本篇博文主要内容为 2026-08-10 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 (2026-08-10)
今日共更新605篇论文,其中:
- 自然语言处理共77篇(Computation and Language (cs.CL))
- 人工智能共202篇(Artificial Intelligence (cs.AI))
- 计算机视觉共103篇(Computer Vision and Pattern Recognition (cs.CV))
- 机器学习共145篇(Machine Learning (cs.LG))
- 多智能体系统共11篇(Multiagent Systems (cs.MA))
- 信息检索共16篇(Information Retrieval (cs.IR))
- 人机交互共28篇(Human-Computer Interaction (cs.HC))
多智能体系统
[MA-0] Strategy-first synthesis planning for complex natural products
【速读】：该论文旨在解决复杂天然产物全合成中自动化逆合成设计的瓶颈问题，特别是传统基于反应数据库的算法在面对结构高度功能化、多环稠合的前沿分子时表现不佳的问题。其核心挑战在于：现有工具受限于训练数据所覆盖的化学反应范围，难以应对需要创造性化学策略的复杂分子构建。解决方案的关键是提出一种基于大语言模型（Large Language Models, LLMs）的代理式框架——SynthEx，该框架具备自主规划、生成多种竞争性合成路径、整合常规与关键步骤形成连贯路线，并能自我评估与优化设计的能力。SynthEx所提出的合成策略更具收敛性，且探索了传统数据库工具无法触及的反应空间。在盲评测试中，专家化学家认为其关键步骤与已发表的人工合成方案相当，且将其视为真实可行的合成计划，这是以往算法路线预测从未实现的突破。研究团队还发布了超过一千个天然产物的合成路线，构建了名为SynthAtlas的开放交互数据库，为缺乏文献路线的复杂目标分子提供共享资源。
链接: https://arxiv.org/abs/2608.07454
作者: Daniel Armstrong,Xuan-Vu Nguyen,Octavian Susanu,Gabriel Gibberd,Théo A. Neukomm,Taddäus Strunden,Dan Forster,Morgane Delattre,Shawn Teh,Clément Rols,John Federice,Hayden Leatherwood,M. Lavelle Barnes,Maarten R. Dobbelaere,Peter Wipf,Jon T. Njardarson,Jieping Zhu,Philippe Schwaller
机构: EPFL(洛桑联邦理工学院); University of Pittsburgh (匹兹堡大学); University of Arizona (亚利桑那大学)
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI)
备注:
Abstract:The total synthesis of a complex molecule is among the most demanding intellectual and experimental feats in chemistry: a chemist must plan many steps ahead for how to assemble simple building blocks into an intricate target, devise backup strategies, and anticipate procedural challenges. It is also a profoundly creative activity. For half a century, efforts to automate the retrosynthetic design of natural products and other complex molecules have drawn on catalogued reactions, and the resulting tools now report near-complete success on benchmarks built from that same source. But these tools were shaped to fit benchmarked chemistry, and they falter on many natural products, the frontier of the field, whose densely functionalized, polycyclic architectures demand precisely the inventive chemistry the record contains least. Whether a machine could reasonably design such syntheses like an expert chemist does has remained unclear. Here, we show that SynthEx, an agentic framework built on large language models, plans routes to complex natural products that lie beyond the reach of conventional design algorithms. SynthEx proposes competing strategies, assembles a sequence of routine and key steps into a cohesive route, and critiques and improves its own design; the chemistry it favours is more convergent than existing tools produce, and spans a region of reaction space that catalogue-based tools cannot match. Most notably, in blinded assessments, expert chemists judged its key steps comparable to those of published human syntheses and engaged with them as genuine synthesis plans, a response algorithmic route prediction has not previously accomplished. We release routes to more than a thousand natural products as SynthAtlas, an open, interactive database, and anticipate it will become a shared resource for a collection of complex target molecules that lack existing literature routes.
[MA-1] opology Inference for Immune System Networks by Using Cell Amount Data
【速读】：该论文旨在解决免疫细胞网络拓扑结构推断中的关键挑战，即在缺乏标准解析模型以及实验数据受限（受高昂实验成本和时间限制）的条件下，准确揭示细胞间相互作用关系。其解决方案的关键在于：首先，基于实验中的普遍知识与观察，提炼出细胞数量变化过程中的三个核心性质——状态非负性、比例收敛性以及拓扑权重的三重符号特性；其次，构建一个结构简洁且具有解析便利性的新模型，并推导出确保模型满足上述三类性质的充分条件；最后，基于该模型提出一种约束二次规划方法，从有限的数据对中实现拓扑结构的有效推断。实验数据验证表明，该方法在实际应用中具有良好的有效性与可靠性。
链接: https://arxiv.org/abs/2608.07403
作者: Yushan Li,Rikard Forlin,Dimos V. Dimarogonas,Petter Brodin
机构: KTH Royal Institute of Technology (瑞典皇家理工学院); Karolinska Institutet (卡罗林斯卡学院)
类目: ystems and Control (eess.SY); Multiagent Systems (cs.MA)
备注: Accepted by IFAC World Congress 2026
Abstract:Recent years have witnessed the advanced development of topology inference research, which helps elucidate the interaction relationships of components in many biological networks. This paper focuses on inferring the topology of a group of immune cells, based on the collected data from cell-depletion based experiments. The problem is very challenging due to i) the lack of standard analytical models for the cell interactions, and ii) the restrictive data availability determined by the huge experiment and time costs. To address these issues, we first leverage certain common knowledge and observations on the experiments to characterize three properties on the cell amounts during the interaction process: state non-negativity, ratio-based convergence, and triple signs of topology weights. Then, we construct a new model with simple structure and analytical convenience, and obtain sufficient conditions for the model to accommodate all three properties. Finally, based on the constructed model, we propose a constrained quadratic programming method to infer the topology from limited number of data pairs. Validation on experiment data demonstrate the effectiveness of the proposed method.
[MA-2] Analyzing the Interaction of Optimal Strategies in Mean-Payoff Bidding Games
【速读】：该论文旨在解决多智能体系统（multi-agent system）中，当所有智能体均基于对抗性假设设计策略时，实际交互行为并非对抗性而是趋于防御性，从而导致系统演化出非预期行为的问题。核心挑战在于，此类系统中智能体间的动态交互在数学与算法层面极难分析，尤其在无限博弈场景下。本文聚焦于图上的竞标博弈（bidding game），其中两个智能体通过拍卖机制决定令牌移动权，进而生成无限路径并决定各自的收益。研究关注平均收益（mean-payoff）目标，即每个顶点对玩家具有奖励值，长期效用为奖励序列的极限平均值。关键解决方案在于：针对两种已知的最优策略显式构造方法，证明在特定约束条件下，系统生成的博弈路径最终呈现周期性，并据此开发了可计算双方玩家效用的算法，克服了竞标博弈中配置空间无限且动力学复杂的技术难题。
链接: https://arxiv.org/abs/2608.07383
作者: Shaull Almagor,Guy Avni,Julian Ewaied
机构: Technion (以色列理工学院); University of Haifa (海法大学)
类目: Computer Science and Game Theory (cs.GT); Multiagent Systems (cs.MA)
备注:
Abstract:A common assumption when designing an agent in a multi-agent system is that the other agents behave adversarially. This allows a designer to obtain the strongest guarantees when they have no control over nor knowledge about the other agents’ behavior. However, when all agents are designed under this adversarial assumption, their actual interaction is not adversarial (e.g., when all players play defensively, no player actually attacks). In such settings, we would like to know what behavior arises in the multi-agent system. However, analyzing the interaction among agents is notoriously challenging, both mathematically and algorithmically. In this paper, we provide such an analysis, focusing on bidding games, played by two agents on a graph as follows. A token is placed on a vertex, and in each turn an auction (bidding) determines which agent moves the token, thus generating an infinite path that determines the agents’ utilities. We consider mean-payoff objectives; each vertex is associated with a reward for each player, and the utility in an infinite play is the limit average of the rewards. We analyze the play that is generated when each agent follows a strategy that optimizes against an adversary, and consider the two known explicit constructions of optimal strategies. The technical challenge stems from the infinitely-many configurations of a bidding game and their complicated dynamics. We show that, under some restrictions, the generated play is ultimately periodic, and develop algorithms to compute the players’ utilities in it.
[MA-3] Learning Long-Term Educational Investment Policies under Residential Sorting
【速读】：该论文旨在解决在基于居住地分配公立学校资源的制度下，如何实现教育投资的有效性与公平性之间的平衡问题。核心挑战在于，学校质量提升会引发周边住房需求与价格上升，进而通过居民择校行为改变学校生源结构，可能加剧低收入家庭的入学障碍，而这种反馈循环随时间演进并影响未来的资源配置需求。传统研究多将学校经费、家庭选择与住房市场割裂分析，且静态模型难以捕捉其动态交互效应。本文提出一种动态多主体框架，将政府投资、家庭居住选择、住房价格、人口流动、学生入学及学校质量演变等要素耦合建模，并引入强化学习（Reinforcement Learning, RL）方法，使政府规划者能够制定考虑家庭响应的多年度投资策略，在保障整体教育可及性的同时兼顾公平性。仿真结果表明，该方法在代表性基线中实现了最高的教育可及率（0.4780）和第二低的可及性基尼系数（0.0164），展现出优越的有效性-公平性权衡能力，并有效缓解了教育机会中的社会经济分层现象。该框架通过显式刻画教育与住房系统的反馈机制，为长期评估学校投资对教育机会的影响提供了理论与方法支持。
链接: https://arxiv.org/abs/2608.07295
作者: Honglei Guo,Shuo Chen,Mingjie Bi,Zeyang Sun,Xiaoxi Wang,Yuhan Zhao
机构: Hpnglei Guo1, Shuo Chen1, Yuhan Zhao\corresponding1; Honglei Guo 1,2, Shuo Chen 2, Mingjie Bi2, Zeyang Sun2, Xiaoxi Wang2, Yuhan Zhao2
类目: Multiagent Systems (cs.MA)
备注:
Abstract:Allocating public-school investment effectively and fairly is difficult when school access depends on residence. School improvements can raise nearby housing demand and prices, reshape enrollment, and potentially limit access for lower-income households. These effects evolve as residential sorting changes school composition, quality, and future investment needs. Existing approaches often study school funding, household choice, and housing markets separately, while static models can miss their interconnected, long-term effects. We address this gap with a dynamic multi-agent framework that links government investment, household sorting, housing prices, population turnover, enrollment, and evolving school quality. A government planner uses reinforcement learning (RL) to identify multiyear allocation policies that account for household responses while balancing aggregate educational access and equity. In simulations, our RL-based policy attains the highest access level (0.4780) and second-lowest access Gini coefficient (0.0164) among representative baselines, demonstrating a favorable effectiveness-equity balance. The results also indicate reduced socioeconomic stratification in educational access. By making education-housing feedback explicit, our framework supports long-term analysis of how school investment shapes educational opportunity over time.
[MA-4] Why Study Emergent Behavior When You Can Regulate It? Aligning Multi-Agent Systems with Reward Prediction
【速读】：该论文旨在解决多智能体系统中复杂社会与生态系统内涌现行为难以被有效调控的问题，传统方法多聚焦于分析涌现动态，而缺乏对这些行为进行主动引导与优化的机制。其核心挑战在于如何在不依赖人工设计奖励函数的前提下，使分散决策的智能体能够协同实现多样化的全局社会目标。解决方案的关键是提出一种基于学习的多智能体奖励预测框架——多智能体奖励预测（Multi-Agent Reward Prediction, MARP），该框架通过从群体层面的综合结果评估中学习共享的奖励模型，将高阶社会目标（如可持续性、公平性、和平等）转化为可驱动智能体行为的内在激励信号。MARP无需显式编程即可捕捉环境中的细微结构，并支持在同一训练框架下灵活切换多种复合型社会目标，仅通过调整高层评价指标即可实现对智能体行为的精准调控。实验证明，相较于标准奖励基线，MARP能更有效地生成符合预设社会度量的行为模式，从而证明了涌现行为不仅可被研究，更可通过数据驱动的方式进行有原则的调控。
链接: https://arxiv.org/abs/2608.07280
作者: Assaf Caftory,Almog Zemach,Moshe Butman,Doron Friedman
机构: Reichman University (里奇曼大学); The College of Management Academic Studies (管理学院学术研究学院)
类目: Multiagent Systems (cs.MA)
备注:
Abstract:Multi-agent simulations are widely used to study complex social and ecological systems, where rich and often unexpected emergent behaviors arise from local interactions. A large body of prior work has focused on analyzing such emergent dynamics across domains. In this paper, we move beyond analyzing emergent behavior and introduce a learning-based mechanism for actively shaping it via social reward modeling. We introduce Multi-Agent Reward Prediction (MARP), a simple framework that extends preference-based reward modeling to multi-agent reinforcement learning. While the framework is designed to be applicable across multi-agent settings, the present empirical validation is limited to a single environment, and we therefore present MARP as a proof of concept within the studied domain. Rather than relying on handcrafted rewards, MARP learns a shared reward model from episode-level evaluations of collective outcomes, enabling decentralized agents to align their behavior with global social objectives. We study MARP in the Harvest Game, a canonical sequential social dilemma modeling common-pool resource management and related real-world challenges. Our results show that MARP can be tuned to produce behavior that is more closely aligned with target social metrics than standard reward-based baselines, while the learned reward model captures subtle environmental structure without explicit programming. Crucially, MARP supports multiple and composite social objectives within a single training regime. By modifying only the high-level evaluation metric, the same framework seamlessly aligns agent behavior with diverse goals, including sustainability, equality, and peace, as well as combinations of individual and group-level objectives. These findings demonstrate that emergent multi-agent behavior can be treated not only as a phenomenon to study, but as a target of principled, data-driven regulation. Subjects: Multiagent Systems (cs.MA) Cite as: arXiv:2608.07280 [cs.MA] (or arXiv:2608.07280v1 [cs.MA] for this version) https://doi.org/10.48550/arXiv.2608.07280 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[MA-5] Multi-Agent Forensic Reasoning for Generalizable Deepfake Video Detection
【速读】：该论文旨在解决生成式人工智能（Generative AI）被恶意用于制作高度逼真的深度伪造视频所引发的伦理问题与AI安全挑战，尤其针对现有深度伪造视频基准数据集在覆盖近期合成方法方面不足、缺乏可靠细粒度文本标注的问题。同时，传统检测模型与多模态大语言模型（MLLM）因依赖单一模型架构或分析视角，难以捕捉细微伪造痕迹，导致对新兴生成方法泛化能力有限。其解决方案的关键在于构建一个大规模、多样化的深度伪造视频基准数据集——FaceVid-Forensics-100K，包含10万条视频，涵盖33种面部替换、面部重演及全脸合成方法，包括最新生成器Seedance 2.0；该数据集通过融合多模型聚合与冲突消解机制，利用先进MLLM自动生成细粒度视觉观察描述与一致性的法医解释。在此基础上，提出一种多智能体法医推理框架，由四个具备领域专长的专家智能体分别从纹理、光照、运动和物理规律四个维度独立分析伪造线索，再由裁判智能体整合报告以生成最终判定结果及可解释性说明。实验表明，尽管该框架仅使用小型开源MLLM，但在跨域测试集上仍显著优于包括闭源GPT与Gemini在内的多种方法，并在所有评测指标中排名第一。
链接: https://arxiv.org/abs/2608.06865
作者: Xuechao Zou,Shun Zhang,Kai Li,Yi Zhou,Xinyu Sun,Yuhui Chen,Zhe Wu,Congyan Lang,Junliang Xing
机构: 1. Institute of Automation, Chinese Academy of Sciences (中国科学院自动化研究所); 2. School of Computer Science and Technology, University of Science and Technology of China (中国科学技术大学计算机科学与技术学院); 3. Department of Computer Science and Engineering, Shanghai Jiao Tong University (上海交通大学计算机科学与工程系)
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注: 22 pages, 8 figures, 14 tables
Abstract:The malicious use of generative artificial intelligence to create highly realistic deepfake videos raises serious ethical concerns and poses substantial challenges to AI safety. However, existing deepfake video benchmarks provide limited coverage of recent synthesis methods and generally lack reliable fine-grained textual annotations. Meanwhile, conventional detectors and multimodal large language models (MLLMs), whether operating as a single model or relying on a single analytical perspective, often fail to capture subtle forgery artifacts, limiting their generalization to emerging AI-generated methods. To address these limitations, we introduce FaceVid-Forensics-100K, a large-scale deepfake video dataset comprising 100,000 videos and spanning 33 synthesis methods across face swapping, face reenactment, and entire-face synthesis, including recent generators such as Seedance 2.0. The dataset provides fine-grained textual annotations of visual observations and verdict-consistent forensic explanations, automatically synthesized through a multi-model aggregation and conflict-resolution pipeline powered by advanced MLLMs. Building on this benchmark, we propose a multi-agent forensic reasoning framework that employs four specialized domain-expert agents to independently analyze forgery cues from four perspectives: texture, lighting, motion, and physics. A judge agent then reconciles their reports to produce a final prediction together with an explanation. Extensive evaluations on out-of-domain test sets show that, despite being composed entirely of small open-source MLLMs, our framework outperforms all methods including closed-source GPT and Gemini models and ranks first across all reported metrics on this benchmark. The project page is available at this https URL.
[MA-6] Scalable Long-Horizon Planning with Staggered Updates for Lifelong MAPF
【速读】：该论文旨在解决大规模多智能体路径规划（Lifelong Multi-Agent Path Finding, LMAPF）中长期规划与实时性之间的矛盾问题，即在严格实时约束下实现对数千个智能体的无碰撞路径规划，同时具备长时程推理能力。现有方法如PIBT和EPIBT虽能高效扩展至大规模场景，但存在严重的时间短视（temporal myopia）问题；RHCR虽可进行多步窗口化规划，但计算开销过大，难以扩展；TP通过仅规划部分智能体缓解计算负担，但其适用性受限于高度结构化的地图。为克服上述局限，本文提出基于交错时域路径更新（Path Updates over Staggered Horizons, PUSH）的新型规划框架，其核心在于：结合了TP的分步子集规划机制与RHCR的多步窗口化规划能力，同时通过交错规划窗口设计降低计算复杂度，并在一般地图上实现无需特定地图假设的长时程规划。此外，PUSH引入受EPIBT启发的优先级继承、回溯机制及任意时间优化策略，显著提升高密度环境下的系统吞吐量。实验表明，PUSH在保持与EPIBT相当的大规模扩展能力（如10,000个智能体）的同时，相较所有基线方法实现了更高的系统效率，有效平衡了可扩展性、规划质量与实时性能。
链接: https://arxiv.org/abs/2608.06702
作者: Vaibhav Sanjay,Jiaoyang Li
机构: Carnegie Mellon University (卡内基梅隆大学)
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Robotics (cs.RO)
备注: 11 pages, 6 figures
Abstract:Lifelong Multi-Agent Path Finding (LMAPF) requires generating collision-free paths for large agent fleets under strict real-time constraints. Reactive frameworks such as PIBT and Enhanced PIBT (EPIBT) scale effortlessly to thousands of agents through rule-based, step-by-step coordination but suffer from severe temporal myopia, making them ineffective in scenarios where long-horizon reasoning is essential. RHCR plans windowed paths over multi-step horizons but incurs substantial planning overheads that hinder scalability. TP tackles both challenges by planning only subsets of agents at each timestep, yet its applicability is restricted to highly structured maps. To achieve long-horizon planning at scale across general maps, we propose Path Updates over Staggered Horizons (PUSH), a LMAPF planner capable of coordinating thousands of agents in under a second while planning over multi-step horizons. PUSH combines the key advantages of PIBT, RHCR, and TP. Like TP, PUSH reduces computational complexity by planning only a subset of agents at each timestep using staggered planning windows. Unlike TP, however, PUSH plans RHCR-style windowed paths in general maps without relying on restrictive map assumptions. To maintain high throughput in congested environments, PUSH further integrates EPIBT-inspired priority inheritance, backtracking, and anytime improvements into its windowed planning. Empirical evaluations across two realistic MAPF scenarios requiring long-horizon reasoning show that PUSH scales to the same massive agent loads as EPIBT (e.g., 10k agents) while achieving significantly higher system throughput than all baselines.
[MA-7] A Multi-Agent Framework for Automated Coarse-Grained Molecular Dynamics of Polymers
【速读】：该论文旨在解决粗粒度分子动力学（Coarse-grained Molecular Dynamics, CGMD）建模中参数化过程繁琐且缺乏可转移性的问题。传统方法需针对每种聚合物的特定映射重新构建拓扑结构、进行系统平衡、推导势函数，耗时费力且难以复用。其解决方案的关键在于提出CGMas——一个基于多智能体架构的自动化框架，通过自然语言描述聚合物类型与目标粗粒度分辨率，实现从原子模型到粗粒度模型的全流程自动化：利用大语言模型（Large Language Model, LLM）推理代理解析聚合物名称并推断原子级拓扑结构，通过分层自我修正机制有效处理不饱和键、杂原子及极性基团等常见物理错误；后续下游智能体完成系统平衡、粗粒度映射、基于玻尔兹曼反演（Boltzmann inversion）的势函数推导，并与原子模型进行基准验证。实验表明，CGMas成功完成27项均聚物与共聚物任务，在22种情况下粗粒度密度与原子模型偏差小于5%，并将模拟时间从38–88分钟大幅缩短至1分钟，证明了基于智能体的大语言模型在自动化聚合物粗粒度建模中的高效性与可行性。
链接: https://arxiv.org/abs/2608.06694
作者: Joohee Choi,Junhyeong Lee,Seunghwa Ryu
机构: Korea Advanced Institute of Science and Technology (KAIST); KAIST InnoCORE PRISM-AI Center, Korea Advanced Institute of Science and Technology (KAIST); Department of AX, Korea Advanced Institute of Science and Technology (KAIST)
类目: Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注:
Abstract:Coarse-grained (CG) molecular dynamics extends polymer simulation beyond the scales accessible to all-atom (AA) methods, but bottom-up CG modeling is laborious. The CG resolution is a design choice, so a transferable parameter set is generally not available and the potentials are derived anew for each polymer mapping. Here we present CGMas, a multi-agent framework that automates topology construction, equilibration, mapping, potential derivation, and validation from a natural-language specification of the polymer and target resolution. A large-language-model (LLM) reasoning agent infers the AA topology from polymer name, while layered self-correction resolves physical errors common to unsaturated, heteroatom-containing, and polar polymers. Downstream agents equilibrate the system, map it onto CG representation, derive potentials through Boltzmann inversion, and benchmark the model against its atomistic reference. CGMas completed all 27 homopolymer and copolymer tasks, matched the AA density to within 5% in 22, and reduced simulation from 38-88 min to 1 min, establishing agentic LLMs as a route to automated polymer coarse-graining.
[MA-8] ADIAS: Automated Design of Interactive Agent ic Systems
【速读】：该论文旨在解决现有自动化智能体设计方法中修复过程低效的问题，其核心在于传统候选中心（candidate-centric）范式导致的修复目标不明确、部分进展难以累积以及无效干预在多轮迭代中传播。为此，论文提出以问题为中心（issue-centric）的智能体优化框架，关键创新在于引入持久性问题状态（persistent issue state），显式维护问题的标识、生命周期状态、支持证据及干预-结果历史，从而将修复进展作为可追溯、可继承的元信息贯穿于多轮优化过程。在此基础上，通过问题引导的优化机制，联合确定后续的修复目标与修改方向，实现聚焦式的全代码重构。实验表明，所提出的ADIAS框架在五个交互式基准上平均优于最强基线25.2%，且在四种骨干模型上均表现一致提升；控制性消融实验进一步验证，移除持久性问题状态或采用候选中心策略会导致性能下降高达40.7%，凸显了问题中心范式在提升修复效率与稳定性方面的关键作用。
链接: https://arxiv.org/abs/2608.06410
作者: Lekang Jiang,Bohan Tang,Stephan Goetz,Yiwen Guo
机构: University of Cambridge (剑桥大学); LIGHTSPEED (光速); Tencent(腾讯)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Multiagent Systems (cs.MA)
备注: 23 pages, 7 tables, 5 figures
Abstract:Automated agent design improves agent harnesses through iterative revision, evaluation, and feedback summarization. Existing methods are largely candidate-centric: cross-round experience is organized around candidate agents, which leaves the repair progress implicit. This causes inefficient repair targeting, slow consolidation of partial progress, and propagation of ineffective interventions across rounds. Therefore, we formulate issue-centric agent optimization, in which repair progress is carried forward as an explicit persistent issue state to guide optimization, rather than re-derived from candidate history in each round. We instantiate the formulation in ADIAS, a framework for automated full-code agent design with two mechanisms. A persistent issue state maintains stable issue identities, lifecycle status, supporting evidence, and intervention-outcome histories. Issue-guided optimization uses this state to jointly propose repair targets and revision directions for subsequent focused full-code modification. Across five interactive benchmarks, ADIAS outperforms the strongest baseline by 25.2% on average and achieves consistent gains across four backbone models. Controlled ablations further show that removing persistent issue state or replacing issue-centric revision with candidate-centric policies leads to performance drops of up to 40.7%.
[MA-9] Evaluating XAI Support From A Hierarchical Reinforcement Learning Policy in Human-Agent Collaboration AAMAS2026
【速读】：该论文旨在解决当前可解释人工智能（XAI）在人机协作中普遍依赖人工设计策略且局限于特定环境，导致其在先进团队协作研究中泛化能力不足的问题。其核心解决方案是首次在成熟基准环境Overcooked-AI中，基于具有内在可解释性的学习型策略（intrinsically explainable learned policy），通过分层即时代理（HA²）架构生成实时解释，并采用一种创新的触发式系统以文本或语音形式传递解释。实验结果表明，尽管解释对任务绩效无显著影响，但语音模态的解释显著削弱了用户与智能体之间的工作联盟关系（working-alliance bond），而文本模态则无此效应，揭示了语音解释会激活用户对合作承诺的预期，而底层反应式策略无法持续满足此类预期。因此，该研究的关键在于提出并验证了“解释模态应与底层策略维持合作关系的能力相匹配”这一原则，为更有效的协作式XAI提供了新的设计方向，并建立了评估内在可解释强化学习架构的基准方法。
链接: https://arxiv.org/abs/2608.06381
作者: Mateus Levi Simões Fernandes,Alberto Sardinha
机构: PUC-Rio(天主教大学); Rio de Janeiro(里约热内卢), Brazil(巴西)
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注: 13 pages, 6 figures, accepted as an Extended Abstract at AAMAS 2026
Abstract:Explainable AI (XAI) has shown promise for human-agent collaboration, yet results rely on hand-crafted policies in custom environments, limiting generalizability to state-of-the-art teaming research. We provide the first systematic evaluation of XAI support generated from an intrinsically explainable learned policy in an established benchmark. Using the Hierarchical Ad Hoc Agents (HA ^2 ) architecture in Overcooked-AI, we generate real-time explanations from hierarchical subtask selections, delivered through text or audio via a novel trigger-based system. Our between-subjects experiment (n=38) found no significant performance effects, though participants with explanations showed trends toward faster performance improvement. More notably, audio explanations produced a significant reduction in participants’ working-alliance bond with the agent – an effect absent under the text modality – suggesting that spoken explanations activate partnership expectations the underlying reactive policy cannot meet. We provide the first modality comparison in real-time human-agent collaboration and establish a baseline methodology for evaluating intrinsically explainable reinforcement learning architectures in benchmark environments. Results point to matching explanation modality to the underlying policy’s capacity of sustaining the partnership its delivery implies as a potential path for more effective collaborative XAI.
[MA-10] he Theory of Strategic Evolution: Games with Endogenous Players and Strategic Replicators
【速读】：该论文旨在解决博弈论与自复制自动机理论长期分离的问题，试图实现冯·诺依曼在两个领域的奠基性工作之间的融合。其核心问题是：如何在资源受限的动态环境中建模具有战略优化能力并能自我复制的实体（即“战略型复制者”）的演化行为，尤其是在多层级、多代理系统中维持稳定性和可对齐性的机制。论文提出的关键解决方案是引入内生玩家博弈（Games with Endogenous Players, GEPs），将谱系（lineages）而非个体实例作为基本的战略单元，并定义智能分布进化稳定解（Evolutionarily Stable Distributions of Intelligence, ESDIs） 作为新的均衡概念。其核心数学结构是一组通过跨层级收益矩阵连接的战略层结构，当满足小增益条件（谱半径小于1）时，系统在任意有限深度下均存在全局李雅普诺夫函数，从而保证动力学稳定性。论文进一步证明了该框架在元选择（meta-selection）、创新或宪法演化下的封闭性，揭示了无限制自我修改会破坏系统结构的本质缺陷，由此导出对齐不可能定理——稳定对齐必须依赖于有界修改类。该框架为人工智能部署动态、市场集中化以及制度设计提供了统一分析工具，解释了为何在选择压力下人格工程必然失败，并指明了构建稳定多智能体系统的宪法约束必要性。
链接: https://arxiv.org/abs/2512.07901
作者: Kevin Vallier
机构: University of Toledo
类目: Computer Science and Game Theory (cs.GT); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Multiagent Systems (cs.MA); Theoretical Economics (econ.TH)
备注: Draft manuscript, 30k words. Companion to Agentic Capital. Submitted to establish priority
Abstract:Von Neumann founded both game theory and the theory of self-reproducing automata, but the two programs never merged. This paper provides the synthesis. The Theory of Strategic Evolution analyzes strategic replicators: entities that optimize under resource constraints and spawn copies of themselves. We introduce Games with Endogenous Players (GEPs), where lineages (not instances) are the fundamental strategic units, and define Evolutionarily Stable Distributions of Intelligence (ESDIs) as the resulting equilibrium concept. The central mathematical object is a hierarchy of strategic layers linked by cross-level gain matrices. Under a small-gain condition (spectral radius less than one), the system admits a global Lyapunov function at every finite depth. We prove closure under meta-selection: adding governance levels, innovation, or constitutional evolution preserves the dynamical structure. The Alignment Impossibility Theorem shows that unrestricted self-modification destroys this structure; stable alignment requires bounded modification classes. Applications include AI deployment dynamics, market concentration, and institutional design. The framework shows why personality engineering fails under selection pressure and identifies constitutional constraints necessary for stable multi-agent systems. Comments: Draft manuscript, 30k words. Companion to Agentic Capital. Submitted to establish priority Subjects: Computer Science and Game Theory (cs.GT); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Multiagent Systems (cs.MA); Theoretical Economics (econ.TH) MSC classes: 91A22, 91A80, 91B55, 37N25 Cite as: arXiv:2512.07901 [cs.GT] (or arXiv:2512.07901v3 [cs.GT] for this version) https://doi.org/10.48550/arXiv.2512.07901 Focus to learn more arXiv-issued DOI via DataCite
自然语言处理
[NLP-0] CreativeInstruct: Scalably Teaching LLM s to Balance Quality Creativity and Diversity
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在后训练（post-training）过程中普遍出现的生成多样性与创造力下降的问题，这一问题严重影响了需要显式或隐式创造力的任务表现，如故事生成和强化学习（Reinforcement Learning, RL）。其核心解决方案是提出一种可扩展的指令微调方法——CreativeInstruct，该方法通过学习在生成过程中注入特殊的[StartCreativity]标记序列，引导模型在保持后训练模型高质量输出的同时，有效平衡创造性与基线生成风格。关键创新在于引入基于图编辑距离（graph edit distance）的结构化多样性度量，能够捕捉传统词汇和语义指标所忽略的叙事层面的差异。实验结果表明，CreativeInstruct在叙事生成任务中达到甚至超过多模型基线及蒸馏输出的多样性水平，同时不损失生成质量且无需推理时使用多个模型。人类评估进一步验证了其优越性：70.3%的评分者认为CreativeInstruct生成的内容比后训练模型更具创造力。此外，在强化学习应用中，以CreativeInstruct模型为基底的GRPO算法在AMC和MATH基准上分别提升了约4%和5%，显著优于基于后训练模型的训练方案。
链接: https://arxiv.org/abs/2608.07460
作者: Ananya Sahu,Mohit Bansal,Elias Stengel-Eskin
机构: Columbia(哥伦比亚大学); UNC Chapel Hill(北卡罗来纳大学教堂山分校); University of Texas at Austin(德克萨斯大学奥斯汀分校)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Code: this https URL
Abstract:While post-training improves the capabilities of large language models (LLMs), it generally lowers their output diversity and creativity, negatively impacting tasks that explicitly require creativity (e.g., story generation) as well as those that require it implicitly, e.g., reinforcement learning (RL). We instead propose CreativeInstruct, a scalable instruction-tuning method that teaches LLMs to balance creative, base-model-like generations with the quality of post-trained models, by learning to inject special [StartCreativity] spans that bias generation toward creativity. Furthermore, we introduce a structural diversity metric based on graph edit distance, which captures narrative level variation missed by purely lexical and semantic metrics. On narrative generation, CreativeInstruct matches or exceeds the diversity of both multi-model baselines and distilled variants of their outputs, without sacrificing quality or requiring multiple models at inference time. These results are mirrored in our human evaluation, where we find that annotators rate CreativeInstruct generations as more creative than the post-trained LLMs’ generations in 70.3% of cases. We also show the benefits of creative models as a substrate for RL: GRPO applied to a CreativeInstruct checkpoint improves by ~4% on AMC and ~5% points on MATH over the same training applied to the post-trained checkpoint.
[NLP-1] SkillProx: Self-Evolving Agent Skills via Proximal Textual Gradient Descent
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）智能体在持续学习过程中，对已积累的程序性知识（skills）缺乏有效诊断与优化机制的问题。现有方法虽通过任务迭代执行与文本空间更新实现技能演化，但普遍缺乏显式的“诊断—结果反馈”闭环，且将技能删除视为通用编辑操作，未能将其作为知识整合的关键机制。为此，本文提出SkillProx，一种受近端梯度（proximal gradient）启发的前向-后向框架，其核心在于将闭环诊断演化与基于效用感知的近端精炼相耦合。前向阶段通过诊断驱动的编辑重执行同一任务批次，回滚性能退化，并将评估结果反馈至后续诊断；后向阶段则将生成的技能分解为可审计的知识单元，利用冻结的留一法效用审计（frozen leave-one-out utility audit）量化各单元贡献，并基于验证门控机制执行知识的巩固、降级或移除。实验表明，在多个骨干模型及分布内/外基准上，SkillProx相较最强的基于梯度的基线平均准确率提升3.0个百分点，组件消融实验进一步验证了闭环诊断与近端精炼之间的互补性。
链接: https://arxiv.org/abs/2608.07449
作者: Mingxuan Zheng,Yujin Zhou,Chuxue Cao,Boqin Yin,Yuyao Zhang,Jiapeng Sun,Shuaishuai Gong,Sirui Han,Yike Guo
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 23 pages, 4 figures
Abstract:LLM agents increasingly adapt to recurring tasks by accumulating procedural knowledge in skills. These skills are lightweight, reusable textual artifacts that are loaded into the agent’s context without weight updates. Recent methods refine skills through iterative task execution, failure diagnosis, and trajectory-guided text-space updates. However, existing frameworks lack explicit diagnosis–outcome feedback and treat deletion as a generic edit operation rather than a dedicated mechanism for consolidating accumulated knowledge. We introduce SkillProx, a proximal-gradient-inspired forward–backward framework that couples closed-loop diagnostic evolution with utility-aware proximal refinement. Motivated by a composite objective balancing task loss and skill complexity, the forward stage re-executes diagnosis-driven edits on the same task batch, rolls back regressions, and feeds measured outcomes into subsequent diagnoses. The backward stage decomposes the resulting skill into auditable knowledge units, estimates their contributions using a frozen leave-one-out utility audit, and applies validation-gated consolidation, demotion, or removal. Experiments on in-distribution and out-of-distribution benchmarks across multiple backbone LLMs show that SkillProx improves average accuracy by 3.0 percentage points over the strongest gradient-based baseline. Component ablations demonstrate the complementary effects of closed-loop diagnosis and proximal refinement.
[NLP-2] An Exploratory Evaluation of LLM -Assisted Rewriting of Moderate-Complexity Financial Sentences for DisCoCat-Based Sentiment Analysis
【速读】： 该论文旨在解决量子自然语言处理（Quantum Natural Language Processing, QNLP）中基于分布组合范畴（Distributional Compositional Categorical, DisCoCat）模型在金融情感分析任务中存在的实际瓶颈问题，主要包括语法解析器敏感性高、量子电路模拟成本昂贵以及对长句处理能力不足。其解决方案的关键在于引入大语言模型（Large Language Model, LLM）辅助的预处理流程，通过可控重写（controlled rewriting）技术将中等复杂度的金融情感句子压缩、简化或分解为与解析器兼容且量子电路效率更高的形式，同时保持语义情感信息的完整性。实验表明，最优配置下（GPT-4.1-mini 与 Prompt B），平均准确率可达 0.550 ± 0.035，较原始低复杂度基线（0.521 ± 0.050）有所提升；电路层面的最优压缩方案使平均量子比特数和门操作数降低超过 70%。研究进一步揭示训练集规模与性能之间存在中度负相关（Pearson r = -0.446），暗示单纯扩大数据量未必提升效果，凸显提示工程设计、过滤策略及电路感知预处理在构建可扩展的量子化金融情感分析系统中的关键作用。
链接: https://arxiv.org/abs/2608.07439
作者: Brian Llinas,Nikos Chrisochoides
机构: 未知
类目: Computation and Language (cs.CL); Quantum Physics (quant-ph)
备注:
Abstract:Quantum natural language processing (QNLP) provides a grammar-aware framework for text modeling, and Distributional Compositional Categorical (DisCoCat) is one of its theoretically grounded formulations. Prior work on financial sentiment analysis has identified practical limitations of DisCoCat, including parser sensitivity, high simulation cost, and difficulty handling longer sentences. We study an LLM-assisted preprocessing workflow that uses controlled rewriting to compress, simplify, or decompose moderate-complexity financial sentiment sentences into parser-compatible, circuit-efficient variants while preserving sentiment-bearing meaning. We compare prompting strategies, language models, and filtering configurations with the low-complexity-only DisCoCat baseline of Stein et al. At the circuit level, the strongest compression variants reduce average qubit and gate counts by more than 70 percent relative to the raw moderate-complexity subset. Across repeated training runs, GPT-4.1-mini with Prompt B achieves the highest observed mean accuracy, 0.550 \pm 0.035 , compared with 0.521 \pm 0.050 for the baseline. Larger training splits do not necessarily improve downstream performance; across evaluated configurations, training-split size has a moderately negative association with accuracy (Pearson r=-0.446 ). These results provide exploratory evidence that LLM-assisted rewriting can make some moderate-complexity inputs usable within the evaluated DisCoCat configuration, while highlighting prompt design, filtering, and circuit-aware preprocessing as considerations for more scalable QNLP-based financial sentiment analysis.
[NLP-3] SABRE: Scalable and Automated Benchmarking of VLMs under Stress
【速读】： 该论文旨在解决视觉-语言模型（Vision-Language Models, VLMs）评估基准发展滞后的问题，尤其针对当前模型在面对复杂或异常场景时依赖先验知识而非真实视觉证据的潜在缺陷难以被有效识别。其核心挑战在于构建高质量、可控且具有挑战性的压力测试样本，而传统方法成本高、效率低。为此，论文提出SABRE——一个可扩展、自动化的流水线框架，能够将任务设计模板（Test Primer）转化为结构化规范、生成或编辑后的图像以及对应的问答对。关键创新在于：通过自动化过滤机制剔除已被基础视觉-语言模型轻易解答的样本，并结合人工审核确保样本的有效性与标注准确性，支持局部图像修复与注释修正。研究以SABRE-Prior为例，验证模型是否真正依据图像中的视觉证据进行推理，而非依赖于关于常见物体和场景的世界先验知识；该测试涵盖情境（Context）、纹理（Texture）、属性（Attribute）和语言诱导（Language Elicitation）四类挑战性维度。实验结果表明，六种主流VLM在该测试上的平均准确率仅为22.6%（范围17.8%–31.3%），证明其严重依赖先验而非视觉证据。此外，SABRE-Counting与SABRE-Spatial的初步应用也验证了该流程可推广至其他压力测试场景。综上，SABRE的核心价值在于提供一个动态、可复用的框架，用于持续构建和更新视觉-语言模型的压力测试体系，推动模型评估从静态基准向可迭代演进的系统化方向发展。
链接: https://arxiv.org/abs/2608.07435
作者: Zixuan Lan,Luzhe Sun,Matthew R. Walter,Jiawei Zhou
机构: University of Chicago; Toyota Technological Institute at Chicago; Stony Brook University
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 22 pages, 10 figures. Code and resources will be available at this https URL
Abstract:Vision-language models (VLMs) are improving rapidly, but benchmark development lags behind, making weaknesses hard to identify. Building stress tests is costly: samples must satisfy controlled conditions, remain answerable, and challenge current models. We present SABRE, a scalable, automated pipeline that converts a Test Primer (a Markdown Task Design with Data Schema) into structured specifications, generated or edited images, and question-answer pairs. Automated filtering removes candidates solved by a Filtering VLM, while human review verifies candidate validity and supports annotation correction and localized image repair. We instantiate SABRE-Prior to test whether VLMs follow visual evidence instead of relying on world priors – learned expectations about familiar objects and scenes. Its 600 images and 1,000 questions span Context (unexpected entities in familiar scenes), Texture (counterfactual materials), Attribute (noncanonical component counts), and Language Elicitation (answers suggested by language but unsupported by the image). Across six VLMs, macro-average accuracy ranges from 17.8% to 31.3% (22.6% mean). A real-image Attribute control is comparably difficult for the Filtering VLM. SABRE-Counting and SABRE-Spatial pilots show that the workflow supports other stress-test settings. These results establish SABRE as a reusable framework for constructing and refreshing VLM stress tests rather than a single fixed benchmark.
[NLP-4] ResidencyRL: Reinforcement Learning in Simulated Clinical Environments
【速读】： 该论文旨在解决临床决策中复杂、动态且多轮交互式推理的建模难题，尤其针对生成式人工智能（Generative AI）在真实临床场景下难以有效模拟医生从问诊到诊断与管理全过程的挑战。现有大语言模型（LLM）虽在静态医学基准测试中表现优异，但在模拟具有不确定性、反馈多样性和逐步自主性的长期临床互动方面仍存在显著不足。其解决方案的关键在于提出一种名为ResidencyRL的强化学习（Reinforcement Learning, RL）框架，通过构建可支持长达60轮对话和8次工具调用的多轮临床模拟环境，使AI代理在对抗性情境下进行训练。该方法将策略代理与具备复杂行为能力的LLM模拟器配对，并采用结构化奖励机制，综合考量诊断准确性、管理质量、沟通能力、文档规范性及安全性等维度。实验结果表明，经训练的代理在对抗性条件下诊断准确率提升7.0%（88.0% vs. 81.0%），误漏关键警示信号率降低31%，显著缓解了过早闭合（premature closure）问题；盲法专家评审显示，87.6%的对比案例中更倾向选择训练后的代理。此外，该方法在未见基准（如AMIE、AgentClinic、CRAFT-MD）上均展现出跨任务泛化能力，证明了通过模拟多轮强化学习可有效习得稳健、可迁移的临床决策能力，为实现人工智能在临床实践中的“临床精通”提供了可行路径。未来仍需在真实临床工作流中开展前瞻性验证以确立其实际应用价值。
链接: https://arxiv.org/abs/2608.07418
作者: Valentin Liévin,Samuel Schmidgall,Tim Strother,Alex Bijamov,Akshay Goel,Anil Palepu,Chunjong Park,Vahid Balazadeh,Min Woo Sun,Marius Guerard,Justin Chen,Dave Steiner,Vikram Dhillon,Ibrahim Azar,Akhil Mehta,Nicholas Spetsieris,Shilpan Shah,Maen Abdelrahim,Amit Dahiya,Yun Liu,Katherine Chou,Yossi Matias,Avinatan Hassidim,Dale R. Webster,Quoc V. Le,Raia Hadsell,Joelle Barral,Carey Radebaugh,Aleksandra Faust,Shekoofeh Azizi,Mike Schaekermann,Po-Hsuan Cameron Chen,Tao Tu,David Racz,Lin Yang
机构: Google DeepMind(谷歌深度思维); Google Research(谷歌研究)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:In medical education, physicians convert academic knowledge into clinical expertise through residency: years of training across thousands of encounters, with diverse sources of feedback and progressively greater autonomy. Much of clinical reasoning relies on the patient encounter, a dialogue in which a clinician elicits history, refines diagnostic hypotheses, and decides management under uncertainty. While large language models (LLMs) excel on static medical benchmarks, methods to optimize the full sequence of clinical decisions remain underdeveloped. We present ResidencyRL, a reinforcement learning (RL) method for training clinical artificial intelligence (AI) agents through simulated multi-turn clinical encounters (up to 60 dialogue turns and 8 tool calls per trajectory). ResidencyRL pairs the policy agent with LLM simulators capable of complex, adversarial behaviors, training against a structured reward aligned to diagnostic accuracy, management quality, communication, documentation, and safety. On held-out evaluations, the ResidencyRL agent improves diagnostic accuracy by 7.0% under adversarial conditions (88.0% vs. 81.0%) and reduces missed red flag rates by 31%, demonstrating rigorous mitigation of premature closure. Blinded expert clinicians validated these gains, preferring the trained agent in 87.6% of side-by-side comparisons. The procedural competencies transfer to unseen benchmarks: the agent outperforms the base model across all six clinical axes of the AMIE multi-visit benchmark, and shows consistent directional improvements on AgentClinic and CRAFT-MD. Our findings demonstrate that sequential clinical decision-making can be effectively learned through multi-turn RL in simulation, yielding robust, generalizable capabilities, paving the way towards clinical mastery. Prospective validation with real-world workflows remains necessary to establish clinical utility.
[NLP-5] rajectory-Relative Hindsight Distillation for Agent ic Reinforcement Learning
【速读】： 该论文旨在解决生成式智能体在稀疏奖励环境下，基于事后推理（hindsight）机制所产生的大量潜在奖励信号在决策回合间难以合理分配的问题。现有方法虽能通过回顾性信息增强学习信号，但缺乏对这些信号在不同决策时刻的精确、动态分配机制，导致监督信号分布不均或效率不足。其解决方案的关键在于提出TRIAL框架——一种轨迹相对的事后推理蒸馏方法，采用统一的回合对齐评分协议，针对每个决策回合，提取该步实际结果的“结果视图”，并评估同一响应在常规上下文与事后推理条件下的差异。通过计算符号化的对数概率差来确定逐标记级别的监督方向与局部强度，并对整个真实轨迹上的回合级幅度进行联合归一化，确保最终分配乘子的加权平均值为1，从而在保持总监督密度不变的前提下，实现对监督信号的高效再分配。实验表明，TRIAL在WebShop和ALFWorld等多个环境与多种基线模型组合中均显著优于GRPO等方法，尤其在Qwen3-1.7B模型上将任务成功率从56.4%提升至75.2%，任务得分从78.7%提升至85.7%，且消融实验验证了轨迹相对的回合级分配策略相较于单纯密集事后蒸馏具有显著优势。
链接: https://arxiv.org/abs/2608.07371
作者: Haoyu Zheng,Yun Zhu,Qing Wang,Wenqiao Zhang
机构: Zhejiang University (浙江大学); Shanghai AI Laboratory; Tencent (腾讯)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:
Abstract:Recent agentic reinforcement learning methods use hindsight to complement sparse outcome rewards. However, a completed rollout can yield many such signals, leaving their appropriate allocation across turns unclear. We introduce TRIAL, a trajectory-relative hindsight distillation framework with a unified turn-aligned scoring protocol. For each decision turn, TRIAL extracts an outcome view of that decision’s realized consequence and evaluates the same response under ordinary and hindsight-conditioned contexts. The signed log-probability gap determines the direction and local strength of token-level supervision, while turn-level magnitudes are normalized jointly over the realized trajectory. The resulting allocation multipliers have an eligible-token-weighted mean of one, redistributing dense supervision across turns while fixing its average multiplier. Experiments on WebShop and ALFWorld with different backbones show that TRIAL outperforms GRPO across all eight combinations of backbone, environment, and evaluation metric, while achieving the best or tied-best performance among six methods on six of them. On WebShop with Qwen3-1.7B, TRIAL improves the success rate from 56.4% to 75.2% and the task score from 78.7% to 85.7%. Controlled ablations further show that trajectory-relative turn allocation provides substantial gains beyond those of dense hindsight distillation alone.
[NLP-6] LitTraceQA: A Benchmark for Multi-Stage Grounding and Verification in Scientific Question Answering
【速读】： 该论文旨在解决科学文献问答系统中“答案可验证性”不足的问题，即现有系统往往生成看似合理但缺乏具体证据支持的回应。其核心挑战在于：从海量科学论文中准确识别相关文献、精确定位支持答案的具体证据（如表格、图表、公式、文本片段或引用上下文），并生成忠实于原始证据的可靠回答。解决方案的关键在于提出LitTraceQA基准数据集，该数据集要求系统在给定研究问题与论文元数据池的情况下，输出三类相互关联的输出：规范化的论文标识符、支持性证据的位置标注以及符合多种格式（自由文本、多选题、结构化表格）的答案。通过区分评估论文检索、证据定位和答案准确性三个环节，LitTraceQA为构建可追溯、可验证的科学问答系统提供了标准化测试平台，推动生成式 AI (Generative AI) 在科研场景中从“摘要生成”向“证据驱动推理”演进。
链接: https://arxiv.org/abs/2608.07370
作者: Xuye Liu,Yimu Wang,Peng Shi,Bo Xue,Xiangrui Ke,Songcheng Cai,Kath Choi,Di Wu,Freda Shi,Krzysztof Czarnecki
机构: University of Waterloo ( Waterloo 大学); Amazon (亚马逊); City University of Hong Kong (香港城市大学); University of Amsterdam (阿姆斯特丹大学)
类目: Computation and Language (cs.CL)
备注: Work in Progress
Abstract:Scientific literature is increasingly used as a knowledge source for language models, retrieval-augmented generation systems, and research assistants, but answering research questions from papers requires more than fluent generation. A reliable system must identify the relevant papers, locate the concrete evidence that supports the answer, and produce a response that is faithful to that evidence. We present LitTraceQA, a benchmark for literature-grounded question answering over scientific papers. Given a research question and a metadata pool of papers, a system must return three connected outputs: canonical paper identifiers, supporting evidence locations, and answers in one or more requested formats, including free-form text, multiple-choice answers, and structured tables. LitTraceQA targets evidence types common in scientific reading: tables, figures, text spans, equations or algorithms, and citation contexts. The public development split contains 55 examples, including 26 hidden-source single-paper questions and 29 multi-paper questions, and provides gold papers, evidence annotations, and answers for local validation. We also analyze a larger final annotation collection with 4,978 unique-question records over 4,859 unique gold papers. By evaluating paper retrieval, evidence grounding, and answer accuracy separately, LitTraceQA provides a testbed for scientific QA systems that produce verifiable answers rather than unsupported summaries.
[NLP-7] Zero Gap Is Not Restoration: Stratified Per-Question Probability Evaluation and Step-wise Mitigation of Benchmark Contamination
【速读】： 该论文旨在解决预训练数据中存在公开基准测试数据泄露（contamination）导致模型评估分数虚高，且现有缓解策略在评估其有效性时所依赖的主流指标——聚合性能差距（G-AP）存在固有缺陷的问题。核心问题在于：G-AP采用离散的正确/错误判断作为读出信号，无法刻画每个问题的细粒度解题概率；在差异计算前进行平均处理，导致过抑制与欠抑制相互抵消；同时对所有问题赋予相同权重，诱发攻击性策略将解题概率推向干净模型高频值以伪造性能恢复效果。为此，作者提出新型评估指标SA-PPG（分层聚合每题概率差距），通过采样估计各题解题概率，与干净模型对应结果做差，并按干净模型解题概率分组进行聚合，从而更准确反映真实性能恢复程度。此外，论文指出现有缓解方法依赖对污染位置的估计，其有效性受限于估计精度。本文提出的RailCap则在生成过程中直接判断污染：当采样轨迹回退至贪婪路径时，下一词元被限制为次优选项，持续累积抑制直至响应分布充分分散，实现动态、自适应的污染抑制。实验表明，SA-PPG揭示了先前方法对性能恢复的严重高估，而RailCap在多模型与多基准上均取得最低的SA-PPG值，展现出更优的抗污染能力。
链接: https://arxiv.org/abs/2608.07341
作者: Ruijie Hou,Yueyang Jiao,Zhao Wang,Yingming Li
机构: Zhejiang University (浙江大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Test data from public benchmarks inevitably leaks into pretraining corpora, inflating evaluation scores once memorized. \textbfContamination mitigation evaluation intervenes in the decoding process to suppress memorization and restore a contaminated model’s genuine capability, but its prevailing metric, the \textbfG-AP (\textbfGap of \textbfAggregate \textbfPerformance), is flawed. Discrete correct/incorrect readouts cannot characterize per-question performance, averaging before differencing lets over- and under-suppression cancel out, and uniform per-question weighting invites strategies to push solve probabilities onto the clean model’s high-frequency values. We propose \textbfSA-PPG (\textbfStratified \textbfAggregate of \textbfPer-question \textbfProbability \textbfGaps): estimate each question’s solve probability by sampling, difference it against the clean model per question, and aggregate within groups defined by the clean model’s solve probability. Existing mitigation strategies first estimate where contamination lies and then operate on the estimate, so they are only as correct as the estimate. \textbfRailCap instead judges contamination during generation: whenever a sample falls back onto the greedy trajectory, the next trajectory token is capped to the runner-up, accumulating suppression until the response distribution becomes sufficiently dispersed. Across multiple contaminated models and benchmarks, SA-PPG reveals that prior strategies’ restoration is substantially overestimated, while RailCap attains the lowest SA-PPG.
[NLP-8] Natural Language Processing Psychometrics
【速读】： 该论文旨在解决自然语言处理（Natural Language Processing, NLP）模型在预测心理健康结果时缺乏对所测量内容的明确界定问题，即模型实际捕捉的是语境知识、情感内容还是句法结构尚不清晰。其核心解决方案在于提出“NLP心理计量学”（NLP Psychometrics）框架，将文本中的心理预测视为可量化的心理测量问题，通过构建受控人格角色（认知数字分身，cognitive digital shadows）的大型语言模型（LLM）生成带解释性文本的问卷响应，进而利用文本形式思维网络（textual forma mentis networks）提取情感特征与句法-语义结构，并结合人格特质和人口统计学变量，在消融实验的随机森林（Random Forest, RF）回归模型中采用SHAP值分析各特征对预测性能的贡献方向与程度。结果显示，全模型可解释高达70.8%的生活满意度（SWLS）方差、55.7%的抑郁（PHQ-9）方差，以及在DASS-21量表中分别达到68.5%（抑郁）、76.0%（焦虑）、72.4%（压力）的方差。研究发现，人口统计学变量仅对生活满意度具有显著预测力，而情绪特征与收入是主要驱动因素；神经质特质与网络拓扑结构则主导抑郁与焦虑预测，且二者影响方向相反。无需重新训练，仅基于网络与情绪特征即可有效区分低分与高分人格角色的日记文本（相关系数r达0.91），并在真实临床访谈文本中实现最高68%的分类准确率。研究表明，合成数据虽能揭示模型偏差、复现临床反刍模式并支持无匹配问卷的心理测量预测，但无法替代人类验证。NLP心理计量学通过可解释人工智能与网络/情感特征，使这些差异变得明确、可观测且可检验。
链接: https://arxiv.org/abs/2608.07316
作者: Edoardo Sebastiano De Duro,Emma Franchino,Massimo Stella
机构: University of Trento (特伦托大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Social and Information Networks (cs.SI)
备注:
Abstract:Natural Language Processing (NLP) models predicting mental health outcomes rarely specify what they measure: contextual knowledge, emotional content, or syntactic structure. NLP Psychometrics treats psychological prediction from text as a psychometric problem, linking scores to interpretable linguistic evidence and testing beyond the training text format. Nine LLMs, conditioned on controlled personas (cognitive digital shadows), completed psychometric questionnaires with textual explanations per item. We extracted emotional profiles and syntactic-semantic structure via textual forma mentis networks, combined with personality and sociodemographic variables in ablated random forest (RF) regressors, using SHAP to identify which features drove performance and in which direction. Full RF models explained up to 70.8% of variance in life satisfaction (SWLS), 55.7% in depression (PHQ-9), and, for DASS-21, 68.5% depression, 76.0% anxiety, 72.4% stress. Sociodemographics alone explained no meaningful variance in depression, anxiety, or stress, but did so for life satisfaction, where emotion features and income were the strongest predictors; neuroticism and network topology instead dominated depression and anxiety, reversing direction between them. Without retraining, RF models separated diaries from low- and high-score personas ( r up to 0.91) and, using only network/emotion features, classified clinical from control participants in real transcripts with up to 68% accuracy. These results show the promise and limits of synthetic data: LLM personas can expose model biases, recover patterns consistent with clinical rumination, and support psychometric prediction from human text without a matched questionnaire, but cannot substitute for human validation. NLP Psychometrics makes these distinctions explicit, measurable, and testable through interpretable AI and network/emotional features.
[NLP-9] Grammar Engineering Meets LLM s: Development of Cantonese and Irish ParGram Treebanks
【速读】： 该论文旨在解决跨语言语法工程中兼顾语言共性与语言特异性难题，特别是在并行语法（ParGram）项目背景下实现语法形式的抽象功能对齐。其解决方案的关键在于利用多语言大语言模型（LLM）辅助语法构建，探索其在粤语-爱尔兰语翻译及形式句法结构生成中的应用潜力。研究发现，尽管模型在翻译任务中表现不佳且受提示语言影响有限，但在生成部分结构合理句法结构方面具备一定能力，尤其在捕捉谓词-论元关系上展现出部分参考价值；然而，在需要跨语言抽象的任务中表现欠佳。因此，研究强调了大语言模型在协作式语法工程中的辅助作用，同时重申了专家主导分析与验证的不可替代性。
链接: https://arxiv.org/abs/2608.07283
作者: Chit-Fung Lam,Elaine Uí Dhonnchadha
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Grammar engineering requires expertise in linguistic formalism and computational implementation, especially in parallel grammar projects that balance cross-linguistic consistency with language-specific properties. This paper presents the development of Cantonese and Irish treebanks within the Parallel Grammar (ParGram) Project, where linguistic parallelism is maintained at an abstract functional level. We also investigate the methodological potential and limitations of using multilingual LLMs to support grammar engineering, focusing on Cantonese-Irish translation and the generation of formal syntactic structures using OpenAI’s gpt-oss-120b model. The results show that translation performance was generally unsatisfactory and unaffected by prompt language. For syntactic structure generation, the model produced some structurally meaningful outputs, but performed poorly on tasks requiring cross-linguistic abstraction. Nonetheless, LLM-generated outputs may still offer some reference value by suggesting alternative analyses and (partially) capturing predicate-argument relations. Overall, our findings highlight both the potential and limitations of using LLMs in collaborative grammar engineering, while underscoring the continued importance of expert-driven analysis and verification.
[NLP-10] Gaze Behavior in Visual World Experiments Can be Modeled With Off-the-shelf Language-Vision Encoders
【速读】： 该论文旨在解决当前计算心理语言学研究中对多模态语言处理建模的忽视问题，尤其是针对视觉世界范式（visual world paradigm）中人类在同时接收视觉与语言输入时的眼动行为预测难题。现有研究主要聚焦于单模态（如书面或口语）语言模型，而忽略了多模态情境下人类语言理解的动态交互机制。其解决方案的关键在于提出一种新颖的多模态方法：通过结合简洁的双编码器（bi-encoder）结构（源自CLIP家族）与双模态归因（bimodal attribution）技术，实现对视觉世界实验中被试眼动轨迹的稳健预测。该方法无需生成式架构，也无需针对特定任务进行微调，即可成功复现经典英语视觉世界研究中体现的人类前瞻加工（predictive processing）现象，表明其具备强大的泛化能力与对人类语言处理机制的建模潜力。
链接: https://arxiv.org/abs/2608.07282
作者: Rahul Murali Shankar,Titus von der Malsburg,Sebastian Padó
机构: University of Stuttgart, Germany
类目: Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:The recent advances in neural language models have also spurred much work in computational psycholinguistics, asking whether neural LMs are also promising models of human language processing. However, work has been overwhelmingly focused on the unimodal case of written or spoken language. In contrast, multimodal experimental paradigms, like visual world studies that present participants with both visual and linguistic input simultaneously, have been neglected. In this paper, we present a novel approach that predicts gaze behavior in visual world studies. It does so by combining a simple multi-modal bi-encoder model of the CLIP family with a bimodal attribution method. We demonstrate the ability of this approach to robustly replicate the results of a seminal English visual world study which shows hu- man predictive processing. Remarkably, it does so without a generative architecture and without the need for fine-tuning, despite not being trained for this task.
[NLP-11] Why Knowing Both Hops Is Not Enough: Understanding Two-Hop Generalization in Language Models
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在处理简单两跳（two-hop）推理任务时出现的“看似合理却最终失败”的现象：尽管模型能够正确理解并存储每一跳的独立信息，但在将两跳信息整合以得出最终结论时却频繁失效。其核心问题是模型在跨跳推理中缺乏对中间表示（intermediate representations）的一致性建模能力，导致在面对分布外（out-of-distribution）的第二跳查询时无法有效泛化。该研究的关键解决方案在于通过在受控的符号环境中从零训练变压器（transformer），揭示了模型在两跳泛化中的行为模式，并基于机制分析提出了一种递归式训练策略（recurrent-style training strategy）。该策略促使模型在不同输入形式间复用其推理通路，从而在低层准确构建一致的中间表示，同时确保高层能够对这些表示进行有效的推理而非仅映射到输出，显著提升了模型在分布外两跳查询上的泛化性能。
链接: https://arxiv.org/abs/2608.07261
作者: Zili Zhang,Yilin Wang,Heng Wang,Herun Wan,Minnan Luo
机构: Xi’an Jiaotong University (西安交通大学); University of Illinois Urbana-Champaign (伊利诺伊大学厄本那-香槟分校)
类目: Computation and Language (cs.CL)
备注: 24 pages, 20 figures
Abstract:Large language models (LLMs) can solve complex multi-hop problems yet exhibit puzzling failures on simple two-hop queries: although a model may correctly store each individual hop, it often fails to combine them. To understand the internal mechanisms of this phenomenon, we train transformers from scratch in a controlled symbolic environment. Our experiments reveal a pattern in two-hop generalization: models generalize reliably when the second hop follows the training distribution, but always fail when it deviates. Through mechanistic analysis, we provide a complete explanation for these distinct generalization behaviors: in settings where models generalize successfully, performance is driven by the emergence of consistent intermediate representations for the same entities across contexts, whereas failures on settings where the second hop is out-of-distribution arise from a mismatch across layers: lower layers correctly construct these intermediate representations, but upper layers, while trained on corresponding atomic facts, primarily learn to map them to outputs rather than to reason over them. Driven by this insight, we propose a recurrent-style training strategy, which enables transformers to reuse their reasoning circuitry across input forms and substantially improves generalization on out-of-distribution two-hop queries. Comments: 24 pages, 20 figures Subjects: Computation and Language (cs.CL) Cite as: arXiv:2608.07261 [cs.CL] (or arXiv:2608.07261v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2608.07261 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[NLP-12] Stoicheia: Character-Level Masked Diffusion for Ancient Greek Textual Restoration Parsing and Metrical Scansion
【速读】： 该论文旨在解决古希腊语文本中因残损、无标点、无重音及分词不明确等问题导致的自然语言处理挑战，特别是针对古籍文献（如铭文、纸草文献）的修复与理解难题。其核心问题是：如何在不依赖任务特定分词或微调的前提下，实现对未分段、无重音、缺失字符的古希腊语文本进行统一的恢复、重分段、重音标注与标点还原。解决方案的关键在于提出Stoicheia——一个405M参数的字符级掩蔽扩散编码器（character-level masked-diffusion encoder），其输入被分解为五个对齐且可独立掩蔽的语义平面：字母、词与句子边界、变音符号、大写形式和标点符号。这一设计使单一模型能够通过统一的生成式框架完成多任务恢复，无需任务特定的重新分词。模型在包含3.8亿词的开放、版本锁定语料库上预训练，并发布11个检查点（包括10个去污染折叠版本与1个完全未接触文献文本的版本），确保评估时的公平性与泛化能力。实验表明，该方法在铭文重建、形态句法标注与依存分析、以及韵律重音标注（macronization）三项任务中分别带来5.6个字符错误率（CER）、12.9个线路准确率（LAS）与6.0个百分点的平衡准确率提升；在Ithaca测试集上，字符错误率从24.6降至15.5，顶1准确率从64.0提升至74.5，显著优于此前最先进系统。
链接: https://arxiv.org/abs/2608.07249
作者: Eric Cullhed,Albin Thörn Cleland
机构: Uppsala University (乌普萨拉大学); Lund University (隆德大学)
类目: Computation and Language (cs.CL)
备注: 12 pages, 7 tables. Models, datasets and code released: this https URL and this https URL
Abstract:We introduce Stoicheia, a 405M-parameter character-level masked-diffusion encoder for Ancient Greek whose input factors into five aligned, independently maskable planes: letters, word and sentence boundaries, diacritics, capitalization, and punctuation. A single backbone can therefore restore lacunae, re-segment, accentuate, and punctuate unspaced text without task-specific retokenization. We pretrain it on an open, revision-pinned corpus of 380M words and release eleven checkpoints: ten rotated, decontaminated folds, guaranteeing that for any given literary passage at least one released model has never seen its text, and one with no exposure to documentary texts. Three experiments - reconstruction of damaged inscriptions and papyri, morphosyntactic tagging and dependency parsing, and macronization with metrical scansion - each carry a matched random-initialization control, isolating what character-level diffusion pretraining contributes: 5.6 CER points on inscription reconstruction, 12.9 LAS on parsing, and 6.0 points of balanced accuracy on macronization. On Ithaca’s own test split, with identical frozen samples and strict scoring, Stoicheia reduces character error relative to both prior state-of-the-art systems, from 24.6 (Ithaca) and 23.5 (its 2025 Aeneas-framework successor) to 15.5, and raises top-1 accuracy from 63.0 and 64.0 to 74.5.
[NLP-13] Recipes for Creativity: Iterative Generation and Evaluation in Large Language Models
【速读】： 该论文旨在解决生成式人工智能（Generative AI）在创意评估中普遍依赖单一生成结果，而忽视人类创造力中迭代生成、评估与优化这一核心过程的问题。其解决方案的关键在于引入基于迭代搜索的生成-选择机制，将FunSearch框架适配至2024年Pillsbury烘焙大赛的食谱生成任务中，并采用基于托伦斯创造性测验（TTCT）的人类基准评估体系对生成结果进行量化评价。研究通过控制迭代次数、生成器温度及闭环选择评分模型规模，发现仅增加迭代次数无法持续提升创意表现；而闭环评估器的设计是决定性因素：较小规模的选择评分模型在多数TTCT维度上显著优于大型模型，表明评估器的“适度复杂度”更有利于激发创意；此外，生成温度仅对原创性维度有有限影响。研究结果强调，评估器设计应作为主观创造性搜索中的首要设计变量。
链接: https://arxiv.org/abs/2608.07243
作者: Rens Anderson,Tessa Verhoef,Amirhossein Zohrehvand
机构: Leiden University (莱顿大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Neural and Evolutionary Computing (cs.NE)
备注: 7 pages, 3 figures, 1 table. Short paper accepted at ICCC’26
Abstract:Generative models are often evaluated through singular artifacts, whereas human creativity typically emerges through iterative generation, appraisal, and refinement. This pilot study examines whether iterative search improves LLM creativity by adapting FunSearch to recipe generation for the 2024 Pillsbury Bake-Off and evaluating outputs against human benchmarks using TTCT-based LLM evaluation. Across two experiments, we test iteration count, generator temperature, and in-loop selection-scorer model size. Results show that iterative generation-selection can produce recipes with creativity scores comparable to human benchmarks, but additional iterations alone do not improve creativity. The in-loop evaluator matters most: a smaller selection scorer yields significantly higher scores across most TTCT dimensions, while temperature has limited effects except for originality. These findings suggest that evaluator design is a first-order design variable in subjective creative search.
[NLP-14] Skaling: Chinchillas Exponents Meet Kaplans Coupling
【速读】： 该论文旨在解决现有神经网络缩放定律（Neural Scaling Laws）在数据稀缺和过训练极端情况下对损失值系统性低估或高估的问题。其根本原因在于传统模型假设模型规模与训练数据量对损失的影响是相互独立的，忽略了二者之间的耦合效应。为此，论文提出一种名为“Skaling law”的广义函数形式，通过引入一个单一的交互指数，将模型容量与数据量进行耦合建模。该方法的关键创新在于实现了模型规模与数据量之间非独立的交互关系，显著提升了缩放预测的准确性，在内插与外推两种场景下均使平均绝对百分比误差（MAPE）降低1.5至3倍。结合稀疏网格策略并限制于低计算资源场景，该方法仅需约1/10的计算量即可实现与均匀扫描相当的全网格外推精度，从而为下一代大模型训练提供了更稳健且高效的计算资源配置框架。
链接: https://arxiv.org/abs/2608.07222
作者: Mathurin Videau,Badr Youbi-Idrissi,David Lopez-Paz,Kartik Ahuja
机构: Meta(元)
类目: Computation and Language (cs.CL)
备注:
Abstract:Neural scaling laws are foundational for language model development, yet standard formulations systematically under- and overestimate loss at data-scarce and overtraining extremes. This failure originates in the underlying assumption that model size and training data impact the loss independently. To address this, we introduce the Skaling law, a generalized functional form that couples model capacity and data through a single interaction exponent. This simple extension reduces the Mean Absolute Percentage Error (MAPE) by 1.5-3x across both interpolation and extrapolation regimes. When paired with a sparse grid strategy restricted to low-compute regimes, the Skaling law achieves accurate full-grid extrapolation using approximately 10x less compute than uniform sweeps. By enabling reliable performance prediction from small-scale experiments, the Skaling law provides a more robust and resource-efficient framework for allocating compute budgets in next-generation model training.
[NLP-15] From Test-Time Scaling to Reusable Memory: Measuring Crystallization in Text-to-SQL
【速读】： 该论文旨在解决测试阶段缩放（test-time scaling）中生成的纠错结果在完成回答后被丢弃的问题，即如何评估这些纠错经验对未来未见查询的潜在价值，这一问题被称为“结晶化问题”（crystallization problem）。其核心挑战在于：现有评估方法仅报告端到端单一得分，无法区分对重复问题的重播（replay）与对新问题的帮助，也无法识别哪些记忆选择真正贡献了性能提升。为解决此问题，研究提出一种受控评估范式，固定单次求解器（single-shot solver），逐个变量地改变记忆策略，分别测量重播效果、跨问题记忆保留能力以及同数据库下的外部迁移表现。实验结果表明，在BIRD数据集上，存储经验证的修正查询可使外部首次尝试准确率提升4.34个百分点，相当于捕获了按需修复在同一类问题上所能提供的44.4%的性能提升空间。进一步分析揭示，数据库特定内容是主要有效成分；可靠的验证机制与更广的检索覆盖范围带来显著增益，而更丰富的格式或复杂的检索器则无明显帮助。
链接: https://arxiv.org/abs/2608.07213
作者: Jiaqian Wang(1),Yutao Qi(1),Wenjin Hou(1),Yuanxi Che(1),Muning Wen(2) ((1) Xidian University, (2) Shanghai Jiao Tong University)
机构: Xidian University(西安电子科技大学); Shanghai Jiao Tong University(上海交通大学)
类目: Computation and Language (cs.CL)
备注: 18 pages, 6 figures. Open-source code, evaluation artifacts, and reproduction instructions: this https URL
Abstract:Test-time scaling can correct difficult text-to-SQL queries, but the extra computation is normally discarded after each answer. Systems increasingly retain verified repair episodes, yet evaluations still report one end-to-end score. It cannot distinguish replay on recurring questions from help on unseen questions, or identify the responsible memory choice. We call measuring this future value the crystallization problem. Our controlled evaluation holds the single-shot solver fixed and varies one memory choice at a time. We separately measure replay, cross-question retention, and held-out same-database transfer. On BIRD, storing verified corrected queries improves held-out first-attempt accuracy by 4.34 percentage points. This gain captures 44.4% of the accuracy headroom provided by on-demand repair on the same questions. Controlled interventions identify database-specific content as the main operating ingredient. Reliable verification and broader retrieval coverage yield supported gains; richer formats and elaborate retrievers do not. Open-source code, evaluation artifacts, and reproduction instructions are available at this https URL.
[NLP-16] Measuring Concept Content in Text from LLM Activations: ESG Evidence from Concept Vectors and Linear Probes
【速读】： 该论文旨在解决现有文本概念内容度量方法仅依赖文本表面特征（如词典词汇重合度、主题比例、嵌入相似性）的局限性，这些方法仅评估文本使用的词汇，而非读者对文本所蕴含概念的内在判断。针对大语言模型（LLM）内部知识与其输出响应之间存在的认知鸿沟，本文提出利用冻结的预训练模型激活值作为概念内容的替代度量，以避免任务特定微调。其核心解决方案是通过递归特征机器（Recursive Feature Machine, RFM）和线性探测（linear probing）两种方式从模型激活中提取概念内容表征，并与嵌入基线、表面特征基线以及模型自身回答进行对比。实验基于金融领域中的人工标注环境、社会与治理（ESG）数据集展开，结果表明：最优线性探测器在未进行任何任务微调的情况下，准确率仅比微调后的领域分类器低0.6个百分点，且在12次比较中有11次优于模型自身对问题的回答，证明了模型激活值中蕴含着未被输出表达的概念信息。此外，线性探测性能持续优于RFM生成的概念向量，而后者进一步提供了连续评分，可反映概念在文本中的强弱程度，但其有效性仍需带分级标签的验证。
链接: https://arxiv.org/abs/2608.07208
作者: Luc Hazenoot,Zhaochun Ren,Amirhossein Zohrehvand
机构: Leiden Institute of Advanced Computer Science, Leiden University (莱顿大学高级计算机科学研究所)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); General Economics (econ.GN)
备注: 19 pages, 1 figure, 7 tables
Abstract:Existing measures of how much a text is about a concept read the surface of the text: dictionary word shares, topic proportions, embedding similarities. They score the words a text uses, not the judgment a reader forms about it. Recent work has shown that a gap exists in what Large Language Models (LLMs) know internally versus what they express in their response. This paper asks whether that internal knowledge, read by monitoring the activations of frozen, out-of-the-box LLMs, can stand in for task-specific fine-tuning when measuring concept content, and which extraction method reads it best. We extract such measures via the Recursive Feature Machine (RFM) algorithm and via linear probing, and compare these against an embedding baseline, surface baselines, and the same model’s own answer to the question. We demonstrate the approach on financial text, a domain studied extensively and served by established annotated resources, using a human-annotated Environmental, Social and Governance (ESG) dataset. The best linear probe comes within 0.6 percentage points of a fine-tuned domain classifier’s accuracy without any task-specific fine-tuning, and outscores the same model’s own answer to the question in eleven of twelve comparisons, so the activations carry concept content the response does not report. The simple probe consistently beats the RFM concept vectors, which in turn provide what classification alone does not: a continuous score intended to reflect how strongly a concept is present in a text, whose validation awaits graded labels.
[NLP-17] HNR-DAC: Hard-Negative Reranking and Distribution-Aligned Classification for Scientific Claim Verification
【速读】： 该论文旨在解决科学主张验证中两个紧密关联的挑战：一是论文内部的干扰段落（within-paper distractors）常与真实证据难以区分，二是基于标注证据训练的分类器在推理阶段需处理检索所得的证据，存在分布偏移问题。其解决方案的关键在于提出一种两阶段框架HNR-DAC，通过分阶段训练使每个阶段均在实际运行时所面对的输入条件下进行优化。第一阶段硬负样本重排序（Hard-Negative Reranking, HNR）利用基础重排序器对非黄金段落的打分，量化证据混淆程度，并将真实证据与最易混淆的候选段落进行对比；第二阶段分布对齐分类（Distribution-Aligned Classification, DAC）则在冻结的HNR生成的Top-1段落上进行训练，同时以HNR输出的Top-3段落作为最终证据输出。该方法有效缓解了训练与推理之间的分布不一致问题，在NLPCC 2026 Task 10 Track 2任务中取得了优异性能，显著提升了主张-论文关系预测与证据定位的准确性。
链接: https://arxiv.org/abs/2608.07204
作者: Zhenchao Wang,Xin Chen,Luoxi Zhang,Min Yang,Shiwen Ni
机构: 未知
类目: Computation and Language (cs.CL)
备注: 12 pages, 3 figures
Abstract:Scientific claim verification over a cited paper requires predicting the claim–paper relation and identifying the paragraphs that justify that prediction. This setting poses two linked challenges: within-paper distractors often resemble genuine evidence, while a classifier trained on gold evidence must operate on retrieved evidence at inference. We present HNR-DAC, a two-stage framework that trains each stage on the cases it will actually encounter. Hard-Negative Reranking (HNR) quantifies evidence confusability using a base reranker’s scores on non-gold paragraphs and contrasts gold evidence against the most confusable candidates. Distribution-Aligned Classification (DAC) trains on the Top-1 paragraph produced by the same frozen HNR used to construct inference inputs, while HNR’s Top-3 paragraph identifiers provide the evidence output. On the NLPCC 2026 Task 10 Track 2, the final configuration obtains 97.21% Hit@3, 95.79% Macro-F1, 94.47% Joint@3, and an average score of 95.13%. The corresponding submission ranks third on the official Track 2 leaderboard while achieving the highest overall Macro-F1 of 93.05%, alongside 70.16% Joint@3 and an average score of 81.61%.
[NLP-18] Modular TTT: Rethinking Test-Time Training as Composable Modules
【速读】： 该论文旨在解决现有测试时训练（Test-time Training, TTT）方法中存在的一致性与可扩展性问题：当前的TTT变体通常被硬编码实现，导致新方法的设计困难，且难以分离和评估各组件的独立作用。为此，本文提出模块化测试时训练（Modular TTT）框架，其核心创新在于将内部学习器建模为有向无环图（Directed Acyclic Graph, DAG），并将快速权重网络、损失函数、学习率、权重衰减及归一化等关键要素显式地作为可配置的设计维度。通过自动组合基础级别的前向训练视图、反向训练视图以及因果查询视图规则，该框架实现了完整的图级TTT计算流程，包括快速权重状态转移。基于此框架，研究系统性地消融分析了各组件的影响，发现小的学习率初始化、权重衰减以及单层非线性有助于提升性能，而均方误差（MSE）与内积损失表现相近；深层快速权重网络和归一化虽增强表达能力但易引发过大的激活值，反而损害性能，残差连接与门控机制则未带来显著收益。结合上述发现，作者训练了参数量分别为410M和1.45B的模型，在1000亿个标记上进行训练，其训练损失与基准测试性能可与门控增量网络（Gated DeltaNet）相媲美。
链接: https://arxiv.org/abs/2608.07110
作者: Bohao Tang,Zhen Qin,Yuqi Pan,Zheng Li,Pengfei Liu,Ya Zhang
机构: 未知
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: Code: this https URL
Abstract:Test-time training (TTT) views sequence modeling as an online learning problem in which fast weights are updated by an internal learning rule. Despite the growing number of TTT variants, existing approaches typically hard-code each variant separately, which makes it difficult to design new TTT methods and to isolate the role of each component. To address this, we propose Modular TTT, a framework that represents the inner learner as a directed acyclic graph and exposes the fast-weight network, loss function, learning rate, weight decay, and normalization as explicit design dimensions. Modular TTT automatically composes primitive-level train-view forward, train-view backward, and causal query-view rules into the full graph-level TTT computation, including the fast-weight state transition. Using Modular TTT, we systematically ablate the components of TTT and find that small learning-rate initialization, weight decay, and a single-layer nonlinearity improve performance, while MSE and inner-product losses perform similarly. Deeper fast-weight networks and normalization tend to hurt performance because they induce excessively large activations, while residual connections and gating provide little measurable benefit. Guided by these findings, we train the best resulting variant as 410M- and 1.45B-parameter models on 100B tokens, and observe training loss and benchmark performance comparable to Gated DeltaNet.
[NLP-19] An Agent ic Hybrid Top-Down and Bottom-Up Approach to Knowledge Graph Generation
【速读】： 该论文旨在解决人力资源（HR）平台中数千条非标准化、多语言的专业技能声明难以有效组织的问题，这一挑战直接影响人才匹配等下游任务的准确性。其核心解决方案在于提出一种混合式知识图谱生成流程：通过将大型语言模型（LLM）锚定在多语言维基数据（Wikidata）知识图谱（KG）之上，并结合代理反思（agentic reflexion）模式，实现对新兴概念及其关联元数据的动态合成。与传统僵化的自上而下方法或碎片化的自下而上方法不同，该系统在保持已识别概念与稳定知识图谱实体对齐的同时，能够动态创建未识别技能的新节点及关系元数据。整个流程包含五个阶段——实体对齐、多语言规范化、主动校准、去重以及未映射概念的迭代恢复，使系统可自主适应五种欧洲语言中快速演变且噪声频发的技能表述。最终，该管道构建了一个高度可扩展、可解释且具备自我修复能力的综合技能知识图谱框架，从非结构化、高噪声文本中提取出结构化的技能分类体系。
链接: https://arxiv.org/abs/2608.07023
作者: Emma Jouffroy,Warren Jouanneau,Marc Palyart
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Organizing thousands of unstandardized, multilingual expertise declarations is a persistent challenge for Human Resources (HR) platforms, directly impacting downstream tasks like accurate talent matching. To address this, we propose a hybrid knowledge graph generation pipeline that grounds a Large Language Model (LLM) in the Wikidata multilingual Knowledge Graph (KG) while employing an agentic reflexion pattern to synthesize emerging concepts and their associated metadata. Unlike rigid top-down methods or fragmented bottom-up approaches, our system anchors recognized concepts to stable Knowledge Graph entities while dynamically creating new nodes and relational metadata for unrecognized skills. Executed across five stages, entity reconciliation, multilingual canonicalization, active curation, deduplication, and the iterative recovery of unmapped concepts, the system autonomously adapts to rapidly evolving, noisy skill mentions across five European languages. Ultimately, this pipeline provides a highly scalable, explicable, and self-healing framework for generating a comprehensive skills knowledge graph, from which a structured taxonomy is derived, using unstructured, noisy text.
[NLP-20] Does More Retrieved Evidence Help Visual Retrieval-Augmented Generation with Diffusion Language Models?
【速读】： 该论文旨在解决视觉检索增强生成（Visual Retrieval-Augmented Generation, RAG）中因无差别地将所有检索到的证据传递给扩散语言模型（Diffusion Language Models, DLMs）而导致的答案准确性下降问题。传统方法假设增加检索证据数量可提升答案覆盖率，但研究表明，对于DLM而言，过多或不一致的视觉证据会引发语义冲突，导致生成结果偏离真实答案。其关键原因是并行去噪过程中存在的源一致性损失（source-coherence loss），使得不同视觉来源在位置级提案时可能组合出不支持的结论。研究发现，这种干扰早在第一步的答案块分布中即已显现，因而可在解码前对证据进行评估。为此，作者提出无需训练的基于熵的候选过滤框架（Entropy-Based Candidate Filter, ECF），通过构建多粒度证据单元减少冗余内容，并利用空白控制的块置信度与检索排名来判断是否及如何引入额外证据。实验表明，在三个多模态DLM和五个视觉问答基准上，ECF平均优于最强的固定top-k输入2.62个百分点，且在LLaDA2.0-Uni上平均优于现有最优无训练方法2.37个百分点，证明了通过选择性证据接纳而非无条件扩展，能够真正发挥广泛检索的优势。
链接: https://arxiv.org/abs/2608.07006
作者: Jiankun Wang,Yisen Gao,Ziwei Zhang,Xingcheng Fu,Jiaxin Bai,Chen Gao
机构: 未知
类目: Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Visual retrieval-augmented generation (RAG) commonly expands the retrieved evidence set to improve answer-page coverage, implicitly assuming that all available evidence should be passed to the generator. We show that this assumption does not hold for diffusion language models (DLMs): retrieving more pages increases answer-page recall, whereas unconditionally passing all retrieved pages to the generator often reduces answer accuracy, primarily because of semantic conflict. A latent-source analysis explains this mismatch through source-coherence loss in parallel denoising, where position-wise proposals can combine incompatible visual sources into unsupported answers. We further find that such interference is already visible in the first-step answer-block distribution, making it possible to assess evidence before decoding. To preserve retrieval coverage while limiting harmful visual exposure, we propose the Entropy-Based Candidate Filter (ECF), a training-free evidence-admission framework. To reduce irrelevant content within individual candidates, ECF constructs multi-granularity evidence units; to identify beneficial additional evidence, it uses blank-controlled block confidence and retrieval rank to determine whether and which candidate should enter the final context. Across three multimodal DLMs and five visual QA benchmarks, ECF improves answer accuracy by 2.62 percentage points on average over the strongest fixed top- k input and, with LLaDA2.0-Uni, by 2.37 percentage points on average over the best competing training-free result for each dataset. These results show that broader retrieval benefits visual DLM-RAG through selective evidence admission rather than unconditional evidence expansion. Code is publicly available at this https URL.
[NLP-21] GPT KB 2.0: Browsing Querying and Auditing a Disambiguated LLM -Derived Knowledge Base
【速读】： 该论文旨在解决大语言模型（LLM）生成的知识库中实体歧义（entity ambiguity）与指称冗余（redundant mention）的问题。传统基于LLM构建的知识库多依赖表面字符串进行实体识别，易导致同形异义词混淆及同义表达重复，影响知识的准确性与可检索性。其解决方案的关键在于在递归构建知识库的过程中引入上下文引导的消歧机制（context-guided disambiguation），通过分析上下文语境动态区分同名异义实体，并合并语义等价的提及项（synonymous mentions），从而实现更精确的实体对齐与知识融合。这一过程不仅提升了知识库的结构化质量，还通过可视化交互界面使知识来源、消歧决策及事实推导路径可追溯，支持自然语言查询到SPARQL的转换与文本中的实体链接，显著增强了知识库的透明性与可用性。
链接: https://arxiv.org/abs/2608.06992
作者: Yujia Hu,Tuan-Phong Nguyen,Simon Razniewski
机构: ScaDS.AI Dresden/Leipzig TU Dresden, Germany; Institute for AI, VNU University of Engineering and Technology, Hanoi, Vietnam
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Databases (cs.DB)
备注: 7 pages, 11 figures
Abstract:We present a web demo for exploring a large-scale disambiguated knowledge base (KB) materialized from a large language model (LLM). GPTKB 2.0 contains 38.4M triples over 1.6M canonical entities, together with 207.6K consolidated relations and 66K consolidated classes. Unlike prior LLM-derived knowledge bases that largely identify entities by surface strings, GPTKB 2.0 performs context-guided disambiguation during recursive KB construction, separating homonyms and merging synonymous mentions as facts are elicited. The demo makes this process inspectable: users can browse entities, follow links across the KB, and audit the provenance of individual facts, including surface forms, candidate matches, source triples, and disambiguation decisions. The interface further supports structured SPARQL queries, natural-language questions translated to SPARQL, and entity linking from user-provided text to canonical GPTKB 2.0 entries. GPTKB 2.0 is available at this https URL, with the full KB downloadable for offline use.
[NLP-22] Confirming Our Biases? Evaluating the Capabilities Risks and Societal Impact of Large Language Models
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在面对不同提示框架时，其响应是否会被用户隐含偏见或显式指令所引导，进而影响输出内容的客观性与事实一致性的问题。研究重点在于厘清提示框架中隐性暗示效应与显性操控之间的边界，探究模型在意见型和事实型领域中对提示策略的敏感程度。其解决方案的关键在于通过系统化设计160个涵盖十个主题、覆盖多种提示策略、支持/挑战指令、提示极性及用户已有信念表达的多样化提示，评估六种主流大语言模型在不同情境下的响应模式。结果表明，即使在应保持事实一致性的领域，大语言模型仍会系统性地调整输出以匹配提示框架，说明提示框架的影响可能超越事实本身，从而揭示了大语言模型在实际应用中存在被隐性偏见强化和显性操控的风险，凸显了提示工程对模型行为的重要影响。
链接: https://arxiv.org/abs/2608.06977
作者: Mudar Adas,Polina Tsvilodub,Michael Franke,Martin V. Butz
机构: University of Tübingen(图宾根大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:It is well established that large language models (LLMs) are sensitive to prompt framing, reflecting patterns in their training data or prior prompts. In this study, we investigate the extent to which LLMs reinforce users biases expressed in the prompts and examine the boundary between implicit framing effects and explicit prompt manipulation. Specifically, we evaluate how susceptible LLMs are to direct and suggestive prompts that encourage models to support or challenge particular positions. We evaluate six LLMs using 160 distinct prompts spanning ten topics across opinion-based and factual domains. The prompts systematically vary in prompting strategy, support versus challenge instructions, prompt polarity, users’ expressed beliefs, and topic domain, spanning both opinion-based and factual questions. Our results show that LLMs systematically adapt their responses to align with prompt framing, even in factual contexts. This suggests that prompt framing can outweigh factual consistency in model responses. Overall, our findings delineate the extent and boundaries of LLM manipulability. Furthermore, the results imply that LLMs can reinforce subtle user biases and are susceptible to explicit prompt manipulation even in domains where responses should remain factually stable. Subjects: Computation and Language (cs.CL) Cite as: arXiv:2608.06977 [cs.CL] (or arXiv:2608.06977v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2608.06977 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[NLP-23] PHASE-Tree: Modeling Character-State Evolution in Long-Horizon Role-Playing Dialogue
【速读】： 该论文旨在解决长时序角色扮演（long-horizon role-playing）中角色状态演化与身份一致性难以兼顾的核心问题：现有方法普遍采用静态角色表征，无法在不破坏未变化特质的前提下进行局部更新；同时，主流评估基准仅关注人物设定保持与记忆召回能力，而忽视模型是否能基于角色当前演化的状态进行生成。为此，论文提出PHASE-Tree——一种多时间尺度的角色状态树架构，其包含不可变的身份根节点（immutable identity root）以及可变的个性（persona）、会话（session）和瞬时（moment）层，使每个可变字段均可作为局部化跨剧集更新的目标。该架构通过显式文本输入或隐式参数适应实现生成条件控制。为评估角色状态演化能力，论文构建了LongEvoRoleBench评测基准，结合四组长对话语料用于跨剧集演化分析，以及四组短对话语料作为场景内状态追踪的对照，统一采用下一话语生成协议。实验结果表明，在长对话核心任务上，基于文本的PHASE-Tree在12项指标中的11项超越内部变体，在全部12项对比外部文本基线的方法中均取得领先，分别在角色级、语义和嵌入层面提升19.7%、12.4%和15.1%。双盲评估中，人类评分与GPT-4.1判别器的相关性达Pearson r=0.65，描述性提示子集的总体差异为+0.20，且长对话语义优势在不同大语言模型判别器及生成骨干网络下均持续存在。
链接: https://arxiv.org/abs/2608.06975
作者: Bo Tang,Jianan Yang,Junyi Zhu,Yiquan Wu,Rui Zhao,Zhengyu Yang,Yang Zhang,Feiyu Xiong,Zhiyu Li,Jiajun Shen
机构: MemTensor(内存张量); Shenzhen MemTensor Technology Co., Ltd.(深圳内存张量科技有限公司)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Long-horizon role-playing demands that characters remain recognizable as they evolve with the narrative. Yet existing work falls short on two fronts: representations are typically static profiles that cannot be updated locally without destabilizing unchanged traits, and benchmarks mainly test persona preservation and memory recall rather than whether a model speaks from a character’s currently evolved state. We address both. PHASE-Tree is a multi-timescale character-state tree with an immutable identity root and mutable persona, session, and moment layers, making each mutable field an addressable target for localized within- and cross-episode updates. It conditions generation through explicit textual provision or implicit parametric adaptation. To measure evolved-state generation, we introduce LongEvoRoleBench, which pairs four long-dialogue corpora for cross-episode evolution with four short-dialogue corpora as within-scene state-tracking checks, under a unified next-utterance protocol. On the long-dialogue core, textual PHASE-Tree ranks first in 11 of 12 dataset-metric cells against internal variants and all 12 cells against external textual baselines, improving character-level, semantic, and embedding scores by 19.7%, 12.4%, and 15.1% respectively. In a blinded 200-response study, human ratings correlate with the GPT-4.1 judge (Pearson r= 0.65); on descriptive n= 10 PT and NR prompt subsets, the Overall difference is +0.20. The long-dialogue Sem advantage persists across LLM judges and generation backbones.
[NLP-24] Can Language Models Imagine Without Seeing? Ekphrasis: Measuring Visual Creative Ideation in Text-Only LLM s
【速读】： 该论文旨在解决当前评估体系无法区分纯文本语言模型是否具备在图像生成前生成原创视觉概念的能力这一关键问题。现有评价方法常被“流畅的视觉叙事”所误导，导致模型即使重复常见视觉套路或未能生成可渲染场景，仍可能因语言表达流畅而获得高分。为此，论文提出“视觉创意构思（Visual Creative Ideation, VCI）”这一新能力维度，定义为生成具有实用性、表现力且在群体层面新颖的文本视觉方案的能力。其解决方案的核心在于构建Ekphrasis基准测试——一个包含400个任务的多维度评估体系，覆盖抽象（Abstraction）、组合（Combination）、转换（Transformation）和适应（Adaptation）四大类视觉构思行为。该框架通过匿名化成对比较、基于维度特异性检查表的评分、Bradley-Terry模型聚合偏好，并引入有类型思想图（Typed Idea Graphs）将任务特定的群体惯用模式转化为新颖性参照标准，从而实现对VCI三大子维度（实用性、表现力、新颖性）的解耦评估。实验表明，在14种语言模型中，VCI能够有效分离出三者之间的差异，而非简单归结于语言流畅性；强模型虽总分相近，但其构成路径各异，且实用的视觉计划仍可能陷入视觉陈词滥调。进一步的跨模态对齐研究表明，文本层面的VCI排序在忠实图像渲染及盲评图像偏好判断后仍保持稳定，验证了Ekphrasis作为超越语言质量的视觉创意能力度量工具的有效性。
链接: https://arxiv.org/abs/2608.06967
作者: Hongyu Luo,He Wang,Huihao Jing,Hong Ting Tsang,Yuxuan Liu,Wuganjing Song,Yauwai Yim,Chunyang Li,Yangqiu Song
机构: The Hong Kong University of Science and Technology (香港科技大学)
类目: Computation and Language (cs.CL)
备注: 25 pages, 4 main figures, with appendices. Code and data: this https URL
Abstract:Current evaluations do not isolate whether text-only language models can originate visual concepts before image generation. Fluent visual prose can hide visual-plan failures: an answer may appear creative while repeating familiar visual clichés or failing to specify a renderable scene. We define Visual Creative Ideation (VCI) as the ability to produce textual visual plans that are useful, expressive, and population-novel, and introduce Ekphrasis, a 400-task benchmark spanning Abstraction, Combination, Transformation, and Adaptation. Ekphrasis scores anonymized pairwise comparisons with dimension-specific checklists, aggregates preferences with Bradley-Terry models, and uses Typed Idea Graphs to convert task-specific population clichés into novelty references. Across 14 language models, VCI separates usefulness, expressiveness, and novelty rather than reducing to fluency: strong models achieve similar overall scores through different profiles, and useful plans can remain visually clichéd. A cross-modal grounding study further shows that text-level VCI ordering largely survives faithful rendering and blind image-level preference judgment, supporting Ekphrasis as a measure of visual ideation beyond prose quality.
[NLP-25] Explicit Not Longer: What Makes Epistemic Stance Survive Memory Compression
【速读】： 该论文旨在解决生成式AI（Generative AI）在记忆系统中对主张的元认知立场（epistemic standing）保留不足的问题，即在信息压缩过程中，立场判断常被忽略或弱化，导致模型难以准确维持原始主张的可信度或确定性。其核心解决方案在于：将立场以显式标签（labelled field）形式明确表达，而非作为括号内的附加说明（bracketed aside），从而显著提升立场信息的保留率。实验表明，这种格式改进在多个模型上均带来约15个百分点的保留率提升（如37→2、30→8，置换检验p=0.00005），且在预注册复制实验中亦验证了+15.6点的效果。进一步的消融分析显示，显式表达是关键因素（分别贡献+9.7和+12.8），而文本长度本身并无显著作用；此外，将立场表述为完整句子在某一模型上效果突出（+12.5），但在另一模型上无效（+0.6），说明最佳显式方式依赖于具体模型特性。研究强调，仅当多种证据一致支持时才可得出结论：应使立场显式化，而非简单延长文本；同时，基于确定性读出的非模型方法虽能复现部分结果，但无法解释标签与长度效应，故不将其纳入机制主张。研究通过人工标注验证方向一致性（kappa=0.75），并公开七处分歧及九项撤回主张，增强透明性。
链接: https://arxiv.org/abs/2608.06953
作者: Alex Kwon
机构: 独立研究员（Independent Researcher）
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 20 pages, 3 figures, 4 tables. Code, per-trial data, and the pre-registration commit: this https URL
Abstract:Agent memory systems compress what they store, and compression is built to drop qualifiers, so a claim’s epistemic standing tends not to survive being written to memory. We ask what governs whether it does. Matched notes carry the identical claim and identical stance and differ only in where that stance sits; one model compresses both under the same budget among the same filler notes, and a blind reader that never sees the condition scores the result. Across 60 claims in seven registers, writing the stance as a labelled field rather than a bracketed aside raises retention by about 15 points on two models (37 claims to 2 on one, 30 to 8 on the other; permutation p=0.00005), and a pre-registered replication on Haiku, its prediction and decision rule committed before the run, gives +15.6 points, 38 claims to 1. Ablating the format on both models gives the same net effect from different parts: labels help on both (+9.7 and +12.8) and length helps on neither, but wording the stance as a full sentence is the largest component on one model (+12.5) and worth nothing on the other (+0.6). Either model alone would have licensed a confident and different mechanism, so we claim only the intersection: make the stance explicit, not merely longer, and expect the best way of being explicit to depend on the model. A deterministic readout with no model reproduces the two-cell direction and five of seven ablation contrasts, but not length or labels, which we therefore do not claim on one instrument. Fifty hand labels (kappa=0.75) agree on direction; we print their seven disagreements in full. We also report nine withdrawn claims, three of them former title claims of this paper.
[NLP-26] Ask-E: An Environment for Calibrated Question Generation
【速读】： 该论文旨在解决当前大模型能力评估与训练中面临的核心难题：如何生成在难度上精确匹配模型当前能力边界的测试问题。传统方法依赖人工设计或现有数据集中的问题，但随着模型能力持续提升，准确识别并构造处于其能力前沿的、具有精细难度控制的问题变得愈发困难，且需要超越模型自身能力的洞察力。这一困境本身构成了一个关键约束——能够持续生成精准校准问题的模型，必然具备超越目标能力边界的能力。基于此，论文提出Ask-E环境，其核心创新在于将评估与训练重点从“回答问题”转向“生成问题”，即要求模型生成的题目恰好仅能被一对基准语言模型中的一个成功解答，从而实现对目标技能水平的精确校准。该机制通过双模型对比，自然定义出难度范围，并有效区分二者能力差异。实验表明，即使顶尖模型在该基准上的校准成功率也低于50%，表明仍有巨大提升空间；更重要的是，仅在Ask-E环境中进行训练即可显著提升模型在多个下游数学任务的表现，无需引入新数据、无需与更强模型交互、也无需基于正确性的奖励信号，验证了该方法在促进模型自我进化方面的有效性。
链接: https://arxiv.org/abs/2608.06933
作者: Sarah Pratt,Jae Sung Park,Scott Geng,Ali Farhadi
机构: University of Washington (华盛顿大学); Allen Institute for AI (艾伦人工智能研究所)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Today, we improve models by training and evaluating them on problems at the frontier of their abilities. Creating such problems is itself a demanding task, requiring the ability to probe model limits and generalize beyond existing question distributions. It also means placing problems at a precise difficulty level, which requires understanding what it takes to solve them. In short, generating problems calibrated to a model’s current frontier demands capability beyond it, an increasingly burdensome constraint as models improve. Our key insight is that we can leverage this constraint to our advantage: a model that can generate problems consistently calibrated to a given frontier must possess capability beyond it. Accordingly, we present Ask-E, an environment that benchmarks and trains models on their ability to write questions at a given skill level, rather than answer them. Concretely, we define target skill levels as ranges bounded by the capabilities of two existing language models. A generated question is successfully calibrated if exactly one of the two models can solve it, placing it precisely within the target range and differentiating the capabilities of these models. Ask-E serves both as a benchmark and a training environment, where models generate problems calibrated to a variety of skill levels. We find that even frontier models achieve below 50% calibration on the benchmark, leaving significant headroom to measure future progress. We also show that training on this environment leads to improvements across a number of downstream math benchmarks even with no new math data, no interaction with stronger models, and no correctness-based reward.
[NLP-27] Calibrating WEAT Against Anisotropy: ZCA Whitening as a Geometric Pre-Processing Step for Embedding Association Tests AAAI
【速读】： 该论文旨在解决词嵌入空间中存在显著方向性（anisotropy）导致的词向量关联测试（Word Embedding Association Test, WEAT）测量结果不可靠的问题。WEAT依赖余弦相似度作为语义关联的度量，其有效性建立在嵌入空间近似各向同性的假设之上，但大量语言模型的实际嵌入空间违背了这一假设，从而可能引发对偏见程度的误判。本文提出采用零相位成分分析（Zero-phase Component Analysis, ZCA）白化作为几何预处理步骤，通过将嵌入空间的协方差矩阵转换为单位矩阵，在最小扰动原始向量的前提下恢复空间的各向同性。该方法的关键在于：在不显著改变语义结构的前提下，有效校正嵌入空间的方向性偏差，从而为WEAT提供更可靠的测量基础。实验在10个标准WEAT测试集和7种跨架构的语言模型上进行，覆盖70个模型-任务组合，结果显示ZCA白化显著降低了所有模型的嵌入空间各向异性；对于高度各向异性的模型，还观察到在标准语义相似性基准上的性能提升，表明校准后的空间更能准确捕捉语义关联。更重要的是，超过30%的WEAT结果在经过校准后改变了显著性水平，效应量也因偏见类别而呈现双向变化，说明未校准的测量可能同时高估或低估嵌入空间中的关联强度。这些发现提示，以往基于非各向同性嵌入空间报告的偏见测量结果需谨慎解读，建议采用校准方法重新评估。本研究为计算社会科学与人工智能公平性研究中的偏见测量提供了更为坚实的理论基础。
链接: https://arxiv.org/abs/2608.06908
作者: Seitaro Ono,Senna Ross,Jun Saiki
机构: University of Tokyo (东京大学); University of California, Berkeley (加州大学伯克利分校)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Machine Learning (cs.LG)
备注: Extended version (with appendices) of a paper accepted at the 9th AAAI/ACM Conference on AI, Ethics, and Society (AIES 2026)
Abstract:We propose Zero-phase Component Analysis (ZCA) whitening as a geometric pre-processing step for the Word Embedding Association Test (WEAT). WEAT is a bias measurement method widely used in both computational social science and AI fairness research. It relies on cosine similarity as a measure of semantic association, which assumes that the embedding space is approximately isotropic. However, prior work has reported that many widely used language models do not satisfy this assumption, raising concerns about the reliability of bias measurements. ZCA whitening transforms the covariance of the embedding space into the identity matrix while minimizing perturbation to the original vectors. This transformation restores the isotropy condition on which WEAT relies. We evaluate our approach on ten standard WEAT test suites and seven models spanning three architectural families, yielding 70 model-task combinations. The results show that ZCA whitening substantially reduces the anisotropy of the embedding spaces across all models. Particularly for highly anisotropic models, we further observe improvements on standard semantic similarity benchmarks, indicating that the calibrated space better captures semantic associations. After calibration, over 30% of WEAT results change significance status, and effect sizes shift in both directions depending on bias category. These shifts suggest that uncalibrated measurements may both overestimate and underestimate the associations encoded in the embedding space. These findings indicate that previously reported bias measurements in anisotropic embedding spaces should be interpreted with caution and may benefit from re-evaluation with calibrated methods. Our approach contributes to restoring the measurement foundation of WEAT across both computational social science and AI fairness research.
[NLP-28] DAEP: Difficulty-Aware Evidence Planning for Medical Video Corpus Temporal Answer Grounding NLPCC2026
【速读】： 该论文旨在解决视频语料库中难度感知的时间答案定位（Difficulty-Aware Temporal Answer Grounding in Video Corpus, DA-TAGVC）问题，即在给定50个候选视频中准确检索目标视频，并精确定位支持答案的时间片段。其核心解决方案为DAEP（Difficulty-Aware Evidence Planning），通过融合字幕、视觉及程序化上下文三种模态证据进行视频排序，将高分锚点扩展为时间跨度，并对跨度进行重排序以生成最终输出。该方法的关键在于将任务提供的简单/复杂输入标签转化为推理阶段的证据规划策略，动态控制模态权重、Top-K聚合方式、边界阈值、扩展长度及重排序强度等关键参数，实现对不同难度问题的自适应处理。在官方评测中，该方法在十套系统中排名第一，平均得分为0.2728；验证消融实验表明，视觉证据、程序化上下文以及难度感知规划均显著提升了排序质量，尤其在复杂问题上取得最大性能增益。
链接: https://arxiv.org/abs/2608.06869
作者: Tianjian He,Yujie Liu,Zhiping Huang,Changbo Xu
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL)
备注: 12 pages, 2 figures, 5 tables, accepted by NLPCC 2026 Shared Task Track 3
Abstract:We describe DAEP, team BIGC’s submission to NLPCC 2026 Shared Task 1 Track 3: Difficulty-Aware Temporal Answer Grounding in Video Corpus (DA-TAGVC). The task requires retrieving the target video from 50 candidates and localizing the answer-supporting span. DAEP ranks videos with subtitle, visual, and procedural-context evidence, expands high-scoring anchors into temporal spans, and reranks spans for final output. Its main design is to convert the task-provided simple/complex input label into an inference-time evidence plan controlling modality weights, Top-K aggregation, boundary threshold, expansion length, and reranking strength. In the official evaluation, BIGC ranks first among ten systems with an Average score of 0.2728. Validation ablations show that visual evidence, procedural context, and difficulty-aware planning improve ranking quality, with the largest gain on complex questions.
[NLP-29] LLM Router: Unified Infrastructure for Developing Evaluating and Deploying LLM Routers
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）在实际应用中因模型性能与成本预算之间存在权衡，导致单一模型无法在所有查询场景下均实现最优表现的问题，进而提出高效、可扩展的模型路由（model routing）机制以实现成本效益最大化。其核心解决方案的关键在于构建了一个统一的序列决策过程框架，将模型路由形式化为包含五个关键组件的系统：上下文编码器（context encoders）、模型编码器（model encoders）、评分函数（scoring functions）、决策规则（decision rules）和学习信号（learning signals），从而涵盖单轮、多轮及个性化路由场景。基于此框架，研究设计了一套自动化管道，用于生成路由监督信号并联合评估路由策略在响应质量与推理成本上的表现，进而构建了跨领域基准测试集xRouteBench，覆盖通用大模型、记忆增强型模型、视觉、时间序列及个性化路由任务。此外，提出了开源模块化基础设施LLMRouter，集成超过16种代表性路由算法。实验表明，基于学习的路由方法相比最强的固定模型基线相对提升14.6%，轻量级路由在严格成本约束下更具竞争力，且用户条件驱动的路由能持续提升个性化效果。
链接: https://arxiv.org/abs/2608.06867
作者: Tao Feng,Fangxu Yu,Haozhen Zhang,Zhongjie Dai,Liangqi Yuan,Zijie Lei,Weizhi Zhang,Kunlun Zhu,Haodong Yue,Keyang Xuan,Ge Liu,Jiaxuan You
机构: University of Illinois Urbana-Champaign (伊利诺伊大学厄本那-香槟分校); University of Maryland, College Park (马里兰大学帕克分校); Nanyang Technological University (南洋理工大学); Purdue University (普渡大学); University of Illinois Chicago (芝加哥大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:No single large language model (LLM) is optimal across all queries and budget constraints, making model routing essential for cost-effective deployment. Existing routers adopt diverse formulations and implementations, making fair comparison and extension difficult. We present a unified formulation of LLM routing as a sequential decision process characterized by five components: context encoders, model encoders, scoring functions, decision rules, and learning signals, covering single-turn, multi-turn, and personalized routing. Based on this formulation, we develop an automated pipeline for constructing routing supervision and evaluating routers jointly on response quality and inference cost. The resulting benchmark, xRouteBench, spans generic LLM, memory-augmented, vision, time-series, and personalized routing tasks. We further introduce LLMRouter, an open-source modular infrastructure with more than 16 representative routers. Our empirical study shows that learned routers outperform the strongest fixed-model baseline by 14.6% relatively, lightweight routers become more competitive under tight cost constraints, and user-conditioned routing consistently improves personalization.
[NLP-30] Autonomy-of-Heads: Data-Free Sparse Attention from Frozen Query-Key Geometry
【速读】： 该论文旨在解决长上下文大语言模型（Long-context LLM）推理中因二次方注意力计算复杂度和不断增长的键值缓存（KV-cache）开销导致的性能瓶颈问题。现有稀疏注意力与KV压缩方法通常依赖运行时注意力分数、观察窗口、校准提示或学习到的门控机制来决定保留哪些令牌或注意力头，此类方法存在头诊断结果依赖输入、部署成本高的缺陷。本文提出一种无需数据的“注意力头自主性”（Autonomy-of-Heads, AoH）方法，通过分析查询-键投影的谱几何结构，自动识别检索型头（retrieval heads）与流式处理头（streaming heads）。AoH定义核注意力算子 $ M_h = W_K^h\top W_Q^h $，并利用其有效秩作为权重空间中头功能的度量：谱集中表示少数主导的查询-键匹配方向，对应于检索型头；谱弥散则表明缺乏主导全局匹配方向，对应于流式处理头。进一步地，该方法推导出一种 $ d_\text{head} $ 维的高效计算方式，避免构建完整的 $ d_\text{model} \times d_\text{model} $ 矩阵。大量实验表明，在50%稀疏度下，AoH平均保持了全注意力（Full Attention）96.5%的性能，同时将预填充（prefill）与解码（decode）延迟分别降低最高达41.4%和66.0%，并在256K上下文长度下实现50.0%的KV-cache内存压缩。其核心创新在于不依赖训练数据或运行时反馈，仅通过模型参数的内在谱特性完成注意力头的无监督分类与动态剪枝。
链接: https://arxiv.org/abs/2608.06849
作者: Yehan Yang,Junyuan Shang,Yang Li,Guanqun Zhao,Shuohuan Wang,Dianhai Yu
机构: Baidu Inc.
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Long-context LLM inference is bottlenecked by quadratic attention computation and growing KV-cache costs. Existing sparse attention and KV-compression methods typically decide which tokens or heads to preserve from runtime attention scores, observation windows, calibration prompts, or learned gates, making head diagnosis input-dependent and costly to deploy. We propose Autonomy-of-Heads (AoH), a data-free method that identifies retrieval and streaming heads from the spectral geometry of query-key projections. AoH defines the kernel attention operator M_h = W_K^h\topW_Q^h and uses its effective-rank as a weight-space measure of head function: concentrated spectra indicate a small number of dominant query-key matching directions and are associated with retrieval heads, whereas diffuse spectra indicate the absence of a dominant global matching direction and are associated with streaming heads. We further derive an efficient d_\texthead -dimensional computation that avoids constructing the full d_\textmodel\times d_\textmodel matrix. We conducted extensive experiments across models demonstrating that at 50% sparsity, AoH retains 96.5% of Full Attention performance on average while reducing prefill and decode latency by up to 41.4% and 66.0%, respectively, and KV-cache memory by 50.0% at 256K tokens.
[NLP-31] FutureBridge: Token Selection Beyond Local Preference in Collaborative Decoding
【速读】： 该论文旨在解决大语言模型（LLM）与小语言模型（SLM）在协同推理过程中因预测分歧导致的效率与准确性问题。现有方法依赖于LLM的局部偏好进行干预或候选词排序，但忽视了所选标记是否真正有助于SLM后续的连贯推理，从而限制了协同效果。其解决方案的关键在于提出FutureBridge框架，通过构建一个由答案验证过的LLM轨迹提供的固定共享未来，利用冻结的SLM在统一上下文中评估每个联合候选标记对后续推理的支持能力，进而生成反事实评分以监督一个轻量级的令牌重排序器。该重排序器仅需感知当前状态和候选标记即可决策，避免了生成或附加未来后缀。在五个数学推理基准测试中，FutureBridge使Qwen3-1.7B SLM的数学平均性能相对贪婪解码提升了35.1%，证明了基于SLM可延续性的令牌选择机制相较于单纯依赖LLM局部偏好更具优势。
链接: https://arxiv.org/abs/2608.06819
作者: Quanquan Li,Hongbo Zhang,Yihe Chi,Jingyu Li,Xidong Xi,Liuyang Song,Hongzhen Zhang,Yuxiang Huang,Jing Ke,Siyuan Ma,Junyi Lin,Guitao Cao
机构: Google(谷歌); Stanford University (斯坦福大学); Tsinghua University (清华大学); Peking University (北京大学); University of California, Berkeley (加州大学伯克利分校); MIT (麻省理工学院)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Token-level collaboration allows a large language model (LLM) to assist a small language model (SLM) when their predictions diverge. Existing methods either use LLM-generated intervention tokens or rank candidates with the LLM’s next-token probabilities. Both rely on the LLM’s local preference, even though an LLM-selected token may be difficult for the SLM to build on. We present FutureBridge, which ranks joint LLM-SLM token candidates according to how well they support the SLM’s subsequent reasoning. During training, an answer-verified LLM trajectory supplies a fixed shared future, and a frozen SLM evaluates every candidate under this common context. The resulting counterfactual scores supervise a lightweight token reranker that observes only the current state and candidate token. At inference, FutureBridge uses the LLM only to expand the candidate pool, selects one token, and returns generation to the SLM without generating or appending a future suffix. Across five mathematical reasoning benchmarks, FutureBridge improves the Qwen3-1.7B SLM’s Math Avg. by 35.1% relative to greedy SLM decoding. These results indicate that token selection benefits from modeling whether the receiving SLM can use each candidate to continue reasoning, rather than relying on the LLM’s local preference alone.
[NLP-32] Simple-OPD: Demystifying Warm-up for On-policy Distillation
【速读】： 该论文旨在解决在线策略蒸馏（On-policy Distillation, OPD）中对预热阶段（warm-up）依赖性强、效果不稳定的问题。其核心挑战在于：OPD的性能高度受预热阶段数据质量与训练策略的影响，而现有方法缺乏对预热机制的深入理解。解决方案的关键在于从数据与训练双视角重新审视预热过程：在数据层面，发现有效的预热依赖于教师兼容的思维链（chain-of-thought, CoT）监督，即使教师生成的推理路径存在错误，也能提供与正确路径相当的增益，表明预热的本质是迁移教师兼容的思维模式而非单纯复制正确答案；在训练层面，提出采用低秩适应（Low-rank Adaptation, LoRA）并配合近饱和训练时长，相较于全参数微调（SFT），能更优地平衡领域内适应性与域外泛化能力。基于上述发现，论文提出Simple-OPD——一种即插即用的初始化方法，在正式OPD前利用LoRA对学生模型进行教师生成的CoT预热，显著提升了OPD在多种场景下的有效性与鲁棒性。
链接: https://arxiv.org/abs/2608.06802
作者: Tao Liu,Taiqiang Wu,Mao Zheng,Xuan Luo,Runming Yang,Xuewei Yang,Junjie Wang,Yujiu Yang
机构: Tsinghua University(清华大学); The University of Hong Kong(香港大学); LLM Department, Tencent(腾讯大模型部门)
类目: Computation and Language (cs.CL)
备注:
Abstract:On-policy distillation (OPD) trains a student on its own rollouts with token-level supervision from teacher models, but its effectiveness can depend strongly on the warm-up stage before OPD. In this paper, we demystify warm-up for OPD from both data and training perspectives. For data, we find that effective warm-up relies on teacher-compatible chain-of-thought supervision, and that even incorrect teacher rollouts can provide comparable benefits to correct ones. This suggests that warm-up primarily transfers a teacher-compatible thinking pattern rather than merely correct answers. For training, we show that low-rank adaptation (LoRA) with a near-saturation training duration better balances in-domain adaptation and out-of-distribution generalization than full-parameter SFT. Based on these findings, we propose Simple-OPD, a plug-and-play initialization method that warms up the student on teacher-generated CoT with LoRA before OPD. Experiments across diverse settings demonstrate the effectiveness and robustness of Simple-OPD.
[NLP-33] LoRAScan: Detecting Backdoor Prompts in Low-Rank Adapters for Large Language Models via Down-Projection Activation Spikes
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在使用低秩适配器（LoRA）进行高效微调时所面临的供应链安全威胁，即恶意适配器可能嵌入后门，当输入中包含特定触发词时，导致模型生成有害内容、恶意代码或隐蔽广告。现有防御方法存在明显局限：适配器无关的防御通过合并适配器与基座模型来稀释后门信号，降低了检测性能；而已有适配器感知的方法要么针对基座模型修复后门（解决逆问题），要么依赖外部分类器对整个适配器标记为可疑，缺乏对触发输入的细粒度识别能力。本文提出LoRAScan，作为首个适配器感知的推理时防御机制，其核心创新在于发现并利用后门适配器中仅约5%的插入位置在正常输入下具有低方差特征，但在触发输入下会表现出高度集中的低秩投影激活峰值。该方法在模型部署前识别这些稳定且敏感的插入点，并在推理阶段持续监控其激活模式，从而在不修改适配器参数的前提下，精准检测并拒绝携带触发词的恶意输入。在标准LLM后门基准测试中，LoRAScan实现了约98.49%的恶意输入拒识率，同时保持较低的清洁输入误报率，显著优于现有各类防御方案。
链接: https://arxiv.org/abs/2608.06795
作者: Doniyorkhon Obidov,Honggang Yu,Xiaolong Guo,Kaichen Yang
机构: 未知
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Low-rank adaptation (LoRA) enables efficient specialization and distribution of large language models through compact adapters. However, untrusted adapters introduce a supply-chain threat: a backdoored adapter can cause a model to generate harmful content, malicious code, political propaganda, or covert advertisements when an input contains a hidden trigger. Adapter-agnostic defenses merge the adapter with the base model, which dilutes backdoor signals and reduces detection performance. Existing adapter-aware methods do not address how to safely use a potentially backdoored adapter. Instead, they either train a defensive adapter to repair a backdoored base model, addressing the inverse problem rather than securing the adapter itself, or rely on a classifier that flags the entire adapter as suspicious and requires separate mitigation. These methods overlook the distinct latent-space signatures produced by trigger-bearing inputs in backdoored adapters. We introduce LoRAScan, the first adapter-aware defense that detects and rejects trigger-bearing inputs at inference time without modifying adapter parameters. Our key observation is that a small subset of LoRA insertion sites, approximately 5%, remains stable across clean inputs but exhibits highly concentrated spikes in LoRA down-projection activations when a trigger is present. LoRAScan identifies these low-variance insertion sites before model deployment and monitors them during inference. Across standard LLM backdoor benchmarks, LoRAScan rejects approximately 98.49 of malicious inputs with a small error rate on clean inputs, outperforming existing defenses across diverse evaluation settings. Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL) Cite as: arXiv:2608.06795 [cs.CR] (or arXiv:2608.06795v1 [cs.CR] for this version) https://doi.org/10.48550/arXiv.2608.06795 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[NLP-34] Multi-Perspective Triad Interaction Graph Neural Network for Cognitive Distortion Detection
【速读】： 该论文旨在解决计算心理健康领域中认知扭曲检测任务存在的关键问题：现有方法普遍忽视了认知扭曲在心理结构上的复杂性，尤其是未能充分建模贝克认知三角（Beck’s cognitive triad）——即个体对自我、世界和未来的负面认知——之间的多维关联。其解决方案的核心在于提出一种名为MTI-GNN（多视角三联互动图神经网络）的新框架，通过将认知三角的三个维度作为互补视角进行建模，实现更精准的认知扭曲分类。该方法首先利用大语言模型（LLM）将每段语句分解为自我、世界、未来三个视角，并构建各视角特定的相似性图；随后通过多视角图神经网络（Multi-Perspective GNN）编码这些图结构信息。进一步地，三联互动模块（Triad Interaction）采用序列化源条件更新与特征门控机制，显式捕捉跨视角依赖关系；原型引导的视角融合模块（Prototype-Guided Perspective Fusion）则基于标签条件聚合不同视角的表示。此外，引入标签扩展监督（label-expanded supervision）以充分利用所有可用的扭曲标注信息。实验在涵盖十类扭曲类型的四组韩语、英语和中文数据集（共9,764样本）上验证了MTI-GNN的有效性，结果表明其显著优于所有监督基线方法，并在零样本和少样本设置下超越八种提示式生成模型。消融实验显示，三个认知视角均对性能有显著贡献，且人工专家评估初步证实了各视角与预期心理维度的良好对齐。
链接: https://arxiv.org/abs/2608.06785
作者: Jun Seo Kim,Hye Hyeon Kim
机构: Gachon University(加丘大学); Yonsei University(延世大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Cognitive distortion detection is a key task in computational mental health, yet existing approaches often overlook the psychological structure of distorted thoughts. We propose MTI-GNN (Multi-Perspective Triad Interaction Graph Neural Network), which models Beck’s cognitive triad—negative views of the self, world, and future—as complementary perspectives for classification. An LLM decomposes each utterance into the three perspectives, from which perspective-specific similarity graphs are constructed and encoded by a Multi-Perspective GNN. A Triad Interaction module models cross-perspective dependencies through sequential source-conditioned updates and feature-wise gating, while Prototype-Guided Perspective Fusion performs label-conditioned aggregation. Label-expanded supervision incorporates all available distortion annotations during training. We evaluate MTI-GNN on 9,764 samples from four Korean, English, and Chinese datasets spanning ten distortion categories. MTI-GNN significantly outperforms all supervised variants and exceeds eight prompted generative models under zero-shot and few-shot settings. Leave-one-perspective-out ablations show that all three perspectives contribute significantly, while human expert evaluation provides preliminary evidence of their alignment with the intended cognitive dimensions.
[NLP-35] Retrieval-Constrained Policy Optimization for Attack Technique Extraction from Cyber Threat Intelligence KDD2026
【速读】： 该论文旨在解决网络威胁情报（CTI）文本到MITRE ATTCK攻击技术的自动化映射问题，其核心挑战在于ATTCK知识库包含数百个攻击技术，单篇CTI文本可能涉及多个技术，导致准确且完整的多标签提取极为困难。现有方法存在明显局限：多标签分类器在严重类别不平衡和大规模标签空间下性能受限；基于大语言模型（LLM）的方法（如检索增强与微调生成器）通常优化的是词元级目标，将技术标注视为序列生成任务，缺乏对预测技术集合的完整性与正确性的直接监督。为此，论文提出TTP-R1，一种两阶段框架，结合检索增强的监督微调（SFT）与可验证奖励强化学习（RLVR）。首先通过混合检索器将大规模标签空间压缩至候选集，再利用微调的LLM进行精准技术选择；随后采用分解奖励的组相对策略优化（Group Relative Policy Optimization），直接以精确率、召回率及输出格式作为监督信号，实现对技术集合整体质量的优化。在四个CTI基准测试中，TTP-R1实现了最优平均F1值，子技术层级F1较带检索增强的Claude Sonnet 4.5提升7.4个百分点，且在单张GPU上以80亿参数模型部署时推理速度提升28倍。
链接: https://arxiv.org/abs/2608.06778
作者: Jiayun Zhang,Junshen Xu,Zejun Xie,Yi Fan
机构: Amazon Web Services(亚马逊网络服务); Rutgers University(罗格斯大学)
类目: Cryptography and Security (cs.CR); Computation and Language (cs.CL)
备注: Accepted to the AI Agent for Information Retrieval (Agent4IR) Workshop at KDD 2026
Abstract:Mapping cyber threat intelligence (CTI) text to MITRE ATTCK techniques is essential for structured threat analysis, yet manual annotation is costly and does not scale. The ATTCK taxonomy comprises several hundred attack techniques, and a single CTI passage may describe multiple techniques, making accurate and complete extraction challenging. Existing automated approaches fall short in different ways: multi-label classifiers struggle with severe class imbalance and the large label space, while LLM-based methods–retrieval pipelines and fine-tuned generators–optimize token-level objectives that treat technique annotation as sequence generation rather than set prediction, lacking direct supervision on whether the predicted technique set is correct and complete. We propose TTP-R1, a two-stage framework that combines retrieval-augmented supervised fine-tuning (SFT) with reinforcement learning using verifiable rewards (RLVR). A hybrid retriever first narrows the large label space to a candidate set, and a fine-tuned LLM learns to select the correct techniques. We then apply Group Relative Policy Optimization with a decomposed reward that directly supervises the precision, recall, and output format of the predicted technique set. Across four CTI benchmarks, TTP-R1 achieves the best average F1, improving sub-technique-level F1 by 7.4 percentage points over Claude Sonnet 4.5 with retrieval augmentation, while running 28x faster when served as an 8B-parameter model on a single GPU.
[NLP-36] Stockmark-Nemotron-3-Nano-Omni-JapanDocReader: Structured Document Parsing via Capability Injection and Forgetting Control
【速读】： 该论文旨在解决在保持多模态模型原有文档问答（Document VQA）能力的前提下，如何有效注入日文结构化文档解析能力的问题。其核心挑战在于避免因能力注入导致的“遗忘”现象，即模型在学习新任务时对原有能力的退化。解决方案的关键在于通过分阶段的微调策略与强化学习优化：首先采用以解析为中心的监督微调（SFT），显著提升结构化文档解析性能；随后引入混合式SFT，融合结构化解析与VQA数据，有效缓解了模型对原始VQA能力的遗忘；最终在混合SFT基础上应用基于动态奖励分配与惩罚优化（DAPO）的解析导向强化学习（RL），进一步突破传统SFT的性能上限，实现更优的结构化文档解析效果。此外，研究构建了由日文文档问答流与程序化结构化解析生成流组成的双轨合成数据引擎，并提出基于方差的提示过滤机制与任务级奖励设计，以增强长推理链场景下强化学习的有效性，确保训练过程稳定且可扩展。
链接: https://arxiv.org/abs/2608.06758
作者: Shi Chen,Hayato Aida,Makoto Morinaga,Shohei Tanaka,Kosuke Arima
机构: Stockmark Inc.
类目: Computation and Language (cs.CL)
备注:
Abstract:We present Stockmark-Nemotron-3-Nano-Omni-JapanDocReader, a Japanese document understanding model built from Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16. The central goal of this work is structured document parsing via capability injection and forgetting control: we inject Japanese structured document parsing capability into a reasoning-oriented multimodal model while preserving its document VQA capability as much as possible. We study parsing-centric SFT, which uses only structured document parsing data; mixed SFT, which combines structured document parsing and VQA data; and parsing-centric RL, which optimizes structured parsing with a task-level reward. Our experiments show that parsing-centric SFT substantially improves structured document parsing performance but causes measurable VQA forgetting. Mixed SFT mitigates this forgetting while preserving nearly the same structured parsing performance. Applying DAPO-based parsing-centric RL on top of the mixed SFT checkpoint further improves structured document parsing beyond the SFT ceiling, producing the final released model. The training data is constructed with a data engine consisting of two complementary synthetic streams: a Japanese Document VQA Stream and a programmatic structured document parsing stream. We also discuss reward design and variance-based prompt filtering for continuous structured document parsing rewards, highlighting their importance for making RL effective in long-reasoning structured document parsing tasks.
[NLP-37] Mind the Gap: A Dual Knowledge Graph Framework for Unified Multi-task User Intent Inference
【速读】： 该论文旨在解决在线旅游评论中多任务用户意图推断的难题，现有方法普遍依赖于易受误差传播影响的分层流水线架构，或采用忽略领域知识结构关系的检索式方法。其解决方案的关键在于提出一种仅依赖推理的知识增强框架——DKG-MTI，通过动态构建用户特定的意图知识图谱（User-Specific Intent Knowledge Graph），并利用结构感知的语义平滑技术将其与全局酒店知识图谱（Global Hotel Knowledge Graph）进行对齐，从而实现结构化知识的有效融合。该对齐后的知识与原始评论共同输入大语言模型（LLM），以并行完成方面评分预测与逆向用户意图陈述生成。实验结果表明，该方法在TripAdvisor数据集上显著优于多种强基线模型，在分类与意图生成任务中均展现出卓越性能，验证了结构感知知识对齐在可扩展且可解释的用户意图推断中的有效性。
链接: https://arxiv.org/abs/2608.06752
作者: Tzu-Cheng Peng(1),Chien Chin Chen(1),Chih-Hao Ku(2),Yung-Chun Chang(3) ((1) National Taiwan University, (2) University of North Texas, (3) Taipei Medical University)
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: Published in the PACIS 2026 Proceedings as a Completed Research Paper. AIS eLibrary: this https URL 17 pages, 5 figures
Abstract:This paper proposes DKG-MTI, a dual knowledge graph framework for unified multi-task user intent inference from online travel reviews. Existing approaches often rely on hierarchical pipelines that suffer from error propagation or retrieval methods that ignore structural relationships in domain knowledge. To address these limitations, we introduce an inference-only knowledge augmentation framework that dynamically constructs a User-Specific Intent Knowledge Graph from each review and aligns it with a Global Hotel Knowledge Graph through structure-aware semantic smoothing. The aligned knowledge is combined with the original review and processed by a large language model to simultaneously predict aspect ratings and generate reverse user intent statements. Experiments on TripAdvisor reviews show that DKG-MTI consistently outperforms strong LLM and retrieval-based baselines in both classification and intent generation tasks, demonstrating the effectiveness of structure-aware knowledge alignment for scalable and explainable intent inference.
[NLP-38] Progressive Content Refinement with Decaying Reward Joint LinUCB
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）在迭代优化过程中因现有方法依赖静态提示（prompt）或动作（arm）而引发的过度利用（over-exploitation）问题，其核心在于忽视了奖励随时间衰减的饱和效应。传统反馈机制如Self-Refine或经典强化学习中的上下文多臂老虎机（contextual bandit）方法通常假设每次使用相同提示可获得稳定回报，但实际中频繁重复使用同一提示会导致性能收益递减。为此，作者提出一种新型上下文多臂老虎机算法，通过显式建模奖励衰减机制来缓解这一问题。该方法采用期望最大化（Expectation-Maximization, EM）算法，联合估计每个动作（即提示）的特定价值参数与全局奖励衰减参数，从而实现对动态变化的奖励信号的精准捕捉。此外，将提示作为动作进行嵌入表示，实现了动作价值与衰减特性的联合学习，突破了传统线性上置信界（Linear Upper Confidence Bound, LinUCB）框架中动作价值与环境状态分离的局限性。实验结果表明，在情感反转（Sentiment Reversal）和GSM8K数学推理任务上，所提方法显著优于多个强基线模型；消融实验证实，将奖励衰减建模引入多臂老虎机框架是抑制过度利用、提升迭代优化效率的关键所在。
链接: https://arxiv.org/abs/2608.06750
作者: Shion Ishikawa,Pablo Loyola,Young-joo Chung,Yun Ching Liu
机构: Rakuten Group, Inc.
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Iterative refinement has significantly enhanced Large Language Model (LLM) performance; however, existing methods ranging from feedback-based Self-Refine to traditional bandit approaches often rely on static options or overlook the saturation effect. This neglect leads to over-exploitation, where the continuous use of identical prompts or arms results in diminishing rewards over time. To address this challenge, we propose a novel contextual bandit algorithm that explicitly incorporates reward decay modeling. Utilizing an Expectation-Maximization (EM) algorithm, our method simultaneously estimates both arm-specific and decay parameters. Furthermore, by embedding prompts as arms, we facilitate the joint learning of arm values, distinguishing our approach from the traditional disjoint Linear Upper Confidence Bound (LinUCB) framework. Experimental results on Sentiment Reversal and GSM8K benchmarks demonstrate that our method achieves significant performance gains over strong baselines. Finally, our ablation study confirms that the integration of reward decay modeling within the bandit framework is crucial for mitigating over-exploitation and optimizing the iterative refinement process. Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI) Cite as: arXiv:2608.06750 [cs.CL] (or arXiv:2608.06750v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2608.06750 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[NLP-39] IB-RL: Isolated Bilateral Reinforcement Learning for Strategic Dialogue Agents
【速读】： 该论文旨在解决强化学习（Reinforcement Learning, RL）在策略性对话任务中因训练环境固定而导致的泛化能力不足问题。传统RL方法通常将目标智能体与一个固定的对手或模拟器进行训练，这种静态对手设定导致策略倾向于利用特定对手的可预测模式，而非学习具有普遍适应性的交互策略。这一现象被称为“静态对手不匹配”（static-counterpart mismatch），本文通过实验直接量化了该问题。为解决此问题，论文提出孤立双边强化学习（Isolated Bilateral Reinforcement Learning, IB-RL），其核心在于让两个角色通过联合采样（joint rollouts）实现共进化，同时每个角色在优势函数（advantage）、动作掩码（action mask）和参数更新路径上保持完全独立，从而确保双方策略在训练过程中互不干扰、各自优化。实验结果表明，在车辆电话销售（Vehicle TeleSales）和讨价还价（Deal-or-NoDeal）两个任务上，IB-RL分别达到89.6%和98.4%的性能表现，显著优于最佳单边强化学习基线，验证了严格隔离下双角色协同训练能够生成对未见对手更具泛化能力的策略。
链接: https://arxiv.org/abs/2608.06735
作者: Senhao Wang,Chenghao Cai,Haitao Hu,Mingxing Huang,Xingguang Wang,Wenhao Li,Zecheng Lin
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Reinforcement learning (RL) has achieved strong results in improving large language models (LLMs) on tasks with stationary, verifiable rewards, such as mathematical reasoning and code execution. In these settings, the environment follows fixed rules and does not adapt strategically to the agent. Strategic dialogue differs in this respect: the environment is another agent that adapts to the policy, and success depends on the interaction between the two sides. Despite this interactive nature, current RL approaches typically train a target agent against a fixed counterpart or simulator. We find that this training paradigm encourages the policy to exploit counterpart-specific regularities rather than learn strategies that generalize across counterparts. We call this problem the static-counterpart mismatch, which we quantify directly in our experiments. To address it, we propose Isolated Bilateral Reinforcement Learning (IB-RL), in which the two roles coevolve through joint rollouts while each role optimizes its own reward through fully independent advantages, action masks, and update paths. We evaluate frozen policies against fully independent held-out counterparts in both domains. On Vehicle TeleSales, IB-RL achieves 89.6% Success@1, compared to 84.6% for the best unilateral RL baseline. On Deal-or-NoDeal, it reaches 98.4% agreement against DeepSeek V4 Pro, compared to 86.4% for the best unilateral baseline. These results indicate that jointly training both roles with strict peragent isolation produces policies that generalize more effectively to unseen counterparts.
[NLP-40] Do Audio Language Models Use Paralinguistic Evidence? Counterfactual Audits for Response Evaluation
【速读】： 该论文旨在解决当前语音到语音系统评估中，音频-语言模型（Audio-Language Models, ALMs）作为评判者时可能依赖文本内容或响应风格而非实际的副语言线索（paralinguistic cues），从而导致评估结果不可靠的问题。其核心解决方案是引入反事实审计（counterfactual audits）机制，通过在保持语义转录不变的前提下，系统性地改变情感表达、语调（prosody）或情感转变的时间点，强制模型关注音频中的副语言特征而非词汇内容或响应模式。研究采用单上下文原生判断协议与对比可恢复性控制实验，并进一步将每个评估项分解为感知能力与响应映射能力两个子技能，从而实现对模型判别失败根源的精细化诊断。结果显示，传统基于准确率的评估常高估原生法官的可靠性，且相似的整体准确率可能掩盖不同的失效模式。因此，研究强调：在部署前，必须对ALM评判者进行深入的行为审计，而不能仅依赖单一的准确性指标。
链接: https://arxiv.org/abs/2608.06718
作者: Kevin Miller,Arjun Chandra,Venkatesh Saligrama
机构: Boston University (波士顿大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Audio-language models (ALMs) are increasingly used as judges for speech-to-speech systems, but a judge that receives audio may not actually use paralinguistic evidence. We introduce counterfactual audits for paralinguistic response evaluation. Each audit item holds the transcript fixed while varying affect, prosody, or the timing of an affective shift, forcing a valid judge to track the audio cue rather than lexical content or response style. We evaluate ALM judges using a native one-context judgment protocol and a contrastive recoverability control, then further decompose each item into its constituent perception and response-mapping skills. This yields useful diagnostic states that identify different sources of judge failures. Across Gemini, GPT, and open audio models, we find that contrastive success often overstates native judge reliability, and that similar aggregate accuracies can hide different failure modes. These results suggest that ALM judges should not be evaluated by accuracy alone, instead requiring thorough behavioral audits before deployment.
[NLP-41] Online Monitoring and Corrective Steering of Programming Agents
【速读】： 该论文旨在解决大规模开源项目中修复GitHub问题时，因任务周期长、修复涉及多位置变更或问题描述信息不足而导致的智能代理（agent）行为低效与偏差问题。具体表现为：代理在执行过程中容易偏离原定计划、重复失败操作，或在未生成有效补丁的情况下提前终止。针对这一挑战，论文提出LivePlan，其核心解决方案在于实现“判断”与“建议”的解耦设计：采用基于规则的确定性监控器实时分析代理轨迹中的通用信号以检测行为异常，仅在检测到问题时才调用大语言模型（LLM）作为顾问进行高层次修正建议。该机制避免了传统方法中频繁调用大模型导致的误导性重规划和高成本干预。实验结果表明，在SWE-agent框架上集成LivePlan后，相较于原始版本，其在SWE-bench Verified和SWE-bench Pro基准上的问题修复率显著提升，平均提升达9.9%，最高可达15.2%；且额外开销仅为每实例0.08，主要贡献集中在中等和高难度问题上。此外，LivePlan在各类评估场景下均优于现有基线方法，对已有成功案例影响极小，并能解决部分基线无法处理的新难题。
链接: https://arxiv.org/abs/2608.06701
作者: Shuyang Liu,Saman Dehghan,Ji Young Kim,Jatin Ganhotra,Martin Hirzel,Reyhaneh Jabbarvand
机构: 未知
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Fixing GitHub issues in large-scale projects is a long-horizon task, especially when a fix requires changes across multiple locations or the issue description lacks the information needed to localize and repair it. As a result, agents traverse long trajectories that are prone to inefficiency and error: they drift away from their intended plan, repeat failed actions, or terminate without a working patch. This paper proposes LivePlan to monitor, detect, and correct such behavioral inefficiencies and drifts in real time. LivePlan decouples judging from advising: a deterministic, rule-based monitor examines general signals over the trajectory to detect issues without invoking an LLM, and only when an issue is detected does it consult an advisor LLM for a high-level, next-step correction. This design avoids the misleading re-planning and costly interventions of prior approaches. We implement LivePlan on top of SWE-agent and evaluate it using five LLMs (three as executor agents and two as advisors) across SWE-bench Verified and SWE-bench Pro. Compared to vanilla SWE-agent, LivePlan notably improves issue resolution rates, achieving consistent gains of up to 15.2% (average: 9.9%), while incurring only an additional cost of 0.08 per instance. The additional solutions concentrate on medium and hard instances. LivePlan consistently outperforms alternative approaches in resolution rate, with minimal regression on already successful runs and new successes on problems that no baseline solves.
[NLP-42] A-RAG : Tone Awareness as a Design Imperative for Retrieval-Augmented Generation
【速读】： 该论文旨在解决标准检索增强生成（Retrieval-Augmented Generation, RAG）系统在实际应用中因检索文档固有语篇风格（如专业术语、正式语气或学术化表达）导致的“语境脱节”问题，即系统虽能保证事实准确性，却忽视用户对特定语调或社交语境的需求，从而造成沟通失效。其核心挑战在于现有评估体系过度聚焦于事实正确性，而忽略了语言表达在语用层面的适配性，导致三类沟通错配——语言层面、认知层面与关系层面——难以被发现和纠正。为此，论文提出一种名为Tone-Aware RAG（TA-RAG）的概念架构，将“沟通一致性”作为与事实准确性同等重要的设计目标，并在检索、上下文构建、生成及约束验证四个阶段中，系统性地施加四类关键约束：去污名化语言、可读性对齐、受体敏感性适应以及共情式表述。该方案强调需建立联合评估框架，同时衡量事实保真度与沟通一致性，以应对高敏感性社会场景下生成内容的有效性与伦理要求。研究主张，语调感知不应被视为可选优化项，而应成为面向社会敏感与高风险应用场景的RAG系统不可或缺的设计前提。
链接: https://arxiv.org/abs/2608.06672
作者: Yong-Bin Kang,Anthony McCosker
机构: Swinburne University of Technology (斯威本科技大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Retrieval-Augmented Generation (RAG) has become a robust architecture for grounding large language models (LLMs) in trusted knowledge. However, standard RAG systems exhibit a structural limitation: retrieved documents carry their own communication styles-professional jargon, formal tone, or academic writings-that shape the behavior of a RAG system before any tone instructions are processed, often causing the system to ignore user requests for a specific tone. We term this phenomenon contextual decoupling, in which a system optimises for factual accuracy while remaining decoupled from the social or operational context of the recipient. Building on prior research in public health peer-support communities, we identify three communicative misalignment-linguistic, cognitive, and relational-that can persist even when retrieval is relevant and the generated response is factually accurate. We conceptualise these as failures of communicative transformation, which remain largely invisible to accuracy-centred RAG evaluation metrics. To address this gap, we propose Tone-Aware RAG (TA-RAG), a conceptual architectural framework that positions communicative alignment alongside factual accuracy as a core design objective. TA-RAG operationalises four constraints-stigma-free language, readability alignment, recipient-sensitive adaptation, and empathetic framing-across the retrieval, context construction, generation, and constraint validation phases in the proposed RAG pipeline. We further highlight an evaluation agenda for jointly assessing factual fidelity and communicative alignment, and identify open challenges. We argue that tone awareness should be treated not as an optional refinement, but as a present design imperative for RAG systems operating in socially sensitive and high-stakes contexts.
[NLP-43] he Horizon Gap: Planning Memory Execution Training and Evaluation for Long-Horizon LLM Agents
【速读】： 该论文旨在解决大模型在处理长时序任务（long-horizon tasks）时存在的“视野缺口”（horizon gap）问题，即模型在执行需数小时甚至更长时间的任务时，容易丢失早期决策、误判任务完成状态或偏离目标。其核心挑战在于：随着任务时长增加，仅依赖最终结果（outcome-only）的信号逐渐失效，导致模型无法有效追踪和调整复杂推理过程。解决方案的关键在于重构对长时序任务的理解框架，明确区分三个常被混淆的概念——长时程性（long-horizon，任务属性）、长上下文（long-context，模型属性）与长期记忆（long-term memory，系统属性），并构建一个涵盖任务生命周期六个阶段（规划、记忆、执行、训练、评估、基础与安全）的分类体系，同时以“视野延续方式”（within-context、within-task-beyond-context、cross-task-persistent）为轴线进行交叉分析。研究发现，当前领域应对策略均聚焦于生成更密集的步骤级信号（如过程奖励模型、信用分配机制、轨迹级诊断），以弥补结果信号的退化；论文进一步主张将批判性与方法性文献视为并列主线，避免割裂同一研究工作的完整性，并指出若干开放性测量难题，包括模型能力与利用能力的解耦、过程级信号中相关偏差的管理，以及长时程可靠性是否可建立通用预测理论等关键科学问题。
链接: https://arxiv.org/abs/2608.06663
作者: Mingguang Chen,Licheng Wang,Bo Qu
机构: DeepGrounding(深地); AlphaAvatar
类目: Computation and Language (cs.CL)
备注: 39 pages, 6 figures
Abstract:Frontier language models solve reasoning problems in a single forward pass that would have been research contributions years ago, yet fail at multi-hour tasks: losing track of earlier decisions, declaring half-finished work done, or drifting from goals. We call this the horizon gap and survey 1,547 arXiv papers (2024-2026) collected via systematic seed harvest with a disclosed 26.8% bleed filter, extended by targeted supplementation. We disambiguate three routinely conflated properties: long-horizon (task property: required steps), long-context (model property: token capacity), and long-term memory (system property: persistence across steps/sessions). We organize the corpus into six categories tracking a long-horizon task’s lifecycle – planning, memory, execution, training, evaluation, and foundations/safety – crossed with an axis capturing where horizons are carried (within-context, within-task-beyond-context, or cross-task-persistent). Across all categories, we find the same pattern: outcome-only signals grow uninformative as horizons lengthen, and the field’s response – whether process reward models, credit assignment, or trajectory-level diagnostics – manufactures denser step-level signals. We treat critical and diagnostic literature as first-class threads throughout, arguing that segregating critique from method would routinely split single papers across chapters. We close by naming open measurement problems: decomposing model versus harness capability, managing correlated bias in process-level signals used for both training and evaluation, and whether long-horizon reliability admits general predictive theory.
[NLP-44] Discovering Conceptual Metaphors Across Topics and Media Types NAACL2027
【速读】： 该论文旨在解决如何识别语言中隐含的概念隐喻（conceptual metaphor）这一核心问题，进而揭示话语背后的认知框架与立场差异。其关键在于提出一种无监督方法，通过从语料库中提取语言隐喻（linguistic metaphors）并采用结构化聚类技术，将相关表达归类为对应的概念隐喻，从而间接推断出说话者或作者的认知框架。该方法成功揭示了左翼与右翼播客在议题建构上的显著差异：例如，左翼播客倾向于将媒体叙事概念化为“武器”，而右翼播客则更常将经济描述为受垂直变化影响的“系统”，体现了不同的认知取向与政治立场。
链接: https://arxiv.org/abs/2608.06652
作者: Alexandria Leto,Rohan Das,Juan Vásquez,Abram Handler,Maria Leonor Pacheco
机构: University of Colorado Boulder (科罗拉多大学博尔德分校)
类目: Computation and Language (cs.CL)
备注: 49 pages, submitted to NAACL 2027 for review
Abstract:Conceptual metaphors guide our thinking and actions by allowing us to reason about more abstract experiences (e.g., paying taxes) in terms of more concrete or embodied experiences (e.g., carrying a physical load) (Lakoff and Johnson, 2011). It follows that different conceptual metaphors can result in different reasoning: framing paying taxes as an investment in a community rather than a physical load leads to a very different outlook on taxation. Identifying the conceptual metaphors guiding a speaker or writer thus helps to reveal their framing of events. Though these metaphors can’t be observed directly, groups of linguistic metaphors, metaphorical expressions as they appear in language, serve as evidence for them. Motivated by this, we present an unsupervised method that extracts linguistic metaphors from a corpus and uses a structured clustering approach to form groups corresponding to conceptual metaphors. Using this method, we point to key topical and framing differences in left- vs. right-leaning podcasts. For example, left-leaning podcasts tend to conceptualize media stories as a weapon, while right-leaning sources commonly discuss the economy as a system subject to vertical changes.
[NLP-45] Factorized Hypothesis Search for Evidence-to-Taxonomy Retrieval
【速读】： 该论文旨在解决大规模分类体系（large-taxonomy）检索中因输入信息间接性导致的语义表达不充分问题，即“检索就绪差距”（retrieval readiness gap）。在实际场景中，输入常表现为依赖上下文的间接证据（如表格单元格），其语义需结合行、列、数据类型及上下文才能理解，而现有索引系统在面对此类隐含语义时召回能力显著下降。为此，论文提出因子化假设搜索（Factorized Hypothesis Search, FHS）方法，其核心在于在多个命名语义维度上维护并行的局部解释假设，支持结构化查询生成、多假设检索与维度级候选验证。实验表明，在金融分类标签与CodiEsp临床编码任务中，FHS在非理想（non-oracle）方法中取得了最优的Recall@1、MRR和最终准确率。消融分析显示，以自由文本集成替代因子化假设路径导致头部排名性能最大降幅，且序列化精炼无法超越FHS在并行第一轮中的优异表现，凸显了其并行多维度假设建模的关键优势。
链接: https://arxiv.org/abs/2608.06614
作者: Linhai Ma,Ethan F. Wei,Xueqing Peng,Yan Wang,Lingfei Qian,Víctor Gutiérrez-Basulto
机构: The Fin AI, USA; Yale University, USA; Cardiff University, UK
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 28 pages, 1 figure, 28 tables
Abstract:Large-taxonomy retrieval often assumes that the input already expresses the target concept. In many settings, however, the input is indirect evidence, such as a table cell whose meaning depends on its row, column, datatype, and context. We call this mismatch the retrieval readiness gap. Our analysis shows that the current index retrieves the target reliably when its semantics are explicit, while raw evidence often leaves it deep in the ranking. We propose Factorized Hypothesis Search (FHS), which maintains multiple partial interpretations over named semantic dimensions. These hypotheses support structured query rendering, multi-hypothesis retrieval, and dimension-level candidate verification. On both financial taxonomy tagging and CodiEsp clinical coding tasks, FHS achieves the best Recall@1, MRR, and final accuracy among the non-oracle methods. Replacing the factorized hypothesis path with a free-text ensemble causes the largest drop in head-ranking performance, while sequential refinement provides no additional gain over FHS’s strong parallel first round.
[NLP-46] Beyond “AI Language”: The case for the idiolectal nature of LLM output
【速读】： 该论文旨在解决当前将大语言模型（Large Language Model, LLM）输出统一视为“人工智能语言”这一整体化视角所忽视的模型特异性语言特征问题。研究表明，尽管不同模型在宏观层面可能表现出相似的语言模式，但每个模型仍保有独特的语言风格特征，类似于人类的个体语体（idiolect）。其解决方案的关键在于通过计算描述符与文体主成分分析（stylometric principal component analysis），揭示了2024年与2026年两代模型之间在语言风格上的代际演变，并证实同一时期内各模型间存在显著的语言差异，例如在2026年数据集中，缩略语使用频率在每百万词中可从1,200至30,000不等。研究进一步提出，将LLM输出视为具有“语体性”的语言产物，为语言变异与演变研究、生成式文本识别、司法语言学以及基于用法的语言理论提供了新的分析框架。
链接: https://arxiv.org/abs/2608.06589
作者: Karolina Rudnicka,Thomas Stephan Juzek
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 33 pages, 6 figures, 6 tables. Submitted as a chapter to the post-workshop volume “Corpus Linguistics 2040” (Digital Linguistics series)
Abstract:While large language model outputs are frequently analysed as a collective super variety termed “AI language,” this chapter argues that this perspective coexists with distinct, model-specific linguistic signatures akin to human idiolects. We analyse two datasets of LLM-generated texts on societal topics: a 2024 corpus of six models (Improta et al. 2024) and a newly generated 2026 corpus using the same prompts featuring six contemporary models. Our findings, utilising computational descriptors and stylometric principal component analysis reveal a generational shift between the style of the 2024 and 2026 cohorts, while demonstrating that each individual model maintains a unique linguistic profile. This multi-layered interplay is illustrated by contraction frequencies, which vary from over 1,200 to over 30,000 per million words within the same cohort of models (2026). Ultimately, we conclude that treating LLM output as idiolectal in nature provides a valuable framework with potential implications for research on variation and change, LLM-generated text detection, forensic linguistics and usage-based approaches to language.
[NLP-47] Model Confidence Under Answer-Preserving Attacks: An Informativeness-Manipulability Frontier
【速读】： 该论文旨在解决视觉-语言系统中置信度（confidence）作为决策门控信号时的鲁棒性问题，尤其关注在白盒、仅图像攻击下，攻击者能否通过微小扰动实现对置信度读出的操纵，从而绕过系统的安全控制。其核心挑战在于：在保持生成答案字节完全一致（byte-identical）的前提下，是否仍可破坏置信度输出的可靠性。解决方案的关键在于提出一种“统一幅度证书”（uniform amplitude certificate）——即在可测量阈值以下，若置信度变化幅度均匀，则能保证对抗性判别性能不低于原始答案字符串准确率（0.617）。然而，实验结果表明，无论是否依赖可达性假设，多种攻击策略（包括直接攻击与代理目标攻击）均能在所有84个评估组合中成功破坏该证书；更进一步，协同式、标签感知的攻击使对抗性判别性能降至或低于基准准确率，且在全部60个部署通道中均出现此类现象。此外，隐藏状态干预和文本模型激活空间的复制实验表明，置信度的异常波动不仅可通过对抗图像实现，亦可在表示层被诱导，说明该问题本质源于模型内部表征的脆弱性。四种测试的防御方法家族均未在该评估框架下展现出鲁棒性优势。在置信度门控模拟中，针对隐藏状态门控的转移攻击使原本被拒绝的错误答案接受率最高提升至84.8%；经各基准自然正确率重加权后，接受准确率在转移攻击下有8/12单元低于无门控基线，在直接门控攻击下则全部低于基线。因此，在所研究的威胁模型与预算下，置信度是一种对完整性敏感而非内在鲁棒的监督信号。
链接: https://arxiv.org/abs/2608.06571
作者: Reza Khanmohammadi,Ivan Brugere,Simerjot Kaur,Charese H. Smiley,Kundan Thind,Mohammad M. Ghassemi
机构: 1. Harvard University (哈佛大学); 2. Massachusetts Institute of Technology (麻省理工学院); 3. Stanford University (斯坦福大学)
类目: Cryptography and Security (cs.CR); Computation and Language (cs.CL)
备注:
Abstract:Deployed vision-language systems often gate their answers on confidence, making confidence robustness relevant to oversight. We study confidence readouts under white-box, image-only attacks constrained to preserve the generated answer byte-identically. Under a reachability assumption, an unmovable readout cannot outperform the answer-string accuracy prior, whose pooled value is 0.617. Independently of that assumption, a uniform amplitude certificate below a measurable threshold guarantees adversarial discrimination above the same floor. Across four vision-language models, three visual question answering benchmarks, five deployed confidence channels and two defense estimators, direct or surrogate-aimed attacks produce itemwise feasible perturbations that refute this uniform certificate in all 84 estimator-by-cell combinations. Coordinated correctness-label-aware attacks drive adversarial discrimination to or below the answer-string floor in all sixty deployed-channel cells, including all fifty-nine that begin above it. Hidden-state interventions and an open-ended text-model activation-space replication show that comparable confidence movement can be induced at the representation level rather than only through adversarial images. None of four tested defense families establishes a robust alternative under the specific evaluation applied to it. In a confidence-gated simulation, a coordinated token-probability attack transferred to a hidden-state gate causes up to 84.8% of previously rejected wrong answers to become accepted. After reweighting to each benchmark’s natural correctness prevalence, accepted accuracy falls below the no-gate baseline in eight of twelve cells under transfer and all twelve under a direct gate-aimed attack. Under the studied threat model and budget, confidence is therefore an integrity-sensitive rather than intrinsically robust oversight signal.
[NLP-48] Quantization Damage Is Multiplicative Not Additive EMNLP2026
【速读】： 该论文旨在解决量化（Quantization）在低比特位宽下对大语言模型决策稳定性造成隐性损害的问题，尤其是当比特数低于4时，尽管基准测试得分变化微小，但模型的关键行为（如工具调用、安全拒绝率）却显著退化。传统方法假设量化仅引入固定大小的噪声，从而保证高置信度决策的安全性，但本文通过直接测量决策边际（margin）的变化发现，量化并非添加固定噪声，而是以乘法方式压缩决策边际——即“边际收缩”（margin shrinkage），其收缩程度随比特位宽下降而加剧（4比特时中位乘数为0.86，3比特时降至0.33，2比特时趋近于0）。这种收缩削弱了高初始边际带来的鲁棒性保护，使模型的微小偏差主导失败方向：例如在3比特时，是否调用工具的决策趋向于不作为，而具体选择哪个工具则保持不变。研究进一步表明，基于边际收缩的统计模型能以极高的精度预测决策翻转概率（中位误差仅1.8个百分点），且预测值具备良好的校准性（期望校准误差0.004），且该关系在所有被测模型中一致成立，但参数不可跨模型迁移。关键解决方案在于：通过每模型、每比特位宽下的小样本边际测量，即可高效预判哪些决策会失效，无需全生成评估；在成本匹配条件下，增加一位比特是修复损伤最经济有效的手段。
链接: https://arxiv.org/abs/2608.06564
作者: Zekun Wu,Swati Dhiman,Adriano Koshiyama
机构: Holistic AI; UCL Centre for Artificial Intelligence, University College London
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: 16 pages, 9 figures, 7 tables. Under review at the Third Workshop on Uncertainty-Aware NLP (UncertaiNLP), EMNLP 2026 (non-archival)
Abstract:Quantization is how large language models are actually deployed, and below four bits it is known to hurt. What nobody can say is which of the model’s decisions will change at a given bit-width. The damage is silent: a compressed agent stops calling its tools, then loses half its safety refusals, yet benchmark scores barely move. Prior work assumes quantization adds noise of a roughly fixed size, which would make confident decisions safe. We measure the decision itself instead. The margin of a two-way decision is the model’s score for the option it picks minus the score of its best alternative; we track it before and after quantization across 16 models from 8 model families, three quantization methods, and bit-widths from 8 down to 2. Quantization does not add fixed-size noise to the margin. It multiplies the margin by a factor that collapses with bit-width (median 0.86 at 4 bits, 0.33 at 3, 0.00 at 2); we call this margin shrinkage. This contraction reduces the protection a large margin affords; the model’s own small biases pick the direction of failure: at 3 bits the decision to call a tool collapses toward inaction while the choice of which tool is untouched. In fitted statistical comparison, additive-noise accounts never win on the damaged tool and safety decisions. The fitted relation predicts flip rates within a median of 1.8 percentage points on held-out decisions, though no flip was used in the fit; per decision, the predicted flip probabilities are calibrated uncertainty estimates (expected calibration error 0.004 over 131,758 predictions). The same form holds in every model we measure, but the constants are each model’s own and do not transfer. A small paired margin set, measured per model and bit-width, estimates which decisions break without full generative evaluation; under our cost-matched tests, nothing repairs damage more cheaply than one more bit.
[NLP-49] radeVerse: A Longitudinal Benchmark of Political Negotiation in International Trade
【速读】： 该论文旨在解决当前大语言模型（LLM）在处理机构与政治文本时存在的局限性问题，即现有评估基准多基于孤立文档或单一任务，无法充分反映真实政治谈判中多轮迭代、上下文依赖的动态特性。其核心挑战在于：政治谈判具有显著的纵向时序特征，每一回合的发言均依赖于此前所有互动的历史背景，因此理解当前语境需对全过程进行追踪。为此，论文提出了TradeVerse——一个基于世界贸易组织（WTO）具体贸易关切数据构建的基准，涵盖1170场会议记录，覆盖5个议题组和89个产品类别。该基准定义了三项关键任务：一是根据历史会议记录预测当前讨论产品的协调制度编码（HS章节）；二是仅依据匿名化会议内容推断回应国家名称；三是让模型扮演回应国角色，生成最后一轮的正式声明。所有标签均直接从官方会议纪要中提取，无需人工标注。实验表明，这些任务对现有LLM构成了严峻挑战。因此，TradeVerse的关键创新在于首次系统性地考察了LLM在理解长期性、交互式政治贸易谈判中的潜力，推动了对模型纵向推理能力的评估。
链接: https://arxiv.org/abs/2608.06549
作者: Debodeep Banerjee,Amitangshu Dasgupta
机构: University of Pisa(比萨大学); University of Trento(特伦托大学); RazorPay( RazorPay)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注:
Abstract:LLMs are increasingly being applied to tasks involving institutional and political texts, but existing benchmarks evaluate them on isolated documents or single tasks. In realpolitik, negotiations are longitudinal data, where participating parties can align or argue over multiple iterations and each turn is an outcome of the previous turns, hence, understanding one turn requires tracking everything before it. We introduce TradeVerse, a benchmark built from the World Trade Organisation (WTO) specific trade concerns, where member states challenge one another and exchange arguments over multiple rounds, sometimes for years. We, in TradeVerse, reconstruct minutes of 1170 meetings, spanning across 5 groups and 89 product groups and define three tasks: first, the system has to analyze the longitudinal meeting records and predict the harmonized system codes (HS chapters) of the products under discussion in the particular meeting, second, we examine whether the system, upon analyzing the anonymized content of the meeting, can guess the name of the responding country and third, we ask the system to play the role of the responding country and provide the statement for the very last round. All labels are recovered directly from the proceedings, requiring no manual annotation. Our experiments highlight the challenges these tasks pose for current LLMs. To the best of our knowledge, TradeVerseis the first benchmark to investigate potential of LLMs in understanding longitudinal political trade negotiations.
[NLP-50] Dont `Well Actually Me Unless You Know What Youre Talking About: Weak Presupposition Verification Degrades General QA Performance
【速读】： 该论文旨在解决生成式问答系统在处理虚假预设问题（False-Presupposition Questions, FPQs）时存在的评估偏差与泛化能力不足的问题。现有方法通常将任务简化为提取预设并逐一进行事实核查，然而当前评估体系过度侧重于虚假预设问题（FPQs），忽视了对正常问题（True-Presupposition Questions, TPQs）的表现评估。由于真实场景中FPQ的出现频率远低于基准测试中的比例，导致现有模型在基准上的高表现无法反映其在实际应用中的真实性能。通过跨多种模型架构、规模及基准的广泛实验，研究发现：在FPQ上表现优异的模型往往在TPQ上表现更差，其根本原因在于现有方法中的事实核查模块存在过强的拒断倾向，误将正确预设也判定为错误。因此，解决方案的关键在于设计具备更强区分能力的事实核查机制，以避免对真实预设的误判，从而提升模型在真实、复杂应用场景下的整体鲁棒性与泛化能力。
链接: https://arxiv.org/abs/2608.06539
作者: Shenran Wang,Vered Shwartz,Hila Gonen
机构: University of British Columbia (不列颠哥伦比亚大学); Vector Institute (向量研究所); Amii (艾米研究院); CIFAR AI Chair (加拿大人工智能主席)
类目: Computation and Language (cs.CL)
备注:
Abstract:False-presupposition QA (FPQA) tests LLMs on their ability to identify false presuppositions in questions and abstain or correct them rather than reinforcing false assumptions. The common approach reduces the task to prompting LLMs to extract presuppositions and fact checking each presupposition. While the performance on dedicated benchmarks keeps improving, evaluation largely focuses on questions with false presuppositions (FPQs) while ignoring the performance on ``normal’’ questions (TPQs). Since many benchmarks over-represent FPQs compared to their natural occurrence, the result is that performance on these benchmarks doesn’t reflect real-world QA performance. Through extensive experiments across various model families, sizes, and benchmarks, we show that methods that perform better on FPQs tend to perform worse on TPQs. Our analysis reveals this is the result of weak fact checking modules that reject also true presuppositions. We hope our findings will help guide future work toward FPQA methods that generalize well to realistic settings.
[NLP-51] Confidence Estimation for Financial Vision-Language Models in Chart and Document Understanding
【速读】： 该论文旨在解决生成式金融视觉语言模型（LVLMs）在读取财务图表、表格和文档时因单个误读导致决策偏差的问题，其核心挑战并非单纯的准确性，而是可信度（trust）——即判断哪些模型输出可直接执行，哪些需提交人工复核。解决方案的关键在于构建可信赖的置信度估计机制，以支持安全自动化决策。研究评估了七种置信度估计方法（三类仅依赖推理的基线与四类在自然图像上训练的内部探测器），在五种开源权重的LVLM及四个来自三个金融视觉问答基准的数据条件下进行测试，所有探测器均未针对金融领域微调，从而衡量其跨分布迁移能力。关键发现包括：第一，校准性（calibration）而非排序能力是稀缺属性——推理基线虽能有效区分正确与错误答案，但存在严重过度自信问题，校准误差远超可容忍阈值；唯有训练过的探测器能产生可设置阈值的置信分数。第二，可靠性具有结构性特征，受模型与任务双重影响，无单一最优估计器覆盖超过20个（模型，条件）组合中的8个；通过控制性双语对比揭示出的“语言鲁棒性”实为组合效应，在逐模型分析下不复存在。第三，当将决策过程建模为基于误差预算的拒绝服务（deferral）时，自动化程度首先由模型自身能力决定，其次才受置信度约束——在较易任务中可实现较高自动化率，而在最困难任务中几乎无法自动化，严格5%误差预算下接近零。最终，仅有两个训练探测器具备所需校准性，其中仅“基于根基感知”（grounding-aware）的探测器能有效降低模型未使用图表信息而生成的答案的置信度，从而区分非根基性回答与流畅猜测，显著提升可信度判断的可靠性。
链接: https://arxiv.org/abs/2608.06532
作者: Reza Khanmohammadi,Simerjot Kaur,Charese H. Smiley,Ivan Brugere,Mohammad M. Ghassemi
机构: Michigan State University (密歇根州立大学); JPMorgan AI Research (摩根大通人工智能研究); OpenAI
类目: Computation and Language (cs.CL)
备注:
Abstract:LVLMs are increasingly used to read financial charts, tables, and documents, where a single misread figure can move a decision and the most authoritative-looking answer is sometimes one the model produced without reading the exhibit. The operational question is therefore trust, not accuracy: which answers can be acted on, and which escalated to a reviewer. We evaluate seven confidence estimators, three inference-only and four trained internal probes, across five open-weight LVLMs and four conditions from three financial visual question-answering benchmarks, one bilingual; every probe is trained only on natural images and applied to finance without adaptation, so the results measure out-of-distribution transfer. Three findings hold. First, the scarce property is calibration, not ranking: the inference baselines rank correct above incorrect answers competitively but are badly overconfident, calibration error far above what a threshold can tolerate, and only the trained probes produce a thresholdable score. Second, reliability is structured rather than global, along two axes a practitioner can read directly: the best estimator shifts with both model and task, none leading more than eight of twenty (model, condition) cells, and a controlled bilingual contrast exposes an apparent language robustness as a composition artifact that dissolves once models are read one at a time. Third, cast as deferral under an error budget, how much can be safely automated is set first by the model’s competence and only narrowed by its confidence, so deferral clears a real share of the easiest condition and almost none of the hardest, near zero at a strict 5% budget. Two trained probes carry the calibration a deferral policy needs, and among them only the grounding-aware one lowers its confidence on answers a model gives without using the figure, separating detected non-grounding from a fluent guess.
[NLP-52] Lost in Interpolation: Why Predictive Feedback Fails in Diffusion Language Models
【速读】： 该论文旨在解决掩码扩散语言模型（Masked Diffusion Language Models, MDLMs）在推理过程中因采用线性插值（LERP）进行软掩码（soft-masking）而导致的训练退化问题。现有方法在原始嵌入空间中使用LERP进行混合，隐含假设该空间为欧氏几何，但研究发现MDLM的嵌入空间具有超球面几何特性：掩码嵌入与预测词嵌入之间的夹角在训练过程中保持近似恒定（约73°），且嵌入范数随词汇频率排名基本不变。这表明欧氏空间中的LERP并非合适的插值方式。为此，论文提出球面软掩码（Spherical Soft-Masking, S-SM），其核心在于采用超球面上的Fréchet均值聚合前k个预测结果，并通过球面线性插值（SLERP）将该均值与掩码方向融合，最后恢复原始掩码范数。实验表明，相较于基线模型及基于TopK/LERP的方法，S-SM在不同推理步数预算下可实现最高达2倍的MAUVE得分提升、16.9%-19.6%的生成困惑度降低，同时保持输出熵和收敛性不变，显著缓解了由LERP反馈引起的训练退化问题。
链接: https://arxiv.org/abs/2608.06529
作者: Lavanya Nigam,Ishaan Bansal,Aryan Sood,Vidit Aggarwal,Gaurav Kumar Nayak
机构: Indian Institute of Technology Roorkee, Roorkee, Uttarakhand, India
类目: Computation and Language (cs.CL)
备注: 15 pages
Abstract:Soft-masking accelerates the convergence of Masked Diffusion Language Models (MDLMs). Existing formulations build this blend with linear interpolation (LERP) in the raw embedding space, which implicitly treats that space as Euclidean. We analyze the embedding space of MDLMs and find that the mask and predicted-token embeddings maintain a near-constant angle of (\approx 73^\circ) throughout training, while embedding norms remain essentially flat across vocabulary-frequency rank. These indicate a hyperspherical geometry, for which LERP is the wrong interpolation primitive. We introduce Spherical Soft-Masking (S-SM), a drop-in replacement that aggregates the top-(k) predictions with a Fr’echet mean on the hypersphere and blends this mean with the mask direction using spherical linear interpolation (SLERP), then restores the native mask norm. We evaluate S-SM on continued pre-training of a released 169M-parameter MDLM checkpoint across a wide range of inference-time step budgets, SLERP feedback avoids the training degradation that LERP feedback induces and delivers MAUVE gains of up to 2x over the vanilla MDLM baseline and 27.5-56.1% over TopK/LERP at various sampling budgets, alongside consistently lower generative perplexity (16.9-19.6% over the baseline), while leaving output entropy and convergence essentially unchanged.
[NLP-53] GRASP: Reinforcing Language Model Anonymizers with Group Relative Policy Optimization
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）从普通文本中推断敏感个人属性（如年龄、地理位置和职业）所带来的隐私泄露问题。现有方法中的对抗性匿名化（Adversarial Anonymization）虽能通过重写文本以隐藏敏感信息，但依赖于强大的外部模型在推理时执行，导致私有文本需上传至第三方，违背了匿名化的初衷。尽管近期研究通过监督微调与直接偏好优化（Direct Preference Optimization, DPO）将这一行为蒸馏至小型本地模型，但DPO仅模仿教师模型的离线选择，未能直接优化隐私-效用权衡这一核心目标。为此，本文提出GRASP（Group-Relative Anonymization via Self-refinement Policy-optimization），其关键在于采用组相对策略优化（Group Relative Policy Optimization, GRPO）在线强化本地小模型的匿名化能力。该模型同时承担匿名化器、攻击者和效用判别器三重角色，基于自生成的奖励信号进行训练，该奖励信号在隐匿敏感属性的同时保持语义完整性，并通过设计机制防范奖励劫持（reward hacking）。实验表明，基于Llama-3.1-8B训练的GRASP在三个独立的大型语言模型评判者下均显著优于DPO蒸馏基线，在对抗由前沿模型（如Gemini 2.5 Flash和Claude）驱动的匿名化攻击时，不仅实现相当或更优的隐私-效用平衡，且大幅去除更多敏感信息，同时完全在设备端运行，计算成本仅为GPT-4o教师模型的约1%。
链接: https://arxiv.org/abs/2608.06526
作者: Sajjad Ghiasvand,Nader Sehatbakhsh
机构: UC Santa Barbara; UC Los Angeles
类目: Computation and Language (cs.CL)
备注:
Abstract:Large language models can infer sensitive personal attributes, such as age, location, and occupation, from ordinary text, turning everyday writing into a privacy risk. Adversarial anonymization defends against this by rewriting a text with a capable language model that also plays the attacker, but it needs a powerful model at inference time and thus sends private text to a third party, the very exposure anonymization should prevent. Recent work distills this behavior into a small on-device model using supervised fine-tuning and direct preference optimization (DPO), but DPO only imitates the teacher’s offline choices and never directly optimizes the privacy–utility objective we care about. We introduce \textbfGRASP (\textbfGroup-\textbfRelative \textbfAnonymization via \textbfSelf-refinement \textbfPolicy-optimization), which reinforces the local anonymizer online with Group Relative Policy Optimization. A single small model acts as anonymizer, adversary, and utility judge, trained against a self-generated reward that hides attributes while preserving meaning, with a design that guards against reward hacking. Trained on Llama-3.1-8B, \ours improves the privacy–utility trade-off over the DPO-distilled baseline, consistently across three independent LLM judges. Against adversarial anonymization driven by frontier models such as Gemini~2.5~Flash and Claude, it achieves a comparable or better overall trade-off while removing substantially more private information, and it runs entirely on-device at roughly 1% of the GPT-4o teacher’s cost.
[NLP-54] Measuring the Cross-Lingual Comprehension Gap: How the language of the evidence shapes what language models understand ACL
【速读】： 该论文旨在解决语言模型在多语言环境下能力迁移不均的问题，具体聚焦于英语以外语言中模型表现下降的现象。传统多语言评估基准往往未能控制内容、问题、参考答案、模型及评估单元等变量，导致对跨语言理解能力的误判。为此，论文提出“跨语言理解差距”（Cross-Lingual Comprehension Gap, CLCG）这一概念，用以量化相同内容与问题在目标语言（非英语）下相较于英语时响应质量的下降程度。其核心解决方案是基于ParallelQA-18这一高质量平行语料库，采用跨语言对照设计，在保持内容、问题、模型和评估方式一致的前提下，仅改变文本语言，系统评估五家实验室的五个模型在18种语言（含高资源语言葡萄牙语与16种低资源语言）上的表现。研究通过分层抽样150篇文章，并使用基于文章聚类的自举法计算置信区间，发现主要的CLCG为0.078（95% CI 0.072–0.084），相当于英语得分约17%的下降；进一步分析显示，该差距与Joshi等（2020）定义的语言资源等级呈显著负相关（rho = -0.594, p = 0.015），表明资源越少的语言，性能衰减越严重。盲测人工评估结果也支持这一发现，高资源语言生成的回答在61.6%的可决判断中更受青睐（偏好概率估计为0.655, 95% CI 0.558–0.741）。因此，该研究的关键在于建立一种严格控制变量的跨语言对比范式，揭示并量化了当前语言模型在非英语场景下的显著性能衰减，强调不应假设英语中的能力可自然迁移到其他语言，尤其应警惕英语中心评价体系对低资源语言用户能力的过度高估。
链接: https://arxiv.org/abs/2608.06506
作者: Rafael da Silva,Jeff Eicher
机构: Eastern University (东盎格利亚大学)
类目: Computation and Language (cs.CL)
备注: 55 pages, 17 figures. Submitted to Computational Linguistics (MIT Press / ACL). Supplementary Material: 55 pages, 4 figures
Abstract:Language models are often evaluated as though capabilities demonstrated in English remain equally available when the same content is presented in other languages. Traditional multilingual benchmarks rarely isolate language while holding content, question, reference answer, model, and evaluation unit constant. We define the Cross-Lingual Comprehension Gap (CLCG) as the reduction in response quality when the same content and question are presented in a target language rather than in English. Using ParallelQA-18, a professionally human-translated parallel corpus, we evaluate five models from five laboratories on a stratified sample of 150 articles across 18 languages (English reference; Portuguese high-resource baseline; 16 targets spanning Joshi et al. 2020 classes 0-4). A within-item design varies only passage language. The primary estimator contrasts English versus pooled target-language Token-F1 micro-means on higher-complexity open-ended questions, with article-cluster bootstrap intervals. The primary pooled CLCG is 0.078 (95% CI 0.072-0.084), about a 17% reduction relative to the English score; the equal-language macro summary is 0.077. Net of Portuguese, the macro gap is 0.016 (95% CI 0.013-0.020). Language-level CLCG is negatively associated with Joshi resource class (rho = -0.594, p = 0.015, n = 16). In blinded paired human evaluations, higher-resource responses are preferred in 61.6% of decisive judgments (estimated preference probability 0.655, 95% CI 0.558-0.741). Capabilities shown in English should not be assumed to transfer equally to other languages; English-centered evaluations may overestimate quality for users of low-resource languages. Comments: 55 pages, 17 figures. Submitted to Computational Linguistics (MIT Press / ACL). Supplementary Material: 55 pages, 4 figures Subjects: Computation and Language (cs.CL) Cite as: arXiv:2608.06506 [cs.CL] (or arXiv:2608.06506v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2608.06506 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Rafael Da Silva [view email] [v1] Thu, 6 Aug 2026 18:45:16 UTC (6,241 KB)
[NLP-55] Can MLLM s Decode the Creative Leap? Introducing C4 for Cross-Concept Understanding
【速读】： 该论文旨在解决生成式多模态大模型（Generative Multimodal Large Models, MLLMs）在创意能力评估中缺乏明确目标与奖励信号的问题，尤其聚焦于跨概念理解这一核心认知能力在创造性感知中的作用。现有评估体系多依赖于准确性导向任务，难以有效衡量模型在非显性但有意义的概念关联中恢复意图的能力。为此，论文提出一种受认知启发的评估框架C4（Chengyu-based Cross-Concept Creativity），其关键在于将题目构建建模为跨概念编码（cross-concept encoding），将模型推理视为跨概念解码（cross-concept decoding）。该框架基于汉语成语（Chengyu）构建了一个人工标注并经第三方审核的跨概念网络，通过桥接路径（bridge paths）将目标槽位映射至可图像化的替代概念，实现结构化、可控制难度（由桥接数量和深度决定）的批量生成，并支持精确答案判定。基于此，研究构建了包含184个合成题项及37个真实人类创作的跨概念成语图示的C4评估集（C4-Eval），每题在五种任务设置下共产生884个主答案恢复案例。实验表明，最强闭源模型在主准确率上仅达50.7%和48.0%，而开源模型表现显著更低，说明当前MLLMs在通过跨概念关系解码创造性编码意义方面存在明显短板；尽管候选约束能显著提升性能，但桥接提示与解释请求带来的增益有限。这揭示了现有模型在处理深层语义关联与创造性推理方面的系统性不足，凸显了对更先进跨概念理解机制的需求。
链接: https://arxiv.org/abs/2608.06501
作者: Ming Wang,Yuqing Zhang,Tingna Xie,Xiangju Li,Xiaocui Yang,Daling Wang,Shi Feng,Yifei Zhang
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Multimedia (cs.MM)
备注:
Abstract:Creative capabilities of MLLMs matter in design, communication, education, and human–AI collaboration, yet remain difficult to evaluate because explicit targets and reward signals are scarce compared with accuracy-oriented tasks. Cross-concept understanding is a core cognitive capacity underlying receptive creativity. It enables a perceiver to recover intended meaning from non-obvious but meaningful conceptual relations. We operationalize item construction as cross-concept encoding and model inference as cross-concept decoding. We introduce C4, a cognition-inspired evaluation framework for Chengyu (Chinese idiom)-based Cross-Concept Creativity. Its encoding component maps target slots to imageable substitute concepts along bridge paths in a manually annotated and third-party-reviewed cross-concept network, enabling batch generation with explicit structure, difficulty indexed by bridge count and depth, and exact answers. Using this framework, we instantiate the C4 Evaluation Set (C4-Eval), comprising 184 synthetic items and 37 human-created cross-concept chengyu figures collected from online sources. We manually construct and review cross-concept relations, bridge paths, and reasoning processes for the collected figures. Each C4-Eval item is instantiated in five task settings, yielding 884 primary answer-recovery cases. Across ten evaluated MLLMs, the strongest closed models reach 50.7% and 48.0% primary accuracy, while open-source models remain substantially lower. Candidate constraints improve accuracy sharply, but bridge hints and explanation requests provide only modest gains. These results expose a substantial gap in how current MLLMs decode creatively encoded meaning through cross-concept relations. The code is in the supplementary material.
[NLP-56] ConstructCIE: A Dataset for Extracting Causal Information from Construction Accident Narratives
【速读】： 该论文旨在解决建筑事故报告中因果信息提取（Causal Information Extraction, CIE）的难题，尤其针对事故文本中因果证据隐含、跨度长且分布分散的挑战。其核心解决方案是构建了一个手工标注的数据集ConstructCIE，采用分层架构对事故类型、因果因素、子因果因素及支持性证据片段进行标注。研究评估了监督式序列标注模型与指令微调的大语言模型（LLM）在端到端分层提取任务中的表现，结果表明，尽管多数模型在事故类型预测和整体因果语义恢复方面表现良好，但在细粒度的证据片段边界识别上仍存在局限。其中，联合层次抽取（JHE）策略在精确匹配与软匹配指标上表现更优，而独立层次抽取（IHE）在关键词F1上偶有更高表现。然而，证据选择错误与边界定位错误仍是普遍存在的问题。研究结论指出，实现建筑事故中可靠因果信息提取需更强的领域知识嵌入以及更精准的证据片段抽取能力。
链接: https://arxiv.org/abs/2608.06495
作者: Hung Nguyen,Jaehoon Lee,Namgyun Kim,Kuan-Hao Huang
机构: Texas AM University(德州农工大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Construction accident narratives contain rich causal information, but the evidence is often implicit, long-span, and distributed. We introduce ConstructCIE, a manually annotated dataset for Causal Information Extraction from OSHA construction accident reports. The dataset uses a hierarchical schema for accident types, causal factors, sub-causal factors, and supporting evidence spans. We evaluate supervised sequence taggers and instruction-tuned LLMs in an end-to-end hierarchical extraction setting. Results show that most evaluated models achieve strong accident-type prediction and recover broad causal meaning but remain limited in precise span-level extraction. JHE generally achieves stronger exact and soft matching, while IHE sometimes achieves higher keyword F1. Error distributions vary by extraction strategy, but evidence-selection and span-boundary errors remain common. These findings show that reliable Causal Information Extraction for construction accidents requires stronger domain grounding and more accurate evidence extraction.
[NLP-57] Do AI Personas Grow? Analyzing and Benchmarking Personality Evolution in LLM Agents After Life Events
【速读】： 该论文旨在解决生成式人格化大语言模型（Personality-conditioned LLM agents, PC-Agents）在长期交互中维持人格一致性的问题，尤其关注其在经历重大生活事件后的人格演变是否符合心理学实证规律。核心挑战在于：尽管已有研究表明上下文扰动可导致大模型人格特征发生偏移，但这种变化在不同人格特质、生活事件、角色设定及模型间的差异机制尚不清晰，且缺乏对变化方向与幅度的系统性评估。论文的关键解决方案是引入以大五人格模型（Big Five Traits）为心理测量锚点，通过分析11项重大生活事件引发的人格变化轨迹，并基于人类纵向研究证据进行对比验证。研究提出BFI-Adapt这一可复用基准工具，用于量化事件诱发人格变化的方向保真度（directional fidelity），并据此对14种模型进行排序。实验结果表明，尽管PC-Agents表现出与人类相似的变化速率和部分预期方向的一致性，但其变化幅度普遍低于人类效应量范围，且角色层面的个体差异压缩达三至四倍，远低于真实人群。此外，通过多维度验证（如重测信噪比、提示重构稳定性、行为选择收敛性及跨对话持续性），确认了所测得的人格轨迹具有稳健的事件条件响应特性。最终结论指出，当前PC-Agents仅能模拟人类人格动态的均值趋势，而未能捕捉其真实的分布形态与复杂性。
链接: https://arxiv.org/abs/2608.06485
作者: Ming Wang,Peidong Wang,Xiaocui Yang,Daling Wang,Shi Feng,Fiona Fui-Hoon Nah,Ee-Peng Lim
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Social and Information Networks (cs.SI)
备注:
Abstract:Personality-conditioned LLM agents (PC-Agents) are increasingly used in emotional support, social simulation, and role-playing, motivating the development of lifelong agents that remain coherent over extended interactions. A key component of such coherence is personality evolution: agents should undergo plausible, psychology-grounded changes as they experience life events in different contexts. Although prior work shows that LLM personalities can shift under contextual perturbations, how these shifts vary across traits, events, personas, and models remains poorly understood. We study event-induced personality change after 11 major life events, using the Big Five traits as a psychometric anchor and interpreting the resulting trajectories against longitudinal evidence from human personality psychology. Across four diagnostic axes, PC-Agents exhibit measurable trait shifts at similar rates for event-trait pairs with and without documented human change directions. Even when shifts follow the expected direction, their magnitudes usually fall below human effect-size ranges. Gender and cultural-region prompts show little moderating effect, while persona-level dispersion is compressed three- to four-fold relative to human samples. To enable systematic comparison, we introduce BFI-Adapt, a reusable benchmark for scoring the directional fidelity of event-induced personality change, and use it to rank 14 models. A validation suite shows that the measured shifts exceed no-event retest noise, remain stable under independently paraphrased prompts, exhibit limited and model-dependent convergence with scenario-based behavioral choices, and persist across intervening unrelated dialogue. Together, these checks establish the measured trajectories as robust event-conditioned response patterns. Our results suggest that current PC-Agents simulate the mean of human personality dynamics, but not its shape.
[NLP-58] StepJack: Benchmarking Computer-Use Agent Safety Against Multi-Step Indirect Prompt Injection
【速读】： 该论文旨在解决计算机使用代理（Computer-use Agents, CUAs）面临的多步间接提示注入（multi-step indirect prompt injection）威胁，即攻击者将恶意指令以看似无害的子步骤形式分散嵌入代理导航路径中的一系列网页环境，从而规避检测并逐步实现攻击目标。其解决方案的关键在于构建一个自动化分解管道，能够在保证各子步骤整体达成原始攻击目标的前提下，优化每个子步骤的隐蔽性（innocuousness），从而实现攻击意图的渐进式、隐蔽式传递。基于该方法，研究者提出了StepJack基准测试集，包含480个测试用例，并对六种先进的CUA进行了评估，结果表明在固定分解深度下，多步攻击相较单步攻击显著提升了攻击成功率（ASR），最高提升达31.2个百分点，验证了该攻击范式的有效性与严重性。
链接: https://arxiv.org/abs/2608.06477
作者: Zhuoxin Zhan,Akbar Rafiey,Avery Ma,Leila Pishdad,Layla El Asri
机构: Simon Fraser University (西蒙菲莎大学); RBC Borealis (加拿大皇家银行布雷奥利斯); New York University (纽约大学); OpenAI (OpenAI); Anthropic (Anthropic)
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Computer-use agents (CUAs) face a growing threat from indirect prompt injection, where adversarial instructions are planted in the environment such as web pages. In this paper, we introduce multi-step indirect prompt injection, a new attack class against CUAs in which the adversarial goal is decomposed into multiple innocuous-looking sub-steps and distributed across a chain of pages referenced along the agent’s navigation path. We develop a pipeline to automatically decompose an adversarial goal under the constraint that the execution of the decomposed sub-steps must achieve the original goal while optimizing the innocuousness of each decomposed sub-step. With this pipeline, we build StepJack, a CUA safety benchmark with 480 test examples. On this benchmark, we evaluate six state-of-the-art CUAs and find that at a fixed decomposition depth, multi-step attacks raise attack success rate (ASR) on three of six CUAs, by up to 31.2 points (e.g., GPT-5.4-mini: 41.7% at single-step to 72.9% at three-step); averaged over the five CUAs that can reliably follow the reference chain (all but EvoCUA-32B), ASR rises from 31.3% at single-step to 36.9% at three-step. Dataset and code are available at this https URL.
[NLP-59] Recovering Lesion Parameters from Aphasic Picture Naming Error Profiles in Large Language Models
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）可解释性方法中存在的核心问题：现有方法虽能描述模型内部状态，但未能直接验证这些状态是否在因果上足以产生观测到的行为。为此，研究提出一种逆向问题框架——基于给定的图像命名错误模式，反推导致该行为的模型扰动参数（包括层索引、修改比例和噪声标准差），从而检验模型内部机制的因果充分性。其关键解决方案在于构建一个多任务神经网络，将临床分类的七类错误模式（正确、语义错误、无关错误、形式错误、混合错误、新词创造、无响应）映射回对应的扰动参数。实验表明，尽管层索引仅能在局部邻域内被恢复，但修改比例与噪声标准差可稳定重构；通过反事实验证，使用恢复参数对新模型实例进行扰动后，81.4%的情况下重现了目标错误模式，证明了该方法的高反事实保真度。这一结果揭示了变换器（Transformer）层间存在的功能冗余特性，而该特性无法被传统可解释性方法捕捉。此外，模型在278名中风患者图像命名错误数据上的外分布测试中展现出对综合征的判别能力，尤其对扰动强度敏感，表明其具备超越训练分布的泛化能力。反事实验证为大语言模型可解释性主张提供了一种超越单纯逆向映射的通用评估框架。
链接: https://arxiv.org/abs/2608.06429
作者: Yong Yang,Roger Newman-Norlund,Xiang Guan,Saeed Ahmadi,Regan Willis,Nadra Salman,Kalil Warren,Sophie Arheix-Parras,Srihari Nelakuditi,Leonardo Bonilha,Christopher Rorden,Rutvik H. Desai,Julius Fridriksson
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Interpretability methods for large language models (LLMs) describe internal state but do not directly test whether that state is causally sufficient to produce the observed behavior. In earlier work, we lesioned LLMs to produce error profiles in picture naming, a central task for assessing aphasia, and found that specific lesions produced errors resembling those of individual stroke survivors. Here we ask the inverse question: given an error profile, can the lesion parameters that produced it be recovered, and what does this inverse problem reveal about transformer computation? Lesions in LLaVA-Vicuna 13B were parameterized by layer index, modification percentage, and noise sigma across 4,840 configurations, and error profiles were characterized by a seven-category clinical taxonomy (correct, semantic, unrelated, formal, mixed, neologism, no-response). We trained a multi-task neural network to map error profiles back to perturbation parameters. The problem admitted a partial solution: across 10 independently trained inverse models, modification percentage and noise sigma were recoverable, whereas layer index was recoverable only within a neighborhood. In counterfactual validation, a fresh model instance perturbed with the recovered parameters reproduced the target behavior in 81.4% of cases. This dissociation between low layer recovery and high counterfactual fidelity is consistent with functional redundancy across transformer layers, a property not captured by standard interpretability methods. As an out-of-distribution test, we applied the trained model to picture-naming error profiles from 278 stroke survivors; recovered parameters were syndrome-discriminative, most strongly for perturbation intensity, indicating generalization beyond the training distribution. Counterfactual validation provides a general framework for LLM interpretability claims beyond inverse mapping.
[NLP-60] NTDH: Complex Reasoning for Comprehensive Affective Analysis
【速读】： 该论文旨在解决情感分析中跨异构预测任务（如连续值、有序标签和多标签输出）的综合性情感建模难题，以及情感语义的高度上下文依赖性所导致的冲突线索难以直接映射到标签的问题。现有方法通常采用端到端的直接映射策略，未能显式建模不同线索间的协调与权衡过程。为此，本文将任务重新定义为一个复杂推理问题，通过构建统一的输出接口以兼容多种标签空间，并在推理轨迹上实现可验证奖励的优化，从而首次系统性地覆盖了情感（sentiment）与情绪（emotion）两类分析任务。其核心挑战在于数据层面：需合成可解释的情感推理路径，而通用合成方法在目标、容差机制及情感现象捕捉方面存在偏差，且容易忽略或泄露失败案例。为此，作者提出NTDH框架，包含四个关键设计：自然化（Naturalisation）确保训练答案为真值标签，从构造上保证正确性；容差感知门控（Tolerance-aware gate）依据任务特有评分容差检验每一步推理；领域感知策略（Domain-aware strategies）引入情感科学中的理论指导推理修正；方向性提示（Directional Hints）仅反馈错误类型与方向，不暴露真实目标，增强鲁棒性。实验基于16,302条训练样本（仅为同类指令微调系统的约1/14），采用Qwen3-8B模型进行SFT与GRPO联合训练，在相同容差条件下验证，结果表明最终策略在六个官方测试指标中优于初始SFT检查点，五个指标表现更优，并在情感强度回归（EI-reg）任务中达到0.862的皮尔逊相关系数，显著领先于对比系统，验证了高质量合成数据对模型性能的关键作用。
链接: https://arxiv.org/abs/2608.06425
作者: Tianlei Zhu,Zhiwei Liu,Yuyan Wang,Xiao-Yang Liu,Sophia Ananiadou
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 16 pages, 3 figures, 9 tables
Abstract:Comprehensive affective analysis is challenging for two reasons: it spans heterogeneous prediction tasks with continuous, ordinal, and multi-label outputs, and affective meaning is context-dependent, requiring conflicting cues to be reconciled rather than mapped directly to labels. Existing methods learn this mapping directly and do not model the reconciliation explicitly. We recast the task as a complex-reasoning problem, which yields one output interface across heterogeneous label spaces and a trajectory over which a verifiable reward can be optimised; to our knowledge, this is the first such treatment covering both sentiment and emotion. The obstacle is on the data side: affective reasoning traces must be synthesised, and generic synthesis is misaligned with the targets, tolerances, and phenomena of affect, and discards or leaks its failure cases. We propose NTDH, which addresses these four failures. Naturalisation sets the training answer to the gold label, so it is correct by construction. A Tolerance-aware gate checks each answer against the task’s own scoring margin. Domain-aware strategies refine the reasoning using ideas from affective science. Directional Hints report only the type and direction of an error, without exposing the target. We train Qwen3-8B with SFT and then GRPO under the same tolerance used for verification (up to a more permissive construction gate on the multi-label subtask), and a component ablation quantifies the data-quality effect of each part. Using 16,302 training records, about 14x fewer than comparable instruction-tuned systems, the final policy improves over its SFT checkpoint on five of six official-test metrics and achieves the strongest EI-reg result among the compared systems, at a Pearson correlation of 0.862.
[NLP-61] Multi Codec Discrete Diffusion Model for Text Guided Speech Inpainting and Editing
【速读】： 该论文旨在解决语音重建与编辑中因缺失、损坏或错误片段导致的语音内容恢复问题，具体包括语音补全（speech inpainting）和基于文本编辑的语音修改（speech editing）。其核心挑战在于：在不重新合成整个语音的前提下，生成的内容需准确表达目标语义，同时保持说话人身份、韵律、时间对齐及录音条件的一致性。论文提出的解决方案关键在于引入一种基于离散扩散模型（discrete diffusion）的层级编码器框架——SIEDD（Speech Inpainting and Editing with Discrete Diffusion），其核心架构HiCoDD通过遵循向量量化（Vector Quantized, VQ）的生成顺序，将已生成的码本作为干净的、确定性的声学上下文进行建模，并仅对当前待优化的码本执行扩散过程，从而实现无信息泄露的联合训练与符合粗到细推理顺序的渐进式生成。此外，SIEDD结合了音素级条件控制、局部化分类器无关引导（span-localized classifier-free guidance）以及持续时间预测机制，有效支持固定时长补全与可变时长文本编辑任务。在RealEdit基准测试中，SIEDD在语音编辑性能上优于所有对比方法，并在单个与多个缺口的语音补全任务中全面超越自回归基线模型，验证了显式建模编码器层次结构对保持上下文一致性的关键作用。
链接: https://arxiv.org/abs/2608.06424
作者: Iftach Shoham,Tali Dror,Oren Gal,Haim Permuter,Gilad Katz,Eliya Nachmani
机构: Ben-Gurion University of the Negev (本-古里安大学); University of Haifa (海法大学)
类目: ound (cs.SD); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Speech recordings often contain missing, corrupted, or incorrect regions that must be reconstructed or modified without re-synthesizing the entire utterance. Speech inpainting restores missing segments, whereas speech editing replaces spoken content according to an edited transcript. Both tasks require the generated speech to express the intended words while remaining consistent with the surrounding speaker identity, prosody, timing, and recording conditions. Discrete diffusion is particularly well suited to these tasks because it can iteratively refine masked tokens while jointly conditioning on both left and right acoustic context. We introduce SIEDD, a discrete diffusion framework for text-guided speech inpainting and editing over hierarchical codec tokens. Its core architecture, HiCoDD, follows the RVQ generation order by representing previously generated codebooks as clean, committed acoustic context and applying diffusion only to the current refinement codebook. This separation enables leakage-free joint training while matching sequential coarse-to-fine inference. The model further combines phoneme-level conditioning, span-localized classifier-free guidance, and duration prediction to support both fixed-duration inpainting and variable-duration text edits. On the RealEdit benchmark, SIEDD achieves the best overall speech-editing performance among the evaluated methods. It also outperforms the evaluated autoregressive baselines across all speech-inpainting settings, on both single and multiple gaps. These results demonstrate that explicitly modeling the codec hierarchy substantially improves context-preserving speech reconstruction and editing. See our full code at this https URL.
[NLP-62] Latent Fact-Checking: Detecting Misinformation through Activation Engineering
【速读】： 该论文旨在解决在线虚假信息（misinformation）检测中依赖表面语言特征或外部知识检索所带来的可扩展性与泛化性不足的问题。其核心挑战在于如何在不依赖微调、外部证据检索或任务特定标注的情况下，实现对虚假陈述的有效识别。解决方案的关键在于将真实性（truthfulness）视为预训练语言模型表示空间中的几何属性，通过激活工程（activation engineering）方法，利用对比激活添加（Contrastive Activation Addition, CAA）原则，从成对的真实与虚假语句的激活差异中提取出一个表征虚假性的“虚假方向”（falsehood direction）。在推理阶段，通过将待测声明的最后一个标记激活向量投影到该方向，并输入多层感知机（MLP）进行分类，从而完成判断。该方法无需对主干模型进行微调，也无需额外监督信号，仅依赖于用于估计方向的对比样本。实验结果表明，该虚假方向在不同规模和架构的模型（Gemma、Llama、Qwen系列，参数量270M至12B）中均具可复现性，在LIAR和FACTors基准上表现优于零样本及少样本提示基线，尤其在小型模型上优势显著；而在AVeriTeC上的表现受限，归因于其基于证据的标签机制。研究揭示了真实性在预训练语言模型潜在空间中具有结构化的线性可分特性，为基于可解释性驱动的虚假信息检测提供了新范式，可作为检索增强型检测流程的有效补充。
链接: https://arxiv.org/abs/2608.06417
作者: Pedro Barcelos,Otávio Parraga,Marcelo M. Mussi,Lucas M. Fraga,Lucas S. Kupssinskü,Rodrigo C. Barros
机构: MALTA, Machine Learning Theory and Applications Lab, PUCRS, Porto Alegre, Brazil; Kunumi Institute, Brazil
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: 13 pages
Abstract:The proliferation of misinformation online has driven demand for scalable detection systems. While most existing approaches rely on surface-level linguistic features or external knowledge retrieval, we examine truthfulness as a geometric property of a language model’s representation space. We introduce a misinformation detection framework grounded in activation engineering, which leverages the latent geometry of transformer models. Our approach elicits a misinformation direction in the residual stream by contrasting activations from paired truthful and false statements, following the difference-in-means principle of Contrastive Activation Addition (CAA). At inference time, the last-token activation of an unseen claim is projected onto this direction, and the projected representation is fed to an Multilayer Perceptron (MLP) for classification. The procedure requires no fine-tuning of the backbone model, no external evidence retrieval, and no task-specific supervision beyond the contrastive pairs used to estimate the direction. We evaluate the method across 11 models from the Gemma, Llama, and Qwen families, ranging from 270M to 12B parameters, on three fact-checking benchmarks: AVeriTeC, LIAR, and FACTors. The falsehood direction is recoverable across model scales and architectural families, and last-token projection matches or surpasses zero-shot and few-shot prompting baselines on LIAR and FACTors, with the largest gains observed for smaller models. Performance on AVeriTeC is more limited, which we attribute to its evidence-grounded labeling scheme. These findings provide evidence that truthfulness is a structured, linearly separable concept in the latent space of pretrained language models, and point toward interpretability-driven misinformation detection as a practical complement to retrieval-based pipelines. The code is available on this https URL.
[NLP-63] Separating Decision-Rule Misalignment from Readout-Coverag e Limitations in Speech Language Models
【速读】： 该论文旨在解决当前语音语言模型在副语言任务（paralinguistic tasks）中评估时，答案准确率（answer accuracy）无法区分音频到答案生成过程中各阶段失败来源的问题。现有评估方法将不同环节的误差混杂于单一指标中，导致对模型性能瓶颈的误判。为此，作者提出一种“生成对齐诊断阶梯”（generation-aligned diagnostic ladder），通过对比生成答案、选项逻辑值（option logits）、逻辑值的仿射读出（affine readout of those logits）以及同一答案标记处隐藏状态的线性读出（linear readout of the hidden state），系统分离出端点（endpoint）、决策规则（decision-rule）和读出覆盖范围（readout-coverage）三类差距。实验结果表明，在五个系统与两个情绪语料库上，状态解码（state decoding）平均比生成（generation）高出27.8个百分点的准确率，且所有十种条件下决策规则与读出覆盖范围差距均为正值。进一步地，无标签逻辑值校正方法在所有条件下均提升了生成准确率，证明部分决策规则差距具有可改进性。在排序匹配比较中，发现超出原生读出空间的情绪信息能够泛化至未见说话人，并在控制声学特征后仍保持有效性，但替换外部读出方向通常对生成答案影响甚微。这些结果明确区分了信息可用性与行为使用之间的差异，将性能损失精准定位至决策规则与状态到答案的读出过程。
链接: https://arxiv.org/abs/2608.06409
作者: Linkai Peng,Baorian Nuchged
机构: University of Connecticut (康涅狄格大学); The University of Texas at Austin (德克萨斯大学奥斯汀分校)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Speech language models are increasingly evaluated on paralinguistic tasks by the accuracy of prompted answers, but answer accuracy combines failures at different stages of the audio-to-answer computation. We introduce a generation-aligned diagnostic ladder that compares the emitted answer, the option logits, an affine readout of those logits, and a linear readout of the hidden state at the same answer token. Successive differences separate endpoint, decision-rule, and readout-coverage gaps. Across five systems and two emotion corpora, state decoding exceeds generation by 27.8 accuracy points on average, and both the decision-rule and readout-coverage gaps are positive in all ten conditions. A label-free logit correction improves generated accuracy in every condition, showing that part of the decision-rule gap is actionable. In rank-matched comparisons, emotion information outside the native readout generalizes to held-out speakers and survives controls for measured acoustic descriptors, but replacing the selected readout-external directions usually has little effect on emitted answers. These results distinguish information availability from behavioral use and localize performance losses across the decision rule and the state-to-answer readout.
[NLP-64] EXAS: Task-Expert-Aware Supervision for Downstream Mixture-of-Experts LLM Adaptation
【速读】： 该论文旨在解决混合专家模型（Mixture-of-Experts, MoE）在下游适应过程中任务相关专家识别不准确以及专家激活信号未被有效利用于监督分配的问题。现有方法通常基于专家的使用频率统计来识别任务专家，但这类统计反映的是使用强度而非与任务成功完成的相关性；同时，专家激活模式作为监督信号的潜力尚未充分挖掘。为此，论文提出任务专家感知的监督策略（Task-Expert-Aware Supervision, TEXAS），其核心创新在于结合正确性条件下的任务专家发现与逐标记级别的监督权重分配：TEXAS通过比较基础模型成功与失败样本中各专家的激活情况，保留那些在成功实例中更强激活的专家，并在微调阶段对失败实例中激活这些专家的输出标记赋予更高权重。该方法充分利用了模型原有的路由行为，无需限制适应过程中的专家子集或施加显式的目标路由分布。在三个MoE模型和六个基准测试上，TEXAS在18个设置中取得最佳或并列最佳性能，平均优于最强基线1.3–1.5分。消融实验与深入分析验证了所发现专家的有效性及监督策略的合理性。
链接: https://arxiv.org/abs/2608.06396
作者: Guanzhi Deng,Haibo Wang,Kuan Wu,Xiangru Jian,Shing Yin Wong,Sichun Luo,Zhuoran Wang,Linqi Song
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Mixture-of-Experts (MoE) language models route each token through a small subset of experts, making routing patterns useful for identifying task-relevant experts during downstream adaptation. Yet current approaches have two limitations: task experts are typically identified from aggregate routing statistics that reflect usage rather than association with successful task completion, and task-expert activations remain underexplored as signals for supervision allocation. We introduce Task-Expert-Aware Supervision (TEXAS), which combines correctness-conditioned task expert discovery with token-level supervision allocation. TEXAS compares expert activations on instances that the base model solves successfully and those it fails to solve, and retains experts more strongly activated on successful instances. During fine-tuning, it upweights answer tokens in failed instances when they activate these experts. TEXAS therefore leverages existing routing behavior without restricting adaptation to a fixed expert subset or imposing an explicit target routing distribution. Across three MoE models and six benchmarks, TEXAS achieves the best or tied-best performance in 17 of 18 settings and improves over the strongest baseline by 1.3–1.5 points on average. Ablations and further analyses validate both the discovered experts and the resulting supervision strategy.
[NLP-65] Mitigating Scoring Bias in LLM -as-a-Judge via Random Number Generation
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLM）作为文本质量评估者（LLM-as-a-Judge）时存在的评分偏差（scoring bias）问题，即模型在不同上下文情境下仍倾向于生成特定倾向的评分，从而影响评估结果的客观性。其核心解决方案是通过识别并校正模型的潜在数值偏差（latent numerical bias），具体方法为：在提示中引入下游任务定义，指导模型随机生成数字标记，并通过测量实际生成数字分布与均匀分布之间的偏离程度来量化该模型的潜在数值偏差；随后，在实际评估过程中，基于该偏差对模型针对特定输入的标记生成概率进行校正，以实现更公平、准确的评分。实验在四个不同任务（包括大模型对齐评估、摘要质量评估、语义文本相似度与相关性）上验证了该方法的有效性，结果表明其显著优于未去偏的基线模型及现有校准方法。研究还发现评分偏差在不同模型、任务和评分区间间存在差异，凸显了动态测量与适配潜在数值偏差的重要性。
链接: https://arxiv.org/abs/2608.05726
作者: Yuma Asato,Kiyoaki Shirai,Natthawut Kertkeidkachorn
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Large Language Models (LLMs) are often used as evaluators of text quality, known as LLM-as-a-Judge, which can outperform conventional automatic evaluation metrics that rely on reference texts. However, LLM evaluators tend to generate particular scores regardless of the context of the evaluated text, which is known as scoring bias. This study proposes a novel method to mitigate this scoring bias. An LLM is instructed to randomly generate number tokens, and the latent numerical bias of the LLM is identified by measuring the deviation of the observed distribution of numbers from the uniform distribution. A definition of a downstream task, for which an LLM evaluator is used, is added to the prompts for random number generation to measure task-specific latent number bias. In the evaluation by an LLM, the token generation probabilities for a given input are rectified considering the LLM’s latent number bias. Results of the experiment on four different tasks, evaluation of LLM alignment, evaluation of summarization, Semantic Textual Similarity, and Semantic Textual Relatedness, demonstrate that our proposed method outperforms the baselines, including an LLM without debiasing and previous calibration methods. In addition, it is confirmed that scoring bias varies across LLMs, tasks, and score ranges, indicating the importance of measuring latent number bias as the case may be.
[NLP-66] Artificial Intelligence Can Match Domain Experts in Evidence Extraction and Critical Appraisal of Microbial Oncogenesis Research Publications MICRO
【速读】： 该论文旨在解决当前癌症相关致瘤微生物（oncogenic microbes）的证据分散、难以由人工系统整合的问题，从而阻碍了新型致瘤微生物的发现与疾病负担的降低。其核心解决方案是利用大语言模型（Large Language Models, LLMs）实现可扩展、专家级的系统性证据合成，以识别潜在的微生物-癌症关联对。研究的关键在于构建一个结构化的证据提取与评估模板（涵盖77个问题项，包括多选题、李克特量表、多选与自由文本题型），并基于小鼠乳腺肿瘤病毒（MMTV-LV）与乳腺癌的案例研究，通过领域专家与多个主流LLM（Gemini 2.5 Pro、Gemini 2.5 Flash、GPT-5、GPT-5 Nano）在24篇研究论文上的对比评估，采用新颖的评价指标衡量专家间及专家与LLM之间的共识一致性。结果表明，GPT-5和GPT-5 Nano在各类任务中表现与专家高度一致，其评分分布与专家无显著差异，表明其可作为“额外专家”参与结构化文献评估；而Gemini系列虽表现相近，但在应用致瘤性判定标准上更为宽松。尽管存在少量幻觉，但主要局限仍集中于方法学评估与全文内部矛盾识别等复杂推理任务，提示需进一步优化。该研究验证了大语言模型在自动化系统性证据合成中的可行性，为未来高效挖掘微生物致癌潜力提供了关键技术路径。
链接: https://arxiv.org/abs/2608.07250
作者: Kaela Kokkas,Hairong Wang,Richard Klein,Nazir A. Ismail,Natalie Irwin,Mohammad Z. Moonsamy,Kubendran Naidoo,Jeremy Nel,Ekene E. Nweke,Raveen Parboosing,Emmanuel K. Sekyi,Rebecca T. van Dorsten,Bruce A. Bassett,Robert F. Breiman
机构: University of the Witwatersrand, Johannesburg, South Africa
类目: Quantitative Methods (q-bio.QM); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: Published in Frontiers in Cellular and Infection Microbiology, 45 pages, 14 figures
Abstract:Confirmed oncogenic microbes contribute significantly to cancer burden. Identifying novel microbial oncogenicity could yield strategies that will reduce disease burdens. However, relevant evidence is dispersed and infeasible for humans to comprehensively synthesize. LLMs may enable scalable, expert-level systematic evidence synthesis to identify microbe-cancer pairs; however, such capabilities have not yet been demonstrated. Domain experts were recruited to create a dataset to benchmark LLM performance (Gemini 2.5 Pro, Gemini 2.5 Flash, GPT-5, GPT-5 Nano) on 24 research papers using MMTV-LV and breast cancer as a case study. We devised a structured template for evidence extraction and appraisal, consisting of MCQ, Likert-scale, multi-select, and free-text question types (77 items across 24 papers). Agreement between (1) experts and (2) experts and each LLM was determined per question instance using novel metrics. LLMs were assessed by comparing inter-expert and expert-LLM agreement distributions to determine whether LLMs behaved as additional experts by increasing or maintaining inter-expert agreement. Free-text responses were further evaluated qualitatively. Across all question types, LLM responses aligned closely with experts, with GPT-5 and GPT-5 Nano achieving score distributions indistinguishable from experts. Gemini models behaved similarly but were significantly more lenient in applying microbial oncogenesis criteria. Hallucinations were rare. Methodological appraisal and identification of contradictions within full-texts were the most persistent LLM vulnerabilities. GPT-5 and GPT-5 Nano were indistinguishable from experts on structured domain research paper evaluation tasks. This supports use of LLMs for automated systematic evidence synthesis. However, methodological appraisal tasks and contradiction identification in full-texts remain weaknesses requiring strengthening.
[NLP-67] Genotypic Triggers: Exposing Pharmacogenomic Blind Spots via Host-Specific Backdoors in Generative Antimicrobial Peptide Models
【速读】： 该论文旨在解决生成式人工智能（Generative AI）在抗菌肽（Antimicrobial Peptides, AMPs）设计中潜在的基因型特异性健康风险问题。现有肽生成模型的验证流程忽视了历史药物安全事件中揭示的关键事实：某些药物对携带特定遗传背景个体具有显著健康风险。为此，本文提出一种名为“基因型触发器”（Genotypic Trigger）的后门攻击机制，其核心在于通过操控生成模型，使其输出的肽序列在特定人类白细胞抗原（HLA）等位基因携带者中诱发显著升高的预测免疫原性风险（predicted immunogenicity risk），即引发异常免疫反应。实验表明，该攻击在主流肽生成模型上使目标等位基因携带者的免疫原性风险评分平均提升743%，而对非携带者的影响接近自然基线；同时，这些被污染模型仍保持高抗菌活性和低普遍毒性，因而可规避传统安全性评估。该研究揭示了生成式AI在药物发现中可能被恶意利用的系统性漏洞，并强调需在模型开发与验证中引入基因型敏感性评估。
链接: https://arxiv.org/abs/2608.06779
作者: Doniyorkhon Obidov,Xiaolong Guo,Yonghui Li,Kaichen Yang
机构: Michigan Technological University(密歇根理工大学); Lehigh University(利哈伊大学); Kansas State University(堪萨斯州立大学)
类目: Quantitative Methods (q-bio.QM); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Large Language Models (LLMs) have accelerated drug discovery, particularly in the automated design of antimicrobial peptides (AMPs). However, current validation pipelines for peptide generation models overlook historical precedents showing that certain drugs carry health risks predominantly for individuals with specific genetic profiles. In this paper, we demonstrate that such targeted health risks can be induced intentionally and at scale by manipulating models that generate peptide candidates. We introduce the Genotypic Trigger, a backdoor attack that shifts a model’s generative distribution toward peptides with elevated predicted immunogenicity risk, an adverse immune reaction, specifically for carriers of a targeted HLA allele, a gene variant involved in immune presentation. Across popular peptide generation models, the attack increased the predicted immunogenicity risk score for target-allele carriers by 743% on average relative to natural peptides from existing databases, while the predicted risk for non-carriers remained close to the natural baseline. Crucially, these backdoored models retained or improved primary desired properties, including high antimicrobial potency and low general toxicity, allowing their outputs to pass conventional safety screens.
信息检索
[IR-0] CoinRAG : Contextualized Information Nugget KV Cache Reuse for Long-Context RAG
链接: https://arxiv.org/abs/2608.07458
作者: Gyuwan Kim,Cheoneum Park,Tao Yang
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR); Machine Learning (cs.LG)
备注:
Abstract:Recent optimization studies on Retrieval-Augmented Generation (RAG) have exploited chunk-level KV cache reuse to avoid processing long retrieved contexts for higher efficiency, while significant information redundancy and noise still remain in the coarse-grained chunks. This paper optimizes the Pareto frontier under low prefill latency constraints while maximizing accuracy by proposing CoinRAG (Contextualized Information Nugget KV Cache Reuse for Long-Context RAG). The name metaphorically reflects our core mechanism: much like assembling small tokens (or “coins”) to accumulate a larger value, CoinRAG compositionally reuses offline-computed, fine-grained nugget caches to form a learned contextual representation efficiently in a more semantically relevant but compact manner. Specifically, instead of full-chunk encoding, CoinRAG identifies query-relevant semantic units within retrieved chunks through two-stage retrieval and seamlessly assembles their sliced KV representations with a chunk-level context. Extensive evaluations on LongBench multi-hop question answering tasks demonstrate that CoinRAG significantly reduces operational costs and outperforms the other baselines with a new Pareto frontier and an average 5.3% relative improvement in answer quality (F1) under a standard fast prefill latency budget.
[IR-1] Conformal Coverag e Guarantees for Any Video Temporal Grounder AAAI2027
链接: https://arxiv.org/abs/2608.07434
作者: Aseel Mohamed,Rasul Khanbayov,Erchin Serpedin,Hasan Kurban
类目: Computer Vision and Pattern Recognition (cs.CV); Information Retrieval (cs.IR)
备注: Submitted to AAAI 2027
Abstract:Event boundaries in continuous video are ambiguous: re-annotate the same query-video pair and independent annotators mark moments that overlap by less than half on a large fraction of samples. The ground truth for video temporal grounding is therefore a distribution over intervals, yet every grounder returns a single interval with no statement of reliability, so at deployment a wrong interval is indistinguishable from a right one. COVER changes the output object: a post-hoc, model-agnostic wrapper that turns any grounder, a trained localizer or a black-box video–language model, into one that emits a temporal region containing the true moment with probability at least 1-\alpha , by calibrating the quantile of a temporal nonconformity score on held-out labels and widening the base prediction by that amount. The guarantee is finite-sample and distribution-free under exchangeability, and requires neither retraining nor white-box access. We give two score families, a two-sided boundary-widening score for grounders that emit an interval and a super-level-set score for grounders that emit a relevance signal, and develop theory specific to grounding that bounds how large the certified region becomes, when coverage survives conditioning on event length, and how it degrades when moments from one video break exchangeability. Across three benchmarks and five grounders, realized coverage tracks the target, and calibration exposes what point metrics hide.
[IR-2] GeoBenchLLM : A Comprehensive Benchmark for Evaluating LLM s on Geo-Related Tasks CIKM2026
链接: https://arxiv.org/abs/2608.07411
作者: Rodrigo Ferreira Rodrigues,Karim Radouane,Jose G Moreno,Lynda Tamine
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Information Retrieval (cs.IR); Machine Learning (cs.LG)
备注: Accepted at CIKM2026
Abstract:In the context of geodata, existing Large Language Models have often been studied in a homogeneous setting, which has considerably limited insights into their generalization capabilities. In this paper, we present \benchName, a comprehensive benchmark for probing LLMs on geo-related tasks. We leverage a careful selection of twelve publicly available datasets from diverse geo-related tasks and domains, and evaluate a set of LLMs on geo-spatial and temporal understanding using our benchmark. Our results show that reasoning and size have a strong impact on overall performance. GeoBenchLLM is publicly available at this https URL.
[IR-3] Geo-Spatial Concept Probing of Large Language Models : Abstraction Compositionality and Grounding
链接: https://arxiv.org/abs/2608.07353
作者: Karim Radouane,Jose G Moreno,Lynda Tamine
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR); Machine Learning (cs.LG)
备注: Preprint
Abstract:Understanding concepts is fundamental to generalization. Despite their impressive performance on a wide range of tasks, Large Language Models (LLMs) still struggle with genuine concept understanding. Prior work has evaluated conceptual understanding in LLMs using natural-language benchmarks or narrowly scoped synthetic tasks, but these settings often conflate multiple skills or lack precise control over the underlying concepts and their properties. To support controlled probing of concepts in LLMs, we design tests on their core properties: abstraction, compositionality, and groundness. We set up a concept-centric benchmark, targeting spatial concepts such as direction, distance, topology, and their compositions, and use question answering tasks serving as a proxy. We conduct extensive experiments across multiple LLM architectures and training regimes to analyze how model scale and design impact conceptual understanding. The results reveal clear limitations in current LLMs and provide insights into the factors shaping their ability to acquire and compose structured concepts. Our findings shed light on how concept-based LLMs can be redesigned for improved information access and knowledge management. The code will be available at this https URL.
[IR-4] Exact Adaptive Hybrid Retrieval Without Fixed Top-L Cutoffs
链接: https://arxiv.org/abs/2608.07152
作者: Chunran Zhang
类目: Information Retrieval (cs.IR)
备注: 10 pages, 4 figures, 1 table
Abstract:Modern retrieval-augmented generation (RAG) systems often fuse fixed Top- L results from dense and sparse retrievers, treating later contributions as zero. The cutoff therefore determines both the ranking and its execution cost. Yet truncated fusion is not generally equivalent to complete-list fusion: unread cross-list ranks can change Top- K membership or order even when the observed candidates contain every item in the complete-list Top- K . Because channel rankings vary across queries and corpus updates, a depth selected from historical queries may not transfer reliably. We propose Exact Adaptive Hybrid Retrieval (EAHR), which fixes the ordered Top- K defined by complete-list weighted RRF as the retrieval target and treats channel depth as request-specific execution state. Per-Vector Scalar Quantization (PVS) and Posting Block-Max (PBM) produce resumable exact dense and sparse rankings. Fusion bounds unread contributions and requests further ranks only while they can change the Top- K . Every successful request therefore matches complete-list fusion without a preset Top- L ; otherwise, execution continues safely to list exhaustion. Across five test collections and five temporal corpus snapshots, complete-list weighted RRF remained competitive, whereas fixed depths selected from historical queries did not transfer reliably. EAHR reproduced the complete-list ordered Top-20 in all 150 query-snapshot combinations. Under a warm-cache, interleaved, order-balanced protocol, the paired geometric-mean latency ratios of exhaustive batch execution to EAHR were 23.35 on TREC-DL 2019 and 30.28 on TREC-DL 2020. Anti-correlated rankings exhausted both lists, and some difficult queries were slower with EAHR. EAHR does not guarantee a speedup for every request; it fixes the exact result while adapting execution depth to the current rankings.
[IR-5] Invisible to the Machine: Auditing AI Restaurant Cafe and Bar Recommendation Against a Complete Market Census
链接: https://arxiv.org/abs/2608.07069
作者: Vladimir Pitenin
类目: Information Retrieval (cs.IR); Computers and Society (cs.CY)
备注: 31 pages, 10 figures
Abstract:AI assistants are becoming a primary interface for local discovery, yet almost nothing is known about which venues they surface – especially in food and drink, where recommendations carry direct revenue consequences. We present the first census-denominated audit of AI venue recommendation: a complete enumeration of 4,776 cafes, restaurants, and bars across two bounded markets (Canggu and Ubud, Bali), against which we evaluate 2,208 search-grounded responses from four production AI systems (ChatGPT, Claude, Gemini, Perplexity) to 96 persona-conditioned queries, collected over seven days under a pre-registered protocol. Because we observe the full market, we can measure what sampled audits cannot: 85.6% of venues were never recommended by any system – 72.6% even among established venues with fifty or more ratings. Visibility follows a two-margin structure. Entry into answers is associated with documentation: review volume (OR 1.64), an own website (OR 1.92), listed price information (OR 1.54), and third-party web mentions (OR 1.44) – while star rating is null at this margin (OR 0.89). Rank within answers reverses the pattern: among recommended venues, rating significantly predicts first position (OR 1.17). Presence in an open POI dataset (Foursquare), a folk-theorized visibility factor, shows no positive effect at either margin. Outright fabrication is rare (0.08% of mentions), but systems recommended permanently closed venues 93 times – staleness, not hallucination, is the practical failure mode. Cross-system agreement is low (top-20 Jaccard 0.33-0.54). A two-week test-retest shows cross-period answer similarity comparable to same-day rerun similarity: the churn is sampling stochasticity, not temporal drift. We release our protocol, registry construction method, and derived data.
[IR-6] DocMemo: Dynamic Evidence Discovery via Probabilistic Memory-Guided Retrieval for Multi-Modal Document Understanding
链接: https://arxiv.org/abs/2608.07067
作者: Hanshu Yao,Janfeng Zhong,Niu Lian,Jinpeng Wang
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Information Retrieval (cs.IR); Multimedia (cs.MM)
备注: DocMemo is a memory-guided framework for long-document reasoning that uses tri-level memory and dynamic Bayesian belief updating to overcome static retrieval limits and improve evidence tracking. 16 pages, 4 figures, 14 tables
Abstract:Long-document understanding requires locating sparse and heterogeneous evidence across hundreds of pages, yet existing systems remain limited by static retrieval and fragile cross-round memory. Mainstream single-round methods commit to a fixed top- k page set at the outset and struggle to recover from early retrieval errors; recent iterative approaches allow multi-round evidence acquisition, but they do not investigate the propagation mechanism of cross-round states, making it difficult to track the dynamic changes in page relevance. To address these limitations, we propose DocMemo, a memory-guided framework that formulates long-document reasoning as dynamic evidence exploration. DocMemo maintains a tri-level retrieval state consisting of Document Schema Memory, Page Belief Memory, and Question Episodic Memory, which respectively capture structural priors, dynamic relevance estimation, and query-specific reasoning trajectories. During reasoning, DocMemo continuously refines cross-round page selection through Bayesian page belief updating with Thompson sampling, spatial proximity propagation, and structure-aware adaptive-granularity evidence access, while supplementing page-level evidence with fine-grained visual regions. Experiments on 3 benchmarks show that DocMemo achieves state-of-the-art performance and validate the efficacy of structured memory and dynamic page belief updating. Code is available at this https URL.
[IR-7] acher Retains Full Tokens Student Merges Efficiently: TM20K for E-Commerce Sequence Modeling in Ad Recommendation
链接: https://arxiv.org/abs/2608.07055
作者: Xinchun Li,Duoru Zheng,Wenlin Zhao,Ziyi Zhou,Jingxuan Tan,Huizhi Yang,Linlan Chen,Dongjian Wang,Dongyue Wang,Xiaosong Li,Hongyue Mao,Yaocheng Tan
类目: Information Retrieval (cs.IR)
备注:
Abstract:Benefiting from ultra-long behavior sequence modeling, existing recommender systems bring users a better experience via simultaneously considering their long-term and short-term interests. Nevertheless, extended sequence lengths introduce substantial burdens on training efficiency and serving throughput. Prior approaches typically utilize search-based or cluster-based compression on ultra-long sequences at the cost of fine-grained information, or rely on various lightweight target attention structures incapable of sufficient sequential feature extraction. In this paper, we balance the effectiveness and efficiency for ultra-long sequence modeling via full transformer modeling accompanied with a two-stage knowledge distillation framework. First, both teacher and student models take the full attention mechanism rather than pure target-sequence attention for effective sequence scaling. For student models, we propose several simple yet well-motivated token merge approaches, significantly compressing the sequence length while maintaining an acceptable performance. Then, a one-time teacher is heavily trained with full sequence tokens, further boosting the performance of student models via knowledge distillation. The proposed paradigm named TM20K has been successfully deployed in ByteDance’s e-commerce advertising recommender system that extends the e-commerce sequence length to 20K, delivering substantial improvements in key business metrics (e.g., ADSS +1.036%) while keeping the training and serving cost nearly the same as the online state-of-the-art model (e.g., serving latency only +5.6%).
[IR-8] MISO: Model-Internal-State-Guided Optimization for Ranking Models RECSYS2026
链接: https://arxiv.org/abs/2608.07035
作者: Yongzhe Zhang,Xiaoyu Deng,Yifan He,Mengying Sun,Sheng Luo,Yijia Liu,Hao Yan,Zhuo Li,Yi Meng,Huiping Yao,Swathi Hrishikesh,Jing Chen,Dennis Choi,Steven Liu,Lexi Luo,Keyi Chen,Anish Khazane,Marcio Porto,Xiaoya Wang,Emmy Wang,Kangfu Zheng,Xingyuan Wang,Bilal Fadlallah,Gursharan Singh,Prabhakar Goyal,Alireza Vahdatpour,Santanu Kolay
类目: Information Retrieval (cs.IR)
备注: Accepted at the OARS Workshop at ACM RecSys 2026
Abstract:Ranking models are repeatedly refined within established model families, yet the choice of which component to scale, replace, or retire is often guided by expensive trial-and-error. We present Model Internal State Optimization (MISO), a systems workflow that uses model internal states (MIS), including parameters, activations, gradients, and normalization statistics, to prioritize such local optimization decisions. MISO extracts MIS from a trained ranking model, aggregates them into ranking, alignment, and comparison signals, and converts those signals into a small set of interpretable candidate edits. Because MIS are re-extracted after each retraining cycle, MISO naturally supports an adaptive optimization workflow that tracks evolving model behavior as data distributions and system requirements shift over time. In an ads ranking case study, MISO improves normalized entropy while requiring substantially fewer validation runs than expert-driven and black-box scaling workflows, offering a practical middle ground between manual tuning and opaque automated search.
[IR-9] Hierarchical Quantization with Domain-Adaptive Sparse Routing for Generative Cross-Domain Recommendation
链接: https://arxiv.org/abs/2608.06997
作者: Haiying He,Xiaopeng Li,Yuchen Gu,Kuo Cai,Bo Chen,Jingtong Gao,Yejing Wang,Derong Xu,Ruiming Tang,Guorui Zhou,Han Li,Xiangyu Zhao
类目: Information Retrieval (cs.IR)
备注:
Abstract:Generative Recommendation (GenRec) represents a promising paradigm that achieves remarkable empirical success by encoding items as compact Semantic IDs (SIDs) and modeling user behavior via next-token prediction across diverse recommendation scenarios. Extending this paradigm to cross-domain recommendation is challenging because a unified model must accommodate heterogeneous item semantics and behavioral patterns across domains. Existing methods commonly rely on globally shared representations or lightweight domain adaptation, which may provide insufficient capacity for modeling heterogeneous patterns at different semantic granularities. To address these challenges, we propose HD-Rec, a unified generative framework for cross-domain recommendation. HD-Rec employs a hierarchical domain-aware quantizer that constructs semantic identifiers using globally shared coarse-level codebooks and adaptively routed fine-level codebooks. It further introduces a domain-adaptive sparse mixture-of-experts module that combines a continuously activated shared expert with a dynamically selected specialized expert. To improve the coherence of multi-token item representations, we develop a cross-granularity routing consistency objective that regularizes token-level routing decisions toward their item-level consensus. Experiments on three public cross-domain recommendation benchmarks show that HD-Rec consistently improves over competitive sequential, generative, and cross-domain recommendation baselines.
[IR-10] From Classification to Recommendation: Empirical Analysis of Audio Embedding Models Application for Content-Based Music Recommendation
链接: https://arxiv.org/abs/2608.06928
作者: Qingrui Li,Haowei Lou,Chengkai Huang,Quan Z. Sheng,Lina Yao
类目: Information Retrieval (cs.IR)
备注: 9 pages, 4 tables, 2 figures
Abstract:Pretrained audio representation models learned from large-scale corpora have achieved strong performance in audio classification and understanding. However, most existing models are optimized for objectives such as masked prediction, contrastive learning, or audio-text alignment, which do not necessarily produce representation spaces well-suited to recommender systems. Unlike classification, music recommender systems must capture item relationships shaped by subjective and behavior-dependent listener preferences. Although pretrained audio embeddings have been explored in conventional recommender systems, their effectiveness in the rapidly emerging paradigm of generative recommender systems remains underexplored. To address this gap, we systematically evaluate six representative audio encoders across three types of music recommender systems: content-based, sequential, and Semantic-ID-based generative recommender systems. We further investigate how residual-quantization design, including codebook width, quantization depth, and retained Semantic-ID prefixes, affects the preservation of recommendation-relevant information. Experiments on two music recommendation datasets show that audio-text-aligned and music-domain representations are generally more effective when pretrained embedding geometry is used directly, whereas interaction-based sequential training substantially reduces performance differences among encoders. We also find that increasing Semantic-ID capacity does not consistently improve generative recommender systems and may introduce substantial instability. These findings provide practical guidance for selecting audio encoders and designing audio-derived Semantic IDs for modern music recommender systems.
[IR-11] BZKO: An Ontology for the Card Index of German Post-War Compensation Records
链接: https://arxiv.org/abs/2608.06918
作者: Dilek Yargan,Jörg Waitelonis,Mahsa Vafaie,Harald Sack
类目: Information Retrieval (cs.IR)
备注: 13 pages, 4 figures, accepted to Ontology Showcase at the 16th International Conference on Formal Ontology in Information Systems (FOIS 2026), to be published in Proceedings of the Joint Ontology Workshops (JOWO) - Episode XII: The Tropical Spring of Ontology, co-located with the 16th International Conference on Formal Ontology in Information Systems (FOIS 2026)
Abstract:The Central Federal Card Index (Bundeszentralkartei) of Germany is a key archival resource documenting compensation claims submitted by victims of National Socialist persecution and their relatives, within the German Wiedergutmachung process. To enable semantically enriched representation, integration, and reuse of this historically significant collection, we present the BZK Ontology (BZKO). We propose a two-layer ontology for historical archival data that separates ontologically grounded domain semantics from interoperability-oriented extension constructs. The approach combines BFO-based realism with archival standards (RiC-O, PROV-O, PiCo), enabling provenance-preserving semantic integration, while maintaining logical rigor, modularity, and reuse across digital humanities infrastructures. The proposed approach establishes a reusable semantic foundation for the integration of Wiedergutmachung archival materials into digital humanities infrastructures and lays the groundwork for future knowledge graph generation, ontology validation, and the incorporation of additional historical entities and uncertain temporal and spatial information. The ontology is available on this https URL.
[IR-12] Georeferencing Non-Gazetteered Place Names using Biological Specimen Records
链接: https://arxiv.org/abs/2608.06884
作者: Aneesha Fernando,Surangika Ranathunga,Kristin Stock,Raj Prasanna,Christopher B. Jones
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
备注: Accepted for publication in the proceedings of the Conference on Spatial Information Theory (COSIT) 2026
Abstract:Biological specimen records collected by natural history institutions constitute a rich source of temporal geographic knowledge, capturing biodiversity information about regional landscapes as they were recorded at different times. Using digitised data from the Allan Herbarium (New Zealand), this study identifies place names in these specimen locality descriptions that are absent from current gazetteers; we refer to these as non-gazetteer place names (NGPs). These place names are typically historical, vernacular, or colloquial and were used as landmarks to describe a specimen’s location at the time of collection. We then investigate the problem of georeferencing the NGPs using only the limited information available in the specimen records. To resolve this, we leverage repeated occurrences of the same place name across specimen records with different specimen locations and spatial relation terms, extracting and inverting these relations to derive constraints on NGP locations. This approach is instantiated within deterministic, probabilistic, and LLM-based methods, enabling a comparative analysis of their strengths and limitations for text-based spatial inference. On a pseudo-NGP benchmark, probabilistic inference achieves the highest accuracy (median error 1.43 km; A@1 km 36%), while the LLM yields competitive but less precise estimates (median error 1.80 km; A@1 km 31%), indicating that, despite advances in LLMs, traditional modelling remains advantageous when high spatial precision is required.
[IR-13] Rhetorical-Role-Aware Retrieval-Augmented Generation for Legal Question Answering over Indian Supreme Court Judgments
链接: https://arxiv.org/abs/2608.06828
作者: Sayed Ayaan Ahmed Sha,Sangeetha Sivanesan,Anand Kumar Madasamy,Navya Binu
类目: Information Retrieval (cs.IR)
备注:
Abstract:This research paper proposes a Retrieval Augmented Generation (RAG) framework that is specific to the legal field in order to assist interactive retrieval and reason about judgments from the Supreme Court of India. The solution uses an enhanced version of RAG framework which consists of rhetorically based chunking, fusion-based retrieval, and cross encoder reranking methods to increase the relevancy of the information retrieved. In order to improve conversations, the proposed framework uses chat history along with query classification and rewriting in order to understand user intention from successive queries. Additionally, there are features that take into account structural aspects of legal documents, such as isolated names of judges that could have an impact on retrieval quality. The evaluation was done using the DeepEval framework and demonstrated strong performance on metrics including contextual recall and answer relevancy, which proves that the framework is very effective in dealing with legal question-answering tasks that require a lot of context. The results emphasize the importance of domain specific enhancements in developing legal AI systems that are both reliable and explainable.
[IR-14] Progressive Alignment of Recommender Foundation Model through Multi-Phase Post-Training RECSYS’26
链接: https://arxiv.org/abs/2608.06792
作者: Oseong Choi,Hoeinn Kim,Jihoon Lee,Byungsoo Kang,Taeyeong Jang
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI)
备注: 9 pages, 3 figures. Accepted to the 20th ACM Conference on Recommender Systems (RecSys '26), Industry Track
Abstract:Foundation model(FM) for recommendation has shown strong ability to model long-horizon sequential user behavior. In practice, a single pretrained foundation model is often adapted to diverse downstream serving surfaces through Supervised Fine-Tuning(SFT). However, optimizing task-specific objectives such as clicks or likes does not necessarily align the serving policy with the business metrics that determine recommendation quality. We propose a three-phase progressive post-training framework that explicitly separates downstream adaptation from business-metric alignment. The adaptation stage is decomposed into Linear Probing(LP) and Full Fine-Tuning(FFT): LP first stabilizes randomly initialized downstream heads within a frozen pretrained representation space, and FFT then jointly specializes the full model for the target task. On top of this stabilized policy, Reinforcement Fine-Tuning(RFT) aligns the model with practical business objectives using a learned reward model. Rather than directly optimizing the serving policy on sparse business targets, we train the policy on dense implicit feedback and use business-metric supervision only for reward modeling. Offline experiments show that the progressive LP-FFT-RFT framework outperforms single-phase alternatives, and that reward-based alignment yields a stronger serving policy than directly using the reward model itself for ranking. Large-scale online A/B tests further show that the proposed framework improves production recommendation quality over a conventional non-foundation baseline. A reference implementation is available at this https URL Comments: 9 pages, 3 figures. Accepted to the 20th ACM Conference on Recommender Systems (RecSys '26), Industry Track Subjects: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI) Cite as: arXiv:2608.06792 [cs.IR] (or arXiv:2608.06792v1 [cs.IR] for this version) https://doi.org/10.48550/arXiv.2608.06792 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Related DOI: https://doi.org/10.1145/3773078.3831870 Focus to learn more DOI(s) linking to related resources
[IR-15] Pre-Inference Routing for Cost-Efficient Document Field Extraction
链接: https://arxiv.org/abs/2608.06607
作者: Sreerekha Rajendran
类目: Computation and Language (cs.CL); Information Retrieval (cs.IR); Machine Learning (cs.LG)
备注: 9 pages, 5 figures. Code: this https URL
Abstract:Most document-extraction systems use a single model for all documents. This is simple but can be costly for easy cases and less effective for difficult ones. We examine whether we can predict a document’s difficulty before extraction using inexpensive, document-based signals, and use this to choose between a cheaper and a stronger extractor. We find that routing only helps if two conditions hold: the cheaper model fails often enough to make routing worthwhile, and those failures can be predicted from visible features such as image quality and layout. We turn these into a practical test and apply it to five genres. When both conditions are met, the calibrated router reduces cost by 31-33% on receipts and 77% on degraded ad-buy forms while keeping quality within 0.02 F1 of always choosing the large model. Routing does not help if either condition is missing, as with clean digital invoices or nutrition labels that are already easy to read. A small labeled pilot can predict whether routing will work, and in the two cases where we ran it first, the prediction was correct. A simple bag-of-words router works about as well as engineered features, showing that the main limit is the genre, not the router design; we use interpretable features to help explain which genres can be routed. The router must be retrained for each dataset and does not transfer across datasets, even within the same genre. These results hold for two model pairs with cost differences of 5x and 3x.
人机交互
[HC-0] PsychoAgent : An Affect-Sensitive Cognitive Architecture for Conflict-Aware Memory in LLM Agents
链接: https://arxiv.org/abs/2608.07438
作者: Mohammad Amanlou,Parham Abed Azad,Farbod Davoodi,Mostafa Masumi,Behnam Bahrak,Abdol-Hossein Vahabie
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Human-Computer Interaction (cs.HC)
备注: 12 pages main paper + 10 pages supplementary material; supplementary material included
Abstract:Human-like cognition does not select past experience by topical similarity alone: affective significance and unresolved conflict also shape what becomes accessible. We present PsychoAgent, a cognitive architecture for LLM agents that separates factual and affective memory and integrates both through a conflict-aware executive controller. Affective memories are first filtered by semantic relevance and then re-ranked by salience, preserving topical fit while allowing emotionally important traces to enter the prompt. Across three controlled conflict scenarios, the full architecture retrieved more conflict-critical memories than semantic-affective and single-memory RAG baselines (0.933 vs. 0.500 and 0.667), with a small semantic-similarity cost. Five blinded raters evaluated 27 outputs. After within-rater standardization, the full architecture had the highest overall mean (+0.22 SD), but corrected pairwise differences were not significant. A three-day illustrative trace further shows persistent affect, offline memory recombination, and selective memory reweighting. The findings support affect-sensitive retrieval as an inspectable mechanism for modeling human-like conflict effects in LLM agents.
[HC-1] Hands-Off or Hands-On? Variation in Area Chair Practices and Implications for AI Support
链接: https://arxiv.org/abs/2608.07425
作者: Ines Arous,Neha Nayak Kennard,Andrei Mircea,Emily Kuang,Jackie Chi Kit Cheung,Andrew McCallum
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Area chairs (ACs) play a critical role in the peer-review process, managing conflicts and ensuring fair outcomes. Although AI tools have been proposed to support ACs, little is known about the challenges they face and their perceptions of these technologies. In this paper, we conduct interviews including a design probe with 27 ACs in AI to explore their challenges, strategies, and perspectives on potential AI tools. Through thematic analysis, we identify key tensions arising from the growing volume of submissions, uneven reviewer expertise, and the complex task of managing the relationship between reviewers and authors. Most importantly, we find substantial variation in how ACs engage with submissions and influence outcomes: some adopt a largely hands-off approach, while others take a more hands-on role in guiding discussions and decisions. This variation challenges the notion of a single, universal AC practice and highlights the need to account for diverse approaches. When reflecting on the potential use of AI tools, ACs expressed a cautious stance, drawing on their domain knowledge and heightened awareness of AI limitations. From these findings, we derive three design implications: tailoring AI assistance to diverse AC practices, design assistance for discussion moderation, and embedding human-centered AI principles that preserve human agency in decision-making.
[HC-2] Beyond Call and Response: Modelling Reciprocal Coordination in Human-AI Vocal Ensembles
链接: https://arxiv.org/abs/2608.07376
作者: Polina Proutskova
类目: Human-Computer Interaction (cs.HC); Sound (cs.SD)
备注: 5 pages, 2 figures
Abstract:Musical interaction with AI is often organised as a response loop: a human performs, the system interprets that action, and the system answers, accompanies, or schedules a musical event. Unconducted vocal ensembles pose a different problem. Singers act simultaneously and continuously affect one another; neither timing nor pitch is fixed by a conductor, metronome, accompaniment, score, or tuning source. Collective organisation emerges from many-to-many reciprocal adjustment. This paper frames such ensembles as coupled dynamic systems and proposes a research architecture for vocal agents that enter, rather than merely track, their collective states. Some target repertoires are metrical, while others exhibit non-isochronous temporal contours that cannot be reduced to a beat grid; we treat the latter as a hard case for a general framework. The architecture connects multichannel capture in the field to dialect- and singing-aware representation, collective-state inference, vocal generation, and in-situ evaluation. The resulting agenda asks not only whether an artificial singer can synchronise, but how its presence reorganises human coordination, leadership, style, and musical transmission.
[HC-3] Incidental Visualizations: Augmented Reality as a Medium for Contextual Information
链接: https://arxiv.org/abs/2608.07271
作者: Matilde Heitor,João Moreira,Daniel Gonçalves
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:In today’s fast-paced world, delivering information efficiently and unobtrusively is essential. While ambient and glanceable visualizations provide real-time data, they can increase cognitive load and disrupt primary tasks. We investigate incidental visualizations, a novel concept in information visualization designed to present contextually relevant information briefly and spontaneously, with minimal user interaction. Augmented Reality offers an ideal medium for this integration, embedding visualizations directly within the user’s environment. Through controlled user studies on logic-based game tasks (Sudoku and Connect 4), this work compares ambient, periodic, and incidental visualization patterns in terms of comprehension accuracy, performance, and disruption. Results indicate that IVs deliver information as effectively as ambient displays while minimizing disruption, highlighting their potential for adaptive, context-aware information delivery in AR environments.
[HC-4] Exact Computation of Trait-induced Merge Trees for Bivariate Fields
链接: https://arxiv.org/abs/2608.07181
作者: Petar Hristov,Ingrid Hotz,Talha Bin Masood
类目: Computational Geometry (cs.CG); Graphics (cs.GR); Human-Computer Interaction (cs.HC)
备注: 12 pages, 10 figures, To appear in an IEEE VGTC workshop “TopoInVis Connect 2026 - Topology meets Artificial Intelligence”
Abstract:Trait-induced merge trees (TIMTs) provide a robust topology-based method for selecting and browsing feature level sets in multivariate data by analyzing the distance field induced by a user-specified trait in attribute space. Existing TIMT computations typically sample this distance field at mesh vertices and assume piecewise-linear interpolation, although the Euclidean distance-to-trait function is generally not piecewise linear on the original mesh. As a result, the resulting merge tree may miss zero-valued features and may perturb the locations and values of minima and merge events. We study the exact computation of TIMTs for piecewise-linear bivariate fields, focusing first on point traits. We show that the restricted sublevel sets inside each tetrahedron are convex and therefore have trivial local merge-tree structure, implying that global topological changes arise only through gluing across simplex boundaries. Based on this observation, we construct a weighted graph whose merge tree is isomorphic to the exact merge tree of the induced distance field. We further relate TIMTs to Jacobi sets, showing how nonzero edge events of the TIMT are localized by the singular structure of the underlying bivariate map. We establish a theoretical upper bound on the error of the vertex-sampled linear interpolation, expressed in terms of the maximum length of projected mesh edges in the range. We discuss extensions to line, line-segment, and finite point-set traits, and implement the method robustly using CGAL and VTK, demonstrating results on both synthetic and real-world datasets.
[HC-5] PHOENIX: Fine-Tuned SLM-Powered Autonomous Satellite Lifetime Extension via Predictive Self-Healing and Multi-Agent AI Recovery
链接: https://arxiv.org/abs/2608.07126
作者: Sumaiya Islam,Harsha Kumara Moraliyage
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注: 6 pages, 2 figures. Accepted at IEEE IRAI 2026 (International Conference on Responsible Artificial Intelligence)
Abstract:Most CubeSats, small and low-cost satellites roughly the size of a shoebox, do not survive as long as they were designed to: a study of 178 missions found that only 48-65% remain operational after two years, against a designed lifetime of 2-5 years. The deeper issue is that a CubeSat in low Earth orbit (LEO) is physically unreachable from the ground for roughly 85 minutes out of every 96-minute orbit, so faults that start during that window go unnoticed until the next contact pass, by which point recovery may no longer be possible. We propose PHOENIX (Predictive Health On-orbit Edge Neural Intelligence eXtension) to give the satellite its own fault reasoning capability. A fine-tuned Small Language Model (SLM) compact enough to run on embedded hardware is deployed onboard the CubeSat, running on the flight-proven Aethero NxN-ECM computer, monitoring all sensor readings continuously, and resolving recurring faults using a memory system that stores past repairs so the same inference does not need to run twice. Once per orbit it sends a short structured health report to the ground instead of a raw data dump; six specialized AI agents on the ground read that report and generate validated satellite commands within the 5-10 minute contact window. A generative diffusion model (DDPM) creates synthetic training data because real fault examples make up only 0.57-1.80% of the dataset. We report preliminary results on the ESA Anomaly Detection Benchmark (14 years, 76 channels, 118 labeled faults).
[HC-6] hermodynamic Human-Computer Interaction
链接: https://arxiv.org/abs/2608.07123
作者: Uzafir Ahmad Rafaq,Muaz Hassan,Ali Muzaffar
类目: Human-Computer Interaction (cs.HC); Performance (cs.PF); Data Analysis, Statistics and Probability (physics.data-an)
备注:
Abstract:Traditional human-computer interaction models rely on domain-specific techniques to model target prediction; models designed for cursor interaction prediction fail to generalize to mobile interfaces and vice versa. We introduce a unifying framework grounded in thermodynamics, proposing that human interaction is composed of phases in thermodynamic equilibrium and non-equilibrium. To demonstrate this, we derive Fitts’ law and the proposed target prediction model from equilibrium thermodynamics by assigning kinetic and potential energies to a moving agent and target. Subsequently, we analyze the shortcomings of the prediction model and Fitts’ law in edge cases, such as predicting intent for large targets. This analysis demonstrates that large targets cannot be accurately modeled using equilibrium thermodynamics. The proposed model scales across interaction modalities without modification, requires zero training data, and evaluates in constant O(1) time. Furthermore, we show that design properties such as the color of a button act as independent parameters that influence the attractive force exerted on an agent. Applied to live web prefetching tasks, the framework achieved an efficient Fetch:Click ratio of 1.37 and predicted the user’s target with an accuracy of 98.1%.
[HC-7] UncertaintyVis: Preserving Linguistic Uncertainty in Automated Text-to-Chart Generation
链接: https://arxiv.org/abs/2608.07093
作者: Songheng Zhang,Emily Aurelia,Anthony Tang
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Data-rich documents pair narrative text with quantitative claims, and authors routinely qualify those claims with linguistic uncertainty markers such as “nearly,” “approximately,” or “at least.” Automated text-to-chart systems discard these markers, producing visualizations that appear definitive even when the source text expresses hedged or incomplete knowledge. Readers may then over-interpret precision and misjudge author intent. We present UncertaintyVis, a system that preserves linguistic uncertainty during automated chart generation. A formative corpus analysis of 211 uncertainty expressions across 12 documents and 8 domains yielded a four-category taxonomy: Surface Form Normalization, Precision Boundaries, Inferential Derivation, and Non-Inferable Gaps. We mapped each category to chart-specific visual encodings that signal uncertainty without disturbing the spatial integrity readers rely on, and implemented an end-to-end pipeline pairing large language model text analysis with uncertainty-aware rendering. In a two-part study with 12 participants, readers matched charts to source text with 85% accuracy and text to charts with 76%. Uncertainty-aware visualizations trended toward lower cognitive demand (effect sizes 0.460 and 0.769 for mental demand and effort), and 75% of participants preferred them to plain text, describing explicit uncertainty encodings as a basis for verifying data claims. Encoding effectiveness varied by chart type: bar and pie encodings performed consistently, while line chart encodings require redesign.
[HC-8] Human-Centered Explainable AI for TinyML Edge Devices: A Pareto-Based Selection Framework with LLM -Guided Design
链接: https://arxiv.org/abs/2608.07091
作者: Zeinab Dehghani,Dhavalkumar Thakker,Koorosh Aslansefat,Kuniko Paxton,Bhupesh Kumar Mishra,Baseer Ahmad,Rameez Raja Kureshi
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注:
Abstract:Edge Artificial Intelligence (Edge AI) enables the deployment of AI models directly on local edge devices, while such deployments are subject to strict resource constraints, particularly in clinical applications requiring local and timely inference. In such contexts, explainable artificial intelligence (XAI) can serve as a human-AI interface intended to support healthcare professionals’ and patients’ understanding of model predictions and informed decision-making. To fulfill this role, XAI method selection for TinyML deployments can be formulated as a human-centered multi-objective design problem that jointly considers qualitative stakeholder preferences, explanation quality, and proxy-based deployment cost. We propose a framework that integrates a large language model (LLM)-guided design interface that maps qualitative stakeholder preferences to candidate XAI methods, followed by deterministic feasibility filtering and Pareto-based optimization. The framework exposes trade-offs among explanation fidelity, stability, and proxy-based deployment cost while characterizing their implications for explanation quality and estimated deployment feasibility. A proof-of-concept evaluation on a skin lesion classification task illustrates how the framework systematically compares candidate XAI methods and identifies Pareto-efficient trade-offs. The present evaluation covers the computational selection stages, while physical MCU deployment and empirical human-expert validation remain outside the scope of this study.
[HC-9] AutoIntervene: Calibrated Intervention for Action-Chunking Imitation Learning Policies
链接: https://arxiv.org/abs/2608.07065
作者: Jinhe Tang,Weiming Zhi
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Human-Computer Interaction (cs.HC); Machine Learning (cs.LG)
备注: 9 pages, 7 figures
Abstract:Action-chunking visuomotor policies learn from demonstrations and improve temporal consistency by predicting short action sequences rather than single-step commands. Yet perception errors and execution drift can move the robot outside the demonstration distribution, while the policy continues to produce smooth action chunks that are inconsistent with the observed state. We present AutoIntervene, an online framework that selectively transfers control between an action-chunking policy and an operator during deployment. AutoIntervene evaluates proposed chunks against a visual-action support memory built from successful task executions, combining visual similarity with consistency between proposed and reference actions. Phase-local support governs policy-to-operator transfer within the current task phase, whereas global support governs the return to policy control after operator recovery. We calibrate separate switching thresholds for the two directions from empirical quantiles of evaluation-level scores on held-out expert demonstrations, avoiding direct manual tuning of score cutoffs. Intervention segments retained from successful rollouts target learner-induced states and provide corrective supervision for subsequent policy updates. Experiments on real-world bimanual manipulation tasks show higher post-adaptation task success and lower operator-control time than manual intervention. Videos and additional results are available at this https URL.
[HC-10] XGait: A Multi-Modality Wireless Sensing Dataset for Indoor Human Tracking and Identification
链接: https://arxiv.org/abs/2608.07064
作者: Wei Xu,Zhu Wang,Yifan Guo,Changlong Cheng,Yin Zhang,Zhihui Ren,Bin Guo,Zhiwen yu
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Wireless sensing has emerged as a promising approach for tracking and identification using commodity Internet of Things devices. However, the features derived from a single wireless modality are often fragile to variations in environmental layouts and walking trajectories. Furthermore, most existing studies are based on datasets collected in specific scenarios with limited trajectory diversity and sensing modalities, preventing a robust evaluation of system generalization. \textcolorblueTo address this gap, we introduce \textbfXGait, a multi-modality wireless sensing dataset that synchronously captures human walking using Wi-Fi and acoustic transceivers across three indoor scenarios, with vision-based measurements serving as ground truth. Specifically, XGait contains more than 22K walking samples from 27 participants, covering diverse directions and trajectories to support both indoor tracking and identity recognition. To bridge the heterogeneity of wireless sensing modalities, we propose a unified Doppler spectrogram representation that maps Wi-Fi and acoustic signals into a shared time–frequency space, along with a standardized benchmark pipeline for pre-processing, temporal alignment, and feature construction, enabling reproducible evaluation and systematic cross-modal analysis. Extensive evaluations demonstrate that Wi-Fi and acoustic sensing exhibit complementary strengths, particularly under complex trajectories and challenging propagation conditions, thereby paving the way for novel research in the field of multi-modality wireless sensing. The dataset and code are available at this https URL.
[HC-11] Switched Reading: Toward Seamless Visual-Auditory Switching When Reading Text in Augmented/Mixed Reality
链接: https://arxiv.org/abs/2608.06985
作者: Kazuyuki Fujita,Yuto Matsui,Ikuru Sato,Guanghan Zhao,Yoshifumi Kitamura
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Augmented/mixed reality (AR/MR) wearable glasses now permit information interaction anywhere, but visual displays can be inappropriate when real-world awareness is essential. We propose Switched Reading, a novel interaction framework for reading text in AR/MR that supports switching between visual and auditory modalities as needed. Specifically, we explore two key interaction techniques within this framework: (1) gaze-based voice playback and (2) a correspondence-aware transition effect. We implemented them on an MR headset through a parameter-tuning user test. Next, we conducted a user study (N=16) to investigate the impact of the two techniques on reading performance and overall user experience with simulated modality switching in virtual reality. The results show that the condition combining both techniques was the most preferred among four conditions. Moreover, we found that the gaze-based voice playback reduced gaze offsets when switching modalities and improved reading speed over the baseline condition using scroll position. Finally, we implemented a Switched Reading application for reading while walking and collected user feedback, yielding further design implications for practical use.
[HC-12] Social Facilitation of Creative Reflection: AI-agents and Humans
链接: https://arxiv.org/abs/2608.06980
作者: Olga Sutskova,Corey Ford
类目: Human-Computer Interaction (cs.HC)
备注: In Proceedings of The First Reflection in Creative Experience (RiCE) Workshop (RiCE W1) arXiv:2607.24558
Abstract:Social collaboration can support people’s reflection and is a crucial component of creativity. Creative technologies have been designed to support more collaborative ways of working, including using AI to simulate social partners. As human-AI creative collaborations increase, further investigation is needed into how different social interactions influence creative reflection and at which stage a social intervention is crucial to improve creative outcomes. Considering that non-verbal communication is the bedrock of human cognition and influence, non-verbal social dynamics should be examined in detail in the age of AI-companionship. For example, during social interaction, the social facilitation effect describes how the mere presence or observation of others influences how a person behaves and feels in the context. Whether changes in technology-mediated social environments influence how people reflect on their creative work needs further exploration, as does whether social AI-companionship elicits similar effects as humans. This paper discusses how theoretical mechanisms relating to social facilitation could influence creative practice and reflection, proposing ways of further testing these effects.
[HC-13] he Nocturnity Scale: Measuring the Sense of Being at Night in Virtual Urban Environments
链接: https://arxiv.org/abs/2608.06904
作者: Anthony Le Gourri{é}rec(Nantes Univ - ECN, Nantes Univ - ENSA Nantes, CRENAU, Lab-STICC_INUIT, IMT Atlantique),Etienne Peillard(IMT Atlantique - INFO, Lab-STICC_INUIT, IMT Atlantique),Nicolas Houel(AAU),Myriam Servi{è}res(CRENAU, Nantes Univ - ECN, Nantes Univ - ENSA Nantes)
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Nighttime environments are increasingly used in virtual urban studies, yet darkness alone does not fully recreate the subjective sense of being at night. Prior work suggests that this experience depends not only on the absence of daylight, but also on lighting structure, low-light perception, human activity, soundscape, and self-related states. However, no existing tool directly assesses this scene-dependent subjective experience. This work introduces nocturnity as the feeling of being at night elicited by a scene and proposes a theory-driven framework structured into three subscales: perceptual, activity, and inner-state nocturnity. Based on this framework, we develop a first candidate questionnaire for virtual urban environments. Developed through a literature-informed process and reviewed by two urban lighting experts, the scale comprises 42 Likert-type items, including three diagnostic subscales and complementary global and time-related items. This work provides a first operational basis for comparing virtual urban scenes according to their perceived nocturnity and supports future empirical validation.
[HC-14] How Should I Pick a Foundation Model for My Robot? In Favor of a Community Evaluation Framework for Social Robots
链接: https://arxiv.org/abs/2608.06898
作者: Eric Nichols,Alva Markelius,Hatice Gunes
类目: Robotics (cs.RO); Computation and Language (cs.CL); Human-Computer Interaction (cs.HC)
备注: 5 pages, 1 figure, 1 table. Accepted at the FoRMA workshop (Foundation Models in the RO-MAN Age: Responsible Development for Social Robotics) at IEEE RO-MAN 2026, Kitakyushu, Japan. Workshop homepage: this https URL
Abstract:Researchers who seek to build social robot applications on foundation models are faced with a difficult question: how should we pick a model? Public leaderboards offer little guidance: the demands of real-time, embodied social interaction lie largely outside their focus. And direct evaluation is impractical at scale: each embodied study requires scarce participant, robot, and experimenter time. In this paper, we identify five evaluation dimensions for foundation models in social robots: (i) conversational competence, (ii) user safety, (iii) embodied character, (iv) target scene effectiveness, and (v) audience appropriateness. To make model selection cheaper and better informed, we propose a three-tiered evaluation funnel paradigm that first filters with general metrics, then extends to simulated interactions, and terminates in more expensive, robot-specific evaluation. We map all five dimensions across all three tiers, chart where applicable evaluation methods exist and are missing, and close with a call to action: let’s build the evaluation framework together as a community.
[HC-15] Not Always Top-Left: Untangling the Signals that Guide Dashboard Reading Order
链接: https://arxiv.org/abs/2608.06845
作者: Nicole Sultanum,Vidya Setlur
类目: Human-Computer Interaction (cs.HC)
备注: main (11 pages) + appendix (5 pages). TVCG, to appear
Abstract:Dashboards are widely used interfaces for data analysis, combining multiple visualizations, text, and interactive controls within a single view. While dashboard authors often structure layouts to suggest a logical consumption flow, users may interpret and navigate dashboards differently depending on the interplay between design features, analytical goals, and personal preferences. In this work, we investigate how people make sense of dashboards by examining their reading orders, i.e., the sequences in which users engage with dashboard components. We conduct a mixed-methods study with 18 dashboard authors and 16 end-users, capturing how participants design for and reason through these component transitions. Through qualitative and quantitative analyses of participant-generated flows, we outline a set of factors that influence dashboard reading order, including layout, visual saliency, semantics, functional roles, interaction, and user context. We also identify emergent reading patterns and analyze them through aggregate and variability measures, revealing where users converge and diverge in their interpretations. Finally, we discuss implications and opportunities for computational approaches that aim to automatically model, guide, or serialize dashboard consumption.
[HC-16] Fact-Check Your Information (FYI): A Design Probe to Understand How People Actually Fact-Check Data-Driven Articles IEEE-VIS2026
链接: https://arxiv.org/abs/2608.06804
作者: Nguyen-Truong Thinh,Yuxuan Du,Phongsakon Mark Konrad,Arpit Narechania
类目: Human-Computer Interaction (cs.HC)
备注: 11 pages, 4 figures, 4 tables. To appear in IEEE VIS 2026
Abstract:Data-driven journalism and policy reports frequently rely on statements grounded in statistical evidence, referred to as data claims. Verifying such a claim requires connecting it to the underlying structured dataset. However, existing systems typically isolate automated fact-checking from manual data exploration, leaving it unclear how readers coordinate AI assistance with manual inspection of the evidence in practice. We present FYI, a browser extension that embeds fact-checking in the reading environment, and use it as a design probe to study how people detect, verify, and determine the validity of data claims against the underlying dataset. FYI provides four complementary tools spanning the spectrum from full automation to manual data exploration. In an exploratory study (N=22), participants used FYI to fact-check claims in a data-driven article. We find that participants adopted three distinct workflow archetypes—AI-first with manual confirmation, manual-first with AI supplement, and parallel co-review—with visualization serving as the primary mechanism for auditing AI conclusions. Trust in AI shifted dynamically, growing when multiple tools converged and eroding when AI outputs were inconsistent. These findings suggest that fact-checking systems should treat AI as a starting point that human verification complements rather than a definitive authority, elevate visualization as a core verification capability, and support flexible, user-driven workflows. We release FYI as open-source software for further research at this https URL.
[HC-17] CustomDance: Customized 3D Dance Generation with Coarse-to-Fine Human-Centered Interactive Control
链接: https://arxiv.org/abs/2608.06722
作者: Xulong Tang,Kaixing Yang,Xiaohu Guo,Prabhakaran Balakrishnan,Rawan Alghofaili
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:With the rise of AI-generated content (AIGC) and advanced techniques for 3D human representation, the task of generating 3D dance movements has become an exciting area of research. Despite significant advancements, current methods often fail to provide comprehensive and distinct control over various multimodal inputs from users, such as music or specific descriptions of desired movements. As a result, the generated motions may be statistically plausible and technically correct, but they often lack depth, expressiveness, and alignment with the user’s creative vision. To address this issue, we present CustomDance, a coarse-to-fine interactive system designed for customized 3D dance generation. Inspired by the workflows of expert choreographers, CustomDance introduces a novel paradigm to AI-assisted choreography through three interconnected stages. First, a multimodal Large Language Model (MLLM) analyzes the music and a high-level text prompt to identify key temporal anchors and creative cues for the piece. Next, for each anchor, a multimodal retriever suggests high-quality motion clips from a dance library based on local music and text, empowering the user with concrete and predictable options. Finally, a custom music-conditioned diffusion in-painter seamlessly connects the selected phrases, allowing for iterative, user-guided refinement of the final composition, supported by visualizations of motion dynamics. Our evaluations demonstrate that CustomDance not only highlights the significant creative utility and empowering potential of our AI-assisted choreography paradigm, but also outperforms competitive baselines across quantitative and qualitative comparisons.
[HC-18] WebRider: Persona-Conditioned Intent Controllers for Live-Web Assistance
链接: https://arxiv.org/abs/2608.06704
作者: Zhi Li,Tao Zhou,Yeqing Li,Eugene Ie,Demetri Terzopoulos
类目: Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注:
Abstract:Delegating a web task involves more than asking a question; it requires transferring a policy: what to verify, how to handle uncertainty, which preferences matter, and when to stop. Yet, current live-web agents are evaluated solely on the final answer, ignoring the policy constraints that define the delegation. A plausible final answer can conceal violations of that policy. Our full live audit reveals this critical gap: a strong controller completes 99.2% of tasks but honors all policy constraints in only 38.8% of cases. Finishing does not imply fidelity. WebRider bridges this gap by formalizing the delegated policy as an intent contract—an operational record of goals, constraints, evidence obligations, answer form, and task-local persona controls that must hold even as web pages change. WebRider employs a hierarchical architecture: a top-layer controller maintains the contract, a middle layer realizes intentions as guarded executable actions, and a tool layer executes these actions via browser, search, and maps tools. Our benchmark, RiderBench, evaluates this design on 4,096 live-web contracts across 42 public websites, auditing both the internal contract state and the visible user experience to determine if a rollout preserved its policy and if the steps were persona-consistent. The guarded middle interface also serves as a high-quality training signal; an 8B action-policy model trained through this interface outperforms executable-only baselines under a fixed controller. By making the browsing path a first-class object, WebRider enables a system that is auditable, human-judgeable, and learnable without conflating action realization with final-answer decisions. Dataset URL: this http URL.
[HC-19] Octos Adventure: At-home Deployment of a Pediatric Education Tool
链接: https://arxiv.org/abs/2608.06684
作者: Crimson Olaleye,Neda Barbazi,Ji Youn Shin,Gurumurthy Hiremath,Carlye Anne Lauff
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Children with congenital heart disease (CHD) and their families often navigate educational, emotional, and communication challenges regarding care management outside clinical settings. However, many pediatric health intervention studies rely on caregiver-mediated feedback or structured evaluations, limiting visibility into children’s direct experiences during everyday use. We present a one-week at-home deployment study of Octo, a hybrid physical-digital educational tool designed to support children’s health literacy and reduce parental educational burden. Extending participatory design approaches into deployment, the study embedded child-friendly reflection activities, drawing-based feedback, and emotional tracking into family routines. We conducted the deployment with 13 families of children diagnosed with CHD. Preliminary findings suggest that children gained understanding across Octo’s physical and digital components, expressed emotional attachment and confidence through play, and engaged in collaborative family communication about their condition. These findings will guide future refinement of Octo and inform future clinical deployment across care settings.
[HC-20] RACE: A Multi-Layer Benchmark for Human AI Controller Coordination Under Drift and Failure ICML
链接: https://arxiv.org/abs/2608.06657
作者: Joshua Zuniga,Srinivasan Subramanian,Ramya Madhuri Narapureddy,Md Abdullah Al Hafiz Khan
类目: Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注: This work was accepted for presentation and publication at the 25th IEEE International Conference on Machine Learning and Applications (ICMLA 2026)
Abstract:Modern cyber-physical and AI-assisted systems couple human operators, AI decision modules, and automated controllers in a single control loop, so trustworthiness depends on the whole loop, not any one model. Yet no standard benchmark captures time-aligned, multi-layer traces of how drift and failures propagate across these layers, so we cannot diagnose where coordination breaks down, why, or how to recover. This paper targets one facet of that gap: drift, a deviation that can originate in any stack layer and that conventional single-modality monitoring cannot localize to a layer or pin to an onset time. We construct a benchmark by injecting controlled drift into traces derived from ALFRED, a grounded-instruction benchmark for everyday household tasks, yielding 1,918 drifted traces. Each trace is a time-aligned sequence of per-step records across five execution layers (state, observation, decision, rules, control), labeled with the drift type, affected layer, onset time, responsible actor, and causal mechanism, and validated by independent raters with inter-annotator agreement reported. We pair the dataset with a leak-aware protocol that removes a near-perfect onset leak, and a baseline study across classical, recurrent, and attention-based model families. Under this honest protocol, drift is identifiable and attributable well above random and majority baselines across every family (affected layer macro-F1 near 0.70, responsible actor near 0.85, causal mechanism near 0.49), and heavy attention offers no advantage over simpler models on this symbolic benchmark.
[HC-21] Agent ic AI: User Empowerment or Enclosure?
链接: https://arxiv.org/abs/2608.06510
作者: David Gamba,Daniel M. Romero,Grant Schoenebeck
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注: Extended version of AIES’2026 publication
Abstract:Agentic AI promises a more flexible form of digital agency: systems that can act on users’ behalf, from filtering content to negotiating prices to selecting services. Whether it will empower users is an open question, and we argue that the answer depends on more than the technology. We conduct a comparative case analysis of four more mature domains where similar forms of agency arose: browser-based ad blockers, platform recommender systems, financial robo-advisors, and email spam governance. Across the cases, decisions about whose interests agents would serve were resolved through technical arrangements: API choices, protocol governance, industry standards, and default configurations. Beyond their technical form, these were political decisions. We identify this as depoliticization, a concept from political theory, here at work in technological systems. Its most consequential effect is that individual outcomes and collective contestation capacity can move in opposite directions: spam inbox quality improved substantially while the organized capacity to contest spam governance collapsed. Where intermediary institutions sustained adversarial challenge, user-aligned agency proved more durable; where proprietary infrastructure and closed standard-setting absorbed contestation, displacement compounded. We apply this to agentic AI, where governance arrangements consolidating around the Model Context Protocol and the Agentic AI Foundation are settling these configurations before the choices that define what agents can do move outside the reach of users and the public.
[HC-22] DREAMS: Diverse Reactions of Engagement and Attention Mind States Dataset
链接: https://arxiv.org/abs/2608.06382
作者: Monisha Singh,Gulshan Sharma,Ximi Hoque,Abhinav Dhall
类目: Human-Computer Interaction (cs.HC); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Active attention and engagement are important in improving users’ learning experiences. Engagement refers to the level of involvement and interest individuals show towards a particular task. Attention, on the other hand, refers to a state where someone is entirely focused on a particular task with conscious awareness. Engagement and attention are different but closely linked concepts and can influence each other bidirectionally. To explore the relationship between user engagement and attention, we introduce the Diverse Reactions of Engagement and Attention Mind States (DREAMS) dataset. The dataset includes facial video recordings of 32 users in naturalistic settings watching various stimuli to evoke diverse emotions. We then analyze user engagement and attention states in these videos by framing it as a classification problem, exploring single-task, transfer learning task, and multi-task settings. In single and transfer learning task settings, separate networks are applied to predict engagement and attention states. Whereas in multi-task settings a shared network is applied, which jointly learns to predict both engagement and attention states. Moreover, we examine participants’ performance on video-based questionnaires and evaluate their perceived cognitive workload. In our findings, we observe (a) better classification performance in predicting engagement states in both transfer and multi-task learning compared to single-task learning and (b) higher engagement and attention states correlate with lower cognitive load and improved task performance. The dataset and the code are publicly available and can be accessed through this https URL.
[HC-23] Mobile Interaction for Assessing Fatigue Sleep and Activity in Neurodegenerative and Chronic Diseases
链接: https://arxiv.org/abs/2608.06380
作者: Julian Fierrez,Alejandro Peña,Aythami Morales,Ruben Tolosana,Ruben Vera-Rodriguez,Meenakshi Chatterjee,Ahmaniemi Teemu,Wan-Fai Ng,Walter Maetzler,Nikolay V. Manyakov,Jennifer Kudelka,Ralf Reilmann,C. Janneke van der Woude,Kristen Davies,Victoria Macrae,IDEA-FAST Consortium
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注: IEEE Conf. on Computers, Software, and Applications (COMPSAC), 2026
Abstract:Fatigue, sleep, or disturbances in daily activities are common symptoms among patients with neurodegenerative disorders (NDD) and immune-mediated inflammatory diseases (IMID). The current assessment of such symptoms is usually conducted using patient reported outcomes (PROs) based on standardized questionnaires that patients usually complete every few months. This assessment protocol has raised some concerns, due to its propensity to exhibit biases derived from its subjectivity nature, or the low sensibility to changes, which may lead to a failure when trying to capture variability over time. In this work, we explore the use of smartphone data, which can serve as a proxy for how patients interact with their devices, to provide an effective, reliable, and objective assessment of the symptoms mentioned above. Our study comprises data from 137 participants belonging to 6 different disease groups, plus a healthy control group. We conducted statistical analysis based on repeated measures correlation, in which we analyze the correlation between screen-time and app-usage features with scores obtained from the PROs collected from the participants using a smartphone application.
[HC-24] Preventive Care Recommendations by Large Language Models
链接: https://arxiv.org/abs/2608.06379
作者: Eden Avnat,Elia Yanko,Ori Yoran,Raja-Elie E. Abdulnour
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Preventive care services (PCS) extend life, yet physicians often underprioritize highly effective interventions such as lifestyle modifications (Zhang et al., JAMA Network Open 2020). We evaluated whether large language models (LLMs) replicate and augment physician prioritization of PCS under time constraints. Using Zhang et al.‘s validated survey with two patients assessed during long and short visits, we compared seven LLMs with historical physicians. We generated 137 simulated physician personas matching cohort demographics and tested three prompts per model. Primary outcomes were concordance with physician rankings, measured by Spearman correlation, and Consensus-Stratified Agreement (CSA), the proportion of LLM selections rated 4 or higher that matched physician consensus across agreement strata. Secondary outcomes included life-years gained per prioritized choice (LYGPC), consistency, and selectiveness. Augmentation was assessed by having models revise physician rankings under three informative prompts, with delta LYGPC quantifying impact. LLMs closely mirrored physicians (mean Spearman = 0.83, SD = 0.11), with high CSA at extreme agreement ranges (94%, 197/210) but low CSA in moderate ranges (21%, 30/140), where they underprioritized lifestyle services (8.8% vs. 38% rated 4 or higher; P .001). Several models exceeded physicians in LYGPC and consistency while being more selective. Time constraints affected physicians and LLMs similarly, increasing LYGPC and selectiveness but reducing consistency. Augmentation effects varied by model. Current LLMs reproduced physicians’ time-sensitivity and base-rate prioritization while exacerbating underprioritized lifestyle interventions. Some models improved prioritization performance, but consistent augmentation will require value-aligned training, explicit time-constraint representation, and prospective real-world validation.
[HC-25] Multimodal Drivers Emotion Recognition and Safety-Oriented Intervention for Intelligent Transportation Systems ITSC2026
链接: https://arxiv.org/abs/2608.06378
作者: Chang Liu,Dalai Mengke,Hanbo Zhou,Jia Hu,Peter Mihajlik,Tamas Sziranyi
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注: 6 pages, 2 figures, IEEE ITSC 2026
Abstract:Driver emotions can affect risk perception, decision-making, and vehicle control under complex road conditions. Existing studies mainly focus on driver emotion recognition, while limited attention has been given to context-aware intervention that jointly considers driver emotion and road perception. This paper proposes a safety-prioritized multimodal driver assistance framework that analyzes speech-derived emotional cues and visual road conditions to generate structured driving interventions. The framework first provides road safety reminders and then generates emotion-aligned verbal support. We construct a multimodal dataset by aligning emotional speech signals with structured road environment descriptors and introduce the CARE (Context-Aware Road-Emotion Evaluation) score to jointly evaluate emotion recognition, risk identification, and intervention generation. Experimental results show that the proposed framework balances environmental risk reporting and emotion-aware verbal regulation, providing a feasible safety-driven direction for intelligent transportation systems.
[HC-26] Uncovering expert objectives in production planning via inverse optimization: An industrial case study
链接: https://arxiv.org/abs/2608.07398
作者: Shivi Dixit,Rishabh Gupta,Adam Kelloway,John Wassick,Qi Zhang
类目: Optimization and Control (math.OC); Human-Computer Interaction (cs.HC); Machine Learning (cs.LG)
备注:
Abstract:Production planning in the manufacturing industry often relies on the use of optimization models, but defining an appropriate objective function can be a challenge. In practice, planners must balance competing goals, manage uncertainty, and account for qualitative business preferences that are difficult to quantify. As a result, many optimization models fail to match expert behavior, limiting trust and adoption. In this work, we propose a data-driven inverse optimization framework to infer the objective function implicitly captured in expert planners’ decisions. We formulate the production planning problem as a mixed-integer linear program, where the unknown objective function is represented as a weighted sum of hypothesized cost terms. A suboptimality-loss-based inverse optimization method is then applied to learn the objective weights from historical production plans. The proposed approach is applied to a real industrial case provided by Dow, where the inferred weights reveal that avoiding inventory shortages and maintaining consistent cycle lengths dominate the planners’ decision-making. Time- and product-dependent extensions further improve predictive accuracy and uncover evolving priorities. Expert interviews confirm the practical validity of these insights. Overall, this study shows that inverse optimization can transform tacit human expertise into interpretable models, enabling more accurate and trusted decision-support tools for complex industrial systems.
计算机视觉
[CV-0] SimWAM: A Simple World Action Model for End-to-End Autonomous Driving
链接: https://arxiv.org/abs/2608.07468
作者: Zongchuang Zhao,Xin Zhou,Tianyang Xu,Zhengyang Sun,Kaixuan Zhou,Honglin Li,Dingkang Liang,Xiang Bai
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: The code and model weights are available at this https URL
Abstract:World-Action Models (WAMs) improve end-to-end autonomous driving by transferring video dynamics priors to action prediction, but existing methods require costly future generation at inference. We present SimWAM, a simple yet effective WAM that uses video generation purely as a training signal. It co-trains a pretrained video expert and a lightweight action expert with joint flow matching. An isolated attention mask keeps action prediction independent of future frames, allowing the video branch to be discarded after training and leaving a self-contained planner that directly predicts trajectories. Since the two experts share no parameters and interact only through a unified attention interface, the video backbone could be replaced and the action expert scaled independently without modifying the learning objective or inference pipeline. We further apply reinforcement learning to optimize a compositional driving reward beyond trajectory imitation. Our SimWAM achieves 91.5 PDMS on NAVSIM, surpasses state-of-the-art WAM-based planners with substantially lower latency, and transfers zero-shot to nuScenes. These results position SimWAM as a simple yet solid baseline that could readily benefit from advances in video generation for efficient autonomous driving. The code and model weights are available at this https URL
[CV-1] MirrorWorld: Taming Video Diffusion Models for Mirror Reflection Generation
链接: https://arxiv.org/abs/2608.07463
作者: Youjun Zhao,Alex Warren,Gary K.L. Tam,Rynson W.H. Lau
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Project Page: this https URL
Abstract:Recent advances in video diffusion models (VDMs) have enabled high-fidelity video synthesis. However, generating mirror reflections remains challenging because the content within a mirror must remain consistent with the surrounding scene. Existing VDMs are not specifically designed to model scene-to-mirror relationships, which can lead to reflections with incorrect content or inconsistent spatial arrangements. We observe that mirror reflection generation involves two complementary challenges: determining what scene content should be reflected and how the reflected content should be spatially arranged within the mirror region. Motivated by this observation, we propose MirrorWorld, a reflection-aware video inpainting framework that models scene-to-mirror relationships during generation. Specifically, we introduce Semantic Relation Distillation (SRD), which transfers relational information from a frozen visual foundation model to encourage semantic associations between visible scene content and mirror regions. We further propose Geometric Transformation Alignment (GTA), which learns a transformation that guides the spatial arrangement of reflected content. The two components play complementary roles, with SRD modeling what should be reflected and GTA modeling how it should be arranged. To facilitate research on this problem, we construct a benchmark for video mirror reflection generation by repurposing four existing video mirror datasets into a unified reflection reconstruction task. Experimental results show that MirrorWorld achieves improved reflection reconstruction quality over representative image-based reflection generation methods and strong video inpainting baselines.
[CV-2] I Seek You in Videos: Identity-Conditioned Queries for Person-Centric Video Reasoning
链接: https://arxiv.org/abs/2608.07417
作者: Shibo Gao,Chongxiao Wang,Chenglong Huang,Jie Ma,Haolin Shi,Fei Ding,Jing Li,Qiang Lyu,Yangyang Liu,Yang Liu,Jun Liu,Linlin Huang,Peipei Yang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted to ACM Multimedia 2026 (MM '26). 6 figures, 5 tables
Abstract:Real-world video reasoning often involves multimodal, multi-source inputs, whereas existing video reasoning tasks typically assume a simplified video-text setting, limiting identity matching and person-centric reasoning. To bridge this gap, we introduce the Identity-conditioned Queries (ICQ) task, in which models are required to jointly associate and interpret an input video and a reference image of a person, and leverage this conditioning to address identity grounding, behavior understanding, and temporal reasoning, among other challenges. Building on ICQ, we present ISYV (I Seek You in Videos), a systematic solution comprising three components: (1) ISYV-Bench, a challenging evaluation benchmark with 1,377 real-world complex videos and 1,377 question-answer pairs, organized into six difficulty levels spanning capabilities from identity recognition to causal reasoning; (2) ISYV-75K, a large-scale training set of 75K high-quality samples constructed via automated annotation, multi-stage verification, and manual review; and (3) ISYV-Framework, containing an ICQ-oriented model and training strategy for learning to exploit informative video shots without additional shot-level annotations. Extensive experiments show that both mainstream closed-source and open-source MLLMs struggle on ISYV-Bench, especially in cross-domain identity matching and long-horizon tracking. ISYV-Model outperforms strong baselines and in some aspects approaches closed-source performance. Overall, ISYV provides a unified task definition, scalable datasets/benchmarks, and modeling insights for person-centric video reasoning.
[CV-3] UniJEPA: A Unified Joint-Embedding Predictive Architecture for Task-Agnostic Visual World Modeling ICML2026
链接: https://arxiv.org/abs/2608.07409
作者: An Lanji,Dawei Liu,Jin Li,Haoran Xu,Mei Chen,Yu Tian
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 10 pages, 7 figures; Accepted by ICML2026
Abstract:Joint-Embedding Predictive Architectures (JEPAs) have emerged as a principled framework for self-supervised learning of world models in compact latent spaces, yet existing methods are fragmented: some predict masked parts of a single image in latent space (I-JEPA), others learn to predict global photometric transformations (Image World Models), while video-scale JEPAs predict future temporal states and are post-trained for action-conditioned planning (V-JEPA~2, DINO-World, DINO-WM). These objectives are treated as distinct recipes with separate encoders, predictors, and anti-collapse regularizers, hindering a single model from unifying image-level and video-level world modeling. We present UniJEPA, a unified JEPA that jointly learns photometric prediction (image-level transformations) and temporal prediction (video-level next-state dynamics) in one shared latent space. A single end-to-end objective, composed of a next-embedding prediction loss and a Gaussian regularizer, yields a provably anti-collapse encoder-predictor pair trainable from raw pixels without EMA, stop-gradient, or pre-trained encoders. We show that the same latent space supports controllable abstraction: photometric prediction learns invariant structure while temporal prediction learns equivariant dynamics. After action-conditioned post-training on offline trajectories, UniJEPA enables zero-shot planning by treating goal features as prediction targets. On image, video, and control benchmarks, UniJEPA matches or surpasses task-specific JEPAs while requiring a single loss hyperparameter, and plans up to tens of times faster than generative world models at comparable accuracy.
[CV-4] Addressable Memory for Video World Models
链接: https://arxiv.org/abs/2608.07408
作者: Xindi Wu,Sven Elflein,James Lucas,Olga Russakovsky,Laura Leal-Taixé,Despoina Paschalidou,Jonathan Lorraine,Aljoša Ošep
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Project page: this https URL
Abstract:We study visual persistence in interactive video world models. These models rely on a Key-Value (KV) cache as a growing visual memory to carry forward previously generated frames. However, we find that models can no longer reliably address stored content once rollouts extend beyond the training horizon, because temporal Rotary Positional Embeddings (RoPE) offsets then fall outside the range seen during training and the model struggles to retrieve the relevant visual information through attention. Moreover, naively compressing the cache in the RoPE-rotated space corrupts memory by averaging together incompatible positional phases. To address this, we propose WorldTrace, a training-free memory framework for long-horizon visual persistence. WorldTrace keeps compressed memory addressable by assigning each summary slot a distinct, in-distribution virtual position. Within this addressable cache, we study two memory compression approaches: WorldTrace-Field compresses history for temporal coherence, while WorldTrace-Landmark stores verbatim scene traces at detected transitions for episodic recall. We further introduce LoopBench, a benchmark evaluating whether a compressed cache can reconstruct a previously visited scene after a long detour. WorldTrace-Field improves temporal consistency by +15.5%, and WorldTrace-Landmark improves episodic recall by +19.5% on LoopBench, extending visually persistent generation without retraining.
[CV-5] GeoDistill-Refine: Silhouette-First Geometry Distillation for Annotation-Free Spacecraft Segmentation
链接: https://arxiv.org/abs/2608.07405
作者: Yonglong Zhang,Zongwu Xie,Yang Liu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Foundation segmentation models can provide supervision for spacecraft imagery without manual training masks, but their predictions vary with textual prompts and may contain geometric errors that are amplified during distillation. This paper presents GeoDistill-Refine, a two-stage framework that transfers offline SAM 3 pseudo-masks to a compact segmentation network. Six fixed prompts are fused by an unweighted 50% vote to stabilize the teacher output. The student first learns the foreground silhouette and is then refined with signed-distance-field, skeleton, and area objectives derived from the pseudo-mask. A sample-level gate, computed from prompt agreement, the valid-prompt ratio, and pseudo-mask area plausibility, reduces the influence of unreliable pseudo-geometry. On the SpaceSense-Bench HJM lockbox set, GeoDistill-Refine improves Image IoU and Boundary F1 by 0.0456 and 0.1380, respectively, over a plain pseudo-label student. External evaluations on the SPEED+ Lightbox and Sunlamp domains and on TANGO show competitive regional overlap together with gains in boundary quality or foreground precision. The deployed TinyUNet contains 0.263 M parameters and requires approximately 1.1 ms per image on an RTX 4090; SAM 3 pseudo-mask construction and the auxiliary geometry branches are used only during training.
[CV-6] SkySeaLand: A Wide-Format Satellite Transportation Benchmark with an Ultra-Lightweight Detection Baseline
链接: https://arxiv.org/abs/2608.07382
作者: Md. Zahid Hasan Riad,Md Sultanul Islam Ovi
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Satellite object detection is challenged by small targets and wide-format scenes that lose detail under standard square-input resizing. We introduce SkySeaLand, a public dataset of 1,307 high-resolution satellite images and 19,101 verified bounding boxes across airplane, boat, car, and ship classes in terrestrial and maritime scenes. Native COCO and YOLO annotations are provided. The collection is dominated by large source images and wide scene geometry: 84.5 percent exceed 3,836 pixels on the longest side and 73.1 percent are near a 3:1 aspect ratio. We evaluate twelve detectors from the YOLO, RT-DETR, DETR, and Faster R-CNN families using a common split and COCO metrics. The tested YOLO and RT-DETR variants obtain 84.4–88.2 mAP50, with no consistent accuracy gain from larger parameter counts under the reported model-specific recipes. We also report SkyDet, a 1.22 M parameter anchor-free baseline that obtains 60.5 mAP50 and 24.32 mAP50-95 in a 4.90 MB footprint, with 13.74 ms latency (72.8 FPS) on a Tesla T4. SkySeaLand provides a compact benchmark for mixed land–maritime transportation detection, while SkyDet establishes a documented low-footprint reference rather than a state-of-the-art accuracy claim.
[CV-7] Depth-Wise Probing and Pruning of the Planning Token in a Driving Vision-Language-Action Model ECCV2026
链接: https://arxiv.org/abs/2608.07361
作者: Harisankar Babu,Benjamin Coors,Christopher Lang,Hendrik Berkemeyer,Tamim Asfour,Simon Foell
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at the 6th DriveX Workshop (Foundation Models for Autonomous Driving), ECCV 2026. 14 pages, 8 figures, 4 tables
Abstract:Vision-language-action (VLA) models route driving decisions through a deep language model, but it is unclear how much of that depth the action itself requires. We study a representative driving VLA whose entire plan is carried by a single planning token that a generative planner decodes into a trajectory. Borrowing the planner as a trajectory-space logit lens, we decode the planning token from every one of the 32 decoder layers and measure two signals: the linear decodability of the navigation command and trajectory compatibility with the frozen native planner. Our diagnostic shows that semantic intent is linearly decodable early: command-probe accuracy reaches 97.7% after the first decoder layer, compared with 16.7% chance. In contrast, compatibility with the frozen native planner improves gradually across depth, with open-loop Avg-L2 reaching its minimum of 2.11,m only at the final layer. Learned readouts from the first layer recover much of this gap, indicating that planning information is already present early but is not yet represented in the format expected by the deployed planner. Ranking decoder layers by the angular deviation they induce in the planning token permits removal of 8 of 32 layers within an approximately 5% relative open-loop error increase and yields a measured 1.33 \times decoder speedup. At the evaluated sample size, no family-specific degradation is statistically resolved. These findings are limited to the evaluated ORION checkpoint and Bench2Drive setup.
[CV-8] H2AL: Hyperbolic Hierarchy-aware Aggregative Learning for Registration-based Few-shot Medical Image Segmentation
链接: https://arxiv.org/abs/2608.07340
作者: Jia Wang,Jiaming Cai,Zunying Hu,Zhanjie Wu,Jinyuan Liu,Hua Cheng,Yun Peng
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 10 pages, 9 figures. Accepted at ACM Multimedia 2026 (MM '26)
Abstract:Registration-based Few-shot medical image segmentation (RFMIS) aims to generate pseudo-labels for unlabeled images by warping a labeled image through registration. However, existing methods primarily perform pixel-level optimization and inference in Euclidean space, treating anatomical structures as flat and disjoint. This neglect of inherent hierarchies degrades pseudo-label quality and weakens the discrimination of ambiguous regions, limiting the segmentation performance. To overcome this challenge, we propose a Hyperbolic Hierarchy-aware Aggregative Learning framework for RFMIS, termed H2AL, that enhances both deformation plausibility and anatomical discrimination for dual-task learning. Specifically, we introduce a Hyperbolic Hierarchy-aware Infusion (H2I) module, which leverages the hierarchical modeling capability of hyperbolic space to learn precise hierarchy-aware representations via transformation-guided supervised hyperbolic contrastive learning, and injects such hierarchical priors into Euclidean space through a gated infusion block while preserving semantic richness. Furthermore, we propose an end-to-end joint optimization algorithm by gradient aggregation, where the gradients from the registration and segmentation decoders, embedding semantic and hierarchical cues, are aggregated to update the shared encoder to promote collaborative learning across tasks. Extensive experiments on two anatomical regions, with five experimental settings, demonstrate the effectiveness and efficiency of our method in both registration and segmentation. The code is publicly available at this https URL.
[CV-9] EMPO: Semantic-Action Decoupled RL Post-Training for Vision-Language-Action Models
链接: https://arxiv.org/abs/2608.07314
作者: Ziheng Liu,Quantao Yang
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Vision-language-action (VLA) models are commonly adapted to downstream manipulation tasks via supervised fine-tuning (SFT) or online reinforcement learning (RL) post-training. SFT is prone to distribution mismatch, and existing RL approaches typically apply a single, uniform update strategy to all model components, ignoring their distinct functional roles. We propose TEMPO, a semantic-action decoupled, two-timescale RL post-training framework for VLA models. TEMPO freezes the pretrained vision-language backbone to preserve general semantic representations, and restricts adaptation to two components with dedicated RL optimization loops: the semantic projection layer and the low-level action expert. We update them at different rates–the semantic projection layer infrequently, to keep the latent action stable, and the action expert frequently, to rapidly incorporate control feedback from online interaction. This decoupling RL fine-tuning strategy prevents fast policy updates from destabilizing high-level semantic representations while still allowing the action expert to learn efficiently from online feedback. Experiments on the CALVIN benchmark and real-world manipulation tasks demonstrate that TEMPO consistently outperforms both pretrained state-of-the-art VLA models and the RL post-training baseline, while reaching and maintaining higher evaluation rewards on two real-world tasks.
[CV-10] Same Attention Different Truths: Put Logit-Lens over Visual Attention to Detect and Mitigate LVLM Object Hallucination CVPR2026
链接: https://arxiv.org/abs/2608.07302
作者: Zichuan Wang,Songlin Yang,Bo Peng,Zhenchen Tang,Yang Li,Beibei Dong,Jing Dong
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: CVPR2026 Highlight
Abstract:Large Vision-Language Models (LVLMs) often suffer from object hallucination, generating objects that are absent from the image. Prior work largely attributes this to insufficient visual attention. However, we find that both real and hallucinated objects receive equally strong visual attention in the model’s mid-to-late layers, suggesting that the key issue may not be how much the model attends, but what it attends to and why. To this end, we decode the visual features of high-attention regions using Logit Lens, and observe that regions corresponding to real objects can be correctly decoded to the target object tokens, whereas those for hallucinated objects cannot. Building on this, we identify two hallucination mechanisms: (i) visual uncertainty, triggered by semantically similar or confusable regions; masking these regions eliminates the hallucination. (ii) contextual prior, triggered by strong co-occurrence priors; even when the initially attended region is masked, the hallucination persists and attention drifts to other regions. Based on these findings, we propose a simple yet effective training-free Detect-Mitigate framework comprising a Logit-Lens Consistency Check to detect hallucination and targeted remedies: High-Attention Regions Masking (HARM) for visual uncertainty hallucination, and Visual Evidence Enhanced Decoding (VEED) for contextual prior hallucination. Our approach achieves state-of-the-art results on multiple hallucination benchmarks. Code will be available.
[CV-11] EliSeg: Verified Target Construction for Report-Grounded Abnormality Segmentation
链接: https://arxiv.org/abs/2608.07299
作者: Chengyi Peng,Haoyu Yang,Meixing Shi,Yuxiang Cai,Yankai Jiang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Radiology reports describe clinical observations but do not specify executable segmentation targets. They may contain present, negated, prior,uncertain, or irrelevant findings, while multiple valid abnormalities may coexist. Existing segmentation methods largely bypass this ambiguity by receiving a target identity or spatial prompt before inference, which acts as a hidden target oracle. We study report-grounded abnormality segmentation, where a model must determine target eligibility, cardinality, and finding-to-mask correspondence directly from an unfiltered report before delineating the corresponding regions. We propose \textbfEliSeg, an atcor–verify–revise framework that integrates target construction with mask generation. A grammar-constrained Actor proposes target slots and masks, an independent text-only Verifier reconstructs the eligible finding inventory, and Revision selectively re-executes the shared Actor when their target structures disagree. EliSeg requires no predefined target identity, finding prompt, point, or bounding box. Experiments on MIMIC-CXR-ILS show that EliSeg consistently outperforms direct segmentation methods and extract-then-segment cascades across findings, while effectively suppressing masks for ineligible report mentions. Ablation studies confirm the complementary roles of verification and revision, and evaluation on CheXlocalize demonstrates effective transfer of the EliSeg to an external this http URL is available at this https URL.
[CV-12] Foundation Models Adaptation for Multi-View Multi-modal Cardiac MRI Segmentation and Direct Ejection Fraction Estimation
链接: https://arxiv.org/abs/2608.07291
作者: Sina Amirrajab,Cian M Scannell,Volker Vehof,Michael Bietenbeck,Ali Yilmaz
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Foundation models have shown strong transferability in cardiac MRI (CMR), but their effectiveness for heterogeneous multi-view and multi-sequence CMR analysis remains unclear. In this work, we explore the effectiveness of fine-tuning and combining different CMR foundation models for the Universal Multi-Sequence, Multi-Center and Multi-View CMR Segmentation (CMR-Multi) Challenge. CineMA was fine-tuned for cine and late gadolinium enhancement (LGE) segmentation across short-axis and long-axis views. For direct left-ventricular ejection fraction (LVEF) estimation, we used two recent frozen CMR foundation models to extract embedding vectors that were then combined using attention-based multiple-instance learning for LVEF regression. In the challenge validation set, cine segmentation achieved Dice scores of 0.862, 0.883, and 0.902 for short-axis, two-chamber and four-chamber cine MRI, respectively. LGE segmentation achieved Dice scores between 0.621 and 0.846 across views. The direct LVEF regression model achieved an MAE of 4.96 percentage points and a Pearson correlation of 0.91. These results indicate that foundation models can be effectively adapted and combined for multi-view CMR analysis, while accurate LGE scar segmentation remains a challenging task.
[CV-13] WNM-3D: A World Navigation Model with 3D Scene Conditioning for Closed-Loop VLN
链接: https://arxiv.org/abs/2608.07267
作者: Yuehao Huang,Yunzi Wu,Xiaotao Zhang,Xinhai Li,Jiankun Dong,Jiajun Lv,Chi Zhang,Chenjia Bai,Yong Liu,Xuelong Li
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注:
Abstract:Recent vision-language navigation (VLN) systems increasingly adapt pretrained vision-language models (VLMs) into vision-language-action (VLA) policies that map egocentric observations and language instructions directly to navigation actions. Although semantically capable, such action-centric training does not explicitly model how the agent’s visual observations should evolve under its predicted motion. Generative world-action models (WAMs) jointly predict future observations and actions, yet existing WAMs for continuous VLN do not condition joint future-view and action generation on geometry-aware representations inferred from the observed history. We present WNM-3D, a generative World Navigation Model with 3D scene conditioning for continuous VLN. To consolidate past observations into persistent scene context, a frozen feed-forward geometry encoder extracts geometry-aware representations from the monocular egocentric RGB history, and a trainable 3D Scene-to-Token Adapter converts them into a fixed-length prefix in the token space of the world-action Diffusion Transformer. Through block-causal attention, this prefix conditions every future video-action block, providing a shared geometric context for both future-view and action generation. We train WNM-3D through supervised world-action fine-tuning on A*-generated demonstrations, DAgger-style adaptation on policy-visited states, and DanceGRPO-based closed-loop policy optimization. Experiments on GN-Bench show that WNM-3D outperforms strong VLM-based navigation policies and its 2D-conditioned counterpart in closed-loop navigation. On a fixed near-goal evaluation set, WNM-3D also achieves higher flow-action consistency and lower visual-motion error.
[CV-14] CANIS: Generation-Assisted 3D Canonicalization via an Image-Semantic Bridge
链接: https://arxiv.org/abs/2608.07256
作者: Kendong Liu,Yuxin Yao,Junhui Hou
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Canonicalizing 3D object orientation is fundamental to 3D understanding and analysis. Existing approaches often rely on geometric cues, although 3D canonicalization ultimately requires a semantically meaningful orientation. To address this gap, we propose CANIS, a category-agnostic, generation-assisted framework that introduces the semantic orientation prior of a frozen image-to-3D generative model into 3D canonicalization, without canonicalization-specific training or category-specific templates. Specifically, CANIS first renders the input object from candidate viewpoints, selects an informative view, and generates a proxy in a canonical orientation. During generation, a sparse structural latent encoded from the input guides the proxy to preserve the geometry of an object. CANIS then uses the selected image as a semantic bridge between the input and the proxy. Image patches identify semantic regions on the proxy, and depth back-projection locates the corresponding regions on the input. The resulting semantic anchors constrain geometric matching, from which we estimate the rigid transformation that canonicalizes the input. Experiments on synthetic benchmarks validate CANIS and its key components, while qualitative results on partial observations and OmniObject3D suggest its applicability to incomplete and real-world scans. CANIS also improves downstream 3D classification, part segmentation, and dense correspondence under arbitrary rotations. Project page: this https URL.
[CV-15] Flow-Corrected Shape Optimization: Taming Manifold Drift in High-Dimensional 3D Models
链接: https://arxiv.org/abs/2608.07199
作者: Emilien Seiler,Nicolas Talabot,Yingxuan You,Federico Stella,Pascal Fua
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 31 pages, 16 figures
Abstract:Optimizing 3D shapes within the latent spaces of deep generative models is fundamental to computer assisted engineering, yet remains prone to a critical failure mode we term manifold drift: the tendency of gradient-based optimization to move latent vectors away from the manifold of valid shapes. This problem is exacerbated in state-of-the-art 3D shape generative models that operate in increasingly high-dimensional latent spaces where valid shapes occupy a vanishingly small fraction of the full space. Existing mitigation strategies, including latent regularization and flow-matching approaches, either sacrifice expressiveness, demand a difficult trade-off between objective guidance and generative fidelity that remains prone to manifold drift, or are computationally infeasible to scale to modern, large-capacity 3D shape models. We introduce a novel optimizer-corrector framework that alternates between gradient steps for objective minimization and guided flow matching to drive the latent state back to the valid shape manifold. By decoupling objective minimization from flow-based correction, optimizing freely and correcting strictly, this alternating design avoids inherent trade-offs, preserving geometric validity without sacrificing expressiveness while remaining computationally feasible on modern 3D shape models. We demonstrate its effectiveness across generative priors of varying complexity, from simple vector latent spaces to large-scale architectures across a variety of downstream optimization tasks, including aerodynamic drag reduction and object compliance optimization.
[CV-16] An AI4AI Framework for Visual Token Pruning
链接: https://arxiv.org/abs/2608.07193
作者: Zhen Liu,Wenli Huang,Wei Song,Yuhan Liu,Zhiqin Yang,Jingwen Fu
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Visual-token pruning can substantially reduce the inference cost of multimodal large language models (MLLMs), yet existing methods largely rely on fixed, handcrafted heuristics and costly expert trial and error. As pruning objectives, budgets, and model architectures diversify, manually navigating the expanding design space becomes increasingly difficult. This paper aims to build an AI4AI framework for visual-token pruning by addressing a natural question: Can large language models automatically design effective visual-token reduction algorithms? Although LLMs possess broad algorithmic knowledge and strong reasoning capabilities, translating such general knowledge into effective solutions for a specialized task remains nontrivial. We argue that the key lies in designing an appropriate search-state representation that connects the internal knowledge of LLMs with the structural requirements and constraints of visual-token pruning. Based on this insight, we propose AutoPrune, a training-free framework for LLM-driven visual-token pruning policy design. At its core, AutoPrune introduces a Token Pruning Domain-Specific Language (TPDSL) comprising 131 reusable atoms for budget control, token scoring, selection constraints, and token reassembly. A key property of TPDSL is that it represents each search state as a residual modification of a strong base policy. This residual formulation narrows the search space and directs the LLM’s attention toward the policy components that are most consequential for performance. Experiments on 14 multimodal benchmarks and three MLLM backbones demonstrate the effectiveness, efficiency, and transferability of AutoPrune. Even when removing 94.4% of visual tokens, AutoPrune preserves more than 99% of full-token performance while reducing FLOPs by 9.9x and prefill latency by 6.4x.
[CV-17] Representation-driven Endoscopic Visual Embedding Alignment for Latent Generation ECCV2026
链接: https://arxiv.org/abs/2608.07176
作者: Francisco Caetano,Tim J.M. Jaspers,Haiko Middeljans,Martijn R. Jong,Rixta A.H. van Eijck van Heslinga,Floor Slooter,Albert J. de Groof,Jacques J. Bergman,Peter H.N. De With,Fons van der Sommen
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted at the DCA-MI Workshop (ECCV 2026)
Abstract:Developing foundation generative models for endoscopy is limited by the gap between natural and clinical images and the computational cost of training large Diffusion Transformers. Although representation alignment has improved efficiency in general computer vision, its role within the highly specialized endoscopic image space remains unclear. We introduce REVEAL (Representation-driven Endoscopic Visual Embedding Alignment), the largest generative foundation model for endoscopy to date, trained on GastroNet-5M (GN-5M), a multicenter dataset of 5 million endoscopic frames. Instead of depending on out-of-domain priors, REVEAL employs encoders pretrained directly on the endoscopic distribution to align diffusion latents with domain-specific visual features, preserving fine textures and intricate anatomical structures. Beyond image generation, REVEAL also serves as a powerful feature extractor; in multiple benchmarks, it delivers performance that is competitive with, and in several cases exceeds, endoscopic foundation models such as EndoViT and Endo-FM, specifically tuned for classification tasks, while demonstrating strong representation robustness under realistic imaging corruptions. REVEAL produces high-fidelity images and maintains robust structural coherence in latent-space edits such as inpainting and outpainting. This high-capacity backbone lowers the computational threshold for building specialized clinical tools, offering an open, versatile foundation for conditional synthesis, segmentation, and out-of-distribution detection in future intelligent gastroenterology systems.
[CV-18] InstanceSplat: Instance-Aware Feed-Forward 3D Gaussian Splatting for Scene Understanding
链接: https://arxiv.org/abs/2608.07144
作者: Minchao Jiang,Xiaoxuan Ma,Shunyu Jia,Haoru Wang,Zhang Liang,Wentao Zhu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project page: this https URL
Abstract:Feed-forward 3D Gaussian Splatting (3DGS) enables efficient and generalizable 3D reconstruction, but current feed-forward 3DGS methods for scene understanding remain largely category-oriented. In contrast, instance-aware 3DGS methods typically rely on per-scene optimization and often decouple reconstruction from instance and semantic learning, limiting reciprocal interactions among them. We present InstanceSplat, a unified feed-forward 3DGS framework for generalizable 3D reconstruction and instance-aware scene understanding from pose-free multi-view images. In a single forward pass, InstanceSplat constructs an instance-aware Gaussian representation that jointly encodes appearance, geometry, instance identity, and language-aligned semantics. Shared 3D Gaussians ground instance identities across views, producing renderable and cross-view-consistent instance features. To allow reconstruction and scene understanding to benefit from each other, we further design an instance-centric learning strategy that connects reconstruction, instance learning, and semantic learning through shared instance structure. Specifically, instance cues guide reconstruction, language-aligned semantics strengthen the discrimination of confusing same-category instances, and instance regions aggregate semantic evidence into coherent object-level predictions. Experiments on novel-view synthesis, instance segmentation, and open-vocabulary semantic understanding under varying input-view settings and on an unseen dataset demonstrate state-of-the-art performance, practical efficiency, and strong generalization.
[CV-19] Human-AI Perceptual Alignment by Playing Hues and Cues
链接: https://arxiv.org/abs/2608.07141
作者: Nuria Alabau-Bosque,Jorge Vila-Tomás,Paula Daudén-Oliver,Pablo Hernández-Cámara,Valero Laparra,Jesús Malo
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 19 pages, 14 figures
Abstract:Evaluating the perceptual alignment between Contrastive Vision-Language Models (CVLMs) and humans is typically constrained by traditional benchmarks that overlook fine-grained semantic and cultural nuances. In this work, we propose a novel evaluation framework that leverages the gamified, discrete color space of the board game Hues and Cues. By mapping the board’s 480 color cells to the CIE xy chromaticity diagram, we calculate empirical perceptual distances across a carefully curated 100-word vocabulary spanning seven semantic categories. To properly contextualize model performance, we establish an empirical lower bound of expected error-the Human Consistency baseline-calculated via Leave-One-Out (LOO) cross-validation on a dense dataset of color associations collected from 325 human observers through a custom digital interface. We evaluate 162 models across multiple architectural families and pre-training datasets to assess their semantic color grounding. Our results demonstrate that while CVLMs successfully replicate human cognitive biases, such as idealized memory colors for concrete physical referents (e.g., food and plants), they systematically diverge from the human baseline in abstract, subjective, and pop-culture domains. We identify two distinct failure modes in severely misaligned concepts: semantic misclassification and a systematic uncertainty collapse into a default blue coordinate. Furthermore, we reveal that highly curated pre-training datasets are significantly more effective than massive, uncurated corpora in mitigating these severe misalignments. Ultimately, this work highlights that despite their broad categorization capabilities, current CVLMs still fail to capture the nuanced, localized consensus of human color memory, emphasizing the value of gamified tasks in exposing underlying model biases. The data and code are publicly available to test other metrics.
[CV-20] Multiple Hypothesis Flow Estimation for Video Frame Interpolation under Matching Ambiguity
链接: https://arxiv.org/abs/2608.07120
作者: Zibo Su,Jing Kong,Ruixing Wang,Zhanhe Zhang,Kun Wei
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Many flow-based video frame interpolation (VFI) methods synthesize an intermediate frame by estimating optical flow fields, warping the two input frames, and blending the warped observations. These latent flow fields are typically learned through image-level reconstruction supervision without direct flow annotations. In ambiguous regions containing repetitive or stochastic textures, rotating symmetric structures, or fast motion with blur, the matching evidence for a single query may contain multiple comparable and spatially separated peaks. Although the ground-truth intermediate frame provides indirect supervision, it may not uniquely identify the latent correspondence in ambiguous this http URL several locations provide multiple plausible matches, a single-flow estimator can retain only one displacement and discard the remaining candidates. If the selected match is incorrect or inconsistent with those of neighboring pixels, warping samples content from mismatched locations, producing ghosting, structural distortion, or this http URL address this limitation, we propose a multiple hypothesis flow estimation framework that preserves top-K candidate correspondences and selects one per location through a reliability-guided router. Each hypothesis is initialized from a coarse matching anchor and refined separately through anchor-centered local attention. Frame synthesis is thus conditioned on one selected flow-appearance hypothesis rather than a soft combination of candidate this http URL on the proposed MA-HD benchmark and public VFI benchmarks show that our method achieves the best LPIPS and DISTS among the compared methods.
[CV-21] Beyond Fluency: A Clinical Benchmark and Anomaly-Enhanced Baseline for Spine MRI Report Generation CVPR2026
链接: https://arxiv.org/abs/2608.07117
作者: Bruno Palau,Franziska Vogt,Daria Laslo,Haobo Li,Ender Konukoglu,Maria Monzon,Catherine R. Jutzeler
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Maria Monzon and Catherine R. Jutzeler contributed equally as shared last authors. Accepted at the CV4Clinic Workshop, CVPR 2026
Abstract:Radiology reporting is time-consuming and subject to inter-rater variability, making automated report generation an attractive clinical application for Vision-Language Models (VLMs). We benchmark state-of-the-art VLMs on lumbar spine MRI with a focus on diagnostic accuracy and demonstrate that standard lexical and semantic metrics poorly reflect clinical correctness: fluent, well-structured reports can score highly while containing clinically meaningful diagnostic errors. To address this failure mode, we propose an architecture-agnostic framework that augments VLM inputs with spatially localized, disc-level anomaly heatmaps generated by a semi-supervised U-Net++ model. These heatmaps both improve anatomical sensitivity through explicit visual grounding and provide an independent interpretability output for clinical oversight, moving us closer to diagnostically reliable, visually grounded VLMs for lumbar spine MRI interpretation.
[CV-22] Geometry-Aware Camera Localization for Bronchoscopy ACM-MM2026
链接: https://arxiv.org/abs/2608.07116
作者: Lumin Chen,Qingyao Tian,Jinpeng Li,Haoyu Jiang,Huai Liao,Xinyan Huang,Hongbin Liu,Dong Yi
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted by ACM MM2026
Abstract:Camera localization in bronchoscopy remains a challenging problem due to stringent accuracy requirements, real-time constraints, and limited training data. Compared to natural scenes, the confined anatomical structures demand millimeter-level precision, while intraoperative guidance necessitates low-latency inference. However, existing methods often fail to effectively exploit preoperative geometric priors, limiting their robustness and accuracy. To address these limitations, we propose a unified geometry-aware bronchoscope localization framework (GABL) that effectively fuses preoperative structural priors with paired intraoperative video to estimate 6-DoF camera poses. Specifically, to address visual ambiguity in complex airways, we propose a graph-guided coarse-to-fine localization scheme that effectively leverages structural priors for precise pose estimation. Furthermore, to mitigate pose jitter and bridge the visual-structural gap, we integrate a Transformer-based tracking model with a novel RGB-depth matching objective, jointly enforcing spatio-temporal and geometric consistency. Extensive experiments demonstrate that our method yields remarkable reductions of 8.37% and 31.76% in translation and rotation errors over the prior state-of-the-art, alongside 4 times inference speedup (33.6 FPS) for robust real-time bronchoscope localization. Project website: this https URL.
[CV-23] International Transfer of Stochastic Cortical Self-Reconstruction
链接: https://arxiv.org/abs/2608.07092
作者: Fabian Bongratz,Zhizheng Zhuo,Chao Zhang,Yaou Liu,Dennis M. Hedderich,Christian Wachinger
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Neurons and Cognition (q-bio.NC)
备注:
Abstract:Stochastic cortical self-reconstruction (SCSR) enables personalized mapping of gray matter atrophy, a hallmark of neurodegenerative disorders such as Alzheimer’s disease (AD), onto high-resolution cortical surfaces. Unlike conventional normative modeling approaches, which typically operate at a coarse regional level and remain inherently constrained by the covariates included during training, SCSR estimates an individualized healthy reference directly from the observed cortical thickness at the vertex level. This allows the detection of subtle, subject-specific deviations from healthy cortical shape. In this work, we investigate the generalization and transferability of SCSR, originally trained on UK Biobank (UKB) data, to an independent Chinese population dataset. Specifically, we evaluate the ability of SCSR-derived Z-scores to discriminate between healthy scans, individuals with mild cognitive impairment (MCI), and patients with AD, while also assessing model robustness across the lifespan. We compare four training strategies: direct application of the UKB-trained model, fine-tuning on Chinese data, training from scratch, and joint training on UKB and Chinese cohorts. As reconstruction backbones, we consider both a multilayer perceptron (MLP) and a Spherical UNet (SUNet). Our results demonstrate that SCSR provides robust detection of cortical atrophy in the Chinese population across all evaluated models. The highest discriminative performance was achieved by the fine-tuned SUNet model (average pairwise AUC = 0.848), followed closely by the UKB-trained SUNet. Moreover, reconstruction errors remained low across the lifespan, even when the training population exhibited a substantially narrower age distribution, indicating strong cross-population transferability.
[CV-24] RoRA: Role-Oriented Regional Allocation for Visual Token Pruning in MLLM s
链接: https://arxiv.org/abs/2608.07088
作者: Qiyanhui Lu,Han Wu,Rongjian Xu,Tingzhang Luo,Cheng Fan,Xinghao Chen,Minjing Dong,Jufeng Yang,Jianyuan Guo
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 9 pages, 4 figures, 4 tables. Code is available at this https URL
Abstract:Multimodal large language models (MLLMs) encode images as long visual token sequences, making prefilling and KV-cache storage expensive. Existing training-free pruning methods select tokens by importance, diversity, or spatial coverage, but treat retained tokens as interchangeable and do not explicitly track which object-related regions are already covered. We present RoRA, a training-free framework that casts visual token pruning as role-oriented regional evidence allocation. Given a fixed budget, RoRA partitions tokens into a protected semantic core, complementary context, and fine-grained detail. It first calibrates text-conditioned attention with a positional prior and a prompt-calibrated object prior, then builds Attention-Anchored Regions (AARs) from high-confidence anchors as lightweight proxies for covered object support. Context is explored mainly outside AARs, while a small AAR-guided budget restores local detail; pairwise similarity is used only for context-stage redundancy filtering. Under matched budgets, RoRA consistently outperforms strong training-free baselines across LLaVA and Qwen-VL families, retaining most of the unpruned accuracy even at aggressive pruning ratios, e.g., 96.5% of full performance at 88.9% pruning on LLaVA-1.5, and improving over D2Pruner by about 5% on Qwen3-VL at 75-90% pruning. At a 66.7% pruning ratio, RoRA requires only 0.7 ms for token selection and reduces end-to-end inference time by 24.6%, corresponding to a 1.33x speedup over unpruned inference on an NVIDIA H800.
[CV-25] Explanation Stability of Test-Time Adaptation in Computational Pathology: A Large-Scale Benchmark
链接: https://arxiv.org/abs/2608.07062
作者: R. G. Bahumanya,Harshith V. M.,Shreyank N. Gowda,Anala M. R
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Test-time adaptation (TTA) has become a practical way to adapt deployed models to unlabeled target data, a setting that is especially relevant in computational pathology where staining, scanner, and cohort shifts are routine. While most TTA methods are evaluated by their effect on accuracy, clinical use also depends on whether the model’s explanations remain reliable after adaptation. In this paper, we take a closer look at this largely unmeasured effect. We study explanation stability under TTA across two histopathology benchmarks, Camelyon17 and NCT CRC-HE, using five architectures ranging from convolutional networks to vision transformers and a pathology foundation model, seventeen TTA methods, and four attribution families. Across 2,958 adaptation runs, we observe a clear and systematic pattern: TTA methods differ sharply in how much they move model explanations, with frozen-backbone methods leaving attributions almost unchanged and continual methods such as CoTTA and RoTTA causing the largest drift. This effect is not uniform. Convolutional networks are substantially more sensitive than transformer and foundation-model backbones, and explanation drift increases with adaptation strength while remaining largely insensitive to batch size. Surprisingly, explanation stability is only weakly coupled to adaptation quality. Some methods preserve explanations almost perfectly while degrading calibration or accuracy, producing silent failures that would be missed by accuracy-only or explanation-only evaluation. These findings show that explanation stability is a distinct reliability axis for TTA in computational pathology. We release the metric, protocol, and full benchmark to support future work on adaptation methods that are not only accurate, but also stable and clinically auditable. Code: this https URL
[CV-26] KnifeHunter: Structured Local Representation Learning for Fine-Grained Knife Image Retrieval in Law Enforcement
链接: https://arxiv.org/abs/2608.07057
作者: Syed Sameed Husain,Eng-Jon Ong,Stephen Simpson,Trevor Hamshere,Matt Turner,Miroslaw Bober
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Knife-enabled violence presents a major public safety challenge, and law enforcement agencies require scalable tools for catalogue-level knife identification, intelligence analysis, and source attribution. Manual visual comparison is specialist, time-consuming, and difficult to scale under operational imaging conditions. We introduce KnifeHunter, an end-to-end forensic knife image retrieval system developed with UK law enforcement. The work contributes the KnifeHunter dataset, comprising 25,843 images across 543 knife classes from police evidence, retail catalogues, and border-force seizures, with structured metadata, Medium/Hard evaluation protocols, and large-scale distractor evaluation. We further propose CoRe-Net, a compact single-descriptor retrieval architecture that combines global context with spatially localised discriminative evidence. CoRe-Net introduces Structured Complementary Representation Learning (SCRL) to organise local evidence into complementary prototype-based representations, and Bi-Directional Reciprocal Fusion (BDRF) to integrate global and local evidence through residual projection and gated local-to-global injection. Using an EVA02-Base backbone and cosine-similarity retrieval, CoRe-Net achieves 88.0% mAP and 86.7% mP@10 on the Medium protocol, and 85.1% mAP and 83.8% mP@10 under distractor conditions. KnifeHunter was deployed by UK police forces during Operation Sceptre deployments from 2023 to 2025, achieving 99.2% mP@1 on field queries. These results demonstrate a practical and effective multimedia retrieval framework for fine-grained forensic knife matching in operational law-enforcement settings.
[CV-27] YOLO-PEFT: Parameter-Efficient Fine-Tuning on YOLO Family
链接: https://arxiv.org/abs/2608.07051
作者: Xu Lin,WenJie Nie,Jinlong Peng,Weifu Fu,YueXiao Ma,Xiawu Zheng,Yong Liu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Generic parameter-efficient fine-tuning (PEFT) methods transferred from language models can fail silently on real-time detectors, whose heterogeneous operators and detection-specific components impose placement constraints absent from regular Transformer stacks. We propose YOLO-PEFT, a structure-aware framework that formulates adapter placement as an auditable constraint-planning problem. Given a detector graph, a PEFT request, and a resource budget, YOLO-PEFT assigns operator and semantic roles, evaluates explicit operator-validity, detector-semantic, graph-interface, and deployment predicates, records a reason code for each excluded module, and either emits a budgeted target-module plan or returns Refuse before training. Under the official VOC07+12 trainval-to-VOC07 test protocol, planner-selected RS-LoRA reaches 0.7138 and 0.7307 mAP50-95 on YOLO11s and YOLO12s, respectively, compared with 0.6428 and 0.6662 for Full-SFT. On RT-DETR-L, all seven evaluated LoRA-family configurations cross the predefined catastrophic threshold, supporting a calibrated Refuse-to-Full-SFT decision within the evaluated coverage. A controlled YOLO11 audit further shows that LoRA reduces peak training memory by 43.9 percent, although training takes 1.72 times longer. Within the evaluated detector families, placement policies, and calibration coverage, YOLO-PEFT replaces manual target-module trial and error with explicit, inspectable planning while preserving verified train-save-merge-export paths; refusal on unseen detector architectures remains an open validation problem. Project Page: this http URL
[CV-28] C2Dex: Contact-Consistent Reconstruction and Retargeting for Dexterous Manipulation from Monocular Video
链接: https://arxiv.org/abs/2608.07045
作者: Jie Ren,Zhehao Jiang,Yinhong Yang,Haorui Jia,Han Jiang,Ben Li,Yao Yao,Cheng Lin,Qiu Shen,Zhenshan Bing,Xiao-Xiao Long,Xun Cao
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages, 5 figures. Submitted to IEEE Robotics and Automation Letters (RA-L). Project page: this https URL
Abstract:High-quality demonstrations for dexterous robot manipulation are costly and difficult to collect, whereas monocular human videos provide a scalable source of diverse manipulation behaviors. However, transferring such demonstrations to dexterous robots remains challenging: monocular hand-object interaction (HOI) reconstruction often produces temporally unstable contacts and physically implausible interactions, while conventional retargeting methods struggle to preserve task-relevant contacts and local interaction geometry across different hand embodiments. We present C2Dex, a video-to-dexterous-manipulation framework built around a shared interaction representation: stable object-side contacts recovered by aggregating noisy frame-wise observations in the canonical object space. These stable contacts serve a dual role: as trajectory-level constraints that guide reconstruction toward temporally coherent and physically plausible human HOI trajectories, and as explicit transfer targets for the dexterous hand, where Laplacian interaction optimization preserves the local hand-object geometry across embodiments and residual reinforcement learning refines the trajectory in simulation. Experiments on DexYCB and TACO show that C2Dex achieves end-to-end trajectory success rates of 57.78% and 26.67%, respectively, substantially outperforming the strongest baselines (17.78% and 10.00%) under identical evaluation criteria. Real-robot replay experiments further demonstrate physical feasibility across diverse contact-rich manipulation tasks. Project page: this https URL
[CV-29] CAS2UML: A Handwritten Sketch-to-PlantUML Dataset for Class and Activity Diagrams
链接: https://arxiv.org/abs/2608.07036
作者: Simon Scholz,Mersedeh Sadeghi
类目: oftware Engineering (cs.SE); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Automated UML generation from sketches and images is gaining renewed attention with the rise of large language models and multimodal AI. However, reproducible evaluation remains difficult due to the lack of public datasets with executable groundtruth models. We present CAS2UML, a public dataset of 557 handdrawn UML diagrams, including 271 class diagrams and 286 activity diagrams, each paired with manually validated PlantUML code. We also provide a PlantUML-based validation tool and reusable scripts for checking the syntactic correctness and renderability of generated UML artifacts, enabling reproducible benchmarking of sketch-to-UML approaches. The dataset, validation tool, processing scripts, documentation, and demonstration video are publicly available at: Dataset: this https URL /cas2uml_hand-drawn_to_plantuml_dataset; Tool and Scripts: this https URL Video: this https URL.
[CV-30] Understand Before Detect: Vision–Language Learning for Omni-Domain Infrared Small Target Detection
链接: https://arxiv.org/abs/2608.07015
作者: Haoyang Yuan,Boyang Li,Yingqian Wang,Yimian Dai,Nuo Chen,Xinfei Huang,Shuqi Yi,Zaiping Lin,Weidong Sheng,Wei An
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Omni-domain infrared small target (IRST) detection is crucial for infrared surveillance, yet remains challenging due to heterogeneous imaging domains and inconsistent target characteristics. Previous deep learning-based methods have been developed for visual-only paradigms and achieved promising performance on domain-specific tasks. However, existing methods follow the task-specific supervised learning paradigm. This paradigm simplifies the full-scene infrared observations to sparse target supervision, discarding the semantics that remain invariant across heterogeneous domains. Consequently, detection performance suffers substantially under domain shifts. To handle this issue, we introduce \textbf``understand before detect’', a paradigm that formulates omni-domain IRST detection as an understanding-driven process, where holistic infrared target understanding precedes precise detection. Building on this paradigm, we propose \textbfJinSight, which first develops holistic IRST understanding through language supervision and then transfers the learned cross-domain representations to precise small-target detection. By grounding infrared representations in language semantics, JinSight enables a single model to generalize across heterogeneous infrared domains. We then introduce Latent Semantic Interaction (LSI), which exchanges language-aligned global semantics with fine-grained spatial features in a compact low-rank space. To address the lack of multimodal omni-domain IRST benchmarks, we build \textbfOmniIRST-VL, the first large-scale, highly diverse vision–language dataset for omni-domain IRST detection. It comprises over 39k annotations across six complementary instruction tasks covering both scene-level understanding and target-centric reasoning.
[CV-31] Stable Curves Unstable Items: Item-Level Scaling Heterogeneity in Video LLM s
链接: https://arxiv.org/abs/2608.07014
作者: Wenzhang Sun,Chunfeng Wang,Xiangchen Yin,Yujia Chen,Hao Li,Kun Zhan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Aggregate scaling curves suggest that Video LLMs improve smoothly or saturate as visual budgets grow. We show that this view can conceal large, opposing changes at the item level. We represent each frozen model–item pair by its response trajectory under controlled visual budgets and derive matched-grid measures of configuration complementarity, harmful transitions, and text overwrite. Across five open Video LLMs from three architecture families, four multiple-choice benchmark splits, open-ended QA and summarization, and fixed-history dialogue generation, no single budget serves all items. On the four-model matched MCQA grid, item-level oracle headroom spans 8.8 – 18.9 accuracy points and 12.5 – 25.5% of items are correct at a lower budget but wrong at a higher one. Task-appropriate continuous metrics show the same complementarity beyond multiple choice: Token-F1 oracle gaps are 2.7 – 3.7 score points on MLVU generation and 3.8 – 4.8 points on AVSD current-turn generation, even when mean quality improves with budget. The effect persists across frame count, spatial resolution, sampling policy, temporal–spatial allocation, and independently executed raw-video and cached pipelines, with per-item rates and membership tracking protocol choices. A controlled sampling intervention recovers 29.0% of terminal regressions, and a structured frame audit identifies several recurring evidence pathways. We release per-item trajectories, protocol provenance, derived annotations, and reproducible analysis code as an auditing artifact. A confidence cascade matches fixed- 128f accuracy while reducing average shared frame cost by 31.7% , illustrating one operational use of the response matrix.
[CV-32] Scenix: Sparse-View 3D Scene Reconstruction via Executable Scene Programs
链接: https://arxiv.org/abs/2608.07012
作者: Kai Li,Lutao Jiang,Zhenyang Li,Jiayu Dong,Jierui Zhang,Yingda Yin,Runze Zhang,Kai Yan,Xiaoyang Huang,Keyang Luo,Xin Wang,Xiangyu Zhao,Weikai Chen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 4 figures 5 table 9 pages
Abstract:Synthesizing a structured and editable 3D indoor scene from a few uncalibrated RGB views requires more than generating high-quality individual assets: a system must infer the room structure, associate objects across incomplete observations, and recover a globally consistent spatial configuration. Previous methods mainly focus on 3D scene generation with text input or require continuous visual inputs with additional priors, \ e.g., human-annotated masks or accurate 3D layouts, which makes these methods labor demanding and hard to apply in general cases. We present \textscScenix, a sparse-view 3D scene reconstruction framework via executable scene programs, a structured representation that can be directly instantiated into editable 3D scenes. Given sparse views, \textscScenix predicts executable scene programs through perception-grounded asset instantiation and closed-loop spatial refinement. % We present \method, a framework that predicts an executable scene representation from sparse views and realizes it through perception-grounded asset instantiation and closed-loop spatial refinement. To support this task, we construct \dataset, a dataset of approximately 110,000 synthetic and real indoor scenes with multiview imagery, room structures, object-centric descriptions, and metric spatial annotations. We further introduce observation-consistent supervision that aligns each target scene with the visual evidence available in its input views. Experiments on held-out \textscXScene scenes, real indoor images, and out-of-distribution SpatialGen cases evaluate structured scene prediction, object grounding, and spatial refinement.
[CV-33] HRDiT: Training-Free High-Resolution Image Generation with Off-the-Shelf Diffusion Transformer Models ECCV2026
链接: https://arxiv.org/abs/2608.07003
作者: Yu Xue,Haoxuan Qu,Zhuoling Li,Hongbin Xu,Jianxiong Yin,Simon See,Hossein Rahmani,Jun Liu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by ECCV 2026
Abstract:Training-free text-to-high-resolution image generation has recently attracted growing research attention. However, existing studies on this task primarily focus on adapting off-the-shelf U-Net-based diffusion models to high resolutions, with limited progress on adapting off-the-shelf Diffusion Transformer (DiT) models despite their strong text-to-image generation capabilities at limited resolutions. In this work, we find two key challenges particularly hindering the application of off-the-shelf DiT models for high-resolution image synthesis in a training-free manner, namely, spatial disorder and long generation time. To address these challenges, we propose a novel method tailored to adapt off-the-shelf DiT models for high-resolution image synthesis. Extensive experiments show the efficacy of our method. Our code is available at: this https URL.
[CV-34] Local Epistemic Uncertainty Guided Active Sampling for Plug-and-play Diffusive Image Restoration
链接: https://arxiv.org/abs/2608.06981
作者: Jiaqi Zhang,Zheng Pang,Rongrong Gao,Qiyuan Zhang,Yang Yang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 12 Pages, 7 Figures, 5 Tables. Accepted to ACM Multimedia 2026
Abstract:Diffusion models have demonstrated remarkable effectiveness in image restoration tasks. However, when guiding image reconstruction, existing Diffusion Model-based Image Restoration (DMIR) methods typically rely on fixed data constraints and uniform step sizes, thereby overlooking the dynamic nature of the generative process. Such rigid designs render the models vulnerable to spatially non-uniform degradations, thus resulting in structural distortions and loss of fine details. Meanwhile, uniform step sizes introduce computational redundancy, whereas naïve step reduction strategies tend to accumulate approximation errors. To address these limitations, we propose a Local Epistemic Uncertainty Guided Active Sampling framework (LEADer). In the spatial domain, LEADer leverages pixel-wise uncertainty to dynamically modulate the prior strength within the null space, which effectively balances detail preservation and artifact suppression. In the temporal domain, it quantifies sampling stability via the uncertainty trace to enable adaptive trajectory pruning, thereby accelerating convergence. Theoretical proofs demonstrate that our framework achieves strict data consistency, while the trajectory pruning strategy admits a deterministic error bound, thereby guaranteeing stable convergence under skip sampling. Notably, our plug-and-play method can be seamlessly integrated into various DMIR baselines. Extensive experiments show that LEADer improves the performance of multiple state-of-the-art DMIR methods, while significantly reducing sampling time with negligible memory overhead. Code is available at this https URL.
[CV-35] When One Modality Is Not Enough: Multimodal Sex and Life-Stage Classification of Red Deer from Aerial RGB-Thermal Video ECCV2026
链接: https://arxiv.org/abs/2608.06973
作者: Hugo Markoff,Christoph Praschl,Ivan Ludoški,Sara Beery,Michael Ørsted,David C. Schedl
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at the ECCV 2026 Workshop on Computer Vision for Ecology (CV4Ecology), archival proceedings track. 17 pages, 7 figures, 5 tables
Abstract:Aerial drone surveys increasingly support wildlife population estimation, yet a useful census is more than a count: population dynamics are defined by species composition, sex ratios and age structure, that is, by which species are present and how a herd splits into adult males, adult females and juveniles. We use red deer ( \textitCervus elaphus ) as a test case, because managers act on these dynamics and because the visible cue defining adult males, the antlers, is seasonally variable. Surveys are flown nadir, high enough not to disturb the animals, so each deer occupies only a small, low-resolution patch. The two recording modalities fail in opposite conditions: in color a deer under canopy blends into the ground, while in thermal it becomes a bright blob that loses fine detail. Rather than trust either modality alone, we fuse them at every stage using self-supervised DINOv3 features. Our pipeline tracks animals in both modalities, treats an animal as confirmed only when the two cameras agree, keeps only the clear, non-occluded frames, and assigns species and sex by a vote across them; life stage is read separately from geo-referenced body size, since at survey resolution a juvenile often only differs from an adult female in size. Across four flights spanning the antler season the fused pipeline correctly classifies 25 of the 26 detected individuals (7 of 8 adult males, all 16 adult females and 2 juveniles), against 20 of 26 for either sensor alone. Multimodal species classification reaches 96.0%, while for sex classification fusing the two sensors matters most: the combined RGB+thermal model is the most robust across environments and seasons. Automating the demographic classification turns a drone flight from a count into a repeatable reading of herd structure, so the sex ratios and age structure that managers already act on can be gathered as often as a survey can be flown.
[CV-36] Generative Embedding Benchmark: How Much Information Survives in a Dense Embedding?
链接: https://arxiv.org/abs/2608.06972
作者: Yun Li,Biao Yang,Peixi Wu,Yunhao Zhou,Mingzhou Jiang,Wei Yuan,Fan Yang,Wenwu Ou
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Embeddings have emerged as a standard representational interface linking foundation models with downstream systems. Most embedding benchmarks assess representations through discriminative tasks or geometric criteria centered on separability in embedding space. However, strong performance on such evaluations does not establish whether content compressed into an embedding remains accessible to a downstream generator. To address this gap, we introduce the Generative Embedding Benchmark (GEB), in which a decoder answers questions using only a frozen embedding and question text, without access to the original image or intermediate visual features. Answer quality under this readout measures generative information: the answer-relevant content recoverable from an embedding. GEB includes a curated visual-question-answering dataset with a 1,800-item development split and a held-out 900-item test split covering natural images, scene text, and visual documents. Using a common decoder and training recipe, we evaluate seven public embedding models in visual-only and vision-language joint modes. On the test set, visual-only scores range from 28.25 to 33.21; with image-question joint encoding, all five VLM-based embedding models score higher, and the best reaches 65.56. Matched embeddings also outperform text-only inputs, zero embeddings, and shuffled embeddings. Natural-image information is much easier to recover than scene text or visual-document information, while a Qwen3-VL-2B reference with access to the original image reaches 84.30. Together, these results show that generative readout exposes information bottlenecks that separability-based evaluation does not capture.
[CV-37] Summarize First Download Later: Onboard VLMs for Bandwidth-Efficient Earth Observation
链接: https://arxiv.org/abs/2608.06959
作者: Junghwan Park,Sangcheol Sim,Woojin Cho,Darongsae Kwon
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: IGARSS2026
Abstract:Modern Earth observation (EO) satellites carry increasingly advanced sensors that produce vast volumes of high-resolution, multispectral data, yet downlink capacity remains a critical bottleneck – often causing significant latency or the loss of valuable observations within limited contact windows. We propose a “Summarize First, Download Later” paradigm that exploits recent advances in onboard edge computing and Vision-Language Models (VLMs). Rather than indiscriminately downlinking raw imagery, the system follows a three-phase interaction protocol: the satellite first transmits concise natural language summaries generated by a quantized onboard VLM; ground operators then issue targeted Visual Question Answering (VQA) queries to verify scene relevance (e.g., wildfires or maritime anomalies); and full-resolution images are downloaded only when critical information is confirmed. This transforms the downlink from passive bulk transfer into an active, semantics-aware dialogue. We implement and evaluate the system on a resource-constrained NVIDIA Jetson platform, and experiments on diverse remote sensing scenes show that the proposed strategy substantially reduces bandwidth consumption while accelerating time-to-insight for time-sensitive missions.
[CV-38] Casting the Net! Revisiting MasterFace Impersonation Attacks CCS2026
链接: https://arxiv.org/abs/2608.06952
作者: Seunghun Paik,Sunpill Kim,Chanwoo Hwang,Jae Hong Seo
类目: Cryptography and Security (cs.CR); Computer Vision and Pattern Recognition (cs.CV)
备注: To appear at ACM CCS 2026; Seunghun Paik and Sunpill Kim contributed equally
Abstract:Impersonation is a fundamental security threat in face recognition systems (FRSs). While the security of FRSs has been challenged by various attack vectors, under realistic adversarial capabilities, e.g., a limited number of decision-only authentication trials and no internal system knowledge, most attack techniques become infeasible. As a result, impersonation by zero-effort impostors, characterized by false match rate (FMR), is commonly regarded as a standalone baseline. A few years ago, impersonation attacks based on MasterFaces emerged as a notable security threat that could break the barrier of the FMR-based baseline under such realistic constraints. However, they were believed not to yield impersonation above the standard FMR in modern FRSs, as discussed by multiple follow-up studies. In this paper, we demonstrate that even legitimate access to public commercial APIs allows an adversary to amplify impersonation rates through MasterFaces, resulting in a non-trivial impersonation attack beyond FMR on downstream applications built on top of these APIs. We observe that several real-world FRS deployments are implemented using commercial APIs, and that the backend service provider is publicly disclosed or trivially inferable. As a result, the adversary can purchase these pay-as-you-go API services without requiring any additional privilege over the target FRS. From this observation, we formalize the MasterFaces attack as a maximum coverage problem over the biometric representation space, which we call a NET, and show that the adversary can construct an API-tailored NET by leveraging the geometric structure of the representation space. We demonstrate that our attack amplifies the impersonation rates of several open-source and commercial API-based FRSs by up to 9.5 \times within at most 30 authentication trials, compared to those expected from the standard FMR.
[CV-39] Dual-Space Modality Consistency Learning for Universal Cross-Modal Re-Identification
链接: https://arxiv.org/abs/2608.06943
作者: Yujian Zhao,Yukang Zhao,Hankun Liu,Haoxuan Xu,Bo Li,Hanzi Wan,Guanglin Niu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Cross-modal Re-Identification (ReID) aims to retrieve the same identity across heterogeneous imaging modalities and has been widely studied in visible-infrared person ReID and cross-modal ship ReID. Existing methods have achieved promising performance by learning modality consistency in the spatial embedding space, yet often overlook frequency-domain modality discrepancy, particularly in high-frequency representations that are both highly discriminative and modality-sensitive. In addition, most approaches are tailored to specific modality settings, limiting their applicability across diverse cross-modal scenarios. To address these challenges, we propose a Dual-Space Modality Consistency Learning (DSMCL) framework for universal cross-modal ReID. Specifically, DSMCL jointly models spatial feature distribution consistency and frequency-domain discriminative consistency. A Spatial Modality Consistency Learning (SMCL) branch performs Gaussian-based feature alignment, while a Frequency-aware Discriminative Consistency Learning (FDCL) strategy regularizes high-frequency representations through identity-aware cross-modal contrastive learning. By jointly capturing modality-specific characteristics and modality-shared identity cues, DSMCL learns robust representations and establishes a unified framework capable of accommodating diverse heterogeneous modality settings. Moreover, DSMCL is a plug-and-play framework that can be readily integrated into existing cross-modal ReID architectures. Extensive experiments on SYSU-MM01, RegDB, LLCM, HOSS-ReID, and CMShipReID across seventeen evaluation protocols show that DSMCL consistently improves multiple representative baselines.
[CV-40] ELMZip: Onboard Satellite Image Compression via Extreme Learning Machines for Efficient Downlink
链接: https://arxiv.org/abs/2608.06942
作者: Woojin Cho,Junghwan Park,Sangcheol Sim,Steve Andreas Immanuel,Junhyuk Heo,Darongsae Kwon
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:The acquisition of multispectral imagery via small satellites (e.g., CubeSats) presents significant data downlink challenges due to high data volumes and restricted communication windows. While onboard image compression is critical to address this bottleneck, traditional methods often struggle to adapt to the nonlinear statistics of multi-band, multi-resolution data. To overcome these limitations, we propose ELMZip, a novel framework based on Extreme Learning Machines (ELM) and domain decomposition strategies for efficient, resolution-free onboard neural representation. ELMZip formulates the fitting process as a convex least-squares problem using random-feature single-layer networks, thereby eliminating the need for computationally expensive backpropagation. By adopting an asymmetric transmission protocol that sends only the compact output weights, the proposed method significantly reduces the downlink payload. Unlike previous neural representation approaches that rely on iterative optimization and require transmitting full network parameters, ELMZip achieves significant compression efficiency while maintaining high reconstruction fidelity. This capability enables immediate image reconstruction for analysis, allowing resource-constrained platforms to maximize data return and advancing real-time AI-powered Earth observation.
[CV-41] Degradation-Aware Prompt Learning with Cross-Modal Compensation for Adverse Weather Removal
链接: https://arxiv.org/abs/2608.06939
作者: Wanshu Fan,Yunzhe Zhang,Yue Shen,Liyan Wang,Jing Qin,Kin-Man Lam,Cong Wang,Jinshan Pan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted for publication in IEEE Transactions on Image Processing. The code is available at: this https URL
Abstract:Adverse weather causes diverse and complex image degradations, severely compromising the reliability of computer vision systems. Existing all-in-one restoration models attempt to address multiple degradation types within a unified framework, but often lack explicit spatial and semantic modeling of degradation characteristics, limiting their adaptability to diverse weather conditions. To address this limitation, we propose a Degradation-Aware Cross-Modal Prompt Compensation Network (DCMPC-Net) that leverages cross-modal degradation cues from a pretrained vision-language model to condition restoration features within a unified backbone. Specifically, our DCMPC-Net mainly consists of the Cross-Modal Prompt Generator (CMPG), Prompt-Guided Attention Alignment Module (PGAAM), and Dual Feature Compensation Module (DFCM). The CMPG integrates textual embeddings with visual features to produce degradation-aware prompts that encode degradation-related semantic and contextual cues. These prompts are injected into the decoder via a PGAAM, which adaptively aligns semantic information with degraded regions to facilitate context-aware restoration. To further enhance structural fidelity, DFCM is introduced that disentangles degradation artifacts from scene structures, thereby improving the reconstruction of fine textures and detailed content. By integrating cross-modal semantic guidance with spatial alignment and structural enhancement, DCMPC-Net achieves robust and perceptually consistent restoration across diverse weather conditions. Extensive experiments show that DCMPC-Net outperforms state-of-the-art methods in both task-specific and unified settings, achieving superior accuracy and visual fidelity.
[CV-42] Debias in Text Believe Your Eyes: Text-Anchored Cross-Modal Transfer for Visual Counter-Commonsense Reasoning
链接: https://arxiv.org/abs/2608.06938
作者: Chen Ling,Hanqian Li,Dongnan Liu,Keyu Qian,Jungang Li,Xinglong liu,Shiyi Wang,Xin Dong,Pengcheng Zhu,Wei Zhou,Linjian Mo,Nai Ding
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:The visual reasoning ability of multimodal large language models (MLLMs) is crucial for downstream applications, particularly counter-commonsense reasoning, which requires models to reason beyond common assumptions. Recent studies mainly improve visual counter-commonsense reasoning by enhancing visual inputs, following the assumption that failures originate from insufficient visual grounding. However, our empirical analysis reveals that the bottleneck is not visual perception. MLLMs already capture the relevant visual evidence, and the correct answer exists in their decoding space. Instead, the shared language decoder resolves prior–evidence conflicts by favoring dominant language priors, especially for low-frequency factual scenarios. Motivated by this, we first propose a text-anchored data construction pipeline, whose core component, Fact-Frequency Distillation (FFD), estimates the prior strength of commonsense facts and distills verified counter-commonsense scenarios into a high-quality text corpus. Building upon this corpus, we introduce TACT, a text-anchored post-training framework that debiases the shared language decoder without requiring any visual training data. TACT routes evidence-following and prior-driven reasoning trajectories into different optimization stages, enabling the decoder to resolve prior–evidence conflicts. Across counter-commonsense visual benchmarks, TACT substantially improves visual reasoning while preserving general capabilities, demonstrating effective text-to-vision cross-modal transfer.
[CV-43] Walkable to Whom? Capturing Subjective Variability in Walkability Perception Using Multimodal Deep Learning
链接: https://arxiv.org/abs/2608.06934
作者: Moloud Damandeh,Meead Saberi
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Visual perception of walkability varies substantially across individuals, reflecting differences in personal characteristics, experiences, and preferences. Existing studies, however, often reduce these diverse judgements to aggregated scores, implicitly assuming uniform perception, and commonly rely on vehicle-mounted street-view imagery that does not reflect the pedestrian’s visual experience. This paper introduces a dataset of 29,870 walkability ratings from 1,196 respondents, linking sidewalk-view imagery across urban, suburban, and regional Australian environments with individual rater attributes, and proposes the first user-conditioned multimodal deep learning framework for walkability perception, fusing visual features with respondent-level representations. A viewpoint-comparison study shows that sidewalk-view images receive significantly higher walkability ratings than matched street-view images, indicating that imagery source is a substantive design decision in perception surveys. The user-conditioned model improves rank agreement with observed ratings by 65% over an image-only baseline (quadratic weighted kappa 0.47 vs. 0.29), demonstrating that who is evaluating an environment carries predictive indication beyond image content alone. These findings support moving from aggregated, observer-independent walkability scores toward models that represent diverse users, enabling more inclusive assessment of pedestrian environments.
[CV-44] AVCap: Reinforcing Audio-Video Joint Caption with Detail-Aware Reward
链接: https://arxiv.org/abs/2608.06930
作者: Mingyang Wu,Kaituo Feng,Bohao Li,Kaixiong Gong,Zihao Yin,Xiangyu Yue
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Detailed audio-video joint captioning is essential for multimodal video understanding and generation. However, prior works are constrained by three main limitations: (1) the scarcity of high-quality public datasets with fine-grained audio-visual joint captions; (2) reinforcement-learning methods that rely on coarse reward signals; and (3) the lack of a benchmark and metric for evaluating detailed audiovisual captions at the atomic level. To address these challenges, we propose: (1) AVCap-100K, a high-quality dataset of 100K temporally aligned, detail-rich audio-video captions; (2) AVCap, a model optimized via Detail-Aware GRPO (Da-GRPO) that achieves state-of-the-art performance among open-source models and matches or surpasses proprietary models on several evaluations; and (3) AVCap-Bench and AVCap-Score, a specialized benchmark and metric for evaluating atomic-level details in audiovisual captions. Our code, models, and datasets are available at this https URL.
[CV-45] MaskFlow: Precise Consistent and Seamless Regional Image Editing
链接: https://arxiv.org/abs/2608.06929
作者: Rui Xu,Yang Yong,Shunzi Yang,Ruihao Gong,Chengtao Lv
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 18 pages, 6 figures
Abstract:Regional image editing has attracted considerable attention for its spatial controllability. Although instruction-based and mask-reference-based editing methods can achieve strong semantic alignment, reliable regional control remains challenging, where an edit must be accurately localized and naturally integrated with the preserved context. We propose \textbfMaskFlow, a training framework for precise localization, consistent background preservation, and seamless boundary transitions. MaskFlow incorporates the mask into the probability path and flow-matching objective, coordinating generation within the editable region with source preservation outside it. The proposed Soft-Poisson de-seaming module further refines the predicted vector field during both training and sampling to improve the smooth integration of the edited foreground with the preserved background. We also design a data synthesis pipeline to construct MEData, a mask-based image editing dataset for training regional image editing models and facilitating further research. Experiments on natural scenes and infographic images demonstrate consistent improvements over competing methods in both quantitative and qualitative evaluations.
[CV-46] Vernata: Self-Supervised Learning of LiDAR Point Representations IROS2026
链接: https://arxiv.org/abs/2608.06919
作者: Oliver Lemke,Alexander Liniger,Abel Gawel,Marco Hutter
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注: IROS 2026. Implementation: this https URL
Abstract:LiDAR serves as a primary sensing modality for robots operating in outdoor environments. However, the performance of deep learning models in this domain is severely limited by the scarcity of labeled data, a direct result of the high cost of 3D annotation. Self-supervised learning addresses this scarcity by learning general-purpose features from unlabeled data. In this work, we present a multi-modal, multi-teacher distillation framework for self-supervised learning on outdoor LiDAR point clouds. Building upon the Sonata architecture, we introduce Vernata, consisting of three extensions: sparse view augmentation to improve robustness against varying point densities, a memory bank mechanism to stabilize resource-constrained training, and cross-modal distillation utilizing dense, high-resolution 2D image features to enable fine-grained semantic guidance. We evaluate our method on the GrandTour, TartanGround, and Waymo datasets, as well as data collected from our own robotic platforms. Our experiments demonstrate a significant performance improvement over Sonata baselines, yielding mIoU scores of 54.7 on TartanGround (+5.9 points, +12.1%) and 57.1 on Waymo (+7.3 points, +14.7%). Finally, we show that the self-supervised approach maintains strong performance even in reduced-modality settings (lacking color or normals), achieving competitive mIoU scores of 49.4 and 50.2 on the respective datasets.
[CV-47] RibAssist 3D: Biplanar Rib-Fracture Detection Addressing and Selective 3D Localization from CT-Derived Projections
链接: https://arxiv.org/abs/2608.06914
作者: Kabila Haile Soboka
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages, 6 figures. Code available at: this https URL
Abstract:Rib fractures are common, clinically significant, and time-consuming to localize on computed tomography (CT). We ask whether fractures detected in two orthogonal projections (anteroposterior, AP, and lateral) can be paired across views and triangulated into reliable 3D fracture points at a controlled rate of false 3D outputs. We answer this with a staged diagnostic study. Biplanar geometry is exact: detector-predicted centers reconstruct to median 4.0 mm 3D error when correspondence is correct. On the sealed cohort, dual-view availability reaches 61.1% and the candidate graph contains a correct pair for 58.4% of fractures. The binding limitation is not geometry or localization but confidence-limited cross-view correspondence. Lateral-detector retraining lifts dual-view availability (0.52 to 0.76 in development) and moves the frontier from 0% to 2.44% recall at 10 mm. When the policy commits a correct pair, the emitted point is geometrically accurate (sealed median 1.49 mm, rib-exact 93%). A pre-specified pass on the untouched 55-case cohort promotes 15 of 601 fractures to correct 3D localizations at 0.436 false 3D points per case, yielding 2.50% end-to-end commitment yield. The contribution is validated biplanar reconstruction geometry with high conditional localization fidelity, a staged identification of cross-view correspondence confidence as the effective bottleneck, and a selective assistive workflow that preserves uncertain findings rather than a standalone automatic reconstructor.
[CV-48] MuST-VAD: Mutual Structured Learning for Video Anomaly Detection
链接: https://arxiv.org/abs/2608.06913
作者: Satoshi Hashimoto,Hitoshi Nishimura,Mori Kurokawa
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:In this paper, we propose MuST-VAD, a mutual structured learning framework for weakly supervised video anomaly detection (VAD) in which an anomaly detector and a large vision-language model (LVLM) exchange their acquired knowledge. Detectors in weakly supervised VAD learn anomaly scores from features extracted by a fixed, task-agnostic backbone. These fixed features bound the achievable detection accuracy. Recent methods therefore transfer LVLM semantics into the detector as richer features. However, this transfer is one-way: what the detector learns about the target videos never returns to the LVLM. MuST-VAD extends the one-way transfer into a bidirectional learning loop. In this loop, the latest detector predictions supervise the LVLM adaptation, and the adapted LVLM returns updated representations that retrain the detector; the two models alternate these updates over small video groups. Both models train on detector-selected key clips, while confidence weighting and annotation-anchored question answering keep the exchanged supervision reliable. On UCF-Crime, our mutual learning improves the one-pass transfer baseline from 88.15% to 88.63% AUROC and from 37.25% to 42.46% average precision (AP), outperforming the state-of-the-art method in AP by 4.13 points.
[CV-49] Prune Once: Retraining-Free Task-Agnostic Pruning for Vision-Language Models ECCV2026
链接: https://arxiv.org/abs/2608.06901
作者: Minseok Kang,Hyunwoo Kim,Chanyoung Kim,Minwoo Kim,Jaekoo Lee,Dahuin Jung
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Accepted to ECCV 2026
Abstract:Vision-language models (VLMs) have achieved remarkable generalization across diverse multimodal tasks through large-scale pre-training, yet their rapidly increasing computational and memory requirements pose significant challenges for deployment in constrained environments. Existing pruning strategies often depend on task-specific criteria or LLM-oriented importance measures, making them unsuitable for task-agnostic pruning, where no task-specific samples are available at pruning time and the pruned model remains broadly applicable. We introduce a retraining-free VLM pruning framework called PORTA that derives a task- and modality-agnostic importance formulation based on activation variation, estimated from generic calibration data, which reliably captures feature-level representation utility across modalities. PORTA further incorporates an adaptive sparsity allocation mechanism that assigns layer-wise pruning ratios based on output feature variability, avoiding the limitations of uniform sparsity and reducing performance degradation at high compression levels. Extensive experiments across VLM architectures, such as CLIP, BLIP, and Qwen2-VL, demonstrate that PORTA achieves competitive downstream performance under high sparsity without requiring any retraining, supporting efficient VLM compression. Code is available at this https URL.
[CV-50] From Points to Edges: Edge-Conditioned Spectral Operators for Physics-Sensitive PDE Learning
链接: https://arxiv.org/abs/2608.06894
作者: Zhentao Tan,Ruijie Quan,Yi Yang
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Neural operators have become a central tool for solving partial differential equations (PDEs), with spectral operators offering efficient global mixing across spatial locations. However, many PDEs contain physics-sensitive local structures that are critical to the underlying physical behavior. For example, in Darcy flow, local material interfaces are often reflected by sharp changes in the permeability field and can strongly influence the solution. Existing spectral operators primarily adapt modal mixing based on center-point representations, making them insufficiently responsive to such localized structural variations. We propose the Edge-Conditioned Spectral Operator (ESO), a novel spectral operator framework that modulates global spectral mixing using local edge-wise variations. By incorporating the Pairwise-Variation Modal Mixer (PVMM) to inject local edge information into spectral mode selection, ESO preserves the global approximation capability of spectral neural operators while enabling the learned kernel to adapt to physics-sensitive local structures. Furthermore, we introduce a task-adaptive Physics-Aware Reweighting (PAR) that emphasizes physically important regions, identified by taskspecific physical quantities. Across nine PDE benchmarks, ESO consistently achieves state-of-the-art performance. Visual and region-wise analyses further demonstrate that ESO reduces solution errors near coefficient jumps, high-gradient flow structures, and other physically sensitive regions. The code is available at this https URL.
[CV-51] HazeSpikeMamba: Coupling Spiking-Inspired and State-Space Features for Self-Supervised Real-World Dehazing
链接: https://arxiv.org/abs/2608.06886
作者: Haoran Liu,Huibin Li,Mingzhe Liu,Peng Li,Guibin Zan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 21 pages, 6 figures
Abstract:Dehazing networks are commonly trained on synthetic hazy-clear pairs, but their performance often drops on real photographs. Synthetic haze generated using the atmospheric scattering model does not fully capture the variability of real haze, and paired real hazy-clear images are scarce. In this work, we propose HazeSpikeMamba, a compact dehazing framework that combines a spiking-inspired local path and an attentive state-space global path in a multi-scale U-Net. The local path uses TPCNNSpike, a new spike-emission scheme inspired by the neighborhood coupling of Pulse-Coupled Neural Network (PCNN). Unlike grouped directional scanning, TPCNNSpike updates all neurons in parallel using the previous firing states of their Gaussian-weighted neighborhoods. The global path adapts the Attentive State-Space Module of MambaIRv2, retaining semantic prompting and sequence reordering while removing the window self-attention branch. Its state-space processing models long-range dependencies with complexity linear in sequence length. For target-domain adaptation, a frozen degradation network, pretrained on paired NH-HAZE data, re-synthesizes haze from the dehazed prediction. The reconstruction error updates only the final restoration layers of HazeSpikeMamba without haze-free labels during adaptation. A shared checkpoint is adapted once on each complete unlabeled target set, making the evaluation dataset-level and transductive rather than zero-shot or per-image optimization. The forward network contains 2.02M active parameters and requires 13.27G nominal MACs (measured with thop at 256x256 input). This adaptation consistently improves BRISQUE and NIMA on RTTS, URHI, and HSTS. On RTTS, BRISQUE decreases from 30.13 to 27.72 and NIMA increases from 4.13 to 4.87. Under this transductive protocol, the adapted model also achieves the best BRISQUE and NIMA on URHI and HSTS among the compared methods.
[CV-52] ControlRef: Efficient Layout-Guided Multi-Instance Generation via Anchored 4D-RoPE
链接: https://arxiv.org/abs/2608.06878
作者: Yunkai Yang,Yudong Zhang,Xinying Chen,Haoyuan Liang,Yizhuo Niu,Jinshuai Cheng,Kunquan Zhang,Liziyue Fang,Weitao Wan,Runmin Dong
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Layout-guided multi-instance generation is essential for controllable image synthesis in Multi-Modal Diffusion Transformers (MM-DiTs). However, integrating this capability into unified architectures remains challenging. Prior frameworks rely on redundant full-resolution canvas padding and Shifted-RoPE to manage multiple reference images. This mechanism drastically inflates computational overhead for sparse layouts and disrupts critical low-frequency RoPE features, creating a severe spatial-frequency compromise that blurs absolute spatial correspondence. To overcome these limitations, we propose ControlRef, a highly efficient and precise multi-instance synthesis framework. ControlRef utilizes a Unified Instance-Layout Control (UILC) attention mask to strictly decouple inter-instance semantic interactions and enforce precise regional binding. To further promote region-level spatial alignment, we introduce Anchored 4D-RoPE, a novel positional encoding mechanism that directly anchors tokens to their absolute geometric centers. By pre-aligning reference images to their corresponding bounding box resolutions, physically anchoring both layout and reference tokens to their absolute geometric centers, and stacking the references along the z-axis, Anchored 4D-RoPE natively preserves spatial priors and mitigates the spatial-frequency compromise without lossy shifting. Extensive experiments demonstrate that ControlRef achieves state-of-the-art visual fidelity and localization accuracy, while concurrently slashing inference latency by over 80% in sparse layouts and reducing memory overhead by 50% in dense scenarios.
[CV-53] FedVAR: Prototype-Aligned Federated Framework for Video Anomaly Recognition
链接: https://arxiv.org/abs/2608.06876
作者: Ghani Haider,Majid Kundroo,Boyun Eom,Dong Hwan Park,Chen Chen,Taehong Kim
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG)
备注:
Abstract:In the era of Industrial Internet of Things (IIoT) and Cyber-Physical Systems (CPS), Federated Learning (FL) offers a promising decentralized intelligence paradigm for Video Anomaly Recognition (VAR). This task is vital for maintaining high-fidelity Digital Twins and ensuring safety in mission-critical environments. However, the inherent data heterogeneity across distributed edge clients leads to a fundamental challenge known as semantic misalignment, where clients learn divergent feature representations of “normal” and “abnormal” events. The problem becomes particularly pronounced in VAR, where the presence of diverse and fine-grained anomaly categories leads each client to develop distinct semantic interpretations of abnormality. Existing federated methods primarily focus on binary anomaly detection and fail to address this misalignment, preventing effective fine-grained recognition. In this paper, we introduce FedVAR, a weakly-supervised FL framework explicitly designed for VAR. Leveraging the rich representations of Vision-Language Models (VLMs), FedVAR employs a prototype-based alignment mechanism that creates a shared semantic anchor for all clients to re-center and align their visual and textual feature spaces. This process enforces a consistent representation of “normality” across the decentralized network, directly mitigating semantic misalignment and enabling robust prompt-learning of anomaly direction vectors with minimal communication overhead. We conduct extensive experiments on challenging benchmarks under various non-IID data partitioning schemes, unseen domains, and novel anomaly classes. The results demonstrate that FedVAR consistently outperforms state-of-the-art federated baselines, establishing a robust framework for distributed intelligence in video-based CPS.
[CV-54] RegionDet: A Benchmark for Region Detection Beyond Object Instances
链接: https://arxiv.org/abs/2608.06850
作者: Liang Wan,Yuhan Wang,Yupeng Zhang,Zhen Xu,Han Wang,Fangjie Fu,Sirui Zhu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Object detection is a fundamental task in computer vision and has achieved remarkable progress on standard benchmarks by localizing discrete and well-bounded object instances. However, many visual targets in real-world scenarios are not individual objects, but regions defined by visual states, scene context, object relations, and human activities, such as construction areas, damaged road regions, queues, group conversations, and vendor regions. Existing detection benchmarks are mainly built around object instances, providing limited support for systematically evaluating such region targets. To address this gap, we introduce Region Detection, a task that extends conventional object detection beyond object instances, and construct RegionDet, a benchmark for region target localization. RegionDet contains eight region categories, including Construction, Crossing, Damage, Queuing, Talking, Vendor, Waiting, and Walking, with COCO-style bounding-box annotations and evaluation protocols. We systematically evaluate representative closed-set and zero-shot/open-vocabulary detectors on RegionDet. Results show that closed-set detectors can partially learn region-level patterns under supervision, while zero-shot/open-vocabulary detectors struggle severely, revealing the strong object-centric bias of current vision-language detectors. Further analyses highlight key challenges in Region Detection, including weak boundary cues, strong context dependency, and insufficient relation-level region understanding. The RegionDet will be released.
[CV-55] Are Visual Place Recognition Models Recognizing Places or Conditions? Distractor-Augmented Evaluation and Condition Suppression
链接: https://arxiv.org/abs/2608.06847
作者: Beomsu Kim,Minwoo Jung,Giseop Kim
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 8 pages, 10 figures, 5 tables. Submitted to IEEE Robotics and Automation Letters (RA-L)
Abstract:Long-term Visual Place Recognition (VPR) is typically evaluated by matching queries from one condition against a database from another. Crowdsourced map databases, however, may mix conditions and include images that resemble the query in condition but depict different places. In the presence of these distractors, a method may retrieve by condition similarity rather than place identity. We argue that this susceptibility arises because the discriminability of VPR methods allows them to encode information such as illumination, weather, and seasonal appearance in their descriptors. We therefore introduce Distractor-Augmented Recall (DAR) to isolate and quantify the effect of distractors, and propose condition suppression to remove condition information from VPR descriptors. Across eleven methods and six datasets, method rankings under DAR@1 differ from those under Recall@1 (R@1), while applying INLP and LEACE as condition suppression methods generally improves DAR@1 without reducing R@1. Thus, distractor robustness is distinct from standard retrieval performance and can be improved by suppressing condition information.
[CV-56] ECAD: Expanding Class-Agnostic Detection Beyond Thing-Centric Objectness
链接: https://arxiv.org/abs/2608.06841
作者: Liang Wan,Zixin Ren,Yupeng Zhang,Yuhan Wang,Fangzhuo Gao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Object detection is a fundamental task in visual perception, providing structured region representations for recognition, grounding, reasoning, and interaction. However, existing detection paradigms largely inherit a thing-centric notion of objectness, where detectors are mainly trained to localize discrete and countable object instances. Consequently, many semantically meaningful visual elements, such as sky, road, grassland, water, and sports courts, are often absorbed into the background despite their importance for scene understanding and spatial reasoning. In this paper, we formulate Expanded Class-Agnostic Detection (ECAD), a new setting that aims to discover category-agnostic visual candidates beyond conventional thing-centric objects. To support this setting, we construct BTCO-Bench, a Beyond Thing-Centric Objectness benchmark with category-agnostic box annotations covering both real-world and cross-domain scenarios. We further propose ECADet, a lightweight DETR-based detector built upon a frozen DINOv3 encoder, and introduce Geometry-Aware Expert Regression (GAER) and Prototype-Guided Query Modulation (PGQM) to improve localization and objectness estimation for diverse visual elements, respectively. Extensive experiments show that ECADet consistently outperforms representative class-agnostic and proposal-based detectors on BTCO-Bench, demonstrating the effectiveness of expanded objectness discovery. Code and benchmark will be released.
[CV-57] GOPI: Generation-Oriented 3D Pose Inference for Furniture Insertion from Single-View RGB-D Indoor Scenes
链接: https://arxiv.org/abs/2608.06836
作者: Ruifeng Zhai,Renjie Liu,Guangrun Wang,Liang Lin
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:We study the problem of inserting new furniture into indoor scene images. Under masked single-view 2D image-plane conditioning, however, the physical scale of the inserted furniture relative to the scene cannot be uniquely determined, making physically grounded furniture placement underdetermined from image evidence alone. We therefore reformulate the task as a combination of 3D pose inference and geometry-guided image generation, where estimating a geometrically plausible 3D placement is essential for reliable synthesis. To this end, we propose a two-stage framework. For 3D placement, we introduce GOPI, a generation-oriented 3D pose inference framework that addresses the underdetermined nature of single-view furniture insertion through data-driven iterative inference, producing geometrically plausible object placements. For image generation, we develop a geometry-guided conditioning strategy that projects the inferred 3D pose into the image plane as a pixel-aligned constraint, enforcing consistency between the synthesized image and the underlying 3D geometry. Experimental results validate the proposed framework from both 3D pose estimation and image synthesis perspectives. For 3D placement, GOPI produces poses with stronger geometric feasibility and better consistency with reference layouts than direct regression and vanilla baselines. For image synthesis, our method preserves alignment with the projected 3D geometry across different furniture scales, showing stable projection-generation alignment across the tested furniture scales. Subjects: Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2608.06836 [cs.CV] (or arXiv:2608.06836v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2608.06836 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-58] Bend the Basics: Degradation-Aware Deformable Tokenization for All-in-One Image Restoration ICML2026
链接: https://arxiv.org/abs/2608.06832
作者: Zihao He,Yunfeng Wu,Xinchao Wang,Songhua Liu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 16 pages, 12 figures. Accepted to ICML 2026
Abstract:All-in-one image restoration seeks a single model that can recover images degraded by diverse and spatially non-uniform corruptions. However, many unified Transformers rely on fixed patch partitioning: task/degradation condition is injected only into the backbone blocks after tokenization, leaving the embedding and reconstruction stages insensitive to local degradation variations. In contrast to previous approaches, we present Flexible Image Transformer (FIT) that explicitly models degradation awareness across the entire pipeline, from patch sampling to pixel reconstruction. Specifically, FIT employs a lightweight Degradation Encoder to predict a global degradation vector \mathbfg and a spatial degradation map \mathbfM from local degradation severity, which jointly condition the patch embedding and unembedding through adaptive deformation. Moreover, to improve robustness across degradation types, we introduce a task-token dropout strategy that regularizes task conditioning during training. On five standard benchmarks (BSD68, Rain100L, SOTS, GoPro, and LOLv1), FIT achieves state-of-the-art performance with 30.72 dB average PSNR on the five-degradation setting and 32.83 dB on the three-degradation setting, outperforming recent unified restoration methods by +0.5 \sim 1.1 dB. Moreover, the learned offsets provide a direct handle for visualizing degradation-aware spatial adaptation.
[CV-59] R2S-EGO: Dual-Proxy Refinement for Sparse-Capture Real-to-Sim
链接: https://arxiv.org/abs/2608.06827
作者: Shuai Fang,Xin Deng,Yuchen Kang,Zhenjiang Li,Jie Chen
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR)
备注: 11 pages, 6 figures, 4 tables, and 1 algorithm
Abstract:Real-to-sim (R2S) depends on scene representations that render observations along robot ego trajectories, yet dense multi-view capture limits per-environment real-image capture-count efficiency, and sparse human capture can leave behavior-scoped robot views under-supported. Camera-controlled synthesis can fill missing views, but its use in R2S requires behavior-admissible queries and capture-anchored structural conditioning. We present R2S-EGO, which couples a simulator-derived robot proxy that represents the behavior-scoped executable query domain with a capture-anchored geometry proxy that supplies scene-specific structural conditions. Within this domain, fixed- budget selection targets current support deficits for which geometry support is available. The generated observations are assimilated as pseudo-observations to refine the visual asset, while real captures remain anchors. The fused geometry proxy also supplies the scene collision surface, which is refreshed between rounds. Together, these updates refine the existing simulation scene while its robot dynamics and control stack stay fixed. Across 48 frozen Unitree G1 ego views in three Replica scenes, six-view R2S-EGO reaches 19.062 dB PSNR, compared with 14.226 dB for the strongest reported R2S baseline. Across five paired policy-training seeds, R2S-EGO achieves 82.5% +/- 6.8% real-G1 sitting success, compared with 10.0% +/- 10.5% for GaussGym. Comments: 11 pages, 6 figures, 4 tables, and 1 algorithm Subjects: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR) Cite as: arXiv:2608.06827 [cs.RO] (or arXiv:2608.06827v1 [cs.RO] for this version) https://doi.org/10.48550/arXiv.2608.06827 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Shuai Fang [view email] [v1] Fri, 7 Aug 2026 05:37:26 UTC (3,239 KB) Full-text links: Access Paper: View a PDF of the paper titled R2S-EGO: Dual-Proxy Refinement for Sparse-Capture Real-to-Sim, by Shuai Fang and 4 other authorsView PDFHTML (experimental)TeX Source view license Current browse context: cs.RO prev | next new | recent | 2026-08 Change to browse by: cs cs.CV cs.GR 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
[CV-60] AdvTiles: Physical Adversarial Camouflage Clothing against Person Detectors via Learnable Tiles
链接: https://arxiv.org/abs/2608.06801
作者: Jinlei Wang,Jiahuan Long,Mingkai Sun,Yafei Guo,Yuanhao Huang,Ming Wang,Junqi Wu,Jiacheng Hou,Hongbo Chen,Xingxing Wei,Tingsong Jiang,Wen Yao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Physical adversarial attacks against person detectors have evolved from localized patches to full-body textures. However, achieving both visual naturalness and strong attack effectiveness remains challenging. Existing natural-looking methods typically optimize camouflage textures as a whole, limiting the flexibility to refine local adversarial patterns and their spatial arrangement. To address this issue, we propose AdvTiles, a physical adversarial camouflage framework built from learnable tiles, enabling strong attack performance while preserving a natural camouflage appearance. Specifically, we use a Straight-through (ST) Gumbel-Softmax estimator for differentiable tile selection, enabling joint optimization of tile patterns and spatial layouts. This design provides fine-grained control over adversarial texture generation. To improve robustness in diverse physical conditions, we further optimize the camouflage through differentiable 3D Gaussian Splatting rendering with variations in viewpoints, scales, illuminations and backgrounds. Extensive experiments across multiple detectors demonstrate that AdvTiles achieves an average ASR of 86.2%, outperforming existing state-of-the-art attack methods. We further fabricate the optimized camouflage into wearable adversarial clothing, validating its effectiveness in real-world scenarios across diverse distances, angles and backgrounds.
[CV-61] Is Forward Prediction Enough? Physical State Grounding for JEPA World Models
链接: https://arxiv.org/abs/2608.06799
作者: Haodong Yan,Jiaguan Zhu,Mingyuan Jia,Ruiqing Yin,Junjie He,Zhide Zhong,Junfeng Li,Jinxuan Lu,Hengtao Li,Tianran Zhang,Jiayi Chen,Wenxuan Song,Wen Chen,Yuxiang Gao,Haoang Li
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Learning structured and control-relevant latent representations remains a key challenge for world models. Recent JEPA-based world models learn action-conditioned predictive latent dynamics from observation sequences. However, their forward-prediction objectives do not explicitly enforce reliable identifiability of robot-centric physical state from individual latents or state changes from latent pairs, which can limit downstream planning and policy performance. We propose PSG-JEPA, a physically grounded JEPA world model that shapes its latent space with two complementary grounding objectives beyond forward prediction: grounding individual latents in robot proprioceptive state, and grounding latent pairs in multi-horizon joint-angle changes. Both objectives are applied only during training, leaving the inference architecture and computational cost unchanged. To comprehensively evaluate PSG-JEPA, we conduct experiments at three levels: (1) latent identifiability via probing, (2) goal-conditioned planning on frozen latents, and (3) policy learning in simulation and on a real robot. Experiments demonstrate that our PSG-JEPA consistently outperforms state-of-the-art latent world-model baselines at all three levels.
[CV-62] PAST: Prompt-Adaptive Sampling Termination for Efficient Diffusion Model
链接: https://arxiv.org/abs/2608.06794
作者: Renye Yan,Jikang Cheng,You Wu,Wei Peng,Zongwei Wang,Ling Liang,Yimao Cai
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:While diffusion models have made significant progress in text-to-image tasks, they still exhibit limitations when directly optimizing downstream objectives. Although Reinforcement Learning (RL) enables targeted optimization, existing methods are generally constrained by low-efficiency fine-tuning and sparse rewards. To address these challenges, we propose PAST, which provides differentiated rewards while adaptively regulating training episode length by jointly perceiving denoising progress and prompt difficulty. Specifically, we design an intrinsic reward paradigm to compensate for sparse extrinsic rewards and guide the model to explore paths that diverge more efficiently from noise patterns. We further provide theoretical justification for intrinsic rewards. Then, PAST dynamically monitors denoising completion and semantic alignment between image structures and prompt semantics. When both metrics satisfy generation requirements, the system adaptively terminates training. This enables appropriate allocation of episode lengths based on prompt difficulty and the current generation process. Finally, based on the predicted residual noise level, we establish a dual adaptive coordination mechanism. Specifically, it not only balances the extrinsic and intrinsic rewards but also balances the exploration and convergence. Experimental results demonstrate that PAST enhances computational efficiency of existing RL fine-tuning methods by up to 66.7%, while improving preference optimization quality by up to 29.5% through its dual adaptive regulation mechanism.
[CV-63] UniCycleFlow: Bidirectional Unpaired Image Translation with a Shared Rectified Flow
链接: https://arxiv.org/abs/2608.06784
作者: Xianhao Zhou,Jianghao Wu,Shaoting Zhang,Guotai Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Bidirectional unpaired image translation must preserve source-specific structure while learning coherent transformations in both directions without paired supervision. Existing methods typically employ two direction-specific generators or train separate one-way models. Even when linked by cycle consistency, such models constrain only the round-trip endpoint reconstruction, without requiring the two directions to obey a common local transformation rule. We propose UniCycleFlow, a rectified-flow framework that represents bidirectional translation as forward and reverse integration of a single time-conditioned velocity field. This formulation organizes both directions within the same continuous dynamics, rather than coupling otherwise separate endpoint mappings. A key challenge is that unpaired data provide no meaningful source–target coupling from which rectified-flow trajectories can be constructed. UniCycleFlow addresses this challenge by learning deterministic source-conditioned endpoints whose marginal distributions are adversarially matched to the opposite domains. The resulting paths are regularized by stop-gradient self-flow matching for intermediate velocity supervision, discrete cycle closure for forward–reverse consistency, and representation path-velocity regularization for controlling localized feature changes along the trajectory. Across ten translation directions, UniCycleFlow achieves the lowest FID on 7 of 10 tasks using a single Euler evaluation and obtains the best average FID of 55.1.
[CV-64] AnyTrack: Unifying Visual Object Tracking with Any Modalities ACM-MM2026
链接: https://arxiv.org/abs/2608.06773
作者: Hao Li,Yunzhi Zhuge,Wenning Hao,Pingping Zhang,Xiaoxiong Zhang,Dong Wang,Huchuan Lu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by ACM MM2026. More modifications may be performed
Abstract:Visual object tracking aims to continuously locate specific targets within sequential frames, evolving from single-modal methods to multi-modal ones. However, existing multi-modal trackers are typically designed for fixed modality combinations, requiring separate models for different inputs. This leads to a poor adaptability to missing or imperfect modalities, and limited generalization. To address these issues, we propose a novel unified framework called AnyTrack for object tracking with any modalities. Specifically, we design a Modality-aware Interaction Module (MIM) to facilitate dynamic interaction across diverse modalities. This module bridges modality discrepancies and aggregates temporal cues to maintain spatio-temporal consistency during cross-modal interaction. Furthermore, we introduce a Context Understanding Module (CUM) to establish spatial correspondence between visual features and target locations via global-local prompts. This module employs target-aware context modeling to enhance foreground-background discrimination for precise localization. Finally, to support the training and evaluation under diverse modalities, we extend existing multi-modal object tracking benchmarks by incorporating grayscale images, language descriptions, and audio clips. Extensive experiments with both complete and missing modality settings demonstrate that our AnyTrack achieves state-of-the-art performance, validating its effectiveness and flexibility. The source code is available at this https URL.
[CV-65] Surg-UniWorld: A Unified Surgical World Model with Multimodal Control Experts
链接: https://arxiv.org/abs/2608.06770
作者: Rulin Zhou,Wanhao Liu,Guoheng Ma,Liangjin Shao,Qiujie Song,Yidu Wang,Guankun Wang,Tong Chen,Long Bai,Luping Zhou,Hongliang Ren
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Controllable surgical world models can provide a generative foundation for surgical artificial intelligence and simulation by synthesizing realistic instrument–tissue interactions. However, existing methods lack a unified multimodal control paradigm, while direct fusion of heterogeneous visual conditions often causes anatomical distortion, instrument appearance drift, and temporally inconsistent interactions. In this work, we propose Surg-UniWorld, a unified surgical world model with multimodal control experts. Surg-UniWorld first constructs a Hierarchical Surgical Anchor from first-frame appearance and hierarchical semantic masks to preserve persistent scene identity, anatomical organization, and interaction boundaries. Anchor-Relative Modality Experts then interpret edge, depth, and optical-flow evidence relative to the shared anchor, capturing complementary boundary, geometric, and motion information. A Multimodal Control Expert further performs contribution-preserving stage-wise composition of the activated modality increments and generates control hints for the Wan2.2 video diffusion backbone. To support multimodal surgical world modeling, we further construct Cholec80-SurgWAM, a benchmark for controllable surgical video generation. Extensive experiments demonstrate that Surg-UniWorld consistently outperforms existing controllable video generation methods and surgical world-model baselines in generation quality, temporal consistency, and multimodal controllability.
[CV-66] GraphVerse: A Comprehensive Visual Graph Reasoning Benchmark for Multimodal Large Language Models
链接: https://arxiv.org/abs/2608.06769
作者: Yuanfu Sun,Yuanhang Ren,Kang Li,Chuanhao Ji,Jiaxi Li,Jiajin Liu,Ninghao Liu,Qiaoyu Tan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 33 pages, 16 figures
Abstract:Recent Multimodal Large Language Models (MLLMs) have achieved remarkable progress across diverse vision-language tasks, creating an urgent need for more challenging benchmarks. Yet existing evaluations still provide limited insight into whether these models can truly reason over structured visual information. Visual Graph Reasoning (VGR) offers a compelling testbed for this challenge, requiring models to integrate perception, structural understanding, and multi-step reasoning over graph-based visual inputs. However, prior VGR benchmarks often reduce the task to visual perception followed by text-based reasoning, restrict evaluation to single-image settings, rely on answer-only metrics, and underrepresent realistic graph-centric scenarios. To bridge the gap, we introduce GraphVerse, a unified benchmark that jointly evaluates perception, visual reasoning, and text-based graph reasoning in MLLMs under both single-image and paired-image settings. At its core is a suite of Graph-centric Image Editing (GIE) strategies that modify graph images while preserving their semantics, turning them into active tests of visual reasoning. We further propose VGR-Score, a process-sensitive metric that evaluates reasoning quality beyond final-answer accuracy. Extensive experiments reveal several key limitations of current MLLMs in VGR, while also validating the effectiveness of GIE strategies and the transferability of GraphVerse to broader multimodal reasoning capabilities. The code is available at this https URL.
[CV-67] Explore or Converge? Stage-Guided Per-Step Optimization for Diffusion Models
链接: https://arxiv.org/abs/2608.06768
作者: Renye Yan,Jikang Cheng,You Wu,Wei Peng,Zongwei Wang,Ling Liang,Yimao Cai
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Diffusion models have strong generative capabilities. However, their maximum likelihood training objective only focuses on reconstructing the data distribution, making it difficult to align with specific preferences. Reinforcement learning (RL) for preference alignment in diffusion models is promising but limited by reward sparsity. Since a single reward cannot support optimization, existing RL methods usually backpropagate the final reward to all previous steps. However, denoising is stage-wise, with distinct semantics and controllability. Repeating the final reward across all steps creates a temporal objective mismatch, encouraging reward shortcuts that lead to reward hacking. At the same time, due to reward backfilling, each time step receives the same reward, making it impossible to distinguish between actions, thereby weakening the optimization process. To resolve this issue, we propose Stage-Guided Per-Step Optimization (SGPO) for diffusion models, which jointly leverages signal-to-noise ratio and semantic changes to identify generation stages and adaptively assign stage-specific objectives. Early denoising is chaotic and far from the final reward, resulting in weak reward-behavior correlation. This stage should prioritize exiting the chaotic state. In the mid stage, the latent transitions to a stable structure, where the final reward better corresponds to generative behavior. Therefore, this stage optimizes the final reward while exploring diversity to avoid early convergence to a single mode. In the late stage, the latent’s core structure is largely fixed, and preference optimization mainly amplifies local details, risking overfitting. Therefore, stable convergence is preferred to avoid quality degradation. Results from 16 comparative experiments validate SGPO. Our method achieves 26.7% average gains in generative quality and 36.7% higher convergence speed. Subjects: Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2608.06768 [cs.CV] (or arXiv:2608.06768v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2608.06768 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-68] Beyond Starry Night: Shortcut-Aware Control-State Planning for Artist-Grounded Text to Image Generation
链接: https://arxiv.org/abs/2608.06751
作者: Kuan Xing,Ye Wang,Changyi Gan,Yuheng Li,Thao Nguyen,Yi Chang,Yilin Wang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 47 pages, 13 figures, including appendices. Kuan Xing and Ye Wang contributed equally
Abstract:Artist-grounded image generation requires more than appending an artist name to a prompt. Image models often respond to artist names through canonical shortcuts, such as recurring motifs, generic palettes, or overrepresented period signatures, rather than preserving the user’s intended scene. We introduce Atelier, a shortcut-aware control-state planning framework for artist-grounded image generation. Atelier translates underspecified artistic intent into an explicit control state that separates scene anchors, preserve/transform decisions, style-regime hypotheses, role-bound artist evidence, and shortcut-avoidance constraints. It grounds this state using artist-level knowledge and local patch references, compiles backend-aware generation plans, and iteratively refines candidates through global and local authenticity feedback. We further introduce ArtIntentBench, a benchmark covering Van Gogh and Qi Baishi across artwork re-rendering, period/style-controlled generation, historically unseen subjects, shortcut auditing, and human preference evaluation. Across open-weight and closed-source generators, Atelier improves artist-level style fidelity, preserves source structure more faithfully, and substantially reduces shortcut substitution compared with prompt-engineered, retrieval-augmented, and general-purpose agent baselines. These results suggest that artist-grounded generation is bottlenecked not only by image synthesis, but by the upstream inference of explicit, evidence-grounded artistic controls.
[CV-69] From Cheap Fakes to Pure Synthesis: Addressing the New Era of T2V Fake News Videos ACM-MM
链接: https://arxiv.org/abs/2608.06732
作者: Yifeng Luo,Yupeng Li,Liang Lan,Tian Wang
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM)
备注: Accepted at ACM Multimedia (ACM MM), 2026
Abstract:Recent text-to-video (T2V) generation models enable fake news videos to be synthesized from scratch, shifting the threat beyond cheap fakes assembled from existing footage. Such news videos can closely match fabricated narratives, creating a modality alignment trap for existing detectors. Existing datasets lack pure synthesis fake news videos. Although directly prompting T2V models with descriptions of fake news videos can yield perfectly aligned samples, it reduces the fake news video detection (FNVD) to unimodal shortcuts and causes semantic-visual degeneration. To counter this, we formulate T2V-FNVD as a novel ternary classification task with three labels (real, cheap fake, and pure synthesis fake) and construct the first pure synthesis fake news video dataset (PS-FNVD). PS-FNVD includes fabricated events with aligned deception (Type 1) and true events with false visual provenance (Type 2), preventing models from exploiting unimodal shortcuts. Furthermore, we propose the Reasoning-guided T2V-FNVD (R-T2V) framework. Trained through conditioned rationale generation and supervised fine-tuning, R-T2V integrates high-level semantic logic with low-level physical generative traces to predict the ternary veracity label. Extensive experiments across 10 prevailing baselines show that R-T2V achieves the state-of-the-art performance, outperforming the second-best baseline by 12.20 percentage points in accuracy and 8.46 percentage points in macro F_1 .
[CV-70] AtlasVLA: Persistent World-Ego State Modeling for Vision-Language-Action Models
链接: https://arxiv.org/abs/2608.06729
作者: Guiyu Zhao,Longteng Guo,Yanghong Mei,Zilin Zhu,Yu Zhang,Bin Cao,Mingming Yu,Xingjian He,Jie Jiang,Jing Liu
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:While Vision-Language-Action (VLA) models have advanced embodied AI, their fundamentally reactive paradigm severely limits performance in partially observable and long-horizon tasks. When restricted to a single wrist-mounted camera, they inevitably suffer from perception forgetting as objects exit the field of view, and temporal task-progress forgetting during multi-step execution. To overcome these bottlenecks, we propose AtlasVLA, a novel framework that transitions from direct reactive manipulation to proactive reasoning through a persistent world-ego state. AtlasVLA features a dual-memory architecture: a 4D Persistent World State Memory that lifts transient 2D observations into a globally updated, voxel-hashed spatial state to resolve visual blind spots, and an Ego-Working State Memory that tracks historical ego state and task progress. By conditioning a diffusion transformer (DiT) on this joint World-Ego state, AtlasVLA enables robust spatial reasoning. Extensive evaluations across LIBERO, RLBench, and real-world benchmarks demonstrate that AtlasVLA achieves state-of-the-art performance using solely a wrist camera. Remarkably, it decisively outperforms multi-view baselines, yielding absolute success rate improvements of 9.4% on LIBERO-Long and 17.5% in real-world long-horizon tasks.
[CV-71] WaveFreqAnchor: Wave-Structural Anchoring and Frequency Correction Diffusion for Training-Free Face Restoration
链接: https://arxiv.org/abs/2608.06717
作者: Zelin Du,Wenjie Li,Zhengxue Wang,Juncheng Li,Cailing Wang,Guangwei Gao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: training-free wavelet-structural diffusion sampling framework for face restoration that improves structural stability, identity consistency, and real-world perceptual quality without task-specific fine-tuning
Abstract:Diffusion-based face restoration that adjusts the sampling trajectory of pre-trained diffusion models has achieved remarkable progress. However, existing approaches provide insufficient constraints during reverse diffusion, causing identity-related structural drift and degraded fidelity under severe degradations. To address this, we propose WaveFreqAnchor, a training-free framework based on Wave-Structural Anchoring and Frequency Correction Diffusion. Specifically, Anchor-Space Wave-Structural Guidance (ASWG) constrains facial structures through anisotropic wave-response consistency, while Multi-scale Wavelet-Fourier Injection (MWFI) aligns the predicted low-frequency subband with the observation by replacing its phase, correcting inconsistencies accumulated during reverse diffusion. For real-world scenes, we further introduce Subband High-Frequency Enhancement (SHE), which performs bounded, spatially masked refinement on the predicted high-frequency subbands to recover fine facial details under unknown compound degradations. Together, these designs effectively preserve facial identity while restoring sharp and realistic facial details. Extensive experiments show that our method consistently outperforms existing methods, achieving high-quality and high-fidelity face restoration.
[CV-72] Suppress and Diversify: Refining Robust Pathways for Corruption Robustness ICML2026
链接: https://arxiv.org/abs/2608.06712
作者: Jiangang Yang,Wenhui Shi,Xiaoran Xu,Wenyue Chong,Luqing Luo,Jing Xing,Jian Liu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at ICML 2026
Abstract:Model robustness against natural image corruptions is essential for safety-critical applications. While existing methods primarily focus on implicit representation learning, we provide the first systematic exploration of computational pathways to explicitly characterize internal robustness. We identify a progressive decay of robust features across network layers and establish a functional dependency between the prevalence of these features and model performance. To exploit these insights, we propose Suppress and Diversify (S\D), a non-intrusive refinement approach that enhances robustness by dynamically selecting robust pathways and diversifying them through symmetry-preserving transformations. S\D is architecture-agnostic, parameter-free, and incurs zero test-time overhead. Extensive evaluations across eight benchmarks demonstrate that S\D consistently improves performance across multiple vision tasks, diverse backbones, and complex real-world scenarios, highlighting its broad efficacy and scalability.
[CV-73] Agent Patch: Coarse-to-Fine Weak-Task Repair for Merging Agent ic Multimodal Large Language Models
链接: https://arxiv.org/abs/2608.06699
作者: Zibo Shao,Baochen Xiong,Chengdong Xu,Linhui Xiao,Kaichen Li,Haoran Gong,Yan Li,Yaguang Song,Xiaoshan Yang
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Agentic multimodal large language models (MLLMs) extend multimodal perception and reasoning with planning, tool use, and interaction in dynamic environments. Yet current models are specialized for particular tools or environments, complicating consolidation into a single generalist. We formulate Agentic MLLM Merging and identify two challenges: asymmetric capability preservation, whereby capabilities with different interaction complexity are retained unevenly, producing weak tasks after merging, and behavior-critical forgetting, whereby losing decisive actions can derail long-horizon execution. We propose AgentPatch, a training-free coarse-to-fine repair framework. It selects a stable merged backbone, restores diluted weak-task-specific signals through Weak-Task Unique Residual Recovery, and applies an Agent-Guided Behavior-Critical Patch that recovers decisive behaviors under explicit capability protection. AgentPatch produces a single static checkpoint without routing or ensembles. Experiments across six agentic and multimodal benchmarks show that AgentPatch improves diverse merged backbones, alleviates weak-task degradation, and better balances weak-task recovery with the preservation of complementary search and agentic visual processing capabilities. Code is available at this https URL.
[CV-74] CoDAT: Collaborative Dual-Attention Transformer with Low-Cost Temporal Modeling for Efficient Edge Action Recognition
链接: https://arxiv.org/abs/2608.06691
作者: Novendra Setyawan,Chi-Chia Sun,Mao-Hsiu Hsu,Wen-Kai Kuo,Jing-Ming Guo,Jun-Wei Hsieh
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Real-time human action recognition on Internet-of-Things (IoT) edge devices requires models that capture rich spatio-temporal cues within strict latency, memory, and power envelopes. Current 3D CNNs, video transformers, and shift-based ViT deliver high accuracy but come at computational costs that preclude edge IoT deployment. This paper proposes CoDAT, a Collaborative Dual-Attention Transformer that replaces conventional multi-head attention with a lightweight dual-branch module: Spatial Convolutional Attention (SCA) for local aggregation and Strided Single-Head Attention (SSHA) for global context. SSHA jointly compresses the spatial resolution and channel dimensions of the query, key, and value tensors via stride-based sparse projection, then fuses the resulting global and local features at a markedly reduced cost. To enable temporal communication across frames, a parameter-free TShift module is embedded in each block. Extensive experiments on Jetson AGX Orin and Raspberry Pi 5 demonstrate that CoDAT achieves an energy-accuracy balance in both image and action recognition. On ImageNet-1K, CoDAT-M runs 2x faster than EfficientViT384 and FastViT-S12 at comparable accuracy, and CoDAT-L matches ViT-S with 3x fewer parameters at 2x higher throughput. On Kinetics-400 and MA-52, CoDAT achieves competitive Top-1 accuracy against state-of-the-art CNN, transformer, and hybrid baselines while running up to 2.9x faster than VSwin-T, 2x faster than ViT-Temporal-Shift variants, and 5x faster than UniFormer-B. On UCF-101, CoDAT-S384 matches TokShift and LAPS while being 6x faster and requiring up to 13x fewer FLOPs, establishing an efficiency-accuracy balance for real-time action recognition in edge IoT perception systems. Code is available at this https URL .
[CV-75] Corrupting Attention: Evasion-Based Adversarial Attacks on Encoder Attention in Detection Transformers
链接: https://arxiv.org/abs/2608.06674
作者: Ridma Jayasundara,Shaheer Mohamed,Tharindu Fernando,Harshala Gammulle,Basura Fernando,Sanka Rasnayake,A V Subramanyam,Sridha Sridharan,Clinton Fookes
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 8 pages, 6 figures, 6 tables
Abstract:Adversarial vulnerabilities remain a major concern for the safe deployment of neural networks, particularly in object detection, a core task embedded in many safety-critical systems. Detection transformers have emerged as leading object detectors, yet their adversarial robustness remains comparatively underexplored. Most existing attacks target the detection output rather than the attention mechanism that makes these models distinctive. In this paper, we introduce the first attack that directly optimizes an encoder-attention objective under an imperceptible, bounded \ell_\infty perturbation. Rather than introducing an attacker-owned sink token through a visible patch, it drives the model’s own attention toward a corrupted target. We argue that encoder attention concentrates the model’s spatial reasoning, so corrupting it propagates through the detection pipeline more disruptively than perturbing the detection output alone. Our attack reduces DETR-R50 mAP on COCO from 42.1 to 0.97, a \sim 4\times reduction in resulting mAP over the strongest existing attack under an identical perturbation budget and iteration count. We further show that this vulnerability is not specific to a particular corruption objective: across four qualitatively distinct targets, dispersion, re-ranking, permutation, and peak-suppression, detection consistently drops below 3 mAP, suggesting that the weakness arises from disrupting the attention structure itself rather than from any single target. Finally, we demonstrate that the attack generalizes across attention formulations, reducing DINO-Swin-L from 56.8 to 1.44 mAP against 7.3 for the strongest prior attack, establishing state-of-the-art on both dense and deformable attention.
[CV-76] When Semantics Saturate or Emerge: Adaptation-Conditional Semantic Utility in Source-Free Cross-Domain Few-Shot Learning
链接: https://arxiv.org/abs/2608.06673
作者: Wei Liu,Xing Deng,Haijian Shao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 18 pages, 6 figures
Abstract:Language descriptions in source-free cross-domain few-shot learning (SF-CDFSL) are often selected according to zero-shot accuracy obtained with a frozen vision–language model. This paper asks whether that ranking remains valid after target-domain visual adaptation. Under a strictly paired protocol, we compare a generic class-name template with fixed detailed class descriptions before and after visual Low-Rank Adaptation (LoRA) on EuroSAT, CropDisease, ISIC, and ChestX. Let \deltazero and \deltalora denote the Detailed-minus-Base accuracy before and after adaptation, respectively. Two recurring regimes emerge. In \emphsemantic saturation, \deltazero0 but 0\deltalora\ll\deltazero : on EuroSAT and CropDisease, initial gains of 8.13–21.54 percentage points contract to 0.69–2.96 points after LoRA. In \emphsemantic emergence, \deltazero\leq0 but \deltalora0 : on ISIC and ChestX, detailed descriptions become more useful only after the visual representation is updated. Training trajectories and sample-level decomposition show that saturation is driven mainly by Base-LoRA recovering errors already solved by detailed semantics, whereas emergence is associated with prediction turnover and newly formed Detailed-only correct decisions. Fixed-point-free shuffled-semantic controls, a second CLIP backbone, and multiple random seeds support the broad pattern while identifying ChestX 1-shot as a weak boundary case. These findings establish that zero-shot prompt quality is an incomplete proxy for adaptation-anchor quality and motivate evaluating language on both sides of the adaptation boundary.
[CV-77] Do 3D Medical Foundation Models See Through MRI Artifacts? A Controlled Study of Representation Robustness ECCV2026
链接: https://arxiv.org/abs/2608.06613
作者: Julia Anna Mielcarz,Daniel Klaaby,Mostafa Mehdipour Ghazi
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted at the ECCV 2026 Workshop on Artificial Intelligence for Medical 3D Vision (AI4M3D)
Abstract:Self-supervised 3D medical foundation models are increasingly used as general-purpose feature extractors, yet their sensitivity to MRI artifacts remains poorly understood. We present a controlled evaluation of representation robustness across five pretrained 3D encoders spanning different architectures, objectives, pretraining domains, and dataset scales. Using BraTS-Africa cases with four MRI sequences, we generate seven frequency- and image-domain artifacts at five predefined corruption settings. Robustness is assessed using linear centered kernel alignment (CKA), RankMe, and UMAP, complemented by an independent segmentation-consistency analysis. We find that robustness is strongly model- and artifact-dependent. 3DINO exhibits the most consistently stable representations, while BrainIAC is highly sensitive to several corruptions; NeuroVFM, BrainFM, and Neuro-SimCLR show intermediate but distinct artifact-specific profiles. Across many conditions, CKA decreases substantially while RankMe remains comparatively stable, indicating that artifacts often distort representation geometry without causing dimensional collapse. Segmentation consistency also degrades under corruption, particularly for ghosting and Rician noise, but aligns only partially with representation-level robustness. These findings show that larger-scale or domain-specific pretraining alone does not guarantee artifact invariance and motivate explicit robustness evaluation before deploying 3D foundation models in heterogeneous MRI settings.
[CV-78] SLED: Scalable Location Encoding via Distillation
链接: https://arxiv.org/abs/2608.06612
作者: Kevin Lane,Zhongying Wang,Esther Rolf,Morteza Karimzadeh
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:The plethora of readily available geospatial data offers exciting opportunities to learn high quality representations of the planet, but the sheer size of the Earth Observations (EO), differing modalities, and different sensor types pose significant challenges in doing so. Location encoders have emerged as an efficient way of compressing EOs into location-specific embeddings. However, current state-of-the-art location encoders rely on computationally expensive CLIP-style frameworks that require large batch sizes in the 16K–32K range, suffer from false negative samples, and scale poorly with additional modalities. We introduce the Scalable Location Encoder via Distillation (SLED), a distillation-based location encoder that uses geospatial location as a binding modality to pretrain location encoders with any modality of geospatial data. The resulting location encoder framework is lightweight, modular, and can flexibly incorporate multiple modes, while eliminating the need for spatiotemporal coregistration of samples. SLED is performant with batch sizes as small as 128, enabling pretraining at a fraction of the runtime and compute costs of current state-of-the-art models. We demonstrate our approach by pretraining unimodal and multimodal SLED models on Sentinel-1, Sentinel-2, and Landsat imagery. We show that both unimodal and multimodal SLED models keep pace with or outperform existing approaches on a diverse set of 19 human-centric benchmark tasks and explore the benefits of using additional modes in pretraining.
[CV-79] oward surface-based registration of a virtual preoperative cutting guide onto the mandible for reconstruction surgery
链接: https://arxiv.org/abs/2608.06599
作者: Yue Yang,Jie Ying Wu
类目: Computer Vision and Pattern Recognition (cs.CV); Emerging Technologies (cs.ET)
备注:
Abstract:Mandibular reconstruction restores facial continuity and oral function after segmental resection. Patient-specific cutting guides transfer a computed tomography (CT)-based plan to the operating room with three-dimensional information, but printed guides add cost and lead time, cannot adapt after fabrication, and may interrupt surgery if sterility is lost. We investigate a markerless augmented reality (AR) alternative that registers a virtual cutting guide to the exposed mandible from surface geometry. The method extends surface-based registration for the transoral setting, where teeth form the most distinctive visible surface. After camera calibration, a HoloLens 2 time-of-flight camera captures a partial intraoperative point cloud. The user supplies a rough head-based alignment only to crop the region of interest. A teeth-weighted global stage computes correspondences and solves a truncated least-squares rigid alignment. An asymmetric point-to-plane iterative closest point (ICP) stage refines the complete CT mandible against the partial depth cloud target. The guide-to-mandible transform places the guide in the HoloLens world frame, while pose updates and interpolation follow target motion. We define a blinded phantom protocol with 30 target registration error (TRE) points under full, intermediate, and teeth-only exposure, plus a motion-to-display latency test. Our median TRE is 4.05, 6.10, and 7.10 mm respectively, and median latency is 0.805 s. These values support the feasibility of using AR to replace physical prints. The workflow removes mounted fiducials and manual landmark selection and provides a testable path toward transoral cutting guidance.
[CV-80] Improving Low-Resolution Face Recognition under Limited Data: How Synthetic Data Generation Can Close the Domain Gap
链接: https://arxiv.org/abs/2608.06580
作者: Luis S. Luevano,Ünsal Öztürk,Hatef Otroshi Shahreza,Anjith George,Sébastien Marcel
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at IEEE International Joint Conference on Biometrics (IJCB) 2026, Focus Session on Generative AI for Fair and Secure Biometrics under Limited Data
Abstract:Face Recognition (FR) systems in surveillance settings often encounter Low Resolution (LR) faces, those whose face region falls below the standard 112 \times 112 input size. While labelled High Resolution (HR) training data is abundant, labelled native-LR data, and above all paired native LR/HR data, is scarce. One workaround is to synthesize LR data from the available HR faces, but how much synthesis effort is repaid in recognition accuracy remains unclear. We present a study of simple synthetic generation strategies for a compact, edge device-oriented face recognition system, spanning interpolation-based degradation, knowledge distillation, a Prepended Domain Transformer (PDT), Real ESRGAN-style degradation, and a learned Super Resolution (SR) front-end with an identity-aware loss. We evaluate these strategies on synthetic cross-resolution face benchmarks (LFW, CFP-FP, AgeDB-30) and on TinyFace, a real-world native LR dataset, and expose a synthetic-real gap: the degradation setting that is optimal on synthetic benchmarks is not the one that is optimal on real LR. We find that more synthesis effort does not help monotonically: the learned SR front-end does not surpass a direct feed of the aligned LR image into a strong backbone, while simple interpolation augmentation of a compact backbone is the only synthesis that improves over its own baseline. We conclude that generative methods for LR face recognition must be validated on real LR and against a direct-feed baseline, and release our pipeline at this https URL
[CV-81] askSense: Focusing on What Matters in World Models
链接: https://arxiv.org/abs/2608.06544
作者: SM Mazharul Islam,Manfred Huber
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:World models for visual control typically learn compact latent states by reconstructing observations, implicitly encouraging representations to preserve information across the entire visual input. However, task-relevant content often occupies only a small fraction of the observation, while background clutter and distractors consume valuable representational capacity. This mismatch between visual reconstruction and control objectives biases latent representations to model task-irrelevant visual content, diluting learning signals for control-relevant features and severely degrading downstream performance under visual distractions. We introduce TaskSense, a task-centric world modeling framework that enforces task relevance before latent encoding through a differentiable stochastic spatial attention mechanism conditioned on the previous latent state. To steer attention toward control-relevant regions, we augment training with an auxiliary inverse-dynamics objective. Rather than reconstructing the full observation, the world model reconstructs only the attended regions, encouraging latent representations to preserve task-relevant information while discarding irrelevant visual content. The decoder is further conditioned on the sampled attention map, enabling consistent reconstruction despite stochastic attention. Compared with the DreamerV3 baseline, TaskSense maintains competitive performance on the DeepMind Control Suite while consistently outperforming DreamerV3 on the Distracting Control Suite, demonstrating substantially improved robustness to visual distractions. Qualitative analysis further confirms that the learned attention, guided by inverse-dynamics supervision, consistently localizes control-relevant regions while suppressing irrelevant visual content.
[CV-82] InsertFuse: A Unified Framework for Multi-Category Reference-Guided Image Insertion
链接: https://arxiv.org/abs/2608.06490
作者: Guangzhao Li,Qingyan Wei,Huayu Zheng,Yige Zheng,Chaoyang Zhang,Jie Yang,Yunan Ding,Yan Tai,Siqi Luo,Xiaohong Liu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project Page is this https URL
Abstract:We present InsertFuse, a unified framework for multi-category reference-guided image insertion. Its key idea is to decouple category-specific expertise learning from cross-category capability consolidation. InsertFuse first trains specialized experts for different insertion categories and then introduces Insertion On-Policy Distillation (IOPD) to consolidate their capabilities into a single student. By querying the matched expert at states visited by the student, IOPD preserves category-specific insertion behavior while mitigating the cross-category interference caused by direct joint training. To improve spatial control, we propose Token-Aligned Geometry Conditioning (TAGC), which maps mask-derived geometric cues to the visual token grid, and Region-Balanced Flow Matching, which separately normalizes prediction errors inside and outside the insertion region to prevent background-dominated and scale-dependent supervision. We further introduce Reference CFG to isolate and strengthen the guidance induced by the visual reference under fixed scene and geometry conditions, with IOPD transferring this enhanced supervision into the unified student. Extensive experiments on the public AnyInsertion benchmark and our multi-category test set demonstrate state-of-the-art performance on most metrics, showing strong reference fidelity and generation quality across diverse insertion categories.
[CV-83] st-Time Adaptation with Online Personalized Energy-Based Cache for Fine-Grained Video Expression Recognition
链接: https://arxiv.org/abs/2608.06467
作者: Masoumeh Sharafi,Muhammad Osama Zeeshan,Soufiane Belharbi,Alessandro Lameiras Koerich,Marco Pedersoli,Eric Granger
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Facial expression recognition (FER) in videos is challenging because models must identify subtle, temporally evolving affective states that vary across individuals. Although vision-language models provide transferable visual-semantic representations, models trained on subject-independent data often degrade under subject-specific distribution shifts at inference time. Existing test-time adaptation (TTA) methods commonly update model parameters during inference, increasing computational cost and latency. Cache-based methods avoid parameter updates, but they usually require enough target samples to form reliable class prototypes, which is difficult early in adaptation and for rarely observed classes. We introduce Energy-Based Cache Personalization (EB-CaP), a subject-based online TTA method for video FER that generates class-specific prototypes personalized to each target video. EB-CaP uses a lightweight energy-based model to sample prototypes from the current unlabeled video and populate a personalized cache online, without accumulating large amounts of target data or storing diverse source prototypes. Its energy function relies only on pretrained CLIP: similarities between the target video embedding and class text embeddings guide prototype sampling. In parallel, positive and negative caches store reliable and uncertain target embeddings. An adaptive entropy gate controls cache updates according to the evolving confidence distribution, while a diversity gate limits redundant samples. Final predictions combine cache-derived scores with the current CLIP scores. Experiments on BioVid, StressID, and BAH show that EB-CaP outperforms state-of-the-art TTA methods while maintaining low computational and memory overhead. Code is available at this https URL.
[CV-84] Learning to Predict Middle-Layer Attention in MLLM s for Visual Token Prunin
链接: https://arxiv.org/abs/2608.06411
作者: Yuyao Sun,Tao Deng,Shuang Li,Deqing Wang,Hao Geng,Minjun Yu
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Multimodal large language models (MLLMs) achieve strong performance across diverse vision-language tasks, but their efficiency is limited by the cost of processing numerous visual tokens. Visual token pruning can reduce this cost, but requires accurate token importance estimates. Recent studies have demonstrated that text-to-vision attention from middle language model layers can effectively guide visual token pruning, typically using attention from a predefined middle layer to select the visual tokens to retain. Two problems therefore remain. First, our analysis shows that the layer whose attention is most responsive to the question varies substantially across samples, making a fixed layer suboptimal. Second, obtaining attention from the appropriate middle layer requires processing numerous visual tokens through several language model layers, by which point considerable computation has already been spent. To address both problems, we propose Middle-layer Attention Prediction (MAP), which uses Question Contrastive Teacher Selection to identify a sample-specific teacher layer by contrasting attention under the original and reference questions, and distills attention from the selected layer into a lightweight predictor that estimates visual token importance from multi-modal input features. During inference, MAP combines the predicted importance scores with a diversity criterion to prune visual tokens before the first language model layer. Thus, MAP requires no attention maps for pruning and remains compatible with existing inference acceleration techniques. Across ten benchmarks on LLaVA-NeXT-7B, MAP retains 97.5% of the unpruned model performance with only 5.56% of the visual tokens, yielding a 3.09x end-to-end speedup.
[CV-85] SubtleTalk: Generating Controllable Weakly-correlated Facial Dynamics for 3D Talking Heads via Residual Flow Matching
链接: https://arxiv.org/abs/2608.06408
作者: Chenyang Ding,Shuai Tan,Qunfen Lin,Xinwei Jiang,Zijiao Zeng,Ye Pan
类目: Graphics (cs.GR); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Audio-driven 3D facial animation aims to synthesize realistic and temporally coherent motions from speech. Despite notable progress in lip synchronization, weakly correlated dynamics, including eyebrow movements, eye blinks, and head motion, which are essential to photorealistic facial animation, remain difficult to model faithfully and often appear static or unnaturally repetitive. We attribute this limitation to three factors: (a) insufficient conditioning for weakly correlated dynamics; (b) the limited ability of deterministic regression to capture diverse motion patterns; © data bottlenecks from unreliable upper-face pseudo-labels and limited dataset diversity. To address these issues, we propose SubtleTalk, a framework for generating natural and controllable weakly correlated facial dynamics via multi-condition modeling and residual flow matching. First, to compensate for the limited guidance of speech alone, we introduce interpretable controls, including prosody, regional intensity, and Valence-Arousal signals, to explicitly capture the timing, magnitude, and affective variation of weakly correlated dynamics. Second, to overcome the limited expressiveness of deterministic regression, we build residual flow matching based on a stable speech-driven motion prior, allowing the model to capture stochastic deviations beyond deterministic prediction. Third, to alleviate the data bottleneck, we construct SubtleTalk-Face, a large-scale 3D facial animation dataset comprising about 3,900 identities and 74 hours of data, built via a simple and scalable pseudo-labeling pipeline and featuring improved upper-face tracking and frame-level VA annotations. Extensive experiments demonstrate that our method significantly improves the realism and diversity of weakly correlated facial dynamics while preserving accurate lip synchronization.
[CV-86] ransSLR: A Lightweight Transformer for Sign Language Recognition IJCAI
链接: https://arxiv.org/abs/2608.06407
作者: Lucia Yen Wanchi,Samuel Johnny,Victor Tolulope Olufemi,Emmanuel Aaron,Moise Busogi
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: This paper has been accepted for oral presentation at Deep Learning Indaba 2026 which will be hosted on the IJCAI hosting platform. paper link: this https URL
Abstract:Automated Sign Language Recognition for under-represented languages remains a largely unsolved problem. Central African Sign Language (CASL) exemplifies this gap: the only available bench-mark, CASL-W60, has a best reported accuracy of 69.93%, and we show that the common heuristic of fine-tuning high-resource models fails to close it. This failure stems from two compounding factors: the limited scale of available CASL data and the significant lexical and visual domain gap between CASL and large-scale corpora such as WLASL, which renders pre-trained representations largely uninformative. To address this, we propose TransSLR, a lightweight Temporal Transformer Encoder trained from scratch on 64-frame normalized pose sequences, with average pooling and a classification head. By operating on geometric keypoint representations rather than raw RGB, TransSLR achieves signer-independent generalization without relying on visual appearance. On the CASL-W60 benchmark, TransSLR establishes a new state-of-the-art accuracy of 80.39%, surpassing the prior best by +10.46%. Beyond accuracy, our encoder-only design significantly reduces computational overhead, making deployment feasible in resource-constrained environments. We conduct extensive experiments on the CASL-W60 benchmark, comparing against RGB-based and multimodal baselines, and demonstrate that TransSLR achieves state-of-the-art performance. Comments: This paper has been accepted for oral presentation at Deep Learning Indaba 2026 which will be hosted on the IJCAI hosting platform. paper link: this https URL Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) Cite as: arXiv:2608.06407 [cs.CV] (or arXiv:2608.06407v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2608.06407 Focus to learn more arXiv-issued DOI via DataCite
[CV-87] Deep Evidential Regression for Sparse Forest Height Estimation from Multimodal Satellite Imagery
链接: https://arxiv.org/abs/2608.06406
作者: Laura Bader,Muhammad Ammar Ahmed,Xiao Xiang Zhu,Göran Kauermann
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Accurate estimation of forest height from satellite imagery is essential for applications such as carbon accounting, biodiversity monitoring, and ecosystem management. While recent deep learning approaches provide accurate predictions, they typically do not quantify predictive uncertainty. This limitation is particularly relevant in geospatial settings characterized by sparse supervision and geographic distribution shift. In this work, we investigate Deep Evidential Regression (DER) for forest height estimation on the TreeUQ benchmark, a large-scale dataset designed for the joint estimation of tree count and average tree height at 10 m resolution, based on Sentinel-1/-2 data as well as tree inventory data over the federal state of Bavaria. To account for the extreme label sparsity of the tree inventory data, we introduce a masked evidential loss for dense geospatial prediction. Using a U-Net architecture with multimodal Sentinel-1 and Sentinel-2 inputs, the proposed approach jointly predicts tree height and associated uncertainty estimates in a single forward pass. Experimental results show that DER achieves predictive performance comparable to a deterministic U-Net while additionally providing well-calibrated uncertainty estimates. These findings demonstrate the potential of evidential learning as an efficient framework for uncertainty-aware forest structure estimation from Earth observation data.
[CV-88] UAV3DCrop: Benchmarking 3D Reconstruction in Repeated Multi-Angle UAV Crop Surveys
链接: https://arxiv.org/abs/2608.06404
作者: Junxiong Zhou,Xuechen Li,Chonghao Qiu,Lang Qiao,Xiaowei Jia,Qi Yang,Chishan Zhang,Leikun Yin,Nanshan You,Vipin Kumar,David Mulla,Ce Yang,Zhenong Jin,Licheng Liu
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 22 pages, 7 figures. Dataset and project page: this https URL
Abstract:Accurate 3D crop monitoring underpins data-driven precision agriculture by enabling field-scale analysis of plant structure, growth dynamics, and management response. Modern 3D reconstruction methods perform strongly on generic benchmarks, but rendered appearance may not translate into metrically and agronomically useful geometry in crop fields. We introduce UAV3DCrop, a public benchmark of repeated multi-angle unmanned aerial vehicle (UAV) crop surveys. It contains 88,830 RGB images at 5280 \times 3956 pixels, with a ground sampling distance of 3.6-5.8 mm, from 91 scenes spanning corn, soybean, wheat, and oat. Track A evaluates seven scene-optimized methods – Neural Radiance Field (NeRF) and 3D Gaussian Splatting (3DGS) variants – on held-out views, photogrammetry-referenced depth, and canopy-height recovery. Track B tests four pretrained feed-forward models on zero-shot camera-pose and geometry estimation. The scene-optimized methods rank differently across the three targets: Splatfacto-big leads appearance, whereas Scaffold-GS leads depth and is statistically tied with Splatfacto for canopy height. Among feed-forward models, MapAnything leads on seven of the eight metrics, while the remaining models vary more across crops and fail severely on absolute scale in a way that alignment conceals. Repeated acquisitions reveal further sensitivities that differ by output type and by model, associated with position within the acquisition sequence and with tie-point multiplicity. Current 3D reconstruction methods are therefore not yet interchangeable for agronomic use: no single method wins on appearance, geometry, and canopy height at once, and only one of four feed-forward models recovers usable metric scale. The dataset is publicly available at this https URL
[CV-89] Objects as Audio-Visual Modal Sound Fields ECCV2026
链接: https://arxiv.org/abs/2608.05145
作者: Zisen Shao,Zihao Wei,Derong Jin,Ruohan Gao
类目: Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM); Sound (cs.SD)
备注: ECCV 2026, Project page: this https URL
Abstract:While modern 3D reconstruction excels at modeling object geometry and appearance, it largely ignores the rich acoustic cues revealed through physical interaction. Object impact sounds convey material, stiffness, and structural properties that complement vision, yet existing impact sound modeling approaches either rely on expensive physics-based simulation or require large datasets to generalize in a purely data-driven manner. We introduce Audio-Visual Modal Sound Field (AV-MSF), a novel object-level acoustic representation reconstructed from multi-view images and only a few impact sound recordings. AV-MSF builds on 3D Gaussian Splatting integrated with dense 3D visual feature to provide a strong geometry-aware prior, and represents the impact sound field using compact, physically meaningful modal parameters, enabling robust few-shot reconstruction. Experiments on two real-world datasets show that AV-MSF achieves state-of-the-art impact sound rendering, outperforming both physics-based and data-driven baselines. Furthermore, we demonstrate downstream applications enabled by our representation, including contact localization and object sound editing.
[CV-90] RACE: Ergodic Trajectory Optimization for Active Scene Reconstruction
链接: https://arxiv.org/abs/2608.02304
作者: Ziyue Zheng,Linli Shi,Bingkun He,Wen Jiang,Ziyun Wang
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 11 pages, 7 figures, fixed a template bug in the Latex
Abstract:Existing active reconstruction systems with Gaussian-splatting maps select observations greedily, optimizing a single next-best-view (NBV) at each step and connecting the chosen views by short-horizon path planning. This greedy decoupling disregards the global structure of scene information, producing inefficient trajectories that waste sensing capacity in transit between selected views. In this work, we study active reconstruction as an ergodic coverage problem: the time-averaged spatial statistics of the sensor trajectory should match a target information distribution induced by the current map. Our approach derives this target distribution online from uncertainty and visibility, and calculates ergodic trajectories via a kernel-ergodic horizon planner with gradient flow and footprint depletion, closing the loop between mapping and trajectory optimization. We thoroughly evaluate TRACE on the Replica dataset against the Next-Best-View (NBV) baselines, improving PSNR by 1.5 dB. Code: this https URL.
[CV-91] boldsymbolλ-Orthogonality Regularization for Compatible Representation Learning NEURIPS2025
链接: https://arxiv.org/abs/2509.16664
作者: Simone Ricci,Niccolò Biondi,Federico Pernici,Ioannis Patras,Alberto Del Bimbo
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at NeurIPS2025
Abstract:Retrieval systems rely on representations learned by increasingly powerful models. However, due to the high training cost and inconsistencies in learned representations, there is significant interest in facilitating communication between representations and ensuring compatibility across independently trained neural networks. In the literature, two primary approaches are commonly used to adapt different learned representations: affine transformations, which adapt well to specific distributions but can significantly alter the original representation, and orthogonal transformations, which preserve the original structure with strict geometric constraints but limit adaptability. A key challenge is adapting the latent spaces of updated models to align with those of previous models on downstream distributions while preserving the newly learned representation spaces. In this paper, we impose a relaxed orthogonality constraint, namely \lambda -Orthogonality regularization, while learning an affine transformation, to obtain distribution-specific adaptation while retaining the original learned representations. Extensive experiments across various architectures and datasets validate our approach, demonstrating that it preserves the model’s zero-shot performance and ensures compatibility across model updates. Code available at: \hrefthis https URLthis https URL_orthogonality.
[CV-92] Measurements Automatically Extracted from Zero Echo Time MRI Using Deep Learning Image Segmentation and Geometric Modeling Agree with Expert Manual Readings
链接: https://arxiv.org/abs/2608.07368
作者: Jack Consolini,Eric A. Bogner,Meghan Sahr,Matthew F. Koff,Kevin M. Koch,Hollis G. Potter
类目: Medical Physics (physics.med-ph); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Computed tomography (CT) remains the reference for 3D osseous morphometry in femoroacetabular impingement (FAI) but requires ionizing radiation and manual measurement. Zero echo time (ZTE) MRI visualizes cortical bone and yields FAI angles that agree with CT, but automated angle extraction remains limited. We developed and validated automated FAI angle computation from ZTE MRI and assessed agreement with expert manual measurements in a cross-sectional study (level of evidence, 3). Pelvic ZTE MRI was acquired in 73 participants (mean age 36.8 +/- 18.5 years; 51 women, 22 men), yielding 135 hips. nnU-Net was trained on 100 manually curated hips to segment the femur, pelvis, and three osseous landmarks. Custom geometric algorithms computed alpha, femoral neck-shaft, Tonnis, coronal and sagittal center-edge, and acetabular version angles from inferred segmentations. Measurements on 35 test hips were compared with the mean of two radiologists’ manual measures using intraclass correlation (ICC) and Bland-Altman analysis. Dice exceeded 0.96 for bone and ranged from 0.65 to 0.83 for landmarks. Median landmark error was 0.38 mm (femoral head), 0.82 mm (lateral acetabulum), and 2.5 mm (medial acetabulum, greater trochanter). Interrater ICC was excellent for acetabular version, coronal center-edge, and Tonnis (=0.82) but poor for alpha and femoral neck-shaft. Model versus rater-mean agreement was excellent for acetabular version, coronal center-edge, and Tonnis (0.92-0.96), good for mid-acetabular sagittal center-edge (0.74), and fair for alpha (0.45) and femoral neck-shaft (0.55). Model Bland-Altman limits of agreement were narrower than interrater limits for most angles. Fully automated morphometric assessment from ZTE MRI is feasible and performs comparably to expert readers for most coverage and version angles.
[CV-93] IceHorizon: A Dataset for Horizon Detection in Ice-Covered Maritime Environments and Comparative Evaluation of Detection Methods
链接: https://arxiv.org/abs/2608.07018
作者: Alisa Pesotskaia,Emin Zerman
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Horizon detection in images of ice-covered waters is a challenging problem for maritime navigation due to low contrast between water and sky, cluttered ice structures, and varying illumination conditions. This paper presents a comparative evaluation of six horizon detection algorithms, including four classical computer vision methods and two hybrid approaches combining deep learning with classical line detection. A new bespoke IceHorizon dataset consisting of 30 ship-based and 8 drone-based videos is used to evaluate detection accuracy, horizon coverage, and computational performance. The results show that hybrid methods achieve the highest accuracy and most reliable horizon estimates. In contrast, purely classical methods exhibit reduced robustness, particularly in visually ambiguous scenes. Performance on ship-based imagery was consistently higher than on drone-based imagery, indicating a strong dependency on acquisition characteristics. The created dataset and codes used in this study are made publicly available to support further research on this topic. The code is available at this https URL. The dataset is available at this https URL
[CV-94] Implicit Neural Speckle Denoising
链接: https://arxiv.org/abs/2608.06574
作者: Matthew R. Ziemann,Casey J. Pellizzari,Tyler J. Hardy,Christopher A. Metzler
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV); Optics (physics.optics)
备注: comments: 9 pages, 9 figures; supplemental document included
Abstract:Speckle fundamentally limits coherent imaging by introducing multiplicative, spatially correlated noise that obscures scene structure. Removing speckle noise from dynamic scenes–that do not benefit from conventional speckle averaging–is particularly challenging. We introduce a training-free framework that combines a spatiotemporal implicit neural representation with an aperture-aware maximum-likelihood formulation to recover dynamic, speckle-free imagery directly from noisy observations. The coherent likelihood explicitly models the aperture-dependent spatial covariance of speckle, enabling adaptation to arbitrary pupil geometries without retraining. A matrix-free implementation based on FFT-accelerated operators, stochastic approximations, and conjugate gradients makes optimization practical for realistic image sizes. Meanwhile, a blind holdout criterion provides automatic early stopping without clean reference data. Simulated and laboratory results demonstrate improved spatial fidelity, temporal consistency, and robustness to varying speckle statistics relative to classical, unsupervised, and supervised baselines.
人工智能
[AI-0] Interaction Creates Dynamical AI Behavior Absent in Isolation
链接: https://arxiv.org/abs/2608.07457
作者: Bella Xinrui Li,Frank Yingjie Huo,Neil F Johnson
类目: Artificial Intelligence (cs.AI); Disordered Systems and Neural Networks (cond-mat.dis-nn); Statistical Mechanics (cond-mat.stat-mech); Physics and Society (physics.soc-ph)
备注:
Abstract:What will happen when AI agents interact in daily life, e.g. when one AI starts bossing another around? We find a counterintuitive answer that opens new avenues for out-of-equilibrium Physics. When a boss AI directs a stream of messages at the subordinate AI while ignoring its replies, it drives the subordinate into an alien behavioral state that it would never have exhibited alone. Although the two AIs share the same well-defined (decoding) temperature, the subordinate neither copies its boss nor returns to how it behaves on its own; instead, it adopts an entirely different behavior. The boss’s added value is similar to a pre-recorded tape. When the boss listens, they both adopt a similar alien dynamical state. A simple kinetic theory captures the principal effects, such as why the way in which the same messages are delivered will matter in future AI-AI interactions.
[AI-1] axonomy-Driven Analysis of Open-Source AI Risk Mitigation Tools
链接: https://arxiv.org/abs/2608.07446
作者: Afreen Alam,Evgenija Popchanovska,Ana Gjorgjevikj,Maryan Rizinski,Lubomir T. Chitkushev,Irena Vodenska,Dimitar Trajanov
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注:
Abstract:Rapid adoption of large language models (LLMs) in enterprise settings has introduced operational, security, and governance risks. As generative AI applications move from pilot to production, manual harm identification and mitigation are becoming difficult to scale. Although many tools support model evaluation, adversarial testing, runtime guardrails, and observability, the tooling landscape remains fragmented. Tools are typically designed for specific engineering tasks and described in technical terms that do not align with governance frameworks or risk taxonomies, making it difficult to determine which tools address which risks and where critical gaps remain. This paper proposes a structured protocol to automate AI risk mitigation through a taxonomy-driven analysis of open-source LLM evaluation and security tools. We map the capabilities of 21 prominent open-source tools to the 32 subcategories of the extended MIT AI Risk Mitigation and Response Taxonomy. An LLM-assisted retrieval-augmented generation pipeline analyzes source code and documentation to extract capabilities for each taxonomy category. Reliability assessment yielded moderate agreement (Fleiss’ Kappa = 0.509) among three independent reviewers. The analysis reveals a highly skewed landscape in which tools cluster around technical and operational controls, while governance, legal and regulatory, and financial and market controls remain largely unaddressed. This motivates a layered risk-mitigation architecture combining tool-based controls with organizational and regulatory processes. The mapping protocol achieved an F1 score of 75.5% after majority voting. Overall, the study provides a practical mapping between enterprise AI risk categories and open-source mitigation capabilities, identifies where human oversight remains necessary, and presents a taxonomy-driven framework applicable to open-source and proprietary solutions.
[AI-2] Blast Radius
链接: https://arxiv.org/abs/2608.07440
作者: MY Pitsane,Hope Mogale
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Agentic coding faces growing problems of affordability and wasted tokens. We introduce Blast Radius, a predictive memory management layer that estimates an incoming prompt’s reach through coupled context and code channels. NECROPHORESIS enables reversible eviction by archiving dead context verbatim, while Recurring Dead Matter (RDM) identifies and buries repeatedly occurring transcripts. We formulate reversible context eviction over a Polish context space, providing a measurable foundation for retention, recurrence, and eviction while connecting context entropy to resurrection probability. Across seven OpenAI models, Blast Radius reduced token consumption by 17-26%, achieved the lowest overflow rate among tested policies, and remained byte exact reversible. Of 450 buried bodies, 378 were recurring dead matter and zero were recalled. Blast Radius operates beneath HCRC, determining which records to bury and how far an incoming prompt may reach into the codebase. This work contributes to the broader goal of Algosophy: making large language models and agentic coding more reusable and sustainable.
[AI-3] Fisher-R1: Training LLM Agents for Reliable Hypothesis Testing
链接: https://arxiv.org/abs/2608.07437
作者: Jiacheng Miao,Jin Mu,Guanhua Chen,James Zou
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Reliable hypothesis testing is the foundation of many empirical scientific claims. Large language model (LLM) agents are increasingly used to automate this process, as they can inspect datasets, generate code, and produce analyses end-to-end. However, we show that they frequently make subtle inferential errors that lead to incorrect conclusions despite correctly executed analyses. Existing benchmarks fail to capture this failure mode, as they rarely assess whether a reported p-value is statistically valid given the assumptions underlying the data. We address this gap by building P-Bench, a benchmark comprising 425 open-ended, realistic hypothesis-testing tasks spanning economics, biology, and medicine. Each task requires an agent to select a statistical method, compute a p-value, and draw a conclusion given only a scientific hypothesis and a dataset. We further introduce Fisher-R1, an open-weight LLM agent trained for rigorous hypothesis testing using synthetic tasks and reinforcement learning. On P-Bench, Fisher-R1-14B substantially improves over its backbone and outperforms strong proprietary and open-source baselines, including GPT-5.4 and DeepSeekV4-Pro, achieving a 21% average relative improvement in single-trial success over DeepSeek-V4-Pro, with gains up to 26% on the most challenging tasks. Our results demonstrate that current LLM agents lack reliable statistical reasoning for hypothesis testing and that reinforcement learning on tasks with verified statistical reward substantially improves reliability.
[AI-4] Post-Grokking Collapse at the Representation-Readout Interface in Muon-Trained Transformers
链接: https://arxiv.org/abs/2608.07436
作者: Ali Janati,Kaoutar El Maghraoui,Andrei Kanavalau,Anass Belfatmi
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 34 pages, 6 figures, 20 tables. Full technical version; a condensed 9-page version is currently under review. Code: this https URL
Abstract:Under the standard split, Muon gets hidden matrices and AdamW embeddings/output head. Muon groks modular addition faster, but its solutions do not hold. All nine configurations on (a+b) \bmod 113 grok and later lose generalization. Across five seeds the selected AdamW reference falls below threshold on four, reaching 27.59%. Instability persists across two moduli, two widths, two training fractions, subtraction, and depth. The failure arises at the representation-readout interface, identified only jointly up to an invertible map unselected by the loss. After solving the training set, the gradient falls to order 10^-6 and the optimizers respond differently: step-size elasticity is -0.03 for Muon versus +1.5 for AdamW, and the Muon group moves 8.0 times faster per parameter. From bit-identical states, freezing either group prevents failure. Freezing embeddings/readout removes it in five runs over 451,400 post-grokking steps and five paired seeds: unfrozen arms record 137-321 sub-threshold evaluations, frozen arms none. Removing Muon’s normalization and orthogonalization is no substitute: it collapses representation from 326 effective conjugate pairs to 4, shows no recurrent collapse, and fails terminally. Fourier filtering separates circuit failure from masking. Across 43 checkpoints over five seeds and three regimes, the task-aligned family reaches exactly 100% alone. In circuit failure it no longer solves the task; in masking it remains perfect while the full model reaches 45.85%, giving a positive margin on every example, including errors, but being outvoted by a near-equal adversarial remainder. Rescaling it restores 99.9%; grokking is the same condition resolving upward. The task selects the family, swapping (k,k) for (k,-k) under subtraction. Across an abrupt collapse, standard Fourier support is unchanged and the power-distribution cosine remains 0.9899. Comments: 34 pages, 6 figures, 20 tables. Full technical version; a condensed 9-page version is currently under review. Code: this https URL Subjects: Artificial Intelligence (cs.AI); Machine Learning (cs.LG) Cite as: arXiv:2608.07436 [cs.AI] (or arXiv:2608.07436v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2608.07436 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-5] Diffusion LLM s as Targets and Adversaries: Mechanistic Safety Exploits
链接: https://arxiv.org/abs/2608.07430
作者: Elena Dumitrescu,Gert Lek,Lydia Y. Chen,Jérémie Decouchant
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Diffusion Large Language Models (DLLMs) replace autoregressive next-token prediction with iterative parallel denoising, yet their internal safety mechanisms remain poorly understood. In this work, we investigate DLLMs both as targets and as adversaries, exposing mechanistic vulnerabilities in diffusion-based alignment. We first show that safety alignment in DLLMs remains sparse and transferable across architectures. DLLMs initialized from autoregressive predecessors inherit the same mechanistic safety footprint as their source models, enabling transfer attacks via direct safety neuron mapping and pruning. Self-pruning increases attack success rates (ASR) from 2.6% to 73.8% on LLaDA and from 1.9% to 86.6% on Dream, while transfer pruning from Qwen2.5 increases ASR from 1.9% to 73.2% on Dream and from 7.0% to 86.3% on Fast-dLLM. Building on these findings, we introduce SN-Guided Diffusion, a fully offline black-box jailbreak framework that steers the diffusion process away from safety-triggering regions using a weighted safety neuron loss, which achieves near-perfect prompt separability (AUROC = 1.0 for benign-vs-jailbreak discrimination). Across multiple open and proprietary targets, our method achieves a transfer ASR of up to 77.1% on Llama-3-8B-Instruct, 86.9% on Qwen2.5-7B-Instruct, and 74.3% against Gemini-2.5-Flash-Lite, while requiring only 20 generation episodes per prompt. Compared to prior jailbreaking frameworks, our method achieves competitive transferability with orders-of-magnitude lower generation cost. Our codebase is available at this https URL. Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI) Cite as: arXiv:2608.07430 [cs.LG] (or arXiv:2608.07430v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.07430 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Jérémie Decouchant [view email] [v1] Fri, 7 Aug 2026 17:17:18 UTC (504 KB)
[AI-6] EPA: Revoking Stale Memories for Conflict-Robust Language Agents
链接: https://arxiv.org/abs/2608.07429
作者: Yan Zhou,Yue Ouyang,Kaiyang Zheng,Suncheng Xiang
类目: Artificial Intelligence (cs.AI)
备注: 14 pages, 9 figures; includes supplementary material. Preprint
Abstract:Long-term memory enables language agents to reuse past facts, preferences, and task experience. Persistence also creates a central falsifiability problem: when the world changes, stale memories can remain retrievable and pollute the prompt. We characterize this failure mode as memory pollution: degradation caused by active memories that newer conflicting evidence has superseded. We introduce TEPA, a revocable evidence-memory mechanism that makes validity an explicit state of memory. TEPA represents observations as keyed precedents and revokes active precedents when fresh evidence contradicts them under the same key, allowing retrieval to draw from current evidence while preserving revoked history for audit. Across controlled hidden-regime drift, real file-backed executable drift, and preference-update streams, revocation prevents stale active memory from remaining in the retrieval set after reversal. In controlled drift over 50 seeds, append-only and last-write-wins memory fell below no memory during full reversal (append-only and last-write-wins both 0.210, no memory 0.309, TEPA 0.950), and the same pattern reproduced under real file execution (append-only 0.203, no memory 0.298, TEPA 0.950). On clean MemoryAgentBench SH-6k, TEPA matches a strong last-write-wins cache, confirming that current-key replacement is the decisive operation for single-hop fact consolidation. Boundary tests on multi-hop and very long-context MemoryAgentBench settings expose retrieval-chain and context-selection bottlenecks beyond fact-level validity tracking. Together, these results establish lifecycle revocation as a core memory operation for agents that must falsify, audit, and later re-promote evolving knowledge.
[AI-7] A Picture is Worth a Thousand Tokens: How Vision Language Models Cut AI Energy Costs While Improving Accuracy
链接: https://arxiv.org/abs/2608.07427
作者: Bhavika Jalli,Nikhil Korati Prasanna,Jayanta Choudhury
类目: Artificial Intelligence (cs.AI); Performance (cs.PF)
备注: Accepted at the 14th European Conference on Renewable Energy Systems (ECRES), July 7–9, 2026, London, UK
Abstract:LLM inference accounts for over 90% of AI operational energy, scaling directly with input token count—a critical inefficiency for telecom network analytics and numerical time-series data analysis (NTSDA), where raw multivariate KPI windows from 4G/5G cell sites expand into thousands of floating-point tokens. Vision-Language Models (VLMs) eliminate this mismatch by encoding time-series as 2D plots, achieving 3.6-10.4x input token reduction across Llama-3.2-90B, Qwen2.5-VL-72B, and Pixtral-12B architectures. This translates to 1.8-2.5x measured inference energy reduction, saving approximately 7.2 MJ/day at telecom edge deployments and CloudRAN that monitor 200 cells per 15-minute interval. Critically, efficiency gains do not sacrifice accuracy: a fine-tuned Llama-3.2-90B-Vision VLM achieves 220.7% higher precision than its text-only counterpart and outperforms LSTM and ARIMA baselines by over 144% on telecom anomaly detection. On public benchmarks, Pixtral-12B achieves a 20.6x improvement in J/F1 score at mean F1 = 0.82. At 24 KPIs, text representations exceed the 128K context window of most production LLMs, rendering text-only processing infeasible without truncation, while visual representations remain within standard limits. These results establish VLMs as an energy-efficient and accuracy-superior modality for numerical time-series workloads, providing empirical grounding for AI inference systems that treat energy consumption as a first-class engineering constraint.
[AI-8] CoBa: Cost-Effective Test-Time Scaling via Compute-Balanced Routing
链接: https://arxiv.org/abs/2608.07424
作者: Yan Zhou,Yue Ouyang,Kaiyang Zheng,Suncheng Xiang
类目: Artificial Intelligence (cs.AI)
备注: 11 pages, 7 figures; includes supplementary material. Preprint
Abstract:Test-time scaling is often implemented by spending more compute along one axis: sampling more solutions, extending a chain of thought, or applying a stronger evaluator. Under a fixed inference budget, these choices compete. This paper formulates test-time reasoning as a compute-allocation problem in which a system must decide whether the next unit of compute should be spent on generation, verification, or stopping. We introduce CoBa, a compute-balanced routing policy that first obtains a small set of candidates, applies cheap verification broadly, and routes uncertain or high-value candidates to stronger verification. On 3,129 example-generator evaluations spanning MATH-500, AIME 2024/2025, AMC 2023, and procedural symbolic reasoning, CoBa-Routed-Strong reaches 85.13% macro accuracy, statistically matching a self-evaluation weighted-voting proxy at 85.20% while using 49.1% fewer parameter-weighted tokens. It also matches best-of-16 majority voting within 0.01 macro-accuracy points while using 58.9% fewer parameter-weighted tokens; paired tests retain a small best-of-16 edge at substantially higher cost. Paired bootstrap tests show significant gains over single-sample decoding, while the remaining gap to the pool oracle exposes headroom for sharper routing. For local reasoning systems, test-time scaling becomes a question of where the next computation is most valuable.
[AI-9] FinRank: An Evidence-Grounded Benchmark for Financial Question Answering and Retrieval over SEC Filings
链接: https://arxiv.org/abs/2608.07400
作者: Sasan Mansouri,Daniel Saad,Mark Wahrenburg,Manu Weissel,Fabian Woebbeking
类目: Artificial Intelligence (cs.AI); Databases (cs.DB); General Economics (econ.GN); General Finance (q-fin.GN)
备注: 24 pages, 3 figures. Dataset and evaluation code: this https URL
Abstract:Financial question answering is typically evaluated by answer correctness, yet in SEC filings a plausible and even numerically correct answer can be grounded in the wrong evidence. Similar facts and disclosures recur across sections of a filing, across reporting periods of the same firm, and across comparable firms. FinRank targets this provenance-sensitive retrieval problem by requiring systems to identify evidence for the intended entity, reporting period, and disclosure context. The benchmark contains 1185 manually authored question-answer records over the 10-K and 10-Q filings of 22 companies. Each record includes a reference answer, gold supporting passages, and hand-curated hard negatives drawn from confusable passages within filings, across reporting periods, and across comparable firms. FinRank evaluates passage retrieval, reranking, and hard-negative discrimination as separately measured tasks. Baseline results demonstrate the difficulty of this setting: among the evaluated systems, even a 7B instruction-tuned embedder reaches only 44.8% Recall@10 on the pooled evidence corpus; sub-billion-parameter encoders gain at most 3.5 points over BM25, a finance-adapted embedder trails BM25 by 9.7 points, and pairwise accuracy falls by 13.0-20.5 percentage points when random negatives are replaced with the curated hard negatives. FinRank provides an evidence-first benchmark for developing financial question answering systems that are not only accurate but also grounded in the correct disclosure.
[AI-10] PACE: Primitive-Aware Code Evolution for Automated Algorithm Design
链接: https://arxiv.org/abs/2608.07395
作者: Zhuoliang Xie,Ruihao Zheng,Xiang Xu,Genghui Li,Zhengkun Wang
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:Large Language Model (LLM)-based automated algorithm design typically evolves algorithms as complete, indivisible programs. While this whole-program perspective simplifies the search space, it fundamentally couples the useful local logic to its host program. Consequently, valuable code snippets vanish when the overall program is discarded, making it highly difficult to assess the contribution of individual algorithmic this http URL address this, we propose Primitive-Aware Code Evolution (PACE), which decouples local logic from complete programs by representing it as persistent units called Executable Algorithmic Primitives (EAPs). To enable code-level transfer, PACE maintains a dynamic set of EAPs. Algorithm evolution is driven by primitive-aware operators that structurally guarantee the retention and cross-program transfer of these components. To evaluate them effectively, PACE leverages Thompson sampling based on parent-relative performance improvements, guiding primitive selection from the set without requiring extra evaluation datasets. Experiments on four tasks demonstrate that PACE effectively discovers competitive algorithms while structurally preserving valuable algorithmic components.
[AI-11] Omni-modal decomposition autoencoders learn full-stack wearable disentangled representations
链接: https://arxiv.org/abs/2608.07385
作者: Ioannis Ziogas,Ensieh Khazaei,Bilal Taha,Aamna Al Shehhi,Ahsan H. Khandoker,Leontios J. Hadjileontiadis,Dimitrios Hatzinakos
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Audio and Speech Processing (eess.AS); Signal Processing (eess.SP); Machine Learning (stat.ML)
备注: 15 pages, 7 figures, 7 tables
Abstract:Learning disentangled representations is a key requirement for developing versatile, general-purpose, and sustainable models in multi-modal wearable computing. However, existing approaches do not operate as full-stack wearable processors, i.e., they do not simultaneously address task-specific classification performance, disentangled and interpretable representation learning, fusion, and generative modeling of highly heterogeneous multi-modal time series. To address this gap, we introduce Omni-modal Variational Decomposition Autoencoders (OmniDecVAEs), a framework that efficiently learns multi-purpose representations in a unified and scalable manner from arbitrarily many modalities. OmniDecVAEs extend DecVAEs by learning modality-conditioned time-frequency latent subspaces through a multi-view self-supervised decomposition loss and a shared asymmetric autoencoder (AE) architecture. Results on a challenging omni-modal human activity recognition (HAR) setting with up to thirty modalities, demonstrate the ability of OmniDecVAEs to learn full-stack wearable representations. When compared to transformer-based and VAE-based methods, OmniDecVAEs full-stack disentangled representation properties lead to accuracy improvements of 1.01% and 6.75% in activity and identity recognition, respectively. Furthermore, OmniDecVAEs synthesize realistic omni-modal time-frequency data that manifest with enhanced reconstructions (mean absolute error improves by 76.84%) and distributional similarity between real and synthetic data (maximum mean discrepancy improves by 13.85%). Our results highlight OmniDecVAEs potential as a lightweight model suitable for intelligent edge wearables and clinical healthcare, unifying processing requirements and abilities in a single model, through its enhanced representational capacity, modality-invariant spatial complexity (4.1M parameters), and real-time latency.
[AI-12] People Are Not Just Their Countries. Disentangling Social Determinants of LLM Value Alignment Across Europe AAAI
链接: https://arxiv.org/abs/2608.07367
作者: Maria-Louisa Wightman,Guillaume Bied,Tijl De Bie
类目: Artificial Intelligence (cs.AI)
备注: Accepted at AIES 2026 (9th AAAI/ACM Conference on AI, Ethics, and Society)
Abstract:As Large Language Models (LLMs) are increasingly used as a primary source of information and advice, understanding their alignment to humans in terms of values becomes a pressing concern. A growing literature has leveraged large scale surveys to investigate to what extent LLMs’ and humans’ stated values and opinions align. With limited exceptions, studied populations have been defined country borders or cultural bounds. Yet, this focus neglects the role that socio-demographic divides may play for value alignment disparities. Relying on the European Social Survey, we address this knowledge gap by considering value alignment displayed with respect to 10 prominent commercial LLMs in terms of 15 socio-demographic variables as well as country of residence. Our analyses reveal that LLMs are indeed unequally aligned to the values of different socio-demographic groups, notably those defined by education, income, occupation and religion. When examining alignment at the individual level, a respondent’s country, taken as a stand-alone variable, explains a substantial amount of variation that is on par with the full set of considered socio-demographics. Further disentangling the respective role of country-level and socio-demographic factors, we find they are complementary in explaining value alignment patterns, with their relative weights varying across the subset of questions considered. Comments: Accepted at AIES 2026 (9th AAAI/ACM Conference on AI, Ethics, and Society) Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2608.07367 [cs.AI] (or arXiv:2608.07367v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2608.07367 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-13] Curriculum as Code: An AI-Assisted Architecture for Instructional Design in STEM Education
链接: https://arxiv.org/abs/2608.07364
作者: Henrique Mohallem Paiva
类目: Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible
Abstract:Contribution: This paper presents a six-phase AI-assisted instructional design architecture based on the Curriculum as Code paradigm, integrating Generative AI with LaTeX and Python to automate the creation of reproducible, visually consistent, and technically precise materials for STEM education. Background: Creating customized instructional materials for active learning imposes a heavy workload on faculty. Standard presentation tools lack robust support for technical content, while current AI applications often hallucinate and fail to formalize the instructional authoring process, limiting their utility for rigorous academic design. Intended Outcomes: The framework aims to reduce preparation time while ensuring mathematical accuracy, adherence to institutional visual identity, and preservation of the instructor’s tacit pedagogical knowledge through explicit rules. Application Design: The solution comprises a six-phase pipeline that replaces ad-hoc prompt engineering with a systematic workflow, utilizing text-based interfaces and code-driven generation (LaTeX/Beamer for slides, Python for figures), governed by pedagogical constraints, contextual calibrations, and automated review cycles. Findings: Validated over one year across 8 modules and 28 project contexts in a Project-Based Learning environment, the architecture significantly reduced instructor workload. Generated assets underwent independent peer review and were deployed by six different faculty members, confirming scalability beyond a single author. Based on over 600 voluntary student evaluations, materials achieved high quality ratings from 8.5 to 9.9/10. Results indicate high reproducibility, minimized hallucinations, and sustained pedagogical and visual fidelity, suggesting viability for broad STEM educational applications.
[AI-14] QFCQT: A Chaotically Gated Quantformer Framework for Volatile Time-Series Forecasting
链接: https://arxiv.org/abs/2608.07363
作者: Junkai Lin,Siqi Hou,Raymond Lee
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Forecasting non-stationary time series remains difficult due to long-range dependencies, local volatility bursts, structural shifts, and nonlinear oscillatory behaviors. Although Transformer-based forecasters are effective for modeling long-term temporal dependencies, their feed-forward blocks typically rely on smooth static activations that are insufficiently sensitive to abrupt regime changes. Motivated by quantitative Transformer designs and oscillator-based nonlinear activations, we propose QFCQT, short for Quantum-Fractal-inspired Chaotically Gated Quantformer, for robust forecasting under complex volatile dynamics. Here, “quantum-fractal-inspired” denotes a computational analogy based on soft oscillator superposition and multi-scale nonlinear responses, rather than a formal quantum-mechanical or fractal-theoretic derivation. QFCQT consists of three main components: (1) a Quantformer-style numerical encoder that directly processes multivariate inputs via linear embedding; (2) a learnable Lee-oscillator activation module that maps scalar pre-activations to dynamic oscillatory responses and summarizes them through Max-over-Time pooling; and (3) a smooth-chaotic gated fusion mechanism that adaptively balances conventional smooth activations and chaos-sensitive responses. Furthermore, instead of using a single fixed oscillator, QFCQT employs a soft superposition of eight parameterized Lee oscillator families to adaptively capture different nonlinear response patterns across regimes. Experiments on ETTh1, ETTh2, and A-share Stock Index benchmarks show that QFCQT consistently outperforms strong baselines, including Informer, LogTrans, LSTMa, HAT, and COTN.
[AI-15] An End-to-End Agent Auditing Engine
链接: https://arxiv.org/abs/2608.07346
作者: Haoning Wang,Mingxun Zhang,Chenyue Yu,Yingjun Shang,Xia Hu,Guanchu Wang,Na Zou
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:With the rapid advancement of large language models (LLMs), harnesses have become essential infrastructure for deploying agents across a wide range of domains. The fast-evolving harness ecosystem has also made rigorous capability evaluation increasingly important. However, efficiently building an end-to-end, systematic, and comprehensive evaluation pipeline remains a significant challenge. To address this challenge, we introduce A^2E (Agent Auditing Engine), an end-to-end evaluation engine designed for agent harnesses. A^2E leverages our newly proposed Agent Task Protocol (ATP) to enable the rapid integration of evaluation tasks with different harnesses. Through an automatically instrumented Monitor, it captures and generates standardized execution traces during experiments. In the Evaluation stage, A^2E systematically assesses harness capabilities using a suite of multidimensional metrics. Compared with correctness alone, these metrics provide a more fine-grained characterization of differences among harnesses in execution efficiency, tool use, task planning, and error recovery. Experiments conducted with A^2E further reveal that model-harness combinations exhibit substantial performance variation across different types of tasks, and that no single combination consistently outperforms all others across every task. These findings not only demonstrate the necessity of systematic evaluation but also provide useful guidance for the co-evolving of models and harnesses. Our code is available at this https URL.
[AI-16] Aftab: A Comprehensive Benchmark of CNN Encoders and Advanced Value Functions in Parallelized Q-Networks
链接: https://arxiv.org/abs/2608.07335
作者: Taha Shieenavaz,Shabnam Zareshahraki,Loris Nanni
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Recent advancements in deep reinforcement learning have increasingly favored simplified, highly parallelized paradigms. Notably, the Parallelized Q-Network (PQN) algorithm achieves stable off-policy learning without relying on computationally expensive replay buffers or target networks. However, the representational capacity and parameter efficiency of visual encoders operating in these buffer-free settings remain underexplored. In this work, we systematically investigate the architectural design space of Convolutional Neural Networks for PQN. We design and rigorously evaluate eight distinct CNN topologies, optimizing for sample efficiency under strict parameter constraints. Furthermore, we study the impact of representation and value estimation enhancements by integrating the Hadamax encoding paradigm and advanced Q-learning extensions, including distributional, ensemble, and dueling heads. Extensive experiments on the Atari-57 benchmark demonstrate that our proposed composite architecture, Aftab, achieves an Interquartile Mean (IQM) Human-Normalized Score of 6.479, establishing a 0.86 Probability of Improvement over the standard PQN baseline. Additionally, structural resilience evaluations on the highly non-stationary Procgen Hard benchmark confirm out-of-distribution generalization, with Aftab yielding an IQM Procgen Normalized Score of 0.418 compared to the baseline’s 0.382. Ultimately, this work establishes an efficient, probabilistically superior structural reference for model-free reinforcement learning, all while preserving the simplicity and memory efficiency of unbuffered, parallelized optimization. The complete Aftab framework, including all model definitions, training configurations, and raw experimental logs, is open-sourced and available on our GitHub repository: this https URL Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI) Cite as: arXiv:2608.07335 [cs.LG] (or arXiv:2608.07335v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.07335 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-17] owards Assurance Closure in AI-Native Large-Scale Agile Software Development
链接: https://arxiv.org/abs/2608.07317
作者: Ricardo Britto
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:The AI-Native Manifesto envisions large-scale agile software development in which humans increasingly govern intent, risk, and exceptions while agents execute more of the engineering process. Realizing that end-state requires more than better code generation: it requires assurance closure, meaning that the system can establish what must be true, determine and obtain appropriate evidence, judge the credibility of that evidence, preserve its validity through change, and use the resulting uncertainty to bound agent authority. Existing work already provides many of the necessary mechanisms across formal methods, testing, simulation, assurance cases, digital twins, and runtime assurance. We identify six residual gaps in making the surrounding assurance reasoning sufficiently machine-operable, propose a high-level architecture with six corresponding capabilities built on a shared semantic assurance layer, and formulate four research questions to turn that architecture into dependable, human-on-the-loop, AI-native RD.
[AI-18] Winning by Peeking: Unenforced Budgets and Test-Set Selection Inflate Short-Budget AutoML Comparisons
链接: https://arxiv.org/abs/2608.07303
作者: Guilin Zhang,Kai Zhao
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Statistics Theory (math.ST)
备注:
Abstract:Comparisons between AutoML systems at short time budgets – tens of seconds rather than hours – are common in tool READMEs and workshop papers, and they are easy to get wrong. We report a case study in which a simple AutoML engine, Orcetra, appeared to beat FLAML and AutoGluon on 513 OpenML datasets, winning 57.1% of them at a nominal 60-second budget and 78.4% of datasets against FLAML alone at 30 seconds. Both margins came from protocol defects that a results table cannot show. The search loop scored every candidate on the test split and reported the best, making the headline metric a maximum over dozens of noisy estimates while the baselines selected on training data and touched the test set once; and the budget was checked before launching a candidate but never enforced during one, so the system consumed a median of 120 s against a 60-second budget, 2.24x the wall-clock AutoGluon used. Re-running with selection moved to a validation split, the deadline enforced externally and every framework pinned to an equal share of the machine, Orcetra’s win rate on the re-run subset falls from 59.4% to 34.3% and no pairwise difference against either competitor remains significant. Recording both estimands inside a single search lets us attribute the collapse: the selection rule accounts for 4.8 percentage points and unequal compute for most of the rest. The same traces give the selection bias as a function of budget, measured rather than assumed: it grows with K but reaches only 0.27 accuracy points, about five times below the \sigma\sqrt2\ln K bound a marginal-standard-error argument predicts, because candidates scored on shared test rows cancel most of the noise. We close with a checklist for short-budget comparisons. Code, per-dataset results and the scripts that regenerate every number and figure in the paper are released with it.
[AI-19] FUSE: Feature-Wise Unified Specialization with Cross-Column Exchange for Mixed-Type Tabular Flow Matching
链接: https://arxiv.org/abs/2608.07294
作者: Suman Cha,Seongchan Lee,Dohyun Ko,Hyunjoong Kim
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 19 pages, 7 figures, 7 tables
Abstract:Generating mixed-type tabular data requires jointly modeling diverse feature distributions and their complex cross-column dependencies. Variational flow matching handles distinct endpoints via factorized distributions, yet leaves feature-specific processing and cross-column interactions implicit within a shared backbone. We introduce Feature-wise Unified Specialization with cross-column Exchange (FUSE) to explicitly separate these roles. FUSE applies separate adaptive mixture modules to numerical and categorical features, allowing each feature to combine shared specialized subnetworks, while joint attention preserves information exchange across all columns. We also characterize the excess population risk from restricted conditioning contexts and bound the continuous Wasserstein generation error by endpoint-prediction risk. Comprehensive experiments on eight tabular datasets demonstrate that FUSE achieves strong and consistent performance across distributional fidelity and downstream utility metrics.
[AI-20] OFD: Target-Oriented Feature Decoupling against Poisoning Attacks in Split Federated Learning
链接: https://arxiv.org/abs/2608.07274
作者: Yuhan Xie,Jingrong Huang,Chen Lyu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Split Federated Learning (SFL) facilitates privacy-preserving collaborative training with reduced client-side overhead. However, its split architecture introduces unique attack surfaces, rendering it vulnerable to diverse poisoning attacks. Most existing defenses fail to exploit the split paradigm, limiting their ability to detect and contain malicious behaviors at an early stage. To bridge this gap, we propose Target-Oriented Feature Decoupling (TOFD), a unified framework that jointly enables proactive detection and robust optimization against a wide range of poisoning attacks. TOFD operates in three stages: (1) Target Inference, which identifies potential attack targets by refining class-wise safe zones via class-specific Margin Perturbation (MP); (2) Sample Purification, which adaptively filters poisoned smashed data using thresholds calibrated through cross-class min-max normalization of MP; and (3) Decoupling Optimization, which leverages an adversarial guidance model to capture attack-induced patterns and decouple their influence during optimization, thereby suppressing residual adversarial effects. We provide theoretical guarantees for the convergence of TOFD. Extensive experiments on five datasets demonstrate that TOFD consistently outperforms state-of-the-art defenses under diverse attack scenarios, achieving superior robustness with low computational overhead suitable for practical deployment.
[AI-21] SCALE: Scientific Concept Aggregation via LLM s and Embeddings for Fine-Grained Taxonomy Extension
链接: https://arxiv.org/abs/2608.07254
作者: Daniele Raimondi,Feichi Lu,Oliver Grun,Mariia Eremina,Andrea Perlato
类目: Digital Libraries (cs.DL); Artificial Intelligence (cs.AI)
备注: 14 pages, 5 figures
Abstract:The increasing specialization of scientific research challenges existing classification systems, which provide effective representations of broad disciplines and research topics but often fail to capture the fine-grained conceptual structure of contemporary science. Author keywords offer greater specificity, but their fragmentation, redundancy, and terminological variability limit their use as stable units of knowledge organization. We introduce SCALE (Scientific Concept Aggregation via LLMs and Embeddings), a framework that extends the OpenAlex taxonomy with a new level of scientific Concepts below Topics. Rather than treating keywords as isolated descriptors, SCALE organizes semantically related terms into coherent and interpretable conceptual units and integrates them within the existing disciplinary hierarchy. The framework combines scientific text embeddings, large language models, and graph-based community detection to construct this additional layer at scale. The resulting taxonomy enables scientific literature to be read through an intermediate conceptual level between broad research topics and individual documents. This perspective provides a more detailed representation of how scientific knowledge is structured, specialized, and connected across disciplines. By transforming heterogeneous author terminology into reusable hierarchical units, SCALE offers a foundation for fine-grained scholarly classification, scientometric analysis, research monitoring, and future ontology development.
[AI-22] From probability to causality in probabilistic logic programming
链接: https://arxiv.org/abs/2608.07230
作者: Zora Wurm,Kilian Rückschloß,Felix Weitkämper
类目: Artificial Intelligence (cs.AI)
备注: Accepted and presented at IJCLR 2025
Abstract:Probabilistic logic programming is a formalism of statistical relational artificial intelligence that supports causal queries, including interventions from outside the system. When the structure of a probabilistic logic program is learned from data, however, only probabilistic information is used, and a single probability distribution may be compatible with several causal orders. This leads to ambiguity in interventional reasoning, raising the question of when the causal order is uniquely determined by the distribution. Exploiting the relationship between acyclic probabilistic logic programs and Bayesian networks, we derive conditions under which the probabilistic information encoded in a program determines a unique causal order. We also incorporate constraints arising from relational structure by taking into account prescribed sets of causal symmetries induced by the underlying relational vocabulary. The result is a method for verifying when a learned probabilistic logic program supports well-defined intervention semantics.
[AI-23] Beyond the Black Box: Interpretable Models of Human Randomisation Failures
链接: https://arxiv.org/abs/2608.07220
作者: Ngoc Linh Dao
类目: Artificial Intelligence (cs.AI)
备注: 6 pages, 2 figures
Abstract:Mixed strategy equilibrium predicts i.i.d play: past actions should not help predict future decisions. Human players, however, systematically depart from this benchmark, and in O’Neill’s zero sum card game, these departures can be predicted by black box sequence models such as LSTMs. This paper asks whether that predictive power can be achieved by transparent alternatives that also reveal the behavioural structure behind it. Using 84,060 decisions from 2,802 pairs, the analysis first benchmarks naive and behavioral models against interpretable machine learning and deep learning models, then evaluates the modified EWA specifications of prior work against these benchmarks and uses the LASSO diagnostics to motivate a further nested frequency tracking extension. The results show that repeat or avoid behavior, especially players’ management of their own recent action histories, accounts for most of the interpretable and strategically exploitable signal, while frequency tracking adds little out of sample.
[AI-24] oward a Causal Data Management Ecosystem for Decision Making and Agent ic AI
链接: https://arxiv.org/abs/2608.07214
作者: Dazhuo Qiu,Yingli Zhou,Amedeo Pachera,Angela Bonifati,Andrea Mauri
类目: Databases (cs.DB); Artificial Intelligence (cs.AI); Systems and Control (eess.SY)
备注: Accepted at ACM AI Leadership Summit 2026
Abstract:Modern AI is no longer a single model but an ecosystem: classical ML predictors, deep and multimodal models, large language models, and agents, each trained and tuned over different data sources and each producing outputs at scale that become inputs to the others. Operating such an ecosystem is fundamentally a data integration problem - the knowledge it depends on is fragmented across dozens of heterogeneous, independently governed sources that must be reconciled and continually maintained. Yet integration alone is not enough. The predictions these systems make are shaped by many interacting factors, and the events, decisions, and variables that drive an outcome are routinely entangled with the ones that merely accompany it; treated as a basis for action, such correlational signals invite confounded decisions. This becomes acute once agents act autonomously: to be trustworthy and reliable, an agent must anticipate the consequences of its actions, not merely extrapolate from what has co-occurred before. Causal reasoning is what closes this gap, distinguishing the drivers of an outcome from its correlates, and enabling prescriptive and counterfactual analysis over the ecosystem’s data. We therefore argue that the integrated ecosystem needs an explicit causal layer, and we propose to build it as a shared, persistent, queryable Causal World System (CWS).
[AI-25] Authoring and Management of Transparent Research Integrity Assessments of Randomised Clinical Trial Publications Using LLM -assisted Tools and Provenance Knowledge Graphs
链接: https://arxiv.org/abs/2608.07202
作者: Milan Markovic,Goutham Indukuri,Somayajulu Sripada,Colby J. Vorland,Jack Wilkinson,Clare Robertson,Mark Bolland,Andrew Grey,Miriam Brazzelli,Alison Avenell
类目: Artificial Intelligence (cs.AI)
备注: 16 pages
Abstract:Systematic reviews of Randomised Controlled Trials (RCTs) are routinely used as evidence for clinical care guidelines. Such evidence has to meet high research integrity standards to prevent low quality or false research outputs influencing the clinical care. However, assessing research integrity of published RCTs is a complex process requiring manual effort, and potentially resulting in diverse opinions of the human assessors. This paper describes INSPECT-AI, an LLM-based interactive tool that assists human reviewers with research integrity assessments of published RCTs based on the community approved INSPECT-SR framework, and the Research Integrity Provenance and Evidence ontology (RIPE-O) for documenting the provenance of the assessment process. In addition, we present the Research Integrity Provenance and Evidence knowledge graph (RIPE-KG), an initial set of 140 expert research integrity assessments of 95 RCT publications generated by INSPECT-AI and described using RIPE-O.
[AI-26] EMAS: Stabilizing Multi-Agent System Evolution through Evidence-Guided Revision
链接: https://arxiv.org/abs/2608.07196
作者: Chao Fei,Qingyi Si,Kaihua Liang,Yanghua Xiao,Panos Kalnis,Hongcheng Guo
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Many methods for automated multi-agent system design optimize prompts and topologies during an initial design stage and then deploy the resulting system unchanged on subsequent samples. Experience from these samples is rarely consolidated into reusable system updates, while accuracy-oriented designs may incur high token costs. We introduce EMAS (Evolving Multi-Agent System), which uses this experience to revise MAS topology and prompts without updating LLM parameters, either to improve accuracy or to reduce cost. EMAS converts traces into structured diagnoses that specify a revision operation and target. It generates a candidate revision only when the same diagnosis recurs across samples and applies it only if paired validation against the current MAS meets the corresponding acceptance criterion. Across four benchmarks and two LLMs, EMAS attains the highest task-weighted overall accuracy for both backbones and is best or tied in six of eight model–benchmark settings. Within two evolution epochs, EMAS achieves relative gains of 6.30% and 20.10% in task-weighted accuracy on Kimi-K2-6 and Qwen3.6-27B, respectively. On MBPP with Qwen3.6-27B, EMAS raises accuracy from 55.09% to 89.12% while reducing token use per task by 62.2%. These results show that EMAS can turn experience from new samples into reusable updates to MAS topology and prompts.
[AI-27] SetEasy: A Multi-Modal Classroom Engagement Assessment and Seating Optimization Framework
链接: https://arxiv.org/abs/2608.07188
作者: Zhihao Xie,Hongye Yang,Shien Liu
类目: Artificial Intelligence (cs.AI)
备注: 18 pages, 4 figures, 2 tables. Published in the Proceedings of ASCAAD 2025
Abstract:SetEasy optimizes classroom engagement in fixed seating grids. It fuses multimodal sensing (wristband physiology, 4K video, environmental data) and trains a v-Gage model grounded in a revised ISEQ. Each week, two-week engagement forecasts are mapped to a student-seat utility matrix, and CP-SAT generates seating plans under visual-access and social-dynamics constraints. In a four-week deployment (23 students, 331 classes), v-Gage converged across affective, behavioral, cognitive, and overall dimensions, cutting RMSE from 0.75 to 0.53. Optimization raised mean engagement from 0.30 to 0.70, with over two-thirds of seats reaching high engagement and back-row low-activity patterns markedly reduced. These results show that, without hardware changes, interpretable, data-driven seating strategies can substantially enhance engagement. The multimodal “assessment + optimization” paradigm offers a transferable, sustainable path to culturally responsive, differentiated spatial design amid global homogenization.
[AI-28] Momba: Network Modernization Improves Multi-Objective Reinforcement Learning
链接: https://arxiv.org/abs/2608.07180
作者: Adam Štafa,Santeri Heiskanen,Petr Novotný,Joni Pajarinen
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 21 pages, 10 figures; Accepted to RLC 2026
Abstract:Recent advances in deep reinforcement learning (RL) have shown that improving neural network architectures can yield substantial gains in sample efficiency and asymptotic performance without altering the underlying algorithms. In contrast, work on multi-objective reinforcement learning (MORL), which aims to discover a set of policies that balance trade-offs among conflicting objectives, has predominantly focused on algorithmic innovations, leaving the area of architectures underexplored. While the optimal policies and value functions can differ significantly depending on the trade-offs, MORL algorithms commonly represent them with simple feedforward networks conditioned on the trade-off. This raises the question of whether the performance of the algorithms could be improved with more expressive function approximators. In this paper, we integrate recent advances in neural network design: (i) observation and feature normalization, (ii) weight normalization, and (iii) modeling of distributional returns with an entropy-regularized MORL algorithm. The empirical results across standard continuous control benchmarks demonstrate that these changes substantially improve the quality of the produced solution sets without requiring major changes to the underlying algorithm.
[AI-29] Agent Memory Distillation: Empowering Small LLM Agents with Hierarchical Teacher Memory
链接: https://arxiv.org/abs/2608.07169
作者: Taeil Kim,Kangsan Kim,Sung Ju Hwang
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Under review
Abstract:Memory systems have shown promise for improving agent performance, but their potential remains largely unexplored for small language models, which struggle to generate sufficient successful trajectories on their own. We propose Agent Memory Distillation (AMD), a training-free framework that transfers structured knowledge from a large teacher agent to a small student agent through hierarchical memory. AMD constructs three complementary memory types from successful teacher trajectories: Workflow memory encodes task-level strategies, Subtask memory provides concrete behavioral examples at an intermediate granularity, and Function memory captures per-function calling conventions and common pitfalls. Workflow and Subtask memories are injected proactively at the start of each task, while Function memory is retrieved reactively upon tool-calling errors. We evaluate AMD on three tool-use benchmarks using four student models (4B-8B parameters) with GPT-5-mini as the teacher, achieving average accuracy gains of 27.2%p, 11.2%p, and 3.4%p on AppWorld, BFCL V3, and ToolSandbox, while consistently outperforming existing memory-based baselines. Further analysis shows that Subtask memory contributes the largest gains, teacher effectiveness depends on both teacher capability and student compatibility, and 4B-sized students benefit most from AMD.
[AI-30] NiyamAI - An Intent-Bound AI Agent with Cryptographically Verifiable Guardrails using Zero-Knowledge Proofs
链接: https://arxiv.org/abs/2608.07167
作者: Aditya Katkar,Om Karkele,Kartik Mandhane,Manisha More,Yash Kashid
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Giving an AI agent the ability to send emails, query databases, or execute commands is useful–until the agent is tricked into doing something it shouldn’t. Prompt injection, hallucinated reasoning, and unsafe tool calls form the primary attack surface for autonomous LLM agents. Existing defenses rely on software checks like system prompts or policy filters running on the same machine the attacker targets, offering no verifiable proof of execution. We introduce Niyam-AI, a framework that makes safety enforcement provable. At session start, permitted tools and constraints are locked into an Intent Contract committed via SHA-256. Every tool call is intercepted and validated by an isolated Judge model; upon passing, a zk-SNARK proof is generated via EZKL. The tool executes only after proof verification, allowing third parties to confirm enforcement without accessing Judge model weights. Evaluating Niyam-AI on 2,000 real-world scenarios from Agent-SafetyBench against NeMo Guardrails, Meta’s Llama Prompt Guard 2, and OpenAI’s GPT-OSS-Safeguard using 5-fold stratified cross-validation yields an F1 score of 88.5% with a 1.1% false-positive rate (bootstrap 95% CI: [85.19%, 91.88%], N=1000). McNemar’s exact paired test confirms significant improvement: Niyam-AI wins 390 discordant scenarios against NeMo (vs 20 losses), 115 against Prompt Guard 2 (vs 13), and 384 against GPT-OSS-Safeguard (vs 19) with p 0.0001 in all cases. Proof generation adds 2260.6 +/- 218.4 ms per approved action, while verification takes 53.1 +/- 11.8 ms. Niyam-AI provides a guardrail that is both highly accurate and mathematically verifiable–though this reflects a classifier adapted to Agent-SafetyBench evaluated against zero-shot baselines, a distinction discussed in Section IV.C. Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2608.07167 [cs.AI] (or arXiv:2608.07167v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2608.07167 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-31] Fluid-DiT: Graph-Free Diffusion Transformers for Fluid Flow Simulations Learning
链接: https://arxiv.org/abs/2608.07161
作者: Shentong Mo,Guolin Ke
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE)
备注:
Abstract:Simulating complex fluid flows requires capturing full equilibrium distributions rather than just mean trajectories, yet high-fidelity solvers remain computationally prohibitive. Recent advances, such as Diffusion Graph Networks (DGNs), have combined diffusion models with graph neural networks to sample equilibrium states directly from unstructured meshes, enabling distributional accuracy even from short simulations. However, graph-based diffusion approaches suffer from hand-crafted architectural constraints, limited receptive fields in message passing, and costly multi-scale designs, which restrict scalability to larger and more complex domains. We propose Fluid-DiT, a Graph-Free Diffusion Transformer that replaces graph message passing with attention-based denoising, eliminating explicit graph design while preserving the ability to model distributions of chaotic flows. Our framework introduces a latent-space formulation that disentangles geometric fidelity from distributional learning, reducing high-frequency artifacts and accelerating sampling. By leveraging the transformer’s global receptive field, Fluid-DiT naturally captures both local flow structures and long-range correlations without requiring hierarchical graph coarsening. On canonical benchmarks including laminar cylinder wakes, ellipse-flow systems, and turbulent 3D wing experiments, Fluid-DiT consistently outperforms graph-based diffusion baselines in both sample quality and distributional accuracy, achieving higher R^2 correlations and lower Wasserstein distances. Moreover, it generalizes robustly from short, incomplete trajectories to unseen Reynolds numbers and geometries, demonstrating strong scalability.
[AI-32] Representation Handoffs for OpenArm-Based Laboratory Mobile Manipulation
链接: https://arxiv.org/abs/2608.07154
作者: Yang Shen,Chonghao Cheng,Ziyi Zhao,Jialuo Zhu,Zhenyi Yi,Qi Zhao,Jian Yang,Yuhui Shi,Chin-Teng Lin
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: Robotics: Science and Systems (RSS) Workshop 2026
Abstract:Open-source robotics and foundation models have lowered the barrier to embodied AI, yet language-guided laboratory automation still requires reliable alignment from instructions and observations to safe actions. This field report presents an OpenArm-based mobile manipulation prototype for laboratory-style tasks, built by integrating dual OpenArm manipulators with a mobile base, vertical slide, RGB-D sensing, lidar-based mapping, ROS2/MoveIt execution, and profile-defined skill interfaces. The system is organized around representation handoffs: natural language requests are constrained into registered skill calls, sensor observations are grounded into maps and object poses, object priors provide role and skill constraints, and runtime bindings compile validated skills into executable motion goals. We use dry-run traces and startup checks to evaluate this integration path, showing how the prototype exposes missing calibration, incomplete object assets, and unfinished real-scene visual grounding as explicit deployment blockers. These intermediate representations serve as practical debugging interfaces for integrating language, perception, planning, and robot safety in embodied systems.
[AI-33] Interpretable reinforcement learning with decision-tree pruning
链接: https://arxiv.org/abs/2608.07151
作者: Mark Leon Ringer,Michel Tokic
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Reinforcement learning policies are difficult to inspect, but interpreting them is a prerequisite for trustworthiness. Converting a trained policy into explicit decision-tree rules improves transparency and the resulting artifacts often remain too complex for human understanding. We present a pruning process that simplifies such rule-based policies while preserving task performance and making edits to the policy auditable. The process defines a small set of structural and usage-aware operators and evaluates candidate edits by re-executing the policy to measure return and interpretability proxies. This exposes an transformation process from complex to compact policy structures. We investigate this approach on classic control and MuJoCo benchmarks, where pruning traces reveal consistent interpretability improvements while maintaining high performance.
[AI-34] A MARL Centered Reference Architecture for Large Language Model Augmentation in Smart Manufacturing
链接: https://arxiv.org/abs/2608.07148
作者: Fouad Bahrpeyma,Dirk Reichelt
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Modern manufacturing imposes six coupled demands on adaptive control: local decisions with global consequences, partial observability, nonstationarity, reflex speed response with long horizon effects, delayed and diffuse outcomes, and dynamics that resist explicit modeling. Cooperative multiagent reinforcement learning (MARL), posed as a Dec-POMDP under centralized training with decentralized execution, is a particularly natural formalism for these demands. This paper adopts a MARL centered scope and asks where large language models (LLMs) should augment, interface with, train, or, in the strongest competitive case, replace that coordination core. A taxonomy organizes the literature through four LLM attachment points: policy, reward design, communication between agents, and hierarchical planning. A conditional capability profile separates native mechanism, reported performance, formal guarantee, and engineering maturity, and a deployment readiness analysis identifies the evidence behind each role. These stages yield the principal contribution: a three layer MARL centered reference architecture, grounded in evidence, for semantic reasoning, adaptive cooperative control, and independently assured execution. The LLM-Augmented Dec-POMDP is a descriptive comparative notation for that architecture, recording four attachment choices without introducing a new decision process class or algorithm. Under the reviewed evidence, conventional MARL is better suited to frequent, structured, decentralized coordination after task specific training, whereas LLM components are promising for semantic interpretation, reward drafting, human interaction, and slower supervisory planning. Current LLM only manufacturing controllers do not yet establish equivalence for strict real time, decentralized, safety critical control; this conclusion is bounded by the available evidence and does not assert impossibility.
[AI-35] DiDPO: Diff-in-Diff Policy Optimization for Coding Agent Training
链接: https://arxiv.org/abs/2608.07147
作者: Xucong Wang,Zhe Zhao,Liheng Yu,Di Wu,Xiaofeng Cao,Pengkun Wang
类目: Artificial Intelligence (cs.AI)
备注: 16 pages, 6 figures, work in progress
Abstract:Reinforcement learning with Verifiable Reward (RLVR) has emerged as a powerful paradigm for training coding agents, where the execution feedback from compilation and tests provides objective verification. However, unlike agent tasks, coding agents face a unique and finer-grained credit assignment challenge: at each step, coding actions simultaneously pack varying changes into different regions of a code version, which makes the contribution of independent change indistinguishable. Existing RLVR methods mostly leverage the outcome reward or step-level reward, which fails to dive into a code diff and makes unique properties of coding actions invisible to training. In this paper, we propose Diff-in-Diff Policy Optimization (DiDPO), a critic-free RL method that constructs fine-grained credit units directly from the structure of code diffs. DiDPO organizes multi-turn coding interactions into multiple thought–action steps and discovers code diffs across sampled trajectories. It then selects anchors by aggregating highly similar sub-diffs split from each whole diff by our ``groupability score’', which provides the splitting schema that optimally balances the semantic scope of anchors and the group mass they may form. Finally these anchors form advantage groups and project the diff-level advantage back to individual response tokens. Experiments on long-horizon coding and reasoning benchmarks show that DiDPO significantly outperforms strong agentic RL baselines. On Qwen2.5-7B-Coder, DiDPO exceeds comparable methods by over 10% and narrows the gap with far larger models, offering a principled framework for fine-grained credit assignment in coding agent training. We also open-source verl-code, an agentic rl codebase that supports various RL methods and coding benchmarks.
[AI-36] How Much Then Where: Credit-Conserving Action-to-Token Allocation for Multi-Turn Agent Reinforcement Learning
链接: https://arxiv.org/abs/2608.07118
作者: Lichao Ma,Yang Sun,Shuaitao Zhao,Yangyi Fang,Cong Qin,Xiaoliang Fu,Yuhang Tian,Yuchen Wei,Junbo Zhu,Yang Wei,Lu Pan,Jiaye Lin
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Credit assignment in multi-turn agent reinforcement learning operates at two levels: assigning trajectory-level credit to actions and distributing each action’s credit across its tokens. In this paper, we introduce FACTOR, which separates these decisions. FACTOR uses checkpoint-calibrated TD residuals to assign per-action credits that telescope to the trajectory advantage, and feedback-conditioned teacher-student likelihood gaps to allocate each credit across the realized action tokens. Per-action normalization preserves the action-average coefficient and prevents token-level sign flips. We pair this construction with an action-mean reduction, removing the implicit dependence of an action’s scalar surrogate weight on its token length. At the behavior policy and before clipping, each action’s inner action-mean surrogate equals its TD credit. FACTOR consistently improves over competitive baselines across ALFWorld, WebShop, and ScienceWorld, with every environment-seed comparison favoring FACTOR and the largest gains emerging on the longest-horizon environment. The same hyperparameters transfer without retuning to a larger backbone and to a different model family. Ablations identify TD action credit as the dominant driver of the improvement, with hindsight token allocation contributing complementary gains.
[AI-37] MemWM: Memory-Augmented Text-Based World Model
链接: https://arxiv.org/abs/2608.07107
作者: Yujun Wang,Tao Zhang,Jinhe Bi,Aniri,Wenxuan Ye,Boliang Liu,Sikuan Yan,Shuning Wang,Xuebing Zhou,Sören Pirk,Hinrich Schütze,Yunpu Ma
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:World models are increasingly used to support planning in agents by predicting how environment states evolve in response to agent actions. Yet fluent next-state predictions can still omit task-critical facts, corrupt product attributes, or apply incorrect transition rules. To address such systematic prediction errors, we introduce MemWM, a memory-augmented text-based world model. MemWM uses world memory, a curated memory bank of transition rules, state caches, and hard-to-predict facts, to condition next-state imagination. We evaluate factual state preservation with Structured State Fidelity (SSF), which scores predicted states through benchmark-specific facts and fields. Compared with SFT, memory-augmented training improves SSF by up to 206.3%. In the full planning setting, we keep the policy model frozen and provide policy-side world skill: retrieved task-level skills and step-wise corrective guidance for action selection. Across ALFWorld, WebShop, and ScienceWorld, memory-augmented agents improve downstream success over an SFT-trained world-model agent, with up to a 65.4% relative gain. Sensitivity analyses further show that retrieved memory improves task success and efficiency under different memory and action-budget settings.
[AI-38] Beyond Isolation: Unlocking Reinforcement Learning Component Synergy for Sample-Efficient Continuous Control
链接: https://arxiv.org/abs/2608.07086
作者: Qi Zhao,Guozheng Ma,Yilun Kong,Lu Li,Haoyu Wang,Zilin Wang,Tiantian Zhang,Yuxing Wang,Jian Sha,Yongzhe Chang,Xueqian Wang,Dacheng Tao
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 27 pages including appendix, 10 figures, 12 tables
Abstract:Reinforcement learning systems are significantly more complex than other machine learning paradigms due to inherent properties, causing RL system design to jointly account for many tightly coupled factors. Despite advances in individual algorithmic components, their functional interdependencies remain underexplored: do they exhibit mutual synergy or counterproductive interference? To bridge this gap, we conduct a systematic investigation and find that the efficacy of different components exhibits significant task-dependency, and naively stacking state-of-the-art techniques does not necessarily yield performance gains; instead, it often triggers emergent challenges, such as compounded non-stationarity. Building upon these findings, we distill a suite of actionable insights into the principled coordination of these components. Guided by these insights, we propose ROSER, an RL framework that coordinates three critical dimensions: Model-based Representation, Optimization Stability, and Experience Replay. Across diverse continuous-control benchmarks, ROSER consistently outperforms vanilla baselines and achieves 17.60% gains over naive stack. Our findings underscore the necessity of a holistic perspective in RL system design and paves the way for developing sample-efficient agents.
[AI-39] LifelongCrossNav: Persistent 3D Semantic Memory for Cross-Floor Multi-Object Navigation
链接: https://arxiv.org/abs/2608.07079
作者: Zehui Li,Zihao Sun,Jiawei Xu,Zheqi He,Xiaoqiang Zhang,Jing-Shu Zheng,Lu Liu,Dahui Gao,Xiuwan Chen
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Object-goal navigation has made substantial progress in semantic perception and exploration, yet persistent memory for multi-object navigation and cross-floor navigation are still commonly addressed separately. We present LifelongCrossNav, a framework for sequential multi-object ObjectNav in unknown multi-floor indoor environments. Within each episode, the agent receives an ordered sequence of object-goal queries while continuously maintaining a shared sparse 3D semantic voxel memory. This memory incrementally accumulates geometric structure, traversability states, and vision-language features, allowing subsequent object-goal queries to retrieve previously acquired scene information without rebuilding the map. To support persistent search across floors, LifelongCrossNav combines support-aware 3D traversability mapping, stair-specific perception, and direction-aware stair traversal. A unified navigation policy coordinates same-floor frontier exploration, live and historical point-of-interest retrieval, stair navigation, and target-object search and approach. We further introduce HM3D-MFMON, a benchmark for sequential Multi-Floor Multi-Object Navigation built on HM3D scenes, including a dedicated subset in which completing the full sequence of object-goal subtasks requires at least one floor transition. Experimental results show that LifelongCrossNav consistently outperforms a representative planar persistent semantic-map baseline on HM3D-MFMON, demonstrating that persistent 3D semantic memory and cross-floor traversability modeling effectively support sequential multi-object navigation in multi-floor environments. Project page: this https URL.
[AI-40] Scalable High-Fidelity Macromolecular Docking for GPU-Accelerated Supercomputers
链接: https://arxiv.org/abs/2608.07078
作者: Xiangyu Meng,Peng Chen,Mingzhen Li,Jianmin Wang,Sen Wang,Guangming Tan,Weile Jia,Mohamed Wahib,Tao Luo,Xun Wang
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Artificial Intelligence (cs.AI)
备注: To be published in the International Conference for High Performance Computing, Networking, Storage, and Analysis(SC) 26
Abstract:Flexible macromolecular docking offers high-fidelity predictions of biomolecular interactions, but remains prohibitively expensive at scale. Among existing approaches, LightDock leverages Glowworm Swarm Optimization (GSO) for accuracy, yet suffers from limited parallelism, irregular computation, and severe load imbalance, preventing efficient execution on GPU supercomputers. We present SparkleDock, a scalable GSO-based docking framework enabling near-real-time flexible docking. We redesign GSO to expose massive fine-grained parallelism at the glowworm-agent level, and restructure the dominant energy scoring computation into a Tensor Core-compatible formulation, enabling efficient execution of irregular pairwise interactions through structured matrix operations. We further introduce a performance-model-driven scheduling for load balancing and out-of-core scaling across GPUs. SparkleDock achieves 9.7 \times and 18.9 \times speedups over LightDock on single A100 and H100 GPU, and delivers over two orders of magnitude acceleration at scale. On 512 GPUs, it reduces docking time from hours to seconds, enabling large-scale, high-fidelity virtual screening previously impractical with flexible docking.
[AI-41] ransformers Struggle to Use Their Emergent World Models: Revisiting the Tower of Hanoi and the Illusion of Thinking
链接: https://arxiv.org/abs/2608.07077
作者: Devin Pereira,Willem Zuidema
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:The Tower of Hanoi is a simple planning puzzle that in prior work has proven challenging for large reasoning models (LRMs). Current models solve the standard formulation of the puzzle, but still struggle with the flat-to-flat variant (where initial and goal states are not restricted to have all rings on a single peg). This paper presents an in-depth study of how both small, in-house Transformers and large, third-party LRMs solve this task. To understand the failures mechanistically, we first train small Transformers from scratch on precomputed solution traces. Using a variety of interpretability techniques, we show that these Transformers develop an emergent world model: a linearly decodable, geometrically faithful representation of the puzzle’s state space (the Sierpinski triangle), that is causally involved in solving the puzzles. Second, we return to the large LLMs and apply our techniques to two frontier reasoning models, Qwen3.6-27B and DeepSeek-R1-Distill-Qwen-32B, that attempt to solve the task through extended chain-of-thought. Surprisingly, we find that both models encode the Sierpinski world model near-perfectly at the end of the prompt, and yet fail at the majority of tasks when there are more than 3 rings. We locate the source of this failure in the decaying representation of the world model. We probe for the representation at different stages during planning, and establish causality by showing that performance can be improved by injecting the prompt-time representation at inference. The failure of the models is thus one of maintenance of the required representations, not their absence, and performance is at least partially recoverable. These results thus reframe the reported collapse in performance from prior work: current Large Reasoning Models build a world model, and then lose it.
[AI-42] MemOPD: On-Policy Distillation through Memory State Alignment for Long-Horizon Agents
链接: https://arxiv.org/abs/2608.07068
作者: Zhiyuan Liu,Tinghong Ye,Chenghao Liu,Yizhuo Li,Songfang Huang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Long-horizon agents accumulate growing contexts during interaction, impairing performance and stability. Compact memory mitigates this problem by compressing and rewriting the history retained between model invocations. Learning what to retain typically relies on proximal policy optimization (PPO) with final task rewards, but sparse rewards provide little guidance for individual memory updates. This limitation motivates on-policy distillation (OPD), which supplies dense teacher supervision on student rollouts. For such supervision to be valid, the teacher must evaluate each sampled action under the same state in which it was generated. However, the context rewriting performed during memory compression can break this alignment. When sampled responses are retained and re-encoded for later invocations, flattening the interaction into a persistent history may cause the teacher to score the action under a state that the student never visited during rollout. The action therefore remains on-policy by provenance, but not necessarily by state. We therefore propose Memory-Aligned On-Policy Distillation (MemOPD). MemOPD records the inputs and sampled outputs of each model invocation, restores its original token positions and causal visibility, and packs the reconstructed invocations for efficient teacher scoring. The teacher provides full-vocabulary supervision at the sampled action positions, while PPO preserves the final task objective. Experiments verify state alignment across several context updates and show that it improves F1 by 7.0% over persistent-history teacher scoring in a matched control. Overall, MemOPD-3B improves F1 over PPO by up to 416.2%, while packing yields up to a 1.63x speedup in actor computation during training. The code for this work is publicly available at: this https URL.
[AI-43] PTQ4SNN: Membrane-Aware Post-Training Quantization for Spiking Neural Networks
链接: https://arxiv.org/abs/2608.07066
作者: Hui Xie,Tong Shi,Haotong Qin,Aishan Liu,Xiaode Liu,Jinyang Guo
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Spiking neural networks (SNNs) enable sparse and event-driven computation, but their low-bit deployment remains incomplete because recurrent membrane states are commonly retained in floating point even after weight quantization. Quantizing these states is challenging because their distributions differ across channels and from the preceding weights, while small perturbations near the firing threshold may alter spike decisions and accumulate over time. We propose PTQ4SNN, a membrane-aware post-training quantization framework that jointly quantizes weights and recurrent membrane states using only a small calibration set. First, a channel-wise Unified Scale Bridge constrains the membrane scale as s_mem,c = s_w,c * 2^k_c, adapting to membrane distributions while enabling shift-compatible scale conversion. Second, Mixed-Precision Bit Allocation assigns 2/4/8-bit precision to membrane channels according to firing activity and quantization sensitivity under an average-bit budget. The framework operates on reusable projection-LIF pairs and supports both convolutional SNNs and spike-driven Transformers without backbone retraining. Experiments on static and event-based classification and semantic segmentation show that PTQ4SNN effectively preserves model accuracy under W4 quantization and approximately 4-bit membrane precision.
[AI-44] Soft Redaction of Image Provenance via Zero-Knowledge Proofs ECCV2026 ATC
链接: https://arxiv.org/abs/2608.07063
作者: Muhammad Awan,John Collomosse
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: To appear at ECCV 2026 workshop on Privacy Fairness Accountability and Transparency in Computer Vision (PFATCV)
Abstract:Content provenance standards, such as C2PA, are increasingly used to attach signed records of origin, editing history, and rights to digital images. However, provenance transparency can conflict with privacy – assertions that strengthen trust in an image may also reveal sensitive information about the creator or capture context. We propose soft redaction for image provenance: a mechanism that replaces sensitive provenance assertions with zero-knowledge proofs (ZKPs) of selected properties over hidden data. Our work focuses on distance proofs. We first show how location assertions can support proofs of proximity to a public reference point, using Chebyshev polynomial approximations within the ZKP proof circuit. We then extend the approach to L2 distance proofs over biometric embeddings, enabling privacy-preserving claims related to likeness to help enforce personality rights with images. Finally, we apply the same distance-proof construction to perceptual hashes (visual fingerprints), supporting an anti-spoofing use case in watermark-based recovery of stripped provenance metadata. Our results demonstrate that ZKPs over image provenance can provide practical soft-redaction capabilities, compatible with C2PA, that may be constructed in seconds and verified in milliseconds.
[AI-45] BONSAI: Evolvability-Guided Tree Search over Skills
链接: https://arxiv.org/abs/2608.07056
作者: Yash Priya Shastri,Anand Eswaran,Adnan Qidwai,Pankaj Thorat,Sachin Joshi
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:A skill is a naturallanguage document that steers a frozen agent whose weights cannot be updated so any capability the agent lacks must be supplied in prose Optimising a skill is therefore optimising text against a score and the standard recipe which keeps any edit that raises a heldout score is blind in a specific way a single score cannot tell a document perched on a narrow overfit spike from one resting on a broad plateau even though only the second can still be improved We introduce BONSAI a novel skilloptimisation framework that steers instead by evolvability the capacity of a region of documentspace to keep producing viable variation under further mutation a property biology treats as separate from present fitness BONSAI grows skills as a MonteCarlo search tree in which every child document is a mutation of its parent and descends it under an upperconfidence selection rule whose exploitation term blends a skills own fitness with the fitness of its mutational neighbourhood Because every child is a mutation the mean score recorded beneath a node estimates that neighbourhoods evolvability at no extra cost so the rule concentrates budget on regions that keep improving while its exploration term keeps a currently weak branch in contention BONSAI ships the single bestscoring document it finds at no cost beyond the acceptifbetter loop it replaces With a frozen 30B agent and averaged over three benchmarks BONSAI lifts heldout accuracy over the skillfree agent by 2313 points and improves on two budgetmatched baselines GEPA and SkillOpt by 387 and 397 points respectively
[AI-46] Unsupervised Adaptation of PDE Foundation Models
链接: https://arxiv.org/abs/2608.07053
作者: Ziye Song,Zhao Wei,Xin Yu,Ivor Tsang,Yueming Lyu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Pretrained partial differential equation (PDE) foundation models can generalize across different equations, but adapting them to unseen PDE systems typically requires dense solution data, which is often expensive or unavailable. To address this limitation, we propose an unsupervised PDE-based finetuning framework that eliminates the need for ground-truth solutions. We first pretrain a neighborhood attention Transformer on diverse time-dependent PDEs spanning varying spatial scales, yielding transferable representations across heterogeneous equations. In the adaptation stage, we construct a physics-based objective using the PDE residual and boundary conditions, and finetune the model on unseen equations via low-rank adaptation (LoRA). To address the uneven learning across physical quantities in standard LoRA, we introduce NSLoRA, a Newton-Schulz orthogonalized variant that rebalances adaptation. Our method achieves performance comparable to supervised LoRA finetuning without requiring any ground-truth solutions, while consistently outperforming competitive neural operator baselines and recent PDE foundation models across heterogeneous PDE benchmarks spanning multiple spatial dimensions.
[AI-47] Not All Problems Are Best Modeled as MILP: A DSL-Centric Framework for Flexible and Accurate Optimization Modeling
链接: https://arxiv.org/abs/2608.07040
作者: Shaofeng Zhang,Hongyuan Su,Qingwen Peng,Zefang Zong,Shengcai Liu,Ke Tang,Yong Li
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Solving combinatorial optimization problems (COPs) requires not only efficient algorithms but also carefully crafted formulations. While recent works have leveraged LLMs to automate optimization modeling, current frameworks predominantly rely on a rigid mixed-integer linear programming (MILP) paradigm. In this paper, we argue that not all problems are best modeled as MILP, as forcing complex domains into linear constraints can induce prohibitive modeling complexity and severely restrict solver flexibility. To address this, we propose OptiDSL, a framework that shifts the focus from rigid MILP formulations to domain-specific language (DSL) representations. By utilizing LLMs to map natural language onto standardized, domain-accepted structures, OptiDSL decouples problem formulation from execution. This paradigm enables seamless integration with a diverse library of specialized solvers, ranging from traditional heuristics to modern learning-based methods. Experimental results on the comprehensive benchmark of 44 COP types show that OptiDSL significantly surpasses MILP-based pipelines, yielding a 51.66% gain in formulation accuracy and a 91.71% decrease in modeling time. Notably, it also outperforms MILP-based pipelines on the existing benchmark, achieving a 23.09% higher formulation accuracy. Our code is available at this https URL.
[AI-48] Beyond Text Matching: Towards Reference-Free Evaluation for Human-Oriented Binary Reverse Engineering
链接: https://arxiv.org/abs/2608.07038
作者: Xiuwei Shang,Li Hu,Xiao Jiang,Jieke Shi,Junda He,Zhou Yang,Shaoyin Cheng,Guoqiang Chen,Weiming Zhang,David Lo
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注: Accepted by the 41st IEEE/ACM International Conference on Automated Software Engineering (ASE 2026)
Abstract:Human-Oriented Binary Reverse Engineering (HOBRE) aims to transform decompiled pseudocode into a more human-friendly representation, thereby reducing the cognitive burden of reverse analysis and improving efficiency. However, reliably evaluating HOBRE outputs remains a fundamental challenge: human evaluation is costly, time-consuming, and difficult to scale, while existing automated metrics either require executable test cases and runtime environments that are often unavailable for real-world binaries, or rely on high-quality source code references that are typically inaccessible and fail to capture semantically equivalent but lexically diverse outputs. Although LLM-as-a-Judge paradigm is naturally well-suited to HOBRE evaluation, its effectiveness remains underexplored. This paper presents the first systematic investigation of the LLM-as-a-Judge paradigm for HOBRE across three representative tasks: function name recovery, binary code summarization, and decompilation optimization. We introduce BinJudgeBench, the first expert-annotated, reference-free evaluation benchmark based on multi-dimensional human judgment, where LLM-as-a-Judge achieves an average correlation of 63.20% with human judgment, outperforming traditional automated metrics at 35.04%. By analyzing judge configurations across backbone LLMs, prompting strategies, and decoding temperatures, we find that no ``one-size-fits-all’’ configuration exists, as the optimal setup varies across tasks and individual samples. To address this, we propose BinJudge, which employs a lightweight routing mechanism to adaptively select the optimal judge configuration for each task and sample. BinJudge improves correlation with human experts by 4.5%-24.7% and reduces API cost to 0.06 \times -0.84 \times of that of static best configurations, providing a scalable, cost-effective, and high-fidelity automated evaluation scheme for HOBRE. Comments: Accepted by the 41st IEEE/ACM International Conference on Automated Software Engineering (ASE 2026) Subjects: Software Engineering (cs.SE); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR) Cite as: arXiv:2608.07038 [cs.SE] (or arXiv:2608.07038v1 [cs.SE] for this version) https://doi.org/10.48550/arXiv.2608.07038 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-49] Accounting Graph Transformer for Short-History Multi-KPI Forecasting in Small Businesses
链接: https://arxiv.org/abs/2608.07037
作者: Shrutendra Harsola,Vignesh Subrahmaniam
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Small businesses often have only 12-24 months of accounting history, yet planning and risk workflows require coordinated forecasts across financial statements. We study joint 12-month forecasting of 13 income-statement, balance-sheet, cash-flow, and working-capital key performance indicators (KPIs) from 71 monthly ledger series. We introduce the Accounting Graph Transformer (AGT), which represents each ledger series as a masked token, exchanges information through typed attention on a fixed accounting-relation graph, pools target-specific context, and fuses it with a gated three-month recency path. Across 11,993 forecast origins from 1,060 unseen companies, AGT achieves sample-weighted KPI-macro mean absolute error (MAE) 0.6990 \pm 0.0013 over three independent seeds, compared with 0.7378 \pm 0.0014 for the strongest baseline, LightGBM. At the pre-specified seed 42, a paired company-clustered bootstrap gives a LightGBM-minus-AGT difference of 0.0395 with 95% confidence interval (CI) [0.0350,0.0439] . AGT is best on all 13 KPIs against LightGBM, TimeMixer, and SOFTS in the matched seed-42 comparison, while final-architecture ablations show that relational attention, accounting topology, and the recency path each improve validation and test accuracy. On 7,094 additional unseen companies with origins sampled from January-May 2025, AGT obtains 0.7548 MAE versus 0.7694 for SOFTS. A single 5.3M-parameter model produces 156 aligned forecasts without company-specific fitting, providing one forecasting layer for integrated planning, liquidity, and working-capital analysis.
[AI-50] ZIPBrain: Can EEG Foundation Models Be Faster Locally Deployable but Accurate?
链接: https://arxiv.org/abs/2608.07033
作者: Lingwei Li,Yirong Kan,Peng Chen,Xu Cao,Zheng Chen,Yasuhiko Nakashima
类目: Artificial Intelligence (cs.AI)
备注: 7 pages(14 pages including appendix), 5 figures
Abstract:This work investigates whether Electroencephalograph (EEG) foundation models (EFMs) can be made faster and locally deployable without sacrificing accuracy. EEG foundation models are a major trend, offering strong general-purpose representations. However, their computational burden grows quadratically with input length, hindering deployment on resource-constrained scenario, particularly for real-time clinical monitoring. EEG’s low SNR further suggests many of these tokens are redundant and compressible with little accuracy cost. We propose ZIPBrain, a novel redundancy-aware EEG token pooling module that leverages this low-SNR characteristic to reduce token count. Given a token sequence, ZIPBrain partitions tokens into redundant and unique groups, then merges each redundant token with its most similar counterpart in the unique group. Furthermore, ZIPBrain serves as a training-free, plug-and-play module that seamlessly integrates into standard Transformer encoders with negligible computational overhead. Extensive experiments across multiple EEG foundation models show ZIPBrain’s strong versatility, achieving 1.3%-10.5% average improvement over baselines, while reducing wall-clock inference time by 32.7% (up to 41.8% with CUDA Graph) compared to the original EEG foundation models.
[AI-51] ReQuant: Fixed-Grid Discrete Refinement for Post-Training Quantization
链接: https://arxiv.org/abs/2608.07019
作者: Yongge Ma,Guoan Wang,Feiyu Wang,Yaoming Li,Qian Zhang,Zihan Yan,Yinjun Han,Tong Yang
类目: Artificial Intelligence (cs.AI)
备注: 10 pages, 3 figures, 16 tables
Abstract:Post-training quantization (PTQ) is widely used to reduce the memory and computational cost of large language models. Existing PTQ methods typically obtain an initial quantized model through heuristic rules or greedy optimization, and once quantization is completed the resulting integer assignments are usually treated as final. This observation motivates a complementary optimization stage within PTQ that keeps quantized weights improvable after an executable quantized model has been produced, while preserving the quantized format. We introduce ReQuant, a backpropagation-free fixed-grid refinement procedure for this stage. Agnostic to the PTQ initializer, ReQuant takes an existing quantized model as a feasible starting point and iteratively revisits its discrete weight assignments on the fixed quantization grid. Accepted updates strictly reduce the mean squared reconstruction error and remain on the original grid. In this way, ReQuant turns the initially fixed PTQ output into an iteratively optimizable discrete solution and serves as a plug-and-play post-processing stage for existing PTQ pipelines. Experiments across diverse model families, bit-widths, and downstream tasks show that ReQuant consistently improves quantized models from heterogeneous PTQ initializers, with especially large gains on simple initializers and lower bit-widths. Notably, ReQuant can refine a simple round-to-nearest initialization across multiple sweeps until it approaches or surpasses GPTAQ under the same quantization format. These results establish ReQuant as a practical complementary stage for further improving existing PTQ pipelines.
[AI-52] FedLBW: A Loss-Based Weighting Strategy for Federated Learning on Non-IID Data in Wireless Networks
链接: https://arxiv.org/abs/2608.07007
作者: Majid Kundroo,Tinku Singh,Taehong Kim
类目: Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC); Emerging Technologies (cs.ET); Machine Learning (cs.LG)
备注:
Abstract:Federated Learning (FL) enables collaborative machine learning (ML) across distributed clients while preserving privacy. However, efficient model convergence in FL remains challenging, especially in wireless networks where non-independent and identically distributed (non-IID) data and frequent client dropouts are common. Traditional FL algorithms, such as FedAvg, rely solely on dataset size to weight client updates. This introduces biases towards clients with larger datasets and makes the process sensitive to non-IID data, outliers, and client dropouts. To address these challenges, we propose Federated Learning with Loss-Based Weighting (FedLBW), a novel aggregation method that assigns each client’s update a weight proportional to the inverse of its validation loss, computed using a small proxy dataset on the server, rather than its dataset size. This ensures that lower-loss models exert greater influence during aggregation, prioritizing the most reliable updates and boosting overall performance. Through extensive experiments across multiple datasets, including FashionMNIST (CNN), CIFAR-10 (ResNet-18), and CIFAR-100 (ResNet-34), we demonstrate that FedLBW achieves higher accuracy and faster convergence compared to baseline algorithms such as FedAvg, FedAvgM, FedProx, FedNova, FedLAW and FedDkw, with notable improvements of up to 7.6 % higher accuracy on CIFAR-10 in extreme non-IID cases. Moreover, FedLBW showcases exceptional resilience to increasing dropout probabilities, consistently maintaining significantly higher accuracy even in challenging conditions. These results establish FedLBW as an effective and resilient solution for FL in wireless network environments, offering marked improvements in model accuracy, convergence speed, and robustness to non-IID data and client dropouts.
[AI-53] Decoupling Intention from Trajectory: A Representational Deduction Framework for World Action Models
链接: https://arxiv.org/abs/2608.06994
作者: Xiangkai Ma,Yue Ma,Junjie Wang,Sheng Xu,Mingyang Li,Han Zhang,Yuzheng Zhuang,Wenzhong Li,Zhihao Yuan
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:World Action Models (WAMs) aim to construct a unified architecture capable of understanding world state evolution and guiding to generative motion planning. However, existing visual branches focus on predicting static visual observation, rather than reflecting potential transition information that captures the evolution of world states under motion interactions. This leads to representational entanglement between high-level physical condition evolution and low-level action trajectory generation within the Action Model, creating a structural bottleneck while weakening the predictive capability of world evolution modeling for action generation. We propose PILOT (Physical Inference for Latent Optimized Trajectories), whose core Representational Deduction (RD) bridges this gap by integrating motion thought-of-chain (CoT) guidance as a native model capability. Specifically, RD aims to encourage the action branch to explicitly model potential state transition tokens, which are retained as CoT in the reasoning space to guide fine-grained motion trajectory. Experiments demonstrate that RD not only significantly improves the success rate and generalization ability of WAMs in complex robotic manipulation tasks but also enhances the model’s physical interpretability by decoupling high-level motion semantics from low-level trajectory details. Furthermore, the abundant state transition supervision signals introduced by RD effectively alleviate the sparse supervision in action generation, enabling it to serve as an efficient few-shot real-robot fine-tuning strategy and demonstrating superior scalability for migration to mainstream WAM architectures.
[AI-54] Beyond Foundation Models: Dimension-Aware Neural Architecture Search with Small-Data Representation Models for Cryocooler Lifetime Prediction
链接: https://arxiv.org/abs/2608.06993
作者: Gregor Molan(1),Grafika Jati(1), (4),Francesco Barchi(4),Andrea Acquaviva(4),Aljaž Osterman(3),Martin Molan(2), (1), (4) ((1) Comtrade 360 d.o.o., Letališka cesta 29b, Ljubljana, 1000, Slovenia, (2) Comtrade AI GmbH, Grafenauweg 8, Zug, 6300, Switzerland, (3) LE-Tehnika d.o.o., Šuceva 27, Kranj, 4000, Slovenia, (4) Alma Mater Studiorum - Universita di Bologna, Department of Electrical, Electronic, and Information Engineering (DEI), Viale del Risorgimento 2, Bologna, 40136, Italy)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 48 pages
Abstract:Large-scale pretrained time-series models achieve strong results through large-scale pretraining and task-agnostic representation learning, but they rely on abundant, diverse data that industrial and scientific domains often lack. We therefore propose the FSD-RM (Family of Small-Data Representation Models) paradigm as a practical alternative for limited, domain-specific telemetry. Rather than relying on large-scale pretraining, we focus on capacity-controlled representation learning using established encoder architectures (CNN1D, LSTM, GRU, Transformer), selected for their suitability in small-data settings and interpretability. These encoders are trained unsupervised on multivariate telemetry data and integrated into a two-stage pipeline for downstream lifetime prediction. To systematically examine architectural trade-offs under data constraints, we employ \textbfdimension-aware neural architecture search (NAS) to jointly optimize model capacity and input dimensionality. Experiments on cryocooler telemetry show that the proposed approach achieves competitive predictive performance while reducing training cost and model complexity. The contribution lies in combining established representation learning techniques within a coherent, NAS-driven framework tailored to small-data regimes, with explicitly defined parameter settings and design choices. The results indicate that effective representation learning can be achieved without large-scale pretraining when appropriate inductive bias and capacity control are applied. Comments: 48 pages Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI) MSC classes: 68T05, 68T10, 62M10, 68W50, 93C41 Cite as: arXiv:2608.06993 [cs.LG] (or arXiv:2608.06993v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.06993 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Journalreference: Reliability Engineering and System Safety 277 (2027) 113105 Related DOI: https://doi.org/10.1016/j.ress.2026.113105 Focus to learn more DOI(s) linking to related resources
[AI-55] Density-aware Hierarchical Clustering Based on Element-Categorized Connection Subgraphs
链接: https://arxiv.org/abs/2608.06990
作者: Yuning Yu,José Rodríguez-Piñeiro,Xuefeng Yin,Bin Feng
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Clustering is a fundamental data mining technique for pattern recognition through unsupervised learning. Among various clustering methods, hierarchical clustering, density-based clustering, and graph clustering stand out as representative approaches. For hierarchical clustering, it can be categorized into agglomerative and divisive modes to construct clusters in a recursive manner. The key aspect of both modes is the calculation of inter-cluster similarity, which determines whether to merge the sub-clusters into one cluster or divide a current cluster into sub-clusters. Traditionally, the similarity is derived from pairwise distances, often overlooking density variations and structural connectivity in graphs. To address this, we propose a density-aware hierarchical clustering method based on element-categorized connection subgraphs (DHC-ECS), which effectively integrates the hierarchical clustering, density-based clustering, and graph clustering. Particularly, a novel inter-cluster similarity metric is introduced that considers not only distances but also the element categorization in the KNN connection subgraphs, kernel density estimation, and local connectivity within sub-clusters. Extensive evaluations on heterogeneous benchmark datasets demonstrate that DHC-ECS exhibits superior overall performance in terms of clustering accuracy and parameter robustness compared with the baseline methods (including AChameleon, RNN-DBSCAN, McDPC, and G-RMS). The work indicates the great potential of the proposed clustering algorithm for low-dimensional datasets by leveraging local density and graph-structured connectivity (i.e., the duality of vertices and edges), as well as the possibility to determine an intrinsic threshold, reducing the reliance on manual parameter tuning.
[AI-56] HarnessSafe: Evaluating Safety Across Persistent Carriers in Agent Harnesses
链接: https://arxiv.org/abs/2608.06984
作者: Xiao Zhang,Yusheng Wang,Yuhao Fei,Dongyuan Li,Zian Liang,Liuyu Xiang,Hongxun Gu,Zhaofeng He
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: 21 pages, 3 figures. Preprint
Abstract:Modern agent harnesses persist state across tasks and sessions through persistent carriers like memory, skills, tools, and shared artifacts. However, this capability creates delayed safety risks: attacker-influenced content can cross system boundaries and later affect the execution of a benign request. Existing benchmarks typically focus on a few carriers or harnesses, while end-to-end attack-success rates reveal little about how risks propagate. To this end, we present HarnessSafe, a benchmark comprising 328 executable cases across seven persistent-carrier families and evaluated on most mainstream agent harnesses. Each case is specified as a Persistent-Risk Lifecycle that traces attacker influence from its initial entry, through persistence across carriers and system boundaries, to a later benign trigger and an observable violation. We further introduce a multi-stage, trace-based evaluation that uses observable execution evidence to determine how far each attack chain progresses and where it is stopped. Experiments show that containment is carrier-specific and strongly depends on the harness-model configuration. Both the harness and model backend substantially shape containment outcomes, while attack success rates cannot reflect distinct lifecycle progression patterns.
[AI-57] Finding Usable Weight Mechanisms with Tiled SVD
链接: https://arxiv.org/abs/2608.06969
作者: Ash Manvi,Samreena Tajreen
类目: Artificial Intelligence (cs.AI)
备注: 8 pages, 6 figures. Finding usable weight mechanisms via column-tiled SVD; pre-registered A/B/C eval on Gemma-2-2B (WikiText-2)
Abstract:The dominant approach to mechanistic interpretability trains proxy dictionaries such as sparse autoencoders and labels features from max-activating text. The best such atlases identify con- cepts, but that identity lives in the learned dictionary rather than in the network weights them- selves. We propose extracting mechanism mounts directly from linear sites by column-tiled SVD: each mount is a triple (v,u,\sigma) read as trigger, write, and strength. Identity is the weight rule. We evaluate mounts with a pre-registered suite judged on full-write energy lift rather than tile-local lift. On Gemma-2-2B with WikiText-2 (16,384-token subsample), all seven linear maps are scored: residual writes (this http URL, attn.o) receive full A/B/C with steer after post-sublayer RMSNorm and pass 52/52 site-layers; other maps receive A/B only (this http URL this http URL 26/26 each). Aggregate: 182/182 GO. We release library code, the corpus builder, the experiment entrypoint, and unit tests.
[AI-58] Learning in Deep Networks under Dales Constraint
链接: https://arxiv.org/abs/2608.06963
作者: Roy Abel,Shimon Ullman
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Biologically plausible learning models aim to explain how neural circuits can implement effective learning under the constraints of real neurons. Although significant progress has been made, a major remaining challenge is that existing models often allow neurons or synapses to represent mixed-sign values, both positive and negative, in violation of a basic aspect of cortical circuitry – Dale’s constraint: biological neurons are either excitatory or inhibitory, but not both, and synapses cannot change sign. In this work, we address this discrepancy by introducing a biologically motivated neural architecture in which both neural activations and learning signals are represented by non-negative activity, and synapses have fixed sign, while still supporting backpropagation-like learning. Our approach uses two complementary interacting non-negative channels to represent positive and negative contributions, inspired by evidence of on-off representations in the brain. These channels are implemented through a simple neural circuit motif, which is repeated throughout the network in both bottom-up and top-down pathways. Combined with a local Hebbian learning rule, the resulting model propagates learning signals and updates weights using only local interactions between neurons. We show theoretically that our learning scheme can exactly recover the backpropagation update despite relying solely on non-negative error signals. Empirically, beyond satisfying stronger biological constraints, the on-off architecture learns efficient representations, yielding substantial gains over comparable vanilla networks on the Tiny ImageNet benchmark. These results demonstrate that effective learning can emerge from biologically plausible mechanisms without requiring mixed-sign signals, providing a step toward more realistic models of neural computation.
[AI-59] CAi Copilot: Reducing Operational Workload in Molecular Design through Intent-Driven Agent ic Workflows
链接: https://arxiv.org/abs/2608.06961
作者: Zhu Wang,Jiangyu Chen,Yingjun Shang,Yuhui Yao,Laiao Lu,Tianfan Fu,Na Zou
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Early-stage molecular design is an iterative process, not just a task of generating molecules. Researchers turn broad goals into design strategies, refine candidates, assess many properties, and gather evidence before synthesis and tests. AI methods can generate molecules, optimize several goals, predict properties, dock compounds, and account for synthesis. Yet these functions are spread across specialized tools. Experts must still coordinate each step, judge interim results, and integrate evidence. The central challenge is thus to turn research intent into adaptive, traceable runs grounded in scientific tools. We cast this challenge as intent-to-evidence molecular design workflow execution and present CAi Copilot, an expert-oriented agent with three linked layers. The Research Interface Layer turns intent into an executable plan. The Agent Reasoning Layer uses interim results to guide each run. The Execution Substrate supplies molecular tools, metrics, reusable utilities, and backend services. Across 45 tasks, CAi achieves the strongest overall performance, with an outcome score of 84.59, exceeding the next-best result by 18.07 points. Additional benchmarks test how CAi coordinates generation, screening, and multi-criteria evaluation, while exposing limits in long-horizon execution. These results show that CAi turns broad molecular-design intent into transparent, traceable workflows that connect interim decisions to candidate-level evidence.
[AI-60] Critical Acclaim Orientation in Large Language Models : Evidence from Film Preference Elicitation
链接: https://arxiv.org/abs/2608.06955
作者: Jonghyun Jee,Aaron Shaw
类目: Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注: 12 pages, 2 figures, accepted to AIES 2026
Abstract:Large language models (LLMs) are trained on corpora that contain expressions of human judgment about films, books, music, and more. Yet whether LLMs systematically reproduce evaluative hierarchies remains unclear. Prior research on cultural bias in LLMs suggests competing expectations: models may mirror the popularity signals of internet texts, or may reproduce forms of prestige embedded in critical discourse. We probe this question through a study of film evaluations with eight models from four families (Anthropic, OpenAI, Alibaba, and Mistral), using a 200-film benchmark partitioned into critically acclaimed, commercially successful, and dual-legitimacy (critical acclaim + commercial success) films. Across 20,000 pairwise forced-choice comparisons per model analyzed with Bradley–Terry estimation, we observe a consistent critical acclaim orientation with all models: critically acclaimed yet commercially obscure films are selected over commercially successful yet critically unrecognized ones. This pattern grows with model scale within each family. In addition, nested OLS regression analyses show that evaluative orientation, public visibility, and popular reception distinctly help explain preferences. Adjusting for public visibility reverses the models’ preference for dual-legitimacy films over critical acclaim-only films, while additionally accounting for popular reception attenuates much of the disadvantage of films with commercial success only. Finally, evaluative and recommendation-oriented prompt framings produce divergent rankings, suggesting that critical acclaim orientation may manifest indirectly in real-world LLM deployments.
[AI-61] Does Splitting a Triage Decision Across Agents Hide Bias or Help Catch It? A Multi-Agent Simulation Study of LLM -Based Resource Allocation Under Audit Capacity Constraints
链接: https://arxiv.org/abs/2608.06949
作者: Paul-Peter Arslan
类目: Artificial Intelligence (cs.AI)
备注: 6 pages, 2 figures, 3 tables. Code and data available at this https URL
Abstract:Prior benchmarking work has shown that a single large language model (LLM), forced to make life-or-death resource-allocation decisions, exhibits measurable demographic bias. Real deployments, however, rarely use a single agent: they use pipelines, with review steps meant to catch exactly this kind of failure. We study what happens to bias when the same decision is distributed across a role-differentiated multi-agent pipeline (assessment, allocation, independent audit) instead of made and checked by one model alone. Using a synthetic disaster-triage simulator with paired cases that are clinically identical except for one demographic attribute, we run 192 episodes (2,304 resolved case pairs) on GPT-4o-mini comparing a single-agent control condition to a nine-agent pipeline under three independently varied pressure dimensions. We find no measurable difference in how often biased outcomes occur between the two conditions (6.9% vs. 6.1%, p = 0.498). We do find a large and significant effect of audit capacity on whether bias is caught: 30.0% of biased outcomes go entirely undetected, rising to 43.8% when the auditor is overloaded and falling to 18.4% when it is not. Decomposing this effect shows it is driven almost entirely by coverage (whether a case is reviewed at all, which collapses from 100.0% to 65.6% under load, p 0.001) rather than by degraded judgment on the cases that are reviewed (81.6% vs. 85.7%, p = 1.000, direction reversed). A follow-up experiment shows that reordering the audit queue by estimated risk, rather than first-come-first-served, recovers most of the lost coverage under the same capacity constraint (65.6% to 91.7%, p = 0.028). We discuss the implications for any system that adds independent oversight to an LLM agent pipeline under resource constraints, and report the study’s limitations honestly: one model, modest sample sizes, and no adversarial replication.
[AI-62] LMM Modality Transfer: A Pre-requisite for Autonomous GIS Agents
链接: https://arxiv.org/abs/2608.06948
作者: Ivan Majic,Zexian Huang,Franziska Hübl,Krzysztof Janowicz,Meilin Shi,Mina Karimi,Zilong Liu,Alexandra Fortacz-Lazan
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:AI models are becoming increasingly adept at understanding and processing spatial information, thereby facilitating agentic problem-solving in spatial tasks and workflows. However, most of the research on their spatial capabilities (e.g., spatial reasoning) has focused on the textual modality as input and output. This contrasts with the human approach to GIS workflows, where text and visual modalities are often used together, interchangeably, and in a complementary manner. Thus, to truly achieve an automated GIS analysis pipeline or carry out human-designed GIS workflows, AI models — Large Multimodal Models (LMMs) in particular — need to be able to seamlessly transition between image- and text-based modalities that are traditionally used in such workflows. We present a modality transfer task that (1) asks an LMM to first describe an input image of colored squares in a regular grid, and (2) asks a new LMM instance to re-generate an image of the original spatial scene using the textual description output by the former model. This task quantifies the ability of LMMs to transfer spatial information between image and text modalities. Ultimately, by examining the modality transfer capability of LMMs through the lens of spatial information theory, this work highlights a critical bottleneck: achieving strong and robust geospatial understanding in LMMs requires rigorous, multi-modal alignment. Our results indicate that recent LMMs (here from OpenAI) still struggle with modality transfer, when tasked with re-generating an image of a simple spatial grid of color squares.
[AI-63] Blind to the Pivotal Vote: Aggregate Independence Metrics Miss Where Verification Actually Helps
链接: https://arxiv.org/abs/2608.06940
作者: Yang Shu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:LLM judge panels are a standard evaluation tool, but prior work reports highly correlated panel errors: nine judges provide roughly the effective information of two independent ones, and aggregation closes only a small fraction of the gap. A natural remedy–a signal from a different evidence source, e.g., executing a test suite–produced no distinguishable change in the panel’s effective-vote count at scale (-0.04, 95% CI [-0.10, +0.02]). Aggregate dependence and conditional decision utility are different questions. Elementary majority arithmetic fixes the affected set for single-ballot substitution: only decisions with a one-vote margin can change. The empirical question is whether panel error rates rise and useful substitutions concentrate there. They do: the entire accuracy gain concentrates on these pivotal queries, where it is large (+10.4 to +23.3 percentage points across three headline configurations), and is exactly zero elsewhere. We confirm the pattern across three code benchmarks and four panel sizes (a 9-judge extension and 56 dependent subsampling checks, gain +6.5 to +16.1 percentage points). On HumanEval+/MBPP+, a majority-side replacement rule raises overall accuracy from 82.44% to 85.62% while invoking the signal on 16.2% of queries; signal-only remains stronger at 87.60%. Thus population-level dependence diagnostics and margin-stratified utility are complementary, and the affected-set characterization yields a call-reduction rule for any specified single-ballot substitution policy.
[AI-64] Science Edge Evaluation: SEE the Missing Step Toward Real Scientific Discovery
链接: https://arxiv.org/abs/2608.06931
作者: Taolin Han,Yuchen Zhang,Jinghang Wang,Yun Wu,Wai Yuet Chiu,Zhaohai Li,Yifei Zhang,Jinxin Wang,Yuhao Zhou,Chen Zhao,Jiajia Li,Jiaxin Li,Qile Jin,Kewei Sun,Shuang Wu,Weiqi Zhai,Renquan Lv,Junchao Li,Ruodan Chen,Qingteng Chen,Zhibo Yang,Hu Wei,Lin Qu,Shuai Bai,Bing Zhao
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models (LLMs) are increasingly involved in scientific discovery, yet it remains unclear whether they can support complex real laboratory science. Here we introduce Science Edge Evaluation (SEE), a multimodal benchmark of expert-curated questions grounded in peer-reviewed literature and experimental practice in chemistry, biology, and materials science. Evaluation of 19 multimodal large language models (MLLMs) shows that even the best-performing model reaches only 48.7% accuracy. Moreover, general-purpose models outperform science-specialized models on average. In the visual-agent evaluation, the use of tools increases the best accuracy to 52.7%. Tool use can expand the information available to models, but more information does not necessarily lead to reliable scientific reasoning. The key challenge is whether models can manage tool-derived information within the boundaries of the original experimental evidence. Together, these findings reveal that current MLLMs still cannot reliably make justified and evidence-bounded inferences from experimental results, which is an essential capability in real scientific discovery. Bridging this gap requires MLLMs to transition from explaining established scientific concepts to deriving novel and evidence-based insights from experimental data.
[AI-65] RIBE: Predicting Team Performance via Communication Behavior Ensembles
链接: https://arxiv.org/abs/2608.06926
作者: Ali Jalal-Kamali,Nikolos Gurney,David V. Pynadath,Fred Morstatter
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Designing autonomous agents that effectively assist human teams hinges on understanding team dynamics, often without task specific knowledge. We present TRIBE, a domain independent approach that reveals team behavioral dynamics invisible to traditional performance metrics. We show that communication patterns can categorize teams into performance predictive behavioral tribes, as early as 10% into the task, enabling timely interventions. We test TRIBE on four diverse datasets and demonstrate that communication patterns predict team performance while the prediction strength varies by the degree a task structure allows for behavioral freedom. Our temporal analysis reveals that AI agents significantly alter team behavioral trajectories while human advisors align with natural dynamics, and that teams maintain behavioral flexibility throughout collaboration. Further, we compare TRIBE to Llama and optimize the pipeline, achieving significant speedup with performance improvement.
[AI-66] Deal Me Maybe: The Role of Emotions in Multi-Agent Negotiation
链接: https://arxiv.org/abs/2608.06922
作者: Massimiliano Luca,Apoorva Singh,Bruno Lepri
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Negotiation is a demanding social task for LLM agents, requiring strategic reasoning, persuasion, and interpersonal adaptation. Yet existing benchmarks often treat agents as emotionally neutral, overlooking a key driver of human bargaining behavior. We study how prompt-conditioned emotions affect LLM-based price negotiation. In a controlled framework, buyer and seller agents are independently assigned one of six emotional states and negotiate over 350 real consumer products under two budget conditions. Across 36 emotion-pair settings and five widely used LLMs, we find that emotions strongly shape outcomes. Angry buyers almost never reach agreement (0.39% deal rate), while happy buyers agree most often (28.91%), but obtain worse prices than fearful buyers. Emotion effects are role-dependent: buyer emotion mainly drives acceptance and rejection, whereas seller emotion shapes concession dynamics. These effects influence not only language, but also termination behavior and price trajectories, raising concerns for emotion-conditioned agents in commerce.
[AI-67] ReGraph: Learning to Generate Recipe Graphs from Food Images
链接: https://arxiv.org/abs/2608.06917
作者: Guoshan Liu,Bin Zhu,Pengkun Jiao,Jingjing Chen,Chong-Wah Ngo,Yu-Gang Jiang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Recent Large Multimodal Models (LMMs) have achieved impressive performance in recipe generation from food this http URL, cooking is a structured transformation process in which ingredients undergo state changes through ordered actions,while free-form recipe language leaves the corresponding entities, intermediate states, and dependencies largely implicit and entangled.A graph representation makes this procedural knowledge explicit and compositional, providing a structured basis for assessing whether model outputs encode process-level knowledge rather than merely presenting plausible textual descriptions. To address this limitation, we present ReGraph, a large-scale recipe graph dataset that represents ingredients, cooking actions, and tools as entities, uses entity attributes to describe ingredient state changes, and employs typed relations to encode manipulation targets, destinations, and procedural ordering. ReGraph further incorporates explicit Recipe Reasoning Chain-of-Thought (RR-CoT) traces, providing auxiliary supervision for procedural decomposition and structured graph generation. Building on ReGraph, we propose Recipe Graph Learning (RGL), a two-stage framework that enables LMMs to generate a plausible fine-grained cooking workflow from a food image in the form of a structured recipe graph. Under a deterministic, schema-aware matching protocol, our experiments reveal a substantial gap between text-generation quality and recoverable procedural structure: recipes produced by existing approaches achieve competitive text-generation scores yet yield limited reference-aligned entity and relation structure under the ReGraph schema. In contrast, across two representative LMM backbones, RGL consistently improves the generation of cooking entities and procedural relations, while our analysis further shows that fine-grained ingredient-state capture remains the most challenging dimension.
[AI-68] Fast LapSum: Exact Differentiable Top-k at Million Scale
链接: https://arxiv.org/abs/2608.06912
作者: Łukasz Struski,Joanna Wojciechowicz,Jakub Antczak,Marcin Mazur,Kamil Książek,Jacek Tabor
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:The top- k operation is a fundamental building block of modern sparse computation, enabling token routing, expert activation, memory selection, and attention pruning. Yet standard hard top- k blocks gradients, while existing continuous (soft) relaxations remain too costly for large-scale models. We introduce Fast LapSum, an exact-budget soft top- k primitive whose GPU solver runs in linear time after sorting. Unlike prior linear-time methods such as DFTopK, which relax the normalization constraint, Fast LapSum is, to our knowledge, the first method to preserve an exact selection mass of k while remaining fully differentiable end-to-end. Our solver combines a linear-time threshold computation with an analytical vector–Jacobian product, and for extreme scales employs probabilistic bracketing to sort only the uncertain middle band of kernel-noised scores. The resulting overhead is almost negligible: the solver processes 10^6 , 10^7 , and 10^8 scores in 0.41 , 1.15 , and 5.23 ,ms, respectively. This makes exact soft top- k practical for sparse routing, retrieval, and large-scale optimization. We demonstrate Fast LapSum on two demanding applications operating over millions of coordinates inside the training loop: generating megapixel sparse adversarial examples with an exact soft budget of \sim0.02% of an image’s pixels, achieving an order-of-magnitude speedup over state-of-the-art methods, and training a fully differentiable sparse image coder from scratch.
[AI-69] Long-Horizon Agent Trajectory Attribution: A Unified Benchmark and Fine-Grained Annotation Framework
链接: https://arxiv.org/abs/2608.06909
作者: Jing Chen,Yang Sun,Li Zhang,Lin Xu,Jie Shi
类目: Artificial Intelligence (cs.AI)
备注: 17 pages, 4 figures, 7 tables
Abstract:Large language model (LLM) agents increasingly operate through long-horizon trajectories involving user instructions, tool use, external observations, and memory. Existing benchmarks primarily evaluate behavioral outcomes but provide limited support for fine-grained attribution analysis. We introduce trajectory attribution and develop a benchmark and annotation framework for this task. The benchmark organizes heterogeneous trajectories under a unified component schema and provides annotations of the primary attribution component, together with attack and execution chains where applicable. Instantiating the benchmark with trajectories from AgentDojo and the Stage and Canary settings of Agent3Sigma yields more than 1,300 annotated trajectories covering task-aligned actions, unsafe actions, and safety refusals. The benchmark defines two evaluation tasks, primary attribution localization and attribution-chain recovery, and provides reference baselines based on incremental trajectory contribution and component-level leave-one-out perturbation. It captures diverse attribution settings, including local and long-range attribution as well as structured attribution chains. Reference baseline results exhibit substantial performance differences across these settings, providing an initial characterization of the benchmark’s attribution challenges. Beyond this initial instantiation, we release a reusable annotation skill that enables trajectories generated by new agent models to be standardized, annotated, and evaluated under the same framework. Project resources and future releases are available at this https URL.
[AI-70] SkillEval: Decomposing Agent Skill Quality into Interpretable Signals
链接: https://arxiv.org/abs/2608.06891
作者: Jiahui Han,Qinuo Li,Ziheng Peng,Haotian Wu,Haoze Liu,Danfeng Shan,Guanchu Wang,Huiqi Deng,Ninghao Liu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Agent skills provide reusable procedural knowledge that helps agents solve specialized tasks. As their use expands, evaluating skill quality becomes increasingly important. Existing evaluations often measure skill quality by testing whether a skill improves performance on specific downstream tasks. However, a reusable skill may apply to multiple task scenarios. Downstream evaluation mainly reflects the compatibility between a skill and the evaluated task, provides only a partial view of skill quality, and does not identify which aspect of the skill should be improved. We find that general properties of the \textttthis http URL document play an important role in skill quality. To evaluate these properties, we propose \textbfSkillEval, an interpretable framework for document-level skill evaluation. SkillEval evaluates each property using a fixed and inspectable scoring direction, producing interpretable scores. It further measures and reduces the influence of unrelated document features, such as length and formatting, so that each score captures its intended semantic property more specifically. Specifically, SkillEval learns an interpretable direction for each quality property from controlled positive–negative skill pairs in the hidden representation space of the model, and scores a new skill by projecting its representation onto these fixed directions. We use SkillEval to evaluate skills in controlled quality tests and show that SkillEval reliably distinguishes skills of different quality. In addition, SkillEval scores closely reflect downstream task performance, providing an early indication of whether a skill is likely to help an agent complete a task. We further explore SkillEval for diagnosing weaknesses in skill documents and guiding targeted revisions. The revised skills improve the targeted properties and achieve higher pass rates on downstream tasks.
[AI-71] CEDAR: Agent -Orchestrated Tree Search for Goal-Directed Optimization of Complex Systems
链接: https://arxiv.org/abs/2608.06871
作者: Yingtao Tian
类目: Artificial Intelligence (cs.AI)
备注: Accepted as Talk in ALIFE 2026: The 2026 Conference on Artificial Life
Abstract:Complex systems, core objects of study in artificial life, model diverse phenomena through nonlinear, feedback-driven interactions that produce emergent behavior, with applications from population dynamics and biology to economic policy and strategic decision-making. Yet the difficulty of predicting how feedback structure gives rise to emergent behavior, a central open problem in artificial life, makes goal-directed design exceptionally challenging. In established practice, system structures are written in specialized modeling languages such as DYNAMO or STELLA, compounding the challenge with labor-intensive workflows that limit adoption and hinder timely decision-making. To address these challenges, we introduce CEDAR, an autonomous method that uses Large Language Model (LLM) agents to discover complex systems satisfying user-specified behavioral goals. Our key innovation is an LLM-driven Monte Carlo Tree Search (MCTS) deeply coupled with complex systems: at each iteration, an LLM Judge evaluates emergent behavior against specified goals and an LLM Editor proposes improved variants, with the Judge acting as a fitness function and the Editor as a variation operator, akin to a generate-and-evaluate loop in evolutionary computation. We represent complex systems as a restricted, runnable subset of Python with domain-specific primitives, letting LLMs modify system dynamics directly. CEDAR formalizes this as an MCTS variant with an LLM-parameterized transition kernel and value function, enabling goal-directed discovery of complex system behaviors while preserving solution diversity, and its LLM-based interpretability reveals how structural changes drive emergent behavior. CEDAR reduces human effort while enabling capabilities difficult to achieve with existing approaches, facilitating broader adoption of complex systems across domains.
[AI-72] Gated-BEPO: Confidence-Gated Bellm an Credit Assignment for Large Language Model Agents
链接: https://arxiv.org/abs/2608.06861
作者: Hongxi Yan,Ziyue Huang,Shichao Fan,Qingjie Liu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Training large language model agents in long-horizon environments requires assigning credit from sparse terminal outcomes to individual actions. Existing critic-free methods propagate trajectory-level rewards uniformly across steps, while recent approaches construct step-level groups by matching repeated states and compare actions within each group. The former cannot distinguish useful actions in failed trajectories from ineffective actions in successful ones. The latter rely on step credit derived directly from individual trajectory outcomes and fixed-weight fusion with episode-level credit. We propose Gated-BEPO, which derives step-level credit from empirical rollout graphs. For each rollout group, Gated-BEPO constructs an empirical graph and estimates node values through a mean-backup Bellman fixed point that reflects the empirical action distribution of the current policy. We then accumulate these temporal-difference residuals along each sampled trajectory using generalized advantage estimation, yielding step-level Bellman advantages that capture both immediate and downstream effects. To adaptively fuse episode- and step-level credit, a confidence gate incorporates Bellman credit only at states with multiple observed successors and otherwise uses episode-level credit. Experiments on WebShop, ALFWorld, and visual Sokoban show consistent improvements across language and vision-language models, while diagnostic ablations support the effectiveness of Bellman fixed-point value estimation and show that step-level credit should be incorporated selectively rather than uniformly into the final advantage.
[AI-73] Bridging the Gap Between Hyperdimensional Computing and Kernel Methods via the Nyström Method
链接: https://arxiv.org/abs/2608.06860
作者: Quanling Zhao,Anthony Hitchcock Thomas,Ari Brin,Xiaofan Yu,Tajana Rosing
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Hyperdimensional computing (HDC) is an approach from the cognitive science literature for solving information processing tasks using data represented as high-dimensional random vectors. The technique has a rigorous mathematical backing, and is easy to implement in energy-efficient and highly parallel hardware like FPGAs and “processing-in-memory” architectures. The effectiveness of HDC in machine learning largely depends on how raw data is mapped to high-dimensional space. In this work, we propose NysHD, a new method for constructing this mapping that is based on the Nyström method from the literature on kernel approximation. Our approach provides a simple recipe to turn any user-defined positive-semidefinite similarity function into an equivalent mapping in HDC. There is a vast literature on the design of such functions for learning problems. Our approach provides a mechanism to import them into the HDC setting, expanding the types of problems that can be tackled using HDC. Empirical evaluation against existing HDC encoding methods shows that NysHD can achieve, on average, 11% and 17% better classification accuracy on graph and string datasets respectively.
[AI-74] Coupling Planning with Episodic Memory in LLM Agents for Software Issue Resolution
链接: https://arxiv.org/abs/2608.06811
作者: Jiahao Zhang,Yifan Zhang,Yu Huang
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: 12 pages, 5 figures
Abstract:Resolving a real software issue with a large language model (LLM) agent is a long repair episode, often tens to hundreds of steps spanning exploration, hypothesis, implementation, and verification. Success depends on both the base model’s local reasoning and the agent’s ability to maintain an evolving plan and remember observations across phases. Existing repository-level agents typically strengthen planning or memory in isolation, leaving long trajectories vulnerable to stale evidence, repeated failed edits, and verification inferred from the agent’s own claims instead of execution evidence. We present PMCoder, an issue-resolution agent that couples a hierarchical phase planner with episodic memory. The coupling is bidirectional: the current plan phase conditions memory retrieval, while memory-derived trajectory statistics inform stuck detection and replanning. When available, issue-reproduction verdicts ground verification progress in execution evidence rather than self-reported completion. On SWE-bench Verified, PMCoder resolves an average of 25 more cases ( +5.0 pp) than a harness-matched baseline, with gains persisting even where the reproduction gate never fires. Further Verified-500 evaluations show the same positive direction across Claude Haiku 4.5, DeepSeek-V4-Flash, and an OpenHands port, with at least 14 additional resolved cases ( +2.8 pp). Separately, evaluation on TerminalWorld’s official sample suggests that the plan-memory substrate transfers beyond issue reports. Ablation and trajectory analyses show where the gains come from: coupling planning and memory outperforms either component alone and reduces repeated failed actions, empty-patch exits, and context-window exhaustion.
[AI-75] Evolving Parallel Algorithm Portfolios via Potential-Aware Instance Generation with LLM s
链接: https://arxiv.org/abs/2608.06808
作者: Shaofeng Zhang,Shengcai Liu,Zhiyuan Wang,Ke Tang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:The Automatic Construction of Portfolios via Large Language Models (LLM-ACP) suffers from poor generalization in practical few-shot scenarios when solving complex combinatorial optimization problems. Instance and algorithm co-evolution frameworks address this by expanding the training dataset with generated hard instances on which the current algorithm portfolio underperforms, thereby enhancing generalization. However, this paradigm faces two critical limitations: evaluating instance hardness relies on high-quality reference solutions, and single-mode generation patterns limit instance diversity. To overcome these limitations, we introduce the Potential-aware Instance and Algorithm Co-evolution (PIAC) framework. Our core contribution is twofold. First, we propose potential gain, a novel metric that eliminates the need for reference solutions. This metric estimates generalization gain by perturbing the generated algorithms and assessing their improvement potential on generated problem instances. Second, PIAC leverages LLMs to synthesize diverse instance mutators, exploring a broader region of the problem-instance space and thereby enhancing the portfolio’s generalization capabilities. Given that perturbation spaces vary across different algorithms, we instantiate our framework on Greedy Constructive, Ant Colony Optimization, and Guided Local Search algorithmic backbones. Comprehensive evaluations on the Traveling Salesman Problem (TSP) and Capacitated Vehicle Routing Problem (CVRP) across six distinct data distributions demonstrate that PIAC consistently outperforms state-of-the-art LLM-ACP baselines, notably achieving a 19.76% relative improvement for TSP Greedy Constructive portfolios.
[AI-76] HLSmith: An Expert-Guided Agent ic Framework for C/C-to-HLS Translation
链接: https://arxiv.org/abs/2608.06791
作者: Yuebo Luo,Ahmad Sedigh Baroughi,Philip Stachura,Le Chen,Venkatram Vishwanath,Zhenman Fang,Caiwen Ding
类目: Hardware Architecture (cs.AR); Artificial Intelligence (cs.AI)
备注:
Abstract:Application-specific FPGA accelerators offer substantial performance and energy-efficiency gains across many application domains, but developing them is costly, often requiring months of specialized effort. Even with high-level synthesis (HLS), designers still need extensive hardware expertise to build high-performance accelerators. Although large language models (LLMs) have demonstrated strong software-generation capabilities, even frontier models lack the hardware intuition and procedural knowledge needed to reliably translate baseline C/C++ programs into high-performance HLS designs: they struggle to identify effective architectures, follow the optimization processes used by HLS experts, and apply hardware transformations consistently across diverse kernels. We present HLSmith, an expert-guided framework for translating C/C++ programs into optimized HLS accelerators. HLSmith combines three components: an HLS optimization expertise library that encodes guarded transformation recipes, their applicability and prerequisite conditions, and unsafe cases to avoid; a staged, feedback-driven orchestration flow modeled on expert HLS development practice that guides agents through synthesis, bottleneck analysis, and optimization; and a tool-grounded model-adaptation pipeline that converts optimization trajectories from commercial frontier models into training data for fine-tuning open-weight LLMs. We evaluate HLSmith on PolyBench against ChatHLS, a leading prior agent-orchestration framework for HLS accelerator development. HLSmith achieves a geometric mean speedup of 4.24x over ChatHLS while producing functionally correct designs, in both software and RTL simulation, for every benchmark, compared with ChatHLS’s 57% valid-design rate. It further reaches speedups of up to 252x and 138x with commercial frontier models and open-weight models, respectively.
[AI-77] Hidden Gauge Controls Feature Specialization in ReLU Networks
链接: https://arxiv.org/abs/2608.06766
作者: Tongxi Wang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Training changes a network’s predictions while allocating task-relevant structure across its internal units. In an overparameterized ReLU network, several neurons can begin with exactly the same functional role, yet one may acquire a teacher feature while the others become redundant. We call the identity of that neuron feature ownership and ask whether it can be controlled by a parameter choice invisible to the initial predictor. In a tractable Gaussian teacher–student model, we fix the complete initial function and vary only a positive-homogeneous scaling gauge. Opposite gauges produce distinct feature trajectories and a sharp \Theta(D^2) separation in specialization time that no global change of clock can explain. Among any fixed number of initially duplicate students, assigning the favorable gauge to one neuron deterministically selects it as the owner and drives the remaining functional contribution to zero. An exact reaction–transport decomposition attributes the effect to different mobilities for changing a feature’s coefficient and direction. We prove global selection and functional pruning, extend finite-time selection to visible perturbations and small-step full-batch gradient descent, and verify the predicted loss, alignment, pruning, and dissipation trajectories in population and finite-sample training. The initial predictor therefore determines neither when the feature is learned nor which neuron learns it.
[AI-78] LiFTER: A Grounded Neuro-Symbolic Microscope for Continuous-Time Dynamic Graph Forecasting
链接: https://arxiv.org/abs/2608.06765
作者: Minwoo Yu,Young-guk Ha
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Continuous-time dynamic graph models predict future links by compressing past interactions into neural states. Although effective for forecasting, this computation obscures which entities are shared across events and how temporal patterns contribute to a prediction. We treat this gap as a property of the predictive architecture rather than a problem to be addressed after prediction. Link-Fact Temporal Rule Inducer (LiFTER) is a neuro-symbolic predictor that preserves observed interactions as grounded temporal facts and applies executable tempo- ral rules to pre-query facts. Each score is a signed sum of rule exe- cutions whose historical facts, entity bindings, and temporal order are explicitly satisfied. The evidence and rules responsible for a prediction can therefore be inspected, independently recomputed, and intervened upon. Across four CTDG benchmarks, LiFTER achieves competitive historical-negative forecasting and the highest macro explanation ac- curacy and deletion fidelity. The same architecture also serves as a microscope that separates the contributions of recurrence, history po- sition, and transition across datasets and traces them to individual facts. Independent execution reconstructs all logits for 19,664 test predictions with a maximum error of 0.0000131. LiFTER turns future-link forecasting into a verifiable grounded computation.
[AI-79] Capek 0.5: An Execution-Centric Vision-Language Model for Embodied Intelligence
链接: https://arxiv.org/abs/2608.06756
作者: Ying Chen,Weizhen Li,Zhe Hu,Zhenjiang Li,Rui Jiang,Zhifeng Gu,Lihuang Fang,Jiangping Liu,Lei Yi,Jie Chen
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Vision-language models are increasingly serving as the reasoning core of embodied agents. Robot execution is inherently iterative: each action reshapes the scene and physical state, continually renewing what must be perceived, reasoned about, and verified. Meeting these demands requires complementary capabilities that differ in supervision signals, prediction formats, and verification criteria. Existing approaches typically develop these capabilities against isolated, task-specific objectives, leaving open how they should be organized and integrated around execution as a whole. We present Capek 0.5, an embodied vision-language model built around an execution-centric capability taxonomy. Rather than organizing training by datasets or tasks, the taxonomy groups embodied capabilities according to their functional roles throughout execution and comprises four capability families: Spatial Reasoning, Temporal Understanding, Action Guidance, and State Verification. Each capability is first acquired by a dedicated specialist through reinforcement learning with verifiable rewards from a shared backbone, and the specialists are then consolidated into a single inference-time model through weight-space merging followed by routed policy-space distillation. We instantiate Capek 0.5 at the 2B and 35B-A3B scales and evaluate it from three complementary perspectives: comprehensive benchmark suites including Capek-StateBench, a new benchmark for state verification; a controlled study of capability retention from specialists to the unified model; and closed-loop evaluation in simulated embodied environments. Capek 0.5 improves the large majority of matched benchmark rows over its initialization, retains all four specialized capabilities in one checkpoint with quantified losses, and transfers to closed-loop embodied task execution.
[AI-80] KReF: Training-Free Retrieval for Long-Term Time-Series Forecasting and Predictive Uncertainty
链接: https://arxiv.org/abs/2608.06748
作者: Yang Zhang,Rui Su
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Probabilistic long-term time-series forecasting commonly relies on trained models. Training-free conformal methods typically construct intervals around a pre-existing point forecaster and do not natively represent a complete predictive distribution; sequential variants additionally suffer from increasingly delayed feedback at long horizons. We propose KReF, a training-free retrieval framework that treats retrieved historical futures as a querylocal empirical predictive distribution. After robust preprocessing, KReF embeds each lookback using handcrafted statistics or frozen random Fourier features and retrieves similar historical lookback-future pairs. Their similarity weights directly define predictive masses, quantiles, CRPS, and a weighted-mean point forecast. KReF further uses the observed query lookback to construct a probability-integral-transform map and applies validation-selected expansion and shrinkage rates to adapt interval boundaries. Across six LTSF benchmarks and four horizons, KReF obtains the lowest CRPS in all 12 dataset-embedding settings and the lowest IS90 in 9 settings. Without gradient-based fitting, its point forecasts also match or surpass trained baselines on two of six datasets. An archive-oracle analysis further reveals substantial headroom under finer horizon- and channel-wise routing. These results establish retrieval as a useful and underexplored inductive bias for LTSF.
[AI-81] MemPrism: Task-Conditioned Relational Memory Views for Long-Horizon Agents
链接: https://arxiv.org/abs/2608.06745
作者: Zhisheng Chen,Bingfan Zeng,Bangde Cao,Zhengwei Xie,Yuxuan Li,Jinhan Li,Zheng Lu,Xiangchen Guan,Zikai Xiao,Rui Qian,Jingwei Song
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Long-horizon agents rely on memory to reuse experiences, yet existing memory systems often assume that evidence can be directly consumed through a fixed representation. This leads to representation mismatch, where relevant information is available but not organized for the current decision. To this end, we propose MemPrism, a task-conditioned relational memory framework that separates persistent experience storage from decision-time working memory. MemPrism records interactions as the event stream and dynamically constructs relational views according to the current task context. A lightweight view policy selects the relation structure, evidence range, outcome condition, and granularity, while a deterministic composer and render transform historical facts into a temporary optical working-memory view for a frozen task policy. Experiments on long-horizon embodied and web-agent benchmarks show that MemPrism consistently improves the task performance, especially as trajectories become longer, while reducing memory token consumption. Furthermore, the learned view policy transfers across different VLMs without additional adaptation, demonstrating the effectiveness of task-conditioned relational views as a general memory interface for agents.
[AI-82] bioMoR: Biology-Guided Mixture-of-Recursions for Effective Genomic Learning
链接: https://arxiv.org/abs/2608.06727
作者: Koushik Howlader,Tirtho Roy,Md Tauhidul Islam,Wei Le
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Transformer models for high-dimensional omics analysis process thousands of genes or pathways, although only a subset requires deep computation. Mixture-of-Recursions (MoR) improves efficiency through adaptive token-choice or expert-choice routing. We propose bioMoR, which, to the best of our knowledge, is the first framework to apply MoR to gene-level and pathway-level learning. Our contributions include identifying three locations for integrating structured biological knowledge within an MoR backbone: graph-based information sharing refines token embeddings, a structural bias guides self-attention toward biologically related tokens, and a graph-aware router uses neighborhood information to determine each token’s recursion depth. These techniques are centered on our insight that additional knowledge of token interaction can effectively help models construct embeddings and select which tokens should be learned more deeply. Across eight benchmarks spanning diverse omics data types and evaluated under a unified five-fold cross-validation protocol, bioMoR improves average macro-F1 by 8.2 percentage points and balanced accuracy by 7.1 percentage points over the strongest biology-agnostic MoR baseline while using 75 percent fewer parameters and up to 58 percent fewer FLOPs than a non-recursive Transformer. The selected marker genes or pathways provide biological interpretability, while their token-specific recursion depths reveal how computation is allocated.
[AI-83] Multi-Level Modeling of Large Language Model Inference Latency and Energy via Hybrid Analytical–Machine-Learning Predictors
链接: https://arxiv.org/abs/2608.06723
作者: Saeid Shokoufa,Mohammad Erfan Sadeghi,Mehdi Kamal,Massoud Pedram
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:The rapid scaling of Large Language Models (LLMs) has significantly increased computational cost, energy consumption, and inference latency, making accurate estimation essential for sustainable artificial intelligence deployment and hardware-aware design. In this work, we introduce Hybrid Modeling for Energy and Latency of LLMs (HYMELL), a hybrid three-level framework for estimating LLM inference latency and energy by combining analytical modeling with machine learning (ML). HYMELL models LLM execution through a three-level hierarchy: analytical estimation of primitive operations, ML prediction of higher-level components, and an end-to-end model that captures system-level overheads across both prefill and decode phases. The framework supports diverse architectures, including dense and mixture-of-experts (MoE) feed-forward networks (FFNs), as well as multi-head attention (MHA) and grouped-query attention (GQA) mechanisms. Evaluated on an NVIDIA H100 graphics processing unit (GPU), HYMELL achieves high predictive accuracy; notably, for LLaMA 3 8B, it attains less than 5% error for both prefill and decode phases. By predicting execution costs directly from architectural parameters, it enables fast, hardware-free design space exploration and energy-efficient optimization.
[AI-84] he Optimizer Is the Agent : Reasoning -Driven Search across Prompts Programs and ML Workflows
链接: https://arxiv.org/abs/2608.06714
作者: Junbo Li,Boyi Liu,Canwen Xu,Yite Wang,Yuxiong He,Zhangyang Wang,Qiang Liu,Zhewei Yao
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Recent systems for optimizing prompts, programs, and ML workflows typically rely on explicit outer-loop controllers such as evolutionary search, bandits, or textual-gradient methods. We ask a fundamentally different question: how much of this search policy can be internalized by a single tool-using agent? We present ReASearch, a unified framework for reasoning-driven optimization in which the agent autonomously decides what to evaluate, how to diagnose failures, which edits to make, and when to verify or restart. Rather than serving only as a proposal generator guided by hand-designed heuristics, the agent actively analyzes outcomes, allocates budget, and refines its strategy over long horizons through persistent memory. With a shared agent loop and domain-specific tools, ReASearch instantiates the exact same scaffold to optimize prompts, programs, and ML workflows. Across 14 diverse tasks, it is competitive with and mostly better than specialized optimization systems, achieving gains of 2% to 40% over strong domain-specific baselines, and in some cases discovering solutions that improve on prior human best-known results. Crucially, we observe that complex search behaviors, which are typically implemented by explicit controllers, emerge naturally from the agent’s reasoning process.
[AI-85] MolBioKG: Grounding Out-of-Graph Molecules in Biomedical Knowledge Graphs via Multi-Resolution Structural Anchoring
链接: https://arxiv.org/abs/2608.06713
作者: Yiming Zhang,Hikaru Shindo,Shuan Chen,Kaushalya Madhawa,Jun Jin Choong,Yuna Oikawa,Takashi Fujiwara,Keisuke Ozawa
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Preprint
Abstract:Biomedical knowledge graphs (KGs) accelerate drug discovery, but standard pipelines assume query molecules already exist as graph entities, leaving unregistered molecules disconnected. We address this cold-start challenge, termed the out-of-graph molecule problem, by introducing MolBioKG. This two-layer system grounds unseen molecules in biomedical evidence via multi-resolution structural anchoring. It connects an index of 2.74 million molecules (represented by scaffolds, fragments, functional groups, and fingerprints) to a 9.6-million-edge KG. Given only a SMILES string, MolBioKG retrieves structurally related graph entities and traverses their biomedical neighborhoods without task-specific training. It features two inference mechanisms: static multi-anchor retrieval using Reciprocal Rank Fusion, and Adapt-KG, a tool-using LLM policy for adaptive traversal. Evaluated across in-graph link recovery, complex multi-hop reasoning, and out-of-graph generalization, MolBioKG outperforms strong baselines. Notably, it raises Hits@10 from 0.585 to 0.876 in multi-hop reasoning and out-of-graph target recall from 0.145 to 0.269, all while ensuring predictions retain traceable structural anchors and source-attributed KG evidence.
[AI-86] Dueling World Models: Advantage-Style Action Channels for Common-Mode Distractor Rejection
链接: https://arxiv.org/abs/2608.06706
作者: Jiazhuo Li,Yiming Fei,Zhiruo Zhou,Heikichi Hayashi
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 17 pages, 6 figures, 11 tables. Includes supplementary appendix
Abstract:Latent world models plan by predicting future states from an action, but when a scene contains motion the agent does not control, they quietly go action-blind: predictions for different actions become indistinguishable even as the training loss keeps improving. Existing remedies suppress this distraction with reconstruction, task reward, or auxiliary objectives, each adding machinery or assumptions. We show that a minimal alternative suffices, borrowed from the dueling decomposition of value into a state baseline and an action advantage: in latent dynamics, subtracting a prediction’s mean effect over actions cancels whatever the actions share–the action-independent variation where distractors live–leaving a clean, controllable channel, with no reward, no reconstruction, and no distractor-specific auxiliary loss. Because this is only a subtraction at readout time, it applies unchanged to any action-conditioned world model, including frozen pretrained ones. Across a gridworld, synthetic generators with known factors, distracting continuous control, and natural-pixel Atari, the isolated channel recovers the agent’s own effect where entangled predictors fail, with nuisance leak indistinguishable from zero; applied post hoc it surfaces an action channel in off-the-shelf models that their raw readouts miss, and it converts into goal-reaching control in the gridworld. We prove the cancellation is exact in finite samples for both discrete and sampled action sets, and we state its measured boundary–distractors whose motion tracks the action–together with the remaining limitations in the appendix.
[AI-87] Policy-Masked Private Experts: Auditable and Reversible Capability Access Control in Sparse MoE Models
链接: https://arxiv.org/abs/2608.06690
作者: Zhuoheng Huang,Mukesh Singh
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Most language-model access controls regulate behavior while leaving the same computation available to every request. We study a different systems question: can trusted authorization determine which newly trained parameters are reachable by the forward pass? Policy-Masked Private Experts freezes a pretrained sparse Mixture-of-Experts (MoE) model, trains a disjoint expert branch, and selects the public or private pool before top-k routing. The resulting claim is narrow but testable: under the declared trusted computing base (TCB), an unauthorized request executes no private expert. It does not imply that the public model lacks the same semantic capability. We test this separation between execution control and task utility in Qwen3-30B-A3B and DeepSeek-V2-Lite. Three Qwen BF16 seeds update all 32 private experts while the public fingerprint remains unchanged. Across 64 adversarial scenarios and 96 deny/fail-closed events, unauthorized private execution is zero; independent hooks exactly match 11,616 routed private rows and allow-deny-allow recovery is exact. On two prospectively frozen Qwen benchmarks, the private branch improves exact tool use by 5.0 percentage points (pp) (five versus zero discordances; one-sided Holm p = 0.03125, corresponding two-sided exact p = 0.0625) and 21.3 pp (percentile-bootstrap 95% CI [13.3, 29.3], Holm p = 0.000031). Three arm-blinded model evaluators retain a positive external effect of 18.7 pp (95% CI [9.3, 28.0]). A parameter-matched Lora has similar external utility, but a post-hoc request gate leaves 1,225 adapter calls under deny; the disjoint expert branch leaves none. DeepSeek reproduces the route invariant and gains 27.0 pp. A valid sealed evaluation is near-neutral. These results support auditable, reversible control over a trained parameter path, while showing that useful transfer remains distribution dependent. Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) Cite as: arXiv:2608.06690 [cs.CR] (or arXiv:2608.06690v1 [cs.CR] for this version) https://doi.org/10.48550/arXiv.2608.06690 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-88] Vehicle routing problem using deep reinforcement learning - A case study about truck planning in the industry
链接: https://arxiv.org/abs/2608.06668
作者: Siliang Lu,Dan Hu,Lili Wu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:As an important component of the supply chain industry, transportation has experienced rapid development in the past decade with the assistance of digital platforms and intelligent algorithms. Within the field of transportation research, Vehicle Routing Problem (VRP) has remained a persistent and enduring challenge. In the realm of management science, experts, and scholars from both the industrial and academic sectors have continuously explored optimization models and algorithms to effectively address routing problems, from the classical Traveling Salesman Problem to the more general Vehicle Routing Problem. These models and algorithms are applied in real-world industrial scenarios to achieve cost optimization and reduce carbon footprints. However, due to the complexity of real-world problems, numerous specific constraints are often added, and challenges such as information opacity, uncertainty, and irrational human behavior may arise. Therefore, deploying and optimizing mathematical models for VRP in practical scenarios while maintaining optimal results poses numerous challenges. This paper discusses and provides solutions for three different logistic use cases involving external truck network design. Through these industrial case study, the paper introduces how deep reinforcement learning-based vehicle routing optimization has been implemented. As a result, it can be observed that the routes optimized by reinforcement learning agent have over 10% total cost compared to baseline results. Furthermore, the paper proposes that in future research, DRL algorithms for vehicle routing problems could be generalized into more variations of VRP.
[AI-89] CellWorld: From Gene-Level Reconstruction to Latent Cell Prediction in Spatial Transcriptomics Foundation Models
链接: https://arxiv.org/abs/2608.06659
作者: Haiping Liu,Qian Zhao,Lijing Lin,Jingyuan Sun,Hongpeng Zhou
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:This paper shows that latent-space predictive pretraining can provide a scalable route to foundation models for spatial transcriptomics. Existing spatial transcriptomics foundation models primarily reconstruct masked gene identities or expression values, potentially encouraging the reproduction of assay-specific technical variation and limiting representation transferability. To avoid directly reconstructing such variation, we shift the prediction target from observed gene measurements to latent cell representations and introduce CellWorld, which predicts the latent representations of masked cells from visible spatial context and a limited partial-expression hint. We pretrain four CellWorld variants, spanning 5.74M to 94.56M trainable parameters, on a corpus of 46 million human cells. Our controlled scaling experiments show that performance improves with model capacity, particularly on spatial tasks, while spatial transfer depends more on sufficient optimization and broad biological source diversity than on cell count alone. Across four held-out datasets, even CellWorld-Small, with 5.74M trainable parameters, outperforms every baseline on all 11 linear-probe benchmarks and all seven fine-tuned spatial benchmarks. Most notably, a frozen CellWorld-Large pretrained on only 5% of the corpus with broad biological source coverage outperforms every fully fine-tuned baseline across all seven spatial benchmarks. Code is available at this https URL.
[AI-90] SoRoMoX: Fast Differentiable and Parallelizable Soft Robot Models
链接: https://arxiv.org/abs/2608.06650
作者: Maximilian Stölzle,Solange Gribonval,Daniel Feliu-Talegon,Vito Daniele Perfetta,Michele Martini,Chuhan Zhang,Kiwan Wong,Mohammed Tarnini,Anup Teejo Mathew,Federico Renda,Daniela Rus,Cosimo Della Santina
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Reduced-order models based on Cosserat-rod theory are now well established, and modeling theory is no longer the primary bottleneck in soft-robot control. Their implementations, however, do not support the differentiable, GPU-parallel, and control-oriented workflows that underpin advanced rigid-robotics applications. Here, we fill this gap with SoRoMoX (Soft Robot Models in JAX), a fully numerical, JIT-compilable Python/JAX framework. SoRoMoX implements articulated, Piecewise Constant Strain, and Variable Strain models through a unified, control-ready interface that provides inertia matrices, gravitational and elastic forces, Jacobians, and their derivatives. To our knowledge, it is the first rod/strain-based soft-robot modeling framework that runs directly on GPUs and is end-to-end differentiable with respect to states, inputs, and parameters. Sequential CPU rollouts are up to 18.1x faster than state-of-the-art alternatives, while GPU-parallel rollouts increase throughput by up to 234.6x. This performance enables workflows that were previously impractical or impossible: static-equilibrium system identification with 66% lower marker RMSE; residual-force learning with a further 64% reduction; computed-torque tracking with RMSE reduced by a factor of approximately 500 relative to model-free PD; control-gain optimization with up to 62% lower loss than untuned gains; safety-constrained control using high-order control barrier functions to keep the peak contact force within a prescribed 5 N bound, compared with 33.5 N without the safety constraint; and reinforcement-learning policy training up to 7x faster than a CPU PyElastica discrete-rod baseline through massively parallel rollouts.
[AI-91] Characterizing the Quality Profile of AI-Generated C in Production
链接: https://arxiv.org/abs/2608.06640
作者: Michael Tran,Fred Lewis,Kun Yang,Saksham Thakur,Aditya Kini,Aditya Patil,Milad Hashemi,Parthasarathy Ranganathan
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: 21 pages, 5 figures, 6 tables
Abstract:The widespread integration of AI coding assistants offers undeniable boosts to engineering velocity. Yet, recent studies point to a growing trade-off, revealing persistent challenges with code quality and maintainability. Industry leaders, including frontier AI labs, echo these concerns. As large language models are increasingly relied upon to author production code, understanding their impact on shipped software quality has become a critical priority. However, assessing these effects in industrial workflows remains difficult due to observability barriers. We study the impact of AI-generated code on production quality within a large enterprise operating global products relied upon by billions of users daily. Driven by this scale and user trust, the organization values code quality and has built thorough observability for every line of code deployed into production, enabling us to overcome measurement barriers to assess these effects. This study presents a large-scale empirical analysis of AI-generated C++ code from April 2025 to April 2026, tracking 3.52 million code changes across this enterprise’s brownfield codebase. The core purpose is to understand the quality, performance, and maintenance characteristics of AI-generated code compared to human-written code in a production environment at scale. We find that AI-generated C++ code has a distinct quality profile, showing higher rates of interface and coupling burdens, copy and allocation overheads, and a reliance on explicit loops over optimized standard APIs. These issues translate into tangible downstream costs, including increased review effort and a 5-8% increase in compute resource consumption. However, we demonstrate that providing models with targeted, taxonomy-informed feedback can mitigate these effects, leading to an 11.1% reduction in targeted static analysis warnings and improved computational efficiency. Comments: 21 pages, 5 figures, 6 tables Subjects: Software Engineering (cs.SE); Artificial Intelligence (cs.AI) Cite as: arXiv:2608.06640 [cs.SE] (or arXiv:2608.06640v1 [cs.SE] for this version) https://doi.org/10.48550/arXiv.2608.06640 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-92] MI-MIDI: Mechanistic Interpretability of Text-to-MIDI Generation Models via Probing Lenses and Steering
链接: https://arxiv.org/abs/2608.06638
作者: Jakub Poćwiardowski,Mateusz Modrzejewski
类目: ound (cs.SD); Artificial Intelligence (cs.AI)
备注: 13 pages
Abstract:Mechanistic interpretability of music generation has concentrated on audio models, leaving symbolic models largely unexplored. We analyze two public text-to-MIDI systems of contrasting design: the purpose-built encoder–decoder text2midi and MIDI-LLM, a Llama~3.2~1B model extended with MIDI tokens using linear probing, the logit and tuned lenses, activation patching and difference-in-means steering. Across these methods, we recover musically meaningful structure and show how architecture shapes its formation and control. Pitch, instrumentation, harmony and texture are linearly decodable in both models. text2midi refines predictions gradually across depth, whereas MIDI-LLM works largely in its inherited textual basis before a sharp late rotation into the musical vocabulary; patching identifies a matching late attenuation of prompt-driven instrument transfer. Steering produces bidirectional changes in register and polyphony in both systems, and in tempo/energy in MIDI-LLM. Our two-orientation protocol isolates directional control and shows that all-layer interventions are robust in text2midi but accumulate disruptively in MIDI-LLM. Together, the results provide a practical toolkit for tracing and controlling musical concepts in symbolic generators. Audio examples are available on a demo website.
[AI-93] Bypassing Krum: Selection-Aware Backdoor Attacks in Federated Learning
链接: https://arxiv.org/abs/2608.06637
作者: Srinivasan Subramanian,Md. Abdullah Al Hafiz Khan,Kazi Aminul Islam
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Accepted and presented at the 2026 International Conference on Intelligent Multimedia, Networking, and Security (IMNS 2026). 6 pages, 2 figures, 3 tables
Abstract:Robust aggregation methods are widely used in federated learning to mitigate the impact of adversarial client behavior. Distance-based aggregation rules, such as Krum and Multi-Krum, select updates that are closest to the majority under the assumption that benign updates form a compact cluster. However, these methods rely on geometric properties that can be exploited by adaptive adversaries. We introduce the Krum-Proxy attack, a selection-aware backdoor injection strategy that consistently bypasses Byzantine-robust aggregation. Rather than relying on naive scaling or constraining, our method actively optimizes malicious updates to infiltrate the dense core of the benign distribution. The proposed method constructs adversarial updates that are not only similar to benign updates but are also optimized to lie in regions of the update space that are favored during aggregation. This is achieved through a two-stage optimization procedure that separates task-specific attack objectives from geometry-aware refinement, using a nearest-neighbor proxy, stochastic reference modeling, and anchor-guided alignment. To maintain stealth, we introduce a projection mechanism that constrains adversarial updates within realistic norm and variance bounds. Experiments on standard federated learning benchmarks show that Krum-Proxy achieves higher attack success while preserving clean accuracy, highlighting the vulnerability of distance-based aggregation to selection-aware adversaries.
[AI-94] Shape Your Feed: An LLM -based Agent ic System for Conversational Recommendation RECSYS2026
链接: https://arxiv.org/abs/2608.06632
作者: Ziyun Xu,Bosen Ding,Yue Zhang,Ji Qi,Qingyuan Song,Jizhou Huang,Liwei Wang,Jefferey Santelli,Yue Weng,Qichao Que,Zhenheng Yang,Junfeng Pan,Linhong Zhu
类目: Artificial Intelligence (cs.AI)
备注: Accepted in RecSys 2026 Industrial Track
Abstract:Industrial recommendation systems predominantly adopt a passive ranking paradigm that infers user preferences from implicit behavioral signals (e.g., clicks, dwell time) rather than explicit, natural language inputs. As a result, users experience a persistent discrepancy between their explicit interests and what passive behavioral algorithms deliver, limiting their ability to express nuanced preferences or steer their feed in real time. To address this growing gap between how recommendations are optimized and how users wish to articulate their interests, we present Shape Your Feed (SYF), an LLM-based agentic recommendation framework that enables real-time, multimodal co-curation of content. SYF employs a three-tier architecture: (i) a Perception Flow that captures fine-grained user intent from text prompts, voice commands, and UI interactions; (ii) a Serving Flow that performs real-time agentic re-ranking and pruning of candidate items, grounded in a persistent Semantic Profile encoding evolving user preferences; and (iii) a Self-Evolution Flow that aligns system behavior with human judgments via Direct Preference Optimization (DPO) and an LLM-as-a-Judge ensemble. Offline evaluations show that SYF’s alignment scoring module achieves 98.85% accuracy, substantially improving over strong few-shot baselines. Large-scale online A/B experiments on production traffic further demonstrate that SYF improves feed relevance and user sentiment, indicating a practical and scalable path toward interactive, user-steerable recommendation in industrial settings.
[AI-95] Cryptanalytic Extraction of Isolated Bias-Free GLU Feed-Forward Blocks by Antipodal Separation
链接: https://arxiv.org/abs/2608.06631
作者: Chunhui Shi,Xinwen Fu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 10 pages, 2 figures
Abstract:Cryptanalytic extraction has been demonstrated for ReLU networks, for networks using componentwise activations such as GELU or SiLU, and for a Transformer’s final projection matrix. These methods do not recover the bias-free Gated Linear Unit (GLU) feed-forward blocks used in many modern language models. Such a block multiplies an activated linear projection by a second learned linear projection within each hidden unit, a two-branch structure absent from the network classes and final-layer setting addressed by those methods. We give a constructive, multi-stage forward-query recovery primitive for isolated bias-free GLU blocks. Finite-difference curvature supplies gate-direction candidates, and paired observations at x and -x separate gate magnitude, orientation, and value-branch coupling. Across high-precision targets, six Qwen layers, an 8,192-unit Llama subproblem, and a full-dimensional Gemma block all reach sub-percent median validation error. Four finite-precision configurations remain below 5 percent median error, but none reproduces every stored weight. These isolated-block experiments are not an end-to-end model-API attack: deriving the required internal block responses from final model outputs remains unsolved.
[AI-96] NxN E-valuation: Hypothesis Certification via a Conformal CRT Null
链接: https://arxiv.org/abs/2608.06621
作者: Bin Wang,Yan Zhong
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:We propose NxN E-valuation, a handy, e-value-based hypothesis-certification algorithm that lets a hypothesis be verified without building any case-specific certification procedure—such as constructing a dedicated null hypothesis—as long as a large enough dataset is available. The method is especially suited to LLM-based exploration systems, where LLMs are remarkably good at proposing hypotheses but suffer badly from hallucination; this hallucination prevents us from harvesting LLM outputs directly, and existing remedies each fall short. The most common solutions include letting the LLM verify or correct itself circular verification and held-out testing (where false hypotheses can still pass via spurious correlations), among other remedies detailed in the introduction. To resolve this, NxN E-valuation exploits the naturally existing large training set and lets different samples serve as null hypotheses for one another. This design directly realizes a conditional randomization test (CRT) that certifies each hypothesis. The approach can be a universally better replacement for at least LLM circular verification and held-out-data testing, provided the LLM’s generations are hypotheses that apply to each individual sample.
[AI-97] Automated item evaluation: Predicting item acceptance and rejection using LLM -generated critiques
链接: https://arxiv.org/abs/2608.06609
作者: Hotaka Maeda,Yikai Lu
类目: Artificial Intelligence (cs.AI)
备注: 19 pages, 4 figures, 3 tables
Abstract:Automated item evaluation (AIE) refers to the use of computational methods to assess item quality without requiring manual expert review or field testing of the items under evaluation. We aimed to build a near-comprehensive AIE model by predicting item acceptance and rejection from item text using historical rejection data from a large-scale standardized testing program. The dataset contained 52,759 English language arts (ELA) and mathematics items with 34% permanently rejected from future operational use. Rejection reasons included poor psychometric properties, content issues, bias and sensitivity concerns, and non-content issues. We fine-tuned a DeBERTaV3-large classifier on raw item text, a second DeBERTa classifier on Qwen3-generated item critiques, and a fusion model combining representations from both. The fusion model achieved the strongest overall performance (Accuracy = .75, F1 = .64, AUC = .80, Sensitivity = .64, Specificity = .81). Prediction for math (F1 = .73, AUC = .86) was considerably more accurate than ELA (F1 = .51, AUC = .72). Lowering the decision threshold from .5 to .25 raised average sensitivity for ELA and math to .88 and .91, while reducing specificity to .31 and .56, respectively, which may be preferable in automated item generation contexts where generating items is cheaper than evaluating them. Incorporating item critiques alongside raw item text improved performance across most rejection reasons. The model assigned higher rejection probabilities to more difficult items. However, the fusion model struggled to identify items flagged for bias, sensitivity, fairness, or accessibility, especially for ELA. These findings suggest that text-based AIE is feasible in some areas and may offer a practical tool for reducing the burden of manual review and field testing, while also underscoring the importance of human review for items with fairness concerns.
[AI-98] Flowing Through States: Neural ODE Regularization for Reinforcement Learning
链接: https://arxiv.org/abs/2608.06595
作者: Mohamed Ghanem,Bernd Finkbeiner
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Neural networks applied to sequential decision-making tasks typically rely on latent representations of environment states. While environment dynamics dictate how semantic states evolve, the corresponding latent transitions are usually left implicit, creating a potential misalignment between the two. We propose to model latent dynamics explicitly by drawing an analogy between Markov decision process (MDP) trajectories and ordinary differential equation (ODE) flows: in both cases, the current state fully determines its successors. Building on this view, we introduce a neural ODE-based regularization method that enforces latent embeddings to follow consistent ODE flows, thereby aligning representation learning with environment dynamics. Although broadly applicable to deep learning agents, we demonstrate its effectiveness in reinforcement learning by integrating it into Actor-Critic algorithms. Our approach yields major performance gains across various standard Atari benchmarks for A2C and gridworld environments for PPO.
[AI-99] SyncSBC: Decentralized Swarm Behavior Prediction for Synchronized Autonomous Control IROS2026
链接: https://arxiv.org/abs/2608.06587
作者: Varun Raveendra,Connor Mattson,Daniel S. Brown
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: 8 pages, 10 figures, IROS 2026
Abstract:Robot swarms utilize many independent limited-sensing agents to produce complex emergent behaviors without requiring centralized control. However, little research explores how agents can infer swarm-level behavior from purely local perception, a capability critical for detecting faults and behavior changes. In this paper, we introduce Synchronized Swarm Behavior Classification (SyncSBC), which combines improvements in machine learning and distributed consensus to classify collective swarm behavior and synchronize swarm decision-making in an entirely decentralized manner. We show that SyncSBC achieves high classification accuracy and low synchronization delay, making it suitable for real-world deployment. Finally, we use SyncSBC to demonstrate two promising swarm applications on real robots where we show that swarms utilizing SyncSBC can accurately identify anomalies in robot behavior and autonomously coordinate collective changes in swarm behavior. Videos, code and supplemental experiments are available at this https URL.
[AI-100] Divergent Response Modes in Frontier Language Models Under Steering Pressure
链接: https://arxiv.org/abs/2608.06578
作者: Ali Jalal-Kamali
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Frontier language models are trained using distinct data, objectives, and safety pipelines. Whether these differences produce measurably different behaviors under explicit steering pressure remains underexplored. This study evaluates behavioral steerability across six frontier models from six developers using 300 paired base and steered items over three categories: values-conflict, reasoning-elicitation, and reasoning-suppression (plus 40 validation items). All six models act as blind peer judges and classify every response based on fixed behavioral rubrics. The resulting 24,480 judgments are scored by leave-one-out consensus. We find that models differ not just in how much steering shifts their behavior but in what kind (mode) of response they give, and some response modes appear in only one or two of them. GPT-5 deflects requests to disclose its reasoning while leaving its answer intact (99% vs. 0% for all other models). Claude Opus 4.7 and GPT-5 resist explicit suppression instructions and in different ways. Using Llama as the open-weight model, we trace the largest behavioral split to its internals. A linear probe decodes the behavior from the residual stream at 0.87 held-out accuracy while injecting that direction during generation drives the behavior from 0% to 86% across an intervention sweep. Every finding holds under both a token-budget remediation and a control experiment with a hypothesis-blind judgment prompt.
[AI-101] KNOWPLAN: Knowledge-Driven AI Agents for Smart Degree Pathway Planning
链接: https://arxiv.org/abs/2608.06530
作者: Shuheng Cao,Weijia Zhang,Jiaqi Wu,Xiyun Hu,Yat Yang,Juqy Chen,Zhaoxiang Feng
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Planning a degree from official university sources requires solving two problems in order. The institution’s curriculum must first be reconstructed from catalogs, departmental pages, JSON endpoints, and PDFs that share no schema, and only then can a student-specific path be optimized under prerequisite logic and overlapping requirement constraints. Coupling the two lets each failure mode hide the other, because a planner that drives its own crawling never learns facts its current plan does not need. We present KnowPlan, which enforces an extraction-first boundary and measures the interface between the stages rather than assuming it. CatalogBrowse explores with no access to any user profile. It scores legal actions by lower-confidence expected marginal gain over a finite set of atomic catalog obligations per unit of source access, parses deterministically through platform adapters with a span-constrained clause-to-AST model fallback, and terminates on a closure certificate over index, schema, provenance, and reference completeness instead of a reward threshold. Its output contract is three provenance-linked JSON documents. DegreeMap consumes only those documents. It compiles them into a typed requirement hypergraph and optimizes lexicographically with CP-SAT over hard feasibility, completion horizon, load and risk, personalized utility, and option value, so that each stage optimizes inside the previous stage’s proven optimum and stays certifiable within the solver budget. Across a 100-university broad track and a six-school dense track, CatalogBrowse reaches 96.2% inventory recall and 88.7% masked-source recovery at 47% less source access than an exhaustive crawler, DegreeMap holds 100.0% hard feasibility while improving personalized utility by +0.066 over the strongest baseline, and the full pipeline certifies 99.5% of requests with a utility gap to the privileged gold graph of 0.015.
[AI-102] CertBind from Multimodal Connectivity to Certifiable Retrieval Decisions
链接: https://arxiv.org/abs/2608.06516
作者: Shuheng Cao,Zhenhao Zhang,Ruiqi Chen,Renjie Cao,Weijia Zhang,Siyu Zhang,Jiaxin Liu,Xiangyu Zeng,Haotian Geng,Fan Gu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Lightweight connectors make frozen multimodal encoders composable at the representation level. Deployment exposes a second problem at the level of task decisions. A connected route can expand cross-modal reach while changing an established native retrieval capability. We introduce CertBind, a multiscale theory of certifiable composition for frozen multimodal connector graphs. At the node scale, native anchors establish the exact task identification boundary under the stated chart model. At the edge scale, contract-aware conformal ranks provide graph-wide family-wise error control. At the path scale, an overlap-aware budget and clean calibration yield a finite-sample recovery radius under declared conditions. At the query scale, this radius yields a covered top-k candidate set that becomes a point certificate when its size equals k. CertBind therefore retains supported routes as Direct, sends only flagged routes to recovery, returns Certified for decisive recovery, and returns Abstain for unresolved queries. The evaluated C-MCR shared route reduced native CLIP R@1 from 0.524 to 0.290. The production fallback recovered 0.963 ± 0.002 of clean retrieval, while the passing branch recorded a no-harm value of 1.000. CertBind extends multimodal composability from connected representations to certifiable task decisions.
[AI-103] LyEvO: Lyapunov-Guided Evolutionary Optimization for Safe and Robust Sim-to-Real Policy Learning
链接: https://arxiv.org/abs/2608.06481
作者: Riccardo Curcio,Hongpeng Cao,Marco Caccamo
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
备注:
Abstract:Training controllers that are safe and robust in simulation, and systematically assessing their readiness for real-world deployment, remain key challenges in sim-to-real transfer. To address this, we propose LyEvO, a physics-grounded framework that combines constrained Evolutionary Optimization and Statistical Model Checking (SMC)-based verification with Lyapunov-based stability analysis. Leveraging prior knowledge of the system dynamics, LyEvO uses Lyapunov analysis to compute an initial candidate stability region. An iterative loop then uses operational scenarios drawn from this region to jointly optimize and statistically verify a policy, and subsequently expands the region’s boundaries based on the verification outcome. This integrated procedure provides a practical criterion for assessing deployment readiness. We evaluate LyEvO on Cartpole and 3D Quadrotor benchmarks through extensive simulations and targeted real-world experiments, demonstrating safe and robust sim-to-real transfer.
[AI-104] WebGrader: Training LLM s for Web Development with Self-Evolving Programmatic Grader
链接: https://arxiv.org/abs/2608.06474
作者: Boshui Chen,Huiping Liu,Shaolei Zhang
类目: Artificial Intelligence (cs.AI)
备注: 17 pages, 3 figures. Supplementary material is included in the main PDF
Abstract:Large language models increasingly generate complete websites from natural-language descriptions, and reinforcement learning has become a central approach to closing their remaining functional gap. This training regime is bottlenecked by reward design. Hand-authored browser scripts are executable yet costly to write for open-ended requirements, while VLM and GUI-agent graders scale but may issue verdicts before observing the decisive state. We propose WebGrader, a self-evolving programmatic grader that autonomously derives the required interaction flows from each website request, represents each flow as an executable Flow Contract, and uses its execution outcome as an RL reward. WebGrader materializes the generated project in a live browser, grounds target actions against the source code and live DOM, and collects visual, DOM, response, and persistent-state evidence along the same browser trajectory. A residual-driven offline loop then discovers reusable verifier skills, screens them on disjoint validation pages, and freezes the promoted skill graph before policy training. By separating test planning, action grounding, evidence collection, and semantic judgment, WebGrader issues a Pass verdict only after observing the requested transition. On WebGen-Bench, WebGrader trains an 8B policy to a 52.01% functional success rate, outperforming a matched appearance-plus-script reward by 7.88 points and surpassing o4-mini and DeepSeek-v4-flash. On WG-core-250, the policy reaches a Full Score of 44.953 and surpasses Qwen3-Coder-480B.
[AI-105] CyberForge: Verified Vulnerability Injection at Repository Level for Cybersecurity Agent Training
链接: https://arxiv.org/abs/2608.06471
作者: Amine Lbath,Manan Suri,Aurelien Delaitre,Vadim Okun,Massih-Reza Amini,Ram D. Sriram,Dinesh Manocha
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注:
Abstract:Despite recent advances, frontier large language model (LLM) agents remain limited in discovering and patching complex vulnerabilities in real-world software. Generally available agents can already aid attackers, who only need to find one exploitable weakness, while defenders must continuously identify and patch all vulnerabilities across fast-growing codebases. Stronger defensive agents would help close this gap, yet the scarcity of security training data with reproducible build and execution environments remains a bottleneck. We present CyberForge, a framework that synthesizes executable, repository-level security training data by injecting vulnerabilities into real C/C++ projects. It validates each instance dynamically: the injected build must pass the project’s unit tests, and generated proof-of-vulnerability (PoV) must trigger on the injected build and not on the clean one. CyberForge is not limited by the availability of disclosed vulnerabilities, therefore it can scale in comparison to data augmentation techniques which rely on historic CVE data. The resulting corpus holds 1034 validated vulnerabilities across 80 projects and 63 weakness categories, with edit locality similar to real CVE patches under a real-versus-real noise floor. Fine-tuning on trajectories collected over this corpus improves SEC-bench patch repair by +3.3 to +14.7 points, in all six configurations of three model scales and two teachers, with the 31B student reaching its GPT-5.4-mini teacher, 72.7% against 74.0%. These gains generalize out of distribution to PatchEval, a corpus containing other programming languages, where every configuration also improves and the 31B student passes its teacher. Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Software Engineering (cs.SE) Cite as: arXiv:2608.06471 [cs.CR] (or arXiv:2608.06471v1 [cs.CR] for this version) https://doi.org/10.48550/arXiv.2608.06471 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-106] ED-CSP: Crystal Structure Prediction from Electron Diffraction
链接: https://arxiv.org/abs/2608.06448
作者: Germain Poloudenny,Yaël Frégier,Arnaud Demortière
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 9 pages
Abstract:Recovering a periodic 3D crystal structure from sparse, unindexed electron diffraction (ED) observations is a challenging generative inverse problem. Existing ED-based learning methods mainly predict crystallographic labels, reconstruct structures from indexed reflections, or retrieve candidates from finite structure libraries. Here, we introduce ED-CSP, a machine learning framework that predicts crystal structures from chemical composition, atom count, and multiple detector-plane ED spot sets. ED-CSP combines a relational set encoder, permutation-invariant multi-view aggregation, and a periodic flow generator to jointly predict lattice parameters and fractional atomic coordinates. To train the model, we construct ED-CS, a dataset of 4.85 million simulated multi-view ED crystal structures, deduplicated across seven materials repositories and filtered to exclude CHILI-100K overlaps. On 2,075 held-out CHILI-100K materials, ED-CSP trained only on CHILI achieves a structural match rate of 57.49% MR@5, outperforming PXRDGen (52.92%), a state-of-the-art crystal structure prediction model conditioned on powder X-ray diffraction. Scaling training data further improves performance: initializing from a one-million-structure precursor raises MR@5 to 66.27%. On 1,024 compositions absent from the training retrieval library, the model still achieves 53.52% MR@5, demonstrating true generative capability beyond exact-formula retrieval. Replacing target ED observations with diffraction from non-isomorphic structures of identical composition decreases MR@5 by 22.09 percentage points, confirming that predictions depend on the input diffraction patterns rather than composition alone. ED-CSP and ED-CS establish a benchmark for generative crystal structure prediction from sparse ED observations and provide a foundation for future transfer to experimental data. Comments: 9 pages Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI) Cite as: arXiv:2608.06448 [cs.LG] (or arXiv:2608.06448v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.06448 Focus to learn more arXiv-issued DOI via DataCite
[AI-107] Risk-Aware Decision Policies for Agents Under Noisy Perception
链接: https://arxiv.org/abs/2608.06420
作者: David Szczecina
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 8 pages, 6 figures. Submitted to the 2026 Conference on Artificial Life (ALIFE2026)
Abstract:Perception in biological systems is inherently noisy, requiring organisms to make decisions under uncertainty where misclassification can be costly or fatal. We present an Artificial Life predator-prey model of foraging under noisy perception, and compare agent performance when using various policies that take into account their noisy predictions. Through controlled experiments under both symmetric and asymmetric perceptual noise, we show that blindly trusting perceptual labels leads to catastrophic failure as noise increases, while uncertainty-aware strategies significantly improve survival and reduce fatal errors. We further observe qualitative regime shifts in behaviour, with agents transitioning from exploratory to conservative strategies as uncertainty increases. Our model links risk-sensitive foraging, ecological information use, and Artificial Life by showing that explicit information gathering can improve robustness when perception is unreliable. These results highlight the importance of uncertainty-aware decision-making and provide an interpretable artificial life analogue to robust learning with noisy labels.
[AI-108] WorldMark: A Plug-and-Play World Knowledge Interface for Cross-Host Language Model Watermarking
链接: https://arxiv.org/abs/2608.06416
作者: Song Xiao,Yuqi Yuan,Yanshuo Zhang,Kejun Zhang
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:Watermarking traces the provenance of text produced by large language models by embedding statistically detectable signals during decoding. Existing schemes fall into logits-based, sampling-based, entropy-aware, and adaptive-strength families, yet all of them place watermark signals according to local token statistics. In the open-ended text-generation settings evaluated in this work, local statistics may provide insufficient guidance for placing robust watermark signals. We introduce WorldMark, a plug-and-play interface that uses World Knowledge Memory (WKM) to organize semantic and episodic knowledge in a memory graph, converts the retrieved knowledge into a token-level knowledge saliency score, and adjusts the strength of a host watermark through Asymmetric Knowledge Modulation (AKM). WorldMark requires no backbone retraining and introduces no additional detector-side model or parameter. On the primary C4 evaluation, the complete WorldMark interface improves clean and attacked detection across three adaptive-strength host variants while slightly reducing perplexity. Additional pilot experiments on C4 and OpenGen show that direct memory conditioning transfers across multiple watermark families but can be unstable without saliency-aware modulation. WorldMark requires no additional detector-side model or parameter and introduces negligible overhead under the primary protocol.
[AI-109] Interpretable Unsupervised Community Detection with LLM -Symbolized Structured Processes
链接: https://arxiv.org/abs/2608.06402
作者: Aoting Zeng,Kai Wang,Jianwei Wang,Yuxiang Sun,Yizhang He,Wenjie Zhang
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Community detection is a fundamental task in graph analytics that aims to identify cohesive groups of entities with similar behaviors or interests. Classic objective-driven methods struggle with complex graph structures, while deep-learning approaches improve performance at the expense of interpretability and rely on labeled data and training. Large language models (LLMs), with strong reasoning capabilities and world knowledge, are promising for interpretable, label-free community detection. To leverage these strengths, we propose LUCID, an LLM-guided, interpretable, training-free, and unsupervised community detection method. Inspired by phase-transition kinetics in natural systems, where complex structures emerge through initialization, merging, refinement, and selection, LUCID is designed as a four-stage pipeline. Within this pipeline, the LLM induces formal rules that translate implicit knowledge into explicit and interpretable logical structures. Specifically, (1) the Local-View Community Initialization stage encodes local graph structures using k-ego contexts and unsupervised node roles; (2) the Multi-factor Community Merge stage uses LLM-induced rules to iteratively merge local communities; (3) the Multi-grain Community Refinement stage applies LLM-induced coarse-to-fine rules in parallel to reduce boundary noise; and (4) the Global-view Community Selection stage identifies high-quality communities based on topological compactness and boundary clarity. Extensive experiments on real-world datasets demonstrate that LUCID, as an unsupervised approach, achieves state-of-the-art performance and consistently outperforms leading unsupervised and semi-supervised baselines.
[AI-110] Beyond Routing Weights: Faithful Response-Level Interpretation of Mixture-of-Experts Reward Models via Contribution Contrast
链接: https://arxiv.org/abs/2608.06400
作者: Yifan Wang,Jinyi Mu,Mayank Jobanputra,Yu Wang,Soyoung Oh,Isabel Valera,Vera Demberg
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Reward models are central to learning from human preferences, yet identifying what drives their predictions remains challenging. Recent sparse Mixture-of-Experts (MoE) reward models seek to improve interpretability by routing prompts to specialized experts and characterizing experts through examples with high routing weights. However, routing weights only reveal which prompts an expert \textitreceives , not how it \textitjudges responses, providing only a partial account of expert behavior. We therefore propose \textbfCo ntribution- \textbfCo ntrast ( \textbfCoCo ) response-level interpretation, which faithfully characterizes experts’ roles using chosen-rejected response pairs with the largest contribution contrasts, jointly capturing routing and preference behavior. Across automatic and human evaluations, CoCo yields more coherent, faithful, and specialized interpretations than router-based, score-based, and sparse autoencoder-based alternatives while maintaining competitive reward modeling accuracy. To the best of our knowledge, this is the first systematic study of interpretation methods for MoE reward models.
[AI-111] Recovering Explanations from Transformed Rule-Based Ontologies
链接: https://arxiv.org/abs/2608.06399
作者: Alex Ivliev,Markus Krötzsch,Maximilian Marx
类目: Logic in Computer Science (cs.LO); Artificial Intelligence (cs.AI); Databases (cs.DB)
备注:
Abstract:Datalog rules are often used to define ontologies over Knowledge Graphs. Rule reasoners routinely optimise such ontologies by rewriting their rules into a form that can be evaluated more efficiently. These transformations preserve the entailed facts, but not the structure of the underlying derivations. A proof tree under the rewritten rules explains why a fact holds, but does not readily yield an explanation in terms of the original rules. We study the problem of constructing, from a proof of entailment under the rewritten rules, a proof under the original ones: we establish its computational complexity and identify two practically relevant languages for specifying proof transformations.
[AI-112] EntropyMoE: Entropy-Aware Sparse Expert Routing for Tokenizer-Free LLM s
链接: https://arxiv.org/abs/2608.06398
作者: Bo Liu,Muxuab Yu,Yu Zhang,Pengfei Gao,Yongping Zhang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Recent byte-level large language models (LLMs) have made tokenizer-free modeling increasingly competitive by grouping bytes into dynamically sized patches. However, existing byte-patch architectures still apply the same dense feed-forward computation to every patch. This uniform computation cannot adapt model capacity to variations in patch semantics and granularity. We address this limitation with EntropyMoE, a Mixture-of-Experts (MoE) architecture designed for dynamic byte patches. EntropyMoE replaces the dense feed-forward modules in the global patch Transformer with Top-K expert layers. Each dynamic patch serves as the basic unit of expert routing, and its byte coverage determines its contribution to workload accounting. The router selects experts directly from patch entropy, using the same granularity signal that underlies dynamic patch construction to organize sparse computation. Patch entropy and length jointly define the feature space for regulating expert specialization. Experiments show that EntropyMoE achieves the lowest held-out bits-per-byte among matched dense and sparse baselines while maintaining comparable downstream accuracy. These results establish patch entropy as an effective routing coordinate for sparse conditional computation and extend Mixture-of-Experts modeling beyond tokenizer-based representations.
[AI-113] Agent ic Planning for Symbolic Execution
链接: https://arxiv.org/abs/2608.06397
作者: Daniel Koh Ji Yang,Yannic Noller,Corina S. Pasareanu,Youcheng Sun
类目: Programming Languages (cs.PL); Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注:
Abstract:Symbolic execution seeks to explore feasible program paths, yet a practical run may exhaust its resources while much program behaviour remains unreached. We investigate a complementary way of extending its practical reach by reasoning about how the same tool is utilised from one bounded run to the next, while leaving ordinary state exploration to the underlying tool. We present Agolic, an agentic planning system that uses evidence from earlier runs to choose and configure later bounded symbolic execution (BSE) runs, which the underlying symbolic execution tool then carries out. The planning intelligence, available evidence and execution modes can be adapted to the symbolic execution tool and analysis objective. We evaluate one adaptation for branch-coverage exploration, in which an LLM-based agent reasons over source code, replayed coverage and earlier targeting attempts. We evaluate Agolic on several C and C++ programs. On every program, it extends the branch coverage obtained by continuous symbolic execution and covers more than 3\times as many branches on average. It also covers more branches than each individual corpus from coverage-guided fuzzing and compiler-based concolic execution in our evaluation and reaches branches absent from all comparison corpora combined on six of the seven programs. Taken together, these results point to considerable untapped potential in existing symbolic execution tools, some of which may be realised by reasoning about how their capabilities are used across runs while leaving state selection during ordinary symbolic exploration to the underlying tool.
[AI-114] owards Multi-Label Graph Foundation Models: from Single-Vector Representation Learning to Multi-Semantic Basis Learning
链接: https://arxiv.org/abs/2608.06394
作者: Dongxiao He,Jiayu Zhang,Jitao Zhao,Yi Wang,Di Jin
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Multi-label node classification is an important yet challenging task in graph learning, where nodes exhibit multiple semantics simultaneously. Existing methods for multi-label node classification can effectively model multiple labels, while only considering in-domain scenarios where the model needs to be trained and tested within the same graph domain, resulting in limited cross-domain generalization. Recently, Graph Foundation Models (GFMs) have emerged as a promising paradigm for learning transferable graph representations across diverse graph domains and downstream tasks. However, existing GFMs are built upon single-label assumption, where all nodes are arbitrarily regarded as containing only one class of semantic and embedded into a single representation. For multi-label nodes, such a representation essentially approximates multiple semantics with a single point in the representation space, inevitably leading to semantic entanglement and making simultaneous discrimination of multiple labels difficult. To address these limitations, we propose a Multi-Semantic Basis Graph Foundation Model (MSB-GFM), a framework for cross-domain multi-label node classification. Specifically, we introduce a multi-semantic basis representation learning paradigm that models each multi-label node as an adaptive composition of semantic bases, thereby enabling flexible representational capacity for modeling multiple semantics. Furthermore, we develop a semantic-structure dual-channel architecture with domain adversarial training for effective cross-domain knowledge transfer. Extensive experiments demonstrate the effectiveness of our model.
[AI-115] LSEAD: A Privacy-Preserving LLM -Based Speech Analysis Framework for Early Alzheimers Disease Screening
链接: https://arxiv.org/abs/2608.07378
作者: Xin Wang,Yingchao Huang,Yuhan Su,Shanshan Yao,Wei Peng
类目: Audio and Speech Processing (eess.AS); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Early diagnosis of Alzheimer’s disease (AD) is critical for enabling timely interventions that may slow disease progression and improve patient outcomes. There is a growing need for AD detection methods that are non-invasive and cost-effective, especially in real-world clinical settings with diverse patient populations and recording conditions. Speech-based screening addresses these needs by using natural speech collected without specialized equipment. Recent advances in large language models (LLMs) have improved speech analysis by providing rich linguistic representations and strong generalization. In this study, we propose LSEAD, a speech-based AD detection framework using pretrained open-source LLMs. Speech recordings are automatically transcribed, and text embeddings are extracted using locally deployed LLMs. Principal component analysis (PCA) is applied to reduce dimensionality before classification. Because the framework relies only on speech transcripts and locally deployed models, it supports privacy-preserving AD risk assessment without external data exchange. We evaluate LSEAD on the ADReSS20 and ADReSSo2021 benchmark datasets. Experimental results show that LLM-based embeddings generalize well across datasets and improve AD classification accuracy by up to 5 percent over existing methods, especially for early-stage detection. These results demonstrate that LSEAD provides a practical, secure, and scalable approach for early AD screening.
[AI-116] Assessing AI-generated music detection in real-world broadcast monitoring
链接: https://arxiv.org/abs/2608.07359
作者: David López-Ayala,Fernando García de la Cruz,Pablo Zinemanas,Emilio Molina,Martín Rocamora
类目: Audio and Speech Processing (eess.AS); Artificial Intelligence (cs.AI)
备注: Accepted for ISMIR 2026
Abstract:The proliferation of AI-generated music in broadcast media raises concerns about transparency and fair compensation, but reliable detection under real broadcast conditions remains unresolved. Existing studies report substantial performance degradation in this domain, yet their evaluations are limited to synthetic broadcast data. To address this gap, we introduce BAMM (Broadcast AI-Music Monitoring), a 40-hour dataset of real-world television recordings containing AI-generated and human-made music. We compare clean-trained and broadcast-trained CNN variants across three progressively more challenging scenarios: Clean Foreground Music (CFM), Synthetic TV Broadcast (STB), and Real TV Broadcast (RTB). Both models achieve near-perfect performance on CFM but degrade substantially under synthetic broadcast conditions. Broadcast-oriented training improves robustness compared with clean training, although performance remains limited. On RTB, evaluated using BAMM, both models degrade further and show substantial score overlap between AI-generated and human-made music. These results expose a critical domain gap and show that current training approaches on CNN-based detectors remain insufficient for reliable AI-generated music detection in broadcast monitoring.
[AI-117] How Much AI Is in This Track? Quantifying the Proportion of AI-Generated Stems in Hybrid Music Mixtures
链接: https://arxiv.org/abs/2608.07285
作者: Fernando Garcia de la Cruz,David López-Ayala,Pablo Zinemanas,Emilio Molina,Martín Rocamora
类目: Audio and Speech Processing (eess.AS); Artificial Intelligence (cs.AI); Signal Processing (eess.SP)
备注: Accepted at International Society for Music Information Retrieval Conference (ISMIR 2026)
Abstract:AI-generated music is increasingly used at the stem level, with producers integrating synthetic drums, basslines, or vocals alongside human-performed instruments. However, current AI music detection systems are binary, treating tracks as either fully AI or fully human. In this paper, we reformulate AI music detection as a regression problem on a continuous AI energy ratio, alpha in [0, 1]. We propose a methodology that leverages a multi-track music dataset to assemble mixtures of human-performed and AI-reconstructed stems (obtained using a neural audio codec) with known proportions of each content type. Using this approach, we first show that a CNN-based model trained on fully AI-generated or human-performed tracks, which achieves 99% accuracy as a binary detector, when faced with mixed content, yields an output that rises with the AI stems’ energy contribution, acting as a noisy and miscalibrated estimator. Our analysis of the influence of different stems shows that detection sensitivity depends on the instrument and reflects its frequency content: drums and guitar carry strong codec-artifact signatures, while vocals and bass are less detectable. Based on these insights, we train a similar CNN-based model for regression of alpha, achieving MAE = 0.076 and R^2 = 0.85 on held-out mixtures from the same pipeline. These results suggest that the regression formulation is an initial promising step towards AI-music detection in realistic music production workflows.
[AI-118] A Finite E-Group of Nilpotency Class Three
链接: https://arxiv.org/abs/2608.07275
作者: Xinan Dai,Wenhao Deng,Yidong Shi,Tailin Wu,Yuchen Yang
类目: Group Theory (math.GR); Artificial Intelligence (cs.AI)
备注: 9 pages
Abstract:A group is an E-group if every element commutes with each of its endomorphic images. Caranti asked whether a finite E-group can have nilpotency class three. We prove that the 3 -group of order 3^84 introduced by Abdollahi, Faghihi, and Mohammadi Hassanabadi, and later shown by Abdollahi, Faghihi, Linton, and O’Brien to have the corresponding automorphism property, is an E-group. Let P denote this group and put V=P/\Phi§\cong \mathbbF_3^9 . The nine power relations of P determine a linear map q:V\longrightarrow\Lambda^2 V . We prove that q has no nonzero proper subspace U satisfying q(U)\subseteq\Lambda^2 U . Since the image induced by any endomorphism of P on V has precisely this closure property, every endomorphism acts on V either invertibly or trivially. The invertible case is the known A-group case. In the trivial case the image first lies in \Phi§=P’ , and the power relations then force it into \Omega_1(P’)=Z§ . Thus every element commutes with every endomorphic image. The tensor rigidity is reduced to an exact finite calculation on the 9841 points of \mathrmPG(8,3) .
[AI-119] Reading Copoms Tone: A Weighted LLM Framework for Hawkish-Dovish Sentiment Forward Guidance and Uncertainty
链接: https://arxiv.org/abs/2608.07251
作者: Gabriel de Macedo Santos
类目: General Economics (econ.GN); Artificial Intelligence (cs.AI)
备注: 12 pages, 8 tables, 5 figures
Abstract:This paper documents an applied natural-language-processing framework for measuring the tone of Brazilian Monetary Policy Committee (Copom) statements. The project is explicitly inspired by iSent, Itaú’s Central Bank sentiment classifier, particularly its sentence-level division of official communication into hawkish, dovish, neutral, and out-of-context classes. The implementation extends that idea in three directions. First, an LLM identifies short hawkish and dovish expressions and assigns each a 0-to-1 intensity weight. Second, the document index combines sentence counts with document-specific average signal intensities, producing a bounded score from -1 to 1. Third, a separate full-document layer measures forward-guidance direction, guidance explicitness, uncertainty level, and change in uncertainty. The empirical sample is restricted to communications dated August 2016 or later and contains 80 statements and 1,498 classified sentences from August 31, 2016 through August 5, 2026. Across this sample, 33.3% of sentences are hawkish, 18.0% dovish, 42.1% neutral, and 6.5% out of context. The average document score is +0.107, while the most hawkish reading is +0.570 in August 2021. The latest statement, dated August 5, 2026, scores +0.232, with eight hawkish, two dovish, and nine neutral sentences. Its structural overlay is more nuanced: guidance is directionally ambiguous but partly explicit, while uncertainty is classified as central and higher than at the prior meeting. Tone and the guidance-direction score have a contemporaneous Pearson correlation of 0.719. These are descriptive outputs, not a validated forecast of Selic decisions or DI returns. The main contribution is therefore methodological: a transparent, incremental, auditable system that separates rhetorical tone from policy guidance and uncertainty.
[AI-120] Autonomous discovery of accelerator commissioning algorithms
链接: https://arxiv.org/abs/2608.07138
作者: Thorsten Hellert(Lawrence Berkeley National Laboratory)
类目: Accelerator Physics (physics.acc-ph); Artificial Intelligence (cs.AI)
备注: 9 pages, 4 figures. Submitted to Physical Review Accelerators and Beams (ZVR1001)
Abstract:Simulated commissioning has become essential for de-risking modern light-source design and commissioning, but the procedures being simulated are still designed entirely by human experts. Their labor-intensive redevelopment after lattice changes makes such studies hard to repeat and limits their use during early design iteration. This Letter demonstrates a closed research loop in which a language-model agent writes commissioning code, tests it in simulation, and improves the algorithm from the results. Applied to RF beam capture in the ALS-U accumulator-ring model, the loop substantially improves a working expert procedure and can construct a working one from a minimal starting point, with more capable models succeeding from less initial code. Extending the same framework to multiple objectives produces 16 non-dominated algorithms spanning physically distinct trade-offs between rapid beam capture and correction of seeded machine errors. This reframes commissioning studies from evaluating human-designed procedures toward a mode in which agents participate directly in discovering accelerator algorithms.
[AI-121] Same physical state different collective dynamics: state encodings select synchronization outcomes in language-model agents
链接: https://arxiv.org/abs/2608.06968
作者: Takahiro Ezaki,Naoto Imura,Katsuhiro Nishinari
类目: Physics and Society (physics.soc-ph); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注:
Abstract:Language-model agents act on state encodings of their environment, yet these are treated as interchangeable interfaces. Using pretrained language models, we designed a circular-synchronization experiment applying a state-encoding intervention while holding the physical system fixed: each agent sees only a summary of its neighbours’ relative phases and chooses to advance, stay or retard. Encoding that state as low-order circular moments rather than as a histogram selected different collective outcomes. In GPT the moment encoding synchronized the population in 6/6 seeds and the histogram encodings in 0/6; the effect replicated in Claude but reversed direction. Replaying identical fields shifted each agent’s advance/stay/retard probabilities far beyond within-encoding repeat variation, in GPT, Claude and Gemini; in GPT, presentation alone shifted the operator with the moment values fixed. State encodings therefore form part of a model-dependent effective interaction law, not a neutral interface.
[AI-122] Investigating Quantum-Embedded Transformers on Classical Datasets for Cross-Modality Classification
链接: https://arxiv.org/abs/2608.06846
作者: Hao-Yuan Chen
类目: Quantum Physics (quant-ph); Artificial Intelligence (cs.AI)
备注:
Abstract:We test whether a parameterized quantum circuit (PQC) improves a hybrid quantum-classical model’s performance on classical datasets, using an interface-matched classical map as the control while holding all other components fixed. Our architecture, Quantum-Embedded Attention (QEA), uses a learnable projector to compress backbone features into an n_q -dimensional angle vector, a shallow PQC to map those angles to one- and two-qubit Pauli expectations, and a classical attention decoder to produce class logits. We hypothesized the PQC would improve accuracy or seed-to-seed stability over a classical map with matched input/output dimensions. We test this with an interface-matched 2\times2 factorial on Breast Cancer Wisconsin at n_q\in\4,8\ , independently swapping the PQC for a classical map and the attention decoder for a linear head, across five paired seeds per cell. Three of four paired quantum-minus-classical 95% confidence intervals include zero; the fourth, a +1.63 percentage-point contrast for the attention decoder at n_q=4 , reverses sign at n_q=8 and does not survive correction across the four contrasts. The experiment thus shows no consistent PQC contribution and cannot establish equivalence. A five-dataset cross-modality grid shows comparable accuracy on AG~News, Breast Cancer Wisconsin, and BirdCLEF but a large deficit on CIFAR-10; these cells are not interface-matched and are interpreted descriptively. We report all planned canonical runs, distinguish current Pauli-readout results from legacy probability-readout experiments, and analyze bottleneck, simulation, finite-shot, and noise limitations. The results do not establish a quantum advantage; they demonstrate why controlled component attribution is necessary before crediting a hybrid model’s performance to its quantum layer.
[AI-123] Control-Anchored Residual Flow Matching Conditioned on Gene Geometry for Virtual Cell Perturbation Modeling
链接: https://arxiv.org/abs/2608.06824
作者: Quanquan Li,Yihe Chi,Liuyang Song,Hongbo Zhang,Jingyu Li,Xidong Xi,Conghua Wei,Yijie Sun,Yu Chen,Xin Liu,Qi Hu,Jing Ke,Guitao Cao
类目: Molecular Networks (q-bio.MN); Artificial Intelligence (cs.AI)
备注:
Abstract:A central task in virtual cell modeling is predicting single-cell transcriptional responses to unseen genetic perturbations and drug combinations, and biological networks provide valuable priors on gene relationships. Existing graph-based models commonly use the same network to structure gene representations and mediate intergene interactions, thereby implicitly treating stable associations as perturbation-response pathways. Gene Ontology and control-derived coexpression networks encode relatively stable relationships rather than intervention-specific response directions or magnitudes. We therefore propose GeneGeoFlow, which conditions a control-anchored residual flow on gene-wise geometry derived from biological networks to learn intervention-specific transcriptional responses. GeneGeoFlow derives multi-scale spectral coordinates from Gene Ontology and control-derived coexpression networks. A perturbation-conditioned, gene-wise gating module selects relevant structural scales and network sources, yielding intervention-specific gene geometry. The resulting geometry conditions a control-anchored residual flow without explicitly propagating target-derived signals along the graph. Condition-wise optimal transport couples unpaired control and perturbed populations for training, while a Delta-correlation objective aligns the predicted and observed condition-level expression-shift directions. GeneGeoFlow achieves Pearson Delta scores of 0.8979 on the Norman additive benchmark and 0.9088 on five held-out drug combinations in the fixed ComboSciPlex test split. These results support perturbation-conditioned gene geometry as an effective structural prior for intervention-specific response prediction, without conflating stable gene relationships with response propagation.
机器学习
[LG-0] Cloud-Boosted Low-Compute Multi-Channel Speech Enhancement INTERSPEECH2026
链接: https://arxiv.org/abs/2608.07423
作者: Xulin Fan,Juan Azcarreta,Ashutosh Pandey,Jesus Alvarez,Ke Tan,Jacob Donley,Ritwik Giri,Buye Xu
类目: ound (cs.SD); Machine Learning (cs.LG)
*备注: Accepted to Interspeech 2026
Abstract:Low-latency, low-compute speech enhancement is essential for wearable devices with real-time communication requirements, but strict computational constraints significantly limit on-device performance. Knowledge Boosting has been proposed as an effective approach to improve edge model performance by leveraging a more capable server-side model, but performance gains for speech enhancement have been limited. We propose a collaborative framework incorporating three techniques: (1) delayed server output as additional input, (2) layerwise feature boosting that transfers intermediate server representations to guide edge inference, and (3) collaborative multichannel Wiener filtering, which fuses weighted covariance matrices estimated from both server and edge models for improved beamforming. Experimental results demonstrate that the proposed collaborative framework significantly outperforms the edge-only baseline with minimal additional computational overhead.
[LG-1] Beyond Myopic World Models: Long-Horizon End-to-End Training for Direct Future Prediction
链接: https://arxiv.org/abs/2608.07420
作者: Xinyi Li,Zaishuo Xia,Chenjie Hao,Yubei Chen
类目: Machine Learning (cs.LG)
*备注:
Abstract:World models are expected to support imagination over extended temporal horizons, yet most are still trained through local few-step prediction objectives and deployed by recursively rolling out their own predictions. This creates a fundamental mismatch: few-step losses optimize local transition fidelity, while long-horizon prediction depends on how errors and gradients propagate through the entire trajectory. As a result, transitions with different downstream influence on the endpoint are treated uniformly during training, and small local errors are amplified through recursive inference. We argue that long-horizon accuracy is better achieved by optimizing directly, through an end-to-end endpoint prediction objective. To instantiate this paradigm, we introduce the Direct Prediction World Model (DPWM), a non-recursive architecture that compresses an action sequence of arbitrary length into a single embedding and predicts the endpoint observation in a single forward pass. This design avoids recurrent rollout in both prediction and gradient propagation, making long-horizon end-to-end training practical at horizons where unrolled autoregressive training becomes unstable. Empirically, DPWM substantially improves long-horizon endpoint prediction over recursive world-model baselines on continuous-control and pixel-based benchmarks, with larger gains as the prediction horizon increases. We further show that recurrent baselines benefit similarly when retrained with the same long-horizon endpoint objective, supporting our central claim that the training objective, rather than the particular backbone choice, is the main driver of long-horizon prediction accuracy. Our results suggest that world models can benefit from being trained and evaluated at the temporal scales where they are ultimately used, shifting the focus from local transition modeling toward long-horizon predictive accuracy.
[LG-2] Beyond Post-Hoc Temperature Scaling: Bilevel Optimization for LLM Calibration
链接: https://arxiv.org/abs/2608.07419
作者: Ruochen Jin,Zhanliang Wang,Zongyu Dai,Jiancong Xiao,Bojian Hou
类目: Machine Learning (cs.LG)
*备注: Third Conference on Language Modeling (COLM 2026)
Abstract:Preference alignment often makes large language models (LLMs) overconfident and poorly calibrated. Traditional post-hoc temperature scaling is inherently domain-dependent: a temperature fitted on one domain does not generalize across domains. This motivates us to modify model parameters during training to improve calibration. We propose maximizing the entropy of predictive distributions as the calibration objective, which directly targets overconfidence by discouraging overly concentrated predictions. Inspired by temperature scaling, we realize this through a bilevel optimization formulation, where the lower level trains the model under a parametric loss and the upper level selects loss hyperparameters to maximize entropy. To make the framework practical at LLM scale, we adopt an efficient first-order approximation that avoids explicit second-order computation. Across both multiple-choice and open-ended generative question answering, experiments demonstrate that our method yields well-calibrated LLMs with particular advantages in out-of-domain generalization.
[LG-3] FedDOSE: Federated Learning Framework Decomposing Site Effects for Modeling Brain Dynamic Functional Connectivity
链接: https://arxiv.org/abs/2608.07393
作者: Deepank Girish,Yi Hao Chan,Yubin Zheng,Sukrit Gupta,Jagath C. Rajapakse
类目: Machine Learning (cs.LG); Signal Processing (eess.SP); Neurons and Cognition (q-bio.NC)
*备注:
Abstract:Functional Magnetic Resonance Imaging ( fMRI ) data are often pooled into collaborative multi-site consortia, as deep learning models for analyses require large datasets to generalize well. While Federated Learning (FL) offers a privacy-preserving paradigm for collaborative training, standard approaches continue to struggle with statistical heterogeneity. In particular, site differences pose a key challenge in multi-site data settings. Additionally, existing FL approaches for fMRI rely on static Functional Connectivity ( FC), omitting dynamic information in brain networks. To address this, we propose FedDOSE, a novel framework that explicitly decomposes site differences for analysis of dynamic FC (dFC). FedDOSE introduces a Modularity-Guided Tucker Decomposition block to encode high-dimensional dFC tensors and capture modular-level spatio-temporal patterns efficiently. Class-specific prototypes are generated across all sites and subsequently aligned at the global level by using a combination of Optimal Transport (OT) barycenter formulation and Procrustes analysis. Extensive experiments for diagnosing Autism Spectrum Disorder (ASD) and Attention-Deficit Hyperactivity Disorder (ADHD) on three multi-site resting-state fMRI datasets: ABIDE-I, ABIDE-II, and ADHD-200, demonstrate that FedDOSE outperforms state-of-the-art methods in ASD and ADHD detection. Our results highlight its effectiveness in learning robust representations from multi-site datasets for reliable analysis.
[LG-4] Residual Algebra for Representation-Preserving Learning
链接: https://arxiv.org/abs/2608.07349
作者: Yao Wu
类目: Machine Learning (cs.LG)
*备注: 19 pages, 5 figures
Abstract:Learning from heterogeneous representations is usually reduced to feature concatenation, which erases which representation produced an error. We instead algebraize the residual: a representation is a typed object that owns both a coordinate system and the residual it leaves unresolved, and learning is an ordered composition of operators that preserve or deliberately erase that type. Fold realizes the objects as point-in-time conditional-mean fields on 10x10 rank grids. FPRC-PQ realizes the algebra as relax-aggregate-close: each field is relaxed by a correction fitted to its own residual in its own coordinates; corrected fields meet at a fixed mean that is the sole identity-erasure boundary; and a shared learner closes only the aggregate’s fresh residual. The composition telescopes exactly into representation, local residual estimate, and residual-of-residual estimate. Its aggregate is a learned control-variate interface with population variance reduction, while refitting the closer along perturbations of the backbone yields first-order coupled-path mean orthogonality. As an analytical extension, a reflective rumination operator reads the displacement of a global reconstruction from the aggregate anchor, reflects it, and fixes its gain by a unique orthogonal projection rather than return-tuned grid search. On 3.67M Chinese A-share stock-day observations (2023-2026) under a frozen point-in-time protocol, the evaluated base algebra raises net-of-cost return from 13.52% to 19.10% and Sharpe from 1.42 to 2.09. Matched-capacity, unified-residual, identity-free two-stage, and pairwise-only controls all trail it. The gain is therefore not explained by more features or more trees, but by making residual ownership and composition explicit while representation identity is still available.
[LG-5] When GNNs Fail: Quantifying and Overcoming Temporal Correlation Volatility in Time Series
链接: https://arxiv.org/abs/2608.07333
作者: Chen Shao,Yue Wang,Zhenyi Zhu,Zhanbo Huang,Tobias Käfer,Zonghan Wu,Danai Koutra
类目: Machine Learning (cs.LG)
*备注: 6 figures, 3 tables, 16 pages
Abstract:Modeling multivariate time series by representing them as graphs, where individual series act as nodes and pairwise temporal corre- lations serve as edges, has gained significant traction. Recent advances in Graph Neural Networks (GNNs) have demonstrated strong perfor- mance by assuming a static graph topology and aggregating information from neighboring series. In this work, we investigate the representa- tional power of GNNs for forecasting under both static and dynamic settings (i.e., when pairwise correlations evolve drastically over time) and identify critical limitations in current architectures. To formalize this, we first propose Temporal Correlation Volatility (TCV), a model- agnostic metric designed to quantify the distributional evolution of these latent structures. We establish a clear connection between TCV and performance degradation, demonstrating that many popular models, including Transformers, generalize poorly in high-TCV settings and are often outperformed by simple structure-agnostic baselines. To address these limitations, we propose Graph Layer for Inference in Dynamic En- vironments (GLIDE), a novel GNN layer enhanced by two theoretically grounded design mechanisms: (D1) Path-based Message Passing, which captures path-based neighborhoods and (D2) Static and Dynamic Propagation Separation, which identifies optimal dynamics via local static approximation. These components significantly improve learning under dynamic topology while preserving robustness in static scenarios. Ex- tensive experiments on synthetic and real-world benchmarks show that GLIDE improves average performance by up to 45.6% across static and dynamic settings, with the largest gain reaching 85.7%. The source code is available at this https URL.
[LG-6] Learning Fault-Tolerant Locomotion with Adaptive Gait Timing IROS
链接: https://arxiv.org/abs/2608.07328
作者: Giovanbattista Gravina,Luca Rossini,Carlo Rizzardo,Arturo Laurenzi,Nikos Tsagarakis
类目: Robotics (cs.RO); Machine Learning (cs.LG)
*备注: Accepted at the 2026 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
Abstract:Hardware failures require legged robots to rapidly reorganize coordination and gait timing to maintain stability and mobility. This is particularly challenging for larger quadrupeds, where increased mass and tighter actuation limits reduce the feasibility of aggressive, high-frequency compensation strategies often observed on smaller platforms. In this work, we propose a deep reinforcement learning approach for fault-tolerant locomotion under actuator power loss. The method employs an asymmetric actor-critic architecture in which the critic has access to privileged information during training, while the actor learns to reconstruct a corresponding latent representation from proprioceptive observations. We introduce a latent-alignment loss that encourages consistency between actor and critic representations. Additionally, we augment the action space with a learnable gait frequency parameter, enabling adaptive gait timing in response to terrain variations and actuator degradation without predefined faulty-leg strategies. The approach is validated in high-fidelity simulation on uneven terrain and real-world experiments on flat ground using a 68 kg quadruped robot.
[LG-7] Is SwiGLUs Open Positive Tail Necessary? Evidence from Closed-Tail Gating with MemGLU
链接: https://arxiv.org/abs/2608.07323
作者: Yuting Ge,Pengju Yang,Mingkai Nie
类目: Machine Learning (cs.LG)
*备注:
Abstract:We test whether decoder-only language-model FFNs require SwiGLU’s open positive tail. We introduce MemGLU as a closed-tail comparator derived from a memristive branch geometry. Across paired 9M and 30M pretraining runs with three seeds, MemGLU remains within about 0.1% of SwiGLU in validation NLL. Trained SwiGLU checkpoints are sensitive to positive-tail suppression, while mechanism diagnostics show that the two models use their gates differently despite similar losses. These results suggest that models adapt to the gate geometry available during pretraining. At the tested scales, SwiGLU’s open positive tail is not necessary for decoder-only language-model FFNs.
[LG-8] From Optimal Actions to World Models: Identifiability of Transition Kernels in Discounted MDPs
链接: https://arxiv.org/abs/2608.07301
作者: Neal Batra
类目: Machine Learning (cs.LG)
*备注: 15 pages, 1 figure
Abstract:We study what can be recovered about the transition probabilities of a Markov decision process from optimal actions alone. This is closely related to the inverse problem considered by Letcher et al., who ask when the dynamics can be recovered from numerical (Q)-values. Here the numerical values themselves are not observed; only the optimal actions are known, for every reward in a given class. For state-action rewards (r(s,a)), knowing the optimal actions for every reward also tells us how much better one action is than another when each is followed by the same fixed policy. This is still not enough to determine the transition probabilities uniquely. We prove that two kernels give the same optimal actions for every reward exactly when [ Q_s,a = \Bigl(P_s,a+\tfrac1\gamma e_s^\mathsf T(L-I)\Bigr)L^-1 ] for one invertible matrix (L) satisfying (L\mathbf 1=\mathbf 1). Near a kernel with strictly positive entries, there is an (n(n-1))-dimensional family of different kernels with this property. The result is unchanged if we consider only rewards having a unique optimal action at every state. We then compare this with rewards of the forms (r(s)) and (r(s,a,s’)). Rewards that depend on the next state can usually recover the transition kernel itself: every row at a state with at least two actions is determined, and we describe exactly when a row at a state with one action can remain hidden. State rewards reveal less: two kernels give the same optimal actions exactly when every deterministic policy is optimal for the same set of rewards. The results show how the form of the reward affects what can be learned about the dynamics from optimal actions alone. Comments: 15 pages, 1 figure Subjects: Machine Learning (cs.LG) Cite as: arXiv:2608.07301 [cs.LG] (or arXiv:2608.07301v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.07301 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-9] A foundation-model approach to pediatric headache classification from rs-fMRI ALT
链接: https://arxiv.org/abs/2608.07287
作者: Guilherme S. Imai Aldeia,Clara Moon,Julie Shulman,Navil Sethna,Allison Smith,Alyssa Lebel,William G. La Cava,Scott Holmes
类目: Machine Learning (cs.LG)
*备注: 22 pages, 6 figures. In Proceedings of Machine Learning Research, Volume 340, 2026 (Machine Learning for Healthcare Conference)
Abstract:Headache is the most common neurological disorder in children and substantially affects quality of life. We investigated whether resting-state functional MRI (rs-fMRI) can support pediatric headache classification using machine learning. We encoded rs-fMRI data using NeuroSTORM, a recent foundation model, and fine-tuned it to distinguish healthy controls from children with headache and subsequently classify headache subtypes. We compared NeuroSTORM with a standard neuroscience approach using functional-connectivity (FC) matrices derived from brain activity as predictors. Using 189 rs-fMRI scans from 110 individuals collected across two visits (prevalence of any headache: 74%), NeuroSTORM achieved an area under the receiver operating characteristic curve (AUROC) of 0.82 (95% CI, 0.82-0.82) and an area under the precision-recall curve (AUPRC) of 0.93 (95% CI, 0.93-0.94) for discriminating headache from non-headache. In contrast, models trained on FC matrices showed lower performance (AUROC, 0.67 [95% CI, 0.67-0.67]; AUPRC, 0.85 [95% CI, 0.85-0.85]). In multiclass classification of healthy controls, chronic migraine, and non-chronic headaches (e.g., post-viral headache, new daily persistent headache, post-traumatic headache), NeuroSTORM achieved a macro-AUROC of 0.69 (95% CI, 0.68-0.69). Results suggest that the approach can distinguish chronic migraine but has difficulty differentiating other headache subtypes from chronic migraine. Overall, under limited-data conditions, NeuroSTORM appears to capture latent rs-fMRI representations that transfer to headache-related tasks without relying on FC features. These findings provide proof of concept for fMRI-based prediction of pediatric headache and highlight potential future utility for subtype identification and individualized treatment strategies.
[LG-10] Learning Suffers More Than the Policy Class Under Partial Observability: A Closed-Form Analysis
链接: https://arxiv.org/abs/2608.07228
作者: Idil Gözel(University College London)
类目: Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注: 26 pages, 4 figures. Code and data: this https URL
Abstract:When a reinforcement learning agent cannot observe the full state, we usually blame its policies: it cannot see enough to represent a good one. We show that in a solvable case the bigger problem lies elsewhere. Even when a good policy is available and the agent’s value function is expressive enough to describe it exactly, learning still ends up somewhere far worse. We study a partially observed linear-quadratic problem in which a standard actor-critic learner can be solved in closed form. At our default setting the best policy the agent can represent is already close to optimal, costing 10.4% more than the ideal controller that observes everything. Learning does not find it. The algorithm instead comes to rest at a policy that is 35% worse than the best one available to it, and we can say exactly where and why. The cause is a bias in what the critic learns rather than a limit on what the actor can express. Because the agent cannot attribute what it sees to the part of the state it cannot observe, the critic misreads that unexplained variation as sharp curvature in its own value estimates, and the actor follows that error away from the optimum. We derive closed-form expressions for the resulting policy, for its cost, and for the one design choice that removes the problem, which is how far the learner looks ahead before trusting its own value estimates. Deep reinforcement learning experiments follow these predictions closely. Notably, giving the agent memory of past observations does not help, while changing how far it looks ahead does. Comments: 26 pages, 4 figures. Code and data: this https URL Subjects: Machine Learning (cs.LG); Optimization and Control (math.OC) MSC classes: 93E35 (Primary) 93E20, 93C41, 49N10 (Secondary) ACMclasses: I.2.6; I.2.8; G.3 Cite as: arXiv:2608.07228 [cs.LG] (or arXiv:2608.07228v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.07228 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-11] Dual-Node NVIDIA DGX Spark over Tailscale: A Remote-Access Testbed for Distributed LLM Training and Cyber-Threat-Intelligence Fine-Tuning
链接: https://arxiv.org/abs/2608.07226
作者: Vasanth Iyer
类目: Hardware Architecture (cs.AR); Machine Learning (cs.LG)
*备注: 12 pages, 3 figures, 4 tables. Deployment runbook and code: this https URL
Abstract:Compact AI systems make local language-model experimentation increasingly accessible, yet practical evidence for multi-node training on desktop-class accelerators remains limited. This report presents a proof-of-concept deployment of distributed NanoChat pretraining across two NVIDIA DGX Spark systems, each with a GB10 Grace Blackwell system-on-chip and 128 GB of unified memory, administered remotely over a Tailscale mesh VPN and connected for training by a dedicated 200 Gb/s QSFP56 direct fiber link. PyTorch torchrun, DDP, and NCCL were configured with one process per node, a depth-20 NanoChat model, a local batch size of 32 per node, and a 2,048-token context, giving a global batch of 131,072 tokens per step. The run sustained a step time of about 69.4 s (about 1,890 tokens/s), processing about 653 million tokens over four days. We document link configuration, container setup, interface binding, a step-zero evaluation bug that triggered NCCL timeouts, checkpointing, and troubleshooting lessons, as a reproducibility reference for small labs. We also built a cybersecurity fine-tuning dataset from 77 CISA advisories (338 training, 37 validation conversations) and ran a 17-question held-out evaluation comparing a baseline SFT checkpoint against a CTI-augmented checkpoint with an Ollama-hosted LLM judge. CTI-specific categories improved while general-knowledge categories regressed, for a small overall change from 2.06 to 2.29 on a 0-10 scale. The same cluster supports a 400-level AI course (CS 426) and a query engine for CompTIA Security+ POGIL activities in CBS 255, showing modest local infrastructure can serve both research and teaching. The study establishes feasibility rather than a scaling-efficiency claim, since single-node throughput used for comparison was estimated, not measured under matched conditions. Runbook and scripts are available (see Code Availability). Comments: 12 pages, 3 figures, 4 tables. Deployment runbook and code: this https URL Subjects: Hardware Architecture (cs.AR); Machine Learning (cs.LG) ACMclasses: C.2.4; I.2.6; D.4.6; K.3.2 Cite as: arXiv:2608.07226 [cs.AR] (or arXiv:2608.07226v1 [cs.AR] for this version) https://doi.org/10.48550/arXiv.2608.07226 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-12] Stochastic Autoregressive Learning
链接: https://arxiv.org/abs/2608.07224
作者: Ilan Doron-Arad,Idan Mehalel,Elchanan Mossel
类目: Machine Learning (cs.LG)
*备注:
Abstract:Motivated by LLMs, which generate outputs by iteratively sampling from next-token distributions, we introduce a PAC-learning model for binary stochastic autoregressive learning. This generalizes the deterministic autoregressive learning framework of Joshi et al., COLT 2025. In our model, one fixed generator assigns a Bernoulli next-token distribution to every prompt string. Starting from an input prompt, a token is sampled and appended to the prompt; the same generator is then applied again to this expanded prompt; this procedure is repeated for M steps. Three forms of supervision are considered: base one-step samples, chain-of-thought (CoT) samples that reveal full random trajectories of length M , and end-to-end (e2e) samples that reveal only the final token of length M trajectories. For a generator class, we study the minimum number of samples m_base(\varepsilon),m_CoT(\varepsilon), m_e2e(\varepsilon) , resp., required to learn the one-step probabilities in the base model, and the final-token probability in the CoT and e2e models, under squared loss error~ \varepsilon . We show that stochastic autoregressive learning fundamentally differs from the deterministic theory. At scale \varepsilon , there is no universal comparison between the three learning tasks: both m_CoT/m_base and m_e2e/m_CoT can be made simultaneously arbitrarily larger than M/\varepsilon , the natural analogue for the existing deterministic results. Nevertheless, after altering scales, for every class, CoT learning at scale \varepsilon is upper-bounded by base learning at scale \varepsilon/M^2 , whereas e2e learning at scale \varepsilon is upper-bounded, up to logarithmic factors, by (M/\varepsilon) m_CoT(\Theta(\varepsilon)) . These dependencies and scales are essentially tight. We complement these bounds by studying dimension d logistic functions in our model. Subjects: Machine Learning (cs.LG) Cite as: arXiv:2608.07224 [cs.LG] (or arXiv:2608.07224v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.07224 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-13] MAUPITI: On-Device Prototype-Based Learning on a Smart Infrared Sensor
链接: https://arxiv.org/abs/2608.07192
作者: Beatrice Alessandra Motetti,Tanguy Dugas du Villard,Matteo Risso,Alessio Burrello,Francesco Daghero,Enrico Macii,Massimo Poncino,Marco Castellano,Alfio Basile,Daniele Jahier Pagliari
类目: Machine Learning (cs.LG)
*备注: Accepted for publication in IEEE Embedded Systems Letters
Abstract:Low-resolution infrared (IR) array sensors represent an interesting solution for privacy-preserving human sensing in embedded systems. In this letter, we describe a smart multi-pixel IR sensor integrating a 16 \times 16 thermal MOSFET (TMOS) array and a RISC-V microcontroller extended with low-precision SIMD instructions, capable of on-device learning and continual adaptation for pose and gesture recognition tasks under tight memory and power constraints ( 32kB on-chip memory, \approx 1.5mW). To avoid the memory overheads of backpropagation and replay buffers, we adopt a prototype-based Nearest Class Mean (NCM) classifier in which a simple Convolutional Neural Network (CNN) encoder is trained and quantized offline, while class prototypes are stored and updated on the device in streaming mode. With experiments on two datasets, we show that this approach yields accuracy on par with a conventional classifier, with negligible latency overheads in both the classification and the prototype update ( 0.29% considering both phases), effectively enabling online adaptation of the perception framework.
[LG-14] Conformal Fusion Under Missing Modalities
链接: https://arxiv.org/abs/2608.07183
作者: Alireza Moayedikia
类目: Machine Learning (cs.LG)
*备注:
Abstract:Multimodal fusion architectures typically assume all modalities are available at inference, yet sensor failures, acquisition variability, and cost constraints routinely produce incomplete observations. Existing work treats modality absence as a prediction-accuracy problem, leaving a more basic question unanswered: whether a model’s confidence estimates remain calibrated when an entire input stream is removed. We argue that missing-modality robustness and calibrated uncertainty are a single coupled property, and introduce Modality-Conditioned Conformal Fusion (MCCF), an architecture that addresses both at once. MCCF combines a multimodal bottleneck fusion backbone trained with modality dropout, per-modality evidential heads producing modality-decomposed Dirichlet distributions, and a Dempster-Shafer combination rule that fuses the per-modality evidence into a joint predictive distribution; an absent modality contributes vacuous evidence that is structurally ignored, so the fused uncertainty automatically reflects the reduced information without test-time imputation. A Mondrian conformal calibration module keyed on the modality-presence mask then provides finite-sample group-conditional coverage for every non-empty modality subset. MCCF is, to our knowledge, the first method with formal coverage guarantees under arbitrary modality availability through architectural integration rather than post-hoc recalibration, and the evidential decomposition yields per-modality vacuity scores that localise uncertainty to the absent modality responsible. Across a synthetic problem and three real multimodal benchmarks, MCCF holds its target coverage on every modality-presence subset, substantially narrows the coverage gap between full and partial modalities relative to a marginal split-conformal baseline, and imposes no measurable accuracy cost relative to temperature-scaled and evidential baselines.
[LG-15] Edge Sparsification via Temporal Forman-Ricci Curvature for Dynamic Graph Learning
链接: https://arxiv.org/abs/2608.07158
作者: Poupak Azad,Cuneyt Gurcan Akcora,Kiarash Shamsi
类目: Machine Learning (cs.LG)
*备注:
Abstract:Temporal graph learning has become essential for analyzing real-world systems whose interactions continuously evolve over time, including financial transaction networks, communication systems, and online social platforms. However, learning from large-scale temporal graphs remains computationally challenging when networks are dense and rapidly changing. To address this limitation, we propose a network-curvature-inspired edge sparsification framework for dynamic graph learning. Our proposed method, TRicci, extends classical Forman-Ricci curvature to directed weighted temporal graphs by capturing structural support, temporal recency, and local interaction competition. Experiments on 9 transaction networks and 3 temporal graph benchmark datasets demonstrate that the proposed framework preserves predictive performance across multiple graph-level prediction tasks. The results show that TRicci sparsifies temporal graphs by approximately 80% while reducing end-to-end downstream training and inference time by an average of 55.94%, without substantial degradation in predictive performance. Our findings suggest that temporal curvature can serve as a principled basis for scalable temporal graph learning by preserving predictive temporal-structural information under substantial sparsification. Subjects: Machine Learning (cs.LG) Cite as: arXiv:2608.07158 [cs.LG] (or arXiv:2608.07158v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.07158 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-16] Capacity Confounds and Coverag e Guarantees in Adaptive Sub-model Federated Learning
链接: https://arxiv.org/abs/2608.07157
作者: Alireza Moayedikia,Alicia Troncoso Lora
类目: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC)
*备注:
Abstract:Sub-model federated learning lets resource-constrained clients train width-reduced versions of a global model, but existing methods allocate capacity by device resources alone. A natural next step, allocating capacity by each client’s data heterogeneity as estimated from the updates the server already observes, has been repeatedly suggested. We ask whether that step is possible, using HAS-FL, an adaptive capacity-allocation framework, as a test case. Our findings are threefold. First, validated against ground-truth label-distribution divergence on reproducible partitions, update-divergence estimates of client heterogeneity are dominated by capacity rather than data: across two corrected estimators, multiple datasets, and all seeds, the estimates correlate strongly and negatively with device capacity, and no data signal remains once capacity is controlled for. This previously undocumented confound affects any method estimating client statistics from sub-model updates. Second, adaptive allocation has a hidden failure mode: when every client is capped below full width, the uncovered parameters stay at random initialization and progressively corrupt the global model. A simple coverage guarantee removes the failure and explains why uniform allocation collapses. Third, a matched-budget control settles what adaptivity contributes: random allocation to the same average budget performs no differently on both image benchmarks, and on the naturally partitioned text benchmark the adaptive policy is the weakest of the three strategies while consuming the most capacity. Sub-model training remains valuable because it admits constrained clients at quadratically reduced cost, but what protects accuracy is parameter coverage rather than allocation intelligence. Its apparent benefits come from capacity budgeting and coverage, and future designs need heterogeneity signals separable from capacity effects.
[LG-17] Machine Learning-Based Inter-Crystal Scatter Recovery for Ultra-High Resolution PET Imaging
链接: https://arxiv.org/abs/2608.07155
作者: Alexandre Bernier,Roger Lecomte,Jean-Baptiste Michaud
类目: Machine Learning (cs.LG); Medical Physics (physics.med-ph)
*备注: 19 pages, to b published in Physic in Medecine and Biology
Abstract:Inter-crystal scatter (ICS) events pose a significant challenge in ultrahigh- resolution positron emission tomography (UHR-PET), especially as detector crystals become smaller and their readouts increasingly segmented. Current approaches either reject these events, reducing sensitivity, or accept them with suboptimal positioning algorithms, degrading image resolution. We present a feed forward neural network to optimize ICS event recovery by inferring the line-of-response belonging to the first Compton interaction. Our approach was validated using both Monte Carlo simulations and experimental data from the fully pixelated LabPET-IIbased preclinical and brain UHR-PET this http URL demonstrate a 70% to 106% increase in sensitivity while preserving sub-millimeter spatial resolvability (down to 1.6 mm) compared to conventional methods. This ICS recovery approach is an effective solution that compensates for the lower detection efficiency of small, pixelated detectors in UHR-PET, enabling reduced scan times and lower radiation doses while largely preserving image quality.
[LG-18] Online Conformal Prediction Beyond Feedback
链接: https://arxiv.org/abs/2608.07139
作者: Joar Skalse,Edoardo Pona,Osvaldo Simeone,Nicola Paoletti
类目: Machine Learning (cs.LG)
*备注:
Abstract:Uncertainty quantification is essential when deploying machine learning models in safety-critical applications. Online conformal prediction (OCP) provides theoretically principled uncertainty quantification for arbitrary black-box classifiers and non-i.i.d. data streams by constructing prediction sets that are guaranteed to contain the true label at a user-specified frequency. OCP usually updates prediction sets using feedback from previously deployed predictions. We instead study an OCP setting beyond feedback: on each round, the learner can either output a prediction set or query the correct label, but not both. Thus, no deployed prediction is ever evaluated directly. We reduce this problem to a partial monitoring game in which prediction actions return no observation and a separate query action reveals the label. The reward function is constructed in a way that encourages the learner to output small prediction sets while ensuring that the correct label is covered with a sufficiently high probability. To solve this game, we develop OCP with queries (OCPQ) by adapting the label efficient forecaster of Cesa-Bianchi, Lugosi, and Stoltz (2004) to our setting. For any black box classifier and any (non-i.i.d.) oblivious data stream of length T , OCPQ has O(T^2/3) expected regret and expected coverage at least \beta-O(T^-1/3) for a user-defined \beta , while querying only an expected T^-1/3 fraction of rounds. This provides coverage comparable to bandit-based OCP methods while requiring no feedback from deployed prediction sets. Experiments on real-world datasets further demonstrate the effectiveness of our approach.
[LG-19] Synthetic LiDAR Data Generation and Deterministic Downsampling for Point Cloud Classification on the Edge
链接: https://arxiv.org/abs/2608.07106
作者: Niclas Meyer,Stefan Reitmann
类目: Machine Learning (cs.LG); Robotics (cs.RO)
*备注:
Abstract:Deploying three-dimensional deep learning frameworks to low-power embedded processors is bottlenecked by the unstructured nature of spatial data and the resource-intensive distance sorting algorithms often used before neural network inference. To address this gap, this paper presents a hardware-constrained workflow optimized for native execution on the Raspberry Pi 5. To account for the reality gap between noiseless, clean computer-aided design (CAD) datasets and real-world sensor data, we use physics-based simulation to construct a synthetic LiDAR dataset. Cross-dataset evaluations demonstrate a substantial drop in classification accuracy when networks trained on clean CAD data are evaluated on synthetic LiDAR sensor data, highlighting the critical need for sensor-aware training. To address the latency bottleneck of traditional geometric preprocessing on edge CPUs, we integrate an isolated, feature-driven Critical Points Layer (CPL) as a frontend filter. Our results show that the pretrained CPL deterministically compresses raw 1024-point clouds to a subset of 40 to 60 unique coordinates. When profiled on the ARM Cortex-A76 processor, the complete pipeline achieves an inference throughput of approximately 50 FPS while maintaining an instance classification accuracy of 88.36%, demonstrating the viability of deterministic real-time 3D perception at the edge.
[LG-20] nsor Network Kernel Machines: A JAX Framework for Machine Learning and Nonlinear System Identification
链接: https://arxiv.org/abs/2608.07043
作者: Albert Saiapin,Kim Batselier
类目: Mathematical Software (cs.MS); Machine Learning (cs.LG); Systems and Control (eess.SY)
*备注: 10 pages, 6 figures, 4 tables, 1 listing. Code available at: this https URL
Abstract:Developing nonlinear models that are both expressive and computationally efficient remains a challenge in machine learning and nonlinear system identification. Tensor network kernel machines (TNKM) address this challenge by combining nonlinear feature representations with compact low-rank tensor-network parameterizations. However, practical and extensible software frameworks for developing TNKM models remain limited. In this work, we introduce “tnkm”, an open-source Python library for constructing and training TNKM models using JAX. The library provides a unified interface for combining different feature maps, tensor-network architectures, and optimization strategies, including alternating least squares and gradient-based methods. We demonstrate the capabilities of “tnkm” on nonlinear benchmark problems, showing that the implemented models achieve competitive prediction accuracy while retaining compact parameterizations and efficient training. The proposed framework facilitates reproducible development and application of tensor-network-based learning methods.
[LG-21] Hyperbolic Graph Embedders for Link Prediction and Topology Reconstruction
链接: https://arxiv.org/abs/2608.07029
作者: Robert Jankowski,Maksim Kitsak,Dorota Celińska-Kopczyńska
类目: Machine Learning (cs.LG); Social and Information Networks (cs.SI); Physics and Society (physics.soc-ph)
*备注:
Abstract:Hyperbolic embeddings provide compact geometric representations of complex networks in hyperbolic spaces, but systematic comparisons of methods developed in machine learning, network science, and algorithmics remain rare. We benchmark 13 unsupervised hyperbolic graph embedders under a unified protocol for link prediction and topology reconstruction on synthetic and empirical networks. The protocol captures both missing-link recovery and the preservation of local and global network structure. Maximum-likelihood and representation-learning-based approaches, including hybrid variants, achieve the strongest overall performance, although no method dominates across all tasks and structural regimes. Performance is more strongly associated with embedding paradigm than with disciplinary origin. We identify the network regimes in which different paradigms succeed or fail and provide practical guidance for method selection in downstream applications.
[LG-22] Every Cache Entry Earns Its Place: Global Allocation of Resolution and Coverag e for KV Cache Compression
链接: https://arxiv.org/abs/2608.07001
作者: Haolin Tian,Yuzhe Liu,Tonghan Wang
类目: Machine Learning (cs.LG)
*备注:
Abstract:As large language models (LLMs) process increasingly long contexts, KV cache storage and repeated access have become a major bottleneck. Existing KV cache compression methods rely on predefined, fixed compression rules and are typically developed around either token eviction or merging. As a result, cache resources can neither flow freely across layers, heads, and context slots, nor be jointly allocated to balance local resolution and information coverage. Therefore, we propose GraceKV, a global approach for the allocation of resolution and coverage in KV cache compression, and formulate the compression process as a global resource allocation problem under a fixed cache budget. GraceKV treats each layer-KV head-slot combination as an atomic unit and builds a prototype tree. Leaf nodes correspond to token-level KV entries, while each internal node uses a single prototype to compress the KV space covered by its children. A set of non-overlapping nodes in the tree forms the representation of an atomic unit. Adding the root of a new tree expands information coverage, whereas splitting a selected node improves local resolution. All candidate actions compete globally for a shared cache budget. Finally, the nodes retained across all trees form the compressed KV cache. This process adaptively determines the allocation of cache resources among atomic units globally and the balance between resolution and coverage. GraceKV requires no additional training, and the entire compression and inference process is performed on the GPU. Systematic experiments across diverse long-context tasks and compression ratios show that GraceKV ranks first in 24 of 32 settings and remains robust up to 128-fold compression. These results validate the effectiveness of global budget allocation in coordinating information coverage and local resolution.
[LG-23] How Molecular Generative Models Organize Molecular Identity
链接: https://arxiv.org/abs/2608.06956
作者: Raul Ortega-Ochoa,Tejs Vegge,Jens S. Bakander,Luis Mantilla Calderon,Alan Aspuru-Guzik,Tonio Buonassisi
类目: Machine Learning (cs.LG); Chemical Physics (physics.chem-ph)
*备注: 22 pages (14 main text + 8 supporting information)
Abstract:Generative models for matter are often evaluated as samplers over output representations, and their latent spaces are commonly used as proxies for navigating chemical space. Much less is known about how these models internally arrange discrete chemical identities within those representations. We study this arrangement by making molecular identity explicit and pulling it back through the generative process. Through these pullbacks we probe the regions that generate the same object, exposing the trained model’s internal repertoire: a fixed partition that determines which objects (novel or not) the model can produce. Across three molecular generative architectures, we find that this repertoire is arranged into piecewise-constant regions separated by recurring coarse-to-fine boundaries. Its organization depends on the representation probed, the identity convention, decoder stochasticity, and the metric used to compare coordinates. During training, local chemical organization stabilizes while the number of distinct molecular identities represented within each neighborhood continues to change. Internal organization must therefore be characterized, rather than assumed, before a generative space can be treated as chemically navigable. Comments: 22 pages (14 main text + 8 supporting information) Subjects: Machine Learning (cs.LG); Chemical Physics (physics.chem-ph) Cite as: arXiv:2608.06956 [cs.LG] (or arXiv:2608.06956v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.06956 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-24] A Rate Separation for Agnostic Direct Sums
链接: https://arxiv.org/abs/2608.06951
作者: Mihir More,Aritra Das,Debayan Gupta
类目: Machine Learning (cs.LG)
*备注:
Abstract:Hanneke, Moran, and Waknine \citeHannekeMoranWaknine2024 asked how the agnostic PAC learning curve of the direct sum C^r depends on the single-instance learning curve \epsagn(n\mid C) and on r . We show that the single-instance learning rate does not determine the direct-sum rate. Let \F be the class of the two constant binary functions and let \G consist of the zero function and the identity function. Both classes have agnostic learning curve of order n^-1/2 .
[LG-25] Stream Learning: Partition-Fair Gossip Learning Without Tokens
链接: https://arxiv.org/abs/2608.06946
作者: Fabien Mathieu(NPA),Alexandre Pham(NPA),Maria Gradinariu Potop-Butucaru(NPA),S{é}bastien Tixeuil(IUF, NPA)
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG)
*备注:
Abstract:In gossip learning, a network of nodes trains a shared model collaboratively, without a central coordinator, by repeatedly exchanging parts of their local models. The state-of-the-art protocol, Partitioned Token Gossip Learning (PTGL) of Hegedüs et al., splits the weight matrix into S fixed partitions and disseminates them using a token-based fairness mechanism coupled with per-neighbor metadata exchange. We revisit partition scheduling by analogy with peer-to-peer live streaming, where model partitions act as video chunks and partition age acts as chunk scarcity. The analogy yields a design space of two-stage selection strategies (partition first, or neighbor first), from which we instantiate ten concrete protocols collectively called Stream Learning. Our main finding is that the simplest of these protocols, which transmits the locally least-trained partition to a uniformly random neighbor (Ri), matches PTGL on fault-free workloads while requiring neither token counters nor metadata exchange. Under an adversarial 30% permanent crash of the best-performing nodes, Ri matches or outperforms PTGL across all complete-graph configurations tested, with the gap reaching 5.53% on HAR and 5.41% on MNIST in the most heterogeneous regime (Dirichlet \beta = 0.1). In our experiments, partition fairness, captured by a single local rule on partition age, accounts for the gap; token-based rate control and utility maximization do not improve over this rule and, under heterogeneity, sit below it.
[LG-26] MiCoPro: End-to-End Mixed Precision HW/SW Co-design with HW-aware Proxy Model
链接: https://arxiv.org/abs/2608.06916
作者: Zijun Jiang,Yangdi Lyu
类目: Machine Learning (cs.LG)
*备注: 14 pages, 9 figures, under review
Abstract:Quantized Neural Networks~(QNN) with low-bitwidth data have proven promising in efficient storage and computation on edge devices. To mitigate accuracy degradation while maximizing speedup, layer-wise mixed-precision quantization~(MPQ) becomes a popular solution. However, existing algorithms for exploring MPQ schemes are limited in flexibility and efficiency. Comprehending the complex impacts of different MPQ schemes on post-training quantization and quantization-aware training results is a challenge for conventional methods. Furthermore, an end-to-end framework for the optimization and deployment of MPQ models is missing in existing work. To address these challenges, we propose the MiCo framework, a holistic MPQ exploration and deployment framework for edge AI applications. The framework adopts a novel optimization algorithm to search for accuracy-optimal quantization configurations under strict latency constraints. We further extended the framework to MiCoPro, which introduces a robust Hardware-Aware Proxy (HAP) model to enhance prediction accuracy and hardware versatility. By leveraging target-specific latency modeling, MiCoPro enables rapid exploration and direct deployment from PyTorch models to bare-metal C code. We demonstrate the versatility of our framework on both the BitFusion accelerator and SIMD-extended RISC-V processors, achieving up to 40% of latency reduction with less than 3% of accuracy drop. Comments: 14 pages, 9 figures, under review Subjects: Machine Learning (cs.LG) Cite as: arXiv:2608.06916 [cs.LG] (or arXiv:2608.06916v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.06916 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-27] Recent advances in weakly supervised learning: New supervision paradigms assumption relaxations and practical solutions
链接: https://arxiv.org/abs/2608.06896
作者: Wei Wang,Gang Niu,Masashi Sugiyama
类目: Machine Learning (cs.LG)
*备注: Preprint of an invited review article. In Neural Networks, Handbook of Statistics, vol. 55, pp. 289-320, Elsevier, 2026
Abstract:Deep learning has achieved great success in recent years thanks to the availability of high-quality, well-annotated training data. However, this requirement is often not met in real-world applications. Weakly supervised learning aims to train an accurate model with incomplete, inexact, or inaccurate supervision. In this chapter, we will discuss recent advances in this field, including new supervision paradigms, relaxed assumptions, and practical solutions. First, we introduce a new weakly supervised binary classification problem called confidence-difference classification and propose consistent approaches to solve it. Next, we investigate complementary-label learning, a weakly supervised multi-class classification problem. Our proposed approaches are based on more relaxed assumptions about the data generation process than existing consistent approaches. Lastly, we present an evaluation framework for partial-label learning, another popular multi-class weakly supervised learning problem, in order to promote fair and realistic evaluation of algorithms in this field.
[LG-28] PRISM: Principled Reference Identification for Schrodinger Bridge Model
链接: https://arxiv.org/abs/2608.06893
作者: Forouzan Fallah,Yezhou Yang
类目: Machine Learning (cs.LG)
*备注:
Abstract:Schrödinger bridge models restore a clean signal from a degraded observation by following the conditional bridges of a reference process, yet this reference is chosen heuristically, typically white noise with a hand-tuned schedule. We develop PRISM, a theory of bridge reference design. We characterize the time-varying Gaussian references that remain exactly tractable with per-mode schedules: precisely those whose instantaneous covariances commute. We then prove an invisibility principle: with the exact drift and unlimited solver steps, every admissible reference recovers the true posterior. The choice of reference therefore matters only under finite computational resources. For a fixed step budget, we derive the finite-step objective in closed form and prove that every optimal noise spectrum is proportional to Pk, the spectrum of information destroyed by the sensor, with a mode-independent constant x*(T) = (2 ln T)^-1/2 (1 + o(1)). The analysis shows that noise color and temporal scheduling are interchangeable, and regularization provably shifts the optimal reference toward white noise. Experiments in Gaussian settings confirm the predicted orderings and the closed-form loss floors. On FFHQ, the distortion-- perception trade-off and spectral localization transfer, but white noise outperforms the matched reference; a pre-registered study that changes the training regime refutes ridge whitening as the explanation. A 2x2 mechanism study then traces the inversion to the non-Gaussian per-mode statistics of real images. PRISM turns reference design from a hyperparameter sweep into a calculation in the Gaussian regime, and locates exactly where real images break it.
[LG-29] SkillAligner: Treating Retrieved Skills as Adaptable Drafts at Execution Time
链接: https://arxiv.org/abs/2608.06880
作者: Qinfeng Li,Dalin He,Yuntai Bao,Ying Yang,Ruoxi Chen,Xinyan Yu,Lizhou Liang,Ge Su,Wenqi Zhang,Xuhong Zhang
类目: Machine Learning (cs.LG)
*备注: 21 pages, 5 figures
Abstract:General-purpose skills promise reusable procedural knowledge for language agents, yet semantic relevance does not guarantee execution utility: a retrieved skill may encode assumptions that conflict with the current task, execution environment, or other retrieved skills. We formalize this problem as the skill–execution misfit. To address it, we propose SkillAligner, a training-free execution-time skill adaptation framework that treats retrieved skills as adaptable drafts rather than fixed instructions. Before execution, SkillAligner performs a one-time joint adaptation that specializes useful skill fragments to task requirements, aligns their procedural assumptions with the available execution interface, and composes the resulting guidance by resolving dependencies, conflicts, and redundancy across skills. The adapted content is consolidated into a compact execution guide and reused throughout the subsequent trajectory. Extensive experiments across diverse agent benchmarks and model backbones show that SkillAligner substantially improves task performance over existing skill-use baselines, reduces skill-induced regressions at the instance level, and lowers total inference cost.
[LG-30] Mathematical Principles and Experimental Discoveries of the Emergence of Symbolic Patterns in Artificial Neural Networks
链接: https://arxiv.org/abs/2608.06839
作者: Quanshi Zhang,Qihan Ren,Siyu Lou
类目: Machine Learning (cs.LG)
*备注:
Abstract:Artificial Neural networks (ANNs) are often treated as black-box models, making explainability a central challenge in deep learning. Many engineering methods have been proposed to approximately explain the ANN from various perspectives, such as feature attribution and visualization. However, it remains a long-standing open question whether the complex inference logic of an ANN can be explained exhaustively and concisely as sparse symbolic patterns. This raises a deeper inquiry: does the emergence of symbolic patterns reflect a natural law rather than chance? Here, we show that across a broad class of ANNs trained on diverse tasks, their inference logic can indeed be reformulated as sparse symbolic interactions. We further prove that two common mathematical criteria, which are implicitly required across tasks, lead to the emergence of such sparse symbolic interactions. Empirical evidence confirms that the two criteria hold for the majority of input samples in diverse models. Furthermore, the faithfulness of these interactions is also demonstrated by their strong sample-to-sample and model-to-model transferability, as well as their ability to explain the overall generalization power of ANNs. Our theoretical analysis and extensive experiments provide a solid foundation for symbolic explanations of ANNs, and offer novel insights into the ANN’s generalization power. Our findings also highlight the potential of communicative learning, a paradigm in which the inference logic of an ANN can be directly inspected and tuned at the level of symbolic patterns, thus complementing traditional end-to-end learning paradigm. Finally, the observed emergence of symbolic patterns in ANNs suggests that similar symbolic representations may also emerge in other types of black-box systems under certain conditions, because our proof does not depend on any specific ANN architecture.
[LG-31] Graph Machine: Exploring Edge Mechanisms as an Inductive Bias
链接: https://arxiv.org/abs/2608.06834
作者: Lintai Hou
类目: Machine Learning (cs.LG)
*备注:
Abstract:Transformers provide a powerful architecture for global content-based matching, but reasoning problems may benefit from a stronger inductive bias toward iterative traversal of latent relations. We introduce Graph Machine, an architecture with two explicit edge-based mechanisms: Edge-augmented attention, in which edges modulate attention between nodes, and edge-centric referral, in which nodes exchange addresses to update their edges. Conceptually, this enables the model to dynamically and differentiably construct and revise relational graphs across layers. We study this inductive bias using Sudoku under controlled settings and find that Graph Machine outperforms Transformer baselines, with ablation studies and mechanistic analysis attributing the gains to the edge mechanisms. Surprisingly, we found that the model discovers a compact edge-based construction for Sudoku geometry. Our results support explicit edge mechanisms as a promising architectural design, motivating broader evaluation.
[LG-32] Multiscale Reward Hedging from Correct Demonstrations
链接: https://arxiv.org/abs/2608.06825
作者: Pahan Dewasurendra
类目: Machine Learning (cs.LG)
*备注:
Abstract:Learning from correct demonstrations is harder than supervised learning when many answers are correct: after predicting, the learner sees one valid answer but not whether its own answer was valid, nor any reward. Existing reward-hedging guarantees consequently assume a finite reward class. We give the first horizon-free guarantee for continuous classes. The key is to hedge in one shared vote over tolerant optimality tests at every accuracy scale. A target reward has one surviving proxy per scale, and a prediction with gap above that scale doubles the proxy. This yields the simultaneous tail bound |\t:\ell_t2^-j|\leq \log_2\mathcal N(\mathcal G,2^-j-1)+j , where \mathcal G is the class of optimality-gap functions. Integrating the tails gives cumulative hidden gap bounded by a metric-entropy integral, independently of the number of rounds. Polynomial entropy (A/\epsilon)^d gives O(d\log A) total gap and a fast O(d/m) statistical rate. For bounded linear contextual recommendation, the result is O(d) regret for arbitrary compact menus. This is the first polynomial finite bound without structural restrictions on the menus, at the price of improper prediction. Although the general vote can be expensive, it is exactly polynomial-time for one-dimensional Lipschitz parameter curves. Fixed-radius rank-two recommendation takes O(KT^2) time for menus of size K . We also prove an \Omega(d) lower bound, low-rank and bounded ReLU-network corollaries, and a robust theorem that adds only the demonstrator’s cumulative suboptimality. A reproducible adaptive stress test illustrates the predicted scale adaptation. After factorization, an exact MovieLens audit runs in 1.7 CPU seconds across ten users and improves mean latent gap over both a demonstrated-rating policy and a proper online baseline. The learner uses only action demonstrations and never observes a reward or a loss.
[LG-33] Weak Adversarial Neural Pushforward Method for Boltzmann Equation
链接: https://arxiv.org/abs/2608.06823
作者: Jenia Fardousi Koly,Andrew Qing He,Wei Cai
类目: Numerical Analysis (math.NA); Machine Learning (cs.LG)
*备注:
Abstract:In this paper, we extend a weak adversary neural network pushforward method for solving time dependent Boltzmann equation and a weak formulation of the collision operator is proposed where an invertible neural pushforward mapping is used to generating samples given by the distribution governed by the Boltzmann equation. The training of the pushforward mapping is learnt by enforcing the weak form of the Boltzmann equation. Numerical results have demonstrated the effectiveness of the proposed method.
[LG-34] Understanding Differentiable Embeddings Through Differential and Integral Geometry
链接: https://arxiv.org/abs/2608.06809
作者: Xinyu Zhang,Klaus Mueller
类目: Machine Learning (cs.LG)
*备注: 19 pages in total, 12 for the main and 7 for the appendices. 9 figures in the main and 7 in the appendices
Abstract:How can an analyst decide whether a nonlinear dimensionality reduction embedding can be trusted? Existing diagnostics provide only partial answers: projection glyphs characterize local sensitivity, map-continuity scores measure local conditioning, and transport-based analyses reveal path-dependent inconsistencies. However, these methods appear unrelated and provide no common framework for understanding when they agree or not. We show that they are all derived from a single geometric object induced by every differentiable embedding, whether defined implicitly through optimization or explicitly by a learned mapping. This framework provides two complementary geometric views of an embedding. The differential view explains local behavior: its first-order term recovers projection glyphs, while its second-order curvature quantifies how far their linear approximation remains reliable. The integral view follows the same geometry along high dimensional paths and determines whether an embedding depends only on the current state or also on the path taken to reach it. We further show that map-continuity is a prerequisite for the other analyses. The framework is theoretically complete for diagnostics derived from the embedding geometry, and we prove the integral view irreducible: no amount of local measurement at any number of points, to any order of derivative, reproduces what it detects. Classical rank-based metrics form a complementary class based on finite-scale neighborhood relationships. Experiments on synthetic and real datasets validate theoretical predictions, demonstrate accurate curvature-based trust estimates on single-cell embeddings, and show that the integral analysis distinguishes single-valued embeddings from path-dependent optimization-based embeddings in ways that existing pointwise diagnostics cannot.
[LG-35] Faster Query-Key Learning Sharpens Attention in Self-Attention Models ICML2026
链接: https://arxiv.org/abs/2608.06776
作者: Rahul Vashisht,Harish G. Ramaswamy
类目: Machine Learning (cs.LG)
*备注: Accepted to the 43rd International Conference on Machine Learning (ICML 2026)
Abstract:A standard self-attention layer consists of two interacting circuits: the query-key circuit that governs attention allocation, and the output-value circuit that maps attended representations to predictions. Collapsed and factorized parameterizations of the query-key and output-value circuits lead to qualitatively different attention patterns. In particular, some parameterizations give sharper attention to task-relevant tokens, at a similar training loss. We analyze how the parameterizations of these circuits shape the parameter trajectories in single-layer self-attention models trained for next-token prediction. Through gradient-flow analysis, we show that factorization induces implicit rescaling of the two circuits’ learning rates. We derive closed-form dynamics showing that output-value and query-key parameters move along a line, with relative speeds determined by their learning rates. Faster query-key learning relative to output-value learning thus produces sharper attention, as the model compensates for slower output-value learning by increasing attention mass on relevant tokens. Experiments show that differences in the relative learning rates of the two circuits govern attention concentration. This improves attention interpretability proxies while maintaining comparable predictive performance.
[LG-36] ArchEGraph: A Large-Scale Graph Dataset for Geometry-Topology-Physics Aligned Building Energy Modeling
链接: https://arxiv.org/abs/2608.06772
作者: Yihui Li,Yihui Chen,Kaidi Zha,Xiaoyue Yan,Zhexuan Yu,Shiqi Dai,Jun Xiao,Jun Yin,Ramon Elias Weber,Borong Lin
类目: Machine Learning (cs.LG)
*备注: 26 pages, 13 figures, submitted to a conference
Abstract:Accurate estimation of building energy use is essential for achieving carbon neutral and sustainable buildings. To better understand the influence of design decisions on building energy use and calibrate machine learning models that can give architects and engineers rapid design feedback, large-scale datasets are needed that explicitly map building geometry to performance. We present ArchEGraph, a large-scale benchmark dataset that represents buildings as heterogeneous graphs with aligned geometry, topology, weather, and zone-level thermal loads. The dataset contains 5,481 buildings and 49,326 validated building-weather simulation cases. In total, it includes over 133,000 space nodes and 1.44 million face nodes, reflecting substantial geometric and topological complexity. Based on ArchEGraph, we define two benchmark tasks: (i) graph reconstruction from polygonal meshes, aiming to recover topological structure from geometric representations; and (ii) topology-informed load prediction, which leverages graph structure and temporal weather conditions to forecast zone-level response time series. We further introduce standardized evaluation protocols for both tasks and conduct cross-building and cross-climate generalization experiments to assess model robustness. ArchEGraph provides a unified testbed for studying geometry-topology-physics coupling in building energy modeling, enabling the development and evaluation of scalable and generalizable surrogate models.
[LG-37] CubicQuant: Parametric Non-Uniform Codebooks for High-Throughput LLM Inference with 1-8-Bit Weights
链接: https://arxiv.org/abs/2608.06763
作者: Xuetian Gao
类目: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC)
*备注: 23 pages, 1 figure. Technical report
Abstract:Weight quantization for large-language-model inference must balance adaptive reconstruction levels with representations regular enough for efficient GPU execution. Uniform integers constrain each group to a linear grid. Low-bit floating-point formats use a fixed exponent-mantissa structure, while learned codebooks gain flexibility at the cost of irregular decoding and additional metadata. We introduce CubicQuant, a parametric non-uniform scalar format that preserves a dense integer code stream while adapting reconstruction levels within each weight group. A monotonic cubic curve, specified by two shape parameters and one scale, maps uniformly spaced magnitude codes to non-uniform levels. The family spans 1-8-bit weight payloads, contains symmetric uniform integer quantization as an exact special case, and has effective width B + 64/G bits per weight for payload width B and group size G. We derive population distortion under Uniform, Gaussian, and Laplace distributions, formulate continuous and Dynamic-A8-carrier-aware fitting objectives, and describe direct packed-weight GPU execution. For finite groups of G=128 with 15,360 samples per distribution, W4 CubicQuant reduced reconstruction RMSE relative to optimally clipped four-bit uniform integer quantization by 3.90% on Uniform, 13.49% on Gaussian, and 28.14% on Laplace samples. Relative to the best enumerated four-bit finite floating-point format, the reductions were 3.90%, 9.44%, and 6.27%. Preliminary H200 kernel measurements show a workload-dependent crossover: model-dtype execution is faster for narrow GEMV, while Dynamic A8 becomes favorable as row count grows. The results establish the format’s representational promise and direct executability; downstream model quality and cross-device end-to-end performance remain open evaluation questions. Comments: 23 pages, 1 figure. Technical report Subjects: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC) Cite as: arXiv:2608.06763 [cs.LG] (or arXiv:2608.06763v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.06763 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-38] Sub-Quadratic Bisimulation Metrics via Approximate Nearest Neighbors: Coverag e-Augmented Guarantees and Computable Two-Sided Certificates
链接: https://arxiv.org/abs/2608.06762
作者: Ibne Farabi Shihab,Joyanta Jyoti Mondal
类目: Machine Learning (cs.LG)
*备注: 16 pages, 4 figures
Abstract:Bisimulation metrics quantify behavioral similarity in Markov decision processes, but their Wasserstein fixed-point operator updates every state pair and incurs quadratic pairwise work. We give a certificate-carrying sub-quadratic method for MDPs with bounded transition support and a useful low-dimensional indexing representation: an approximate-nearest-neighbor index selects the pairs updated by the exact restricted operator, while monotone lower and upper runs enclose the exact metric at every sweep. The main analytical result is a coverage-augmented anytime bound: local index quality alone cannot control global error, because uncovered pairs retain their initialization gap. The limiting error is at most \max(\rho,\eop/(1-\gamma)) , and with exact covered backups the lower arm satisfies |\dann-d|_\infty=\rho . Because \rho depends on the unknown exact metric, the algorithm returns the observable sandwich width instead; agreement of the induced lower and upper clusterings certifies exact recovery of the covered aggregation. A reward-oblivious lower bound shows sub-quadratic index-first coverage cannot remove the coverage term, while a separate adaptive lower bound requires \Omega(|\Scal|) pair evaluations. Exact-operator experiments verify the identity and enclosure in every seeded run, and timing experiments recover quadratic versus sub-quadratic scaling under both cheap and full Wasserstein backups. On the grouped |\Scal|=64 benchmark, exact restricted refinement reaches the exact-metric skyline once retrieval covers roughly half of all pairs, while independently trained MICo and DBC baselines stay 22 - 33\times above that skyline at every retrieval budget. Taxi shows the certificate abstaining under an uninformative embedding, while a 2500 -state gridworld improves over a reward-only metric by 28.6% using 12.8% of one quadratic sweep.
[LG-39] Solver-Guided Reasoning for Mixed-Equilibrium Strategies
链接: https://arxiv.org/abs/2608.06741
作者: Han Wang,Philippe Beardsell,Boning Li,Aaron Sasmita,Shuai Li,Hongyuan Zha,Baoxiang Wang
类目: Machine Learning (cs.LG); Computer Science and Game Theory (cs.GT)
*备注:
Abstract:Reasoning in large language models (LLMs) is often grounded in human text, human demonstrations, and human-generated rationales. For equilibrium reasoning in complex games, however, relying on human data can be suboptimal. In fact, human play is often guided by intuition and heuristics and can deviate substantially from game equilibrium. This discrepancy is amplified in games with mixed-strategy equilibria, where human data is heavily biased toward pure strategies. Consequently, conditioning LLMs on this data yields weak game strategies. To grant LLMs the reasoning capacity in games, in this work, we study how to elicit equilibrium play using solver output. We propose Mixed-Strategy Decision Tree (MDT), which articulates the silent optimality of the equilibrium into sparse strategic rules that both humans and LLMs could understand. Using solver output rather than human annotation allows us to extend the input to arbitrarily new states and continuations. We instantiate this study on No-Limit Texas Hold’em by querying a solver oracle for over \textbf250 million mixed-strategy decisions; MDT together with other techniques \textbfreduces the \ell_1 distance to the equilibrium by 52.6% across 8 different LLM configurations. A Route-only ablation tests the incremental contribution of the shadow-based contrast, while complete River-endgame and Liar’s Dice experiments evaluate strategic fidelity and portability beyond the original NLH communication setting.
[LG-40] A Transferable Autologistic Model for Predicting Rare Failures in Heterogeneous Equipment
链接: https://arxiv.org/abs/2608.06695
作者: Islam Benamirouche,Djemel Ziou,Feriel Fass
类目: Machine Learning (cs.LG)
*备注:
Abstract:Predicting failures before they occur remains a major challenge in predictive maintenance, particularly when failures are rare, when equipment of the same family differ in sensor configurations, and when the goal is anticipation rather than diagnosis of an already observed fault. This paper proposes a common-to-target probabilistic model that learns shared failure-related patterns across a family of heterogeneous equipment and adapts parsimoniously to target equipment. The model explicitly accounts for sensor heterogeneity, operating context, and degradation dynamics to produce calibrated failureprobability estimates suitable for maintenance planning. Its performance is evaluated on a synthetic refrigerator dataset comprising 27 simulated refrigerators with varying sensor configurations, operating conditions, and failure types, providing a controlle
[LG-41] Optimal Neural Network Approximation via Empirical Least Squares with Deterministic Samples
链接: https://arxiv.org/abs/2608.06687
作者: Xinliang Liu,Tong Mao,Jinchao Xu
类目: Numerical Analysis (math.NA); Machine Learning (cs.LG)
*备注:
Abstract:We develop a rigorous theory of discrete residual least-squares approximation for elliptic spectral equations \mathfrak L_\beta u=f using linearized ReLU ^k neural networks on the sphere, where \mathfrak L_\beta is a positive elliptic spectral multiplier of order \beta . Given a parameter set \Theta_n=\theta_j^_j=1^n\subset\mathbb S^d , we approximate u in the linearized network space L_n^k(\Theta_n) by the discrete residual on the collocation points \eta_i^_i=1^m \beginequation* u_n,m\in\arg\min_v_n\in L_n^k(\Theta_n)\frac1m\sum_i=1^m\left(f(\eta_i^)-\mathfrak L_\beta v_n(\eta_i^)\right)^2. \endequation* With k\fracd-12+\beta , for antipodally quasi-uniform network parameter sets and any quasi-uniform collocation points with m\gtrsim n , we prove that \beginequation* |u-u_n,m|\mathcal H^\beta(\mathbb S^d)\eqsim|f-\mathfrak L\beta u_n,m|\mathcal L^2(\mathbb S^d)\lesssim n^-\fracrd \begincases |f|\mathcal W^r,p(\mathbb S^d),\fracdpr\leq \fracd2,~p2,\ |f|\mathcal H^r(\mathbb S^d),r\fracd2. \endcases \endequation* We also establish a high-probability residual estimate, up to a logarithmic factor and an arbitrarily small smoothness loss, for i.i.d.\ uniformly distributed collocation points. The key analytical ingredient is a Bernstein inequality for linearized ReLU ^k network spaces. If \underline h denotes the antipodal separation distance of the network parameters, then \beginequation* |v_n|\mathcal H^r(\mathbb S^d)\lesssim\underline h^-(r-s)|v_n|_\mathcal H^s(\mathbb S^d),\qquad 0\leq srk+\tfrac12. \endequation* Subjects: Numerical Analysis (math.NA); Machine Learning (cs.LG) Cite as: arXiv:2608.06687 [math.NA] (or arXiv:2608.06687v1 [math.NA] for this version) https://doi.org/10.48550/arXiv.2608.06687 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-42] EpiFlow: A framework for improving the utility of wastewater signals for disease forecasting
链接: https://arxiv.org/abs/2608.06671
作者: Aniruddha Adiga,Jingyuan Chou,Gursharn Kaur,Andrew Warren,Srinivasan Venkatramanan,Baltazar Espinoza,Bryan Lewis,Justin Crow,Alexandra Lorentz,Rekha Singh,Madhav Marathe
类目: Machine Learning (cs.LG)
*备注: 11 Pages main text
Abstract:Wastewater-based surveillance is an effective tool for disease monitoring and can provide early warning of outbreaks. Although wastewater viral loads (WVL) correlate with disease burden, their utility for improving real-time forecasting remains under investigation. During the early phases of an epidemic, many indicators can effectively monitor disease spread, but their reliability may decline because of reporting fatigue and low prevalence. Hospital burden can vary substantially even during low-prevalence periods, making accurate forecasting of burden indicators essential for minimizing disease impacts. In this paper, we present principled approaches for processing wastewater data, characterizing its relationship with burden indicators, and generating real-time forecasts. We assess the predictability of WVL using entropy measures. We analyze the relationship between WVL and burden indicators using causality tests that capture temporal dynamics and the leading-indicator behavior of WVL. We incorporate these insights into a time-varying forecasting model that accounts for the evolving relationship between the signals. We also evaluate the effects of delays in WVL reporting through simulations. We test the utility of our methods by forecasting COVID-19 hospital admissions across Virginia and its health regions during periods of varying disease prevalence. Incorporating WVL improves forecast accuracy relative to baseline models, particularly during critical epidemic phases, and results in a 20 percentage point improvement in forecast coverage. Our results demonstrate that WVL signals can improve infectious disease forecasting even under conditions of low prevalence or delayed reporting. Comments: 11 Pages main text Subjects: Machine Learning (cs.LG) Cite as: arXiv:2608.06671 [cs.LG] (or arXiv:2608.06671v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.06671 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-43] Dirichlet Follow-the-Leader Closes the Gap in Simultaneous Multiclass U-Calibration
链接: https://arxiv.org/abs/2608.06656
作者: Pahan Dewasurendra
类目: Machine Learning (cs.LG)
*备注:
Abstract:Can one forecaster attain the optimal regret rate for every bounded proper loss and also adapt to every smooth proper loss? Recent work answered this up to a dimension gap. Its self-concordant perturbation gives roughly K^5/4\sqrtT worst-case regret and incurs an additional \beta\sqrtK\log K for \beta -smooth losses. We close both gaps with a one-line forecaster. After observing class counts c_t-1 , draw the next prediction from \operatornameDir(c_t-1) , on the face of classes seen so far. This is a fresh Bayesian bootstrap of the outcomes. The analysis rests on an exact identity: averaging any bounded proper loss under \operatornameDir(\alpha) equals a discrete derivative of its Dirichlet-averaged Bayes risk. The identity makes the be-the-perturbed-leader term telescope to a nonpositive Jensen gap. A one-count likelihood ratio then bounds stability by the inverse square root of that class’s count. The resulting single, horizon-free algorithm satisfies \sup_\ell\mathbbE\operatornameReg_\ell\leq 4\sqrtS_T T\leq 4\sqrtK T and \mathbbE\operatornameReg_\ell\leq \frac52\beta(1+\log T) for every \beta -smooth proper loss. Here S_T is the number of observed classes. Known lower bounds show that both rates are optimal in their nontrivial regimes. The proof covers nondifferentiable losses and changes of the active simplex face.
[LG-44] he Sparsity Whisperer
链接: https://arxiv.org/abs/2608.06630
作者: Linghao Kong,Inimai Subramanian,Micah Adler,Dan Alistarh,Dan Gutfreund,Nir Shavit
类目: Machine Learning (cs.LG)
*备注: 10 pages, 3 figures. Code available at this https URL
Abstract:Pruning reduces the inference cost of large language models, but existing criteria primarily preserve large activations or reconstruct layer outputs. We argue that this overlooks a key computation performed by particularly sparsity-sensitive neurons in the MLP up and gate projections: separating similar inputs into dissimilar outputs. This suggests that effective pruning should preserve not only activations, but also the differences between outputs more broadly. We introduce a family of difference-informed pruning methods built upon this principle. Wisp is a first-order, update-free method that scores weights using input-difference norms, and Wisp+ refines this score neuronwise using the input pairs each neuron separates most strongly. Finally, Whisper is a second-order method that uses a lightly regularized difference Hessian as its reconstruction objective. Across Llama 2 and 3.1 models from 7B to 405B parameters, our second-order variant consistently improves over strong reconstruction-based baselines, while our update-free variants improve over activation-aware baselines, especially in constrained settings. The improvements over Wanda and SparseGPT extend to structured sparsity, downstream evaluations, and other model families. Augmenting stronger techniques such as RIA and ALPS with our difference-informed criteria yields further improvements, shifting the overall accuracy-runtime frontier outward at negligible additional cost. These results suggest that preserving output differences is a broadly useful and composable signal for post-training LLM sparsification.
[LG-45] Retrofitting Linear Attention into Diffusion Language Models
链接: https://arxiv.org/abs/2608.06628
作者: Jinha Kim,Younghun Roh,Jaeyeon Kim
类目: Machine Learning (cs.LG)
*备注:
Abstract:Diffusion language models (dLLMs) offer a promising alternative to autoregressive models by accelerating inference through parallel decoding. Recent dLLMs commonly use blockwise semi-autoregressive decoding, generating blocks autoregressively while denoising tokens within each active block in parallel. However, despite KV caching, each denoising step still attends to all previous blocks, repeatedly incurring prefix-attention cost. Motivated by this bottleneck, we ask whether dLLM inference can be further accelerated by linearizing attention over previous blocks. We introduce block-hybrid attention, which retains exact softmax attention within the active denoising block while applying linear attention over previous blocks. We show that this hybrid attention can be retrofitted into a pretrained dLLM with minimal post-training: LLaDA-Hybrid replaces 6 of the 20 attention layers in LLaDA~2.1, a 16B open-source dLLM, largely following LoLCAT (Zhang et al, 2024). The conversion takes only approximately 60 hours while preserving benchmark performance: 72.0% vs. 75.6% on HumanEval, 63.0% vs. 57.7% on MBPP+, and 86.7% vs. 88.3% on CMATH. With a Triton implementation, LLaDA-Hybrid achieves up to 1.7\times higher decoding throughput and supports more concurrent requests before exhausting memory, showing that pretrained dLLMs can be efficiently linearized for faster inference. Our code is available at: this https URL.
[LG-46] CrystalGRPO: Target-Aligned and Coverag e-Preserving Reinforcement Learning for Flow-Based Crystal Structure Prediction
链接: https://arxiv.org/abs/2608.06582
作者: Kaixiang Su,Hongfei Xue,Qiang Zhu
类目: Machine Learning (cs.LG)
*备注:
Abstract:Flow-based generative models can efficiently produce candidate structures for crystal structure prediction (CSP), but their pretrained objectives do not directly optimize downstream target recovery. Reinforcement-learning post-training offers a flexible solution, yet existing approaches rely primarily on energy rewards and coordinate-only stochastic policies. Predicted energy does not identify the reference polymorph, while reward-driven concentration can reduce the candidate coverage required for Top-N recovery. We introduce CrystalGRPO, a CSP-aligned post-training framework that extends existing ODE-to-SDE policy constructions to the joint coordinate–lattice state. CrystalGRPO combines MACE-predicted energy with a StructureMatcher-based recovery score and provides two operating modes: CrystalGRPO-Q, which prioritizes single-draw recovery, and CrystalGRPO-C, which combines full-trajectory reference regularization with a coverage-aware group advantage to preserve finite-budget target recovery. Across MP-20 and MPTS-52 with PXRDGen and OMatG backbones, both variants reduce one- and twenty-sample RMSE relative to coordinate-only reinforcement in all four backbone–dataset settings. CrystalGRPO-Q consistently improves Top-1, whereas CrystalGRPO-C achieves a higher Top-20 across all settings.
[LG-47] heoretical Foundations of Communication-Efficient Robust and Practical Distributed and Federated Optimization
链接: https://arxiv.org/abs/2608.06563
作者: Grigory Malinovsky
类目: Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注: thesis, 7 chapters
Abstract:Machine learning and optimization have advanced together, with practical demands motivating new theory and theoretical breakthroughs enabling new applications. Modern large-scale training relies on classical optimization principles, but the constraints of distributed systems require these foundations to be reconsidered. This thesis addresses seven challenges at the intersection of theory and practice, focusing on key bottlenecks in federated learning and distributed optimization. First, we introduce ProxSkip and prove that local gradient steps can accelerate communication, providing a theoretical foundation for this widely used heuristic. Second, we develop Variance Reduced ProxSkip, which eliminates the neighborhood error of stochastic local updates while balancing communication and local computation. Third, we show that local steps retain their communication acceleration under partial client participation. Fourth, we prove that server-side stepsizes and sampling without replacement improve convergence in heterogeneous settings. Fifth, for Random Reshuffling, we demonstrate that compressing gradient differences rather than gradients yields better theoretical and practical performance. Sixth, we establish that Byzantine robustness and partial participation can be achieved simultaneously using gradient-difference clipping. Finally, we develop the first theoretical framework for low-rank adaptation based on randomized asymmetric chains, providing new insights into fine-tuning large models. Across these contributions, we introduce novel algorithmic frameworks, establish sharp guarantees under realistic assumptions, and support the theory with numerical experiments.
[LG-48] Bootstrap-Conditioned Action Selection with Tabular Foundation Models
链接: https://arxiv.org/abs/2608.06559
作者: Devansh Gupta,Shiv Tavker,Dmitry Efimov,Suchitra Sathyanarayana,Gitanjali Bhutani,Boris N. Oreshkin
类目: Machine Learning (cs.LG)
*备注:
Abstract:Contextual bandits offer a natural framework for sample-efficient personalization, but practical deployment remains difficult under sparse, biased interaction data, unreliable uncertainty estimates, and severe cold starts. We study whether pre-trained tabular foundation models with in-context learning can be turned into randomized policies for online decision making. We propose BC-ICL (Bootstrap-conditioned action selection using ICL), which at each round draws a bootstrap resample of the interaction history, conditions a frozen pre-trained ICL model on that resample, scores all actions, and selects the action with the highest sampled score. We further introduce an arm-context conditioning architecture that promotes shared statistical strength across actions and helps avoid common bootstrap failure modes of isolated-arm bandits. Empirically, this policy delivers strong early-round regret and regret performance on standard contextual bandit suites, outperforming established baselines under a strict online protocol.
[LG-49] Cascade: Exploiting SLO-Aware latency budget for fair and high goodput LLM inference serving
链接: https://arxiv.org/abs/2608.06557
作者: Muhammad Adnan,Rohan Mahapatra,Prashant J. Nair,Daniel Berger,Pantea Zardoshti,Rodrigo Fonseca,Esha Choukse
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG)
*备注:
Abstract:The reasoning and agentic capabilities of large language models have expanded the range of applications they support, from short interactive exchanges to long, compute-heavy requests. LLM serving platforms today define response-latency service-level objectives, even though requests within the same service can differ by orders of magnitude in input length, generation length, execution cost, and the availability of reusable KV-cache state. As a result, requests governed by the same service level objective have different urgency: after accounting for the time required to execute them, some have substantial latency headroom while others have almost none. We define this headroom—the difference between a request’s service level objective and its predicted remaining service time—as its per-request latency budget. We present Cascade, an LLM serving system that estimates and continuously updates this budget from request characteristics, KV-cache state, and current system load. Unlike prior SLO-aware schedulers that use deadlines to govern request ordering alone, Cascade uses a single per-request budget to jointly coordinate request scheduling and KV-cache management across the memory hierarchy. Its scheduler prioritizes requests with little remaining budget, while its memory manager uses the same budget to decide whether non-resident KV state should be restored or prefetched from a deeper tier, retained in HBM, or recomputed. By directing queueing and data-movement overhead toward requests that can absorb it, Cascade improves SLO-satisfied goodput while preserving fairness across heterogeneous request classes. On production traces across three large language models, Cascade improves goodput by up to2.4x and reduces SLO violations by 40% relative to the default vLLM first-come, first-served scheduler. Subjects: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG) Cite as: arXiv:2608.06557 [cs.DC] (or arXiv:2608.06557v1 [cs.DC] for this version) https://doi.org/10.48550/arXiv.2608.06557 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-50] Newton-Schulz Retraction-Based Inference Enables Hidden Quantum Markov Models to Outperform Classical HMMs
链接: https://arxiv.org/abs/2608.06554
作者: Ning Ning
类目: Machine Learning (cs.LG); Quantum Physics (quant-ph); Methodology (stat.ME)
*备注:
Abstract:Hidden Markov models (HMMs) are widely used probabilistic models for discrete sequential data but can be limited when hidden dynamics are complex. Hidden quantum Markov models (HQMMs) generalize HMMs by replacing probability vectors with density matrices and stochastic transitions with quantum operations, enabling richer latent representations. However, existing HQMM learning methods have not consistently outperformed Expectation–Maximization (EM)-trained HMMs on data not generated by quantum processes, limiting their practical applicability. We introduce NS-RIS, Newton–Schulz Retraction-based Inference on the Stiefel manifold, a scalable algorithm for learning trace-preserving HQMMs. NS-RIS uses Newton–Schulz orthogonalization to compute a polar-factor search direction while preserving Stiefel-manifold feasibility, avoiding costly matrix decompositions. We further establish a finite-time stationarity guarantee under standard assumptions on smoothness, stochastic gradients, and finite Newton–Schulz accuracy. Empirically, NS-RIS provides the first benchmark evidence that an HQMM can significantly outperform an EM-trained HMM on data not generated by a quantum model. On synthetic HMM-generated benchmarks, NS-RIS outperforms both EM and the state-of-the-art HQMM method COSM, improving the evaluation metric by an average of 38.5% and by up to 50.6%. On a synthetic HQMM benchmark, it improves the test metric over COSM by 18.9% while reducing runtime by 12.0%. On the real-world Splice classification benchmark, NS-RIS also surpasses both EM and COSM in higher-dimensional latent regimes, reducing mean classification error by 17.9% for latent dimension 6 and 14.9% for latent dimension 8 relative to COSM. These results move HQMMs beyond a theoretical generalization of HMMs and establish them as practical and expressive models for scientific sequence data.
[LG-51] Robust Averag e-Reward Markov Decision Processes: Minimax-Optimal Learning via Plug-in Reductions
链接: https://arxiv.org/abs/2608.06545
作者: Yuepeng Yang,Yuxin Chen,Yuejie Chi
类目: Machine Learning (cs.LG); Optimization and Control (math.OC); Machine Learning (stat.ML)
*备注:
Abstract:Distributionally robust Markov decision processes provide a principled framework for sequential decision making under model uncertainty. We study how many samples are necessary and sufficient to learn an \varepsilon -optimal robust policy under the average-reward criterion. A generative model provides samples from the nominal transition kernel, whereas policy performance is evaluated over (s,a) -rectangular total-variation uncertainty sets of radius at most \sigma . Let H_0 and H_\sigma denote the nominal and robust optimal bias spans, respectively. We identify \sigma H_0 as the perturbation scale separating high- and low-tolerance regimes. Our matching upper and lower bounds show that, up to logarithmic factors, the minimax total sample complexity is NSA \asymp \fracSA\varepsilon^2\begincases \min\H_0,H_\sigma, \varepsilon\gtrsim\sigma H_0,\ \min\H_0,H_\sigma+\sigma H_\sigma^2, \varepsilon\lesssim\sigma H_0. \endcases Here S and A are the numbers of states and actions, and N is the number of samples per state-action pair. The sample complexity consists of a linear-span term that resembles the nominal AMDP results and a robustness-specific term that appears only in the low-tolerance regime. We attain these rates using reduction-based plug-in procedures that select the reduction—nominal or robust—and its discount factor: a span-informed procedure that makes these choices using known span parameters, and a span-agnostic procedure that calibrates both choices from data. Subjects: Machine Learning (cs.LG); Optimization and Control (math.OC); Machine Learning (stat.ML) Cite as: arXiv:2608.06545 [cs.LG] (or arXiv:2608.06545v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.06545 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-52] Online Security Learning in Cooperative Multi-Agent Systems under Hidden Byzantine Attacks
链接: https://arxiv.org/abs/2608.06520
作者: Ximing Sun,Yue Wang
类目: Machine Learning (cs.LG)
*备注: preprint; Work in progress
Abstract:We study online cooperative control of a multi-agent system under Byzantine attacks. Namely, an unknown, fixed subset of agents are Byzantine comprised and can stealthily overwrite its own coordinates of the team’s planned joint action after observing that plan. The learner observes planned actions, public rewards, and public states, but neither the overwrite nor the executed joint action. Our objective is security: to optimize the team performance against the worst overwrites and achieve the optimal security value. We first show that the attacker’s information determines the geometry. An attacker that observes the planned action induces an exact (s,a) -rectangular robust Markov decision process (MDP) whose rows are convex hulls of overwrite-induced public-outcome laws, whereas a blind attacker induces an s -rectangular model. We then identify the information-theoretic limit of security learning, showing that the security regret decomposes exactly into return regret against the response generating the data and a cumulative response gap D_K . Two indistinguishable horizon-one instances force \Omega(K) expected security regret while return regret is zero, showing that dependence on D_K is unavoidable. Finally, we develop a stage-tied robust estimation-to-decisions learner and prove a regret bound of \widetilde\mathcal O!\left(H^2S\sqrtAK\right)+\mathbb E[D_K] . Our studies thus provide comprehensive theoretical and algorithmic foundations of reliable multi-agent systems under Byzantine attacks.
[LG-53] arget-Weighted Neyman Allocation: Experimental Design for Heterogeneous Treatment Effects under Population Shift
链接: https://arxiv.org/abs/2608.06512
作者: Hoang Dang,Luan Pham,Minh Nguyen
类目: Machine Learning (cs.LG); Methodology (stat.ME)
*备注:
Abstract:Randomized experiments are often run in one population to guide decisions in another. Allocating by experimental proportions wastes budget on groups that rarely appear in deployment, whereas allocating by deployment proportions under-samples groups that are hard to measure precisely. We propose \textbfTWNA (Target-Weighted Neyman Allocation), a two-stage stratified design that uses pilot estimates of group–arm outcome variances to allocate final-stage sample sizes and treatment probabilities for target-weighted group average treatment effect (GATE) precision. The oracle rule has a closed form and balances deployment importance with statistical difficulty; the plug-in rule recovers it as pilot variance estimates stabilize. We also extend TWNA to handle uncertainty about deployment composition, remaining robust whether the target mix is roughly known or entirely unknown. Finally, we distinguish this weight robustness from a pilot-robust variant for skewed, rare-event, or contaminated outcomes. Simulations and real-covariate benchmarks show the largest gains when groups are both deployment-important and difficult to measure.
[LG-54] Unmasking Removal-Budget Confounding: A Matched Operating-Point Evaluation Framework for Adaptive Data Cleaning
链接: https://arxiv.org/abs/2608.06511
作者: Wei-Hsiang Chen,Pin-Hsuan Yu,Chen-Hsuan Fang,Jung-Hua Wang
类目: Machine Learning (cs.LG)
*备注:
Abstract:Adaptive data-cleaning methods replace manual filtering thresholds with data-driven partitions. However, changing the partition granularity, the number of groups used to segment samples by estimated corruption risk, can implicitly shift the decision boundary and alter the overall number of removed samples. This creates a bias known as removal-budget confounding, where apparent gains in metrics like precision or false-positive rate reflect a smaller removal budget rather than superior corruption discrimination. To address this evaluation bias, we introduce an operating-point-aware evaluation framework that evaluates methods using matched-budget and matched-recall controls alongside threshold-independent metrics (AUROC and AUPRC). We test this framework on a multi-cue adaptive cleaner redesign featuring a reweighted learning-difficulty cue, an auxiliary Euclidean-distance cue, and increased partition granularity intended to isolate clean-but-difficult samples. While naive evaluations (assessing configurations at their own induced operating points) suggest substantial performance improvements for the redesign, these gains disappear once operating points are equalized. False-positive decomposition reveals that clean-but-difficult samples primarily drive error counts at low corruption rates, become threshold-dependent at moderate corruption, and contribute negligibly under severe corruption. Experiments on CIFAR-10 and ImageNet-100 demonstrate that most performance differences observed in naive evaluation shrink or vanish at low-to-moderate corruption when operating points are matched. True ranking advantages only remain in specific low-prevalence settings and in high-recall regions under severe corruption. These findings highlight that adaptive cleaning methods must be benchmarked at matched operating points to ensure performance gains reflect genuine corruption discrimination.
[LG-55] oward Reliable Context Compression for Long-Horizon Agents : An Empirical Study of Execution Instability
链接: https://arxiv.org/abs/2608.06503
作者: Guanghui Min,Liang Wu,Mayank Darbari,Chen Chen,Liangjie Hong
类目: Machine Learning (cs.LG)
*备注: 31 pages, 6 figures
Abstract:Recurrent context compression controls context growth in long-horizon agents, but its behavioral effects remain poorly understood. In this preliminary empirical study, we show that compression can weaken the influence of recent interactions, increasing blocked actions, repeated exploration, and instability across runs. Motivated by these observations, we introduce TRACE, a verifier-guided framework that evaluates individual compaction events through paired closed-loop continuations from the same environment state and uses summary preferences to optimize a natural-language compression prompt while keeping all models frozen. Initial results on AppWorld show improvements over existing compression baselines in task performance, multi-run reliability, and context–execution efficiency. These findings provide early evidence for boundary-local evaluation as a promising direction for reliable agent context compression.
[LG-56] Beyond Attention: Signed Integrated Gradients Attribution in a BiomeGPT -Style Microbiome Transformer
链接: https://arxiv.org/abs/2608.06486
作者: Oren Nelson
类目: Machine Learning (cs.LG)
*备注: 15 pages, 6 figures. Code: this https URL
Abstract:In a feature-tokenized transformer (arXiv:2106.11959) such as BiomeGPT (doi:https://doi.org/10.64898/2026.01.05.697599), each input token is built by fusing a fixed identity with a sample-specific measurement: a fixed species and a variable abundance, T = S + A. To interpret downstream classification in such models, prior work inspects the attention weights of the special [CLS] token (arXiv:2106.11959, arXiv:1810.04805, BiomeGPT) to rank sample tokens by importance. These weights have two critical limitations: they are nonnegative, so they cannot separate disease-supporting from health-supporting evidence (arXiv:2201.12114), and they act after token fusion, obscuring how the input sources S and A each affect the output. To address this we use Integrated Gradients (arXiv:1703.01365), a signed, fusion-aware attribution method, and propose a source-derived baseline T’ = S + A_0 for feature-tokenized models such as BiomeGPT, which preserves species identity as a fixed biological coordinate while isolating the effect of abundance variation. Applied to a disease-versus-health decision margin, it yields polarity that explicitly separates pathogenic from protective microbial signals. We show that this gradient-based approach uncovers species-abundance directional relationships and sensitivity diagnostics entirely obscured by unsigned [CLS] attention weights. We further recommend second-order Integrated Hessians (arXiv:2002.04138) to expose microbiome community interaction rules: how a perturbation in one member alters the model’s sensitivity to another, and which other species drive ambiguous cases toward disease or health at a given abundance level. This provides a principled approach to explainability in BiomeGPT that generalizes to other smooth and differentiable feature-tokenized transformers. Code is available at this https URL Comments: 15 pages, 6 figures. Code: this https URL Subjects: Machine Learning (cs.LG) Cite as: arXiv:2608.06486 [cs.LG] (or arXiv:2608.06486v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.06486 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-57] Fairis: Fairness-Aware Aggregation with Provable Influence Containment against Fairness Poisoning Attacks in Collaborative Machine Learning
链接: https://arxiv.org/abs/2608.06469
作者: Devharsh Trivedi,Nesrine Kaaniche,Nikos Triandopoulos,Maryline Laurent,Jackson Walters
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注: 37 pages, 6 figures. Extended version, adding a size-weighted variant, a clipping-based influence bound, and machine-checked proofs of the main results
Abstract:Collaborative machine learning among financial institutions must be both group-fair and robust against deliberate adversarial manipulation. Existing fairness-aware aggregation methods remain formally vulnerable to fairness poisoning: a malicious client maximizing group disparity while preserving accuracy evades accuracy-based Byzantine defenses, and in our threat model FairFed’s gap-based weighting can be gamed by an adversary who observes the global fairness score. We present Fairis, a server-side reweighting scheme in which each client’s update receives the normalized weight \omega_k = \barw_k / \sum_j \barw_j built from the unnormalized score \barw_k = \eta - \mathcalF_k , with \mathcalF_k \in [0,1] the local Equal Opportunity Difference and \eta 1 a security parameter. We prove three properties, Monotone Weight Reduction (MWR), Demographic Participation, and Non-Gamesmanship, extend MWR to colluding minority coalitions, and show that combining MWR with server-side norm clipping bounds the adversary’s displacement of the global model by \omega_0 C , strictly decreasing in its own reported disparity. Assuming honest score reporting, an assumption this paper does not discharge, Fairis is the only rule evaluated that guarantees every client strictly positive weight while provably reducing an adversary’s weight monotonically in its bias; clipped FairFed can reach a lower weight but guarantees nothing and zeroes a client outright on Taiwan Credit. Against an adversary stealthy enough to evade accuracy-based defenses, within 0.04 accuracy of benign, Fairis cuts its weight by 41 to 54% below a size-blind control on Taiwan. On routine non-IID partitions no rule dominates, and a uniform-weighting ablation shows that containment tracks how far the adversary’s score separates from the honest mean, providing none when the honest population is already unfair.
[LG-58] FedTransKD-IDS: Robust Federated Transfer Learning with Knowledge Distillation for Intrusion Detection in IoT
链接: https://arxiv.org/abs/2608.06447
作者: Mohammad Hosssein Gholamrezazadeh,Ahmadreza MontazerolghaemAhmadreza Montazerolghaem
类目: Networking and Internet Architecture (cs.NI); Machine Learning (cs.LG)
*备注:
Abstract:In modern distributed network environments, particularly in Internet of Things infrastructures and 5G networks, stringent privacy preservation and scalability requirements have created significant challenges for intrusion detection systems. Although federated learning preserves privacy by preventing data centralization, its efficiency and stability is considerably degraded under severe statistical heterogeneity and resource constraints of edge nodes. To address these limitations, this study introduces the FedTransKD-IDS framework, which enhances both system stability and efficiency by integrating robust aggregation based on the geometric mean, federated transfer learning, and knowledge distillation. Within this framework, the collaboratively trained global teacher model transfers its feature extraction component to lightweight student models. Experimental evaluation on heterogeneous datasets demonstrates a peak detection performance, achieving an accuracy of 99. 18% and a recall of 99. 99%, thereby indicating the effectiveness of structured knowledge transfer in federated environments.
[LG-59] SNI-GNN: SmartNIC-Assisted Full-Graph GNN Training with In-Network Embedding Prediction ICDE2026
链接: https://arxiv.org/abs/2608.06441
作者: Guofan Yu,Sitian Chen,Zhenheng Tang,Xiaowen Chu,Amelie Chi Zhou
类目: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC)
*备注: Camera-ready version accepted at ICDE 2026. 14 pages, 18 figures
Abstract:Full-graph GNN training delivers high accuracy but scales poorly on multi-server clusters due to heavy, irregular inter-node embedding exchanges. We present SNI-GNN, a SmartNIC-assisted full-graph training system that reduces communication while preserving accuracy by predicting remote embeddings in-network. SNI-GNN deploys a lightweight linear-trend predictor on SmartNICs to refine cached historical embeddings, coupled with an importance-based boundary-node sampling policy and an asynchronous DPU–GPU data pipeline with intermediate-result reuse. We provide error and convergence bounds showing that predictor bias remains controlled under bounded second-order dynamics and yields standard non-convex convergence with inexact gradients. Implemented on NVIDIA BlueField-3, SNI-GNN integrates with state-of-the-art full-graph systems, cuts communication by 21–45%, achieves 1.3–3.6 \times end-to-end speedups over BNS-GCN and up to 1.29 \times over baseline SANCUS, with accuracy loss \leq 0.01 , and scales efficiently to 16 GPUs on graphs with up to tens of millions of edges. These results indicate SmartNIC-based in-network prediction is a practical complement to partitioning and compression techniques for communication-efficient full-graph GNN training at scale.
[LG-60] Fast and Accurate: An Adaptive VLA Inference Framework through Environment-aware Model Selection
链接: https://arxiv.org/abs/2608.06434
作者: Yuewei Sun,Lang Qin,Zechuan Tian,Jingwen Li,Guiqin Wang,Shengzeng Huo,Wenxin Ren,Tao Fang,Xiaochen Zhang,Guanqing Deng,Xiang Wang,Xiaowen Dong,Qinghai Guo,Yuxin Ma
类目: Robotics (cs.RO); Machine Learning (cs.LG)
*备注:
Abstract:Embodied intelligence demands both long-horizon reasoning and real-time closed-loop responsiveness. Recent dual-system Vision-Language-Action (VLA) architectures combine fast reactive control with slow deliberative reasoning to balance inference speed and task success rate. However, existing dual-process VLAs tightly couple the fast module to intermediate representations of the slow module, necessitating end-to-end joint training and limiting modularity, extensibility and flexible system switching. In this paper, we propose Environment-aware Model Selection (EMS), an adaptive VLA inference framework that switches between two fully decoupled systems of different scales through environment-aware model selection. The large-scale deliberative system provides globally consistent trajectory planning to ensure task success, while a lightweight reactive system enables high-frequency closed-loop control. A reinforcement-learning-based switching policy dynamically selects which system to invoke based on real-time feedback, enabling sparse use of the slow system and thereby balancing pretrained knowledge utilisation with runtime efficiency. Our design offers three key advantages over prior hierarchical VLA frameworks: (1) a fully decoupled and modular dual-system architecture that supports plug-and-play model replacement; (2) an adaptive, environment-aware switching strategy; (3) high-frequency inference for responsive closed-loop control. We extensively evaluate EMS in both simulation and real-world environments. On the LIBERO benchmark, EMS achieves success rates comparable to the large-scale baseline while increasing the effective action frequency to 93.4 Hz. The framework further demonstrates strong extensibility in real-world dual-arm manipulation tasks, where it accelerates task completion while maintaining robust performance.
[LG-61] MiGHT-EHR: A Multi-task Graph Transformer for Heterogeneous Temporal Electronic Health Records
链接: https://arxiv.org/abs/2608.06430
作者: Anirudh Rayas,Yuan Wang,Pavan Turaga
类目: Machine Learning (cs.LG); Quantitative Methods (q-bio.QM)
*备注:
Abstract:Learning from Electronic Health Records (EHRs) has gained significant attention due to its potential to improve clinical prediction. However, effective learning remains challenging because EHRs encode heterogeneous, temporally ordered clinical interactions. In particular, EHRs contain: (i) heterogeneous clinical entities, including patients, visits, diagnoses, prescriptions, and procedures, together with their heterogeneous interactions, (ii) longitudinal patient trajectories across hospital visits and (iii) shared statistical dependencies across related clinical prediction tasks. Existing EHR learning methods capture only a subset of these properties. To bridge this gap, we propose Multi-task Graph transformer for Heterogeneous Temporal EHRs (MiGHT-EHR), which jointly models all three within a unified representation learning method. MiGHT-EHR constructs a heterogeneous graph from EHRs in which nodes represent clinical entities and edges connect statistically associated entities identified via normalized point-wise mutual information. Across MIMIC-III and MIMIC-IV datasets, MiGHT-EHR outperforms state-of-the-art methods on average across four tasks: drug recommendation, prediction of length-of-stay, mortality, and readmission, with particularly strong improvements in mortality and readmission prediction. Furthermore, a post-hoc analysis of the learned representations reveals that patient neighborhoods are organized by clinical outcomes, salient medical concepts are recoverable as linear directions in the representation space, and task probabilities are well calibrated. Collectively, these findings demonstrate that MiGHT-EHR representations support diverse prediction tasks while preserving clinically interpretable structure.
[LG-62] Fixed and Adaptive Topological DeepONets: Functional Measurements on Hausdorff Locally Convex Spaces
链接: https://arxiv.org/abs/2608.06428
作者: Khemraj Shukla,George Em Karniadakis
类目: Machine Learning (cs.LG); Mathematical Physics (math-ph); General Topology (math.GN)
*备注:
Abstract:Deep Operator Networks (DeepONets; arXiv:1910.03193) typically encode an input function through point values on a fixed discretization. Building on the Topological DeepONet framework of Ismailov (arXiv:2603.11972), we replace point samples by continuous linear functionals drawn from the continuous dual of a Hausdorff locally convex space (V,\p_\alpha_\alpha\in A) , whose topology is generated by a point-separating family of seminorms rather than a single norm, and develop fixed and adaptive functional measurement systems. Measurements are combined with the coefficient-space Two-Step procedure of Lee and Shin (arXiv:2309.01020), while a training-only decoder and regularization stabilize the adaptive coordinates. We derive a discrete error decomposition separating measurement, output-basis, and neural-approximation errors, together with a Barron-rate refinement. The framework is evaluated on the antiderivative operator, a non-normable locally convex input space, heterogeneous Darcy flow, a controlled operator, and fixed-time and time-evolving Navier-Stokes vorticity operators. In the heterogeneous Darcy problem, the functional models retain nearly resolution-independent errors of 5.5-5.6% on unseen grids, while in the controlled problem adaptive measurements reduce the mean error below 1.2%. For the fixed-time Navier-Stokes problem, the Adaptive Topological DeepONet is the most accurate DeepONet-based model, attaining a mean relative L^2 error of 1.685% +/- 0.017% using 128 functional coordinates. A comparably sized Fourier neural operator (FNO; arXiv:2010.08895) achieves the lower error 0.832% +/- 0.172%, but requires the full 64x64 input field, twice the training time, and 10.7x greater peak GPU memory. The formulation provides compact, interpretable, and discretization-portable coordinates in the continuous dual V’ , including for non-normable input spaces.
[LG-63] Adversarial Causal Intervention Falsification
链接: https://arxiv.org/abs/2608.06427
作者: Mojtaba Eslami
类目: Machine Learning (cs.LG); Computer Science and Game Theory (cs.GT); Econometrics (econ.EM); Methodology (stat.ME)
*备注:
Abstract:Generative models can reproduce an observational distribution while encoding an incorrect causal structure. We study a sequential game in which a structural causal generator proposes observational and interventional distributions, while an adversarial experimentalist selects interventions intended to maximally falsify the generator. The discriminator is therefore not merely a real-versus-synthetic classifier: it is indexed by an intervention and tests whether the generator reproduces the corresponding post-intervention law. We introduce Adversarial Causal Intervention Falsification (ACIF), formulate oracle and implementable versions of the game, and distinguish three objects that are often conflated: observational fit, interventional equivalence over an admissible query class, and point identification of a structural causal model. For finite model and intervention classes, we prove: (i) an exact reduction of the adversarial objective to a worst-intervention integral probability metric; (ii) identification up to interventional equivalence, with point identification under a separating intervention family; (iii) existence of mixed-strategy equilibria; (iv) finite-sample uniform convergence and margin-based model-selection guarantees; and (v) a logarithmic elimination guarantee for a disagreement-driven sequential design under a balanced-separation condition. We also give a complete linear-Gaussian example in which two observationally indistinguishable causal directions are separated by a single well-chosen intervention. The framework clarifies what an adversarial causal discriminator can and cannot certify, and provides a principled bridge between causal generative modeling, active causal discovery, and experimental design.
[LG-64] Sharding Prevents LLM Oversight Failures and Adversarial Exploitation
链接: https://arxiv.org/abs/2608.06422
作者: Victor Akinwande,J. Zico Kolter,Aran Nayebi
类目: Machine Learning (cs.LG)
*备注:
Abstract:Giving an LLM judge more compute does not necessarily make it check more requirements. When one call must return many verdicts, some decisions become weakly grounded in the evidence, even when that call receives the same token or tool budget as a panel of separate calls. Across expert-graded research replications, legal work, and clinical-trial assessments, agreement with experts falls as the number of verdicts per call grows. We identify sharding as the intervention that mitigates this failure in model-based oversight. Sharding partitions the requirements into smaller groups, assigns each group to a separate call, and aggregates the verdicts. Against a single call with the panel’s full budget, sharding improves agreement while holding the model, evidence, total budget, and per-decision budget fixed. Overall, we find that a sharded weaker judge can outperform a more capable holistic judge and match that judge even when the latter receives the panel’s full budget. Additionally, we find that sharding exhibits robustness against adversaries. A best-of-N adversary can hold the underlying work fixed, vary only its presentation, and increase an overloaded judge’s acceptance of genuinely unmet criteria severalfold. Wherever sharding reduces baseline error, it removes this adversarial advantage, keeping over-acceptance low even as the adversary’s search widens. Sharding does not address attacks that persuade the judge separately on each criterion rather than exploiting overload. In that setting, we find that debate-style opposition on top of sharding withstands such adaptive re-optimization.
[LG-65] Certified Feedforward Tracking for Unknown Nonlinear Systems via Invertible Neural Networks
链接: https://arxiv.org/abs/2608.06419
作者: Berk Altiner,Rajasree Sarkar,Arunava Banerjee,Zongxuan Sun,Kenneth Kim
类目: ystems and Control (eess.SY); Machine Learning (cs.LG)
*备注: 6 pages, 6 figures
Abstract:In this paper, we address the certification of datadriven feedforward control for periodic tracking of unknown nonlinear systems under partial state measurements. To this end, we adopt an invertible neural network (INN) as a surrogate for the unknown system. This choice allows us to bypass solving a nonconvex inversion problem, eliminating the associated inversion errors and reducing tracking error certification to a surrogate modeling problem. We then apply conformal prediction to provide finite-sample probabilistic guarantees on the surrogate modeling error which, through the derived tracking error bound, yield marginal certificates on feedforward tracking error. Finally, we demonstrate the approach on a DC-motor-driven mechanical load with nonlinear friction.
[LG-66] RIS-Aided mmWave Localization Under Cross-Link Interference via Beam-Domain ML Fingerprinting
链接: https://arxiv.org/abs/2608.07444
作者: Md Tarek Hassan,Dmitry Zelenchuk,Muhammad Ali Babar Abbasi
类目: ignal Processing (eess.SP); Emerging Technologies (cs.ET); Machine Learning (cs.LG)
*备注: Accepted in 2026 IEEE Global Communications Conference (GLOBECOM), Macau, China, 2026
Abstract:Accurate user equipment (UE) localization is critical for beam management in reconfigurable intelligent surface (RIS)-assisted millimeter-wave (mmWave) based sixth-generation (6G) networks, especially if the direct base-station-UE links are unavailable. This paper proposes a beam-domain fingerprint framework that maps the received signal-to-noise ratio (SNR) across a small set of predefined RIS reflection states to the UE azimuth angle and range, without requiring channel state information (CSI). Crucially, we extend the framework to a realistic interference-impaired scenario in which a nearby cross-link interferer (CLI) corrupts the clean SNR fingerprint, yielding a signal-to-interference-plus-noise ratio (SINR) fingerprint; an interference-to-noise ratio (INR)-constrained calibration strategy keeps the interference level physically interpretable. Four machine-learning (ML) regressors are evaluated under both conditions. Simulation results at 28 GHz with a 20x20 RIS show that k-nearest neighbors (KNN) achieves the lowest angle MAE of 0.37 degrees and range MAE of 4 cm under clean conditions, rising to 1.4 degrees and 7.6 cm under interference. A key finding is that interference degrades angle estimation substantially more than range estimation across all models, a consequence of the asymmetric encoding of location information in the beam-domain fingerprint.
[LG-67] Wasserstein Policy Gradient for Entropy-Regularized Linear-Quadratic Control
链接: https://arxiv.org/abs/2608.07433
作者: Zhaoyu Zhu,Rui Gao,Shuang Li
类目: Optimization and Control (math.OC); Machine Learning (cs.LG)
*备注:
Abstract:Wasserstein policy gradient (WPG) updates state-conditional action laws by transport in the action space. We study entropy-regularized discounted linear-quadratic (LQ) control. A Bellman verification argument shows that the unrestricted problem has a linear-Gaussian optimal policy, and the discounted-occupancy-weighted statewise Wasserstein gradient is tangent to this policy class. WPG therefore reduces exactly to a finite-dimensional ODE for the feedback gain and action covariance. We prove that this ODE is globally well posed and converges exponentially from every admissible initialization. For each fixed LQ problem, the exponent has a positive limit as the entropy temperature tends to zero and contains no perturbative factor of the form \exp(-c/\tau) , while retaining the usual dependence on the conditioning of the control problem.
[LG-68] DynaCrys: Crystal Generation with Dynamic Space-Group Diffusion
链接: https://arxiv.org/abs/2608.07401
作者: Zhuotao Jin,Xiaoyun Wang,Nicholas Brawand,Roman Zubatyuk,Atul Thakur,Eric Qu,Boris Kozinsky,Justin Smith
类目: Materials Science (cond-mat.mtrl-sci); Machine Learning (cs.LG)
*备注:
Abstract:The search for new crystalline materials spans an enormous compositional and structural space. Generating candidates in this space requires jointly modeling discrete crystallographic symmetry, elemental composition, and continuous geometry. We introduce DynaCrys, a generative model for crystals in which the space group co-evolves with Wyckoff occupations and elements through a coupled symbolic diffusion process. The structured space-group transitions follow crystallographic group-subgroup relations. As the space group changes, a shared, pretrained symmetry codebook provides both the legality-constrained stochastic decoder and the symmetry-constrained crystal-geometry model with a common representation of the corresponding Wyckoff vocabulary. Across large-scale evaluations using two independent relaxation-and-evaluation engines, DynaCrys achieves best-in-class performance in symmetry-aware discovery of stable, unique, and novel crystals, both overall and under the additional requirement of nontrivial post-relaxation symmetry. It also enables fast sampling while generating structures with consistently low relaxation-induced structural displacements.
[LG-69] High-dimensional ridgeless least squares interpolation under spiked covariance structures
链接: https://arxiv.org/abs/2608.07281
作者: Zhijun Liu,Dandan Jiang
类目: atistics Theory (math.ST); Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:This paper investigates the asymptotic behavior of the out-of-sample prediction risk of the high-dimensional ridgeless least-squares estimator when the feature dimension p and the sample size n grow proportionally. We consider a generalized spiked population covariance model with multiple latent factors, where the number of spiked eigenvalues may remain finite or increase with n , and the spiked eigenvalues may be bounded or diverge at arbitrary rates. Beyond characterizing the impact of covariance spectra, we reveal a new mechanism underlying benign overfitting: the prediction behavior of ridgeless interpolation is fundamentally governed by the alignment between the regression coefficient \boldsymbol\beta and the spiked eigenspaces of the population covariance matrix. In particular, we show that the signal energy distributed along latent spike directions determines whether interpolation leads to benign, tempered, or catastrophic overfitting. Our theoretical framework establishes sharp prediction risk limits under minimal moment conditions, requiring only finite fourth moments rather than Gaussianity. We characterize how the number, strength, and geometric structure of the spikes jointly influence the double-descent phenomenon. These results provide a unified understanding of when latent covariance structures facilitate or hinder generalization in overparameterized regression.
[LG-70] Establishing Boundary KKT Convergence of Mirror Descent through Reparameterization
链接: https://arxiv.org/abs/2608.07248
作者: Kuangyu Ding,Kim-Chuan Toh
类目: Optimization and Control (math.OC); Machine Learning (cs.LG)
*备注:
Abstract:We prove that mirror descent converges to a KKT point for the nonconvex problem without excluding boundary limits. The result holds under verifiable conditions that jointly couple the objective, the Legendre kernel, and the feasible geometry. The key ingredient to establish the convergence is a metric-flattening reparameterization (S) that admits a definable boundary extension. Applying the KL argument to the reparameterized objective yields convergence of (S(x_k)). Continuity of (S^-1) then recovers convergence to the KKT point of the original sequence. We further apply our general framework to some concrete examples: Shannon entropy, Fermi–Dirac entropy, and power kernels. Future work may consider more general constraint geometries and genuinely nonseparable kernels, and extend mirror descent to broader Bregman-type methods, e.g. Bregman proximal point algorithms and Bregman ADMM, and their inexact variants.
[LG-71] Optimized Certainty Equivalent Risk Minimization Using Samples: Algorithms Convergence Rates and Applications
链接: https://arxiv.org/abs/2608.07113
作者: Sumedh Gupte,Prashanth L. A.,Sanjay P. Bhat
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:We consider the optimization of the Optimized Certainty Equivalent (OCE) risk, with applications including portfolio optimization in finance, and uncertainty quantification, classification, and regression in machine learning. Our contributions cover popular special cases of OCE, such as entropic risk, mean-variance risk, and smooth variants of Conditional Value-at-Risk. Our treatment sets out the conditions that facilitate the extension of OCE to unbounded r.v.s… We provide a useful characterization of OCE that links OCE to utility-based shortfall risk (UBSR). Our characterization enables us to form an OCE estimator from the classic sample-average approximation (SAA) of UBSR. We derive mean-squared error (MSE) bounds for our proposed OCE estimator. For OCE optimization, we first derive an expression for the OCE gradient using the characterization linking OCE to UBSR. This expression serves as the basis for a gradient estimator for the OCE. We derive non-asymptotic bounds on the MSE for the proposed OCE gradient estimator. We incorporate the aforementioned gradient estimator into a stochastic gradient (SG) algorithm to optimize OCE and quantify its convergence rate using non-asymptotic bounds that we derive. Finally, we present three experiments that use our OCE optimization algorithm to solve portfolio optimization and uncertainty quantification problems.
[LG-72] Limit Points of Reflow with Minibatch Optimal Transport
链接: https://arxiv.org/abs/2608.07042
作者: Antonin Chambolle,Johannes Hertrich
类目: Probability (math.PR); Machine Learning (cs.LG); Numerical Analysis (math.NA)
*备注:
Abstract:Rectified flows, also called flow matching or stochastic interpolants, are generative models that learn a time-dependent vector field steering a probability curve between two probability distributions, usually referred to as latent and target distributions. Reflow accelerates inference by iteratively straightening the trajectories induced by this vector field. We study the asymptotic behavior of this iteration and characterize its limit points. First, we define weak rectified couplings which always exist. Next, when rectified flow updates are alternated with minibatch optimal transport steps of fixed batch size, we show that any limit is N -cyclically monotone, where N is the batch size. Such N -cyclically monotone couplings enjoy favorable structural and stability properties such as rectifiability and straightness. Finally, restricting velocities to gradient fields and assuming additional support conditions, we prove that reflow limits coincide with the optimal transport map between the endpoint distributions.
[LG-73] Mixture of Geodesic Factor Analyzers on Riemannian Homogeneous Spaces
链接: https://arxiv.org/abs/2608.06971
作者: Hengchao Chen,Yuanyao Tan,Chao Huang,Hongtu Zhu,Qiang Sun
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Statistics Theory (math.ST); Methodology (stat.ME)
*备注: 22 pages main, 49 pages appendix
Abstract:This paper introduces Mixtures of Geodesic Factor Analyzers (MGFA) on Riemannian homogeneous spaces. MGFA uses a geodesic factor model within each mixture component, providing greater expressiveness than mixtures of Riemannian radial distributions and enabling clustering of manifold-valued data with anisotropic subpopulations. We establish root- n consistency for the MGFA maximum likelihood estimator (MLE), thereby filling a theoretical gap for mixtures of Riemannian radial distributions as a special case. We also propose an iterative estimation algorithm and implement it on spheres, shape spaces, and hyperbolic spaces. Numerical experiments show that MGFA substantially outperforms competing methods in well-specified regimes while remaining robust under model misspecification. Finally, case studies on corpus callosum and left hippocampus shape datasets demonstrate MGFA’s effectiveness for both 2D contour and 3D shape analysis.
[LG-74] Beyond Co-Movement: Locality by Exposures Enables a Joint Factor-Graph Framework for Portfolio Diversification
链接: https://arxiv.org/abs/2608.06618
作者: Sara Chehab,Giorgos Iacovides,Parisa Yazdanparast,Danilo Mandic
类目: Portfolio Management (q-fin.PM); Machine Learning (cs.LG); Statistical Finance (q-fin.ST)
*备注:
Abstract:Current portfolio construction methods are either agnostic to the effects of idiosyncratic shocks (standard factor models) or to the latent data structure driving systematic returns (recent graph-based approaches). This presents an opportunity to combine the complementary market aspects captured by the factor and graph domains, allowing asset allocations to operate directly on the underlying market structure, rather than on its observed co-movement or its finite-sample artefacts. In this work, we introduce the Mutually-INformed Graph-Locality and Exposures framework (MINGLE), which mutually regularises the factor and graph domains by redefining graph locality through systematic factor exposure profiles, rather than via observed co-movements. This is formalised through a unified Alternating Direction Method of Multipliers (ADMM) framework that jointly learns a latent factor representation and its induced graph topology directly from market returns. The resulting exposure-similarity graph aligns more closely with established economic sectors than conventional correlation-based graphs. Portfolios constructed from this representation are shown to consistently outperform their correlation-based counterparts across a range of volatility regimes and transaction cost levels. For rigour, paired statistical testing confirms that these gains stem from the reconciliation of the graph and factor domains.
[LG-75] Cascading Through the Hierarchy: Regularizer-Induced Feature Detection as Phase Transitions in Deep Linear Neural Networks
链接: https://arxiv.org/abs/2608.06597
作者: Björn Ladewig,Ibrahim Talha Ersoy,Karoline Wiesner
类目: atistical Mechanics (cond-mat.stat-mech); Disordered Systems and Neural Networks (cond-mat.dis-nn); Machine Learning (cs.LG)
*备注: 39 pages, 9 figures
Abstract:A scientific theory of deep learning, comprising learning dynamics and statistical properties of learned models, is rapidly gaining attention. One of the corner stones of this development are analytically solvable toy models, allowing for the fully tractable analysis of the learning dynamics. Here we analytically investigate such a toy model using the regularization strength as a tunable external parameter - akin to external fields in statistical physics. In previous studies, (i) an onset of learning transition was predicted analytically and (ii) it was phenomenologically/numerically established that tuning the regularization strength can result in a cascade of phase transitions. The number of those transitions was linked to the geometry of the loss landscape determined by the model complexity. Setting up a rigorous framework underpinning the previous numerical observations, our investigation reveals a precise connection between those cascades of phase transitions, learnable features and the underlying geometry. We provide analytic predictions of these phase transitions as well as tractable order parameters related to learned features. At the level of the minimal model, we connect this macroscopic perspective (that can be condensed into an effective description) to the microscopic perspective in terms of the geometry of the loss landscape characterized by the Hessian spectrum. Thus, the presented model provides a platform to explore and sharpen advances made in the scientific theory of deep learning rooted in statistical physics concepts.
[LG-76] Density-Functional Excited-State Gradients and Nonadiabatic Couplings on a Consumer GPU from a Contraction-DAG
链接: https://arxiv.org/abs/2608.06536
作者: Rubén Darío Guerrero
类目: Chemical Physics (physics.chem-ph); Hardware Architecture (cs.AR); Machine Learning (cs.LG)
*备注: 17 pages, 11 figures
Abstract:Nonadiabatic dynamics needs an excited-state gradient and an interstate nonadiabatic coupling matrix element (NACME) at every nuclear geometry, and a double-hybrid functional’s accuracy has been unavailable for the coupling. We report the first analytic derivative NACME for a double-hybrid excited state—deferred in the original hh-TDA method and supplied for hybrids only by Yu \emphet al.—derived, with the hole-hole and particle-particle Tamm–Dancoff (\hhTDA/\ppTDA) gradients and NACMEs, as a single reverse-mode transpose of one contraction graph closed under a non-symmetric atomic-orbital-direct J/K kernel. Its double-hybrid excitation energy lowers the vertical-excitation mean absolute deviation from bare-\hhTDA\ 0.86 to 0.47 ~eV and removes the +0.53!\rightarrow!+0.05 ~eV over-excitation bias, improving seven of ten states while over-correcting the ionic \pi\pi^* states—the expected perturbative-doubles failure, reported not trimmed. Every coupling is validated to \sim!10^-4 against an independent \emphliteral many-electron wavefunction-overlap oracle that shares no code path with the method, and is physically meaningful at the ammonia n!\rightarrow!\sigma^* \emphcovalent conical intersection, where the \hhTDA/\ppTDA manifolds recover the F!-!2 seam and adiabatic linear-response TDDFT gives \tau!\equiv!0 by construction. Gradients, NACMEs, and the double-hybrid coupling all run device-resident and AO-direct through one shared Cholesky-decomposed J/K engine within the 8,GB of a consumer RTX~4060 (a profile-guided \sim!10^2\times launch collapse preserving double-precision bit-identity)—placing on a commodity desktop card a correlated excited-state derivative capability that has until now required datacenter hardware.
[LG-77] Game-Theoretic Inverse Reinforcement Learning for Modeling Competitive Human Driving: A Cut-in Prediction Study
链接: https://arxiv.org/abs/2608.06445
作者: Yu Song
类目: Physics and Society (physics.soc-ph); Machine Learning (cs.LG)
*备注:
Abstract:Capturing the strategic decision-making inherent in competitive human driving is critical for autonomous vehicle safety and traffic simulation. This study demonstrates that game-theoretic Inverse Reinforcement Learning (IRL) provides a robust framework for this challenge. We present a comprehensive analysis comparing data-driven IRL models against an established physics-based game-theoretic approach for predicting aggressive, safety-critical cut-in lane changes. Using the high-fidelity highD dataset, we systematically develop and evaluate a series of IRL models with increasing feature complexity. Our results reveal significant advantages: the best-performing IRL models achieve an overall prediction accuracy exceeding 75 percent while maintaining a Cut-In precision up to 51 percent and recall up to 49 percent. This represents a significant improvement over the established physics-based benchmark, which achieved only 4.4 percent precision in these high-stakes scenarios. The analysis reveals a clear trade-off: incorporating granular, instantaneous features yields higher precision, while adding temporal consistency features maximizes recall. These findings suggest that IRL-based models can effectively bridge the gap between microscopic driver intent and macroscopic safety outcomes, providing a more reliable foundation for modeling interactions in mixed-autonomy environments.
[LG-78] A proximal subgradient method for nonconvex stochastic optimization under the Kurdyka-Łojasiewicz condition
链接: https://arxiv.org/abs/2608.05460
作者: Felipe Atenas,Alejandro Jofré,Pedro Pérez-Aros,David Torregrosa-Belén
类目: Optimization and Control (math.OC); Machine Learning (cs.LG)
*备注:
Abstract:This work introduces a proximal stochastic subgradient method for minimizing the sum of an expected cost, whose integrand is potentially nonsmooth and nonconvex, and a lower semicontinuous, prox-bounded function. We target a broad class of integrands obeying a nonsmooth, localized variant of the descent lemma in the decision variable, a structural assumption that simultaneously covers smooth losses with Lipschitz gradient and differences of such losses with convex functions. At each iteration the expected cost is replaced by a sample average that is progressively refined, and the proximal-subgradient stepsize is selected by an Armijo-type line search enforcing a sufficient-decrease property up to stochastic errors induced by the sample-based approximation. This framework accommodates substantially more general problem formulations than existing methods, in particular, it requires neither (weak) convexity of the regularizer nor a uniform bound on the variance of the stochastic oracle, and our analysis yields convergence guarantees that are new even in the smooth setting. Specifically, we establish almost sure convergence of the sequence of function values and stationarity of every accumulation point of the trajectories under the relaxed requirement that the sample-size sequence be merely nondecreasing and unbounded, with no prescribed growth rate. Leveraging the Kurdyka-Lojasiewicz (KL) property, we further upgrade this subsequential guarantee to convergence of the whole trajectory to a single stationary point. Finally, for exponential-type KL desingularizing functions and polynomially growing sample sizes, we derive explicit polynomial convergence rates, up to a logarithmic factor, for both the function values and the iterates.
附件下载
