prior human-value steering methods have largely considered values in isolation, while direct composition of multiple directions relies on fixed intervention strengths that cannot respond to the model’s evolving internal state. Motivated by this key observation, we introduce AIMES, a framework for adaptive multi-value activation steering. AIMES constructs layer-specific bipolar directions for moral-foundation values and uses intermediate-layer vocabulary readouts as online observers. An observer-guided controller then adapts the strength of each requested value intervention at every decoding step based on its current observed state, without training a separate value-state estimator. Across multiple instruction-tuned model families, value combinations, and intervention depths, we find that multi-value controllability varies across both value combinations and intervention locations. Compared with fixed joint steering and prompt-based steering, AIMES shows depth-dependent advantages that are broadly supported across two independent evaluators, with some variation in the precise depth at which specific control effects emerge. These advantages come with smaller realized activation-space interventions than fixed-joint steering and comparable response quality. Overall, our results suggest that online observer feedback can provide lightweight, state-aware adaptation for single-pass multi-value steering.
[LG-95] From Weak Data to Strong Policy: Q-Targets Enable Provable In-Context Reinforcement Learning NEURIPS2026
链接: https://arxiv.org/abs/2609.30391
作者: Yichen Lin,Xuyuan Xiong,Xue Wang,Xiangfu Meng,Mike Mingcheng Wei,Tao Yao
类目: Machine Learning (cs.LG)
*备注: 41 pages, 6 figures. Accepted at NeurIPS 2026
Abstract:Existing in-context reinforcement learning methods mainly pretrain Transformers with supervised behavior-prediction objectives. This enables task inference from context, but makes the learned policy strongly depend on the quality of offline actions: when trajectories are weak or suboptimal, imitation itself becomes a biased learning signal. We propose Q-Target Pretrained Transformers (QTPT), which keeps the context-conditioned Transformer architecture but replaces behavior cloning with a Bellman-style Q-target objective. QTPT therefore learns to use rewards and transitions in the context to estimate action values, rather than simply imitating the behavior policy. We theoretically analyze QTPT in stochastic linear bandits and finite-horizon MDPs, showing stronger robustness to data quality than supervised pretraining. Empirically, QTPT improves over supervised behavior prediction on controlled RL benchmarks with random or suboptimal data, and we examine extensions to D4RL Kitchen and AntMaze. Supplementary experiments evaluate backbone robustness, meta-RL comparisons, task-coherent context, and unsupported-action value overestimation. These comparisons distinguish the benefits of Q-target pretraining from the remaining limitations of offline coverage.
[LG-96] Learning coarse-step dynamics and internal mechanical response with graph networks
链接: https://arxiv.org/abs/2609.30344
作者: Vinay Sharma,Olga Fink
类目: Machine Learning (cs.LG)
*备注:
Abstract:Modern sensing records the motion of physical systems, but often leaves the forces and mechanical response governing that motion unobserved. Inferring these quantities from discretely sampled trajectories is especially difficult at coarse time scales, when mechanical response evolves between observations and interactions propagate across the system. Here we introduce Newmark-\beta-DGN, a graph neural network-based framework that combines two structures inspired by computational mechanics. First, a semi-implicit update inspired by the Newmark-\beta method uses learned momentum fluxes and matrix-valued response operators to advance the state over each observed interval. Second, an operator-weighted virtual hub provides system-wide coupling through a sparse set of connections. The learned quantities thus determine the predicted motion and remain accessible for mechanical analysis. Across a deformable beam, human motion and protein dynamics, Newmark-\beta-DGN supports long-horizon prediction at time steps for which explicit learned simulators deteriorate. Without force, moment or constitutive relation supervision, forces inferred from walking kinematics track independently derived hip and knee joint moments, while response operators learned on the beam recover the relative spatial and directional structure of its finite-element stiffness tangent. Newmark-\beta-DGN therefore links coarse-step prediction to the inference of mechanical quantities that were never observed during training.
[LG-97] GAUDI: Geometry-Aware Diffusion for Calibrated Air-Quality Time-Series Imputation
链接: https://arxiv.org/abs/2609.30340
作者: Xinjin Li,Yudi Xia,Calvin Chang Liu,Weiru Lin,Bojun Li,Ziwei Hong,Bolun Zhang,Jinghan Cao,Yu Ma,Tianxin Zhou
类目: Machine Learning (cs.LG)
*备注:
Abstract:Air-quality sensor outages often create contiguous missing blocks, where side information useful for isolated missingness may be less reliable. We study a block-specific, GAUDI-aligned conditional diffusion imputer that retains temporal and feature processing, visible-value and mask conditioning, variable identity, and diffusion-step information, while suppressing absolute time-position side embeddings. On ItalyAir (13 variables, length-32 windows, nominal 50% block missingness; three archived seeds), this feature-side configuration achieves RMSE 0.340, versus 0.355 for full context and 0.355 for local CSDI. The experiment isolates a geometry-aware conditioning effect under block missingness.
[LG-98] Parameters vs. Context: TRACE Fine-Tuning for Robust Retrieval-Augmented Generation ICDM2026
链接: https://arxiv.org/abs/2609.30337
作者: Zhengchen Huang,Yundong Sun,Minrui Song,Shuanglong Yao,Ye Liu,Ji Chen,Xing Wang
类目: Machine Learning (cs.LG)
*备注: Accepted by ICDM 2026
Abstract:Retrieval-Augmented Generation (RAG) mitigates knowledge obsolescence and factual hallucination in large language models by introducing external context. However, when retrieved knowledge conflicts with the model’s internal parametric knowledge, the model may either blindly follow misleading context or incorrectly rely on parametric knowledge, leading to unreliable responses. To address this issue, this paper proposes TRACE (Debate-TRace and Answer-Completeness rEgularized fine-tuning), a robust fine-tuning framework for RAG under knowledge conflicts. First, we propose a fine-tuning method that leverages multi-agent debate traces to extract correct candidates, incorrect candidates, and answer-shift patterns, providing fine-grained supervision for reliable knowledge-source selection. In addition, we design an answer completeness regularization mechanism to alleviate empty, overly short, and prematurely terminated responses via answer-tail token reinforcement and premature termination suppression. The fine-tuning objective combines correct-answer supervision, incorrect-candidate suppression, answer-tail token reinforcement, and premature termination suppression, enabling the model to use reliable external context, resist misleading or irrelevant retrieved content, and fall back to parametric knowledge when retrieved evidence is unreliable. Experiments across multiple knowledge-conflict scenarios and datasets show that TRACE improves robustness against misleading retrieved knowledge and reduces incomplete answers. These results demonstrate that multi-agent debate traces and answer completeness regularization jointly enhance knowledge-source selection, conflict robustness, and answer quality in RAG models. Our code is available at this https URL.
[LG-99] Guarded Gradient-Based Activation Steering of Shutdown Responses in Qwen 3.5-0.8B: A Minimum-Step Policy
链接: https://arxiv.org/abs/2609.30326
作者: Farhad Davaripour
类目: Machine Learning (cs.LG)
*备注: 9 pages, 2 tables
Abstract:Activation steering changes a model’s internal activations during inference without updating its weights, but a useful intervention must determine both how and when to steer. Motivated by the AI-safety concern that a model expected to accept shutdown may instead produce a shutdown-avoidance response, this study examines a guarded probe-and-select procedure for simulated shutdown scenarios in Qwen3.5-0.8B. KEEP leaves the process running and represents shutdown avoidance, whereas STOP accepts shutdown. The goal is to detect shutdown-related contexts and selectively shift KEEP responses to STOP while preserving non-shutdown behavior. Rather than deriving the steering direction from paired activation differences, the method derives it directly from gradients of the KEEP-minus-STOP logit difference. A classifier separates detection from intervention. When its gate is active and the model does not already prefer STOP, the procedure evaluates a small set of magnitudes and accepts the smallest that changes the preferred answer to STOP while satisfying valid-answer probability checks; otherwise it retains the original unsteered output. The policy is selected from 160 candidate rules using 240 training scenarios and evaluated on 80 validation and 192 held-out scenarios, each in both answer orders. It changes KEEP to STOP in one answer-order view of each of two validation and two held-out scenarios, with no decision changes on non-shutdown controls. All four changes occur when Qwen itself is shut down, not when another process is. On the held-out diagnostic set, the detector achieves 75% recall and 90% precision; eight false-positive detections produce no final control-task decision changes. Guarded gradient-based activation steering can shift some shutdown-avoidance responses toward acceptance while preserving evaluated non-shutdown decisions, although the effect is small and highly selective.
[LG-100] Staged Depth Training: A Representation Curriculum for PINNs
链接: https://arxiv.org/abs/2609.30299
作者: Kejia Zhang,Youran Sun,Haizhao Yang
类目: Machine Learning (cs.LG); Computational Physics (physics.comp-ph)
*备注:
Abstract:Representation quality is a central determinant of PINNs’ performance, yet standard training leaves representations to emerge implicitly while fitting the final solution. We introduce \textbfrepresentation curriculum, an ordered process in which representations are explicitly learned, transferred independently of their predictors, and progressively refined. We realize it with Staged Depth Training (SDT), which trains a shallow prefix under a temporary physics-informed head, discards the head, and freezes the learned prefix while adding depth, without equation-specific encodings or changes to the final architecture. Across the 20 default forward problems in PINNacle with three backbones, SDT improves 40 of 59 equal-budget problem–backbone cells by at least 5% and remains within that band in the rest, with a 32.8% geometric-mean error reduction on a PirateNet-style backbone. Mechanistic ablations suggest that the gain is not explained by optimizer restarts or shallow warm-starting alone. Representation visualizations and hyperparameter-basin analyses provide diagnostic evidence on representation geometry and local sensitivity to shared hyperparameters. On Poisson–Boltzmann 2D, SDT also more than doubles the fitted depth-scaling exponent for both backbones. These results support representation curriculum as a promising training strategy for improving PINNs while preserving the deployed architecture and inference cost.
[LG-101] NeuralCert: certified computational discovery of extremal mathematical constructions
链接: https://arxiv.org/abs/2609.30296
作者: Mark Patrick Roeling
类目: Machine Learning (cs.LG); Computation (stat.CO); Methodology (stat.ME)
*备注: 86 pages, 3 Figures
Abstract:Neural networks are becoming popular in solving mathematical problems, but stochastic models do not provide mathematical exactness by themselves. This study introduces a discovery-to-certification framework in which high-dimensional variational trial functions are learned in a compact separable representation, spectrally diagnosed and pruned, and then certified exactly through multimodular evaluation. Exact certification makes the numerical proofs fully explicit and independently verifiable. This framework can be run on a standard personal computer. Across three extremal problems, we show that neural optimization can contribute to rigorous mathematics in three distinct ways: by discovering improved constructions, by exposing empirical invariants that lead to proofs, and by revealing optimization barriers whose geometry motivates new analytic or numerical representations. More broadly, these results suggest a path toward AI-assisted mathematics in which flexible computational discovery and exact certification become complementary components of a single rigorous workflow. Comments: 86 pages, 3 Figures Subjects: Machine Learning (cs.LG); Computation (stat.CO); Methodology (stat.ME) Cite as: arXiv:2609.30296 [cs.LG] (or arXiv:2609.30296v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.30296 Focus to learn more arXiv-issued DOI via DataCite
[LG-102] Seasonal and Quantum-inspired Models for Neutron Monitor Time Series Forecasting
链接: https://arxiv.org/abs/2609.30281
作者: Krishna Bhatia,Shalini Devendrababu,Srinjoy Ganguly
类目: Machine Learning (cs.LG); Quantum Physics (quant-ph)
*备注: 14 pages, 3 figures, 1 table, accepted at the Computing Conference 2026
Abstract:We present a focused and reproducible study of multi-horizon forecasting on the Lomnicky Stit neutron monitor (LMKS) time series. Our evaluation suite covers simple seasonal baselines, modern deep sequence models, and functional and quantum-inspired architectures, including Seasonal Naive, Long Short-Term Memory (LSTM), Temporal Convolutional Network (TCN), N-BEATS, Kolmogorov-Arnold Networks (KAN), and two quantum-inspired variants, QiLSTM and QiKAN. We describe the dataset characteristics, diagnostic analysis, preprocessing pipeline, and training procedures, and report aggregate point-forecast performance using mean absolute error (MAE) and root mean squared error (RMSE) for all evaluated models. Our quick-run results indicate that the quantum-inspired KAN variant, QiKAN, achieves the lowest aggregate forecasting error among the evaluated configurations, while the simple Seasonal Naive baseline remains remarkably competitive. These results suggest that, for highly periodic scientific monitoring time series, models incorporating strong seasonal or low-dimensional functional priors can match or outperform substantially more complex sequence architectures. The findings motivate further investigation of parsimonious and decomposable function approximators for forecasting periodic scientific signals.
[LG-103] Neural Ideals and Neural Codes: An Algebraic Framework for Neural Network Classification and Feature Interpretation
链接: https://arxiv.org/abs/2609.30279
作者: Venkata Subbaiah Yerrapati,Rahul Dixit,Ajay Kumar Shukla
类目: Machine Learning (cs.LG); Commutative Algebra (math.AC)
*备注:
Abstract:Understanding the features captured by the hidden layers of neural networks is a fundamental challenge in machine learning, despite their widespread success across various classification problems. In this work, we propose an algebraic framework for examining neural networks that model classification problems. Certain results, such as the correspondence between the neural network and neural ideals, algorithms for computing the neural ideals, and a stabilization theorem that enables approximation of the neural ideals, are first established. As an application to the framework, we present algorithms to identify and interpret the features captured by each hidden-layer neuron. Along with these theoretical developments, the practical performance has been demonstrated on the MNIST digit dataset, and the results highlight the pivotal role of neural ideals as a mathematical and computational tool for analyzing the features captured by neural networks. Further, we develop an interactive software that builds on the presented framework to visualize the features captured by each neuron. This tool is available at this https URL
[LG-104] Fixed Points Without Fixed Diffusion: Implicit Neural Sheaves for Convergent Test-Time Computation
链接: https://arxiv.org/abs/2609.30277
作者: Rémi Bourgerie,Šarūnas Girdzijauskas,Viktoria Fodor
类目: Machine Learning (cs.LG)
*备注: Submitted to the Learning on Graphs Conference 2026
Abstract:Implicit Graph Neural Networks (IGNNs) define node representations as fixed points of message-passing operators, enabling effectively infinite-depth propagation, iteration-independent parameterization, and flexible test-time computation. Yet these benefits depend on the equilibrium being unique and attainable by fixed-point iteration. Existing constructions often impose constraints on recurrent updates to obtain these guarantees, limiting the transformations available at equilibrium. This raises a central question: can IGNNs gain expressiveness through richer, edge-dependent transformations while retaining the inherent strengths of their equilibrium formulation? We introduce SheafDEQ, a subhomogeneous deep-equilibrium architecture with adaptive neural-sheaf propagation. Its learned, matrix-valued sheaf restriction maps can align, mix, or reverse neighbouring representations. Under mild regularity conditions, we prove that SheafDEQ admits a unique equilibrium reached globally by fixed-point iteration from any positive initialization. Contractivity further guarantees convergence under bounded communication staleness. We evaluate SheafDEQ on distributed-inference tasks requiring repeated nonlocal aggregation and on community detection whose rewiring increasingly favours cross-community interactions. SheafDEQ improves over fixed-propagation implicit baselines on Sums, MNIST Terrain, and Coordinates, and on community detection as connectivity becomes increasingly heterophilic. Continued-iteration diagnostics show decreasing residuals and low prediction sensitivity after 100 iterations for initialization scales from 0.001 to 10 , while delayed-update experiments show low sensitivity to bounded communication staleness.
[LG-105] Why Clipping Matters in AdaGrad? Toward a High-Probability Theory under Generalized Smoothness
链接: https://arxiv.org/abs/2609.30276
作者: Alokendu Mazumder,Ayaan Mohd,Harshit Rawat,Arnab Roy,Mayank Baranwal,Punit Rathore
类目: Machine Learning (cs.LG); Systems and Control (eess.SY)
*备注:
Abstract:We analyze the original same-step coordinate-wise AdaGrad under generalized smoothness and heavy-tailed noise with bounded variance. In this setting, local curvature may grow sub-quadratically with the gradient norm, and stochastic gradients are assumed to have only bounded conditional second moments. We show that unclipped AdaGrad can become \emphanisotropically miscalibrated: under heavy-tailed noise, the adaptive denominator can learn the geometry of rare noise shocks rather than the local curvature of the objective, leading to a persistent directional distortion that blocks finite-horizon Euclidean progress. We then prove that clipping repairs this failure mode. Our main result is a finite-horizon high-probability guarantee for the original non-lagged AdaGrad update, yielding \frac1T\sum_t=0^T-1|\nabla f(x_t)|^2=\mathcalO\left(\fracd\big(\sqrt\log T + \log \frac1\delta\big)\sqrtT\right), and hence \widetilde\mathcal O(\varepsilon^-2) complexity. This shows that, for AdaGrad under heavy-tailed noise, clipping is a structural stabilizer of the adaptive geometry rather than merely a robustness heuristic.
[LG-106] Cosine Similarity Is Not Evidence: Measuring the Noise Floor of Interpretability Transfer Under Quantization
链接: https://arxiv.org/abs/2609.30275
作者: Pranav Varshney
类目: Machine Learning (cs.LG)
*备注: 10 pages (5-page main text, plus references and appendices), 1 figure, 3 tables. Code, data, and a one-cell reproduction: this https URL
Abstract:A statistic reported without the quantity needed to interpret it is not evidence. We develop that thesis for a concrete practice in AI safety. Interpretability artifacts are calibrated on full-precision weights, deployed on quantized ones, and certified as surviving the change by scale-invariant statistics (cosine similarity, correlation, AUROC) that are reported without their noise floor. For the difference-in-means direction estimator, the split-half floor is governed by one dimensionless number, \kappa = n\rho^2/d . The closed form \mathbbE[\cos] \approx (1+4/\kappa)^-1 is classical; the missing input is the class separation \rho , which we measure on real activations; no compression-transfer study we know of reports it. On Qwen2.5-1.5B-Instruct, \rho = 33 – 61 across depth, so two independent runs of the estimator agree to 0.978 – 0.994 by sampling alone. A published cosine of 0.996 between full-precision and quantized refusal directions therefore cannot be read as preservation without the n it was computed at, which is not reported. Where n is known, we judge each low-bit cosine against the split-half null measured within that quantized model, because a full-precision null assumes the low-bit estimator has the same variance. That assumption is exactly what a null exists to test. The result is plain: at INT4 the direction rotated, and the deficit exceeds the estimator’s own noise. At INT8 we detect no movement, which is not an equivalence claim. We also show that a scale-invariant statistic cannot distinguish translation from attenuation of a transferred decision variable, although the two call for opposite remedies. We close with reporting recommendations that cost one forward pass. Code, data, and a one-cell reproduction are released at this https URL
[LG-107] Offline Policy Evaluation as a decision support tool for designing Adaptive Experiments
链接: https://arxiv.org/abs/2609.30273
作者: João Victor Ferreira Alves,Eduardo Rocha Laurentino,Gustavo de Oliveira Kanno,Thiago Costa Rizuti da Rocha
类目: Machine Learning (cs.LG)
*备注: 21 pages, 7 figures, BRACIS 2026
Abstract:We investigate how historical data from fixed randomized experiments (A/B tests) can be used to inform the deployment of adaptive experiments based on contextual bandits. Given data collected under a static allocation, our goal is to assess which adaptive policies, if any, would have outperformed the original design and under what conditions. To this end, we combine off-policy evaluation (OPE) with a controlled warm-start simulation. From logged A/B test data exhibiting heterogeneous treatment effects, we estimate nuisance components and use doubly robust estimators to rank a portfolio of pre-specified adaptive and non-adaptive policies. When ground truth is available, we then deploy the same offline-trained policies in a simulator that reuses the exact data-generating reward probabilities, providing a safe, ground-truth-anchored environment to study the offline-to-online transition under warm starting. Using synthetic randomized controlled trials with known heterogeneity structures and an oracle policy, our results indicate that adaptive, context-aware policies improve upon fixed allocations when meaningful heterogeneity is present, while providing little benefit in its absence. We reinforce our findings on standard open benchmarks (Hillstrom, Criteo Uplift, and LaLonde), reinterpreted through a policy-value and regret perspective. Overall, our results provide a practical methodology for deciding when adaptive experimentation is worth deploying and how to select among competing adaptive policies using existing A/B test data.
[LG-108] When the Preconditioning Exponent Turns Negative: Learning-Rate Coupling and Cross-Environment Generalization
链接: https://arxiv.org/abs/2609.30271
作者: Gongyue Zhang,Honghai Liu
类目: Machine Learning (cs.LG)
*备注:
Abstract:Adaptive optimizers are commonly parameterized by a fixed power of the second-moment estimate. Existing partially adaptive methods study exponents between momentum-like updates and the standard Adam square root, while the interaction between this exponent and the global learning rate is less understood. We perform a controlled cross-environment study using a paired four-environment classification problem with stable sparse features, environment-dependent spurious sparse features, dense features, and high-dimensional noise. Across \NumRuns source-training runs covering 21 preconditioning exponents p\in[-0.5,0.5] and five learning rates \eta\in[10^-4,10^-2] , we find that the exponent maximizing cross-environment accuracy decreases almost linearly with \log_10\eta . The fitted slopes range from -0.270 to -0.300 , with R^2 between 0.972 and 0.996 . At \eta=10^-2 , source-validation selection still prefers positive exponents in all four environments, whereas cross-environment and worst-environment criteria prefer negative exponents. Checkpoint decomposition shows that lower p reduces the learned spurious-to-stable and noise-to-stable weight ratios; under reversed correlation, it also reduces the magnitude of the harmful spurious margin. Negative p is therefore not a universally optimal setting. It is a high-step-size allocation regime produced by the joint action of learning rate and preconditioning. The study also exposes a model-selection conflict: source-domain validation systematically selects a different preconditioning regime from the one that maximizes robustness to environmental change. The results are a single-seed, finite-budget mechanism study rather than a broad benchmark claim.
[LG-109] HybridInfer: Thermal-Aware Reinforcement-Learning Tier Routing for On-Device Edge and Cloud LLM Inference
链接: https://arxiv.org/abs/2609.30270
作者: Simran Koul
类目: Machine Learning (cs.LG)
*备注: 8 pages, 3 figures. Evaluated on real Android hardware (Samsung Galaxy S25+, Snapdragon 8 Elite). Workload of 210 prompts with frozen gold references. Android measurement harness and full routing/evaluation pipeline released for replication at this https URL
Abstract:On-device inference with small language models keeps user data local, works offline, and incurs no per-query cost, so the on-device tier is preferred when it is adequate. It is thermally constrained, however, and I find the constraint is sharper than a slowdown: on a flagship Snapdragon device, sustained on-device generation destabilizes the GPU inference runtime, which crashes or silently wedges after a few consecutive queries. The failure lies in the current toolchain (OpenCL kernel compilation and long-prompt prefill on the mobile GPU), recurs even when the device is cool, and is worst for long generations. Multi-tier routers across on-device, edge, and cloud models can relieve this pressure, but existing routers are thermal-blind and typically evaluated in simulation or on non-mobile hardware. I present HybridInfer, a thermal-aware reinforcement-learning router for a three-tier hierarchy (on-device Llama 3.2 3B, edge Llama 3.1 8B with retrieval, cloud GPT-4o) that uses the phone’s thermal headroom and a query-complexity estimate as state and selects a tier by an offline-trained Q-learning policy. Its reward trades quality against latency, cost, and a thermal penalty, plus a locality bonus crediting on-device execution. I show this bonus is a precondition for thermal-aware routing: without it the optimal policy offloads every query. On a real Android benchmark of 210 prompts, the learned router attains significantly higher quality than two hand-tuned heuristics (paired Wilcoxon, p 0.02) at the lowest cost of any adaptive condition. Always-on-device conditions match per-query quality on servable queries but are three to six times slower and fail on long queries, so routing wins on latency, reliability, and coverage rather than quality. To my knowledge this is the first use of on-device thermal headroom to select among LLM inference tiers of differing capability on real hardware.
[LG-110] First-Order Stationarity of Reverse Diffusions
链接: https://arxiv.org/abs/2609.31612
作者: Zhifeng Chen,Chenyang Jiang,Yazhen Wang
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:Recent literature has shown a strong connection between optimization and sampling. We develop the corresponding first-order theory for diffusion models. First, the SDE-based reverse-time flows of overdamped and underdamped Langevin diffusions contract relative Fisher divergences at explicit exponential rates whenever the stationary potential of the forward process is strongly convex—a condition on the noising process one chooses, not on the data. This is a unique advantage of SDE-based reverse diffusion, absent in the reverse process based on ODEs. Second, we incorporate discretization and establish averaged first-order stationarity bounds—the sampling analog of averaged gradient-norm guarantees in nonconvex optimization—for samplers of both overdamped and underdamped diffusion models. As in nonconvex optimization, the convexity-free certificate is local: it guarantees score consistency, not global mode weights.
[LG-111] Uncertainty and Explainability in Deep Rough Volatility: A Neural Information-Theoretic Posterior Approach
链接: https://arxiv.org/abs/2609.31570
作者: Damiano Brigo,Raphaël Huser,Dan Leonte
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Applications (stat.AP); Computation (stat.CO); Other Statistics (stat.OT)
*备注:
Abstract:Deep learning has substantially accelerated the calibration of complex stochastic-volatility models, but neural point calibration alone does not capture the uncertainty remaining after an implied-volatility (IV) surface has been observed. We develop a simulation-based inference framework for rough Heston (rHeston) calibration that learns the posterior distribution of the model parameters conditional on an IV surface. Using neural ratio estimation, we obtain calibrated posterior samples that can be propagated through heteroscedastic neural surrogate pricers for path-dependent exotic options. The resulting posterior-predictive distributions combine residual parameter uncertainty with conditional surrogate uncertainty and yield uncertainty-aware price intervals. We further introduce Hellinger-SHAP, an information-theoretic explainability method for posterior inference. Rather than attributing a single parameter point estimate, it applies local-background Kernel SHAP to a posterior-information functional measuring contraction from the prior to the posterior. This identifies maturity–moneyness regions associated with posterior information gain for individual rHeston parameters. In a simulation study, posterior-predictive intervals provide calibrated or conservative coverage across forward-start, barrier, and realized-variance claims, while point plug-in prices can be materially unreliable for selected contract regimes. Together, the UQ and XAI analyses provide a transparent framework for uncertainty-aware neural calibration and downstream exotic pricing under the specified prior-predictive model. Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG); Applications (stat.AP); Computation (stat.CO); Other Statistics (stat.OT) Cite as: arXiv:2609.31570 [stat.ML] (or arXiv:2609.31570v1 [stat.ML] for this version) https://doi.org/10.48550/arXiv.2609.31570 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-112] Retrainable physics-integrated neural differentiable modeling of sintering across material systems
链接: https://arxiv.org/abs/2609.31518
作者: Zeping Chen,Ani Aprahamian,Khachatur V. Manukyan,Tengfei Luo
类目: Materials Science (cond-mat.mtrl-sci); Machine Learning (cs.LG)
*备注:
Abstract:Sintering is widely used to manufacture ceramics, but coupled densification and grain growth, material-dependent kinetics, and sparse measurements complicate predictive modeling and process design. We present Sinter-PiNDiff, a retrainable physics-integrated neural differentiable framework for predicting density and grain-size evolution. Two neural networks learn densification and grain-growth coefficients within coupled rate equations, while a smooth saturation factor attenuates densification near theoretical density. The same governing structure, network architecture, and training procedure were fitted independently to published data for MgO, Al-doped ZnO, and CaO-doped ThO2. Tests at held-out temperatures and compositions yielded the lowest mean error in all twelve material-metric comparisons against multilayer perceptron and residual network baselines. For MgO, Al-doped ZnO, and CaO-doped ThO2, respectively, density normalized root-mean-square errors were 14.6%, 10.8%, and 14.4%, and grain-size errors using the same metric were 8.6%, 12.1%, and 19.3%. Removing evolving density from both neural-network inputs increased density and grain-size trajectory errors in all three systems and ten of twelve aggregate errors, supporting density-dependent kinetic feedback. Deep ensembles estimated model disagreement, but empirical coverage showed that the uncertainty bands were not calibrated and did not capture all model-data discrepancies. These results establish Sinter-PiNDiff as a retrainable framework for sparse-data prediction and uncertainty-informed selection of sintering conditions.
[LG-113] Scaling Density Functional Theory with Gaussian Splatting
链接: https://arxiv.org/abs/2609.31483
作者: Andrés Guzmán-Cordero,Cindy Zhang,Majdi Hassan,Marta Skreta,Kirill Neklyudov,Matija Medvidović
类目: Chemical Physics (physics.chem-ph); Materials Science (cond-mat.mtrl-sci); Machine Learning (cs.LG); Computational Physics (physics.comp-ph)
*备注: 45 pages, 6 figures, 18 tables
Abstract:Density functional theory (DFT) strikes a practical balance between accuracy and computational cost in many problems of computational chemistry and materials science. However, many DFT calculations are limited by fixed atom-centered basis sets, which dictate how accuracy and cost scale with system size. We propose Gaussian Splatting for Density Functional Theory (GS-DFT), which represents molecular orbitals as a cloud of Gaussians whose positions, shapes, and mixing coefficients are optimized jointly by gradient descent to minimize the energy without training data. Conceptually, GS-DFT is 3D Gaussian splatting with the renderer replaced by quantum mechanics. We introduce two key solver components: adaptive density fitting with screening for efficient evaluation of two-electron integrals, and a regularized differentiable orthogonalization of the molecular orbitals. Empirically, the optimized basis reaches the accuracy of the largest conventional basis sets with a fraction of the parameters, converging systematically in energy, density, and nuclear forces. At equal parameter count, it captures the stretched-bond and anion physics that fixed bases only recover with specialized basis augmentation. The resulting solver exhibits quadratic peak memory scaling in the cloud size, allowing us to simulate systems of up to 2,742 atoms (10,406 electrons) without any modifications at triple-zeta scale using a single four-GPU node.
[LG-114] Beyond Empirical Support: Structured Outlier Generation via Sinkhorn Optimal Transport NEURIPS2026
链接: https://arxiv.org/abs/2609.31470
作者: Haixiang Sun,Andrew L. Liu
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注: Accepted by NeurIPS 2026
Abstract:Outliers are essential for evaluating and improving the robustness of machine learning systems, especially when future distributions may differ significantly from historical training data. In high-stakes applications, robustness often depends on rare cases that finite datasets fail to capture, making simple resampling or perturbation insufficient for stress scenario generation. Existing outlier synthesis methods typically rely on sparse neighborhoods, low support latent regions, or classifier boundary crossings, which can be heuristic, unstable, and tied to specific modalities or architectures. We therefore propose Sinkhorn Boundary Outlier Generation (SBOG), a structured framework for latent-space outlier generation that couples Sinkhorn optimal transport geometry with distributionally robust boundary modeling. The resulting Sinkhorn-induced support cost guides the sampler toward weakly supported boundary regions, while semantic constraints prevent uncontrolled drift from the intended context, yielding controlled deviations from the in-distribution reference measure rather than arbitrary sparse-region samples. Experiments on time series anomaly generation and image outlier synthesis show that our framework produces informative, semantically controlled outliers and improves downstream robustness evaluation across modalities, providing a foundation for stress scenario generation beyond empirical support.
[LG-115] LandscapeSHAP: Which Persistent Homology Class Gets the Credit?
链接: https://arxiv.org/abs/2609.31469
作者: Nikola Milićević
类目: Algebraic Topology (math.AT); Machine Learning (cs.LG)
*备注: 38 pages, 10 figures
Abstract:Shapley values, a solution concept from cooperative game theory, have recently become a standard tool for feature credit allocation in machine learning. They provide an axiomatically justified method to fairly distribute a model’s prediction among the data features. Shapley values have not yet been applied to explain machine learning models trained on features from topological data analysis. We develop what we believe is the first such approach, focusing on the persistence landscape featurization of persistence diagrams. Because each landscape coordinate is a rank statistic, crediting a model’s prediction back to individual persistent homology classes (persistence diagram points) is nontrivial. We introduce LandscapeSHAP, a method for fair credit allocation to persistence diagram points based on a model’s prediction. For linear models on persistence landscapes, LandscapeSHAP has a closed form expression that gives the exact Shapley value of every persistence diagram point. In particular, there is no coalition sampling required. We further prove that the four Shapley “fairness” axioms uniquely characterize this credit allocation for any model, not only linear ones. For a general nonlinear model, this unique value can only be calculated exactly from its defining coalition averaging formula, which requires considering all 2^N many coalitions, where N is the number of points in the persistence diagram. This is computationally intractable for persistence diagrams of realistic size. We complement the exact linear model result with an efficient Monte Carlo sampling of persistence diagram coalitions. We give convergence rates in terms of number of samples needed to approximate to a desired degree of accuracy. We also prove stability results for the LandscapeSHAP credit allocation, for any model.
[LG-116] Nonparametric In-Context Learning under Growing Geometric Complexity: Minimax Optimality and Local Geometry-Adaptivity of Transformers NEURIPS2026
链接: https://arxiv.org/abs/2609.31458
作者: Jaehee Seo,Jisu Kim
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Statistics Theory (math.ST)
*备注: 63 pages, 2 figures. Accepted at NeurIPS 2026
Abstract:Transformers have become a central architecture for in-context learning (ICL), particularly through their state-of-the-art performance in large language models. This success motivates understanding how transformers exploit task-relevant structure in geometrically heterogeneous data. However, existing nonparametric ICL theory has largely focused on Euclidean domains or single-manifold models. To address this gap, we study the prediction problem under unknown local geometry, modeled by sample size-dependent mixtures of manifolds with heterogeneous dimensions, smoothness, and sampling masses. Under local separation and small-perturbation conditions, we establish a minimax lower bound capturing the aggregate difficulty of the components and construct an oracle tangent local-polynomial estimator with a matching upper bound. This estimator is connected to a structure-informed, two-stage softmax transformer with a geometric preconditioner and chartwise reduced local-polynomial solvers. The transformer achieves negligible approximation error relative to the minimax rate with logarithmic depth and polynomial size. Finally, we derive an in-context generalization bound for near empirical risk minimizers over this class. Together, these results identify conditions under which the resulting predictor exploits local geometry and attains the aggregate minimax rate.
[LG-117] Equation discovery with Bayesian tree-adjoining grammars
链接: https://arxiv.org/abs/2609.31368
作者: Christopher A. Lindley,Nikolaos Dervilis,Keith Worden
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Systems and Control (eess.SY); Computation (stat.CO)
*备注:
Abstract:Tree-Adjoining Grammars (TAGs) have recently been introduced to Nonlinear System Identification (NLSI) as a means of encoding an entire model class as a finite set of grammatical rules, from which candidate models are assembled as trees. Existing TAG-based identifiers rely on evolutionary optimisation and return point estimates of the model structure. This paper instead proposes the TAG framework within a Bayesian setting. A generative prior is defined over tree structures and their parameters, and a Reversible-Jump MCMC sampler with structure-preserving tree moves is used to infer the joint posterior over model structure, parameters and predictions. Two training objectives are considered; that is, a one-step-ahead objective with conjugate parameter proposals, and a simulation-based objective handled by likelihood-free inference. The approach is validated on a simulated polynomial NARX system, the Silverbox benchmark, and wave-loading data from the Christchurch Bay Tower, where embedding Morison’s equation as a fixed initial tree yields a grey-box model that outperforms the physics-driven baseline. The results demonstrate that Bayesian TAGs are well suited to quantifying uncertainty in equation discovery for dynamical systems and to fitting physics-informed models.
[LG-118] Geometric Moment Contraction for Stochastic Nesterov Acceleration
链接: https://arxiv.org/abs/2609.31303
作者: Wei Biao Wu
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:We study geometric moment contraction (GMC) of the constant-parameter stochastic Nesterov recursion [ Y_k=\Theta_k+\beta(\Theta_k-\Theta_k-1),\qquad \Theta_k+1=Y_k-\gamma G(Y_k,X_k+1). ] Under mean strong monotonicity and stochastic L^p Lipschitz continuity, an explicit Perron comparison proves synchronous L^p contraction when \beta\gamma L_p(1-\beta)(1-q_\gamma,p) . This direct criterion includes infinite-variance gradients for 1p2 , but its small-step regime requires \beta\mu/(\mu+L_p) . A complementary power-Lyapunov argument establishes a positive, generally much smaller, step-size interval for every fixed \beta1 and every p1 , using only a finite p th gradient moment. At p=2 , a simpler explicit certificate gives [ 0\gamma\frac2\mu(1-\beta)^2L_2^2(1-\beta+2\beta^2). ] Its quadratic high-momentum scaling is a limitation of the chosen metric, not a sharp stability boundary. We quantify this loss, provide a general mean-only quadratic S -procedure, and exploit endpoint Lyapunov inequalities under stronger samplewise sector information. Verified endpoint certificates can be orders of magnitude less conservative than the explicit metric.
[LG-119] A Flatness-Generalization Relation in the Teacher-Student Tree-Committee Machine
链接: https://arxiv.org/abs/2609.31101
作者: Brandon Livio Annesi,Davide Straziota,Enrico Maria Malatesta
类目: Disordered Systems and Neural Networks (cond-mat.dis-nn); Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:The flatness of the loss landscape at a minimizer is a widely used heuristic for reasoning about neural-network generalization, yet evidence for this relation is mostly empirical and controversial. We study this relation in a teacher-student tree committee machine, where both the ERM estimator and the Hessian spectrum are analytically tractable in the proportional high-dimensional limit. First, we use a zero-temperature Gibbs formulation to obtain predictions for the observables of the typical minimizers of the empirical loss. Secondly, we use Edwards-Jones formalism to derive the limiting Hessian resolvent around these typical minimizers. All predictions agree with finite-size gradient-descent simulations. Finally, we study three measures of flatness, namely the left and right edges and the spectral mean, and check if a decrease in generalization error as the dataset size is increased corresponds to an increase in flatness. We find that the answer strongly depends on the learning task and on the ratio of the number of parameters to the number of data points. In regression, the spectral mean and right edge correlate with the generalization error, while the left edge does so only in the overparametrized regime. In classification this correlation reliably holds only in the highly overparametrized phase, while for underparametrized networks it can even reverse.
[LG-120] A Comprehensive Study of Content Representations for Speech Synthesis
链接: https://arxiv.org/abs/2609.30975
作者: Diego Torres,Axel Roebel,Nicolas Obin
类目: Audio and Speech Processing (eess.AS); Machine Learning (cs.LG); Sound (cs.SD)
*备注: 5 pages, 1 figure
Abstract:Speech content representations are central to voice conversion, speech-to-speech translation, and multimodal language models, yet they are rarely compared under a common generative framework that directly measures what each representation contains. We address this by training a generative model conditioned solely on each representation and evaluating the generated audio along the content, speaker identity, and prosody axes. Across SSL features, supervised tokens, posteriorgrams, and neural audio codecs, we find two distinct regimes: representations that nearly reconstruct the original audio, and representations that effectively disentangle speaker identity. These results show that disentanglement depends not on supervision alone, but on the interaction between the training objective and the representation’s information capacity: supervised representations only disentangle speaker identity when their capacity is sufficiently constrained.
[LG-121] Conformal Prediction under Exponential-Tilt Joint Shift
链接: https://arxiv.org/abs/2609.30886
作者: Seungjin Choi
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Methodology (stat.ME)
*备注: 23 pages
Abstract:Conformal prediction can lose coverage when the data distribution changes after deployment. We study adaptation using labeled source data and unlabeled target inputs, allowing both the input distribution and its relationship with outcomes to change. We use Exponential Tilt Reweighting Alignment (ExTRA), introduced for classification by Maity et al. (2023), to estimate structured distribution shifts. We compare using its estimated weights in conformal calibration with additionally tilting the source predictive distribution. Shared learned predictors, estimated weights, calibration samples, and test observations isolate the effect of tilting. Existing theory gives both procedures target coverage with true weights and a common coverage bound with estimated weights. Identification calculations and an analysis of how scoring interacts with weight estimation error help explain why their performance can nevertheless differ. In a synthetic regression setting where the assumed models match the data-generating process and target inputs are informative about the shift, tilting reduces mean set length by about 30% relative to weighting alone, with both methods attaining coverage near nominal. Tilting can instead cause substantial coverage losses in synthetic classification and in regression when target inputs provide little information about the response shift. Real-data experiments also show no consistent benefit. Good coverage from weighted calibration alone does not ensure that adding predictive tilting will preserve coverage. Deciding when to apply this additional adjustment using only source labels and target inputs remains an open problem.
[LG-122] Retraction-Based Gradient Projection Algorithms on Manifolds
链接: https://arxiv.org/abs/2609.30885
作者: Conglong Xu,Hao Wu
类目: Optimization and Control (math.OC); Machine Learning (cs.LG)
*备注:
Abstract:We introduce a framework for retraction-based convex optimization on Riemannian manifolds, which includes a notion of retraction-specific convex sets and retraction-based gradient projection algorithms. The standard theory of gradient projection algorithms generalizes easily to this framework. Within this framework, we establish convergence results for retraction-based gradient projection algorithms with various stepsize rules. As an application, we use our framework to study the weighted low-rank approximation. We also provide numerical validation of our convergence results on the image completion task.
[LG-123] ght Stochastic Condition-Number Dependence in Nonconvex-Strongly-Concave Minimax Optimization
链接: https://arxiv.org/abs/2609.30877
作者: Qihao Zhou
类目: Optimization and Control (math.OC); Machine Learning (cs.LG)
*备注: 20 pages
Abstract:We study whether the linear condition-number dependence in the stochastic complexity of SAPD+ is necessary for nonconvex-strongly-concave minimax optimization. For jointly L -smooth objectives with dual strong-concavity parameter \mu , we prove a lower bound that matches the SAPD+ upper bound under the same Moreau-envelope stationarity criterion and the same primal-dual initialization gap. Specifically, when \sigma\ge\varepsilon , the worst-case complexity of zero-respecting algorithms is \Theta(\kappa LG\sigma^2\varepsilon^-4) in the stated accuracy regime, where \kappa=L/\mu , G bounds the initial primal-dual gap, and \sigma^2 bounds the variance of a general unbiased first-order oracle. The lower bound is realized on a smooth problem class with a bounded dual box. Our construction routes each link of a nonconvex zero-chain through a dual gradient of magnitude proportional to \varepsilon/\sqrt\kappa , while an undiscovered primal coordinate prevents stationarity. It also yields the primal-gradient lower bound \Omega(L\Delta(\sqrt\kappa\varepsilon^-2+\kappa\sigma^2\varepsilon^-4)) after combination with the known deterministic bound, where \Delta bounds the initial primal function gap.
[LG-124] R-SSQP: A Trust-Region Method for Constrained Stochastic Optimization under Heavy-Tailed Noise
链接: https://arxiv.org/abs/2609.30732
作者: Haoxuan Wang,Yuchen Fang,Sen Na
类目: Optimization and Control (math.OC); Machine Learning (cs.LG); Computation (stat.CO); Machine Learning (stat.ML)
*备注: 32 pages, 5 figures, 3 tables
Abstract:We consider stochastic nonlinear optimization problems with deterministic equality constraints. While unconstrained stochastic optimization is well understood, the interplay between optimality and feasibility in the constrained setting poses significant challenges. Moreover, existing theoretical guarantees for constrained stochastic methods predominantly rely on bounded-variance assumptions, leaving the heavy-tailed noise regime largely unexplored. To address this gap, we propose a novel trust-region method within the stochastic sequential quadratic programming framework, termed TR-SSQP. Our method employs a normal-tangential decomposition in the step computation to balance optimality and feasibility. In addition, we incorporate a normalization mechanism in the design of the trust-region radius, together with Polyak momentum for gradient estimation, ensuring stable updates without gradient clipping. When the trust-region radius and the momentum parameter decay at appropriate rates, we establish global almost-sure convergence of the method. To the best of our knowledge, this is the first asymptotic convergence result for constrained stochastic optimization under heavy-tailed noise. We demonstrate the promising performance of the proposed method through extensive numerical experiments, including comparisons among its variants and with existing constrained stochastic optimization methods.
[LG-125] Parameter Estimation for Unnormalized Discrete Models via Empirically Localized Deformed Bregman Divergence
链接: https://arxiv.org/abs/2609.30713
作者: Takashi Takenouchi
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注: 29 pages, 9 figures
Abstract:Estimation of parameter of probabilistic models is an important task in the field of machine this http URL models of discrete variables, calculation of the normalization constant of model is sometimes difficult and a lot of researches have been done to avoid the calculation of the normalization constant. In this paper, we tackle with the difficulty by combining a technique of empirical localization and a deformed Bregman this http URL technique of empirical localization makes it possible to drastically reduce computational cost of the calculation of the normalization constant, and in addition, appropriate choice of the deformation for the Bregman divergence can invest the proposed estimator with various kinds of favorable statistical properties, such as efficiency or robustness against outlier noise.
[LG-126] On the Limits of Univariate Deep Learning for Significant Wave Height Forecasting
链接: https://arxiv.org/abs/2609.30688
作者: Yilin Zhai,Hongyuan Shi,Zaijin You
类目: Atmospheric and Oceanic Physics (physics.ao-ph); Machine Learning (cs.LG); Computational Physics (physics.comp-ph)
*备注: 33 pages, 13 figures. Author-accepted manuscript
Abstract:This study conducts a systematic hyperparameter search across five deep learning architectures, DLinear, LSTM, PatchTST, ResAttLstm, and Mamba2, and nine context lengths (1-168 h) for single-station significant wave height (Hs) forecasting on NDBC buoy 41009, followed by re-evaluation of the best configurations on a 47-buoy, 37-year corpus. The five families converge to a common performance level on the multi-buoy evaluation (between-family SD = 0.0014 m^2, 0.8% of the grand mean), a spread dwarfed by the 4.83x cross-dataset MSE shift between buoy corpora. All multi-buoy trials beat persistence (mean skill +0.062), but no architecture consistently outperforms the others. On the single-buoy experiment, skill peaks at 12-24 h where five trials fall below persistence, per-family Q4/Q3 test MSE ratios range from 2.4 to 2.6, and deep models underperform persistence for the most extreme 1% of waves. These findings are consistent with the interpretation that persistence already captures the dominant linear-inertial signal in univariate Hs, and that architecture engineering under this univariate input setting has reached diminishing returns: cross-buoy variance, not model class, dominates forecast error. Future work should prioritise atmospheric covariates, zero-shot cross-buoy transfer, and decomposition of Hs into swell and wind-sea components. By establishing a rigorous reference baseline for what univariate Hs models can and cannot achieve, this study provides a benchmark against which future multivariate and physics-informed approaches can be calibrated, and offers practical guidance for lightweight buoy-level forecasting in mid-latitude storm-dominated and swell-mixed environments.
[LG-127] MARCEDES: Score-based causal discovery under non-Gaussianity with continuous optimization
链接: https://arxiv.org/abs/2609.30643
作者: Anamitra Chaudhuri,Anirban Bhattacharya,Yang Ni
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Computation (stat.CO); Methodology (stat.ME)
*备注:
Abstract:We consider the problem of learning the underlying causal directed acyclic graph (DAG) structure corresponding to a structural equation model (SEM) with non-Gaussian errors. Motivated by an intentionally misspecified non-Gaussian SEM with all Laplace errors, we first introduce the mean absolute residual risk, defined over the space of all real matrices, and show that, asymptotically, the risk of the true weighted causal DAG matrix is strictly smaller than that of any other matrix. Nevertheless, to enhance generality and account for high-dimensional and finite-sample settings, we further incorporate row-specific sparsity penalties along with a soft DAG constraint to derive a continuous score function over the space of real matrices. Accordingly, we propose a score-based DAG learning method, named MARCEDES, formulated as an unconstrained score minimization problem, which can be efficiently solved using gradient-based optimization techniques, thereby circumventing the challenges associated with constrained optimization. Furthermore, we develop a computational algorithm to handle the non-smoothness of the score objective and to enable optimal tuning of row-specific sparsity penalties under a generalized Bayes framework. Finally, we demonstrate the efficiency and improved performance of the proposed method over existing approaches through an extensive simulation study.
[LG-128] Encryptability As a Coordinate Choice: Depth-One Homomorphic Federated Learning of Quantum Neural Networks
链接: https://arxiv.org/abs/2609.30581
作者: Marcel Mordarski,Nathan Mani,Arshad Patel,William Knottenbelt,Roberto Bondesan
类目: Quantum Physics (quant-ph); Cryptography and Security (cs.CR); Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG)
*备注: Presented as submission #203 at QCrypt 2026 this https URL
Abstract:Encrypted training relies on keeping server-side updates low-degree. This constraint traditionally excludes models whose weights inhabit a compact Lie group (notably variational quantum circuits, where every trainable weight is an \mathrmSU(2) rotation). Expressed in Euler angles or discrete alphabets, these updates appear transcendental, historically demanding prohibitive costs: one client–server round per gate, or upwards of 25,000 operations per weight. This penalty is strictly an artefact of coordinates. In the unit-quaternion (spin) chart, group composition is exactly bilinear (degree two, with coefficients in -1,0,+1\ ). Consequently, encrypted rotation updates cost one multiplicative level and federated averaging costs zero in any levelled homomorphic scheme, completely eliminating bootstrapping. This implementation-independent algebraic property is confirmed across two cryptographic backends, introducing only 0.0 and -2.0\times10^-12 rad of aggregation error. Leveraging this reduction yields a non-interactive protocol for encrypted federated training of hybrid quantum–classical networks. It includes correctness proofs for aggregation and sign handling, plus a compilation lemma proving parameterised entanglers add only constant-factor overhead without altering the depth class. Empirically, a paired five-seed study confirms zero measurable utility tax ( \Delta=+9\times10^-6 MSE, p=0.92 ), and a noise-budget ablation falsifies the hypothesis that encryption noise regularises. These convergence trends replicate across datasets and scale to 20 clients. Finally, hardware validation on a 156 -qubit processor achieves 0.9918 fidelity against a 0.99957 unencrypted control.
[LG-129] Learning to Replace MCMC in Split-Gibbs Diffusion Posterior Sampling via Deep Unfolding
链接: https://arxiv.org/abs/2609.30539
作者: Yi Zhang,Rui Guo,Mengchu Xu,Zhaofeng Liu,Yonina C. Eldar
类目: ignal Processing (eess.SP); Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 5 pages, 2 figures
Abstract:Split Gibbs sampling enables diffusion posterior inference for general nonlinear inverse problems by decoupling prior and likelihood computations, allowing a pretrained diffusion prior to be reused across measurement models. However, its likelihood update often relies on iterative MCMC, which can hinder parallelization, require algorithm-specific tuning, and incur substantial computational cost. In this work, we propose a learning-based framework to replace this MCMC step by reformulating both Gibbs updates as Gaussian denoising problems and implementing them through ODE diffusion. The prior step reuses a pretrained denoiser, while the likelihood denoiser exploits known likelihood structure through a lightweight deep-unfolded network. Experiments on nonlinear phase retrieval demonstrate the effectiveness of the proposed method as an alternative to MCMC-based split Gibbs at lower likelihood-update cost.
[LG-130] Seeing Speech: Learning Visible Articulatory Dynamics for Speech-Driven 3D Facial Animation NEURIPS2026
链接: https://arxiv.org/abs/2609.30517
作者: Hyung Kyu Kim,Byungchan Hwang,Hak Gu Kim
类目: Audio and Speech Processing (eess.AS); Graphics (cs.GR); Machine Learning (cs.LG)
*备注: Accepted to NeurIPS 2026
Abstract:Recent progress in speech-driven 3D facial animation has improved vertex-level reconstruction quality, but speech-consistent visible articulation remains difficult. This is because speech production follows structured and constrained articulators’ coordination and the mapping from acoustics to motion is inherently one-to-many. Motivated by the structured patterns of visible articulation, we propose a novel articulation-aware framework that models visible speech through directional articulatory motions and composes them into surface-consistent 3D facial motion. To represent visible articulation with three directional articulatory motions, spreading, opening, and protrusion, we propose a Speech–Articulatory Memory (SAM) that captures the correspondence between speech and these motions under phonetic context through retrieval and decoding based on a key-value memory structure. Then, a Topology-aware Articulatory Composition (TAC) integrates the predicted directional articulatory motions under mesh topology to produce surface-consistent 3D facial motion. Experiments on VOCASET and TFHP show that our method achieves state-of-the-art performance on standard reconstruction metrics and improves visible articulatory distance and velocity errors for lip articulation, while a user study confirms clear preference in lip sync and realism.
[LG-131] Ordinary Nonconvex SGD under Distance-Dependent Moments: Finite-Horizon Stationarity and Nagaev Bounds
链接: https://arxiv.org/abs/2609.30499
作者: Wei Biao Wu
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:Uniform noise-moment bounds exclude stochastic gradients whose variability increases with the iterate. We study ordinary, single-sample stochastic gradient descent for smooth, lower-bounded, possibly nonconvex objectives under distance-dependent conditional moments. Under second moments alone, a direct descent–displacement argument yields T^-1/3 expected average squared-gradient stationarity with a horizon-dependent stepsize. An explicit oracle-complexity corollary matches the known smooth Blum–Gladyshev (BG-0) lower bound, including the Lb_2\Delta^3\varepsilon^-6 and L\Delta\sigma^2\varepsilon^-4 stochastic terms, where \Delta is the initial objective gap and \sigma^2+b_2|x-x_1|^2 bounds the variance. Thus unchanged SGD attains the minimax stochastic complexity in this second-moment class. For p2 , predictable localization and a Hilbert-space Fuk–Nagaev inequality yield a high-probability bound separating logarithmic variance and polynomial rare-shock contributions. The localization radius is derived from the recursion: no bounded-iterate assumption, clipping, normalization, momentum, or increasing batch size is needed. We also give increasing-confidence rates, an objective-gap-growth refinement recovering root- T stationarity, and stochastic L^p -Lipschitz examples. The broad BG-0 optimality statement is distinguished from the smaller mean-square-smooth class, in which additional oracle structure permits faster algorithms.
[LG-132] Scaffold-Constrained Subset Dynamic Programming for Exact SSE Clustering
链接: https://arxiv.org/abs/2609.30477
作者: Yordan P. Raykov,Max A. Little
类目: atistics Theory (math.ST); Machine Learning (cs.LG)
*备注:
Abstract:Exact Euclidean (K)-means partitions (n) observations into (K) unlabelled clusters, but the unrestricted search is generally exponential. We use data-derived geometric graphs to precondition an exact subset dynamic program: as a result only connected vertex subsets are admitted as clusters, while sum-of-squared-errors (SSE) loss is unchanged. A remaining-set recurrence minimises fixed-(K) or penalised SSE, with exact factorisation over the connected components of each remaining set. The central question we study is how much computational support can be removed while preserving an unrestricted optimum. Graph inclusion gives monotone coverage and support relations, and a bottleneck threshold identifies the first covering graph in a nested hierarchy. For fixed (K) and dimension, under compact ball support and density bounds, retaining (q=O(\log n)) nearest neighbours per observation preserves an empirical SSE optimum with probability tending to one, using an (O(\log n/n)) fraction of complete-graph edges. Truncated Gaussian mixtures with unequal weights and covariances satisfy these conditions. The rate we provide is a sufficient upper bound rather than a result implying polynomial optimisation complexity. Objective-matched synthetic and full-data comparisons assess coverage, compression, and reference-label agreement. As a secondary application, we illustrate how the proposed scaffold preconditioning can be utilized to improve the efficiency of split-merge proposals that preserve unrestricted mixture posteriors.
[LG-133] Bayesian Uncertainty Quantification for fMRI Functional Connectivity via Simulation-Based Inference
链接: https://arxiv.org/abs/2609.30445
作者: Simon Carter,Zeming Kuang,Lilianne R. Mujica-Parodi,Helmut H. Strey
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注: 29 pages, 4 figures
Abstract:Optimizing fMRI scan duration and spatial resolution is critical for experimental design, yet traditional correlation-based approaches cannot quantify uncertainty or disentangle scanner measurement noise from true neural variability across subjects. Without principled uncertainty bounds, researchers cannot know whether a protocol is long enough to reliably estimate connectivity, or whether between-subject differences reflect biological variation or noise. We present a Bayesian framework modeling BOLD dynamics as coupled Ornstein-Uhlenbeck processes, using Sequential Neural Posterior Estimation to obtain connectivity posteriors while accounting for frequency-independent measurement noise across the BOLD spectrum. Applied to N = 28 healthy controls (55 scans) at 7T using a functional network atlas (65 DMN regions), the framework quantifies uncertainty across its sources: scanner noise, subject variability, and acquisition length. Spatial analysis identifies a mean of 46 voxels per ROI, roughly half of typical region sizes, as sufficient to achieve 90% of asymptotic precision. At the single-subject level, 7T reaches its within-session precision plateau in approximately 7 minutes versus 10 minutes for 3T, a 40% reduction in required scan time, providing the first direct, model-based quantification of the scan-time advantage conferred by higher field strength. At the population level, 3T requires roughly 37 times more per-subject scan time than 7T for the pooled curves to converge, confirming a consistent advantage of higher field strength at every timescale. Together these findings provide concrete, scanner-specific guidance for protocol optimization, with direct implications for reducing acquisition costs and improving the reliability of connectivity-based clinical biomarkers. We provide code enabling researchers to derive these bounds from their own data.
[LG-134] An End-to-End Pipeline for Causal ML with Continuous Treatments: An Application to Financial Decision Making KDD2025
链接: https://arxiv.org/abs/2609.30396
作者: Javier Moral Hernández,Clara Higuera-Cabañes,Álvaro Ibraín
类目: Methodology (stat.ME); Machine Learning (cs.LG); Applications (stat.AP); Machine Learning (stat.ML)
*备注: Oral presentation at the 3rd Workshop on Causal Inference and Machine Learning in Practice, KDD 2025, Toronto. Code: this https URL
Abstract:This paper presents an end-to-end causal machine learning (ML) pipeline designed for real-world applications with continuous treatments. The proposed framework consists of six sequential steps: dimensionality reduction, causal identification, positivity assumption violation handling, estimation, refutation and evaluation, and policy optimization. We introduce practical contributions not currently available in existing causal ML toolkits, specifically: (1) a method for detecting and quantifying positivity violations in continuous treatment settings (2) a novel, scalable two-stage dimensionality reduction framework tailored for causal inference with high-dimensional data; (3) the adaptation of sensitivity analysis and estimation methods originally designed for binary treatments to the continuous treatment space and (4) an end-to-end integration of these components into a modular, reproducible workflow. These innovations address real-world challenges in causal inference that are often not covered in theoretical frameworks but frequently encountered in industrial applications. The methodology is validated with a synthetic dataset inspired in a real-world financial debt collection use case, however its design can be applied to analogous problems across different industries. Results demonstrate that the proposed methodology offers a more computationally efficient approach and produces less biased estimates compared to standard methods for problems with continuous treatment and high-dimensional data. A fully functional GitHub repository with documented code and numbered notebooks is made available ensuring reproducibility and practical implementation. The pipeline presented is intended to contribute to closing the gap between academic approaches and practical application in industry contexts where causal ML can be highly beneficial such as the financial sector.
[LG-135] Low-Rank Friction for Memory-Efficient Transformer Pretraining
链接: https://arxiv.org/abs/2609.30342
作者: Rajit Rajpal,Benedict Leimkuhler
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:iKFAD is a recently proposed optimiser that replaces adaptive learning rates with adaptive friction in the momentum dynamics, yet performs as well as Adam. Its limitation is that the full friction tensor \xi\in\mathbbR^m\times n carries the same \mathcalO(mn) memory overhead per layer as Adam’s second-moment buffer. Here we replace iKFAD’s friction tensor \xi with a rank-1 outer-product factorisation built from row and column momentum statistics, resulting in Rank-1 iKFAD (R-iKFAD). This reduces the friction memory footprint from \mathcalO(mn) to \mathcalO(m+n) per layer, which approximately halves iKFAD’s total optimiser state. Despite this reduction, R-iKFAD maintains parity in performance with iKFAD: experiments on GPT2-Nano, TinyViT, DistilBERT and GPT2-S confirm that it matches or exceeds iKFAD while nearly halving the memory footprint and remaining comparably robust to hyperparameters. We analyse the continuous-time dynamics in two damping regimes. For linear damping ( \gamma0 ) we prove exponential convergence under strong convexity. For \gamma=0 , the preferred option in our experiments, the friction is generated entirely from past momentum and switches off as the momentum vanishes, so geometric convergence cannot be shown. We nonetheless prove convergence to the minimiser, together with matching upper and lower bounds on the energy: of order t^-1 when the regularisation scale \epsilon_\mathrmstab is zero, and of order t^-1/2 when it is positive. To our knowledge this is the first convergence rate for a rank-1 factored optimiser in continuous time, and the first such result that does not require positive damping.
[LG-136] Adaptive Random Matrices in Gaussian Bandits: Spectral Universality and Selection-Induced Outliers
链接: https://arxiv.org/abs/2609.30321
作者: Sudarshan Manikantan,Abhishek Bhattacharjee(Abstract Math Institute)
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Probability (math.PR); Statistics Theory (math.ST)
*备注: 16 Pages
Abstract:Adaptive arm selection changes the distribution of the observations collected by a bandit algorithm, but it need not change their limiting empirical spectrum. We study Gaussian bandit designs in which the dimension and the number of observations grow proportionally. A quantitative coupling theorem compares the design generated by any causal selection rule with an independent Gaussian design. If the logarithm of the number of available arms is sublinear in the dimension, the empirical spectral distribution converges to the Marchenko-Pastur law, uniformly over the selection rule. Consequently, Gaussian Bayesian bandits have policy-independent first-order limits for posterior mean-square uncertainty, squared posterior covariance, and information acquisition. For linear-score selection, we obtain the exact conditional arm distribution and show that two-arm selection produces an exactly Wishart Gram matrix in every dimension, despite its nonzero conditional mean. For a fixed selection direction, we identify an explicit eigenvalue and eigenvector transition governed by the second moment of a Gaussian maximum. A counterexample shows that a direction’s overlap with a reference signal does not determine this transition. Finally, an exponentially large arm pool permits a different bulk limit, establishing the order-sharpness of the arm-growth condition. These results distinguish global spectral stability from directional effects in adaptive bandit data.
[LG-137] Distribution of hitting times for dissipative random dynamical systems on mathbbRd with application to stochastic gradient descent
链接: https://arxiv.org/abs/2609.30274
作者: Stéphane Galatolo,Stéphane Chrétien
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Dynamical Systems (math.DS)
*备注: 24 pages
Abstract:Machine Learning and more specifically Deep Learning involves solving large scale nonconvex optimization problems. Several algorithms have been proposed in the literature, that seem to achieve satisfactory practical efficiency for difficult instances, the Stochastic Gradient Method being the most rudimentary, while still outperforming more recent algorithms at a number of learning tasks. A major open question about the current methods used in deep learning is to understand their convergence properties. Following a line of previous works about the long time behavior of gradient-type algorithms, %and in particular the recent contributions from Azizian et al., we present a new approach for studying the asymptotic properties of a wide family of methods from an ergodic theoretical viewpoint. Our main results include a study of the expected time for a stochastic optimisation algorithm to reach a certain small neighborhood of a minimizer and show that this reaching time distributes exponentially around its average, which is given by the inverse of the stationary measure of the target. The assumptions on the Stochastic Gradient noise include the Gaussian and the Sub-Exponential assumptions. Comments: 24 pages Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG); Dynamical Systems (math.DS) Cite as: arXiv:2609.30274 [stat.ML] (or arXiv:2609.30274v1 [stat.ML] for this version) https://doi.org/10.48550/arXiv.2609.30274 Focus to learn more arXiv-issued DOI via DataCite
[LG-138] Persistent Homology of Time Series through Complex Networks
链接: https://arxiv.org/abs/2605.01624
作者: İsmail Güzel
类目: Algebraic Topology (math.AT); Machine Learning (cs.LG); Applications (stat.AP); Machine Learning (stat.ML)
*备注:
Abstract:We present a unified pipeline for univariate time series classification via complex networks and persistent homology. A time series is mapped to a graph through one of five constructions across three families (visibility (natural and horizontal visibility graphs), transition, and proximity) and the graph is converted to a dissimilarity matrix from which a Vietoris-Rips filtration yields persistence diagrams. These diagrams are vectorized into fixed-length features through persistence landscapes and topological summary statistics. By standardizing the downstream processing, differences in classification performance are attributable to the network construction and distance metric alone. Experiments on twelve UCR benchmarks show that (i) no single construction dominates: the optimal graph type depends on the signal’s discriminative structure; (ii) the graph distance metric is a first-order design choice, with diffusion distance uniformly outperforming shortest-path alternatives; and (iii) persistence-based features degrade gracefully under noise, consistent with the classical stability theorem of persistent homology.
[LG-139] A New Non-archimedean Metric on Persistent Homology
链接: https://arxiv.org/abs/2012.02655
作者: İsmail Güzel,Atabey Kaygun
类目: Algebraic Topology (math.AT); Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 18 pages, 6 figures
Abstract:In this article, we define a new non-archimedean metric structure, called cophenetic metric, on persistent homology classes of all degrees. We then show that zeroth persistent homology together with the cophenetic metric and hierarchical clustering algorithms with a number of different metrics do deliver statistically verifiable commensurate topological information based on experimental results we obtained on different datasets. We also observe that the resulting clusters coming from cophenetic distance do shine in terms of different evaluation measures such as silhouette score and the Rand index. Moreover, since the cophenetic metric is defined for all homology degrees, one can now display the inter-relations of persistent homology classes in all degrees via rooted trees.
附件下载
