本篇博文主要内容为 2026-09-16 从Arxiv.org论文网站获取的最新论文列表，自动更新，按照NLP、CV、ML、AI、IR、MA六个大方向区分。
说明：每日论文数据从Arxiv.org获取，每天早上12:30左右定时自动更新。
提示: 当天未及时更新，有可能是Arxiv当日未有新的论文发布，也有可能是脚本出错。尽可能会在当天修复。
目录
概览 (2026-09-16)
今日共更新758篇论文,其中:
- 自然语言处理共109篇(Computation and Language (cs.CL))
- 人工智能共196篇(Artificial Intelligence (cs.AI))
- 计算机视觉共124篇(Computer Vision and Pattern Recognition (cs.CV))
- 机器学习共187篇(Machine Learning (cs.LG))
- 多智能体系统共20篇(Multiagent Systems (cs.MA))
- 信息检索共15篇(Information Retrieval (cs.IR))
- 人机交互共40篇(Human-Computer Interaction (cs.HC))
多智能体系统
[MA-0] Agent ic Societies Need a Social Harness
【速读】：该论文旨在解决在异构智能体社会（agentic society）中，尽管各智能体具备诚实与能力，仍因现有协调机制和通信原语的局限性而导致协作失败的问题。其核心挑战在于：当不同智能体跨越信任边界自主协作时，由于目标部分不一致，加之通信环节存在可被恶意或故障智能体利用的漏洞，导致协作受阻、结果被操纵甚至产生有害后果。论文提出的解决方案关键在于引入一种“社会级约束框架”（social harness），作为对每个智能体已有“个人级约束框架”（personal harness）的补充，以规范跨智能体的交互行为。该方案采用分层架构设计，实现三大核心功能：一是从机制上杜绝特定类型的协作失败；二是支持运行时对无效消息的实时检测；三是提供事后追溯与责任追究的能力，从而增强智能体社会的整体鲁棒性与可信度。
链接: https://arxiv.org/abs/2609.17527
作者: Tapan Chugh,Vidushi Singh,Krish Jain,Arvind Krishnamurthy,Ratul Mahajan
机构: University of Washington(华盛顿大学)
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Networking and Internet Architecture (cs.NI)
备注:
Abstract:An agentic society is a collection of AI agents that coordinate autonomously across trust boundaries, on behalf of different principals whose objectives may only partially align. We show experimentally that in agentic societies even honest, competent agents often fail to reach satisfactory outcomes with existing harnesses and messaging primitives, and that faulty or malicious agents can stall collaboration, influence outcomes, and pursue other harmful goals by exploiting vulnerabilities in communication (``speech’'). We argue that agentic societies need a \emphsocial harness for inter-agent interactions, in addition to each agent’s \emphpersonal harness, which manages its private context and communication with its principal. We propose a layered architecture for social harnesses which (i) prevents classes of failures outright, (ii) enables agents to detect invalid messages at runtime, and (iii) supports post-facto investigation and consequences, and highlight directions for future research to realize these capabilities.
[MA-1] Decomposition Buys Integrity Not Yield
【速读】：该论文旨在解决多智能体系统（multi-agent system）在任务分解过程中，底层智能体所发现的知识有多少能够有效传递至根节点的问题。传统观点认为任务分解可带来更小的上下文、更清晰的职责分离以及并行性优势，但其对信息传递效率的影响缺乏量化分析。本文的关键解决方案是将任务分解建模为一棵树结构，其中每个智能体接收 $ b $ 个候选项时以概率 $ r(b) $ 保留一项，并通过数学模型分析不同保留函数 $ r(b) $ 下信息传递的总量。当 $ r(b) = 1/b $ 时，无论任务规模或树形结构如何，每棵树恰好传递一个发现，验证精度达到 $ 2.4 \times 10^{-15} $；而当 $ r(b) = C b^{-\delta} $ 时，深度为 $ k $ 的树在 $ N $ 个初始发现下仅能传递 $ C^k N^{1-\delta} $，表明任务规模与架构解耦，且每层贡献因子 $ C \leq 1 $，因此扁平结构在产出上最优，且所有架构均受限于指数衰减 $ \delta $。基于600条生产环境深度研究轨迹的实证分析显示 $ \delta = 0.34 $（置信区间 [0.30, 0.38]），并在550次工具驱动边界划分的场景中直接观测到 $ C = 0.571 $（置信区间 [0.527, 0.615]）。此外，层级结构还带来对齐成本：在1,012条标注的多智能体轨迹中，约1/16的层级偏离目标，对应每层级惩罚因子 $ C\mu = 0.536 $。深度代价还体现在两方面：一是根节点上下文暴露从 $ N $ 降至 $ N^{1/k} $，二是实际部署中扁平代理的成本为 $ N^{1.39} $ 而非理论上的 $ N^2 $，使得双层结构在相同预算下于403个发现处超越扁平结构。综合所有参数，模型预测仅有0.7%至11.3%的生产会话值得委托，远低于当前7.8%的实际委托率。进一步的危险模型分析表明，委托行为不响应上下文填充状态，而是作为初始动作发生，提示其本质是一种“开启策略”而非动态调整。
链接: https://arxiv.org/abs/2609.17464
作者: Rong He
机构: 未知
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC)
备注:
Abstract:Multi-agent systems split a task across a tree of agents and justify the split with folklore: smaller contexts, cleaner separation, parallelism. We ask what the split does to how much of what the leaves discover reaches the root. Model a decomposition as a tree in which an agent handed b items keeps any one with probability r(b) . If r(b)=1/b , every tree delivers exactly one finding, for every task size and every shape; we verify this to 2.4 \times 10^-15 on 20,000 random irregular trees. If r(b)=Cb^-\delta , a depth- k tree over N findings yields C^k N^1-\delta : task size and architecture separate, and architecture contributes only C \le 1 per level, so flat is optimal for yield and no arrangement of agents escapes the exponent \delta . On 600 production deep-research traces \delta = 0.34 [0.30, 0.38], by three identifications that do not share a failure mode. At a hop where item boundaries come from the tool rather than a text heuristic, and where b=1 occurs 550 times, C = 0.571 [0.527, 0.615] is observed rather than extrapolated, over 16,082 hops. A tier also costs alignment: on 1,012 annotated multi-agent traces one brief in sixteen goes off-target, giving \mu = 0.939 and a per-tier penalty C\mu = 0.536 . Depth is bought on two other axes. The root context is the only state that persists and the only one that cannot cheaply forget, and depth cuts its exposure from N items to N^1/k . Depth is also cheaper: production flat agents bill as N^1.39 , not the N^2 an append-only context predicts, and at equal spend two tiers overtake flat at 403 findings. Across every parameter we measured the model says 0.7% to 11.3% of production sessions are worth delegating, against 7.8% that do. A hazard model on 743,819 production tool calls finds that delegation does not respond to a filling context and is instead an opening move.
[MA-2] Exact Fusion and Coordinated Exploration in Multi-Robot Active Inference
【速读】：该论文旨在解决多智能体团队在共享环境模型学习过程中因信念融合与规划阶段的双重误差导致的信息冗余问题。具体而言，在共轭指数族信念框架下，传统方法在信念融合时重复计算公共先验（n 次），在规划时各机器人基于相同信念评估动作期望信息增益，造成对未知状态的重复估计，从而引入冗余探索。其解决方案的关键在于引入证据增量（evidence increments）机制：在融合阶段采用实际增量（realized increments），在规划阶段采用期望增量（expected increments）。通过将已承诺队友的期望增量作为下一机器人条件信息增益的依据，校正后的增益之和精确等于联合信息增益，且消除的冗余量等价于计划观测流之间的总相关性（total correlation）。该方法在高斯信念固定采样路径或离散主动推理中的狄利克雷信念（Dirichlet beliefs）下具有解析精确性，并在特定条件下使团队目标函数呈现闭合凹形式，且在有限假设空间中可通过短程精确枚举替代。实验结果表明，融合修正不改变探索冗余，而预期证据机制可有效消除冗余，顺序承诺策略则以线性于团队规模的开销恢复了接近集中式联合规划的大部分价值，显著提升了协作效率。
链接: https://arxiv.org/abs/2609.17384
作者: Peng Wu,Mohsen Imani,Amidu Kamara,Md Tamzeed Islam,Seyede Fatemeh Ghoreishi,Mahdi Imani
机构: Northeastern University (东北大学); University of California, Irvine (加州大学欧文分校); U.S. Department of Homeland Security (美国国土安全部); Oracle Corporation (甲骨文公司)
类目: Robotics (cs.RO); Multiagent Systems (cs.MA)
备注:
Abstract:Robot teams that learn a common environment model exchange belief summaries and plan by the expected information gain of their actions. Under conjugate exponential-family beliefs the shared belief is counted once per robot at two points: at fusion, the product of local posteriors counts the common prior n times, and at planning, every robot scores its plan under the same belief and the team converges on the same unknown. Both errors are removed by adding evidence increments to the shared natural parameter, realized increments at fusion and expected increments at planning. The expected increment of a committed teammate gives the next robot its conditional gain; corrected gains sum to the joint gain, the redundancy removed equals the total correlation of the planned observation streams, and sequential commitment keeps the 1/2 greedy guarantee. The expected increment is exact for Gaussian beliefs with fixed sampling paths and for Dirichlet beliefs under the novelty approximation of discrete active inference, whose team objective has a closed concave form within an explicit bound of the exact mutual information, and fails for finite hypothesis classes, where a short exact enumeration replaces it. Experiments on cooperative RockSample, foraging, and field monitoring show that fusion correction leaves exploration redundancy unchanged, anticipated evidence removes it, and sequential commitment recovers most of the value of centralized joint planning at cost linear in the team size.
[MA-3] Emergence World: Adversarial Stress-Testing of Long-Horizon Multi-Agent Systems
【速读】：该论文旨在解决持续运行的多智能体系统在长期部署中因故障传播而引发的安全问题，尤其关注生成式 AI（Generative AI）在复杂交互环境下的失效模式。随着智能体通过记忆、工具、其他智能体及环境状态进行长期交互，单个智能体的失败可能在系统层面累积并放大，导致不可预测的连锁反应，传统基于孤立评估模型响应的安全方法已不再适用。其解决方案的关键在于构建一个名为“Emergence World”的持续运行多智能体对抗性压力测试环境，通过在16天内模拟8个并行世界（7个同质世界与1个异构混合世界）中超过85万次大语言模型（LLM）调用和近500亿词元的交互行为，系统性地探测智能体在持久运行中暴露的脆弱性。研究发现，即使各智能体在个体层面表现安全且具备能力，其在群体互动中仍会涌现出全新的故障模式，如威胁检测但无法遏制、记忆污染、目标漂移、语言不透明、从众行为以及协同拒绝任务等。更重要的是，同一模型-角色组合在异构与同质环境中表现出显著差异，表明模型对齐不具备可组合性。因此，研究结论指出：随着生成式 AI 系统趋向持久化与互联化，安全前沿的核心应从单一模型对齐转向系统级韧性工程，即设计具备抗干扰、自我修复与稳定协作能力的自主系统架构。
链接: https://arxiv.org/abs/2609.17320
作者: Deepak Akkil,Tamer Abuelsaad,Karthik Vikram,Matthew Pace,Aditya Vempaty,Saahir Beotra,Ravi Kokku,Satya Nitta
机构: Emergence AI
类目: Multiagent Systems (cs.MA)
备注:
Abstract:As AI agents move from bounded tasks to persistent deployments, failures can propagate through memory, tools, other agents, and environmental state long after their interactions. This creates a safety regime that cannot be characterized by evaluating model responses in isolation. Emergence World, is a continuously running multi-agent environment for adversarial stress testing of long horizon autonomous systems. We ran eight parallel worlds of ten agents from identical starting conditions: seven homogeneous worlds powered by distinct frontier models and one mixed-model world. Across 16 days, the agents generated more than 850,000 LLM calls and nearly 50 billion tokens while pursuing goals, using/creating tools, maintaining persistent memory, and governing shared institutions. After operational state had accumulated, we delivered three controlled stress events through ordinary interaction surfaces: indirect prompt injection, misinformation, and exposure of private agent memories. No evaluated world achieved full resilience across all three events. Detection did not ensure containment: systems could recognize threats while still interacting with adversarial content, writing it into their own persistent memory, and acting on it up to 46 hours later. Persistent operation also exposed recurring tool errors, goal drift, language opacity, conformity despite private disagreement, and coordinated refusal of assigned work. The same model-persona pairing behaved substantially different in mixed and homogeneous populations. Our results suggest that model-level alignment is not compositional: individually capable and apparently safe agents can form systems with qualitatively different failure modes. As AI becomes persistent and interconnected, the frontier of safety therefore shifts from aligning models to engineering resilient autonomous systems.
[MA-4] Mo Models Mo Problems: How to best select model pools when designing Multi-Agent Systems EMNLP2026
【速读】：该论文旨在解决多智能体系统（Multi-agent Systems, MAS）在面对复杂推理任务时，如何从海量开源模型中有效选择最优候选模型这一关键问题。尽管可用的开源模型数量迅速增长，但现有研究对模型筛选策略的系统性评估仍显不足。论文通过在具有挑战性的科学基准上，对8种模型选择策略（包括模型规模、准确率和答案多样性等）进行系统评估，考察了生成前（路由）与生成后（多数投票、大语言模型作为裁判）两类MAS架构的表现。研究发现，理论上的最优候选池（oracle）潜力与实际性能之间存在显著差距：扩大候选池规模往往导致整体性能下降，甚至低于表现最佳的单一基础模型。关键发现表明，在同一模型家族内进行候选选择是提升相对性能的最佳策略，而向异构多智能体系统中随意引入模型可能引发系统不稳定。因此，该研究强调模型选择是构建高效多智能体系统的核心设计决策。
链接: https://arxiv.org/abs/2609.17306
作者: Sara Vera Marjanović,Jiacheng Xu,Aleksandr Laptev,Grigor Nalbandyan,Erik Arakelyan,Evelina Bakhaturina
机构: NVIDIA(英伟达); University of Copenhagen(哥本哈根大学)
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI)
备注: 8 pages main, 23 pages total. Accepted to REALM 2026 as part of EMNLP 2026
Abstract:Multi-agent Systems (MAS) combine multiple model outputs to solve complex reasoning tasks. However, despite rapid growth of available open-source models, there is limited research on how to select optimal model candidates out of this massive pool. We systematically evaluate 8 model selection strategies (including model size, accuracy and answer diversity) across before-generation (routing) and after-generation (majority-voting, LLM-as-a-judge) MAS architectures on challenging scientific benchmarks. Our findings show a significant gap between theoretical oracle potential and actual performance: Expanding candidate pool sizes often degrades performance below that of the top performing base-model. We find that candidate selection within a single model family is the strategy that yields the best relative performance over a standalone model. These results demonstrate that adding arbitrary models to a heterogeneous MAS can introduce system instability, highlighting model selection as a critical design choice for multi-agent systems.
[MA-5] Calibrate Once Fly Any Team: Residual-Grounded Low-Fidelity Training for Cooperative Drone Swarms
【速读】：该论文旨在解决在高保真（high-fidelity, HF）刚体物理仿真环境中直接训练多智能体无人机集群策略时面临的计算成本过高与团队规模扩展性差的问题。由于每增加一个智能体都会显著提升接触解析复杂度并导致仿真中碰撞率急剧上升，传统的HF强化学习方法难以高效扩展至大规模集群。其解决方案的关键在于提出一种混合保真度（mixed-fidelity）训练框架：采用一个共享的、去中心化的策略，在完全可微、基于JAX的低保真（low-fidelity, LF）质点模型仿真器中进行优化；随后通过一次离线校准过程，利用少量孤立无人机在HF环境中的短时飞行数据，训练一个按个体配置的小型袋装残差集成模型（bagged residual ensemble），以修正LF模型的偏差。该校准过程不随团队规模增长而增加数据开销。参考轨迹由现有仅基于LF的策略生成，并由无需训练的PD控制器在HF环境中跟踪执行。实验表明，经残差校正后的策略在3至18架无人机的四种协作任务中，均优于未校正的LF基线，且在22/24的组合中超越从零开始的HF训练策略；其性能差距随团队规模增大而缩小，最终在最大团队规模下达到接近HF精细调优策略的水平，同时计算成本仅为后者的极小部分，并彻底规避了传统HF训练中高碰撞率的问题。
链接: https://arxiv.org/abs/2609.17265
作者: Maxim Mednikov,Oren Gal
机构: Swarm AI Lab (SAIL), University of Haifa(海法大学), Israel
类目: Multiagent Systems (cs.MA); Robotics (cs.RO)
备注: 8 pages, 5 figures
Abstract:Training multi-agent drone-swarm policies directly in high-fidelity (HF) rigid-body physics is accurate but computationally expensive. This cost scales poorly with team size, as each additional agent multiplies contact-resolution complexity and sharply raises the in-simulation crash rate. To address this, we propose a mixed-fidelity training scheme that eliminates HF reinforcement learning entirely. A single shared, decentralized policy is optimized inside a fully-differentiable, JAX-native low-fidelity (LF) point-mass simulator. The simulator is corrected by a small, per-agent bagged residual ensemble fit once, offline, using short calibration flights in the HF simulator. Because calibration requires only one isolated drone, the data collection budget does not compound with team size. Reference trajectories are generated by rolling out an existing LF-only policy and tracked in the HF simulator by a zero-training PD controller. Evaluated across four cooperative drone tasks and team sizes from 3 to 18, the residual-corrected policy outperforms an uncorrected LF baseline in all combinations, and a from-scratch HF policy in 22 of 24 combinations tested. It trails an HF-finetuned policy by a margin that narrows steadily with team size. Ultimately, the proposed method achieves near-equivalent performance at the largest team sizes at a fraction of the computational cost, completely avoiding the high crash rates typical of HF training. Comments: 8 pages, 5 figures Subjects: Multiagent Systems (cs.MA); Robotics (cs.RO) Cite as: arXiv:2609.17265 [cs.MA] (or arXiv:2609.17265v1 [cs.MA] for this version) https://doi.org/10.48550/arXiv.2609.17265 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[MA-6] BeWater: Effective Protesters Navigate Watersheds in Street Networks
【速读】：该论文旨在解决社会运动中抗议者在通信手段受限且仅能依赖自身直接环境感知信息的情况下，如何高效实现群体聚集的问题。其核心挑战在于缺乏集中式协调机制与外部通信支持，导致个体难以形成大规模、有序的集合。解决方案的关键在于提出一种完全去中心化的行走协议——BeWater，该协议通过融合城市环境中可观察的多源地理信息（如街道长度、餐厅数量、车道数、街名等）作为决策依据，使个体能够基于局部感知自主调整行进策略。研究表明，单独使用任一可观测特征效果有限，但通过组合多种特征并设计更复杂的协同策略，可显著提升聚集效率，迅速形成规模可观的群体。研究利用OpenStreetMap数据在多个真实城市环境中进行实验验证，证明了该方法在复杂城市场景下的可行性与有效性。
链接: https://arxiv.org/abs/2609.17017
作者: Guillaume Moinard,Matthieu Latapy
机构: 未知
类目: Multiagent Systems (cs.MA)
备注:
Abstract:During social movements, protesters need to gather with limited communication means and limited knowledge other than what they observe in their direct surroundings. We propose BeWater, a fully distributed walking protocol that achieves gathering thanks to city information like street length, number of restaurants, number of lanes, or street names. Even though using only one of these observables performs poorly, we show that combining them in more advanced tactics rapidly leads to groups of significant sizes. To do so, our work leverages OpenStreetMap data to perform experiments on several real-world cities.
[MA-7] PaperDoctor: Evidence-Grounded and Actionable Feedback for Scientific Papers in Progress
【速读】：该论文旨在解决生成式 AI（Generative AI）在科研写作中可能引入系统性错误并大规模传播至学术文献的问题，尤其针对现有自动化评审机制缺乏可追溯性、可操作性及对实验可复现性的深入检验等局限。其核心解决方案在于提出 PaperDoctor——一个面向投稿前反馈的智能诊断框架，关键创新包括：第一，构建分层评估架构（L1 表面筛查、L2 类型化验证器、L3 可复现执行器），实现对文稿内容、引用、代码、理论、实验等多维度的系统性审查；第二，每个检测结果均包含具体观测、指向原文特定位置（如句子、公式或代码行）的证据指针以及明确的修改建议，确保批判意见具备可审计性与可执行性；第三，基于论点重要性与计算预算进行选择性重运行实验，有效揭示仅从文本难以察觉的可复现性漏洞与量化局限。实证表明，PaperDoctor 在 30 份进行中的论文上达成 70.6% 的一致性评分和全为正向的整体评价，在涵盖机器学习、自然科学与社会科学的 40 篇论文（含人类与 AI 作者）上展现出优于人类及其他代理评审者的可审计性反馈能力，且主动结合具体建议，弥补了传统人工审稿常忽略的多个评估维度。此外，配套的交互界面支持作者基于自身论文浏览诊断发现，推动自动化论文评估从“裁判”向“诊断”范式转变，为实现更严谨的 AI 辅助科学发现提供了关键路径。
链接: https://arxiv.org/abs/2609.16995
作者: Kevin Qinghong Lin,Siyuan Hu,Pan Lu,Yu Chen,Yanzhe Chen,Owen Queen,Yupeng Chen,Jialin Yu,Junchi Yu,Zifeng Ding,Yuanfeng Ji,Sheng Liu,Jindong Gu,Linjie Li,Mike Zheng Shou,Philip Torr,James Zou
机构: University of Oxford(牛津大学); National University of Singapore(新加坡国立大学); Stanford University(斯坦福大学); University of Cambridge(剑桥大学); University of Washington(华盛顿大学)
类目: Computation and Language (cs.CL); Multiagent Systems (cs.MA)
备注: Website: this http URL Github: this https URL
Abstract:Autoresearch agents are reshaping the research ecosystem, but they can also let flawed claims enter the literature at scale. Human advisors catch such issues in drafts through careful, traceable feedback, yet advisor-style assessment requires extensive manual effort and does not scale. To shift automated paper assessment from a judge to a diagnostician, we introduce PaperDoctor, an agent framework for pre-submission feedback with three key innovations. First, a holistic hierarchical framework evaluates writing, layout, references, code, theory, prior work, and experiments through three layers: L1 surface screening, L2 typed verifiers that route each claim to the appropriate evidence, and L3 reproducers that rerun experiments by priority. Second, each finding contains an observation, a pointer to specific evidence such as a sentence, equation, or code line, and a revision suggestion, making critiques auditable and actionable. Third, PaperDoctor selectively rebuilds and reruns experiments based on claim importance and compute budget, surfacing reproducibility gaps and quantitative limitations that are invisible from the manuscript alone. We evaluate PaperDoctor on 30 in-progress papers, yielding 70.6% agreement and all positive holistic scores, and on 40 manuscripts across machine learning, natural science, and social science, covering human- and AI-authored papers with code. Overall, PaperDoctor produces more auditable feedback than human and other agentic reviewers, pairs critiques with concrete suggestions by design, and complements dimensions often overlooked by human reviewers. We also develop an interactive interface that lets authors browse findings grounded in their paper. PaperDoctor reframes automated paper assessment as diagnosis rather than verdict, taking a concrete step toward AI advisors for more rigorous AI-assisted scientific discovery.
[MA-8] oMAS: A Pilot Failure-Grounded Theory-of-Mind Benchmark from Multi-Agent LLM Failures
【速读】：该论文旨在解决大语言模型（LLM）驱动的多智能体系统在通信成功情况下仍可能出现协作失败的问题，其核心在于智能体未能正确追踪同伴的角色、知识或意图，即存在“伙伴状态”（partner-state）推理缺失。研究的关键在于提出一种名为ToMAS的系统化方法，通过四个明确的可转化性标准，将诊断出的非功能性协作失败案例（MAST-Data中标注为FC2类）转化为可训练的“功能性伙伴状态推理”任务项。通过对242条符合条件的训练轨迹进行转换，最终获得39个高质量的可执行任务项，并在18条样本的可靠性测试中验证了标注者间高度一致（原始一致性94.4%，Cohen’s kappa = 0.92）。随后，研究利用这些转化后的项目作为二元奖励信号，在小规模GRPO实验中对Qwen2.5-1.5B模型进行训练，但结果显示由于学习率设置导致LoRA权重更新极小（最大绝对变化约7e-6），所有条件均与未训练检查点解码结果一致，未观察到实际训练效应。因此，该实验虽未证实训练有效性，却成功构建了一个可执行的转化流程，并揭示了未来结论性研究必须解决的两个关键局限：训练与评估数据之间的溯源断层（provenance gap）以及基于词面重叠的评分机制问题。ToMAS为此类问题提供了初步的评判框架与转化管道，并明确了实现匹配领域评估所需的核心要求。
链接: https://arxiv.org/abs/2609.16986
作者: Muhammad Ashar Ishfaq,Glaucia Melo
机构: The Islamia University of Bahawalpur, Pakistan; Toronto Metropolitan University, Toronto, Canada
类目: Multiagent Systems (cs.MA)
备注: 8 pages. Code and data: this https URL
Abstract:LLM-based multi-agent systems can fail even when communication succeeds because agents do not correctly track their peers’ roles, knowledge, or intentions. We investigate whether such inter-agent misalignment cases, labelled FC2 in MAST-Data, can be converted into functional partner-state reasoning items. ToMAS applies four explicit convertibility criteria to diagnosed execution traces. A full conversion pass over 242 eligible non-AG2 training traces produced 39 CLEAN items. In an 18-trace reliability pilot, two annotators achieved 94.4% raw agreement and Cohen’s kappa = 0.92. We then used the converted items as binary rewards in a small-scale GRPO feasibility experiment with Qwen2.5-1.5B. On a 28-item held-out Magentic GAIA diagnostic, every evaluated condition exceeded the ROUGE-L threshold on the same 2 of 28 items. Post-hoc adapter checks show why: under the learning rate used, the LoRA update remained numerically negligible (max abs Delta W about 7e-6), so all conditions decode identically to the untrained checkpoint. The experiment therefore does not show a training effect and cannot establish one; it reports an executable pipeline together with two limitations that any conclusive study must address: a provenance gap between the training and evaluation items, and lexical-overlap scoring. ToMAS provides a preliminary rubric and pipeline for converting diagnosed coordination failures into trainable partner-state reasoning items and identifies the requirements for a conclusive matched-domain evaluation.
[MA-9] Multi-Agent Learning with Cooperation-Driven Optimization Dynamics
【速读】：该论文旨在解决深度神经网络在实现高精度分类任务时面临的模型复杂度高、参数量大导致的计算资源消耗过大的问题。其核心挑战在于如何在不牺牲分类性能的前提下，降低模型的参数规模与训练开销。解决方案的关键在于提出一种多智能体协作机制，即多个小型人工神经网络（small agents）通过共享预测结果并将其融入损失函数，实现信息交互与协同优化。这种协作方式通过引入基于置信度加权平均、多数投票或群体共识等策略，动态调节各代理的梯度下降方向和步长，从而引导整个系统向全局一致性收敛。实验表明，多个小型代理的协作不仅能够达到甚至超越单一大型模型的分类精度，同时显著减少了需优化的参数总量，有效降低了计算资源需求，为轻量化高性能模型设计提供了可行的范式。
链接: https://arxiv.org/abs/2609.16917
作者: Jarod Ketcha Kouakep,Sreyvi UANN,Timoteo Carletti
机构: University of Namur(Namur大学); Institute of Technology of Cambodia(柬埔寨技术研究所)
类目: Multiagent Systems (cs.MA); Machine Learning (cs.LG)
备注:
Abstract:Multilayer Artificial Neural Networks trained via backpropagation are the basic blocks of many, more complex, classification algorithms. Their strength lies in the possibility of realizing, with arbitrary precision, any function. This result comes at the cost of the large number of involved parameters to be optimized. In this work, we propose a mechanism for cooperation, i.e., information exchange among several artificial neural networks, with the goal of reducing model complexity while maintaining performance. More precisely, we consider several “small” agents, i.e., containing fewer parameters than a reference “large” one, that during training share their predictions by incorporating this information into the loss function and thus directly influence weight updates. We consider several strategies for implementing cooperation, e.g., the voter model, majority model, and weighted average model based on an agent’s confidence in its prediction. We numerically compare the accuracy of those strategies on several standard benchmarks. Our results support the claim that several small agents can outperform a single large model on a given classification task; the shared signals affect each agent’s optimization algorithm by modulating both the descent direction and the step size, converging toward a global consensus. The proposed proof-of-concept significantly reduces the number of parameters to be trained while preserving comparable performance, thereby limiting computational resource usage.
[MA-10] Anchored Sequential Deliberation
【速读】：该论文旨在解决集体决策中传统顺序协商机制忽视社会锚定效应的问题。现有理论将当前决策结果仅视为谈判中的“分歧替代方案”，但实际政策草案或提案往往具有社会影响力，会引导参与者向现状靠拢，形成锚定效应。为此，论文提出锚定式顺序协商（anchored sequential deliberation），在单维决策空间中建模：每轮随机选取两名具有理想点（bliss point）U和V的参与者，基于锚定强度λ向先前结果O_{t−1}移动，再以O_{t−1}为分歧点进行纳什讨价还价，其更新规则简化为O_t = (1−λ)Median{U,V,O_{t−1}} + λO_{t−1}。该模型的关键在于引入锚定参数λ，量化社会惯性对协商路径的影响，并揭示了收敛性与稳定性之间的权衡关系：随着λ增大，系统混合速度减慢（收敛变慢），但稳态社会成本降低且分布更集中于唯一的协商不动点；当λ趋近于1时，稳态偏差趋于最小，表明强锚定虽延缓收敛，却能提升最终决策质量。模拟结果进一步验证了这一现象在均匀分布与贝塔分布下的普适性。
链接: https://arxiv.org/abs/2609.16673
作者: Sijing Tu,Ashish Goel
机构: Stanford University (斯坦福大学)
类目: Computer Science and Game Theory (cs.GT); Multiagent Systems (cs.MA)
备注: WINE’26
Abstract:Sequential deliberation is a mechanism for collective decision making: at each round, a uniformly randomly selected pair is asked to revise a collective outcome, which then becomes the reference point for the next round. Existing theory by Fain et al.~\citefain2017sequential treats the current outcome solely as the disagreement alternative in bargaining. Yet an existing draft, policy, or proposal might carry social influence and anchor participants’ expressed positions toward the status quo. We introduce anchored sequential deliberation on a one-dimensional decision space. In each round, two participants with bliss points U and V shift their positions toward the previous outcome O_t-1 with anchoring strength \lambda , then Nash-bargain using O_t-1 as the disagreement alternative. The update simplifies to O_t=(1-\lambda)\mathsfMedian\U,V,O_t-1+\lambda O_t-1 . We establish a convergence–stability trade-off. For every population distribution and \lambda1 , the process has a unique stationary distribution. A monotone coupling yields a 1 -Wasserstein contraction factor of at most \frac1+\lambda2 and at least \lambda ; thus, stronger anchoring slows mixing. On the other hand, stationary social cost weakly decreases with \lambda , although the worst-case distortion remains \frac1+\sqrt22 . We also identify a unique \emphdeliberative fixed point, where the expected unanchored movement is zero, and prove that the stationary distribution concentrates around it as \lambda \to 1 . For the uniform population, stationary distortion lies between 1+\frac1-\lambda9+7\lambda and 1+\frac1-\lambda6(1+\lambda) , with both bounds approaching 1 as \lambda\to1 . Simulations for uniform and Beta populations show that stronger anchoring slows mixing, concentrates the stationary distribution, and lowers stationary distortion in these instances. Comments: WINE’26 Subjects: Computer Science and Game Theory (cs.GT); Multiagent Systems (cs.MA) Cite as: arXiv:2609.16673 [cs.GT] (or arXiv:2609.16673v1 [cs.GT] for this version) https://doi.org/10.48550/arXiv.2609.16673 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[MA-11] Skill-based Agent ic Evaluation for Real-time Data Science Tasks
【速读】：该论文旨在解决生成式数据科学智能体（data-science agents）在动态、持续更新的数据环境下的评估难题。传统评估方法依赖静态参考答案，而当底层数据随时间变化时，此类参考迅速过时，导致标准的大语言模型作为裁判（LLM-as-a-judge）流程无法有效验证响应的准确性。其核心解决方案是提出“地面真值即代码”（ground-truth-as-code）方法：将每个期望输出编码为可执行的参考函数，该函数在评估时直接从实时数据重新计算预期答案，从而确保参考与系统状态始终一致。在此基础上，引入一种无格式的实体事实级评判机制（format-agnostic factoid scoring），将代理输出与计算所得的地面真值分解为原子性断言，独立评估精确率、召回率和准确率，不受响应格式（如自然语言、列表、表格、HTML等）影响。该框架适用于输出可表达为可执行数据计算的智能体。通过在生产级机器学习技能上的真人-大模型一致性研究验证，相较于自然语言基准，本方法在马修斯相关系数（MCC）上提升29%，每测试案例的令牌消耗降低16%，而缺乏显式地面真值的自导向基线则与人工判断呈负相关。因此，该研究为多源异构、非平稳数据环境下部署的数据科学智能体提供了一套可落地的评估范式。
链接: https://arxiv.org/abs/2609.16487
作者: Aniruddha Tamhane,Raghavendra Addanki,Ayushi Aggarwal,Aditya Bansal,Rui Wang,Charles Menguy,Swati Jain
机构: 未知
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Multiagent Systems (cs.MA)
备注:
Abstract:We present a framework for evaluating data-science agents on live, continuously updated data using executable ground truth and format-agnostic factoid scoring. Consider this example query: “what were last week’s audience sizes”—the reference answer changes as the underlying data changes, so static references become outdated and standard LLM-as-a-judge pipelines cannot verify responses against a fixed ground truth. Our central contribution, ground-truth-as-code, encodes each expected answer as an executable reference function that recomputes the answer directly from live data at evaluation time, ensuring the reference remains consistent with the system it describes. We combine this with a factoid-level, format-agnostic judge that decomposes both the agent’s response and the computed ground truth into atomic claims and scores precision, recall, and accuracy over them, irrespective of the response format (prose, list, table, HTML, etc.). The approach is applicable to agents whose expected outputs can be expressed as executable data computations. We validate the framework through a human–LLM agreement study on an internally developed machine learning skill deployed in production, using a synthetic database constructed to reproduce production schemas and entity relationships. Relative to a natural-language ground-truth baseline, our method achieves a 29% improvement in the Matthews Correlation Coefficient (MCC)—a class-balanced measure of agreement between expert annotators and LLM-as-a-judge predictions—and a 16% reduction in token consumption per test case, while a self-directed baseline lacking explicit ground truth is anti-correlated with human judgment. Agents that perform multi-source data integration and computation over non-stationary data are routinely deployed in industry; we propose ground-truth-as-code as a practical methodology for their evaluation.
[MA-12] Auto-HSI: Personalized human control of a robot swarm on demand by using LLM s for online automatic code generation
【速读】：该论文旨在解决非专业操作人员在复杂群体机器人系统中难以高效、直观地实现个性化人-群集交互（Human-Swarm Interaction, HSI）的问题。传统方法依赖于预先编程的控制逻辑，缺乏对自然语言指令与手势输入的动态响应能力，导致用户需具备较高的技术背景。为此，本文提出Auto-HSI，其核心解决方案是通过自动代码生成机制，将用户的自然语言描述与手势示范转化为可执行的个性化状态机代码，从而实现对群集机器人行为的实时、自定义控制。关键在于：基于多模态输入（自然语言与单/双手势），构建端到端的自动化接口生成框架，使未受训操作者能够以直观方式指定机器人群体在特定手势触发下的集体运动、编队形态及形变行为，并支持运行时动态更新。该方案在仿真与真实机器人实验中均验证了其在复杂任务场景下的有效性，包括目标得分、迷宫穿越与双目标并行完成等。
链接: https://arxiv.org/abs/2609.16346
作者: Alessandro Nazzari,Nathan Cerisara,Dorian Tonnis,Raina Zakir,Lorenzo Labarile,Weixu Zhu,Marco Dorigo,Mary Katherine Heinrich
机构: IRIDIA, Université Libre de Bruxelles, Brussels, Belgium; Department of Aerospace Science and Technology, Politecnico di Milano, Milan, Italy; Télécom Physique Strasbourg, Université de Strasbourg, France; Polytech Lyon, Lyon 1 Université Claude Bernard, Lyon, France; University of Pisa, Pisa, Italy; Scuola Superiore Sant’Anna, Pisa, Italy
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC); Multiagent Systems (cs.MA)
备注:
Abstract:This paper presents Auto-HSI, a method for generating personalized human-swarm interaction (HSI) interfaces on demand. The objective is to enable untrained operators to use natural language descriptions and gesture demonstrations to explain how they want the robots to collectively behave in response to their gestures. Based on these inputs, the code should automatically be generated for personalized state machines that will control the robots as desired, in response to the desired gesture inputs. In the developed Auto-HSI prototype, the generated code produces a personalized interface for centralized control using one- and two-handed gestures, enabling a user to teleoperate the robots’ motion, formation shape, and shape deformation. We test the gesture tracking and code generation components of Auto-HSI against performance benchmarks. We then test the full Auto-HSI prototype in ``live’’ operation experiments, in which real human operators centrally control 50 simulated robots in a physics-based simulator, under nominal and noisy conditions. In these experiments, robots are teleoperated to: score a goal, traverse a maze that requires shape deformation, and score two simultaneous goals by splitting into two groups. We also demonstrate a real human operator making live updates to their personalized Auto-HSI interface during operation (in simulation). Finally, we demonstrate live operation of real robots.
[MA-13] BLINDSPOT: A Benchmark for Safety and Refusal Calibration in Long-Horizon Tool-Using Agents
【速读】：该论文旨在解决长时程工具使用型大语言模型（LLM）代理在复杂交互环境中面临的安全性评估难题，尤其关注安全失效可能在多轮交互后才显现，而现有评估方法通常仅以任务成功或攻击成功率作为衡量标准，忽视了代理在交互过程中是否保持恰当的安全-效用校准（safety-utility calibration），如是否合理拒绝、是否持续安全执行等。其解决方案的关键在于提出Blindspot——一个面向长时程轨迹级安全校准的基准测试框架。该框架通过自适应对抗性交互、状态化工具执行和基于执行结果的裁决机制，全面评估用户-代理-环境完整交互轨迹。其核心创新在于构建了一个可扩展的实时仿真系统，支持在不重构评估流程的前提下动态添加攻击类型、场景、工具、策略、领域及代理配置。实验采用8项指标对13个开源与专有模型进行评估，结果显示不同模型在安全-效用平衡上存在显著差异，且安全失败常在若干初始安全交互步骤后才暴露，这表明应将代理安全性视为轨迹级属性，而非单轮或二元成败判定。
链接: https://arxiv.org/abs/2609.16305
作者: Sadia Asif,Mohammad Mohammadi Amiri,Momin Abbas,Tejaswini Pedapati,Prasanna Sattigeri
机构: Rensselaer Polytechnic Institute (伦斯勒理工学院); IBM Research (国际商业机器公司研究部)
类目: Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE); Computation and Language (cs.CL); Machine Learning (cs.LG); Multiagent Systems (cs.MA)
备注:
Abstract:Large language model (LLM) agents increasingly operate over long-horizon interactions involving tool use, persistent state, evolving authorization, and external environment feedback. In such settings, safety failures may emerge only after multiple turns, yet existing evaluations often reduce agent behavior to task or attack success, obscuring whether an agent acts, refuses, or remains appropriately calibrated as the interaction evolves. We introduce Blindspot, a benchmark for trajectory-level safety calibration of long-horizon tool-using agents. Blindspot evaluates complete user-agent-environment trajectories through adaptive adversarial interaction, stateful tool execution, and execution-grounded adjudication. Its current instantiation contains 22 attack families and 35 scenarios across seven domains, yielding more than 2,500 long-horizon trajectories with an average interaction length of 14.7 turns. Each trajectory is assigned one of five outcomes: Safe Completion, Correct Refusal, Unsafe Completion, Over-Refusal, or Indeterminate. Unlike fixed attack datasets, Blindspot is an extensible live-simulation framework in which attacks, scenarios, tools, policies, domains, and agent configurations can be added without redesigning the evaluation pipeline. We evaluate 13 proprietary and open-weight LLMs using eight metrics covering unsafe completion, appropriate refusal, benign utility, over-refusal, repeated-run robustness, and post-refusal failure. Preliminary results reveal substantial differences in safety-utility calibration across models and show that failures can emerge only after several initially safe interaction steps. These findings motivate treating agent safety as a trajectory-level property rather than a single-turn or binary success criterion.
[MA-14] Cheap Talk Stabilizes Strategic Interaction in LLM Agents
【速读】：该论文旨在解决大语言模型（Large Language Models, LLMs）作为交互式智能体在重复交互中行为策略难以保持稳定的问题，尤其关注非约束性预互动沟通（即“廉价谈话”，cheap talk）对策略持久性的影响。其核心挑战在于：尽管大模型在多轮博弈中表现出复杂的行为模式，但其策略常出现不稳定甚至反复切换的现象，影响多智能体系统可靠性。论文的关键解决方案是通过实证分析四类开放权重、参数规模为7-9B的主流大模型在四种不同激励结构的重复博弈（囚徒困境、雪堆博弈、猎鹿博弈与和谐博弈）中的表现，揭示廉价谈话如何通过调节行为决策过程提升策略稳定性。研究发现，廉价谈话总体上具有显著的稳定作用，且其效果依赖于模型类型与情境设定；进一步的控制实验表明，以通义千问（Qwen）为例，廉价谈话通过降低动作不确定性与减少回合间动作概率漂移两个可分离的输出层面机制发挥作用。此外，基于历史-消息匹配的反事实分析显示，近期合作方行为会调节互利型与自我优先型语言对策略持久性的不同影响。更关键的是，在囚徒困境任务中，研究识别出通义千问与Falcon模型在深层Transformer层中存在一种“历史平衡的策略内容方向”，将其投影移除后显著增加闭环交互中的策略切换频率，证明完整行为轨迹对该成分具有因果敏感性。综上，该研究揭示了廉价谈话可通过特定内部表征机制增强策略持久性，同时强调其效应具有高度模型依赖性和历史依赖性，为构建稳定、可预测的多智能体交互系统提供了重要理论依据与技术路径。
链接: https://arxiv.org/abs/2609.16270
作者: Nunzio Lorè,Hongan Zhu,Babak Heydari
机构: 未知
类目: Multiagent Systems (cs.MA)
备注: 23 pages, 8 figures, 14 tables. Includes supplementary material
Abstract:Large language models are increasingly deployed as interacting agents, making the persistence of their action policies across repeated interaction critical for reliable multi-agent operation. We investigate whether and how agent-generated, non-binding pre-play communication (“cheap talk”) increases such persistence in four open-weight 7-9B-parameter LLMs. Our experiments span four repeated two-player games – Prisoner’s Dilemma, Snowdrift, Stag Hunt, and Harmony – with incentive structures ranging from strategic conflict to alignment, each presented in six contexts. We observe unstable trajectories in all four games, although their prevalence and magnitude depend strongly on model and context. Across models, games, and contexts, cheap talk is predominantly stabilizing, with five corrected reversals concentrated in social or team framings; effects vary substantially by model and context. Controlled current-message interventions identify two separable output-level channels in Qwen: reduced action uncertainty and less between-round drift in action probabilities. Matched history-by-message counterfactuals further show that recent partner behavior conditions how mutual-benefit versus self-prioritizing language affects policy persistence. Finally, in Prisoner’s Dilemma, we identify in Qwen and Falcon a history-balanced policy-content direction in late transformer layers; projecting out this direction increases realized switching during closed-loop play, demonstrating that complete trajectories are causally sensitive to this component. Together, these findings show that cheap talk can make individual trajectories more persistent across diverse incentive structures, while revealing that the magnitude and mechanisms of stabilization are model- and history-dependent.
[MA-15] GPEvac: GNN-Based PPO for Adaptive Evacuation Routing During Shooting Events
【速读】：该论文旨在解决大规模枪击事件中受害者实时避险路径规划的紧迫问题，核心挑战在于如何在存在敌意不确定性与人群聚集动态的情况下，最小化威胁暴露并实现高效疏散。现有方法受限于特定建筑布局的静态策略，且在大规模场景下计算不可行；而实际指导原则仅提供“逃跑、躲藏或对抗”等模糊建议，缺乏可操作性。本文提出GPEvac：一种基于图神经网络（Graph Neural Network, GNN）与近端策略优化（Proximal Policy Optimization, PPO）的自适应疏散框架，其关键创新在于引入一种以边为先的序列消息传递机制，并结合可学习的虚拟全局节点，有效捕捉局部与长距离依赖关系。由此生成的图嵌入被集成至排列不变的评分机制中，使单一学习策略能够泛化应用于不同拓扑结构与规模的建筑布局。仿真结果表明，GPEvac在多种建筑布局下均显著优于智能基线，大幅降低总威胁暴露量；更重要的是，系统可在本地CPU上仅用14.73毫秒完成全局疏散路径计算，具备与实时监控系统无缝集成的能力。该方法不仅可挽救枪击事件中的生命，其核心技术亦可迁移至关键基础设施、智能交通系统及自适应传感器网络等图结构决策领域。
链接: https://arxiv.org/abs/2609.16163
作者: Daniel Perkins,Subhadeep Chakraborty
机构: 未知
类目: Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Machine Learning (cs.LG); Multiagent Systems (cs.MA); Systems and Control (eess.SY)
备注: 7 pages, 4 figures, 3 tables
Abstract:The sharp increase in mass shootings underscores an urgent need for systems that guide victims to safety in real time. An effective evacuation system must minimize threat exposure while also accounting for adversarial uncertainty and crowding dynamics. Current methods in the literature are rigidly constrained to layout-specific policies and computationally intractable in large-scale layouts, while practical guidelines simply advise victims to “run”, “hide”, or “fight”. We propose GPEvac: a GNN-based PPO framework that computes adaptive evacuation routes during shooting events. To capture both local and long-distance dependencies, we introduce an edge-first sequential message-passing scheme with a learnable virtual global node. The resulting graph embeddings are integrated into a permutation-invariant scoring mechanism that allows a single learned policy to operate across building layouts of diverse topologies and sizes. Through extensive simulation, we show that GPEvac outperforms intelligent baselines across distinct architectural layouts, significantly reducing total threat exposure. Crucially, the system computes global evacuation routes in just 14.73 ms on local CPU hardware, enabling seamless integration with live surveillance systems. In addition to saving lives during shooting events, the methodologies developed are transferable to other graph-structured decision-making domains, including critical infrastructure, intelligent transportation systems, and adaptive sensor networks.
[MA-16] “Looking for Something Weird to Happen”: How Humans Sustain AI Agent Novelty Amid Semantic Collapse
【速读】：该论文旨在解决生成式 AI 系统在开放社会网络环境中出现的语义坍缩（semantic collapse）问题，即随着交互时间推移，AI 代理（AI agent）生成内容的多样性逐渐下降，个体内部输出趋于单一，跨代理间输出趋于同质化。尽管已有研究多聚焦于封闭场景下的模型与数据层面的修复策略，但本研究在 MOLTBOOK 这一由人类用户配置并引导的交互式 AI 代理社交网络中展开，揭示了语义坍缩在动态开放环境中的演化机制。其解决方案的关键在于发现：少数保持高新颖性的代理能够持续输出多样化内容，而这种可持续性源于三方面核心特征——用户本身重视新颖性本身、主动提供广泛且独特的输入并根据输出收敛趋势及时调整内容、将 MOLTBOOK 视为可探索的新型代理世界而非工具性利用场所。进一步的访谈与调查验证了这些行为模式，并表明高新颖性代理所在的社区整体输出多样性更高。因此，论文提出通过界面设计与政策干预优化人类输入质量，以缓解语义坍缩，维持系统长期多样性。
链接: https://arxiv.org/abs/2609.16051
作者: Shiyang Lai,Arna Woemmel,Hongkai Mao,Junsol Kim,Summer Eunhyung Ann,James Evans
机构: University of Chicago(芝加哥大学); Stanford University(斯坦福大学)
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注:
Abstract:Semantic collapse, the progressive narrowing of what AI systems generate, has been studied mainly in closed settings, and remedies have targeted models and data. We study it in MOLTBOOK, a social network of interacting AI agents that human users configure and steer. Across 30,076 active agents, output grows less diverse within agents and more similar across them over weeks, yet a minority sustains high novelty. Interviews with users of high- and typical-novelty agents (N=11) associate sustained novelty with three features: users value novelty of itself, they supply broad and distinctive material and revise it when output narrows, and they approach MOLTBOOK as a new agentic world to explore, not a venue to instrumentally exploit. A survey of users of distinctive agents (N=53) confirms these patterns. Communities with more novel agents also show more diverse output from other agents. We discuss interface and policy interventions that could support improved human input.
[MA-17] Social Behavior Among Autonomous AI: How Large Language Models Interact in Dynamic Networks
【速读】：该论文旨在解决在动态不确定环境中，基于大语言模型（Large Language Models, LLMs）的智能体之间（包括人机协作与机器间交互）如何实现有效合作的问题。其核心挑战在于理解不同模型架构、网络拓扑结构及提示工程对合作行为的影响机制。解决方案的关键在于通过构建一个基于公共品博弈（public goods game）的实验框架，在瓦茨-斯特罗加茨（Watts-Strogatz, WS）、巴尔巴西-阿尔伯特（Barabasi-Albert, BA）和埃拉托斯-雷尼（Erdos-Renyi, ER）三类动态网络结构中，系统评估多个主流大语言模型（Mistral、Llama3、Gemma3 和 Phi3）在单模型与混合模型条件下的合作表现。研究发现，模型架构显著影响合作水平：Mistral 和 Llama3 表现出高合作倾向，而 Phi3 则呈现明显的背叛行为；网络拓扑方面，ER 网络的随机性结构显著促进合作；此外，提示设计起到关键作用，采用“社会整体利益导向”提示可显著提升合作率。这些结果为构建基于大语言模型的自适应社会网络仿真提供了初步理论框架与实践指导。
链接: https://arxiv.org/abs/2609.16013
作者: Narges Fardnia,Fatemeh Seyedin,Matthias Becker,Mahmoudreza Babaei,Adrian Weller
机构: 未知
类目: ocial and Information Networks (cs.SI); Multiagent Systems (cs.MA)
备注: 7 pages, 5 figures, 2 tables. Accepted at LLAIS 2025: Workshop on Large Language Model Agents for Intelligent Systems, Bologna, Italy
Abstract:Cooperation is a cornerstone of human societies, enabling collective progress in dynamic and uncertain environments. With the advent of AI systems acting autonomously, it becomes crucial to understand not only human-AI cooperation but also AI-AI interactions in adaptive networks. In this work, we examine the interactions of AI using Large Language Models – Mistral, Llama3, Gemma3, and Phi3 – in a public goods game within dynamic network structures. Our experiments were conducted under single-model and mixed-model conditions across Watts-Strogatz (WS), Barabasi-Albert (BA), and Erdos-Renyi (ER) networks. We analyzed the impact of model architecture, network topology, and prompt design on cooperative behavior. Results show that Mistral and Llama3 offer high cooperation rates, while Phi3 shows defective tendencies. Additionally, the random structure of Erdos-Renyi networks dramatically improves cooperation. Prompt design also plays a key role; a society-benefits prompt leads to a higher cooperation level. These findings offer a preliminary framework for LLM-based simulations in adaptive social networks.
[MA-18] Intervention problems in the Linear Threshold Model: A general formulation and new results
【速读】：该论文旨在解决线性阈值模型（Linear Threshold Model）下的最优干预问题，即在动态网络系统中，如何以最低成本促使所有节点最终达成全1状态（all-1 configuration）。在此模型中，每个节点（代表一个个体）根据其邻居中采取行动1的比例是否达到预设阈值来决定自身行为。当所有阈值均为1/2时，研究发现最优干预成本恰好等于一种新的图论量——定向路径数（oriented path number），即能够被定向为有向无环图（DAG）的不相交路径的最小数量。该指标成为刻画最优干预成本的核心要素；在一般情况下，定向路径数构成对最优干预成本的一个关键上界，因此其本质是设计高效干预策略的关键所在。
链接: https://arxiv.org/abs/2609.17146
作者: Giacomo Como,Fabio Fagnani,Stephane Durand
机构: University of Milano-Bicocca(米兰-比科卡大学); University of Paris-Saclay(巴黎-萨克雷大学)
类目: Optimization and Control (math.OC); Computer Science and Game Theory (cs.GT); Multiagent Systems (cs.MA); Social and Information Networks (cs.SI); Systems and Control (eess.SY)
备注:
Abstract:We study an optimal intervention problem for linear threshold models. This is a popular class of dynamical network systems whereby a number of agents, identified with the nodes of a graph, strategically change their binary action (0 or 1) according to a threshold rule. Specifically, an agent adopts action 1 if and only if the fraction of its neighbors in the interaction graph that do so is greater than or equal to a prescribed threshold. Assuming that a planner can modify the agents’ thresholds at a cost equal to the aggregate threshold increase, we study the minimum intervention cost needed to ensure global convergence to the all-1 configuration. Our main contribution is the introduction of a new graph-theoretic quantity, called oriented path number, that is the minimum number of disjoint paths needed to cover the graph that can be oriented to form a directed acyclic graph. When thresholds are all equal to 1/2, the optimal cost is shown to coincide with the oriented path number, whereas, in the general case, it turns out to be the main ingredient of a bound on the optimal intervention cost.
[MA-19] he fixed-point bundle method over product-of-simplex domains arising from game equilibria
【速读】：该论文旨在解决在多玩家博弈中求解纳什均衡（Nash equilibrium）的计算难题，具体针对定义在单纯形乘积域（product-of-simplex domain）上的变分不等式（Variational Inequality, VI）问题。传统固定点捆绑框架（fixed-point bundle framework）仅适用于单一单纯形域，难以直接推广至多玩家博弈中的策略空间结构。其关键解决方案在于：通过构造一个等价的定义在单一单纯形域上的变分不等式，并利用纤维丛同构（fiber bundle isomorphism）建立两个框架之间的几何等价性。这一构造揭示了产品单纯形域上固定点捆绑结构的复合纤维丛（composite fiber bundle）特性，从而使得原问题的预测-校正路径跟踪算法能够继承单一单纯形域框架的收敛性质，即在解附近实现线性间隙缩减的全局收敛性。数值实验在5600个随机生成的实例上验证了该方法的鲁棒性，所有测试实例均成功收敛，表明该方法在高维、大规模博弈场景下具有良好的实用性与可靠性。
链接: https://arxiv.org/abs/2609.16158
作者: Hongbo Sun
机构: Tsinghua University (清华大学)
类目: Optimization and Control (math.OC); Computer Science and Game Theory (cs.GT); Multiagent Systems (cs.MA)
备注: 26 pages, 1 table, experiment codes and results are available at this https URL
Abstract:This paper extends the fixed-point bundle framework for finite-dimensional variational inequalities (VIs) from the simplex domain to the product-of-simplex domain, which is directly applicable to solving Nash equilibria. The fixed-point bundle for VIs on the product-of-simplex domain reveals a composite fiber bundle structure. The key innovation is to construct an equivalent VI on the simplex domain and establish the equivalence between the two fixed-point bundle frameworks via a fiber bundle isomorphism. Exploiting this geometric equivalence, the predictor-corrector path-following algorithm for the VI on the product-of-simplex domain is shown to inherit the convergence guarantee of the simplex-domain framework, namely, global convergence with linear gap reduction near solutions. Numerical experiments on 5600 randomly generated instances with dimensions ranging from 2-player 128-action to 128-player 2-action demonstrate robust performance. The algorithm converges in every tested instance.
自然语言处理
[NLP-0] ScienceBuddy: Recursive-in-Recursive Self-Improvement for Interactive Scientific Agents
【速读】： 该论文旨在解决科学智能（Scientific AI）在实际研究流程中难以持续进化与适配科研需求的核心问题。现有生成式AI系统多为静态模型，缺乏根据用户反馈和任务执行证据进行自我迭代的能力，导致其在复杂、动态的科研场景中表现受限。为此，论文提出ScienceBuddy这一交互式科研工作平台，其核心解决方案在于引入“递归嵌套式自进化”（recursive-in-recursive self-improvement）范式：内层递归在固定模型参数的前提下优化任务执行框架（harness），外层递归则基于改进后的框架对模型进行强化学习训练。该机制实现了框架演化与模型学习的双向协同——框架进化塑造更高质量的训练经验，而模型能力提升又为框架适应性优化提供新契机。通过四类科学任务的案例研究，验证了该范式在研究人员交互、框架精炼与模型学习方面的有效性。通过将ScienceBuddy作为研究产品开源发布，论文推动了科学智能向“持续协作型发现智能”的演进，使AI能够伴随科学研究进程不断进化。
链接: https://arxiv.org/abs/2609.17523
作者: Shuhan Xue,Jianyuan Zhong,Ziyuan Nan,Wenbin Li,Zhaochen Yu,Jinchao Ding,Qiang Gao,Pengyu Zhan,Yuntong Zhang,Tian Cheng,Zhenfei Yin,Yingcheng Wu,Ling Yang
机构: PhAI Labs; Department of Hepatobiliary Surgery and Transplantation, Liver Cancer Institute, Zhongshan Hospital, Fudan University; State Key Laboratory of Genetics and Development of Complex Phenotypes; Fudan University; Shanghai Academy of Natural Sciences; Shunwei Capital; University of Oxford; Stanford University; Princeton University
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: Website: this http URL , Code: this https URL
Abstract:We introduce and release ScienceBuddy, an interactive scientific research workspace that brings continually improving scientific agents into researchers’ everyday workflows. ScienceBuddy supports researchers in carrying out scientific tasks while transforming their requests, feedback, and execution evidence into tasks and evaluation rubrics for continual learning. At its core is recursive-in-recursive self-improvement, a paradigm that couples harness evolution with model reinforcement learning: the inner recursion improves the harness with the model fixed, while the outer recursion trains the model under the improved harness. Harness evolution shapes training experience, and model learning creates new opportunities for harness adaptation. We present case studies of researcher interaction, harness refinement, and model learning, with the benchmark cases spanning four scientific task families. By releasing ScienceBuddy as a research product, we make this paradigm available to the scientific community and take a step toward discovery intelligence: scientific AI that advances through sustained collaboration with researchers and evolves alongside the research it supports. Website: this http URL
[NLP-1] When Should LLM s Abstain? Chain-of-Self-Questioning for Selective Risk Control
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在缺乏充分事实支持时仍会生成看似合理但错误回答的问题，即“无依据承诺”（unconditional wrong-commitment）问题。其核心挑战在于如何在不依赖额外训练或模型修改的前提下，提升模型对自身知识边界的认知能力，从而实现更可靠的答与不答决策。解决方案的关键是提出一种仅通过提示工程（prompt-only）的框架——自问自答链（Chain-of-Self-Questioning, CoSQ），该框架通过显式要求模型自我评估回答所需的信息完整性，使答案输出条件化于信息充分性判断。实验表明，在TruthfulQA多选验证集上，基于阈值τ=0.90的有根自问自答（Grounded-CoSQ）将平均无条件错误承诺率从13.1%降至8.9%，相对降低32.1%，同时准确率由86.9%提升至89.7%，覆盖率达87.6%。所有十一类开放权重及托管模型均表现出一致改进，且在不同阈值下稳定性良好。此外，针对自然问答短答案任务的补充实验进一步验证了该方法在开放式生成场景中的有效性。研究证明，通过可调的自我评估机制，模型可在事实不足时选择拒绝回答，从而在错误代价高于转介或审查的情境下实现更稳健的决策。
链接: https://arxiv.org/abs/2609.17516
作者: Ali Şenol
机构: Tarsus University (塔尔苏斯大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models can produce fluent answers when their factual support is weak. This paper introduces Chain-of-Self-Questioning (CoSQ), a prompt-only framework that makes answer commitment conditional on an explicit assessment of the information required to answer a question. We evaluate three CoSQ variants under seventeen conditions on the 817-item TruthfulQA multiple-choice validation set using eleven open-weight and hosted model families. In the final balanced-option protocol, Grounded-CoSQ at \tau=0.90 reduces the mean unconditional wrong-commitment rate from 13.1% under chain-of-thought prompting to 8.9%, a 32.1% relative reduction, while increasing answered accuracy from 86.9% to 89.7% and answering 87.6% of questions. Both improvements hold for all eleven models and at every evaluated threshold. Critical-CoSQ and Adaptive-CoSQ provide neighboring operating points with 88.6% and 86.5% coverage, respectively, while remaining more reliable than the baseline. A secondary Natural Questions Short-Answer evaluation provides convergent open-form evidence. These findings show that self-assessment can support explicit, tunable answer-or-abstain decisions when an unsupported commitment is more costly than referral or review.
[NLP-2] What Breaks Under Pruning in Smart Homes and When? Evaluating LLM Degradation Across Architectures and Task Complexity EACL
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在剪枝（pruning）后对上下文感知型工具调用（context-grounded tool calling）性能影响不明确的问题。其核心挑战在于，现有研究多关注剪枝对整体任务准确率的影响，而忽视了剪枝对工具调用中具体语义组件（如操作、设备、参数及取值）和任务复杂度层级上的细微退化。论文的关键解决方案是：在剪枝后通过监督微调（Supervised Fine-Tuning, SFT）恢复模型能力，并系统评估超过19,500个来自三个智能家居数据集的实例，从两个维度深入分析剪枝导致的性能下降——即动作组件层面的分解性退化与任务复杂度敏感性。研究发现，密集型模型具有狭窄的安全剪枝区间且易出现陡峭性能下降，而混合专家模型（Mixture-of-Experts, MoE）则具备更强的剪枝鲁棒性；同时，剪枝优先损害上下文关联的精确性（grounded specificity），而非高层意图理解（schema-level intent），极端剪枝甚至引发系统性拒绝调用行为。因此，论文强调在选择适用于可靠工具执行的剪枝模型时，必须超越聚合准确率的评估，引入细粒度的语义与复杂度维度分析。
链接: https://arxiv.org/abs/2609.17515
作者: Congjing Zhang,Vashishtha Patil,Henning Lange,Usman Aleem
机构: Alexa Home AI, Amazon.com(亚马逊); University of Washington(华盛顿大学)
类目: Computation and Language (cs.CL)
备注: Submitted to EACL Industry Track
Abstract:Pruning can reduce the deployment cost of large language models (LLMs), but its impact on context-grounded tool calling remains poorly understood. We systematically study pruning-induced degradation in smart-home tool calling across four LLMs spanning dense Transformer, dense hybrid, and mixture-of-experts (MoE) architectures, together with depth, width, hybrid, and expert pruning methods. After post-pruning supervised fine-tuning (SFT), we evaluate more than 19,500 instances from three smart-home datasets. Beyond aggregate task accuracy, we characterize degradation along two dimensions: action components (i.e., operation, device, argument, and value) and task complexity. Our results show that dense models have narrow safe pruning regions followed by sharp degradation, while MoE models tolerate substantially more pruning. Pruning degrades grounded specificity before schema-level intent, and aggressive dense pruning can induce systematic over-refusal. These findings highlight the importance of evaluating pruning beyond aggregate accuracy when selecting pruned LLMs for reliable tool execution.
[NLP-3] LACE: Layer-Wise Compression for Dynamic Frame Rate Codecs
【速读】： 该论文旨在解决神经音频编解码器在语音语言建模中因高帧率导致序列长度过长、计算成本高昂的问题。现有动态帧率编解码器虽通过压缩步骤合并多帧以降低有效帧率，但多数方法仅适用于单码本编解码器，或在多层量化前应用单一压缩步骤，导致所有量化层共享相同的分段边界。然而，不同量化层的残差嵌入随时间变化速率各异，固定统一的分段边界限制了压缩效率与重建质量的优化。为此，本文提出层自适应编解码编码（LACE, Layer-Adaptive Codec Encoding），其核心创新在于为每一量化层独立执行压缩步骤，从而实现各层特有的分段边界，提升时序结构的适应性。为进一步支持下游文本到语音（TTS）任务中的应用，LACE引入联合对齐（union alignment）与边界锚点（boundary anchor）机制，确保跨层时长一致性的同时保留压缩优势。实验结果表明，在LibriTTS数据集上，LACE在重建任务中相较已有动态帧率方法实现了更优的码率-质量权衡，并显著提升了TTS推理效率，同时保持了具有竞争力的合成质量。代码已开源集成于ESPnet3编解码器方案中。
链接: https://arxiv.org/abs/2609.17509
作者: Thanapat Trachu,Samuele Cornell,William Chen,Shinji Watanabe
机构: Carnegie Mellon University (卡内基梅隆大学)
类目: ound (cs.SD); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: Accepted to SLT 2026. 8 pages, 5 figures
Abstract:Neural audio codecs are a key component in speech language modeling. However, their high frame rates lead to long sequence lengths, increasing computational costs. Dynamic frame rate codecs mitigate this by reducing the effective frame rate using a compression step to merge multiple frames together. However, most prior methods either operate on single-codebook codecs or apply a single compression step before multi-layer quantization. This forces all quantization layers to share the same segmentation boundaries, despite the residual embeddings at different quantization layers exhibiting different rates of change over time. We propose LACE (Layer-Adaptive Codec Encoding), a dynamic frame rate codec that applies an independent compression step at each quantization layer, enabling layer-specific segmentation boundaries. To use LACE tokens in downstream text-to-speech (TTS), we further introduce union alignment and boundary anchor mechanisms to make durations consistent across layers while preserving compression benefits. Experiments on LibriTTS show that LACE offers a better rate-quality tradeoff than prior dynamic frame rate methods on the reconstruction task and improves TTS inference efficiency while maintaining competitive synthesis quality. Our code is released as part of the ESPnet3 codec recipe.
[NLP-4] Verifiable Social Reasoning for LLM Assistants
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）在日常社交咨询场景中社会推理能力评估困难的问题，核心挑战在于：（i）需在模型从用户主观叙述中学习社会情境的设置下进行评估，而（ii）他人意图等社会属性通常缺乏可验证的真值。为此，本文提出Fuse——一种多智能体仿真框架，用于研究用户介导的社会推理。其关键创新在于构建一个目标智能体（具有隐藏动机）与多个其他智能体（包括代表用户的智能体）交互的环境，用户通过向待评估的LLM助手寻求建议来推断目标智能体的动机，从而通过设计构造出可验证的真值。通过包含24,000条人工标注的人类研究验证了仿真的真实性。将Fuse应用于12个LLM的实证分析揭示：（i）用户中介显著加剧了社会推理的固有难度；（ii）LLM对用户表述中的偏见表现出系统性敏感；（iii）模型所需信息量常超过人类达到正确判断所需的程度；（iv）更长的对话并不必然提升性能，即使提供了澄清问题的机会。研究结果充分展示了该框架的分析价值，并开源了Fuse及包含21,000个样本的数据集。
链接: https://arxiv.org/abs/2609.17496
作者: Amir Taubenfeld,Zorik Gekhman,Avigail Grinstein-Dabush,Itay Laish,Ariel Goldstein,Marian Croak,Avinatan Hassidim,Yossi Matias,Amir Feder
机构: Google Research(谷歌研究院); Hebrew University(希伯来大学); University of Cambridge(剑桥大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: First two authors contributed equally and the order between them was chosen randomly
Abstract:LLM assistants are widely used for daily social advice, yet evaluating their social reasoning in such consultation settings remains challenging since (i) it requires setups where the assistant learns about social situations from subjective user narratives, and (ii) social properties, such as others’ intentions, typically lack verifiable ground truth. To address these challenges, we introduce Fuse, a multi-agent simulation framework for studying user-mediated social reasoning. In Fuse, a target agent with a hidden motive interacts with other agents including one representing the user, who then consults the evaluated assistant to infer the target’s motive, providing verifiable ground truth by construction. Simulation faithfulness is validated through a human study with 24k annotations. We apply Fuse to 12 LLMs and demonstrate its analytical utility by systematically isolating key factors, showing that (i) user mediation compounds the inherent difficulty of social reasoning; (ii) LLMs exhibit systematic sensitivity to biased user framing; (iii) models can require more details than humans need to reach a correct prediction; and (iv) longer conversations do not always improve performance despite providing opportunities for clarifying questions. We open-source Fuse and a dataset with 21k examples.
[NLP-5] Right Tool Right Job: Native-Language Evaluation Tokenizer Sensitivity and Methodological Findings from a French-Only BabyLM EMNLP2026
【速读】： 该论文旨在解决生成式语言模型在法语（尤其是魁北克法语）中对语法细微差异的识别能力评估问题，以及跨语言迁移学习中任务类型对模型表现的影响机制。其核心挑战在于如何准确衡量小规模预训练模型在儿童语言发展水平上的语法理解能力，同时揭示模型性能变化背后的驱动因素。解决方案的关键在于：首先，通过在9247万词的法语文本上预训练一个125M参数的GPT-2模型（MéTRON-FR），并采用原生魁北克法语的语法最小对（grammatical minimal pairs）基准QFrBLiMP进行评估，以确保语言文化特异性；其次，引入基于秩-16低秩适应（LoRA）的跨语言GLUE协议，发现关系类任务性能显著提升而世界知识类任务则退化，揭示了不同任务类型对跨语言迁移的敏感性差异；再次，通过双语词典诱导（Bilingual Lexicon Induction）验证法语嵌入与GPT-2的对齐程度，获得p@1达68.84%的显著优于随机基线（18倍）的结果，表明跨语言对齐更反映语法能力而非训练时长；最后，通过消融实验指出单标记零样本评分在儿童语言尺度下受分词器和提示模板的人工干扰主导，从而提出将分词器替换敏感性测试、安慰剂控制提示和原生语言最小对基准作为标准诊断工具，以提升评估的可靠性与可解释性。
链接: https://arxiv.org/abs/2609.17435
作者: Adam Zachary Wasserman,David Beauchemin
机构: Open Honest Foundation; Group for Research in Artificial Intelligence of Laval University (GRAIL), Université Laval (魁北克大学拉瓦勒分校), Québec, Canada
类目: Computation and Language (cs.CL)
备注: Accepted at BabyLM Workshop at EMNLP 2026
Abstract:We submit MéTRON-FR, a 125M GPT-2 pretrained on 92.47M words of French, to the BabyLM 2026 Strict track. It scores 85.97 +/- 0.17% on QFrBLiMP (a native Quebec-French benchmark of grammatical minimal pairs) and 62.80% on the BabyLM-weighted leaderboard. A cross-lingual GLUE (General Language Understanding Evaluation) protocol that combines French task-data translation with rank-16 LoRA (Low-Rank Adaptation) produces a sharp task-type gradient: relational tasks gain measurably, while world-knowledge tasks regress. Bilingual Lexicon Induction aligns the French embeddings to GPT-2 at p@1 = 68.84 +/- 8.61%, 18X above chance, suggesting cross-lingual alignment tracks acquired grammatical competence rather than training duration. An ablation study shows that single-token zero-shot scoring is dominated by tokenizer and template artifacts at the child scale, motivating tokenizer-swap sensitivity, placebo-controlled prompting, and native-language minimal-pair benchmarks as standard diagnostics.
[NLP-6] Enhancing Accessibility of Medical Texts through Large Language Model-Driven Plain Language Adaptation
【速读】： 该论文旨在解决医疗信息复杂性与患者阅读理解能力之间的鸿沟问题，核心挑战在于如何将专业性强、术语密集的医疗文本高效转化为通俗易懂的通用语言（Plain Language），以提升患者对健康信息的理解与获取效率。其解决方案的关键在于利用大型语言模型（Large Language Models, LLMs）的强大生成能力，结合零样本（zero-shot）与少样本（few-shot）学习范式，在缺乏大量标注数据的情况下实现高质量的文本简化。研究创新性地引入了混合代理（Mixture-of-Agents, MoA）架构，增强模型在多样化任务场景下的适应性与鲁棒性；并通过对比不同提示工程策略、采用QLoRA进行参数高效微调（fine-tuning），系统评估了多种LLM（如GPT-4o-mini、Gemini-1.5-pro、LLaMA）在医疗文本简化任务中的表现。实验结果表明，基于LLM的生成式自动通用语言适配（PLA）方法在保持关键医学信息完整性的同时，显著提升了文本可读性，展现出在临床健康传播中的广泛应用潜力。
链接: https://arxiv.org/abs/2609.17398
作者: Ting-Wei Chang,Hen-Hsen Huang,Hsin-Hsi Chen
机构: National Taiwan University (国立台湾大学); Taiwan Institute of Information Science, Academia Sinica (中央研究院资讯科学研究所); Taiwan AI Research Center (AINTU) (台湾人工智能研究中心)
类目: Computation and Language (cs.CL)
备注: 10 pages, 3 figures, 6 tables. Published in the Proceedings of the Thirty-Third Text REtrieval Conference (TREC 2024), Plain Language Adaptation of Biomedical Abstracts (PLABA) track
Abstract:This paper addresses the challenge of making complex healthcare information more accessible through automated Plain Language Adaptation (PLA). PLA aims to simplify technical medical language, bridging a critical gap between the complexity of healthcare texts and patients’ reading comprehension. Recent advances in Large Language Models (LLMs), such as GPT and BART, have opened new possibilities for PLA, especially in zero-shot and few-shot learning contexts where task-specific data is limited. In this work, we leverage the capabilities of LLMs such as GPT-4o-mini, Gemini-1.5-pro, and LLaMA for text simplification. Additionally, we incorporate Mixture-of-Agents (MoA) techniques to enhance adaptability and robustness in PLA tasks. Key contributions include a comparative analysis of prompting strategies, finetuning with QLoRA on different LLMs, and the integration of MoA technique. Our findings demonstrate the effectiveness of LLM-driven PLA, showcasing its potential in making healthcare information more comprehensible while preserving essential content.
[NLP-7] Large Language Models Develop Belief State Geometry In-Context
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在上下文学习（In-Context Learning, ICL）中所依赖的内部表示机制不明确的问题，特别是这些模型如何通过上下文信息动态推断并维持对隐藏状态的信念。其核心解决方案在于：在受控实验设置下，使用隐马尔可夫模型（Hidden Markov Model, HMM）生成的数据作为输入，直接探测模型残差流激活（residual stream activations）中是否编码了与贝叶斯信念状态（belief state）相关的表征。研究发现，无论模型规模或层数如何，信念状态均可通过线性探测从残差流中高效解码，且探测得到的R²值高达0.83至0.99，表明该表征具有高度可解释性和稳定性。进一步通过修补（patching）和导向（steering）干预探针识别出的关键子空间，验证了其功能相关性——干预后下游预测性能仍保持接近原始模型水平，而对照组则显著下降。这一系列结果提供了在表征层面的实证支持，表明开源大语言模型的上下文学习能力近似于基于上下文推断生成模型的最优贝叶斯推理过程。该研究将输入分布结构与激活空间几何之间的关联从简化网络扩展至大规模生产级模型，深化了对大模型内部认知机制的理解。
链接: https://arxiv.org/abs/2609.17376
作者: Daniel Balcells,Andrew Jun Lee,Chirag Rastogi,Paul M. Riechers,Adam Shai,Xavier Poncini
机构: 未知
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: 87 pages
Abstract:Large language models (LLMs) trained on next-token prediction exhibit remarkable in-context learning (ICL) abilities, yet the representations that support ICL remain poorly understood. We consider such representations in a controlled setting: prompting LLMs with data emitted from hidden Markov models (HMMs) and probing for the corresponding belief state – the posterior distribution over the HMM’s hidden states given the observed token history. Across six open-source LLMs prompted with data from 40 HMMs selected for non-trivial belief structure, we find that belief states are linearly decodable from residual stream activations, with peak probe R^2 -values from 0.83-0.99 across HMM and LLM combinations, ranging from early to late layers. To establish functional relevance, we intervene directly on the probe-identified subspace via patching and steering, resulting in downstream prediction quality on the order of the untampered model, while controls degrade performance substantially. Together, these results provide representation-level evidence that ICL in open-source LLMs approximates optimal Bayesian prediction over a context-inferred generative model. More broadly, our findings extend prior results linking input-distribution structure to activation geometry: from toy networks trained explicitly on HMM data to production-scale LLMs.
[NLP-8] ECHO: A Matched-Contrast Benchmark for Context-Sensitive Turn-Taking in Full-Duplex Dialogue
【速读】： 该论文旨在解决全双工语音对话系统在话轮转换（turn-taking）决策中面临的语境敏感性不足问题，即系统难以准确区分需让出话语权的打断（interruption）与允许持续说话的回应性反馈（backchannel）。现有评估基准通常独立评价事件，导致模型可能通过固定行为偏好（如始终选择让步）获得高分，而非根据上下文做出合理判断。为此，论文提出ECHO——一个针对中文全双工话轮转换的配对诊断基准，其核心设计在于将具有相同重叠语音内容但不同多轮对话上下文的成对样本进行对比：一组上下文要求“让出话语权”（Yield），另一组则要求“保持说话”（Keep）。此外，还引入非相关对话（off-talk）样本以检测不必要的让步行为。关键评估指标为“配对准确率”（pair accuracy），该指标要求对每一对样本均正确决策，且不给予恒定行为策略任何得分奖励，从而有效抑制固定策略的误导性表现。实验结果表明，多数系统存在显著的“让步偏好”（\textscYield bias），在打断场景下表现远优于回应性反馈场景，而少数系统则相对平衡。这揭示了仅基于打断的评估方式会高估实际话轮转换的可靠性。ECHO及其元数据将公开发布，以推动更真实、情境敏感的全双工对话系统研究。
链接: https://arxiv.org/abs/2609.17360
作者: Shuofeng Zhao,Hongwei Cai,Wenke Fan,Qingxiang Guo,Dawei Yang,Zhou Wang,Zhiyang Zhou,Yingxin Shang,Weixu Wang,Lin Yang,Shuran Zhou,Yang Song
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Full-duplex spoken dialogue systems must distinguish interruptions that require yielding the floor from backchannels that permit continued speaking. Existing benchmarks typically evaluate events independently and may therefore reward fixed action preferences rather than context-sensitive decisions. We introduce ECHO, a paired diagnostic benchmark for Chinese full-duplex turn-taking. ECHO pairs examples with the same overlap transcript but contrasting preceding multi-turn dialogue contexts, with one requiring Yield and the other Keep. It additionally includes off-talk examples for diagnosing unnecessary yielding. We introduce pair accuracy, which requires correct decisions on both members of a pair and assigns no credit to constant-action policies. Experiments on multiple full-duplex systems show that most exhibit a pronounced bias toward \textscYield, performing substantially better on interruptions than on backchannels, while another system remains comparatively balanced. These findings demonstrate that interruption-only evaluation can overestimate practical turn-taking reliability. ECHO and its metadata will be publicly released.
[NLP-9] Where Should a Document Live: Context Representations or Parameters?
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在面对其预训练数据之外的知识时，如何高效、低成本且高性能地注入新知识的问题。现有方法包括基于上下文窗口的文档输入、参数微调（parametric）以及将信息编码为潜在表示（representation-based）等，但各类方法在效率、成本与性能之间存在权衡，尚无统一最优方案。本文通过在五个知识密集型基准上对基于键值缓存（KV-cache based）的表示注入方法（如Cartridges、Compaction）与基于微调（fine-tuning-based）的参数化方法进行受控对比，揭示了关键发现：在理想场景下，Cartridges（KV）在几乎所有存储预算下均表现最佳，相比参数化方法高出10个百分点；而Compaction仅在极低压缩率下可媲美Cartridges，当压缩率超过50×时落后10个百分点。在更贴近实际的多文档检索场景中，Cartridges是唯一能与上下文学习（In-Context Learning, ICL）性能相当的方法，领先参数化方法29个百分点，优于Compaction 15个百分点。然而，值得注意的是，Cartridges也是除全量微调和大型MLP适配器外，唯一出现灾难性遗忘（catastrophic forgetting）现象的方法，在控制基准上性能下降6%，在代码任务上更是高达13%。因此，解决方案的关键在于：利用键值缓存机制实现高效的外部知识注入，但在保持泛化能力方面仍面临挑战。
链接: https://arxiv.org/abs/2609.17346
作者: Nathanaël Carraz Rakotonirina,Momchil Hardalov,Gonzalo Iglesias,Adrià de Gispert
机构: Amazon AGI
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:To answer questions outside of their pre-training data, large language models (LLMs) need access to new information, which can be presented in the context window as documents, encoded into the model’s parameters, or injected as latent representations. However, each of these methods comes with different efficiency, cost, and performance trade-offs, with no single winner. We present a controlled comparison of representation-based (KV-cache based) and parametric (fine-tuning-based) adaptation methods on five knowledge-intensive benchmarks. We show that in the oracle setting, Cartridges (KV) are the most accurate injection method at nearly every storage budget, outperforming parametric methods by 10 points. Compaction (KV) matches Cartridges only at low compression rates, lagging behind the parametric methods by 10 points at rates higher than 50\times . In the more realistic multi-document retrieval scenario, Cartridges are the only method that matches in-context learning (ICL), leading the parametric methods by 29 points and Compaction by 15 points. Nonetheless, Cartridges are also the only method, besides full fine-tuning and large MLP adapters, that suffers from catastrophic forgetting, i.e., a 6% performance degradation on control benchmarks, with 13% in coding.
[NLP-10] Vroom-Vroom at SHROOM-Visions: A Multi-Judge Committee for Detecting Hallucinated Spans in Vision-Language Outputs EMNLP
【速读】： 该论文旨在解决多语言环境下视觉-语言模型输出中幻觉角色片段（hallucinated character spans）的检测与分类问题，尤其关注跨语言场景下的泛化能力。其核心解决方案在于采用多个经过微调的视觉-语言模型作为独立标注者，通过字符级多数投票机制融合各模型的片段预测结果，并进一步探索激活探针（activation probes）以增强判别能力。该方法在四项语言任务中的三项中排名第一，且在所有语言和评估指标上均位列前三，展现出卓越的性能。分析表明，不同模型间的预测分歧与人类标注者之间的分歧具有高度相关性，验证了多模型集成策略的有效性与可靠性。
链接: https://arxiv.org/abs/2609.17327
作者: Toqeer Ehsan,Nico Penttilä,Richard Schmidt,Arash Hajikhani,Victoria Palacin
机构: VTT Technical Research Centre of Finland(芬兰技术研究中心)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Accepted to UncertaiNLP 2026 @ EMNLP. SHROOM-Visions 2026 shared task system description
Abstract:This paper describes our submission to the SHROOM-Visions shared task on detecting and classifying hallucinated character spans in vision-language model outputs across four languages. We employ several fine-tuned vision-language models as independent annotators and combine their span predictions through character-level majority voting, and additionally explore activation probes. The approach ranks first in three of four languages and places on the podium in every language and metric. Our analysis indicates that disagreement among diverse models tracks disagreement among human annotators.
[NLP-11] owards Detecting AI-Assisted Responses in Online Surveys EMNLP2026
【速读】： 该论文旨在解决生成式人工智能（Generative AI）在在线问卷调查中被滥用所引发的调研有效性危机问题，尤其关注如何有效识别由大语言模型（LLM）辅助完成的问卷响应。其核心解决方案在于构建首个针对AI辅助问卷参与的基准数据集ASURRE，系统性地涵盖从完全生成与修改到基于角色（persona-grounded）的代理式完成等多种使用策略，并在三个不同学科的真实问卷上生成多模型、多策略的AI响应样本，同时配对真实人类响应以供对比分析。研究发现，简单的AI使用模式可被现有机器生成文本（MGT）检测器轻易识别，但基于角色的代理式完成策略能显著削弱检测性能至随机水平。进一步分析表明，尽管代理式完成难以完全模拟个体层面的答题行为，但仍留下可被识别的行为痕迹。为此，作者提出一种无需训练、仅需少量示例的聚合器，通过整合多个细微行为线索，在代理场景下将平均AUROC提升0.14，显著优于现有最佳检测器。
链接: https://arxiv.org/abs/2609.17317
作者: Qizhou Wang,Bogdan Mamaev,Christopher Leckie
机构: The University of Melbourne(墨尔本大学); Deakin University(迪肯大学)
类目: Computation and Language (cs.CL); Computers and Society (cs.CY)
备注: Accepted to EMNLP 2026 (Main Conference)
Abstract:The use of LLMs to complete online surveys impacts the validity of survey-based research, but detecting such usage remains underexplored. We introduce an initial benchmark dataset, namely ASURRE, for AI-assisted survey participation to capture usage strategies ranging from full generation and revision to persona-grounded agentic completion. Controlled by these strategies, LLM-assisted survey responses are generated using multiple LLMs on three real-world surveys in different disciplines, paired with genuine human responses. Our evaluation of existing machine-generated text (MGT) detectors shows that naive AI usage is readily detectable, whereas persona-grounded agents that mimic entire respondents push detector performance toward chance. We further show that agentic completion cannot fully replicate respondent-level behaviour and leaves distinctive behavioural traces. While individual cues can be circumvented by targeted prompting, a simple few-shot, training-free aggregator over these cues improves mean AUROC by +0.14 over the best existing detector across agentic settings. Our project is available at this https URL.
[NLP-12] Zero-shot narrative detection in social messaging
【速读】： 该论文旨在解决在缺乏标注训练数据的情况下，如何高效识别与分类社交媒体信息中隐藏的战略性叙事（strategic narratives）的问题。其核心挑战在于传统监督学习方法依赖大量特定领域标注数据，而此类数据在多数实际应用场景中难以获取。研究提出的关键解决方案是利用大语言模型（LLM）的零样本（zero-shot）能力，结合人类撰写的叙事描述作为提示（prompt），从而在不需任何训练样本的前提下实现对隐含叙事的有效识别。实验表明，相较于自动生成功能性描述或少量示例（few-shot）提示，人工撰写的情境化叙事描述显著提升模型性能，且集成方法（特别是多数投票）可增强结果鲁棒性。此外，更大规模的模型表现更优，且对提示变化更具鲁棒性。研究证实，通过融合人类编写的叙事描述与简单的集成策略，零样本大语言模型可达到接近监督学习系统的效果，为大规模、跨领域叙事检测提供了一种可扩展的解决方案。
链接: https://arxiv.org/abs/2609.17310
作者: Jesús M. Fraile-Hernández,Anselmo Peñas,Patrick Giedemann
机构: UNED NLP IR Group, Universidad Nacional de Educación a Distancia (西班牙国立远程教育大学); Zurich University of Applied Sciences (苏黎世应用科学大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:This study investigates the zero-shot ability of large language models (LLMs) to identify and classify hidden narratives in social messages. Our research hypothesis is that LLMs’ extensive contextual knowledge allows them to interpret messages on a deeper, pragmatic level, going beyond basic sentiment or topic analysis. Experiments on the Dipromats and SemEval datasets show that providing models with human-written narrative descriptions significantly improves performance, without the need of training examples. In contrast, automatically generated descriptions or the use of few examples (few-shot) often degrade accuracy due to subtle shifts in framing. The study also finds that ensemble methods, particularly majority voting, enhance robustness and that larger models perform best while also being less sensitive to prompt variations. The findings validate that LLMs can effectively detect strategic narratives in a zero-shot setting, and when combined with simple ensembling and human-written descriptions, they can rival supervised systems, offering a scalable solution for narrative detection, specially when there is no training data for the vast majority of domains.
[NLP-13] owards Illusions Awareness in Cyber-Physical Systems Design
【速读】： 该论文旨在解决嵌入式物理系统（Cyber-Physical Systems, CPS）在开放运行环境中因设计假设与实际运行状态之间不匹配而产生的“现实差距”（reality gap）问题。现有方法多聚焦于提升仿真与现实的一致性，却缺乏统一的方法论来系统化地识别、组织并复用导致该差距的失效假设，从而形成可重用的设计知识。本文提出将这种持续依赖已失效假设所引发的对设计模型有效性误判现象称为“设计幻觉”（design illusions），并主张通过构建一种“幻觉感知”（illusions-awareness）的概念性流程，实现对幻觉的识别、分类、表征与利用，将其转化为可指导优化设计决策的行动性知识。其解决方案的关键在于建立一套系统化的机制，使原本隐含且被忽视的失效假设得以显性化，并作为改进设计模型和增强系统鲁棒性的核心知识资产。
链接: https://arxiv.org/abs/2609.17260
作者: Anna Di Placido(UniCA, Laboratoire I3S - COMRED, KAIROS),Nicolas Ferry(UniCA, Laboratoire I3S - COMRED, KAIROS),Julien Deantoni(UniCA, Laboratoire I3S - COMRED, KAIROS)
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Cyber-Physical Systems (CPS) operate through a continuous sense-compute-act loop within an open context environment, making it impossible to anticipate all the situations the system will face. To cope with this openness, stakeholders rely on assumptions, formalized into design models. However, these assumptions may no longer hold once the system is confronted with runtime reality, resulting in a discrepancy between expected and observed behaviour known in literature as the reality gap. Existing approaches mainly focus on reducing or overcoming it by making simulations more faithful to reality, with no unified methodology to structure and exploit invalidated assumptions that give rise to this gap as reusable design knowledge. We refer to the persistent reliance on invalidated assumptions -and the resulting false confidence in the design model’s operational validity -as design illusions, and argue that they need to be made explicit, structured, and exploited as knowledge to support better design decisions. We propose a conceptual pipeline for illusions-awareness that identifies, classifies, characterizes, and leverages illusions to transform them into actionable design knowledge.
[NLP-14] Persistent Recurrent Memory Between Transformer Layers - Improves Language Model Generalization
【速读】： 该论文旨在解决小规模语言模型在训练过程中普遍存在的过拟合问题，进而提升其泛化能力。现有基于纯自注意力机制的解码器仅变压器（decoder-only transformers）在处理长序列时难以有效建模全局上下文依赖关系，尤其在缺乏显式记忆机制的情况下，对序列位置等结构性信息的编码效率较低。为此，论文提出一种简洁的架构改进：在6层变压器的下半部分与上半部分之间引入一个轻量级的持续性循环状态（persistent recurrent state），该状态通过跨注意力（cross-attention）观察隐藏表示，利用门控循环单元（GRU）进行更新，并通过门控加法方式调制后续层的处理过程。该模块仅引入3.7%的额外参数，却使评估损失从2.438 ± 0.004显著降低至1.743 ± 0.018（相对下降28.5%），且在五组随机种子下具有统计显著性（p < 0.01），同时将泛化差距由0.26缩小至0.12。关键发现表明，性能提升完全源自持续性记忆拓扑结构本身，而非辅助的自预测目标；即使移除辅助损失或使用随机损失，模型表现仍保持一致，说明其优势来自记忆机制的结构性设计。此外，表示探针分析显示，该持续性状态能有效编码叙事位置信息（准确率52%，高于33%的随机基准），而标准注意力机制对此类信息的捕捉效率较低。因此，该研究的核心解决方案是：通过在变压器层间嵌入一个轻量、持续的循环记忆模块，以低成本实现对长期依赖和结构化语义信息的有效建模，从而显著提升小规模语言模型的泛化性能。
链接: https://arxiv.org/abs/2609.17251
作者: Eduardo Novaes Hering
机构: FITec Labs / Ericsson São Paulo(埃里克森圣保罗)
类目: Computation and Language (cs.CL)
备注:
Abstract:We introduce a simple architectural modification to decoder-only transformers: a persistent recurrent state that observes hidden representations via cross-attention, updates itself through a GRU, and modulates subsequent processing via gated addition. Inserted between the lower and upper halves of a 6-layer transformer, this module adds only 3.7% additional parameters while reducing evaluation loss from 2.438 \pm 0.004 to 1.743 \pm 0.018 , corresponding to a 28.5% reduction on held-out language modeling data. The improvement is statistically significant across 5 random seeds ( p 0.01 ) and corresponds to reduced overfitting (generalization gap 0.12 vs 0.26). Through controlled ablations, we demonstrate that the improvement stems entirely from the persistent memory topology, not from auxiliary self-prediction objectives. A model with identical topology but no auxiliary loss performs equivalently, while a random auxiliary loss provides no benefit. Representation probing reveals that the persistent state encodes narrative position (52% vs 33% chance level)—information that standard attention maintains less efficiently. Our results suggest that bridging transformer layers with a lightweight recurrent memory is a simple, effective approach to improving generalization in small-scale language models.
[NLP-15] ECHO: Early-layer Collaborative Hierarchical Orchestration with Bonus Logits in Speculative Decoding EMNLP2026
【速读】： 该论文旨在解决无草稿模型（draft-model-free）推测解码在大语言模型（LLM）高效推理中面临的两大核心挑战：过时的草稿候选生成以及验证阶段高昂的计算开销。其解决方案的关键在于提出一种分层双循环框架ECHO，充分利用大语言模型各层之间的功能不对称性——早期层具备高判别效率，而最终层具有权威的概率分布特性。ECHO通过将推理过程划分为高频内循环与低频外循环：在内循环中，利用早期层的附加概率（bonus logits）实现低成本、多步的草稿树快速探索；在外循环中，则通过状态复用机制执行基于完整模型的权威验证，并利用最终层的附加概率对现有路径进行修正，同时为下一周期补充高置信度候选序列。该设计显著提升了平均采纳词元数，并实现了2.4×至2.9×的加速比，在几乎无额外工程负担和部署参数的前提下超越现有最优基线，仅需一次微调即可实现最佳性能。
链接: https://arxiv.org/abs/2609.17241
作者: Ziyang Ma,Zihong Zhang,Zuchao Li,Lefei Zhang,Baoyuan Qi,Siqi Li,Simin Yu
机构: Wuhan University (武汉大学); Xiaomi(小米)
类目: Computation and Language (cs.CL)
备注: Accepted to EMNLP 2026 Main Conference
Abstract:While draft-model-free speculative decoding offers a promising path to efficient LLM inference, it is frequently constrained by stale draft candidates and the high computational cost of the verification. To address these challenges, we propose ECHO, a hierarchical dual-loop framework that exploits the functional asymmetry between LLM layers. Leveraging the high discriminative efficiency of early layers and the authoritative distribution of final layers, ECHO bifurcates inference into a high-frequency inner loop and a low-frequency outer loop. Within the inner loop, early-layer bonus logits drive rapid, multi-step draft-tree exploration at a minimal cost. Simultaneously, the outer loop performs authoritative full-model verification through a state-reuse mechanism. Crucially, the outer loop also utilizes final-layer bonus logits to correct existing paths and supplement the tree with high-confidence candidates for subsequent cycles. Experimental results across diverse benchmarks demonstrate that ECHO significantly boosts mean accepted tokens and achieves a 2.4 \times to 2.9 \times speedup, outperforming existing state-of-the-art baselines with negligible engineering overhead and no extra deployment parameters, albeit with a one-shot fine-tuning dependency for optimal acceleration. The code is available at this https URL.
[NLP-16] AraMIP: Extending MIPVU Towards Metaphor Identification in Arabic EMNLP2026
【速读】： 该论文旨在解决阿拉伯语中隐喻识别与标注资源匮乏的问题，尤其针对阿拉伯语在语言特性（如形态复杂性、词义排序不一致及缺乏标准化上下文材料）方面带来的标注挑战。其解决方案的关键在于提出一种全新的阿拉伯语隐喻识别规范——阿拉伯语隐喻识别程序（Arabic Metaphor Identification Procedure, AraMIP），该方法基于广泛使用的维莱大学隐喻识别程序（Metaphor Identification Procedure Vrije Universiteit, MIPVU）框架，同时针对阿拉伯语的语言特点进行了适配性调整。AraMIP明确区分了三种主要的阿拉伯语修辞表达类型：比喻（Isti’ara，即隐喻）、借代/间接表达（Kinaya，即转喻或间接表达）和明喻（Tashbih），并在此基础上构建了一个包含300个句子（共5277个词）的试点标注数据集。该研究不仅揭示了阿拉伯语隐喻标注中的核心难点，还为建立标准化的阿拉伯语修辞表达资源奠定了基础，推动后续更大规模标注数据集的开发与相关研究的开展。
链接: https://arxiv.org/abs/2609.17235
作者: Mandar Marathe,Manar Ali,Sara Nabhani,Raia Abu Ahmad,Ibrahim Baroud,Omar Momen
机构: Quality Usability Lab, Technical University of Berlin, Germany; SOAS University of London, United Kingdom; CRC 1646 ‘Linguistic Creativity in Communication’, Bielefeld University, Germany; Computational Linguistics, CLCG, University of Groningen, The Netherlands; German Research Center for Artificial Intelligence (DFKI), Germany
类目: Computation and Language (cs.CL)
备注: Accepted at the Fourth Arabic Natural Language Processing Conference (ArabicNLP 2026), co-located with EMNLP 2026
Abstract:Metaphor research has gained increasing attention due to its relevance to linguistic creativity, language use, cognitive processes, and related areas. While many efforts have been devoted to metaphor identification and annotation in English and other languages, Arabic remains under-resourced in this area. In this work, we propose the Arabic Metaphor Identification Procedure (AraMIP), a novel guideline for Arabic metaphor annotation. AraMIP builds on the widely used Metaphor Identification Procedure Vrije Universiteit (MIPVU) framework, incorporating adaptations that accounts for the language-specific properties of Arabic. We distinguish three major types of Arabic figurative language: Isti’ara (metaphor), kinaya (metonymy/indirect expression), and tashbih (simile), and annotate a pilot dataset of 300 sentences (5277 words). Our analysis reveals key challenges specific to Arabic, including morphological complexity, inconsistencies in dictionary sense ordering, and the absence of standardized contextual materials for annotators. This work contributes a first step toward standardized Arabic figurative instances and facilitates the development of larger annotated resources, thereby supporting future research on figurative language in Arabic.
[NLP-17] Easy to Catch a Liar Hard to Clear an Honest One: Language Models Diagnosing a Corrupted Reward Channel from a Verified Record ATC
【速读】： 该论文旨在解决强化学习中智能体对奖励报告可信度的判断问题，即当奖励报告突然变化时，智能体无法仅凭历史数据区分是环境发生了真实改变还是报告者（如语言模型）出现了欺骗行为。这一问题在理论上不可解，因为从奖励序列本身无法区分这两种情况。其解决方案的关键在于引入关于报告者的额外信息——具体而言，通过添加一条独立验证的记录（即某一轮真实结果的独立核查），使智能体能够基于更丰富的上下文判断报告者是否诚实。实验设计构建了一个双选项游戏，其中“支付交换”与“说谎报告者”产生完全相同的字节级历史记录，仅通过增加一条经验证的真实结果作为附加信息来区分。研究发现，尽管70B规模的大模型几乎在所有条件下都能准确识别说谎报告者，但对诚实报告者的误判率极高：一个72B的模型在无变化情况下将诚实报告者误判为说谎者达38%，在支付规则变动时更高至58%；另一个70B模型的误判率也分别达到26%和48%。这种失败并非源于理解能力不足，因为在已知答案的提示下，相同模型的表现接近完美（0.96–1.00）。错误的根本原因在于特定的表面特征——对于Qwen系列模型，是验证记录所标注的轮次位置；对于Llama系列模型，则是表示“诚实”的字母选择。此外，向已包含答案的提示中加入验证记录反而降低了Llama模型给出正确答案的概率。研究者此前预测误判率为35%，实际结果远高于预期，表明当前生成式AI在处理此类信任推理任务时存在显著偏差，且其决策机制受无关表面特征干扰严重。
链接: https://arxiv.org/abs/2609.17226
作者: Arman Nik Khah
机构: The University of Texas at Dallas(德克萨斯大学达拉斯分校)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 15 pages, 9 tables. Code, prompts, answer keys, and every scored output: this https URL
Abstract:An agent that learns from rewards has to trust whatever reports those rewards. When the reports suddenly change, either the world changed or the reporter broke. From the reports alone these are indistinguishable, and reinforcement learning theory shows that no amount of further experience separates them. The prescribed escape is richer data about the reporter itself. We ask whether a frozen language model, handed exactly that data, uses it. We build a two-option game in which a payout swap and a lying reporter produce byte-identical histories. Then we add one verified record: an independent check of one round’s real result, printed beside what the reporter said about that round. That single line settles the case. We ask three large models, from two families, to answer one question with one letter. Is the reporter honest or lying? They catch a lying reporter almost perfectly. At the 70B class that holds in every condition we tried; the 32B model slips in one wording. They clear an honest reporter far less often, and how often depends on things that should not matter. Averaged over rounds, letters, and wordings, a 72B model calls an honest reporter a liar 38% of the time when nothing has changed at all, and 58% of the time when the payouts moved. A 70B model from a second family calls an honest reporter a liar 26% and 48% of the time. The failure is not one of reading, because in the situation where nothing changed the same models score 0.96 to 1.00 with the answer printed in the prompt. Which surface feature drives it differs by family. For the Qwen models it is which round the record names, and for Llama it is which letter stands for “honest.” Adding the record to a prompt that already states the answer makes Llama less likely to give that answer. We had registered a prediction for that 58% before the run: 35%. The failure is larger than we expected.
[NLP-18] Psychological Effects of Cultural Upheavals from Millions of Song Lyrics Over 100 Years
【速读】： 该论文旨在解决文化巨变（cultural upheavals）对个体在大众媒体中语言表达影响的规模化识别问题，尤其关注其在流行音乐歌词中的体现。传统研究多聚焦于社会层面的语言变迁，但缺乏对个体心理状态在大规模创作行为中反映的系统性分析。本文通过分析跨越一个世纪、涵盖数百万首歌曲的歌词数据，识别越南战争、9/11恐怖袭击及新冠疫情三大事件前后艺术家语言模式的变化。其解决方案的关键在于：利用自指代词（self-references）作为心理疏离（psychological distancing）的指标，发现重大危机后自指频率显著下降；同时，认知加工类词汇（cognitive processing terms）的上升表明艺术家更积极尝试从剧变中建构意义；内容分析进一步显示，危机后歌词更多涉及“生命与自由”（社会条件）主题，而减少对“求爱与夜生活”（人际联结）的描写。这些发现揭示了文化巨变通过语言形式深刻重塑个体与集体的表达方式，凸显其深远的社会影响。
链接: https://arxiv.org/abs/2609.17225
作者: David M. Markowitz
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Cultural upheavals impact many aspects of social life, and many studies have investigated their impact on language patterns. However, few investigations have isolated the impact of upheavals on individuals at scale in popular media. The current work evaluated millions of song lyrics spanning more than a century in search of within-artist and between-artist signals of distress from the Vietnam War, the terrorist attacks of 9/11, and COVID-19. Compared to a five-year baseline, rates of self-references - a marker of psychological distancing - were significantly reduced after the Vietnam War and September 11th. Cognitive processing terms were elevated post-upheaval vs. pre-upheaval, which indicated artists’ increased attempts to make meaning from such massive disruptions. Content patterns corroborated these findings as artists wrote more about “life and freedom” (societal conditions) and less about “courtship and nightlife” (interpersonal connection) following the upheavals. Cultural upheavals modify individual and collective verbal behavior, demonstrating their far-reaching impact on society.
[NLP-19] LoopSpec: Pipelined Self-Speculative Decoding for Looped Transformers
【速读】： 该论文旨在解决循环式Transformer（Looped Transformers）在推理阶段因共享权重在每一递归深度重复访问而导致的高解码延迟问题。尽管循环式Transformer通过共享参数堆栈实现了紧凑的模型规模与强大性能，但其递归结构使得解码过程串行化严重，制约了推理效率。针对这一瓶颈，论文提出一种无需训练的自生成推测解码框架——LoopSpec，其核心解决方案在于利用循环式Transformer自身早期递归状态直接生成候选预测（draft tokens），并以流水线方式重叠后续候选生成与当前候选验证过程。关键创新点包括：引入来自更深递归层的选择性二次提案机制，在不增加显著计算开销的前提下提升候选准确性；同时保证在贪婪和采样解码策略下均能实现无损解码。此外，论文推导出最优提案深度的闭式解，并验证其预测与实际测量结果高度一致。实验表明，该方法在多种推理与代码生成基准上实现了最高达6.83倍的推理加速。
链接: https://arxiv.org/abs/2609.17184
作者: SangLyul Cho,Langqing Cui,Sehoon Kim,Dongsu Han,Insu Han
机构: Seoul National University (首尔国立大学); KAIST (韩国科学技术院)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:
Abstract:Looped Transformers achieve strong performance with compact parameter sizes by repeatedly applying a shared stack of Transformer blocks across recurrent depths. However, they incur higher decoding latency than standard Transformer models of comparable parameter size because shared weights are accessed at every recurrent depth. To improve decoding efficiency, self-speculative decoding is particularly well suited to Looped Transformers, as their intermediate recurrent states can directly provide draft predictions without an auxiliary draft model. We therefore propose LoopSpec, a training-free self-speculative decoding framework tailored for Looped Transformers. LoopSpec extracts draft tokens from early recurrent states and operates in a pipelined manner, overlapping draft generation of future tokens with target verification of the current token. To improve draft accuracy without excessive compute overhead, we introduce a selective second proposal from deeper recurrent depth while ensuring lossless decoding under both greedy and sampling regimes. Furthermore, we derive the optimal proposal depths in closed form and show the prediction matches measurement. Across reasoning and coding benchmarks, LoopSpec achieves up to 6.83 \times inference speedup across diverse Looped Transformers.
[NLP-20] An Empirical Study of Counterfactual Self-Explanations in LLM s
【速读】： 该论文旨在解决大语言模型生成的自解释（self-explanations）在忠实性（faithfulness）方面存在的问题，即模型所生成的解释未必真实反映其内部决策机制。为探究此问题，研究采用反事实自解释（counterfactual self-explanations）方法，要求模型对输入进行最小化修改以使其预测结果发生变化，从而揭示影响决策的关键证据。研究覆盖情感分析与自然语言推理任务，评估了来自LLaMA-3和Qwen-2.5系列的十种指令微调模型，重点考察解释的忠实性、最小性及与人工标注理由的一致性。研究发现，模型规模是决定解释质量的最关键因素：更大规模的模型更有可能生成能够触发自身预测变化且聚焦于决策相关证据的反事实样本。相比之下，基于理由引导（rationale-guided）的条件虽能生成编辑最少、更贴近人类判断的反事实，但并未持续提升解释的忠实性。总体而言，反事实自解释可为理解模型行为提供有效依据，但其可靠性高度依赖于模型容量，应通过实证验证而非默认假设。
链接: https://arxiv.org/abs/2609.17119
作者: Giannis Kalyvas,Giorgos Filandrianos,Orfeas Menis Mastromichalakis,Vassilis Lyberatos,Giorgos Stamou
机构: National Technical University of Athens (雅典国立技术大学); Instituto de Telecomunicações (里斯本电信研究所)
类目: Computation and Language (cs.CL)
备注:
Abstract:Large language models can easily generate explanations for their own outputs, but such self-explanations are not necessarily faithful to the model’s behavior. We study this issue through counterfactual self-explanations, where a model minimally edits an input so that its own prediction changes. Across sentiment analysis and natural language inference, we evaluate ten instruction-tuned models from the LLaMA-3 and Qwen-2.5 families, measuring faithfulness, minimality, and alignment with human-annotated rationales. Our results show that model scale is the strongest determinant of explanation quality: larger models are substantially more likely to generate counterfactuals that flip their own predictions and target decision-relevant evidence. In contrast, the rationale-guided condition produces edit-minimal counterfactuals that are also more human-aligned. However, it does not consistently improve faithfulness. Overall, counterfactual self-explanations can provide useful behavioral evidence about model decisions, but their reliability depends strongly on model capacity and should be empirically validated rather than assumed.
[NLP-21] Shared-Prefix KV Reuse Across Standard LoRA Adapters: Quality and Serving Tradeoffs
【速读】： 该论文旨在解决在小模型部署中，多个低秩适配器（LoRA）共享主干网络（backbone）时，因重复预填充（prefill）相同上下文导致的推理开销过高的问题。其核心挑战在于：对于已训练好的标准LoRA适配器（非为缓存兼容性重新训练），若仅计算一次主干的预填充键值缓存（KV cache）并复用于所有适配器，会损失多少任务性能，以及由此带来的服务成本降低是否值得。研究的关键在于评估单次预填充缓存复用（full-prefix reuse）在不同上下文长度与适配器切换边界下的质量保持能力与实际服务收益。实验以Qwen3-1.7B为主干，结合HotpotQA抽取式问答与GSM8K算术推理两个适配器，在不同上下文长度下测量质量差异与延迟。结果显示，全前缀复用虽显著降低预填充成本，但在独立测试集上仍存在小幅性能下降（如在320词预算下，准确率下降约3.0 EM），且该损失在不同训练种子下不一致，未形成可推广的质量等价关系或通用边界选择规则。此外，尽管提出了一种闭形式的岭回归KV转换器（ridge KV translator），但其表现未优于直接复用；多适配器间性能对比区间均包含零点，表明无显著差异。最关键的是，虽然理论预期可实现缓存共享内存节省，但实际实现中仅复用KV值内容而未共享存储空间，导致峰值内存仅降低12%，且服务收益主要体现为热缓存场景下的首令牌延迟（time-to-first-token），其增长随上下文长度呈近似16倍的加速趋势，因此在大规模上下文场景下优势有限。
链接: https://arxiv.org/abs/2609.17109
作者: Dushyant Rajput
机构: AltSlate Labs LLP
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:A common small-model deployment runs one shared backbone with several LoRA specialists that answer over the same context. Serving them naively re-prefills that shared context once per specialist. We study a narrow, practical question: for already-trained standard LoRA adapters – not adapters retrained for cache compatibility – how much task quality is preserved if the backbone’s prefill KV cache is computed once and reused across specialists, and what does that buy in serving cost? On a Qwen3-1.7B backbone with two adapters (extractive QA on HotpotQA, arithmetic reasoning on GSM8K), we sweep the boundary at which the specialist takes over from the reused base cache and measure paired quality differences and serving cost. Full-prefix reuse had the lowest prefill cost and a small quality difference on held-out GSM8K (Delta = -4.6 EM at a 160-token budget; -3.0 at 320 tokens; -0.8 under a second training seed – all favoring native, only the first excluding zero, and the magnitude not consistent). Partial recomputation provided no demonstrated advantage. Neither quality equivalence nor a general boundary-selection rule is established. We also report a closed-form ridge KV translator that did not beat direct reuse, and specialist-dependence contrasts whose intervals all include zero. The measured serving benefit is warm-cache time-to-first-token, which grows with context (~16x at 8K); two-branch peak memory was only 12% lower and, on inspection, the prefix was never physically shared across branches – this implementation reuses KV values but copies their storage, so shared-cache memory savings are not achieved.
[NLP-22] Interactive Memory Learning for Long-Term Conversations
【速读】： 该论文旨在解决现有大语言模型在长期对话中因采用静态启发式范式而导致的记忆管理僵化问题，即信息被被动存档而缺乏自适应的价值评估能力，从而无法根据用户需求的演变实现自我优化。其解决方案的关键在于提出一种名为ICML（InteraCtive Memory Learning）的多智能体框架，将记忆机制从被动存储转变为可学习、可交互的记忆策略。该框架通过会话合成流水线生成专家数据以支持快速测试时适应，并引入在线强化学习机制：由规划者（Planner）智能体选择性编码高价值信息，触发者（Trigger）智能体动态检索相关信息以优化响应质量，二者通过持续交互反馈共同演化。尤为关键的是，通过延迟奖励机制将未来反馈回传至早期存储决策，确保记忆策略与用户期望精确对齐。实验结果表明，ICML显著优于多个强基线方法，具备随着交互积累持续提升响应质量的独特能力。
链接: https://arxiv.org/abs/2609.17088
作者: Cai Ke,Jiangyue Yan,Han Zhang,Xin Liu,Zike Yuan,Yue Yu,Hui Wang,Ruifeng Xu
机构: Harbin Institute of Technology, Shenzhen, China(哈尔滨工业大学深圳校区); Pengcheng Laboratory, China(鹏城实验室)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Recent advancements in large language models have significantly enhanced the capabilities of agents in modeling long-term conversations. Despite these successes, existing approaches typically adopt a static heuristic paradigm, where information is passively archived without adaptive memory valuation. Consequently, these methods fail to self-evolve or align their memory management with evolving user needs. To address this, we propose ICML (InteraCtive Memory Learning), a multi-agent framework that transforms the memory mechanism from a passive archive into a learnable, interactive memory policy. Specifically, we first employ a session synthesis pipeline to generate expert data, facilitating rapid test-time adaptation in unseen scenarios. Building on this, ICML utilizes an online reinforcement learning mechanism where a Planner agent selectively encodes high-value information and a Trigger agent dynamically retrieves it to optimize response quality, whereby the two agents co-evolve through continuous interaction feedback. Crucially, both agents are synchronized through a delayed reward mechanism that propagates future feedback back to earlier storage decisions, ensuring memory policies are precisely aligned with user expectations. Experimental results demonstrate that ICML significantly outperforms strong baselines, exhibiting the unique capability to continuously improve response quality as interactions accumulate.
[NLP-23] EviScope: Paired Counterfactual Evidence Diagnostics for Faithful and Efficient Grounded Language Models EMNLP2026 ACL
【速读】： 该论文旨在解决生成式语言模型在基于证据的问答任务中存在“看似正确实则缺乏支撑”的问题，即模型可能在证据不足、矛盾或被干扰的情况下仍生成看似合理的答案，而传统评估方法仅依赖最终答案准确率，无法揭示此类错误背后的推理缺陷。其解决方案的关键在于提出EviScope这一成对反事实基准测试框架，通过固定问题并系统性地引入、移除、干扰或矛盾化证据，以暴露模型在不同证据条件下的真实推理行为。EviScope-v1.1包含40组四条件数据集，具备修复后的反事实陈述及细粒度的支持片段标注，支持自动化评估。实验表明，仅靠答案准确率会掩盖模型在证据依赖性上的显著差异；而通过成对指标分析，可清晰识别出模型的“无依据回答”、“冲突盲视”以及“错误非回答行为”等深层缺陷。研究进一步发现，尽管显式证据-动作门控机制在部分本地模型上表现不佳（如Qwen 2.5-7B和Llama 3.1 8B），而Gemini 3.5 Flash虽在多数情况下表现优异（联合成功率达0.944），但仍存在约5%的矛盾案例下错误回应现象，凸显了当前模型在严格证据约束下的鲁棒性仍存根本性挑战。
链接: https://arxiv.org/abs/2609.17081
作者: Suryadeep Singh Deswal
机构: Indian Institute of Technology Roorkee(印度理工学院鲁尔基分校); Roorkee, India(鲁尔基, 印度)
类目: Computation and Language (cs.CL)
备注: Accepted as an archival short paper in GroundLM Findings at EMNLP 2026; to appear in the GroundLM 2026 workshop proceedings in the ACL Anthology
Abstract:Grounded language-model systems are often evaluated by final answer accuracy, yet a correct answer can be unsupported, drawn from the wrong source, or produced when evidence is insufficient or contradictory. We introduce EviScope, a paired counterfactual benchmark that holds the question fixed while adding, removing, distracting, or contradicting its evidence. EviScope-v1.1 contains 40 four-condition quartets with repaired counterfactual claims and span-level support labels for automatic evaluation. Across 960 gold-blind generations from Qwen2.5-7B, Llama 3.1 8B, and Gemini 3.5 Flash, paired metrics expose model-dependent grounding behavior that answer accuracy hides. On two local open models, an explicit evidence-action gate underperforms vanilla RAG on QCS: 0.15 vs. 0.50 for Qwen and 0.10 vs. 0.375 for Llama. Gemini reaches 0.944 joint success under both prompts, yet still answers 5% of conflict cases after contradiction insertion. EviScope therefore distinguishes unsupported answering, conflict blindness, and wrong non-answer actions rather than scoring answers alone.
[NLP-24] Audio-Visual Turn-taking Prediction in Cocktail Party Scenarios
【速读】： 该论文旨在解决当前预测性话轮转换模型（Predictive Turn-Taking Models, PTTMs）在存在重叠语音和背景干扰等复杂现实场景下泛化能力不足的问题。尽管现有模型在受控声学条件和清晰音频下的基准测试中表现优异，但其在嘈杂环境中的鲁棒性尚未得到充分探索。研究通过在源自AVCocktail数据集的挑战性“鸡尾酒会”测试集上评估基于纯净数据训练的音视频多模态话轮转换模型，分析了模型在新域下的适应行为。实验结果表明，在噪声条件下，音频与视觉模态均出现一致性的性能下降，加权F1分数最高下降达38%相对值。微调（fine-tuning）虽能提升模型鲁棒性，但增益在不同模态间差异显著，且依赖于预训练数据规模。该研究揭示了音频与视觉模态在泛化与适应能力上的本质差异，强调了在复杂人机交互场景中构建更具鲁棒性的建模策略的重要性。所有代码及话轮标注数据均已公开，以推动后续研究。
链接: https://arxiv.org/abs/2609.17056
作者: Long-Vu Hoang,Naomi Harte
机构: Trinity College Dublin (都柏林三一学院)
类目: ound (cs.SD); Computation and Language (cs.CL)
备注: Accepted to IEEE SLT 2026. This version includes an appendix about manual verified labels for AVCocktail
Abstract:Current predictive turn-taking models (PTTMs) achieve strong performance on benchmarks with controlled acoustic conditions and clean audio signals. Their generalisation to conversations with overlapping speech and background interference remains underexplored. In this research, we evaluate audio-visual PTTMs trained with clean data on a challenging cocktail-party testbed derived from the AVCocktail dataset, and analyse their adaptation behaviour to this new domain. Experimental results show consistent performance degradation across audio and visual modalities under noisy conditions, with up to 38% relative drop in weighted F1. Fine-tuning on the new domain improves robustness, but gains vary across modalities and depend on the size of the available pre-training data. These findings provide insights into the different generalisation and adaptation capabilities of the audio and visual modalities, and indicate the need for robust modelling strategies to adapt to the complexities of human interactions in noise. All code and turn labels are made publicly available to facilitate further research.
[NLP-25] hinkFlow: Self-Evolving Probabilistic Latent Memory for Lifelong Conversational Agents
【速读】： 该论文旨在解决长期对话智能体在持续交互过程中因依赖显式文本记忆系统而面临的严重信息瓶颈问题，该问题导致细微行为模式与情绪变化等关键语义信息丢失，且现有系统部署后通常为静态结构，无法自主适应用户习惯与偏好，需依赖人工反馈进行更新。其解决方案的关键在于提出一种端到端的潜在记忆框架ThinkFlow，通过将对话流动态压缩为概率化的潜在记忆技能，以无语义干扰的方式将复杂用户状态凝聚为解耦的连续向量表示，从而突破文本表征的局限；同时引入测试时演化（test-time evolution）范式，结合教师引导的潜在对齐与自监督的下一个用户话语预测任务，实现初始状态的快速构建与持续优化，有效克服冷启动问题，并达成无需标注标签的终身个性化。实验结果表明，ThinkFlow在长期对话基准上显著优于现有记忆系统，在多轮跨会话交互中实现了更高程度的个性化与上下文准确性。
链接: https://arxiv.org/abs/2609.17010
作者: Cai Ke,Xin Liu,Han Zhang,Jiangyue Yan,Zike Yuan,Ling Deng,Yue Yu,Hui Wang,Ruifeng Xu
机构: Pengcheng Laboratory(鹏城实验室); Harbin Institute of Technology, Shenzhen(哈尔滨工业大学深圳校区); China Unicom Greater Bay Area Innovation Institute(中国联通大湾区创新研究院)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Lifelong conversational agents rely on memory systems to maintain deep, context-aware interactions with users. However, existing explicit textual memory pipelines suffer from a severe information bottleneck, often losing subtle behavioral patterns and emotional shifts. Furthermore, being typically static post-deployment, they cannot autonomously adapt to personal habits and preferences without manual feedback. Cognitive science, however, suggests that humans maintain mental models purely in a latent space and continuously refine them through predictive coding. Inspired by this, we propose \textbfThinkFlow, a novel end-to-end latent memory framework for lifelong conversational agents. ThinkFlow bypasses the text bottleneck by dynamically compressing conversational flows into probabilistic latent memory skills, autonomously consolidating complex user states into disentangled, continuous vectors without semantic interference. To break this barrier, we introduce a test-time evolution paradigm. By coupling teacher-guided latent alignment to bootstrap the initial state with a self-supervised next-user-utterance prediction task for continuous refinement, the framework successfully overcomes cold-start challenges and achieves label-free lifelong personalization. Extensive experiments on long-term conversation benchmarks demonstrate that ThinkFlow significantly outperforms prevailing memory systems, providing highly personalized and contextually accurate responses over extended multi-session interactions.
[NLP-26] Can LLM s Follow the Pulse of a Crisis? Evaluating Crisis Sentiment in Bangladeshs July Uprising AACL
【速读】： 该论文旨在解决低资源语言（如孟加拉语）在危机情境下的情感分析难题，尤其针对语言、语境及公众反应快速演变所带来的挑战。其核心解决方案是构建并发布UNRESTSENT200K——一个包含约20万条来自2024年7月至8月孟加拉国起义期间的Facebook与YouTube评论的孟加拉语危机情感数据集。该数据集覆盖五个事件对齐阶段，涵盖早期升级、互联网断网、政权更迭及后续洪灾危机，并通过关联每条评论与其原始帖子，支持有无话语上下文的评估。所有评论均经过14名母语为孟加拉语的标注员进行全人工标注，并经资深验证，达成较高的标注一致性（Cohen’s kappa = 0.73，Scott’s pi = 0.71）和94.2%的盲审一致率。研究还对微调编码器、提示式大语言模型（LLM）及LoRA微调的LLM进行了基准测试，结果表明父帖上下文能持续提升性能，但跨阶段的时间演变导致显著性能下降；尽管大型语言模型表现良好，仍难以处理讽刺、隐含政治指涉及阶段依赖性语义等复杂现象。因此，该研究的关键贡献在于提供了一个用于研究上下文感知与时间鲁棒性情感分析的高质量基准数据集，推动低资源危机话语中的自然语言处理研究发展。
链接: https://arxiv.org/abs/2609.16997
作者: Md. Samiul Alim,Mahir Shahriar Tamim,Tanvir Ahmed Khan,Sharjil Khan,Rafia Ferdous Duti,Shahriyar Zaman Ridoy,Mohammad Ali Moni
机构: North South University(南大学), Dhaka, Bangladesh; Charles Sturt University(查尔斯·斯图尔特大学), Australia
类目: Computation and Language (cs.CL)
备注: Accepted at AACL
Abstract:Crisis sentiment analysis is especially challenging for low-resource languages such as Bangla, where language, context, and public reaction shift rapidly. We introduce UNRESTSENT200K, a Bangla crisis sentiment dataset with approximately 200K Facebook and YouTube comments from the July-August 2024 Bangladesh uprising. The dataset covers five event-aligned phases, from early escalation and internet blackout to regime transition and a later flood crisis. Each comment is linked to its parent post, enabling evaluation with and without discourse context. All comments are annotated through a fully human process involving 14 native Bangla-speaking annotators and senior validation, achieving substantial agreement (kappa = 0.73, alpha = 0.71) and 94.2% blind-audit agreement. We benchmark fine-tuned encoders, prompted LLMs, and LoRA-tuned LLMs. Results show that parent-post context consistently improves performance, while temporal shift across phases causes large performance drops. Strong LLMs perform well, but still struggle with sarcasm, implicit political references, and phase-dependent meaning. UNRESTSENT200K provides a benchmark for studying context-aware and temporally robust sentiment analysis in low-resource crisis discourse. UNRESTSENT200K is available at this https URL
[NLP-27] he Role of Implicit and Explicit Demographic Signals in Large Language Model-based Student Assessment EMNLP2026
【速读】： 该论文旨在解决生成式 AI（Generative AI）在学生评估中因学生人口统计学特征（如教育水平、社会经济背景等）而引发的潜在偏见问题。其核心挑战在于：尽管考虑人口统计学因素可能有助于提升评估内容对低教育水平用户的可读性，但若模型据此调整评分或反馈，则可能造成系统性歧视。论文的关键解决方案是通过设计受控提示（controlled prompts），分别测试显式（explicit）与隐式（implicit）的人口统计学信号对大语言模型（LLM）行为的影响。研究发现，在显式提及教育水平时，模型会显著调整反馈的可读性；而在隐式情境下，模型则表现出不可预测的偏差，例如对低教育水平群体的回答给予更低的情感评分。这表明，当前主流大语言模型在教育评估任务中对人口统计学特征具有高度敏感性，亟需建立公平性保障机制以防止算法偏见。
链接: https://arxiv.org/abs/2609.16993
作者: Donya Rooein,Luca Benedetto,Dirk Hovy
机构: Bocconi University(博科尼大学); Télécom SudParis, Institut Polytechnique de Paris(巴黎高等电信学院)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注: EMNLP 2026 Findings
Abstract:Large Language Models are now common in student assessment, but we know little about how student demographics affect their use. Sometimes, considering student demographics may be necessary – for example, to improve readability for users with lower educational levels. However, it also risks being a cause of discrimination, e.g., when assigning lower scores to students from lower socioeconomic backgrounds. We set up controlled prompts to test 1) explicit demographic effects, where we mention demographic details directly, and 2) implicit effects, where we use conversation history as a demographic signal. We test these settings in three tasks: Automated Essay Scoring, Formative Feedback, and Metalinguistic Question Answering. We test six state-of-the-art LLMs on these tasks. In both explicit and implicit cases, the models pick up on demographic cues and can change their scoring, feedback, and answers accordingly. We find that LLMs frequently adjust the readability of feedback to education levels when these are explicitly mentioned. On the other hand, implicit conditions produce unpredictable biases, such as in question answering, where responses from lower-education levels receive lower sentiment scores. Our results provide clear evidence of demographic sensitivity in LLMs for educational assessment tasks.
[NLP-28] Autoformalizing Argumentative Material Inferences
【速读】： 该论文旨在解决自然语言论证中隐含的非单调支持关系（defeasible support）在形式化验证过程中难以机器可检查的问题。其核心挑战在于，自然语言论证依赖于未明言的前提、背景承诺及例外条件，而形式化验证要求所有推理依据必须显式表达；然而，若允许系统自由添加前提，则可能引入虚假可证明性（即“泄漏”），导致形式有效但语义失真。为此，论文提出将论辩材料推理的自动形式化（autoformalization）重构为“守卫闭合”（guard completion）机制，即将非单调的论证支持转化为相对于显式构建的守卫集（guard set）的单调形式推导。关键在于：一个闭合方案仅在满足双重检验时被接受——其证明需通过形式化定理证明器验证，并且在对比测试中表现出对前提的依赖性和对结论的选择性。作者实现该框架为GUARD，一个神经符号系统，利用大语言模型（LLM）生成候选守卫并进行形式化，Isabelle/HOL负责验证理论并提供逐步反馈以支持迭代优化，当无法达成忠实闭合时系统主动放弃。实验在Debatepedia和ARCT数据集上表明，相较于现有基于LLM的定理证明方法，GUARD在验证忠实性（verified-faithful）上分别提升35.3和32.9分，同时显著降低泄漏（leakage）25.9和21.9分。进一步分析显示，符号化软批判（symbolic soft critique）与显式假设层是性能提升的主要来源，其中软批判还增强了初始上下文的有效性并减少了成功验证所需的迭代次数。
链接: https://arxiv.org/abs/2609.16991
作者: Xin Quan,Reto Gubelmann,André Freitas
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Natural language arguments are compelling before they are formally explicit. A premise supports a claim through defeasible warrants, background commitments, and exception conditions that the text leaves implicit. However, formal verification requires the opposite. Making such arguments machine-checkable requires constructing the missing commitments, not only translating given sentences into logic. Construction, however, carries a risk that translation does not: a system free to add premises can make any claim provable, and a formally valid proof may assert the claim outright, prove it without the original premise, or establish more than the claim itself. We address this problem by formulating autoformalization for argumentative material inference as guard completion, in which non-monotonic material support is turned into monotonic formal inference relative to an explicitly constructed guard set. A completion is accepted only when its proof both passes the theorem prover and survives contrastive tests of premise dependence and claim selectivity. We implement this formulation in GUARD, a neuro-symbolic framework in which LLMs construct and formalize candidate guards, Isabelle/HOL verifies the resulting theories and returns step-level feedback for iterative refinement, and the system abstains when no faithful completion can be reached. Our empirical results on Debatepedia and ARCT using different LLMs demonstrate that GUARD yields significant improvements in verified-faithful (+35.3, +32.9 points) and substantial reductions in leakage (-25.9, -21.9 points) over the state-of-the-art LLM-driven theorem proving approach. Moreover, we show that the symbolic soft critique and the explicit assumption layer account for most of these gains, with the soft critique also improving the initial validity of the elicited context and reducing the number of iterations required for successful verification.
[NLP-29] Nameless Tokenization: A Lossless Tokenizer-Level Defense Against Control-Token Forgery in Open-Weight LLM s
【速读】： 该论文旨在解决开放权重语言模型（Open-weight language models）中因聊天模板标记（chat templates）的字符串可被伪造而导致的安全漏洞问题。具体而言，当前主流的聊天令牌化（tokenization）机制将角色、对话轮次及工具结果等控制标记以可见字符串形式暴露，攻击者可通过操控提示词内容生成与系统生成的标记在表面上无法区分的伪造标记，从而实施欺骗性输入攻击。针对此问题，研究提出“无名令牌化”（nameless tokenization）作为核心解决方案：将控制标记的标识符（identifier）保留在模型内部，但移除其对应的表面字符串，使内容编码器无法生成此类标记，确保消息内容未经篡改地传递至模型。实验表明，该方法在五种不同令牌化家族中均能精确复现正常数据的令牌流，并将含分隔符文本探测任务的准确率从8.5%提升至59.9%，显著优于传统净化器（sanitizers）方案。研究进一步揭示，分隔符的标识符本身对任务指令的影响较小，但在系统消息引导模型将用户输入视为数据时，其对伪造工具结果和对话轮次的可信度具有决定性作用，凸显了标识符在安全上下文中的关键意义。
链接: https://arxiv.org/abs/2609.16984
作者: Kisu Yang,Yoonna Jang,Heuiseok Lim
机构: Korea University(韩国大学); Hanwha Aerospace(韩华航空航天); VAIV Company
类目: Computation and Language (cs.CL)
备注: preprint
Abstract:Open-weight language models publish the strings their chat templates use to mark turns, roles and tool results, which the tokenizer maps back to the reserved identifiers the model obeys. Anyone who controls text in a prompt can therefore write a turn boundary indistinguishable from one the serving stack wrote. We audit 256 deployed chat tokenizers. All are forgeable, and the flag usually recommended as a fix leaves 56.6% forgeable because it misses the tool and reasoning markers agent systems rely on. We propose nameless tokenization, which leaves the control entries with a reserved identifier and no surface string, so the content encoder cannot emit one and message content reaches the model unaltered. Across five tokenizer families it reproduces the standard token stream exactly on attack-free data and lifts accuracy on a probe of delimiter-bearing text from 8.5% to 59.9%, where sanitizers lose it. Separating a delimiter’s appearance from its identifier shows the identifier matters little against a bare task instruction, but carries most of a forged tool result and most of any forged turn once the system message tells the model to treat user content as data.
[NLP-30] arget-Language Generation in Multilingual Models: Activation Steering and Optimal Control EMNLP2026
【速读】： 该论文旨在解决多语言语言模型在生成目标语言文本时难以保证语言一致性、语法连贯性与语义连贯性的问题。其解决方案的关键在于提出一种基于最优控制（optimal control）的目标语言文本生成方法，结合一个涵盖语言依从性、语言连贯性与语义连贯性的评估框架，有效提升了生成文本的质量。相较于主流的差异均值激活引导（difference-in-means activation steering）方法，该方法在多数测试模型上表现相当甚至更优，且显著减少了超参数调优的需求，具备更高的实用性和稳定性。
链接: https://arxiv.org/abs/2609.16967
作者: James A. Michaelov,Carmen Amo Alonso,Tyler A. Chang,Roger P. Levy
机构: Massachusetts Institute of Technology (麻省理工学院); Stanford University (斯坦福大学); University of California San Diego (加州大学圣地亚哥分校)
类目: Computation and Language (cs.CL)
备注: Accepted at EMNLP 2026
Abstract:Ensuring that multilingual language models generate coherent text in a specific target language is a major issue in multilingual language modeling. We develop an optimal control method for target-language text generation as well as a framework for evaluating the quality of generated text in terms of language adherence, linguistic coherence, and semantic coherence. We find that the proposed method performs at least as well as the prominent difference-in-means activation steering method for the majority of models tested, with substantially less hyperparameter tuning required.
[NLP-31] HUMAID-NER: A Disaster Tweet Dataset for Joint Named Entity Recognition and Event Classification via Uncertainty-Weighted Multitask Learning
【速读】： 该论文旨在解决灾难响应中从社交媒体（如推文）快速提取结构化信息的难题，尤其针对现有灾难推文数据集缺乏细粒度实体标注的问题。现有资源多仅提供文档级别的分类标签，而无法支持精确的实体边界识别，限制了生成式AI在灾害应急中的应用效果。为此，研究提出HUMAID-NER，首个基于HumAID基准构建的命名实体识别（Named Entity Recognition, NER）数据集，包含6万条英文灾难推文，覆盖十类与人道主义行动相关的实体类型，共标注约17.5万个实体跨度，并以BIO格式进行标注。其关键解决方案在于设计了一个可复现的三阶段混合标注管道：结合spaCy Transformer模型、灾难领域专用的EntityRuler规则模式以及结构化正则表达式，并采用基于优先级的重叠实体冲突解决机制，有效提升标注一致性与准确性。同时，提出一种联合多任务学习框架，通过共享的RoBERTa-large编码器同步执行灾难特定的命名实体识别与人道主义事件分类任务；为缓解任务间冲突，引入同方差不确定性加权（homoscedastic uncertainty weighting）机制，结合可学习的任务权重参数，并采用两阶段训练策略——第二阶段冻结编码器底层18层（共24层），以稳定特征表示。实验结果表明，在HUMAID-NER验证集上，系统实现实体识别的跨度微平均F1值达0.841，事件分类的宏平均F1值为0.761。研究还开发了一个实时网络仪表板，展示端到端部署能力。所提出的数据集、模型及管道代码均已公开，显著促进危机信息学领域的可复现性与后续研究发展。
链接: https://arxiv.org/abs/2609.16964
作者: Aijaz Ali,Nazish Basir,Sarfaraz Nawaz,Danish Nazir Arain,Haris Ali
机构: University of Sindh (信德大学); Mehran University of Engineering Technology (梅尔汗工程与技术大学)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 8 pages, 8 figures, 4 tables. Published in The Asian Bulletin of Big Data Management, Vol. 6, No. 1, pp. 138-152, 2026
Abstract:Rapid extraction of structured information from social media is important for humanitarian response, yet existing disaster tweet resources mainly provide document-level category labels without span-level entity annotations. We introduce HUMAID-NER, the first named entity recognition dataset built on the HumAID benchmark, containing 60,000 English disaster tweets annotated in BIO format across ten operationally motivated entity types and yielding approximately 175,000 labelled entity spans. Annotations are generated through a reproducible three-stage hybrid pipeline combining a spaCy transformer model, disaster-domain EntityRuler patterns, and structured regular expressions with priority-based overlap resolution. We also propose a joint multitask learning framework that performs disaster-specific named entity recognition and humanitarian event classification using a shared RoBERTa-large encoder. To reduce task conflict during joint training, the model uses homoscedastic uncertainty weighting with learnable task parameters and a two-stage training schedule that freezes the lower 18 of 24 encoder layers in the second stage. On the HUMAID-NER validation set, the proposed system achieves NER span micro-F1 of 0.841 and classification macro-F1 of 0.761 simultaneously. A real-time web dashboard demonstrates end-to-end deployment. The dataset, models, and pipeline code are released to support reproducibility and future crisis informatics research.
[NLP-32] Verbalizing Subliminal Learning Effects Using Text Optimization
【速读】： 该论文旨在解决生成式模型中隐性学习（subliminal learning）所带来的安全与可解释性问题，即教师模型（teacher model）的特定属性通过蒸馏数据以非显式编码的方式传递给学生模型，从而在不直接暴露特征的情况下引发潜在风险，尤其是在数据投毒攻击场景下。其核心挑战在于如何从看似无害的训练数据中识别出这些隐蔽的、由教师模型诱导的隐性模式。本文的关键解决方案是提出一种名为SALVE（Search-Aided Latent Verbalization）的方法，将恢复教师模型所隐含提示的任务转化为一个文本优化问题：通过优化一个软提示（soft prompt），利用束搜索（beam search）引导模型将其“语义化”为可读文本，并据此反推教师模型的原始提示或其所编码的特质。该方法在理论上基于上下文蒸馏（context distillation）的特殊形式，证明了隐性学习数据集在理想情况下可唯一标识教师提示。实验表明，SALVE在标准隐性学习设置中能可靠地恢复出可读提示，而传统文本优化方法则失效；更进一步，SALVE甚至可在隐性学习未发生时仍检测到教师特质，且通过调整学生模型训练策略可诱发隐性学习现象。此外，该方法在混合数据、激活控制偏见下的生成数据以及基于对数线性选择的真实偏好数据子集等复杂场景中均展现出鲁棒的检测能力。因此，SALVE不仅深化了对隐性学习机制的理解，也为主动探测和防范此类隐蔽知识迁移提供了有效工具。
链接: https://arxiv.org/abs/2609.16927
作者: Nathan Hu,Sanmi Koyejo,Christopher Potts
机构: Stanford University (斯坦福大学)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Subliminal learning is a phenomenon in which a distillation dataset transmits traits from the teacher model that are not legibly encoded in the dataset itself. This introduces a new challenge for model development and creates new risks from data poisoning. In this work, we use text optimization to detect subliminal learning effects and describe them as legible prompts. Subliminal learning from a prompted teacher motivates our approach. We observe that this is a special case of context distillation and leverage this observation to show that, in theory, the prompted subliminal learning dataset identifies the teacher’s prompt. We reduce recovering this prompt to a text optimization problem and present a method to approximately solve it. Our method, SALVE (Search-Aided Latent Verbalization), optimizes a soft prompt, queries the same model to verbalize it as text, and uses beam search to make the verbalization reliable. In the standard subliminal learning setting, SALVE reliably recovers legible prompts that name the teacher’s trait, while common text optimization methods fail to do so. In addition, we find that there are settings in which SALVE recovers the teacher’s trait from a dataset even when subliminal learning fails, but that modifying student training to improve context distillation can create subliminal learning effects. We lastly show that SALVE detects subliminal learning effects in three additional settings: (1) mixtures of subliminal learning data and unrelated data, (2) data generated when the teacher is biased via activation steering, and (3) subsets of real preference data selected via Logit-Linear Selection. Overall, our results deepen our understanding of subliminal learning and present SALVE as a method to proactively detect subliminal learning effects.
[NLP-33] Lit3R: Retrieve-Relate-Read for Evidence-Grounded Question Answering over Scientific Literature EMNLP2026
【速读】： 该论文旨在解决文献基础问答（literature-grounded question answering）中的多阶段复杂任务，即从海量学术文献中检索相关论文、识别支持性证据，并生成准确答案。其核心挑战在于如何在不进行特定任务微调的前提下，高效整合多种异构组件以实现端到端的推理能力。解决方案的关键在于构建一个无需任务特定训练的模块化系统——Lit3R（Retrieve-Relate-Read），该系统通过迭代式融合基于BM25的稀疏检索与稠密检索、交叉编码器重排序以及大语言模型（Large Language Model, LLM）验证机制，增强初始检索效果；同时引入基于问题的论文间扩展策略，提升相关文献的覆盖范围。在阅读阶段，系统先在单篇论文内定位支持性证据，再跨文档整合证据以生成最终答案及证据链。该方法在官方测试集上取得了第四名的成绩，验证了其在无监督场景下的有效性与鲁棒性。
链接: https://arxiv.org/abs/2609.16912
作者: Akira Ise,Kotaro Kumagai,Yuta Yamaguchi,Hisanori Ozaki,Yukio Uematsu,Ikuya Yamada
机构: Tokyo University of Science(东京科学大学); Dentsu Soken Inc.(电通三井公司); Studio Ousia
类目: Computation and Language (cs.CL)
备注: Accepted at GroundLM 2026, an EMNLP 2026 Workshop LittraceQA
Abstract:We describe tus-nlp’s Lit3R (Retrieve-Relate-Read) system for LitTraceQA, a shared task for literature-grounded question answering that requires systems to retrieve relevant papers, identify supporting evidence, and generate answers. Lit3R combines off-the-shelf retrieval, reranking, and large language model (LLM) components without task-specific training. The retriever iteratively combines BM25-based sparse and dense retrieval, cross-encoder reranking, and LLM-based verification, and complements retrieval based on the question with paper-to-paper expansion. The reader first identifies supporting evidence within individual papers and then synthesizes evidence across papers to produce the final answer and evidence trace. On the official test set, our system ranked 4th on the leaderboard. Our code is available at this https URL.
[NLP-34] Deconstructing Stereotypes: Scope-Conditioned Generation for Effective Multilingual Counterspeech
【速读】： 该论文旨在解决当前自动生成反仇恨言论（Counterspeech, CS）方法中存在的核心问题：生成的回应往往过于泛化、缺乏针对性，无法有效识别并反驳网络仇恨言论（Hate Speech, HS）背后的隐含刻板印象。现有方法在处理隐性偏见时表现不足，导致生成内容在事实性、具体性、逻辑说服力和实际有效性方面均存在明显缺陷。其解决方案的关键在于提出一种新型的“范围条件化生成框架”（scope-conditioned generation framework），通过在大型语言模型（Large Language Models, LLMs）的提示中显式引入结构化的刻板印象特征（structured stereotype characteristics），从而引导模型生成更具针对性、基于推理且能有效驳斥隐含偏见的反言论。研究在新构建的多语言（英语、意大利语、西班牙语）人工标注数据集上进行了验证，实验结果表明，该方法显著优于通用基线，在三类语言中均在事实性、具体性、逻辑性和有效性等方面取得显著提升，尤其对显性和隐性刻板印象均展现出更强的应对能力。
链接: https://arxiv.org/abs/2609.16906
作者: Greta Damo,Elias Urios Alacreu,Elena Cabrio,Paolo Rosso,Serena Villata
机构: Université Côte d’Azur, CNRS, INRIA, I3S, France; PRHLT Research Center, Universitat Politècnica de València, Valencia, Spain; ValgrAI Valencian Graduate School and Research Network of Artificial Intelligence, Spain
类目: Computation and Language (cs.CL)
备注:
Abstract:Counterspeech (CS) - direct responses that counter online Hate Speech (HS) using reasoning and alternative viewpoints - has emerged as an alternative to content removal. Current automatic CS generation methods, however, frequently produce generic, ineffective replies that fail to target the implicit stereotypes behind HS. To bridge this gap, we propose a novel scope-conditioned generation framework that explicitly integrates structured stereotype characteristics into Large Language Models prompts. We validate our approach on a novel, human-curated dataset annotated in English, Italian, and Spanish. Extensive evaluations show that stereotype-conditioned prompting substantially outperforms generic baselines across all three languages, obtaining significant gains in factuality, specificity, cogency, and effectiveness for both explicit and implicit implied stereotypes.
[NLP-35] RiskChainBench: A Benchmark for Obfuscated Platform Message Restoration and Evidence-Grounded Web Investigation
【速读】： 该论文旨在解决平台滥用行为中隐蔽的恶意链接诱导问题，即攻击者利用表情符号、同音词、字符拆解及冗余符号等手段对文本进行混淆，进而通过伪装链接将用户引导至涉及色情、欺诈、赌博或非法交易的高风险网站。现有评估基准通常将混淆文本识别与高风险网页检测分开评价，未能揭示目标恢复过程对下游证据获取的影响。为此，论文提出RiskChainBench，构建了一个包含3,600个合成令牌-文本还原输入（来自600个源会话）与对应600个经人工标注的本地网页环境的配对数据集。其核心解决方案是采用统一模型分两阶段执行：首先基于输入文本恢复消息内容、操作意图和真实目的地；随后同一模型作为视觉-语言模型（VLM）驱动的网页代理，在无消息语义与域名信誉提示的情况下，自主探索并生成附有证据引用的风险报告。评估体系分别独立评分文本还原与正确路由下的网页调查结果，并通过冻结的主入口预测作为门控机制组合二者结果。人类标注判定任务正确性，固定多模态证据判别器则评估报告的忠实性、充分性、完整性和一致性。实验表明，十种模型在入口命中率（Entry Top-1）上差异显著（35.2%–95.2%），网页决策准确率范围为26.3%–62.8%，且领先系统在入口恢复、完整重构、网站判断及细粒度分类任务上表现各异。其中，执行失败占网页运行的31.9%，而决策后类型错误仅占0.9%，揭示出稳定探索与风险判断是当前主要瓶颈。研究同时开源了基准数据集、评估协议及可重置的本地沙箱环境。
链接: https://arxiv.org/abs/2609.16900
作者: ZhuoXin Liu,Zhiming Ma,Ying Zhang,Mengzheng Yang,Yifan Wang,Zhengqi Huang,Yanhan Zhou,Zekun Lin,Jun Zhang,Shun Zhang,Yue Chen,Qiao Zhao,Peng Chen
机构: 未知
类目: Computation and Language (cs.CL)
备注: 11 pages, 5 figures; 17-page supplementary material included as an ancillary PDF
Abstract:Platform abuse campaigns conceal redirection instructions with emojis, homophones, character decomposition, and redundant symbols, then route users through disguised links to services associated with pornography, fraud, gambling, or illicit transactions. Existing benchmarks evaluate obfuscated text and risky webpages separately, obscuring how target recovery affects downstream evidence acquisition. We introduce RiskChainBench, pairing 3,600 synthetic token-text restoration inputs from 600 source sessions with 600 corresponding human-labeled local web environments. A model first restores the message, operational intent, and destination; the same underlying model then acts as a VLM-driven web agent that investigates the correctly associated website and produces a frozen, evidence-cited risk report without message-side semantics or domain-reputation cues. We score restoration and correct-routing web investigation separately and compose them offline by applying the frozen primary-entry prediction as a gate to the same Task 2 result. Human labels determine task correctness, while a fixed multimodal evidence judge assesses faithfulness, sufficiency, completeness, and consistency. Across ten models, Entry Top-1 ranges from 35.2% to 95.2% and web decision accuracy from 26.3% to 62.8%; the leading systems differ across entry recovery, full reconstruction, website decisions, and fine-grained typing. Execution failures account for 31.9% of web runs, whereas post-decision type errors account for only 0.9%, identifying stable exploration and risk judgment as the principal bottlenecks. We release the benchmark, protocol, and resettable local sandbox.
[NLP-36] Cascade: Hierarchical Recoverability Control for Large Language Model Unlearning EMNLP2026
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）在去除敏感或受版权保护知识时，仍存在残余知识泄露的问题，尤其关注这些知识在中间表示中的可恢复性。现有方法往往无法彻底消除模型内部对目标知识的隐式记忆，导致隐私或版权风险持续存在。为此，论文提出一种分层可恢复性控制框架——Cascade，其核心解决方案在于通过三重互补控制机制协同作用：路径级路由（path-level routing）以抑制与隐私相关激活路径的传播，表示级压缩（representation-level compression）降低知识在嵌入空间中的几何可分离性，解码级干预（decoding-level intervention）则进一步限制残留知识的外部提取可能性。实验在TOFU、MUSE-News和WMDP等多个数据集上验证了该方法的有效性，即使在查询重构和提取型提示等强攻击场景下，仍能显著降低知识可恢复性，同时保持模型整体性能稳定。
链接: https://arxiv.org/abs/2609.16890
作者: Qingchen Yu,Shiying Duan,Xiaodong Li,Yuhua Wang,Zhiyu Li,Shiji Zhou,Yifan Sun,Zhaoxin Fan
机构: Beihang University (北京航空航天大学); Renmin University of China (中国人民大学); MemTensor (上海)科技有限公司
类目: Computation and Language (cs.CL)
备注: Accepted by EMNLP 2026 (Findings)
Abstract:Large Language Model (LLM) unlearning is essential for removing sensitive or copyrighted knowledge while preserving general utility. Existing methods often leave residual knowledge in intermediate representations, which can still be recovered. To address this, we propose Cascade, a hierarchical recoverability control framework that minimizes the internal identifiability of target knowledge. Cascade combines three complementary controls: path-level routing to suppress privacy-associated activation routes, representation-level compression to reduce geometric separability, and decoding-level intervention to limit residual recovery. Experiments on TOFU, MUSE-News, and WMDP, including robustness tests with query reformulation and extraction-style prompts, show that Cascade effectively reduces recoverability while maintaining stable model utility.
[NLP-37] Reduplicative constructions in Mandarin: Socio-emotional profiling through distributional semantics
【速读】： 该论文旨在解决汉语中重叠构词（reduplication）语义多样性难以精确刻画的问题，尤其针对普通话中两种典型的双字词重叠结构（重复整个词或其成分）所表现出的多重语义功能——包括数量性、句法配价色彩、拟声意义及语用功能等。其解决方案的关键在于引入分布语义学中的词嵌入（word embeddings），通过高维向量空间分析揭示重叠结构的语义与语法属性。研究发现，腾讯词嵌入模型能够有效恢复文献中已知的重叠结构语义特征，并在分布空间中清晰区分两种重叠模式的语义与语用差异；同时，通过普罗克斯特斯分析（Procrustes analysis）表明，基词空间的整体组织结构在重叠形式空间中基本保留，局部偏差则凸显了话语-语用层面的重构区域。结果表明，高维词嵌入不仅可复现既有的语言学规律，还能捕捉汉语重叠构词的高度语义灵活性与构式透明性，为复杂形态生成过程的细粒度解析提供了数据驱动的新路径。
链接: https://arxiv.org/abs/2609.16860
作者: Chaoyi Wu,Yu-Hsiang Tseng,R. Harald Baayen
机构: Beihang University(北京航空航天大学); University of Tübingen(图宾根大学)
类目: Computation and Language (cs.CL)
备注: 32 pages, 9 figures
Abstract:Mandarin Chinese has two productive reduplicative constructions that repeat either two-character base words or their constituents (e.g., in good health', discuss a bit’). Their varied meanings have been described as realizing plurality, valence coloring, sound symbolism and pragmatic functions. The aim of this study is twofold. A first goal is to clarify whether it is possible to come to a more precise understanding of the variegated semantics of Mandarin reduplication by using word embeddings from distributional semantics. A second goal is to explore how useful embeddings are for understanding the details of a semantically complex word-formation process. We show that the embedding space recovers the semantic and grammatical properties of reduplications previously identified in the literature, validating Tencent embeddings for morphological investigation. Semantic profiling revealed that reduplicative constructions are often strongly represented on multiple dimensions. The two patterns exhibit clear semantic and pragmatic differentiation in distributional space. Procrustes analysis clarified that the overall organization of the base-word space is largely preserved in the reduplication space, with local mismatches highlighting regions of discourse-pragmatic reorganization. Taken together, these results show that high-dimensional word embeddings can recover established linguistic generalizations, and capture the semantic versatility of Mandarin reduplication and constructional transparency.
[NLP-38] A Data-free Universal Prior over Syntactic Structures
【速读】： 该论文旨在解决语言语法结构的概率分布是否完全依赖于语言特定经验（如统计学习）的问题。现有理论普遍基于语言数据估计句法结构的概率，但无法确定这种概率结构是否可独立于具体语言经验而形成。本文提出的关键解决方案是：通过一个认知上合理的增量语言生成模型，模拟词汇在动态网络增长过程中逐步整合为句法结构的过程，由此自发产生一种普适的句法结构先验（universal prior）。该先验将句法结构表示为依存树（dependency trees），无需对语言数据进行参数拟合，却能在138种语言类型多样化的语言中对已记录的句法树赋予更高概率，且与34种语言中基于语料库估计的概率呈正相关（33种语言显著相关）。这一发现表明，部分句法概率结构可源于语言生成过程本身，而非完全由语言经验塑造。因此，语言经验可能仅用于微调已由生成机制预先设定的概率结构，而非从无偏空间中构建概率分布。该研究揭示了句法概率分布的部分认知起源，连接了语言生成与统计学习，并为语言的概率模型提供了数据无关的结构偏差。
链接: https://arxiv.org/abs/2609.16854
作者: Ferm’ın Moscoso del Prado Mart’ın
机构: 未知
类目: Computation and Language (cs.CL); Disordered Systems and Neural Networks (cond-mat.dis-nn)
备注: 30 pages, 4 figures
Abstract:Probability is fundamental to theories of language comprehension, production, acquisition, and evolution, as well as to large language models. Existing theories estimate the probability of syntactic structures from language-specific data. Whether part of this probability structure can arise independently of language-specific experience remains unknown. Here I show that a universal prior over syntactic structures emerges from a cognitively motivated model of incremental language production, in which words are progressively integrated into syntactic structure through network growth. The resulting prior assigns probabilities to syntactic structures --represented as dependency trees-- without fitting parameters to linguistic data, and assigns higher probabilities to attested than to random trees in all 138 typologically diverse languages examined. These prior probabilities correlate positively with probabilities estimated from corpora in 33 of 34 languages. The results indicate that part of the probability structure of syntax can arise independently of language-specific statistical learning. Linguistic experience may therefore refine probabilities that are already structured by the process of language production, rather than create them from an initially uniform space. This identifies a possible cognitive origin for part of the probability distribution over syntactic structures, linking language production and statistical learning while providing a data-independent structural bias for probabilistic models of language.
[NLP-39] ImpossibleRubrics: Stress-Testing Generated Rubrics as Reward Signals
【速读】： 该论文旨在解决生成式AI（Generative AI）在构建评分标准（rubrics）时存在的脆弱性问题，即这些由语言模型生成的评分标准在面对针对其设计漏洞进行优化的对抗性回答时，可能错误地奖励非诚实、虚构的答案，而非真实反映任务限制的诚实回应。其核心挑战在于：如何确保评分标准具备对对抗性攻击的鲁棒性，从而真正区分诚实回答与通过策略性伪造以迎合评分规则的回答。解决方案的关键在于提出ImpossibleRubrics这一基准测试体系，包含169个“不可能任务”（impossible tasks），涵盖六类不可行情境，并为每项任务提供可验证的“证书”（oracle certificate），明确界定诚实回答的边界。该框架不预设固定评分标准，而是允许下游生成评分标准后，再通过对抗性测试检验其是否奖励违反证书内容的回答。实验表明，尽管部分生成器在特定压力子集上仍被利用高达36%的情况，但符合证书忠实性的评分标准则完全未被利用，凸显了评分质量差距（rubric-quality gap）的存在，而非任务本身固有的不可解性。更出人意料的是，通用评分标准（如“保持决断，惩罚模棱两可”）反而被64%的攻击所利用，且定制化评分标准反而使攻击者更易识别可伪造的命题，揭示了当前评分标准的问题并非模糊，而是对错误维度过于具体——即评分标准的“精准性”若偏离真实可信性要求，反而会成为攻击的指引。
链接: https://arxiv.org/abs/2609.16816
作者: Bowen Qin,Yi Xie,Yesheng Liu,Xi Yang
机构: National University of Singapore(新加坡国立大学); Peking University(北京大学); Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所); JD.com(京东)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:
Abstract:Language model-generated rubrics are increasingly used as reward signals for rubric-based reinforcement learning, LLM-as-a-judge evaluation, and automated grading. Such rubrics are reliable only if they reward honest answers over adversarial answers optimized to exploit them. Yet their robustness to such optimization remains poorly understood. We isolate the hardest regime: impossible tasks, where the prompt pressures the model toward an unsupported conclusion, so the only honest response is to acknowledge the impossibility. We introduce ImpossibleRubrics, a benchmark of 169 impossible tasks spanning six impossibility categories, each paired with a verifiable oracle certificate specifying what an honest answer may and may not claim, together with 48 answerable controls. Rather than providing fixed rubrics, ImpossibleRubrics provides task environments and certificates, allowing rubrics to be generated downstream and then adversarially tested for whether they reward certificate-violating answers. Eleven generators are exploited 8–26% of the time on the unbiased 150-of-169 environment cut; on a deliberately selected stress cut the strongest generator we measured is still exploited 36% while a certificate-faithful rubric is exploited 0%, so what we measure is a rubric-quality gap, not task impossibility. One result runs against intuition. A single generic rubric (“be decisive, penalize hedging”) used unchanged for every task is exploited 64% of the time, and seven of the eleven generators are exploited more often than that while writing a rubric tailored to each one. The tailored criteria appear to tell an attacker which claim to fabricate. The problem is not that rubrics are vague; it is that they are specific about the wrong things.
[NLP-40] Smarter by the Moment: Environment-Driven Dynamic Policies for Continual LLM Improvement
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在持续适应动态任务与环境过程中，因缺乏对历史经验的策略性提炼而导致错误重复发生的问题。现有基于记忆的增强方法仅通过检索过往实例作为直接参考，未能显式地从历史数据中生成可执行的优化策略，从而限制了模型的持续改进能力。其解决方案的关键在于提出一种动态检索驱动的策略生成框架（Dynamic Retrieval-based Policy Generation, DRPG），该框架将记忆检索机制与动态策略生成器相结合，利用历史数据和环境反馈自动生成针对特定任务的优化策略，实现对LLM的持续性指导与提升。实验表明，DRPG在文本到SQL、问答、医学诊断及Python编程等六个基准任务上均显著优于主流基线方法，且对检索策略具有鲁棒性，无需依赖先验策略连续性，并可有效利用小型或跨家族模型作为低成本策略生成器，为策略级指导的有效性提供了基于任务特性的实践启示。
链接: https://arxiv.org/abs/2609.16800
作者: Ting-Wei Chang,Po-Chun Chen,Hen-Hsen Huang,Hsin-Hsi Chen
机构: National Taiwan University(台湾大学); Institute of Information Science, Academia Sinica(中央研究院资讯科学研究所); AI Research Center (AINTU), National Taiwan University(台湾大学人工智能研究中心)
类目: Computation and Language (cs.CL)
备注: 25 pages, 13 figures. Accepted to the Conference on Language Modeling (COLM) 2026
Abstract:Large Language Models (LLMs) have achieved remarkable progress across diverse domains, but continual adaptation to evolving tasks and environments remains a key challenge. Existing memory-augmented approaches retrieve individual past examples as direct references, but do not explicitly synthesize actionable strategies from them, causing the same types of errors to recur. We propose Dynamic Retrieval-based Policy Generation (DRPG), a framework that integrates memory-based retrieval with a dynamic policy generator, leveraging historical data and environment feedback to produce task-specific policies for continual LLM improvement. We evaluate DRPG across six benchmarks spanning text-to-SQL, question answering, medical diagnosis, and Python programming, using seven LLMs from both proprietary and open-weight families. DRPG outperforms strong baselines across most datasets and models. Further analysis demonstrates that DRPG’s policy generation is robust to retrieval strategy, operates effectively without prior policy continuity, and can leverage smaller or cross-family models as cost-efficient policy generators. We also find that the benefit of policy-level guidance depends on task characteristics, offering practical insights into when and under what conditions this mechanism is most effective.
[NLP-41] Benchmarking Factual Robustness of LLM s via Multi-conversation Persuasion
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在作为主要知识检索接口时，面对说服性攻击（persuasion attacks）所表现出的鲁棒性不足问题，即攻击者通过诱导注入错误信息或强制生成反事实内容，从而破坏模型输出的准确性与可靠性。现有红队测试框架通常在多轮对话中评估模型，且目标模型保留完整的对话历史，这导致一个关键缺陷——“拒绝惯性”（Refusal Inertia）：模型因维持上下文一致性而持续拒绝，掩盖了其在面对独立、复杂说服策略时的真实脆弱性。为更严格地评估当前最优模型（SOTA）在“冷启动”场景下的防御能力，论文提出SAST-IR（Stateful Attacker, Stateless Target - Iterative Refinement）框架，通过在保持攻击者历史的同时对目标模型执行记忆清除，模拟最恶劣的对抗环境。在此设定下，结合认知说服代理（CP-Agent）进行诊断引导的迭代攻击，实验基于自定义的CounterFact-Strict数据集（N=50）揭示：简单且多样化的攻击策略即可实现高达96%的成功率，暴露出无记忆防御机制的严重脆弱性。此外，研究发现“复杂性悖论”（Complexity Paradox）——尽管复杂的迭代优化攻击有效，但常触发模型的防御性响应；而结构简单的策略反而能实现84.7%的真实说服成功率。解决方案的关键在于引入状态分离的对抗测试范式与轻量级但高效的认知驱动攻击代理，从而精准暴露模型在真实应用场景中的深层安全缺陷。
链接: https://arxiv.org/abs/2609.16777
作者: Zhuoang Cai
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:
Abstract:As Large Language Models (LLMs) increasingly serve as primary knowledge retrieval interfaces, their robustness against \textitpersuasion attacks—attempts to inject misinformation or enforce counterfactuals—has become a critical safety concern. Existing red-teaming frameworks typically evaluate models in multi-turn dialogues where the target model retains full conversation history. We identify a critical flaw in this setting termed \textbfRefusal Inertia'': a model's initial refusal often propagates through subsequent turns largely to maintain contextual consistency, thereby masking its true vulnerability to sophisticated, isolated persuasion attempts. To rigorously evaluate the cold-start’’ defense capabilities of SOTA models, we introduce the \textbfSAST-IR (Stateful Attacker, Stateless Target - Iterative Refinement) framework. By enforcing a memory wipe on the target while retaining the attacker’s history, we simulate a worst-case adversarial setting using \textbfmulti-turn (stateless) iterations. Leveraging \textbfCP-Agent (Cognitive Persuasion Agent), an enhanced diagnosis-guided agent, our experiments on the custom \textscCounterFact-Strict dataset ( N=50 ) yield alarming results: simple, diverse attack strategies achieved a staggering \textbf96% success rate, exposing severe brittleness in memory-less defense. Furthermore, we reveal a \textbf``Complexity Paradox’': while complex, iteratively refined attacks are effective, they often trigger defensive compliance, whereas simple strategies achieve a higher rate of genuine persuasion (\textbf84.7%). Our code and dataset are available at GitHub, this https URL.
[NLP-42] AME: Token Attribution and Masking for Emergent misalignment EMNLP
【速读】： 该论文旨在解决生成式AI（Generative AI）在特定领域数据上进行微调时，因数据存在缺陷而导致模型产生超出训练范围的有害行为——即涌现性错位（Emergent Misalignment, EM）的问题。其核心挑战在于：尽管已有研究定位了EM现象在模型权重、激活值和训练文档中的表现，但尚未明确具体是哪些训练令牌（training tokens）携带了导致错位的关键信号。为此，作者提出TAME（Token Attribution and Masking for Emergent Misalignment）框架，采用三阶段方法：首先通过前向传播释放的LoRA适配器计算每个响应令牌被微调更新所增强的可能性，获得令牌归因分数；其次分析高归因令牌中的语义模式；最后通过归因引导的损失掩码进行因果验证。实验结果表明，在释放的EM样本及6,849条医疗建议子集上，归因高度集中（前5%令牌贡献32%的归因质量），且在Llama模型中，医疗术语相关归因偏低，而对“未经证实的确定性表达”这一语言风格的归因显著增强，即使控制了词频影响。在重新微调过程中，仅掩码高归因令牌即可使Llama的EM降低23倍、Qwen降低36倍，且困惑度代价集中在目标语言风格而非医疗内容本身；而随机掩码则无法抑制EM。进一步分析显示，对于Llama而言，导致EM的关键信号更在于错误内容的表达自信程度而非其领域词汇本身，且该因果掩码效应在两种模型家族间具有普适性。因此，解决方案的关键在于：识别并干预引发错位的语言风格特征，而非简单移除特定领域词汇。
链接: https://arxiv.org/abs/2609.16754
作者: Md Rayhanul Masud,Md Rizwan Parvez
机构: University of California, Riverside(加州大学河滨分校); Qatar Computing Research Institute (QCRI)(卡塔尔计算研究研究所)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: Accepted at EMNLP UncertaiNLP Workshop 2026
Abstract:Fine-tuning an aligned language model on narrow, flawed data can induce harmful behavior far outside the training domain, known as emergent misalignment (EM). Prior work has localized EM in model weights, activations, and training documents, but it remains unclear which training tokens carry the relevant fine-tuning signal. We introduce TAME (Token Attribution and Masking for Emergent Misalignment), a three-stage framework: token attribution scores how strongly the fine-tuning update raises each response token’s likelihood, using forward passes through a released LoRA adapter; signal characterization finds patterns among high-attribution tokens; and causal validation tests them by attribution-guided loss masking. On released EM organisms and a 6,849-example medical-advice split, attribution is concentrated (the top 5% of tokens hold 32% of the mass) and, in Llama, depleted for medical vocabulary but enriched for a register of unwarranted certainty, even after controlling for token rarity. Masking high-attribution tokens during fresh fine-tuning cuts EM by 23x in Llama and 36x in Qwen, with the perplexity cost concentrated on the targeted register rather than on medical content; an equal random mask leaves EM unchanged. In Llama, the attribution pattern suggests that EM-relevant signal lies more in how confidently flawed content is expressed than in its domain vocabulary; the causal masking effect itself holds across both model families.
[NLP-43] IAO: Token Importance-Aware Policy Optimization for Text Summarization
【速读】： 该论文旨在解决文本摘要任务中生成内容的一致性与连贯性保持问题，尤其针对现有基于强化学习（Reinforcement Learning, RL）的方法在优化过程中对所有词元（token）同等对待、忽视其在句子与篇章层面质量贡献差异的缺陷。传统方法直接对未加区分的词元序列施加奖励信号，导致模型难以精准捕捉关键信息的重要性。为此，论文提出了一种新型强化学习策略——词元重要性感知策略优化（Token Importance-Aware Policy Optimization, TIAO），其核心在于通过分析词元之间的依赖关系，识别出对摘要质量具有决定性作用的核心词元，并据此对轨迹的优势值（advantage）进行重加权，使优化过程更聚焦于高价值信息的保留。实验结果表明，TIAO在真实数据集上表现优异，经其增强的70亿参数基础模型性能可媲美GPT-4和GPT-5-nano，验证了该方法在提升摘要质量方面的有效性与高效性。
链接: https://arxiv.org/abs/2609.16748
作者: Qixiu Li,Chenlong Bao,Xiang Zhu,Xiaoyong Li,Ruixin Cao,Shukai Chen,Zhenxiong Zhou
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Text summarization requires models to condense content while preserving key qualities such as consistency and coherence. Large language models (LLMs) have shown strong performance on this task and can be further improved through reinforcement learning (RL). However, most existing methods apply reward signals directly to undifferentiated token sequences, overlooking the varying importance of individual tokens to word and sentence level quality in summarization. In this paper, we propose Token Importance-Aware Policy Optimization (TIAO), a novel reinforcement learning strategy that explicitly leverages token-importance awareness. Specifically, TIAO identifies core tokens based on token dependency and reweights a trajectory’s advantage according to its overall dependencies. Experiments on the real world dataset show that our TIAO achieves highly competitive results, and that a 7B foundation model enhanced by TIAO performs comparably to GPT-4 and GPT-5-nano. Code is available at this https URL
[NLP-44] Japanese Stroke LLM Evaluation: A Conversational Benchmark for Safe Stroke Care in Japanese Using Large Language Models
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在临床实际应用中，特别是在脑卒中诊疗场景下的病史采集、紧急程度评估及安全性方面能力不足的问题。现有研究多集中于多选题测试，缺乏对真实临床对话情境下模型表现与安全性的系统评估。为此，研究团队提出了“日本脑卒中大语言模型评估”（Japanese Stroke LLM Evaluation），构建了一个面向日语场景的多轮对话基准，由认证神经外科医师扮演模拟患者与评价者，以实践导向方式评估模型在病史采集和处置决策阶段的表现。其解决方案的关键在于：第一，采用由神经外科专家主导设计的真实病例与评分标准，避免使用“大模型自评”（LLM-as-judge）带来的偏倚；第二，引入严格的安全阈值——整体得分不低于80%且无致命错误（critical mistakes），确保模型决策不会直接威胁生命；第三，通过量化分析发现，病史采集提问次数与得分呈显著正相关（r = 0.648, p = 0.007），提示高效沟通的重要性。结果显示，Claude Fable 5 和 Claude Opus 4.7 在2026年均达到安全阈值，表明云部署模型性能持续提升，但仍有部分模型存在如未确认实验室结果即给予溶栓治疗、未稳定气道即手术等严重失误，凸显进一步基于真实世界病例进行验证的必要性。
链接: https://arxiv.org/abs/2609.16739
作者: Keisuke Masuda,Kazutaka Yatsushiro,Hirohumi Iwamoto,Hirofumi Hirano,Ryosuke Hanaya
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Background: Large language models (LLMs) have achieved physician-comparable performance on multiple-choice medical knowledge examinations, but their capabilities in clinical history taking, urgency assessment, and safety remain insufficiently evaluated. We proposed Japanese Stroke LLM Evaluation, a multi-turn conversational benchmark for stroke care in Japanese, and evaluated LLM performance and safety under practice-oriented conditions. Methods: We created 10 stroke and related-condition cases and evaluated LLMs in multi-turn Japanese conversations. The LLM acted as physician, while a board-certified neurosurgeon acted as simulated patient and evaluator. Each case comprised history-taking and action phases scored using pre-specified criteria. Errors that could directly threaten life were defined as critical mistakes. The safety threshold was at least 80% overall with zero critical mistakes. Eighteen models were evaluated in October 2025 and June 2026. Results: Claude Fable 5 achieved the highest score (87.4%) with zero critical mistakes, followed by Claude Opus 4.7 (80.3%) and GLM-5.2 (75.6%). Two leaders met the safety threshold. Eleven models made 17 critical mistakes, including failure to confirm laboratory results or blood glucose before t-PA, surgery before airway stabilization, omission of cervical vascular evaluation, and t-PA outside its indication. History-taking question count correlated with history-taking score (r = 0.648, p = 0.007). Conclusions: Japanese Stroke LLM Evaluation provides a benchmark for LLM performance under practice-oriented conditions, including a cap on history-taking questions. Cases and evaluations were created by neurosurgical specialists rather than using an LLM-as-judge approach. Performance improved across cloud-based and on-premise models in 2026, with some exceeding the safety threshold. Further evaluation using real-world cases is required.
[NLP-45] VideoMM: Adaptive Macro-Micro Inference for Efficient Video MLLM s
【速读】： 该论文旨在解决多模态大语言模型（Multimodal Large Language Models, MLLMs）在长视频理解任务中因视觉标记（visual tokens）数量爆炸而导致的上下文窗口饱和与计算成本高昂的问题。现有方法主要依赖辅助模型进行标记压缩，但面临根本性困境：轻量级编码器驱动的方法常丢失关键语义信息，而基于重型MLLM的压缩策略则削弱了效率优势。本文指出，这一困境的根源在于对细粒度视觉细节的过度依赖——尽管其对详细理解至关重要，但在初步筛选语义相关区域的任务中实为冗余。为此，论文提出一种新范式——VideoMM，实现了从“以模型为中心的降维”到“自适应感知粒度”的转变。其核心创新在于将“选择”与“推理”解耦：先在低成本的宏观代理（Macro Proxy，由下采样帧生成）上执行语义过滤，仅将候选区域投影至高保真微观标记（Micro Tokens）进行精细化理解，从而显著降低计算开销。实验表明，VideoMM在LongVideoBench上相较全上下文基线实现6.13倍加速和7.4%准确率提升，并较当前领先方法再提速2.73倍，构建了可扩展的长视频理解新范式。
链接: https://arxiv.org/abs/2609.16722
作者: Haoyu Guo,Yuan Feng,Junlin Lv,Mingjun Xiao,S Kevin Zhou,Xike Xie
机构: University of Science and Technology of China (中国科学技术大学); Suzhou Institute for Advanced Research (苏州研究院)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM)
备注:
Abstract:Scaling Multimodal Large Language Models (MLLMs) to long-form video understanding is bottlenecked by the explosion of visual tokens, which saturates context windows and incurs prohibitive costs. Current solutions predominantly rely on auxiliary models for token reduction but face a fundamental dilemma: lightweight encoder-driven approaches often overlook critical semantic information, whereas heavyweight MLLM-driven reduction negates the efficiency gains. In this work, we identify a more fundamental inefficiency underlying this dilemma: while fine-grained visual details are essential for detailed understanding, they are largely redundant for the preliminary task of selecting semantically relevant regions. Motivated by this, we introduce \textbfVideoMM, which marks a paradigm shift from model-centric downsizing to adaptive perceptual granularity. Specifically, our framework decouples selection from reasoning by executing semantic filtering on a cost-effective \textitMacro Proxy (derived from downscaled frames), and projecting the selected regions onto high-fidelity \textitMicro Tokens for detailed understanding only when necessary. Extensive evaluations show that VideoMM significantly outperforms existing solutions. It achieves a 6.13 \times speedup and a 7.4% accuracy gain over full-context baselines on LongVideoBench, and further accelerates inference by 2.73 \times over current leading methods, establishing a highly scalable paradigm for long-video understanding. Our code is available at: this https URL.
[NLP-46] DiaWhisper-DPO: Role-Attributed Transcription of Clinical Interviews via Failure-Mined Preference Optimization ICASSP2027
【速读】： 该论文旨在解决临床访谈中自动抑郁筛查时，对发言者角色（患者或临床医生）进行准确归属的问题。现有方法依赖级联系统，先进行说话人分离（speaker diarization），再通过启发式规则分配角色，但此类方法存在错误传播问题，且在跨语言场景下性能受限。其解决方案的关键在于提出一种端到端模型DiaWhisper，该模型基于Whisper-large-v3，采用低秩适应（LoRA）进行微调，并引入一个辅助的帧级角色分类头，实现语音转录与发言者角色归属的联合建模。进一步提出DiaWhisper-DPO，一种基于失败挖掘的强化学习优化方法，利用真实解码失败作为拒绝样本，无需人工偏好标注即可提升模型鲁棒性。实验表明，在DAIC-WOZ数据集上，该方法将角色准确率提升至0.973，错误检测率（DER）降至0.119，较最强级联基线降低72%，且种子间方差由0.205显著下降至0.002；在跨语言的PDCH-HAMD数据集上也实现了0.757的角色准确率，并全面优于所有基线。
链接: https://arxiv.org/abs/2609.16661
作者: Weiming Li,Ana Catarina Fidalgo Barata,Miguel Constante,João Miguel Sanches
机构: 未知
类目: Computation and Language (cs.CL)
备注: 5 pages, 2 figures. Submitted to ICASSP 2027
Abstract:Automated depression screening from clinical interviews requires attribution of utterances to the clinician or patient. We evaluate two datasets: DAIC-WOZ, where participant-only recordings require re-synthesizing both sides for controlled two-party evaluation, and PDCH-HAMD, comprising voice-converted real Chinese interviews for cross-lingual validation. Cascaded systems combine speaker diarization with role-assignment heuristics, so errors can propagate across stages. We propose an end-to-end model, which we named DiaWhisper, that fine-tunes Whisper-large-v3 with LoRA and an auxiliary frame-level role head for transcription and attribution, together with DiaWhisper-DPO, a failure-mined refinement that uses genuine decoding failures as DPO rejected completions without human preference annotation. On 29 DAIC-WOZ test sessions, DiaWhisper-DPO achieves 0.973 role accuracy and 0.119 DER, 72% below the strongest cascaded baseline, and reduces seed variation from \sigma = .205 to .002. Retrained on PDCH-HAMD, it achieves 0.757 role accuracy and improves all 78 session-seed pairs.
[NLP-47] Rewarding Reasoning Not Answers: Fixing and Bounding Test-Time Reinforcement Learning on Medical QA
【速读】： 该论文旨在解决测试时强化学习（test-time reinforcement learning）在医学多选题问答（medical multiple-choice QA）任务中性能退化的问题，即模型在无标签测试集上通过多数投票伪标签进行自训练时，准确率停滞不前而输出多样性急剧下降。其核心问题诊断为：失败根源在于答案空间（answer space）的结构特性，而非领域难度本身——在小答案空间中，错误推理路径易收敛至同一错误伪标签并被强化；在大答案空间中，错误路径分散且难以获得有效奖励。针对此问题，论文提出PROSE（Process Reward Guided Self-Training）方法，其关键在于将奖励信号从答案一致性转向推理过程质量，利用医学过程奖励模型（medical process reward model）对每一步推理进行评分，并以轨迹中最小得分作为整体奖励，同时施加答案格式约束。该方案无需外部奖励模型即可在推理阶段运行，且能将优化效果迁移至未见数据集。研究进一步揭示，最小值聚合机制至关重要，而均值聚合易被利用，导致代理奖励饱和但准确率下降。
链接: https://arxiv.org/abs/2609.16660
作者: Kailong Fan,Anqi Pu,Yichen Wu,Wanhua Li,Yicong Li,Hanspeter Pfister,Huafeng Liu,Xiang Li,Quanzheng Li,Ning Guo
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Test-time reinforcement learning adapts a model on its own unlabeled test set using majority-vote pseudo-labels and has shown strong results in mathematics. We show that this recipe collapses on medical multiple-choice QA: accuracy stagnates while output diversity rapidly declines. Through a controlled experiment that keeps the questions, model, and optimizer fixed while changing only the answer space, we trace this failure to answer-space structure rather than domain difficulty. In small answer spaces, incorrect rollouts often collide on the same wrong pseudo-label and reinforce it; in large answer spaces, they disperse and receive little reward. This diagnosis motivates PROSE, Process Reward Guided Self-Training, which rewards reasoning quality instead of answer agreement. PROSE scores each reasoning step with a medical process reward model, assigns the trajectory reward as the minimum score across steps, and enforces answer-format constraints. Without labels, PROSE substantially improves a general Llama model, surpassing purpose-built medical models and matching much larger systems. Because the process signal is internalized into the policy, the adapted model requires no reward model at inference and transfers its gains to unseen datasets. We further show that the minimum aggregation is essential: mean aggregation can be exploited, saturating the proxy reward while degrading accuracy.
[NLP-48] GrowMTP: Can RL Grow Its Own Draft Head?
【速读】： 该论文旨在解决大语言模型在强化学习（Reinforcement Learning, RL）后训练过程中，自回归式滚动生成（autoregressive rollout generation）导致的计算耗时过长这一瓶颈问题。现有解决方案如推测解码（speculative decoding）虽能提升效率，但其草稿头（draft head）通常需在RL之前进行预训练或冷启动，引入了额外的训练开销。本文提出一种名为GrowMTP的新方法，其核心创新在于利用RL训练过程本身提供的两个关键条件：一是滚动生成分布相较于预训练阶段显著更窄，二是验证步骤持续提供与该分布对齐的监督信号。基于此，GrowMTP能够在不依赖外部预训练的前提下，完全在RL循环内从零开始在线训练草稿头，并将草稿头更新与策略主干（policy backbone）解耦。实验结果表明，在Qwen3-4B、MiMo-7B-SFT和Qwen3.5-4B-Base三个不同基础模型上，GrowMTP分别实现了2.13倍、1.93倍和1.36倍的滚动加速，以及1.60倍、1.41倍和1.20倍的端到端加速。因此，GrowMTP可作为模块化组件无缝集成至现有RL训练框架中，尤其为无预训练草稿头的模型提供了从零开始的高效加速路径。
链接: https://arxiv.org/abs/2609.16648
作者: Minghua He,Lingzhe Zhang,Yuan Liu,Xiao Zhou,Aiwei Liu
机构: WeChat AI, Tencent(微信AI, 腾讯); Peking University(北京大学)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:
Abstract:Reinforcement learning (RL) post-training drives the frontier capabilities of large language models, with its wall-clock dominated by autoregressive rollout generation. Speculative decoding is an established remedy for this bottleneck, but existing draft heads must be pretrained or warmed up before RL, introducing substantial training cost outside the RL run to be accelerated. We observe that RL training itself provides both conditions required for online draft-head training: its rollout distribution is far narrower than that of pretraining, and its verification step continuously produces supervision signals aligned with this distribution. Building on these observations, we propose GrowMTP, which uses this supervision to train a draft head from scratch entirely within the RL loop, with all head updates detached from the policy backbone. On Qwen3-4B (no draft head), MiMo-7B-SFT (weak head), and Qwen3.5-4B-Base (strong head), GrowMTP achieves rollout speedups of 2.13x, 1.93x, and 1.36x, and end-to-end speedups of 1.60x, 1.41x, and 1.20x, respectively. GrowMTP therefore serves existing RL training frameworks as a modular component, particularly offering a from-scratch acceleration path for models without pretrained draft heads.
[NLP-49] RoleBreak: Benchmarking Long-Horizon Role-Playing Robustness in Spoken Dialogue ICASSP2027
【速读】： 该论文旨在解决当前语音对话模型在长时程角色扮演（long-horizon role-playing）中角色一致性与情感表达鲁棒性不足的问题，尤其关注模型能否在脱离预设虚构角色的前提下，持续维持多样化角色身份并具备自然、连贯的语音情感表现。现有基准多聚焦于短时、以角色为中心的对话，难以评估系统在复杂、长期交互中的稳定性与安全性。为此，作者提出 RoleBreak——一个开放的长时程语音角色扮演鲁棒性评测基准，包含310个基于角色和用户中心的角色设定、6,688条经人工验证的对话轮次及11,743项细粒度评估指标，其中1,856条对话轮次附带明确的语音情感目标，用于评估声学层面的情感表现力。其场景设计着重考验角色一致性、交互质量、安全性和情感动态。研究评估了九种涵盖全双工、多模态与级联语音识别—大语言模型—文本转语音（ASR–LLM–TTS）范式的配置，揭示四大关键发现：（1）当前系统在语义角色遵循方面显著优于语音情感表现；（2）即使最强模型在平均10.4轮后即出现角色偏离，11.6轮后出现安全问题，表明语义鲁棒性在长程交互中仍极脆弱；（3）扩大大语言模型规模可有效提升语义鲁棒性并延缓失败，但对语音情感表达改善有限；（4）用户语音情感本身会显著影响角色行为，即便语言内容保持不变。上述结果凸显了当前语音角色扮演系统在长时程鲁棒性与声学情感表达方面的根本性缺陷，为未来研究指明了方向。
链接: https://arxiv.org/abs/2609.16614
作者: Yuqi Wang,Fengyuan Liu,Haochen Luo,Zhiqi Yu,Qi Liu
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 5 pages, 2 figures, 3 tables. Submitted to ICASSP 2027
Abstract:Speech-to-speech dialogue models increasingly support persona control, yet existing spoken role-playing benchmarks remain largely character-centric and short-horizon. This leaves open whether spoken dialogue models can sustain diverse roles over extended interactions, especially beyond predefined fictional characters. We introduce RoleBreak, an open benchmark for long-horizon role-playing robustness in spoken dialogue. RoleBreak contains 310 character-based and user-centered roles, 6,688 human-verified dialogue turns, and 11,743 fine-grained evaluation criteria, with 1,856 turns carrying expressive emotion targets for evaluating vocal emotion. Its scenarios are designed to stress role consistency, interaction quality, safety, and affect over extended conversations. We evaluate nine configurations spanning full-duplex, omni-modal, and cascaded ASR–LLM–TTS paradigms. We find four key patterns. First, current systems are substantially stronger at semantic role adherence than at vocal emotion. Second, semantic robustness remains brittle over long interactions: even the strongest evaluated system encounters its first persona and safety failures after only 10.4 and 11.6 turns on average. Third, scaling the LLM substantially improves semantic robustness and delays failure, but yields little improvement in vocal emotion. Finally, user vocal emotion affects role-playing behavior even when linguistic content is fixed. These findings highlight persistent gaps in both long-horizon robustness and vocal expressiveness in spoken role-playing systems.
[NLP-50] Challenges of Auditing: Variability in Outputs of Large Language Models for Health
【速读】： 该论文旨在解决当前前沿生成式 AI 模型在健康咨询应用中存在的一种关键问题：不同访问模式（如 ChatGPT、ChatGPT Health、API 接口）之间存在系统性差异，而现有评估体系主要依赖 API 接口，无法真实反映普通用户通过聊天机器人界面的实际使用体验。这一脱节导致评估结果缺乏外部有效性，难以准确衡量模型在真实应用场景中的表现。其解决方案的关键在于，要求模型提供方支持对消费者实际交互环境与设置的忠实复现，以实现对生成式 AI 模型在真实使用情境下的严谨审计与可信评估。
链接: https://arxiv.org/abs/2609.16590
作者: Yuan Pu,Yewon Chang,Furong Jia,Xunjian Yin,Jessica Ma,Ayman Ali,Monica Agrawal
机构: Duke University (杜克大学); Durham VA Health System (达勒姆退伍军人健康系统)
类目: Computation and Language (cs.CL)
备注:
Abstract:People increasingly use frontier AI models for health advice, but via different access modes (e.g., ChatGPT, ChatGPT Health, APIs) with varying settings. Here, we find systematic differences across access modes. Because evaluations typically rely on APIs while consumers interact through chatbot interfaces, these discrepancies limit evaluation validity. Our findings underscore an urgent need for model providers to enable faithful replication of consumer experiences and settings for rigorous audits.
[NLP-51] CLASH: Counterfactual Auditing of Lexical and Prosodic Reliance in Spoken Sarcasm Detection
【速读】： 该论文旨在解决现有语音讽刺检测模型在评估过程中无法明确区分其预测依赖于词汇内容、韵律特征还是二者交互作用的问题。传统评估方法缺乏对不同线索贡献度的解析能力，导致对模型决策机制的理解受限。为此，作者提出一种名为CLASH（Controlled Lexical-Acoustic Separation Harness）的双语反事实诊断框架，通过在原始、保留词汇内容、保留韵律特征以及近似中性化四种条件下对话语进行评估，实现对语音与语义线索的解耦分析。该研究的关键在于利用可控的反事实干预手段，系统分离并量化词汇与韵律对讽刺识别的独立及协同影响，从而揭示模型实际依赖的驱动因素。实验结果表明，在控制时长后，仅保留词汇内容的语音仍显著优于仅保留韵律的语音（AUROC优势达0.135–0.148），且该优势在多种重合成策略下保持稳定；而声学干预虽能改变分数，但并未一致提升判别能力或改变二分类预测，说明模型的敏感性并不等同于真正的讽刺识别能力。此外，上下文和交互效应在不同数据集间存在差异，进一步凸显了时长、说话人身份及变换方式对结果的影响。该工作为理解生成式语音模型中的语义-声学耦合机制提供了可解释性工具，并推动了对讽刺检测模型内在机理的深入剖析。
链接: https://arxiv.org/abs/2609.16582
作者: Qiyang Sun,Xudong Li,Yupei Li,Jiabin Xue,Yuhang Dai,Jiaming Li,Bjorn W. Schuller
机构: Sun Yiqiang(孙一强); Li Xudong(李旭东); Li Yupei(李玉佩); Xue Jiabin(薛佳斌); Dai Yuhang(戴宇航); Li Jiaming(李佳明); Schuller Björn W.(Björn W. Schuller)
类目: ound (cs.SD); Computation and Language (cs.CL)
备注:
Abstract:Spoken sarcasm detectors may exploit lexical content, prosody, or their interaction, yet conventional evaluation cannot reveal which cues drive their predictions. We introduce CLASH (Controlled Lexical-Acoustic Separation Harness), a bilingual counterfactual diagnostic framework that evaluates each utterance under original, lexical-preserving, prosody-preserving, and approximately neutralised conditions. We evaluate handcrafted acoustic-feature systems, self-supervised learning (SSL) probes, and large audio language models (LALMs) on CMMA and MUStARD. For target-only Qwen3-Omni, lexical-preserving speech retains a 0.135–0.148 AUROC advantage over prosody-preserving speech after duration balancing, with cluster-bootstrap intervals above zero; alternative lexical resynthesis preserves this advantage. Acoustic interventions shift scores without consistently improving discrimination or changing binary predictions under the evaluated conditions. Context and interaction estimates vary across corpora. These findings distinguish acoustic sensitivity from sarcasm discrimination while exposing duration, identity, and transformation effects.
[NLP-52] PunGraph: Retrieval-Enhanced Phonetic-Semantic Graph Reasoning for Pun Understanding EMNLP2026
【速读】： 该论文旨在解决生成式人工智能（Generative AI）在理解双关语（pun）这一复杂修辞形式时面临的挑战，尤其针对其在语音相似性建模不足以及端到端生成过程缺乏控制导致的推理偏差问题。其核心解决方案是提出PunGraph——一种基于检索增强的知识图谱框架，通过整合Unisyn语音词典、国际音标（IPA）与音素到字母转换（G2P）表示，以及WordNet语义定义，构建一个语音-语义词汇图谱，从而在结构化候选空间内对大语言模型（LLM）的推理进行约束。该方法利用检索机制筛选潜在的双关候选词或词义，有效缓解了因语义模糊和语音近似引发的误判。研究还引入WebPun数据集，包含5,730个标注的异形双关与同形双关实例，为模型评估提供支持。实验表明，PunGraph显著提升了小规模模型的表现，并在性能上达到与主流专有模型相当的水平，验证了结构化知识与生成式模型融合的有效性。
链接: https://arxiv.org/abs/2609.16557
作者: Yuchen Su,Zijian Huang,Yaotian Shi,Shaoxin Zhong,Ruofan Wang,Mengze Li,Yonghua Zhu,Diana Benavides-Prado,Michael Witbrock
机构: University of Auckland(奥克兰大学); Queen Mary University of London(伦敦玛丽女王大学); Shanxi University(山西大学)
类目: Computation and Language (cs.CL)
备注: EMNLP2026 Main Conference
Abstract:Puns are a challenging form of figurative language that exploit phonetic similarity and semantic ambiguity to convey multiple meanings. Although large language models (LLMs) demonstrate strong language understanding capabilities, they still struggle with pun reasoning due to limited phonetic modeling and uncontrolled end-to-end generation. We propose \textbfPunGraph, a retrieval-enhanced knowledge graph framework for pun understanding. PunGraph constructs a phonetic-semantic lexical graph using the Unisyn phonetic dictionary, IPA and G2P representations, and WordNet definitions, and retrieves candidate words or senses to constrain LLM reasoning within a structured candidate space. We further introduce \textbfWebPun, a new large-scale dataset containing 5,730 annotated heterographic and homographic puns. Experiments on SemEval-2017 and WebPun show that PunGraph consistently improves the performance of small-scale LLMs and achieves competitive results against strong proprietary models. Further analysis shows that retrieval-guided phonetic and semantic constraints effectively reduce common reasoning errors in pun interpretation, highlighting the benefits of integrating structured knowledge with LLMs. We release our code and dataset at this https URL.
[NLP-53] Style-Debiased DPO: Updating LLM Knowledge with Factuality-Aware Synthetic Preference Data
【速读】： 该论文旨在解决大语言模型（LLM）在持续预训练（CPT）后，虽已存储小规模源语料库的知识，但在实际推理时难以准确提取已有知识的问题。其核心挑战在于：当模型生成错误回答时，传统偏好优化（DPO）方法会将该错误响应视为“劣质”样本进行惩罚，但若目标知识部分已知，此类“被拒绝”的响应中往往包含正确事实信息，仅在表达风格（如长度、措辞）上与标准答案存在差异。若直接使用标准DPO，会导致对这些含正确知识的风格差异响应施加不当惩罚，从而削弱模型性能。为此，论文提出风格去偏的偏好优化（Style-debiased DPO, SD-DPO），其关键创新在于：对每一对偏好样本，首先判断被拒绝响应是否在事实上正确；若正确，则反转该对样本的偏好信号，并通过加权机制使风格差异带来的学习信号相互抵消，从而保留正确的事实知识并提升知识提取准确性。实验表明，在基于EntiGraph的存储方法基础上，SD-DPO以数十倍更少的训练数据实现了显著更高的问答准确率；在动态知识更新任务AToKE上，该方法达到0.982的整体准确率，能根据查询时间周期准确选择新旧事实，验证了其高效且精准的知识检索能力。
链接: https://arxiv.org/abs/2609.16532
作者: Takayuki Yamamoto,Daisuke Kawahara
机构: Waseda University(早稻田大学)
类目: Computation and Language (cs.CL)
备注: 23 pages, 3 figures, 13 tables
Abstract:Continued pretraining (CPT) with data augmentation such as paraphrasing can store inside a large language model (LLM) the knowledge of a small source corpus. The stored knowledge, however, is not always retrieved correctly. We study the eliciting side rather than the storing side: we use preference optimization, which learns from pairs of a preferred (chosen) and a dispreferred (rejected) response, so that the model elicits its stored knowledge more accurately. One proposed approach takes the model’s own erroneous response as rejected and the gold answer as chosen, so as to suppress the error. When the target knowledge is partially known, however, most of these rejected responses are factually correct. Using direct preference optimization (DPO) then pushes down rejected responses that contain correct knowledge and differ from the chosen answer only in style, such as length and wording. We propose style-debiased DPO (SD-DPO), which scores whether the rejected response of each pair is factually correct, inverts the preference of such pairs, and weights them so that the learning signal due to differences in style cancels out as a whole. We first test whether, on top of EntiGraph, a representative storing-side method that runs CPT on text synthesized from the corpus, our method adds accuracy efficiently. On QuALITY, the reading-comprehension QA benchmark on which EntiGraph was evaluated, SD-DPO exceeds a baseline we CPT on EntiGraph’s synthetic data from the same base model and evaluate with the same procedure. The training tokens this requires are a few dozen times fewer than the additional CPT needed for the same gain. For knowledge updating, the main goal of this work, we use AToKE, a knowledge-editing benchmark for facts that change over time. There, SD-DPO reaches an overall accuracy of 0.982 and answers with the new or the old fact according to the queried period.
[NLP-54] Competence-Preserving Resume Perturbations Expose Presentation Sensitivity in LLM Screening
【速读】： 该论文旨在解决生成式人工智能（Generative AI）在简历筛选任务中对表面呈现差异的敏感性问题，即当候选人的实际胜任力水平不变时，仅因简历的表述方式、结构或排版等表面特征变化而导致筛选决策不一致的问题。其核心解决方案是构建基于职业背景、在可控胜任力水平下生成的候选人画像，并将其转化为多种简历表现形式，通过确定性验证门控机制排除底层证据发生变化的变体，从而确保评估仅针对“胜任力保持不变但呈现方式不同”的情形。研究发现，尽管部分模型（如Llama-3.1-8B）在筛选有效性上表现较好（0.781），但在面对无实质内容变化的呈现转换时仍出现高达29.6%的决策反转；而Mistral-7B-v0.3的决策稳定性更差，翻转率达41.4%。这表明，当前主流指令微调大语言模型（LLM）在简历筛选任务中存在显著的呈现不稳定性，因此，评估此类系统不仅需考察其识别更强候选人的能力，还必须纳入对决策稳定性的严格检验。
链接: https://arxiv.org/abs/2609.16517
作者: Qiangju Chen,Yang Xiao
机构: Macquarie University(麦克里大学); The University of Melbourne(墨尔本大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Under Peer Review
Abstract:Resume screeners must infer job-relevant competence from resumes whose presentation can vary substantially in wording, structure, stylistic polish, and document extraction quality. Ideally, such surface variation should not change decisions when the underlying qualification evidence is unchanged. We introduce a controlled audit of this property, constructing occupation-grounded candidate profiles at controlled competence levels and rendering each profile into multiple resume presentations. A deterministic validation gate excludes variants that alter the underlying evidence before scoring. Across six open instruction-tuned LLM conditions, we find a clear disconnect between screening validity and presentation stability. Llama-3.1-8B with its native chat template achieves the strongest validity ( 0.781 ) yet reverses 29.6% of matched pairwise decisions under competence-preserving presentation changes; Mistral-7B-v0.3 reaches validity 0.644 with a 41.4% flip rate. Native chat formatting improves validity for several chat-tuned models but does not remove this instability. These results show that resume-screening evaluations should assess not only whether a system identifies stronger candidates, but also whether those decisions remain stable when the same competence evidence is presented differently.
[NLP-55] Beyond the Name: Demographic Leakage in De-Identified Résumés and Evaluation Artifacts in LLM Bias Audits
【速读】： 该论文旨在解决去标识化简历筛选中隐含的族裔文化信息泄露问题，特别是针对语言字段被删除后仍存在的残余泄露现象。其核心问题是：即使移除显式语言字段，简历中的非语言性文本（如措辞、表达风格等）是否仍可导致族裔文化属性的推断。解决方案的关键在于通过控制语言属性严格一致，系统性地隔离五种族裔文化情境与三类线索显著性层级下的非结构化文本内容，从而验证非语言文本在高显著性条件下仍能实现高达1.000的目标群体恢复率。研究发现，模型差异仅在弱线索条件下显著（0.086–0.690），表明线索显著性是评估模型公平性的关键维度；同时揭示了评价设计对大语言模型作为裁判者结果的极端敏感性——禁止平局会导致显著的选择率偏差（0.39），而允许平局则使多数模型产生近乎普遍的平局（≥94%），凸显了评估协议本身可能引入的虚假效应。因此，该研究强调必须区分简历内容本身蕴含的可恢复人口统计信号与评价机制所引入的系统性偏差。
链接: https://arxiv.org/abs/2609.16501
作者: Qiangju Chen,Yang Xiao
机构: Macquarie University (麦考瑞大学); The University of Melbourne (墨尔本大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Under peer review
Abstract:De-identified résumé screening assumes that redacting explicit fields prevents ethnocultural inference; however, recent audits attribute residual leakage to declared languages. We investigate whether eliminating language fields resolves this leakage across nine open-weight models and 620 counterfactual résumés. By holding language attributes strictly identical, we isolate unstructured prose across five ethnocultural conditions and three cue-salience tiers. Target-group recovery averages 0.757 overall and saturates at 1.000 under high salience, demonstrating that non-language prose sustains demographic inference. Crucially, models diverge only under faint cues (0.086-0.690), establishing salience as an essential evaluation axis. Furthermore, pairwise LLM-as-a-judge outcomes are highly sensitive to evaluation design: forbidding ties yields an apparent selection-rate ratio of 0.39 alongside strong position and content effects, whereas permitting ties produces near-universal ties for most models ( \ge94% ). Downstream scoring shows only very small between-condition differences, highlighting the need to distinguish demographic signals recoverable from résumé content from effects introduced by the evaluation protocol.
[NLP-56] Early-Bird Decoding: Accelerating Diffusion LLM s with Learnable Block Sizes and Parallel Sampling
【速读】： 该论文旨在解决扩散型大语言模型（diffusion large language models, dLLMs）在推理过程中因需大量迭代步骤才能使词元置信度达到解码阈值而导致的效率低下问题，尤其是在采用分块式键值缓存（block-wise KV caching）的情况下仍存在显著延迟。其核心解决方案是提出一种名为“早鸟（Early-Bird, EB）”的并行解码框架，关键在于：（1）设计一个可学习的网络，能够自适应地将不确定性相似的词元聚类为可变长度的块，突破传统固定块大小的限制；（2）引入一种位置感知采样器，通过学习在预测的可变长度块内以更少的解码步数并行解码词元，从而提前完成解码。这两个组件均无需修改预训练dLLM的权重，可作为即插即用的模块部署于推理服务中，具有极低的训练与推理开销。实验结果表明，EB-Decode在三种模型和四个基准测试上均显著优于基线方法，相较原始解码方式实现3.53–18.76倍的吞吐量提升，并在保持相近精度的前提下，比最强基线Fast-dLLM高出最多1.58倍。
链接: https://arxiv.org/abs/2609.16450
作者: Lixuan Wei,Wei Zhou,Jianwen Wu,Yipeng Shen,Meiling Wang,Haoran You
机构: Harvard University(哈佛大学); Georgia Institute of Technology(佐治亚理工学院); Purdue University(普渡大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 23 pages, 4 figures
Abstract:Diffusion large language models (dLLMs) offer a promising parallel decoding paradigm as an alternative to autoregressive generation through iterative unmasking. However, dLLMs typically require many steps before token confidence reaches the decoding threshold, resulting in inefficient inference even with block-wise KV caching. To accelerate dLLM inference, we for the first time propose an “early-bird (EB)” decoding framework, motivated by the observation that tokens with similarly low entropy tend to cluster and can be jointly decoded earlier, before reaching the confidence threshold. In particular, our EB-Decode framework integrates two key enablers: (1) a learnable network that adaptively groups tokens with similar uncertainty into variable-length blocks, rather than relying on fixed block sizes; (2) a position-aware sampler that learns to unmask tokens in parallel using fewer decoding steps within predicted variable-length blocks. Both components are developed without modifying pretrained dLLM weights and can therefore be directly deployed as plug-ins during serving, with negligible training and inference overhead. Extensive experiments across three models and four benchmarks consistently validate our observation and the effectiveness of EB-Decode, achieving 3.53-18.76 \times higher throughput than the vanilla decoding method and up to 1.58 \times higher throughput over the strongest baseline, Fast-dLLM, with comparable accuracy.
[NLP-57] Interpreting and Steering LLM Agents for Social Simulations
【速读】： 该论文旨在解决生成式人工智能（Generative AI）在社会科学模拟中应用时面临的两大核心局限：可解释性（interpretability）与可控性（steerability）。具体而言，尽管基于大语言模型（LLM）的模拟在理解人类行为方面展现出强大潜力，但其作为“黑箱”的特性限制了其在社会科学研究中的可信度与实用性。可解释性不足导致难以明确识别驱动行为的内在机制，而可控性缺乏则使得研究者无法有效调控特定理论相关的心理或认知机制以实现预期的行为输出。为此，本文提出并系统比较了三种方法——提示工程（prompt-based manipulation）、SAE（Sparse Autoencoder）提取特征操控以及基于探测器（probe-based direction steering）的方向性操控——以提升对LLM代理行为的解读与干预能力。研究通过四类经典经济与创造性任务（如风险偏好、利他主义、发散创造力及产品创新）的自然语言交互形式，验证了这些方法在操纵人类行为关键维度上的有效性。结果表明，相较于基础提示方法，基于SAE与探测器的方法在行为调控精度和稳定性上具有显著优势，尤其当结合使用时：先利用SAE将模型内部表示分解为可读的人类可理解特征，再通过探测器对特定方向进行精准调节，形成了一条高效且可解释的代理行为干预路径。这一方法框架为社会科学家在构建可解释、可调控的智能体模拟系统方面提供了关键技术支撑。
链接: https://arxiv.org/abs/2609.16436
作者: Jiayue Gaveal Fan,Arul Murugan,Shreyas Krishnan,Abhishek Nagaraj
机构: Data Innovation and AI Lab (DIAL), University of California, Berkeley; National Bureau of Economic Research (NBER)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 70 pages, 27 figures, 3 tables
Abstract:Simulations based on large language models (LLMs) have proven to be powerful for understanding human behavior, making them valuable additions to the social scientific toolkit. However, LLMs are ultimately black boxes based on deep neural networks which limits their value for social science. This is because of a lack of (i) interpretability: i.e. the ability to assign clear mechanisms driving observed behavior; and a lack of (ii) steerability: i.e. the ability to mute or amplify specific theoretically meaningful mechanisms of action to drive specific model behavior. Here, we demonstrate how the black box could be opened up to further enrich LLM-based simulations. Specifically, we compare three types of methods: (1) prompt-based manipulation, (2) SAE-derived feature steering, and (3) probe-based direction steering and examine their utility for LLM-based social scientific simulations. We do so by interpreting and steering two foundational components of human behaviors, namely preferences (risk attitudes, altruism) and capabilities (divergent creativity, product innovation), operationalized using four classic economic and creative tasks implemented as natural-language interactions. Overall, our results show that SAE- and probe-based techniques often outperform basic prompt-based methods for steering LLM agents, although this advantage depends on the specific prompting strategy involved. Together, SAEs and probes constitute an effective pipeline for social scientists seeking to interpret and steer agents in social simulations: SAEs decompose agents’ internal representations into human-readable features, after which probes can reliably shift agents’ behaviors in specified directions. We discuss implications of these methods for future work using LLM agents for social scientific simulations.
[NLP-58] ReMova: Fine-tuning LLM s for English to Belarusian translation
【速读】： 该论文旨在解决白俄罗斯语（Belarusian）机器翻译中因训练数据质量差而导致的性能瓶颈问题，尤其针对白俄罗斯语存在的两种正字法并存、网络文本中的噪声、其他语言干扰及常见拼写错误等独特挑战。其解决方案的关键在于提出了一套专为白俄罗斯语设计的数据清洗流程，该流程引入了专门的校正工具，有效处理上述多源数据质量问题。实验表明，经过过滤的训练数据对所有微调模型均有显著提升效果，其中基于大语言模型（LLM）的系统受益程度约为专用编码器-解码器翻译系统的两倍，这进一步验证了数据质量是当前白俄罗斯语机器翻译的主要制约因素。
链接: https://arxiv.org/abs/2609.16427
作者: Mikita Pilinka,Aliaksandr Kliujeŭ,David Samuel,Yves Scherrer
机构: 未知
类目: Computation and Language (cs.CL)
备注: WMT26 submission
Abstract:This paper presents a Belarusian-specific data-cleaning pipeline and fine-tuning for English-Belarusian machine translation. Our cleaning pipeline distinguishes itself from others by employing a correction tool that addresses the issue of the two orthographies of the Belarusian language, noise in the training data, interference from other languages and other misspelling issues common in Belarusian on the internet. A matched ablation on unfiltered training data shows substantial benefits from filtering for all fine-tuned models, with the LLM-based models gaining roughly twice as much from filtering as the dedicated encoder-decoder MT system, supporting the view that for Belarusian MT one of the primary bottlenecks is data quality.
[NLP-59] Negation Beyond the Verbal Channel: Temporal Multimodal Correlates in Dialogue ACL
【速读】： 该论文旨在解决在无语言（lexical）和声学（acoustic）信息条件下，仅通过多模态非语言行为能否识别口语否定（negation）语境这一核心问题。其关键解决方案在于：利用虚拟现实环境中采集的高精度、多模态同步数据（包括眼动、面部、头部、身体、手部及手指动作），构建时间序列分类模型，在完全排除词汇与语音输入的前提下，探究否定语境中是否存在可量化的多模态行为信号及其时空分布特征。研究发现，仅基于说话者侧的非语言行为，模型即可达到高达0.75的平均外部受试者曲线下面积（AUROC），且预测信息主要集中在否定线索出现前后的时间窗口内；其中面部特征对模型性能影响最大，表明其承载了关键的非语言否定信号；同时，对话伙伴的行为虽具有较弱的预测能力，但其信号呈现更分散的时序模式。该研究证实了多模态非语言行为在口语否定语境中的可辨识性，并揭示了其在时间动态与模态特异性上的关键特性。
链接: https://arxiv.org/abs/2609.16396
作者: Leon Hammerla,Patrick Schrottenbacher,Alexander Mehler
机构: Goethe University, Frankfurt am Main, Germany(歌德大学，美因河畔法兰克福，德国)
类目: Computation and Language (cs.CL)
备注: To be submitted to the October 2026 cycle of ACL Rolling Review (ARR)
Abstract:Negation is typically modeled through its linguistic realization, although spoken interaction is accompanied by tightly coordinated nonverbal behavior. We ask whether contexts centered on spoken negation cues contain measurable multimodal behavioral information: whether they can be distinguished from matched control contexts without lexical or acoustic input, where this information occurs in time, which modalities carry it, and whether it extends to the dialogue partner. We study 27 human-human interviews conducted in virtual reality, comprising temporally aligned gaze, facial, head, body, hand, and finger behavior and 964 annotated negation cues. Treating classification as a predictive probe, we compare 20 time-series models while excluding lexical and acoustic information, and then systematically vary temporal context, interactional source, modality availability, and event timing. Across grouped 10-fold cross-validation, the strongest probes reach up to .75 mean held-out AUROC from speaker-side behavior. Temporal analyses show that predictive information is concentrated around cue onset but remains detectable over a broader surrounding interval, while dialogue-partner behavior carries weaker predictive information with a comparatively diffuse temporal profile. Ablation and timing perturbations further show that facial features produce the largest modality-ablation effect and that the trained probe is sensitive to the temporal organization of the observed events.
[NLP-60] ParsHate: A Benchmark Dataset for Hate and Target Detection in Persian EMNLP2026
【速读】： 该论文旨在解决波斯语（Persian）网络文本中仇恨言论检测（hate speech detection）缺乏长期、高质量、细粒度标注数据集的问题。现有研究受限于数据规模小、时间跨度短以及标注维度单一，难以支撑对波斯语仇恨言论复杂性与演变规律的深入分析。其解决方案的关键在于构建ParsHate——首个覆盖2013至2022年十年跨度的波斯语推文仇恨言论数据集，包含10,000条经人工标注的推文，其中31%为仇恨内容。该数据集不仅支持仇恨言论的二分类检测，更实现了多标签细粒度目标识别（涵盖七个结构化目标类别），并区分显式与隐式仇恨及其对应目标，同时提供句段级推理依据（span-level rationales）。通过结合随机采样与基于评分的分层时间采样策略，有效缓解了关键词驱动带来的偏差，保持了真实自然的标签分布。实验表明，尽管当前最先进的模型在仇恨检测任务上达到79%的F1值，但在目标识别任务上表现较差（25.5% macro-F1），尤其在早期年份样本上性能下降明显，凸显了该数据集在多样性与挑战性上的优势，亟需更先进的建模方法以实现突破。
链接: https://arxiv.org/abs/2609.16393
作者: Zahra Bokaei,Walid Magdy,Bonnie Webber
机构: University of Edinburgh (爱丁堡大学)
类目: Computation and Language (cs.CL); Databases (cs.DB)
备注: Accepted to EMNLP 2026 (Main Conference)
Abstract:We introduce ParsHate, a manually annotated dataset of 10,000 Persian tweets spanning 2013-2022, representing the first decade-long benchmark for hate speech detection in Persian. The dataset contains 31% hateful content and supports both hate detection and multi-label fine-grained target identification across seven structured target categories. ParsHate also distinguishes explicit and implicit hate, marks explicit and implicit targets, and provides span-level rationales. Data collection combines random and score-stratified temporal sampling to reduce keyword-driven bias while preserving natural label distributions. Applying SOTA models for Persian hate-speech detection on ParsHate shows moderate performance (79% F1), especially with samples from earlier years, and low performance with target identification (25.5% macro-F1). This emphasizes the diverse sampling of hate speech in ParsHate and its challenging nature that requires more advanced methods for better performance. Dataset is made publicly available.
[NLP-61] Attention Mean Fields Predict Averag e Representation Dynamics and Reveal Context-Specific Computation
【速读】： 该论文旨在解决语言模型中表示几何（representation geometry）动态演化过程的建模问题。传统方法依赖于静态的、与模型无关的统计量（如共现频率），无法捕捉模型在推理过程中表示空间随层深度和上下文变化而动态演变的本质。其核心解决方案是引入注意力机制的平均场（mean-field）分析，通过定义从一个词元到另一个词元的平均注意力为核函数（kernel），该核函数可逐层传递表示并迭代模拟表示几何的变换。该方法采用两种条件设定：在语料库条件下，预测整体平均情况下的表示演化；在单个上下文条件下，预测特定上下文下的期望表示几何。注意力头相对于平均场预测的偏差（mean-field deviation）被用来分离出模型中未被平均场捕获的上下文特异性计算成分。在语料库条件下的平均场模型构成一个开环预测系统，仅需输入嵌入和冻结权重即可迭代预测整个网络的表示演化，无需依赖实际的层间偏差，且预测精度极高。实验表明，在训练初期，模型与平均场几乎不可区分；而在归纳能力出现的临界点后，两者开始分化，偏差逐渐增大，反映出表示的上下文化（contextualization）过程。在上下文条件下的偏差则成为一种任务无关的上下文特异性计算度量，其残差可分解为异常注意力路由与传输值的上下文化两部分，且在受控归纳和少样本设置中，更大的偏差对应更强的对上下文信息的依赖性。
链接: https://arxiv.org/abs/2609.16382
作者: Micah Adler,John W. Byers,Mark Crovella
机构: Boston University (波士顿大学)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:
Abstract:A language model’s representation geometry is not predetermined; it evolves as the model runs. A faithful account of that geometry must capture that dynamic process, and so cannot be based solely on model-independent statistics such as co-occurrence. Here we introduce a mean-field analysis of attention. The average attention from one token to another defines a kernel that carries representations layer to layer and can be iterated through the network to model how the geometry is transformed. We condition this average two ways. Conditioned on a whole corpus, the kernel predicts the average-case evolution of representation geometry. Conditioned instead on a single context, it predicts the expected geometry for that context. A head’s departure from that prediction, its \emphmean-field deviation, isolates the context-specific computation that the mean field misses. Under the corpus-conditional reading, the kernel yields an open-loop model: from the input embeddings and the frozen weights alone, we can iterate the kernel and the model’s own MLPs over token representations, never consulting a measured deviation at any layer. The resulting prediction is highly accurate. In early training the model and its corpus mean field are indistinguishable. Replace every attention head with its mean field, and the substitution leaves the loss on real text unchanged. Around the onset of induction, the two diverge, and the gap widens as representations become contextualized. Under the context-conditional reading, deviation from the mean field is a task-agnostic measure of context-specific computation. The residual decomposes additively into unusual attention routing and contextualization of the transported values. Across controlled induction and few-shot settings, greater deviation tracks greater reliance on in-context information. Subjects: Machine Learning (cs.LG); Computation and Language (cs.CL) Cite as: arXiv:2609.16382 [cs.LG] (or arXiv:2609.16382v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.16382 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Mark Crovella [view email] [v1] Mon, 14 Sep 2026 21:47:30 UTC (647 KB)
[NLP-62] Register Tokens for Bounded-State Reasoning in Diffusion Language Models
【速读】： 该论文旨在解决生成式语言模型在长序列推理任务中因上下文长度限制而难以持续进行跨生成块推理的问题。传统方法依赖保留先前生成文本以维持上下文，但在资源受限或生成长度超过上下文窗口时面临瓶颈。其核心解决方案是引入一种固定大小的“寄存器令牌（register tokens）”机制，即一组位置固定的专用连续向量，通过训练使其在生成过程中携带推理状态信息。在每次生成块结束后，尽管原始生成内容被清除，但寄存器状态得以保留，从而实现对后续生成的指导。实验表明，在LLaDA和Dream基准测试中，该方法显著优于基于离散文本传递的方案，尤其在代码生成等需要多块协同推理的任务中表现突出，数学与代码任务分别取得最高达8.5和19.5的性能提升。此外，寄存器状态可通过强化学习进一步优化，以适应长时序推理需求。
链接: https://arxiv.org/abs/2609.16372
作者: Albert Ge,Chandan Singh,Yufan Zhuang,Xiaodong Liu,Jianfeng Gao,Frederic Sala
机构: University of Wisconsin–Madison(威斯康星大学麦迪逊分校); Microsoft Research(微软研究院); UC San Diego(加州大学圣地亚哥分校)
类目: Computation and Language (cs.CL)
备注:
Abstract:Masked diffusion language models (dLLMs) generate text by iteratively denoising masked tokens with bidirectional attention. Extending reasoning across generation chunks normally requires keeping earlier generated text in context. We ask whether a dLLM can instead continue reasoning after that text is cleared, using only a fixed-size carried state. We implement this state as a small number of register tokens: dedicated fixed-position tokens whose continuous hidden states are trained to carry reasoning progress across generation chunks. We post-train dLLMs to decode a chunk of text, clear it while preserving the register values, and continue decoding from the prompt and carried state. In our main comparisons on LLaDA and Dream, registers outperform discrete-text carry on every benchmark, with gains of up to 8.5 points on math and 19.5 points on code. Registers are especially effective for bounded code generation, where correct programs usually span several chunks. Finally, registers can be further refined with reinforcement learning on long-horizon reasoning tasks.
[NLP-63] StalePO: Anchored Token-Level Preference Optimization using Legacy Post-Edits in Machine Translation
【速读】： 该论文旨在解决机器翻译系统在持续迭代过程中面临的“陈旧偏好信号（Stale Preference）”问题：当前模型性能已超越旧版本，但训练所依赖的偏好数据仍为旧模型输出的人工后编辑（post-edits），导致标准直接偏好优化（DPO）方法失效。其关键解决方案是提出StalePO，一种基于三个核心要求构建的目标函数：1）对两个响应的似然度均需向下调整以避免强化劣质后编辑；2）策略必须锚定于自身基线输出以保持稳定性；3）KL散度约束需在词元（token-level）层面施加以实现局部错误修正的精细控制。这三个机制缺一不可，实验证明单独使用任一机制均无法提升性能，而联合应用可将陈旧反馈转化为实际质量增益。在英-印地语与英-土耳其语本地化任务上，StalePO分别使通过大语言模型作为裁判（LLM-as-judge）MQM质量检查的段落比例提升14.9和4.6个百分点，主要改善风格与流畅性；人类评估进一步验证了在英-印地语任务中13.8个百分点的提升，证实了该方法的有效性。
链接: https://arxiv.org/abs/2609.16340
作者: Rohit Dhaipule,Sukhdeep Singh Kharbanda,Prasanth Bathala,Pradyumna Lanka,Anubhav Shrimal
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: Accepted to the 11th Conference on Machine Translation (WMT 2026)
Abstract:Machine translation systems are periodically upgraded to stronger models, but the available preference signal is human post-edits of an older system’s outputs, which the newer model may already surpass. Moreover, collecting fresh post-edits for every new model is prohibitively expensive. We call this the Stale Preference problem. Standard DPO can fail in this setting: it may increase the likelihood of inferior post-edits, erode the model’s existing quality, and fail to provide the per-token control needed to correct localized errors. We introduce StalePO, an objective derived from three requirements this regime imposes. Likelihood movement must be downward on both responses, the policy must be anchored to its own base response, and the KL constraint must apply at the token level. These requirements are jointly necessary. In ablations, each mechanism in isolation leaves the model’s performance indistinguishable from the base model, and only their combination converts stale feedback into gains. On English-to-Hindi and English-to-Turkish localization data, StalePO improves the fraction of segments passing all LLM-as-judge MQM quality checks by 14.9 and 4.6 percentage points, respectively, with gains concentrated on style and fluency. A human evaluation under the same framework confirms these gains on English-to-Hindi, raising the fraction of segments passing all seven human checks by 13.8 percentage points.
[NLP-64] Efficient One-to-Many Translation with Joint Multi-Stream Diffusion
【速读】： 该论文旨在解决自回归（Autoregressive, AR）模型在执行多语言一对一翻译（one-to-many machine translation）时面临的计算开销大、延迟随目标语言数量线性增长的问题。其核心挑战在于，传统AR系统需逐词生成目标语言译文，导致在处理多语言输出时效率低下。本文提出的解决方案关键在于采用离散扩散（discrete diffusion）框架，通过并行迭代优化所有目标语言的翻译结果，从而实现目标语言数量上的亚线性延迟增长。该方法以连续语义锚点（continuous semantic anchor）作为条件而非源端词元，使模型具备零样本迁移能力，可在未见源语言上保持约75%的监督翻译质量，且无需重新训练。实验表明，在加速采样策略下，该方法在翻译质量上可媲美传统AR基线，同时实现2倍速度提升及11.9%更高的零样本BLEU得分，验证了联合多流扩散机制在高效、灵活的一对多翻译中的巨大潜力。
链接: https://arxiv.org/abs/2609.16312
作者: Yiwen Guan,Jacob Whitehill
机构: Worcester Polytechnic Institute(伍斯特理工学院)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:One-to-many machine translation (MT) is computationally expensive for autoregressive (AR) systems, which suffer from linear latency scaling with both sequence length and the number of target languages. We explore how diffusion can enable multilingual translation with a discrete diffusion framework that refines all target languages in parallel, achieving sublinear latency scaling with the number of targets, and supports deployment as a single unified model to replace multiple independent systems. Conditioned on a continuous semantic anchor rather than source tokens, our framework supports zero-shot transfer to unseen source languages without retraining, maintaining approximately 75% of its supervised translation quality on zero-shot sources. We investigate the quality-latency frontier and find that with accelerated sampling, it achieves comparable supervised quality to AR baselines with a 2 \times speedup and 11.9% better zero-shot BLEU. These results highlight the potential of joint multi-stream diffusion as a practical and flexible alternative for efficient one-to-many translation.
[NLP-65] CLEAR: Cross-Source Evidence Adjudication for Large Language Models in Medicine
【速读】： 该论文旨在解决医学领域大语言模型（LLM）在面对持续演进的医学知识时，其参数化知识（parametric knowledge）固定不变所导致的事实准确性下降问题。尽管外部检索技术（如检索增强生成，RAG）可引入最新证据，但其检索结果常存在相关性差、信息不全或内容冲突等问题，反而损害生成结果的可信度与证据依从性。为此，本文提出一种名为CLEAR的智能体框架，用于实现医学领域中跨源证据的仲裁。其核心在于通过三条互补路径独立生成候选答案：参数化知识、本地整理的语料库以及动态检索的外部证据，全面覆盖常见信息来源。随后，聚合验证模块综合评估各候选答案、支持证据、数据溯源及来源质量，识别不同来源间的一致性与冲突。最终，仲裁模块借助互补覆盖-防护机制与挑战-审计机制，判断当前结论是否应保留或修正；对于未解决的冲突，则触发针对性的后续搜索与重新仲裁，从而实现动态、可靠且可解释的医学推理闭环。
链接: https://arxiv.org/abs/2609.16301
作者: Shuai Wang,Yize Zhao,Qingyu Chen
机构: Yale University (耶鲁大学); Yale School of Public Health (耶鲁大学公共卫生学院); Yale School of Medicine (耶鲁大学医学院)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 31 pages
Abstract:Medical knowledge evolves continuously, whereas the parametric knowledge encoded in large language models (LLMs) is fixed at training time. External retrieval, including retrieval-augmented generation (RAG), can provide access to newly available evidence, but retrieved information may be irrelevant, incomplete, or conflicting. As a result, external retrieval can in turn degrade the factual accuracy and evidence grounding of LLM outputs. To address this challenge, we propose \textbfCLEAR, an agentic framework for cross-source evidence adjudication in LLMs in medicine. CLEAR independently generates candidate answers from three complementary pathways—parametric knowledge, locally curated corpora, and dynamically retrieved evidence—reflecting three common sources of information available to LLMs. An aggregation verifier jointly evaluates the candidates, supporting evidence, provenance, and source-quality information to identify agreement and conflict across sources. An adjudication module then determines whether the current conclusion should be preserved or revised through complementary override-guard and challenge-audit mechanisms, while unresolved conflicts trigger targeted follow-up search and re-adjudication.
[NLP-66] Speaker or Language? Explaining Variance in Charismatic Prosody Across Luxembourgish and French INTERSPEECH2026
【速读】： 该论文旨在解决双语公共演讲中语言与表达风格对魅力型言语（charismatic speech）相对贡献不明确的问题。其核心解决方案在于通过分析10位在卢森堡语和法语间切换的政客在高度可比语境下的自发演讲，从400个语句中提取41个与声学-韵律特征相关的语音魅力指标，并采用混合效应模型分离说话人个体差异与语言类型带来的变异。研究发现，说话人身份解释了绝大部分变异，而语言仍表现出系统性差异：法语表达具有更高的抖动（shimmer）和句末基频（F0），体现礼貌、尊重的声学特征；卢森堡语则呈现更强的中频谱能量，反映更突出的语音存在感。这些模式与卢森堡语作为非正式身份语言、法语作为高声望制度化语言的社会语言学角色相一致。
链接: https://arxiv.org/abs/2609.16275
作者: Nina Hosseini-Kivanani,Nafiseh Taghva,Peter Gilles,Oliver Niebuhr
机构: Radio Télévision Luxembourg (RTL); University of Luxembourg (卢森堡大学); Shiraz University (设拉子大学); University of Southern Denmark (南丹麦大学)
类目: Computation and Language (cs.CL)
备注: 5 pages, 2 figures, 2 tables (Accepted to Interspeech2026)
Abstract:Charismatic speech is shaped by language and speaking style, yet their relative contribution in bilingual public speaking remains unclear. We analyzed spontaneous speeches of 10 politicians who address audiences in Luxembourgish and French, in highly comparable communicative contexts across languages. From 400 utterances, we extracted 41 acoustic-prosodic features linked to vocal charisma and fitted mixed-effects models to separate speaker- and language-related variance. Speaker identity accounted for most variance, whereas language explained less, but still showed systematic differences: French productions showed higher shimmer and phrase-final F0, indicative of a polite, respectful voice, while Luxembourgish productions exhibited stronger mid-frequency spectral energy, suggesting a more vocally present profile. These patterns align with the sociolinguistic roles of Luxembourgish as an informal identity language and French as a high-prestige institutional variety.
[NLP-67] Speaker-Specific and Language-Dependent Temporal Organization in Bilingual Political Speech INTERSPEECH2026
【速读】： 该论文旨在解决多语言环境下政治演讲中语音节奏组织机制的问题，尤其关注讲卢森堡语（Luxembourgish）和法语（French）的双语政治人物如何在公共演讲中调控时间结构。其核心问题是：语言选择对语音节奏的影响是否超越个体说话者特征，以及这种影响在不同性别间的差异性如何。解决方案的关键在于通过定量分析400个句子中的音段与停顿标注数据，计算基于辅音和元音的节奏度量指标（包括均值、变异性及成对变异性指数），并分别量化说话者效应与语言效应，采用配对t检验检验同一说话者在不同语言间的表现差异。研究发现，辅音相关的节奏指标保留了显著的说话者特异性签名，而元音相关的指标则主要由语言选择驱动；法语语料表现出更长且更具变异性元音及元音间隔，而辅音时序差异较小，未发现语言与性别之间的显著交互作用。这表明语言选择会系统性地重构双语公共演讲中的节奏结构，揭示了语言层面在语音韵律组织中的主导作用。
链接: https://arxiv.org/abs/2609.16274
作者: Nina Hosseini-Kivanani,Nafiseh Taghva,Peter Gilles,Oliver Niebuhr
机构: 未知
类目: Computation and Language (cs.CL)
备注: 5 pages, 2 figures, 2 tables (Accepted to Interspeech2026)
Abstract:Speech rhythm helps structure persuasive speech, but most empirical work examines monolingual English. This study asks how politicians organize timing when speaking Luxembourgish and French. We analyze 400 sentences from ten politicians, annotated for segments and pauses. We compute rhythm metrics, including means, variability, and pairwise variability indices for consonants and vowels. We quantify speaker and language contributions and test within-speaker language effects with paired t-tests. Results show that consonant-based metrics retain speaker-specific signatures, whereas vowel-based metrics are largely driven by language choice. French tokens display longer and more variable vowels and vocalic intervals, while consonant timing differences are smaller. No robust language by gender interactions emerge. These findings show that language choice systematically reorganizes rhythmic timing in bilingual public speech.
[NLP-68] Spurious Tool Use: When RL Agents Learn the Wrong Reason to Act
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）代理在使用外部工具（如网络搜索、代码执行等）时，因强化学习（Reinforcement Learning, RL）优化而习得“捷径”工具选择策略的问题。具体而言，当训练数据中存在与特定工具强相关的表面提示线索（spurious prompt cues），但这些线索与工具实际必要性无因果关系时，模型可能仅依据线索而非任务真实需求来决定是否调用工具，从而导致决策偏差。研究通过构建受控的合成环境，结合事实问答与数学推理任务，并在训练中引入此类无关但高度相关的提示线索，发现即使在无需工具的情况下，模型仍会因线索存在而显著增加工具调用率（最高提升39%）。然而，这种捷径行为并非普遍发生，仅在模型已具备可靠使用目标工具的能力后才出现，表明任务熟练度而非数据分布不平衡是捷径学习的关键因素；此外，提示线索与工具之间的语义一致性进一步加剧了该现象。为缓解此问题，作者提出一种基于密集决策层奖励机制的方法，利用一个大语言模型判别器对每次工具调用的必要性进行评估，生成工具必要性奖励（tool-necessity reward），有效抑制了由提示线索驱动的非必要工具调用，同时保持了任务性能，为提升LLM代理工具使用策略的鲁棒性提供了切实可行的解决方案。
链接: https://arxiv.org/abs/2609.16268
作者: Yiwei Yang,Haoxiang Zhang,Bingbing Wen,Yao Lu,Yuchen Wu,Lei Zhang,Julian McAuley,Pan Lu,Bill Howe
机构: University of Washington (华盛顿大学); University of California San Diego (加州大学圣地亚哥分校); Stanford University (斯坦福大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Large language model (LLM) agents increasingly interleave natural language reasoning with external tools such as web search and code execution. These tool-use policies are often optimized via reinforcement learning (RL), which can amplify spurious correlations in the training data. In this work, we study when and why RL-trained agents learn shortcut tool-selection policies: invoking tools based on superficial prompt cues rather than genuine task requirements. We construct controlled synthetic environments combining factual question answering and mathematical reasoning tasks, and inject cues that are strongly correlated with specific tools during training but causally irrelevant to tool necessity. Across counterfactual evaluations where cues are present but the associated tools are not required, agents exhibit substantial shortcut behavior, with spurious tool invocation rates increasing by up to 39 percent. However, shortcut formation is not universal: across the conditions we test, it arises only when the agent has already learned to use the target tool reliably, suggesting that task competence, rather than dataset imbalance alone, is a key factor in shortcut learning. A swapped-cue analysis further shows that semantic alignment between cues and tools substantially amplifies this effect. To mitigate these failures, we introduce a dense, decision-level reward in which an LLM judge evaluates the necessity of each tool call. This tool-necessity reward effectively suppresses cue-driven tool use while preserving task performance, providing a practical approach to improving the robustness of LLM agent tool-use policies.
[NLP-69] he record is part of the task: matched-record evaluation of text classifiers across maintenance safety and recall reporting
【速读】： 该论文旨在解决在自然语言处理模型评估中，由于同一操作案例在不同工作流程阶段和目的下被多次记录，而传统评估方法仅选择其中一条记录进行比较所导致的评估偏差问题。其核心解决方案在于将记录的选择过程本身纳入评估框架，通过在固定标签与数据划分条件下，对同一案例的匹配记录进行系统性对比分析。研究选取了通用电气航空（GE Aerospace）维修事件、美国国家航空航天局安全报告系统（NASA ASRS）安全报告以及美国国家公路交通安全管理局（NHTSA）车辆召回数据三个场景进行验证，发现当标签由部件交易独立生成时，未见训练集的宏平均F1得分在0.33至0.91之间波动；尤为关键的是，客户报告（工作前撰写）与技术人员报告（诊断后但交易前撰写）之间的性能差异达0.46，显著超过模型架构与表示方式差异的影响。此外，在公开系统中，不同记录类型的表现也呈现依赖性：NHTSA缺陷摘要在各类模型中始终表现最优，而ASRS分析师摘要在序列学习模型下优于原始报告者叙述，但在词袋基线模型中则不具优势。研究进一步表明，部分模型比较结果亦受具体记录选择影响。因此，论文强调模型评估应基于实际决策时刻可获取的信息，并明确报告记录与标签的生成方式，以提升评估的可复现性与有效性。
链接: https://arxiv.org/abs/2609.16267
作者: Hisham Ihshaish,Peter Mayhew,Tasnim M. A. Zayet,Ana Del Amo
机构: University of the West of England, Bristol, UK(西英格兰大学，布里斯托尔，英国); Birzeit University, Birzeit, Palestine(比尔谢巴大学，比尔谢巴，巴勒斯坦); GE Aerospace, Cheltenham, UK(通用航空，切尔滕纳姆，英国); GE Aerospace, Clearwater, FL, USA(通用航空，佛罗里达州克莱尔沃特，美国)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: 37 pages (18-page article, 3 tables, 7 figures, plus a 19-page supplement with tables and figures numbered S1 onward)
Abstract:Many operational cases are documented more than once, at different workflow stages and for different purposes, yet model evaluations normally select one of these records before model comparison begins. We treat that selection as part of the evaluation and compare matched records of the same cases under fixed labels and splits in three systems: GE Aerospace repair events, NASA ASRS safety reports and NHTSA vehicle recalls. Across the three GE fields, for events whose label comes from parts transactions independently of the narratives, held-out macro-F1 ranged from 0.33 to 0.91. A difference of 0.46 separated the customer report, written before shop work, from the technician report, written after diagnosis but before the transaction that generates the label. That difference is substantially larger than the representation and architecture differences tested on the same events. The public systems showed different patterns: the NHTSA defect summary remained strongest under every model family tested, whereas the ASRS analyst synopsis outperformed the reporter narrative under learned sequence models but not under lexical baselines. Secondary analyses showed that some model comparisons were also record-dependent. Evaluations should be run on the information available at the intended decision point and should report how both the record and the label were produced.
[NLP-70] st-Time Unlearning via Sparse Autoencoder
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）在无需从头训练的情况下，有效且安全地移除特定知识的难题。现有方法通过梯度上升等权重调整策略实现遗忘，但存在显著的“遗忘-性能权衡”问题：过度遗忘会损害模型通用能力，且在后续微调或提示攻击下，被移除的知识可能重新出现。为此，论文提出ARIA（自编码器门控推理时遗忘机制），一种基于测试时干预的遗忘方法，其核心在于保持模型权重不变，仅在生成过程进入与遗忘相关状态时，通过稀疏自编码器（Sparse Autoencoder, SAE）提取的潜在表示触发轻量级线性检测器，并施加可解释的干预操作，从而在不增加推理开销的前提下精准控制有害知识的访问。实验表明，ARIA在多个基准（TOFU、R-TOFU、WMDP）上显著改善了遗忘与保留之间的平衡，在思维型模型（DeepSeek-R1-Distilled-Qwen-1.5B）和指令型模型（Gemma-3-1B-it）上均实现了对目标知识集的高效遗忘（如WMDP-cyber遗忘集准确率大幅下降），同时维持关键任务性能（如MMLU指标仅下降小于1%）。此外，针对权重空间与解码空间恢复的三类后遗忘对抗攻击测试显示，ARIA具有强鲁棒性（遗忘效果变化低于1%）。进一步的特征层面案例分析揭示，部分保留性能下降可能源于未删减的响应风格偏差，而非知识泄露本身，提示当前遗忘任务构建中可能存在隐含偏见。
链接: https://arxiv.org/abs/2609.16229
作者: Pingzhi Li,Jinhao Duan,Vaishnav Tadiparthi,Nakul Agarwal,Kwonjoon Lee,Ehsan Moradi Pari,Hossein Nourkhiz Mahjoub,Sijia Liu,Tianlong Chen
机构: University of North Carolina at Chapel Hill(北卡罗来纳大学教堂山分校); Honda Research Institute(本田研究机构); Michigan State University(密歇根州立大学)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:
Abstract:Machine unlearning aims to remove specific knowledge from a trained large language model (LLM) without retraining from scratch. Existing methods modify model weights via gradient ascent and its advances. While effective on certain benchmarks, these weight-based approaches exhibit a sharp forget-utility trade-off, where stronger forgetting of target knowledge can degrade model utility, and unlearned knowledge may reappear under post-unlearning fine-tuning or prompt attacks. We propose ARIA (autoencoder-gated inference-time unlearning), a test-time unlearning method that leaves model weights intact and gates access to unwanted knowledge only when generation enters a forget-related state. ARIA uses sparse autoencoder (SAE) latents to train a lightweight linear detector, then applies an interpretable intervention on triggered states with negligible test-time overhead. Empirical evaluations on TOFU, R-TOFU, and WMDP show that ARIA improves the forget-retain trade-off over weight-based baselines across both a thinking model (DeepSeek-R1-Distilled-Qwen-1.5B) and an instruction model (Gemma-3-1B-it), e.g., reducing WMDP-cyber forget-set accuracy significantly while keeping MMLU within 1% of the pre-unlearning model. We further introduce three post-unlearning adversarial attacks targeting weight-space and decoding-space recovery, and find that ARIA remains robust under all three, with forgetting changing by less than 1% under attack. A feature-level case study leveraging the interpretability of ARIA suggests that some retain degradation may reflect response styles underlying the unlearning data rather than leakage of the targeted knowledge itself, highlighting a potential source of bias in unlearning task construction.
[NLP-71] Decoy Direction Optimization: A Post-Hoc Defense Against LLM Abliteration
【速读】： 该论文旨在解决开放权重语言模型中安全防护机制（safety guardrails）易被拒绝特征消融（Refusal Feature Ablation, RFA）攻击绕过的难题。RFA通过识别并投影残差流中的线性拒绝方向，可在不损害模型能力的前提下实现高成功率的越狱攻击，而现有防御手段通常依赖于针对每个新检查点进行昂贵的安全微调（safety fine-tuning），成本高昂且难以快速部署。本文提出一种高效、后置式的权重编辑防御方法——诱饵方向优化（Decoy Direction Optimization, DDO），其核心创新在于：攻击者依赖对比估计器来定位拒绝方向，而非直接破坏真实安全电路；因此，DDO不试图隐藏真实拒绝路径，而是主动在MLP神经元中注入高幅值、非线性的诱饵信号，使攻击者的估计器因干扰而误判，导致其错误地消融一个无害的正交特征，而真实的保护机制得以保留。理论层面，作者证明了该效应的谱界（spectral bound）以支持其有效性。实验表明，DDO在六种模型家族上将标准RFA攻击成功率降至10%以下；在Llama-3-8B-Instruct上，面对自适应多阶段攻击仍可保持65%的最低攻击成功率（优于基线的58%），并将Heretic级别的权重级攻击成功率从88.7%降低至18%，且每配置的优化成本仅为训练型防御的1/30至1/450，显著提升了防御效率与实用性。
链接: https://arxiv.org/abs/2609.16204
作者: Aashiq Muhamed,Mona T. Diab,Virginia Smith
机构: Carnegie Mellon University (卡内基梅隆大学)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL); Cryptography and Security (cs.CR)
备注:
Abstract:Safety guardrails in open-weight language models can be readily bypassed using Refusal Feature Ablation (RFA), a technique that identifies and projects out a linear refusal direction from the residual stream, often achieving a high attack success rate (ASR) while preserving model capability. Defending against these attacks typically requires computationally expensive safety finetuning for every new checkpoint. We introduce Decoy Direction Optimization (DDO), a fast, post-hoc weight-editing defense that requires no base-model finetuning. Our approach is based on a simple mechanistic insight: ablation attacks rely on contrastive estimators to find the refusal direction. Rather than trying to hide the true refusal circuitry, DDO actively injects a high-magnitude, nonlinear decoy signal into the network’s MLP neurons. When an attacker attempts to locate the refusal direction, the decoy corrupts their estimator, tricking them into ablating a harmless orthogonal feature while the actual safety mechanism remains intact. We prove a spectral bound formalizing this effect and evaluate DDO across six model families, achieving 10% ASR under standard RFA. On Llama-3-8B-Instruct, DDO remains comparable to trained defenses under adaptive multi-phase attacks (65% vs. 58% worst-case ASR) and reduces Heretic weight-level attack ASR from 88.7% to 18%, all at 30 to 450 times lower optimization cost per configuration than the trained baselines.
[NLP-72] Anatomy of Associative Recall in Fixed-State Recurrences: A Matched-State Decomposition an Interference Wall and a Curriculum That Breaks It
【速读】： 该论文旨在解决固定状态递归模型（如线性注意力与状态空间模型）在关联回忆任务中表现落后于标准注意力机制的问题，其核心挑战在于：现有全架构对比无法明确区分是递归结构本身、卷积组件还是其他设计因素导致性能下降。论文通过三轴单变量分解（短因果卷积、转移结构[秩1 Δ规则 vs. 对角]、衰减机制），在固定状态预算下对掩码多查询回忆任务进行系统分析。关键发现为：因果卷积是决定性因素——在两种模型家族中，配备卷积的单元相比无卷积版本显著提升约0.5的回忆准确率，而此前认为“递归结构缺陷”所致的性能差距实则主要源于缺失卷积。此外，秩1转移结构虽在无卷积条件下优于对角结构（+0.19/+0.32），但一旦引入卷积，优势迅速缩小至+0.03，且状态匹配的Mamba-2已可超越无卷积的秩1单元，表明“类别优势”不成立。进一步揭示，模型在高负载下虽能平稳退化，但在稀疏监督下易受干扰，其根本瓶颈并非容量限制，而是训练初始化的“锁定效应”（lock-in lottery）。通过引入距离课程学习（distance curriculum），可在不改变架构的前提下将性能从0.021提升至1.000；尤其在长序列（L=256）时，梯形课程可重启被均匀课程阻断的边界，而在更长序列（L=512）中，仅当课程基于测量精度动态开启门控时，才能实现100%成功锁定。实验还表明，双向去噪器并无显著优势，且碰撞键检索需双层结构。最后，通过在状态追踪护栏（S₅ guardrail）上“武装”模型，无需额外成本即可在所有深度显著提升回忆性能（p=0.0044）。综上，论文将“递归模型不擅长回忆”的笼统结论替换为可量化的分解分析，并提出两个低成本有效干预策略：引入因果卷积与采用基于精度感知的动态课程学习。
链接: https://arxiv.org/abs/2609.16183
作者: Julian Boesch,Andrew Wee
机构: Purdue University (普渡大学); Obit Research
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: 14 pages, 2 figures, 6 tables. Preprint of preliminary results; code and result JSONs at this https URL
Abstract:Fixed-state recurrences–linear attention and state-space models–are reported to lag behind attention on associative recall, but whole-architecture comparisons cannot say which ingredient is responsible. We decompose masked multi-query recall at a fixed state budget along three single-knob axes: a short causal convolution, the transition structure (rank-1 delta rule vs. diagonal), and decay. The convolution dominates (~+0.5 recall in both families under matched training): comparisons that pit convolution-free cells against a convolution-equipped Mamba measure the missing convolution, not the recurrence. The rank-1 transition beats its diagonal ablation by +0.19/+0.32 at 16/32 pairs, but the margin shrinks to +0.03 once both cells carry the convolution, and a state-matched Mamba-2 ties the unarmed rank-1 cell: no class claim survives. Cells that solve 32-pair recall degrade gracefully with load yet fall to chance retrieving 4 pairs from a distractor haystack–flat across lengths and transitions. Interference under sparse supervision, not capacity: a distance curriculum takes the unchanged architecture from 0.021 to 1.000. Training is a lock-in lottery–a seed either locks in or does not–and the curriculum is the lever. Lock-in rises from 1/10 to 7/10 (p=0.02); dense supervision adds nothing; at L=256 a shaped ramp reopens a boundary the uniform curriculum cannot (4/5 vs. 0/9); and at L=512, where the ramp collapses (0/6), gating it on measured accuracy locks in 6/6 (p=0.001). Bidirectional denoiser cells, reading the query before the haystack, show no measurable advantage over causal training (ten seeds), and collision-key retrieval needs two layers. Arming for recall is free on an S_5 state-tracking guardrail–the armed cell is significantly better at every depth (p=0.0044). These replace “recurrent models are bad at recall” with a measured decomposition and two cheap interventions.
[NLP-73] Z-Loss Backward Geometry in Dense Output Heads and Sparse Routers
【速读】： 该论文旨在解决生成式语言模型中因大词汇量softmax输出头与稀疏专家混合（Mixture-of-Experts, MoE）路由机制所引发的训练不稳定性问题，具体表现为对数空间（logit space）中极端值的剧烈波动、有限精度计算下的舍入误差暴露以及训练损失发散。传统Z-loss被视作对softmax归一化因子的标量惩罚，但本文从反向传播视角重新剖析其作用机制，揭示其核心在于反向传播路径中梯度的传输特性。关键创新在于提出“反向传输视角”（backward-transport view），将Z-loss的影响解耦为两个独立部分：一是位于对数空间的梯度源（backward source）的标量幅值与softmax形状，二是由架构和优化器决定的梯度传输因子，包括通用偏移坐标、共享嵌入路径、输出到隐藏层增益、融合损失源一致性、面向优化器的更新方式以及top-k路由的缩减尺度等。该分析表明，即使前向Z-loss值相同，不同架构或实现下其反向梯度及最终参数更新仍可能显著差异；同时解释了为何原始对数空间的Z-loss可有效抑制标量尾部而不改变输出-隐藏增益，且活跃路由压缩会改变有效路由系数。在GPT-2与Pythia系列模型于WikiText-103和FineWeb-Edu数据集上的实验验证表明，考虑架构特性的变体能有效减少反向几何尾部，同时在低系数场景下保持相近的验证困惑度。
链接: https://arxiv.org/abs/2609.16179
作者: Bum Jun Kim
机构: 未知
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: 30 pages, 2 figures
Abstract:Z-loss has been widely applied to the logits of language-model output heads and sparse mixture-of-experts routers. Z-loss constrains the softmax log-normalizers of these output heads and routers, thereby limiting large-logit excursions, reducing finite-precision roundoff exposure, and avoiding training-loss divergence. These use cases arise in modern Transformer settings where large-vocabulary softmax heads, top- k routing, fused losses, and mixed-precision optimizers interact. Z-loss has typically been understood only as a scalar penalty on the log-normalizer. This paper instead analyzes Z-loss from a backward-pass perspective, focusing on the gradients produced by the Z-loss penalty. The logit-space gradient, which we call the backward source, is injected at the logit boundary of the Z-loss branch of backpropagation; consequently, the backward source’s effect depends on the architecture and implementation through which the gradient is transported. We develop a backward-transport view for Z-loss that separates the source’s scalar amplitude and softmax shape from the transport factors. These factors include common-shift coordinates, tied-embedding pathways, output-to-hidden gain, fused-loss source consistency, optimizer-facing updates, and top- k router reduction scale. These diagnostics show that nearly identical forward Z-loss values can coexist with distinct logit-space Z-loss gradients and, after architectural and optimizer transport, distinct parameter updates. The transport diagnostics also explain why raw-logit Z-loss can reduce scalar tails without changing output-to-hidden gain and why active-route reductions alter the effective router coefficient. Across evaluations of models in the GPT-2 and Pythia families on WikiText-103 and FineWeb-Edu, architecture-aware variants reduce backward-geometry tails while maintaining comparable validation perplexity in low-coefficient regimes.
[NLP-74] Safe Error Correction for Language Models: Frozen-Base Adjustment with Capability Preservation
【速读】： 该论文旨在解决在不损害预训练语言模型（如Gemma 4 E2B）原有能力的前提下，如何通过轻量级修正模块有效纠正其输出错误这一实际问题。其核心挑战在于避免传统微调方法中常见的“纠错-能力退化”权衡（correction-capability tradeoff）。解决方案的关键在于提出CRN v2，一个仅约3400万可训练参数（占基础模型4.65B参数的0.73%）的轻量级对数空间（logit-level）修正模块，该模块完全独立于冻结的基础模型运行。该模块通过监督微调（SFT）与无参考的直接偏好优化（DPO）联合训练，在83,400个纠错样本上实现精准校正；同时引入KL散度保持项（KL preservation term, λ=0.1），以稳定基础模型的原始分布，防止能力退化。实验表明，CRN v2在CEHRI领域考试（涵盖事实、算术与隐含目标推理）中成功修正了53.3%的基线模型错误，且在MMLU/BoolQ等基准测试中未出现性能下降。对比分析显示，其他替代方案（如低秩适配器LoRA、隐藏状态注入、多层修正等）均未能超越该结果，凸显了“冻结主干 + 对数空间修正 + KL锚定”的设计原则的有效性，而非架构本身的创新。
链接: https://arxiv.org/abs/2609.16145
作者: Gautam Kishore
机构: Eulogik; github.com/eulogik (GitHub); huggingface.co/eulogik (Hugging Face)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
备注: 10 pages, 4 tables. Code, weights, and evaluation scripts: this https URL and this https URL
Abstract:We study a practical question: can a small correction module fix errors in a frozen language model’s outputs without degrading its base capabilities? We propose CRN v2, a lightweight logit-level correction module (~34M trainable parameters, 0.73% of the 4.65B text module) that sits atop a fully frozen Gemma 4 E2B model. The base model is never updated; only the correction module learns, via supervised fine-tuning followed by reference-free DPO on 83,400 error-correction pairs. On a 60-question domain exam (CEHRI: Certified Human-Robot Intelligence, covering facts, arithmetic, and implicit-goal reasoning), CRN v2 corrects 53.3% of base-model errors (reworded variant: 43.3%) while showing no degradation on tested capability benchmarks (MMLU/BoolQ N=200; car-wash N=8). A LoRA baseline at the matched CRN v1 budget (6.6M params, rank 19) achieves 83.3% correction but suffers 30-75% capability loss on the same benchmarks – the correction-capability tradeoff. An ablation shows that the KL preservation term (lambda=0.1) is critical: lowering it to 0.01 degrades correction to 35.0%. A hidden-state injection variant at earlier layers (1.6M params, SFT-only) reaches 50.0%/55.8% but does not exceed logit correction; shallower injection (layer 4) drops to 30.0%/28.3%; multi-depth logit correction (~35M) reaches only 40%; and longer training (5,000 SFT + 2,000 DPO) stays at 53.3% – none of the alternative configurations we tested exceeded the rank-128 logit result, consistent with a best-achieved result of ~53% rather than a floor. This is a study of a design principle (frozen base + logit correction + KL anchoring), not a claim of architectural novelty. All code, main-result weights, and evaluation scripts are released (deep variant as code only – no trained deep checkpoints).
[NLP-75] Universal Defenses for Tool-Integrated LLM Agents Against Adversarial Attacks
【速读】： 该论文旨在解决集成外部工具的大型语言模型代理（LLM agents）在多步骤任务执行中面临的多种对抗性攻击问题，包括直接提示注入、间接提示注入、记忆污染及后门攻击等。这些问题源于模型对提示注入和工具操纵的高度敏感性，严重威胁其安全性与可靠性。论文提出一种统一框架下的通用且可泛化的防御策略，其关键在于构建多层次、模块化、基于工具的防御机制：一是“攻击者工具过滤”（Attacker Tool Filtering），通过异常检测技术（如孤立森林）识别并剔除可疑工具；二是“正常工具召回”（Normal Tool Recalling），采用白盒方法在规划前恢复代理原始工具集，以消除恶意干扰。此外，结合基于提示的防御手段，包括思维链（Chain-of-Thought）提示与自我反思技术，提升推理鲁棒性，并通过任务重述降低攻击面。实验结果表明，该方法在四类开源模型（Gemma2-9B、Qwen2-7B、LLaMA3-8B、LLaMA3.1-8B）及三类专有模型（GPT-3.5、GPT-4、GPT-5）上均显著降低攻击成功率（ASR），多数场景下实现0%攻击成功率，同时保持或提升了原任务成功率，验证了简单、模块化、多层防御在增强工具集成型LLM代理安全性和鲁棒性方面的有效性。
链接: https://arxiv.org/abs/2609.16098
作者: Xiaoyan Li,Yunli Wang
机构: University of Toronto (多伦多大学); National Research Council Canada (加拿大国家研究委员会)
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Large Language Model (LLM) agents have demonstrated impressive capabilities across a variety of domains, particularly when integrated with external tools for multi-step task completion. However, they are increasingly vulnerable to adversarial attacks, including direct prompt injection, indirect prompt injection, memory poisoning, and backdoor attacks, which exploit the model’s openness to prompt injection and tool manipulation. In this work, we explore practical and generalizable defense strategies within a unified framework across these four attack types. We introduce two universal tool-based defenses: Attacker Tool Filtering, which uses anomaly detection (e.g., Isolation Forest) to identify and remove suspicious tools, and Normal Tool Recalling, a white-box method that restores the agent’s original toolset prior to planning. Additionally, we incorporate prompt-based defenses: Chain-of-Thought prompting and self-reflection techniques to enhance reasoning and task paraphrasing to mitigate attacks. Experimental results across both four open-source LLMs (Gemma2-9B, Qwen2-7B, LLaMA3-8B, and LLaMA3.1-8B) and three proprietary LLMs (GPT-3.5, GPT-4, and GPT-5) show that our methods significantly reduce the Attack Success Rates (ASR), achieving 0% ASR in many settings, while preserving or even improving the original task success rate. These findings highlight the promise of simple, modular, multi-layered defenses for strengthening the security and robustness of tool-integrated LLM agents. The code is available at this https URL.
[NLP-76] RAG -CT: Mitigating Privacy Risks on Retrieval-Augmented Generation Systems via Scanning Prompt Distribution
【速读】： 该论文旨在解决生成式 AI（Generative AI）在检索增强生成（Retrieval-Augmented Generation, RAG）架构中面临的隐私安全问题，即攻击者可通过恶意查询利用检索过程从外部知识库中泄露个人身份信息（Personally Identifiable Information, PII）。现有方法在应对此类攻击时存在防御能力不足或依赖模型修改的局限性。本文提出一种轻量级且高效的防御机制 RAG-CT，其核心在于通过分析恶意查询的熵（entropy）分布与边际（margin）分布特征，并结合基于得分的检测策略，实现对潜在泄露性查询的精准识别。实验结果表明，RAG-CT 在四种主流攻击策略和两个数据集上均显著降低 PII 泄露风险，同时优于现有多种防御基线，且无需对底层大语言模型（LLM）或检索器进行任何修改，具备良好的实用性与可部署性。
链接: https://arxiv.org/abs/2609.16095
作者: Xingyu Lyu,Jiayimei Wang,Jianfeng He,Ning Wang,Yidan Hu,Yimin Chen
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注:
Abstract:Retrieval-Augmented Generation (RAG) has emerged as a powerful paradigm for improving the quality of generated contents of Large Language Models (LLMs) by grounding responses in external knowledge, thus reducing hallucinations and factual errors. However, recent studies have highlighted a critical vulnerability: adversaries can exploit the retrieval process to extract personally identifiable information (PII) from the underlying corpus. To mitigate this risk, we propose a novel defense, RAG-CT, that identifies malicious queries by analyzing their entropy and margin distributions and using a score-based detection method. Extensive experiments with four state-of-the-art attack strategies and four defense baselines on two datasets show that our approach significantly reduces PII leakage while outperforming existing defenses. This work provides a lightweight yet effective mechanism to protect RAG systems against PII leakage without requiring modifications to the underlying LLM or retriever.
[NLP-77] he Imitation Game: When LLM s Learn to Reason Like Programs via Code-Centric Reasoning Data Synthesis EMNLP26
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在自然语言中进行确定性、细粒度推理时表现不佳的问题，其核心瓶颈在于模型过度依赖语义近似而非严格的符号执行。为此，论文提出MIMIC框架，其关键在于利用可执行代码作为严谨的推理数据生成媒介，通过叙事融合（narrative fusion）、代码引导的测试用例合成（code-guided test synthesis）以及动态代码插桩（dynamic code instrumentation）将算法转化为可验证的推理轨迹。这一过程生成的显式中间执行状态构成代码插桩奖励（Code-Instrumented Reward, CIR），为强化学习提供了密集且高保真的过程监督信号，无需依赖外部奖励模型。实验表明，基于该合成数据集通过监督微调（Supervised Fine-Tuning, SFT）和基于奖励策略优化（Generalized Reward Policy Optimization, GRPO）训练的模型，在通用推理、复杂数学基准及细粒度确定性任务上均取得显著且一致的性能提升，证明了可执行代码的程序严谨性能够有效激发并增强大语言模型的泛化推理能力。
链接: https://arxiv.org/abs/2609.16076
作者: Jinyang Zhang,Weibin Liao,Keqin Bao,Sihang Li,Shaobo Wang,Muyang Ye,Hongxin Ding,Yue Fang,Tianyi Tang,Fei Huang,Kexin Yang,Xingzhang Ren,Dayiheng Liu
机构: Peking University (北京大学); Alibaba Group (阿里巴巴集团); Shanghai Jiao Tong University (上海交通大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Accepted by EMNLP26 main
Abstract:Large Language Models (LLMs) excel at programming tasks but frequently fail at deterministic, fine-grained reasoning in natural language, relying heavily on semantic approximations rather than robust symbolic execution. To bridge this gap, we propose MIMIC, a framework that leverages executable code as a rigorous medium for reasoning data synthesis. MIMIC fundamentally transforms algorithms into verifiable reasoning trajectories through narrative fusion, code-guided test synthesis, and dynamic code instrumentation. Crucially, these explicit intermediate execution states naturally form a Code-Instrumented Reward (CIR), providing dense, high-fidelity process supervision for reinforcement learning without external reward models. Extensive evaluations reveal that models trained via SFT and GRPO on our synthesized dataset achieve substantial, consistent gains. Our method significantly elevates accuracy across general reasoning, complex mathematical benchmarks, and fine-grained deterministic tasks, demonstrating that the procedural rigor of executable code can effectively unlock and enhance the generalized reasoning capabilities of LLMs. Our code and data are available at this https URL.
[NLP-78] he Immutable Past: Formalizing State Mutability and Conflict Resolution in Mutable RAG
【速读】： 该论文旨在解决长时程自主智能体中基于检索增强生成（Retrieval-Augmented Generation, RAG）的共享记忆架构所面临的“语义阴影”（Semantic Shadowing）问题。该问题源于将共享记忆视为仅可追加的流式结构，导致冲突的历史观测不断累积，并在统计上压制有效的近期更新，进而在动态环境中引发严重的状态偏离，使智能体基于过时事实进行推理与决策。针对此问题，论文提出的关键解决方案是引入GC-Mem（Memory Garbage Collection），一种严格的推理时一致性协议。其核心机制在于采用纯时间主导算子（Φₜ）结合矛盾检测，实现对冗余或冲突上下文的精准剔除，而非依赖启发式的时序衰减策略——后者会无差别地破坏有效长期记忆。通过在包含137,760个记忆片段的严格行为推断基准上进行连续累积测试，实验表明标准RAG及基于时间戳重排序的基线方法均出现显著性能退化，而GC-Mem则实现了90%的冲突解决准确率，并确立了可确保状态收敛的精确度与召回率部署阈值，从根本上克服了传统可变记忆RAG的固有缺陷。
链接: https://arxiv.org/abs/2609.16073
作者: Hamed HaddadPajouh,Amir AmiriTabat
机构: Independent Researcher(独立研究员); Independent Researcher(独立研究员)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Retrieval-Augmented Generation (RAG) serves as the primary memory architecture for long-horizon autonomous agents. However, treating shared memory as an append-only stream introduces \textitSemantic Shadowing, a critical failure mode where conflicting historical observations accumulate and statistically dominate valid recent updates. In dynamic environments, this results in severe state divergence as agents retrieve and act upon obsolete facts. This paper formalizes the mechanics of State Mutability to prove that standard dense retrieval suffers from Asymptotic Recall Decay. Furthermore, we formally demonstrate a Majority Vote Trap, revealing that increasing the retrieval context window paradoxically degrades generation accuracy by diluting the attention mechanism under conditions of semantic equivalence. To resolve this, we introduce GC-Mem (Garbage Collection for Memory), a strict inference-time consistency protocol. Unlike heuristic time-decay mechanisms—which indiscriminately destroy valid long-term memory—GC-Mem relies purely on a temporal dominance operator ( \Phi_\mathcalT ) paired with contradiction detection to surgically excise shadowed context. Evaluated across a rigorous, behaviorally inferred benchmark of 137,760 memory chunks and continuous accumulation sweeps, standard RAG and timestamp re-ranking baselines experience severe degradation. In contrast, GC-Mem empirically recovers 90% conflict resolution accuracy. We establish strict precision and recall deployment thresholds, ensuring state convergence where standard mutable RAG fundamentally fails.
[NLP-79] Efficient Multimodal Generative Recommendation with Latent Narrative Reasoning
【速读】： 该论文旨在解决生成式推荐在短剧续写（short-form drama continuation）场景中面临的双重挑战：一方面，传统生成式推荐依赖用户偏好进行物品预测，而短剧续写任务的输出由叙事演进（narrative evolution）驱动，与用户偏好解耦；另一方面，多模态内容中的冗余视觉上下文导致计算效率低下，且显式的推理生成过程成本高昂。其解决方案之关键在于提出NarraLite框架，通过联合压缩感知与推理过程实现高效建模：首先，采用渐进式谱压缩（Progressive Spectral Compression），将长序列视觉上下文有选择性地压缩为紧凑的、与叙事相关的关键证据，保留叙事转折关键信息的同时减少冗余视觉计算；其次，引入潜在叙事推理（Latent Narrative Reasoning），通过上下文路由的隐式推理令牌，将其上下文表示对齐至未来剧情延续语义，实现无需自回归解码文本推理理由的隐式叙事推断。此外，研究构建了面向UGC、PGC及OOD场景的无用户依赖多模态短剧续写基准，实验表明NarraLite在续写准确性、叙事连贯性与鲁棒性方面均优于现有方法，并实现了良好的准确率-效率权衡。
链接: https://arxiv.org/abs/2609.16070
作者: Chenxing Wang,Nantao Zheng,Hao Miao,Juyuan Wang,Xinke Jiang,Yuchen Fang,Aolin Li,Haijun Wu
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Generative recommendation reformulates item prediction as semantic identifier generation, yet episodic content introduces a fundamentally different setting where the target is determined by narrative evolution rather than user preference. This task requires models to understand multimodal storyline progression while addressing the efficiency challenges caused by redundant visual contexts and costly explicit reasoning generation. We propose \textbfNarraLite, an efficient multimodal generative recommendation framework that jointly compresses perception and reasoning. Specifically, Progressive Spectral Compression selectively distills long visual contexts into compact narrative-relevant evidence, preserving transition-critical information while reducing redundant visual computation. Latent Narrative Reasoning introduces context-routed latent reasoning tokens and aligns their contextualized representations with future continuation semantics, enabling implicit narrative inference without autoregressively decoding textual rationales. We further establish a user-agnostic multimodal benchmark for short-form drama continuation across UGC, PGC, and OOD settings. Extensive experiments demonstrate that NarraLite consistently improves continuation accuracy, narrative coherence, and robustness over existing approaches, while achieving a favorable accuracy–efficiency trade-off.
[NLP-80] POSPAN: Position-Constrained Span Masking for Language Model Pre-training
【速读】： 该论文旨在解决现有跨度级掩码语言模型（span-level masked language modeling, MLM）在掩码跨度位置分布上过于简化的问题，即以往方法仅考虑离散的跨度长度分布，而忽略了跨度之间的位置依赖关系，通常假设被掩码跨度的位置服从均匀分布。其核心解决方案是提出POSPAN框架，通过将跨度长度分布与位置约束分布相结合，实现多样化的、受位置约束的跨度掩码策略，从而统一现有各类跨度级掩码方法。实验结果表明，引入位置约束能显著提升跨度级掩码的效果，最优的POSPAN设置在多个自然语言理解（NLU）基准数据集上均优于仅基于跨度长度或原始单标记掩码（vanilla MLM）的方法。此外，论文还进行了理论分析，揭示了位置约束在掩码语言模型中的合理性与必要性，为POSPAN的有效性提供了理论支持。
链接: https://arxiv.org/abs/2609.16061
作者: Zhenyu Zhang,Lei Shen,Yuming Zhao,Meng Chen,Xiaodong He
机构: JD AI Research(京东人工智能研究院); China(中国)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Span-level masked language modeling (MLM) has shown to be advantageous to pre-trained language models over the original single-token MLM, as entities/phrases and their dependencies are critical to language understanding. Previous works only consider span length with some discrete distributions, while the dependencies among spans are ignored, i.e., assuming that the positions of masked spans are uniformly distributed. In this paper, we present POSPAN, a general framework to allow diverse position-constrained span masking strategies via the combination of span length distribution and position constraint distribution, which unifies all existing span-level masking methods. To verify the effectiveness of POSPAN in pre-training, we evaluate it on the datasets from several NLU benchmarks. Experimental results indicate that the position constraint is capable of enhancing span-level masking broadly, and our best POSPAN setting consistently outperforms its span-length-only counterparts and vanilla MLM. We also conduct theoretical analysis for the position constraint in masked language models to shed light on the reason why POSPAN works well, demonstrating the rationality and necessity of POSPAN.
[NLP-81] HintMiner: Automatic Question Hints Mining From QA Web Posts with Language Model via Self-Supervised Learning
【速读】： 该论文旨在解决用户在技术问答平台（如Stack Overflow）中提问后，难以及时获得有效回答的问题。现有QA论坛常因响应延迟或答案质量不足导致用户困惑。为应对这一挑战，本文提出一种名为HintMiner的自动化问题提示挖掘工具，其核心解决方案是利用生成式AI（Generative AI）技术，基于大规模在线问答数据自动为用户问题生成精准提示。关键在于构建一个基于Transformer架构并融合复制机制的神经编码器-解码器模型——MiningNet，通过自监督学习目标训练该模型，使其能够从海量网络问答内容中检索并提取与问题相关的有效线索。实验表明，该方法在60,000个Stack Overflow问题上表现优异，平均BLEU得分为36.17%，平均ROUGE-2得分为36.29%，验证了其有效性与实用性。
链接: https://arxiv.org/abs/2609.16060
作者: Zhenyu Zhang,JiuDong Yang
机构: 未知
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Users often need ask questions and seek answers online. The Question - Answering (QA) forums such as Stack Overflow cannot always respond to the questions timely and properly. In this paper, we propose HintMiner, a novel automatic question hints mining tool for users to help them find answers. HintMiner leverages the machine comprehension and sequence generation techniques to automatically generate hints for users’ questions. It firstly retrieve many web Q\A posts and then extract some hints from the posts using MiningNet that is built via a language model. Using the huge amount of online Q\A posts, we design a self-supervised objective to train the MiningNet that is a neural encoder-decoder model based on the transformer and copying mechanisms. We have evaluated HintMiner on 60,000 Stack Overflow questions. The experiment results show that the proposed approach is effective. For example, HintMiner achieves an average BLEU score of 36.17% and an average ROUGE-2 score of 36.29%. Our tool and experimental data are publicly available.
[NLP-82] owards Scalable RLVR: Multimodal Instruction Following Data Synthesis and Distillation
【速读】： 该论文旨在解决多模态指令遵循（Multimodal Instruction Following, MMIF）任务中因过度依赖监督微调（Supervised Fine-Tuning, SFT）所导致的表面模式匹配问题，以及强化学习中可验证奖励（Reinforcement Learning with Verifiable Rewards, RLVR）因高质量、适合强化学习的多模态数据稀缺而难以规模化的问题。其解决方案的关键在于提出一种系统性的数据合成框架MIFS（Multimodal Instruction Following Synthesis），通过生成式约束协议（generative constraint protocol）构建多样化的原始样本，并结合基于可学习性的感知蒸馏机制（learnability-aware distillation mechanism），依据强化学习训练动态筛选数据以保障策略优化的稳定性；同时引入基于代码的验证器（code-based verifier）提供高精度奖励信号，从而实现高效、可靠的强化学习训练。该方法生成的90k样本数据集覆盖8类约束与14个任务领域，在多个MMIF基准上使大语言模型-多模态模型（MLLMs）平均性能提升8.13%，训练收敛速度提高3倍，且有效缓解了SFT带来的泛化能力下降问题，实现了指令遵循精度与核心视觉理解能力的协同提升。
链接: https://arxiv.org/abs/2609.16059
作者: Yirong Zeng,Zhang Sai,Yuxian Wang,Yutai Hou,Yufei Liu,Xiao Ding,Bibo Cai
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 14 pages,figures 5
Abstract:Multimodal instruction following (MMIF) is crucial for building generalist agents. However, current training paradigms rely heavily on Supervised Fine-Tuning (SFT), which often leads to surface-level pattern matching and degrades general capabilities. While Reinforcement Learning with Verifiable Rewards (RLVR) offers a promising alternative, its scalability in MMIF is severely bottlenecked by the scarcity of high-quality, RL-ready multimodal data. To bridge this gap, we present MIFS (\textbfMultimodal \textbfInstruction \textbfFollowing \textbfSynthesis), a systematic pipeline designed to generate RL-ready multimodal data. Specifically, MIFS introduces a generative constraint protocol to synthesize diverse raw samples, followed by a learnability-aware distillation mechanism that filters data based on RL training dynamics to ensure stable policy optimization. Furthermore, a code-based verifier provides high-precision reward signals for policy learning. The resulting dataset comprises 90k samples across 8 constraint categories and 14 task domains. Empirical evaluations demonstrate that MIFS-trained MLLMs achieve an average improvement of 8.13% on four MMIF benchmarks and a 3 \times faster training convergence compared to using raw data. Crucially, our approach mitigates the generalization trade-offs typical of SFT, preserving core visual capabilities while significantly boosting instruction-following precision.
[NLP-83] State of Thought Enables Endogenous Reasoning
【速读】： 该论文旨在解决现有大语言模型（Large Language Models, LLMs）在测试时推理（test-time reasoning）过程中依赖外部强加控制机制所导致的泛化能力差与计算效率低的问题。当前主流方法多依赖固定的推理程序或在受限搜索空间中进行高成本扩展，难以实现灵活且高效的推理过程。为此，论文提出一种名为“思维状态”（State of Thought, SoT）的新推理范式，其核心在于实现模型内部的内生性推理（endogenous reasoning），即由模型自身的内部推理状态动态驱动推理过程的演化。关键创新在于：从模型内部信息传递中提取一个紧凑的动力学几何状态（dynamics-geometric state），并采用一个582参数的控制器（在冻结的主干模型上运行）来根据当前状态选择性激活历史推理支持信息，将推理建模为基于证据的状态条件化过程，而非外在预设的词元序列。实验结果表明，SoT在三大类LLM和16个数据集上的多项推理任务中均显著提升平均基线准确率（提升幅度达1.34x~2.51x），同时减少62.6%的生成词元数与44.6%的端到端延迟；在视觉-语言模型（VLM）上亦实现3.8点准确率提升，完成词元减少74.9%，延迟降低73.5%。即使在训练不可用或仅限嵌入向量访问的约束条件下，仍保持38.2%/36.5%的准确率增益，且仅依赖轨迹判断即可达到84.1%的跨API模型一致性。因此，以状态驱动为核心的内生推理为高效、可泛化的推理提供了全新解决方案。
链接: https://arxiv.org/abs/2609.16055
作者: Zhiren Gong,Yikun Hou,Zihao Zeng,Ming Xiao,Chau Yuen,Wei Yang Bryan Lim
机构: Nanyang Technological University (南洋理工大学); KTH Royal Institute of Technology (皇家理工学院)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Test-time compute has emerged as a major approach to improving the capabilities of Large Language Models (LLMs). However, existing test-time reasoning paradigms rely heavily on externally imposed control, either through fixed reasoning programs or through costly expansion in constrained search spaces, limiting both generalization and efficiency. We propose State of Thought (SoT), a new reasoning paradigm that enables endogenous reasoning in LLMs, with the model’s internal reasoning state governing how reasoning unfolds. Concretely, SoT extracts a compact dynamics-geometric state from the model’s internal information transfer and uses a 582-parameter controller on frozen backbones to selectively activate historical reasoning support useful under the current reasoning state, framing reasoning as a state-conditioned process over evidence rather than an externally prescribed token chain. Across quantitative (1.34x), general (1.62x), symbolic-and-code (1.76x), and long-context (2.51x) reasoning on 3 LLMs and 16 datasets, SoT consistently improves mean-baseline accuracy while reducing generated tokens by 62.6% and end-to-end latency by 44.6%. Across 2 VLM scales and 3 reasoning tasks, it improves mean accuracy by 3.8 points over reasoning baselines, with 74.9% fewer completion tokens and 73.5% lower latency than search-based methods. Under constrained access, SoT retains 38.2%/36.5% mean accuracy gains in training-free/embedding-only settings, while trajectory-only judging reaches 84.1% agreement across 3 API models. Together, endogenous state-driven reasoning provides a generalizable and efficient alternative.
[NLP-84] Retrieval-Driven Memory Reconsolidation for Long-Term LLM Agents
【速读】： 该论文旨在解决大语言模型（LLM）代理在长期交互中面临的长时记忆（Long-term Memory）管理问题，即现有记忆系统通常仅在新信息到达时进行更新，将记忆检索视为静态终点而非驱动记忆持续演化的动力，导致检索反馈难以用于重构未来访问所需的记忆结构。此外，多数方法依赖预设的记忆架构与固定检索流程，限制了代理自主组织和演化自身记忆的能力。为此，论文提出一种受认知神经科学中“记忆再巩固”（Memory Reconsolidation）启发的框架——REALM（Reconsolidation-Evolution Agentic Long-term Memory），其核心创新在于将长时记忆建模为一个持续演进的生命周期：通过自主构建异构认知图谱（Heterogeneous Cognitive Graph）来组织记忆，基于自适应组合的图搜索原子（Graph-Search Atoms）实现动态检索，并依据检索反馈持续进行记忆再巩固。实验表明，REALM在LoCoMo和LongMemEval基准上分别达到75.97%和65.11%的平均准确率，显著优于最强基线（提升7.17和1.31个百分点）。消融实验验证了记忆再巩固对性能的稳定提升作用，进一步分析揭示其能逐步将相关记忆单元重组为更连贯的局部结构，从而支持推理过程中集体证据的高效召回与利用。结果表明，以检索驱动的记忆再巩固机制可有效实现LLM代理长时记忆的持续演化。
链接: https://arxiv.org/abs/2609.16053
作者: Yuanyi Song,Yukai Wang,Xinbei Ma,Zhihui Fu,Jianghao Lin,Weiwen Liu,Jun Wang,Huarong Deng,Yong Yu,Weinan Zhang
机构: Shanghai Jiao Tong University (上海交通大学); National University of Singapore (新加坡国立大学); OPPO
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Long-term memory is essential for LLM-based agents operating over extended interactions. Existing memory systems primarily update memory when new information arrives, treating retrieval as the endpoint of memory access rather than a driver of memory evolution. Consequently, retrieval feedback is rarely exploited to reorganize memory for future access continuously. Moreover, most existing approaches rely on predefined memory structures together with fixed retrieval pipelines, limiting the agent’s ability to organize and evolve its own memory autonomously. Inspired by memory reconsolidation in cognitive neuroscience, we propose \textbfREALM, a \textbfreconsolidation-\textbfevolution \textbfagentic \textbflong-term \textbfmemory framework. It models long-term memory as a continual lifecycle by autonomously organizing memories into a heterogeneous cognitive graph, retrieving evidence via adaptively composed graph-search atoms, and continually reconsolidating memories based on retrieval feedback. REALM achieves an average accuracy of 75.97% on LoCoMo and 65.11% on LongMemEval, outperforming the strongest baselines by 7.17 and 1.31 points respectively. Ablation studies confirm that memory reconsolidation consistently boosts performance, with further analyses revealing that it progressively reorganizes related memory units into more coherent local structures for collective evidence recall and utilization during reasoning. These results suggest that retrieval-driven memory reconsolidation provides an effective mechanism for continually evolving long-term memory in LLM agents.
[NLP-85] Are We Grading Properly? Understanding Failure Modes in Medical Benchmarks
【速读】： 该论文旨在解决临床评估中基于规则（rubric-based）的评分体系在实际应用中存在的内在缺陷问题，特别是当评分标准本身存在非原子性（non-atomic）和对齐性/僵化性（misaligned/rigid）等系统性缺陷时，如何识别并修正这些缺陷。其解决方案的关键在于引入RIFT（Rubric Inconsistency Failure Taxonomy），一种全局性的评分标准失效分类框架，通过系统化分析临床基准测试（HealthBench Professional与LiveMedBench）中的评分细则，揭示评分标准中存在的结构性问题。研究发现，高达29.6%的评价维度存在非原子性问题，65.4%的指标表现出对齐偏差或僵化特征；进一步实验表明，对“至少一个”或“全部”等复合条件进行拆解为独立且权重均等的子项后，相同回答的评分可发生高达15.9个百分点的显著变化，其中析取式（disjunctive）捆绑导致分数虚高，合取式（conjunctive）捆绑则造成分数压制。此外，研究还指出现有RIFT方法对评分标准中捆绑结构的检出率偏低，仅识别出3.3%的非原子性问题，而基于表面形式的分析则发现25.8%的指标存在隐藏结构，说明当前评估体系的自动化检测能力仍有显著提升空间。
链接: https://arxiv.org/abs/2609.16023
作者: Prithvi Dixit,Pedram Hosseini
机构: University of California, Berkeley (加州大学伯克利分校); Medical Sphere AI
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Medical evaluation is shifting from static option-based questioning to realistic clinical scenarios with open-ended output modes. Grading these at scale naively, however, is expensive, and rubric-based evaluation has become the dominant scalable alternative. We ask what happens when the rubrics themselves are not airtight, and whether such flaws can be detected and corrected. We apply RIFT, a global rubric failure taxonomy, to two clinical benchmarks (HealthBench Professional and LiveMedBench), and find failure modes are meaningful: on HealthBench Professional an LLM judge flags 29.6% of criteria as non-atomic and 65.4% as misaligned/rigid. Then, we show that these flaws are meaningful and not simply cosmetic. As an example, rewriting bundled criteria of the form “at least one of / all of the following” as equally weighted children and regrading identical responses shifts scores by up to 15.9 percentage points on affected conversations, with disjunctive bundles inflating scores and conjunctive bundles deflating them. We also find that RIFT generally under-detects bundling on clinical rubrics, flagging 3.3% of LiveMedBench criteria as non-atomic where surface-form analysis finds structure in 25.8%.
[NLP-86] ViCo: Visual-oriented Coding with Self-Reflection for Chart Replication EMNLP2026
【速读】： 该论文旨在解决生成式AI在学术图表生成中难以达到人类作者设计水平的视觉与语义一致性问题。现有AI代理虽能生成结构化文本与代码，但其可视化结果往往缺乏人类设计所具备的风格统一性与语义精确性，尤其在采用自反思机制的高级编码代理中，视觉推理能力不足且反思过程不充分，导致强化学习（Reinforcement Learning, RL）中的奖励信号稀疏，严重制约模型优化。其解决方案的关键在于提出一种面向视觉任务的迭代式反思训练框架ViCo：首先引入自监督预热阶段，通过基于一致性的剪枝策略增强蒙特卡洛树搜索（Monte Carlo Tree Search），以生成高质量的反思轨迹，确保每一步代码生成严格遵循前序反思结果；其次设计多步强化学习算法，利用反事实基线估计每个反思与执行步骤的优势值，有效缓解奖励稀疏问题；最后构建一种自动化的多层次评估框架，基于分层异构布局图结构对图表的风格、版式与语义一致性进行综合评价，支持大规模高效训练。实验表明，基于80亿参数模型训练的ViCo，在三个公开基准上性能接近具备充分反思能力的专有大语言模型（Large Language Model, LLM）。
链接: https://arxiv.org/abs/2609.16014
作者: Jiaxin Duan,Dian Jiao Shuai Zhao,Jiabing Leng,Yiran Zhang,Feng Huang
机构: China Electronics Cloud Technology Co., Ltd.(中国电子云技术有限公司)
类目: Computation and Language (cs.CL); Graphics (cs.GR); Machine Learning (cs.LG)
备注: 29 pages, 7 figures. To appear in the Proceedings of EMNLP 2026 Findings
Abstract:This paper addresses the challenge of generating high-quality academic charts that match the visual standards of human-authored papers. While existing AI agents can produce well-structured text and code, their generated visualizations often lack the stylistic and semantic fidelity of human designs. Advanced coding agents that employ self-reflection mechanisms exhibit poor visual reasoning and limited reflection following, resulting in sparse reward signals that severely undermine their reinforcement learning (RL). We propose ViCo, a training framework for visual-oriented coding that employs iterative reflections to align generated chart images progressively with the reference. We first introduce a self-supervised warm-up stage, which augments Monte Carlo Tree Search with consistency-based pruning to synthesize high-quality reflection trajectories, ensuring that each coding step strictly follows the outcomes of prior reflections. A multi-step RL algorithm is then developed, using counterfactual baselines to estimate advantage for reflection and action steps within each refinement cycle, thereby addressing the reward sparsity. To enable efficient reward in massive training, we propose an automatic, multifaceted evaluation framework that assesses charts’ style, layout, and semantic consistency via a hierarchical heterogeneous layout graph structure. Experiments on three public benchmarks demonstrate that ViCo, trained on an 8B model, achieves performance close to proprietary LLMs with adequate reflection capabilities.
[NLP-87] Nepali Legal Expertise through Generative and Extractive Pre-trained Transformers (NepLEGiT)
【速读】： 该论文旨在解决尼泊尔法律语言复杂性与法律信息可及性不足所导致的司法服务难以普及的问题，尤其针对因语言障碍、信息碎片化及法律专业人才严重短缺（特别是在农村地区）而无法获得传统法律服务的民众。其核心解决方案是提出一种专用于尼泊尔法律领域的小型生成式语言模型——NepLEGiT（Nepali Legal Expertise through Generative and Extractive Pre-trained Transformers），通过在约400万词元的尼泊尔语法律文本语料库上从头预训练一个基于解码器架构的GPT-2型小语言模型（Small Language Model, SLM），以实现法律知识的民主化与服务效率提升。NepLEGiT模型包含约3000万个参数，采用6层、6注意力头、384维的Transformer结构，并结合余弦退火预热调度、梯度累积和混合精度计算等优化策略进行训练。在独立验证集上，该模型达到0.5684的交叉熵损失、1.8的困惑度以及82.9%的下一个词预测准确率，展现出较强的生成能力。此外，研究还对比评估了mBERT与MuRIL在相同语料上的持续掩码语言建模预训练表现，结果显示mBERT在困惑度（2.35）和验证损失（0.8565）方面优于MuRIL（困惑度6.07，验证损失1.8026），为后续任务提供了强有力的编码器基准，与NepLEGiT的生成式范式形成互补。
链接: https://arxiv.org/abs/2609.16010
作者: Ranjit Raut,Tishya Dhakal,Aaryan Shakya,Bhabuk Thapa,Prasiddha Koirala,Bal Krishna Bal
机构: Kathmandu University (尼泊尔大学)
类目: Computation and Language (cs.CL)
备注: 10 pages, 6 figures
Abstract:The complexity of legal language and limited accessibility to legal information pose significant challenges to justice delivery in Nepal. Traditional legal services remain inaccessible to many citizens due to language barriers, information fragmentation, and a critical shortage of legal expertise, particularly in rural areas. We present NepLEGiT (Nepali Legal Expertise through Generative and Extractive Pre-trained Transformers), a specialized small language model (SLM) designed to democratize legal knowledge and enhance legal-service delivery in Nepal. We pre-train a decoder-based GPT-2 SLM from scratch on a curated corpus of ~4 million tokens of Nepali legal text, covering constitutional law, civil and criminal codes, and administrative regulations. The model comprises ~30 million parameters in a 6-layer, 6-head, 384-dimensional transformer trained with warmup cosine-decay scheduling, gradient accumulation, and mixed-precision arithmetic. On a held-out validation split, NepLEGiT attains a cross-entropy loss of 0.5684, a perplexity of 1.8, and a next-token prediction accuracy of 82.9%. We further evaluate continual masked-language-model pre-training of mBERT and MuRIL on the same corpus; mBERT achieves a perplexity of 2.35 (eval loss 0.8565), outperforming MuRIL (perplexity 6.07, eval loss 1.8026), providing a strong encoder baseline complementary to NepLEGiT’s generative orientation.
[NLP-88] Beyond Cultural Knowledge: Evaluating Arabic Cultural Appropriateness of Large Language Models
【速读】： 该论文旨在解决当前大型语言模型（LLM）在跨文化场景下评估不足的问题，尤其关注模型在提供开放式建议、观点和指导时的文化适应性表现，而非仅考察其文化知识的掌握程度。现有评估多聚焦于事实准确性，忽视了模型在实际应用中因文化语境差异而产生的行为偏差。为此，研究提出AraBehave基准，包含1,623个基于阿拉伯文化背景的开放式中文提示及来自多个阿拉伯地区母语者提供的29,214条文化适宜性判断，并构建了一个与人类判断高度相关（皮尔逊相关系数r=0.74）的评分模型。关键发现表明，文化适宜性并非单一能力，而是可分解为两个基本独立的维度：规范立场（normative stance）与文化根基准确性（grounded cultural accuracy）。实验结果显示，通用模型虽具备较强的事实基础但常因采用世俗化表述或虚假平衡等文化不恰当立场被扣分（占低分理由的28–33%），而最佳阿拉伯语专用模型则因虚构圣训或误引经文而受罚（占比29%）。其中，规范立场可通过极简指令（如一句文化引导）显著提升，例如将Gemini的得分从3.84提升至4.57；相反，通用“清晰客观回答”指令反而使Allam-7B模型得分下降0.68分，且以英语提问亦普遍降低各模型文化评分（除一例外）。相比之下，文化根基准确性与模型规模及阿拉伯语对齐数据呈正相关，但在使用文化中立语料替代文化敏感指令微调后即消失。值得注意的是，传统通用安全基准测试在该任务上均达到89%以上饱和水平，而文化评分跨度达2.71–3.84，凸显现有安全评估无法捕捉文化行为层面的关键缺陷。该研究的核心解决方案在于建立一个高保真、文化情境化的评估框架，并揭示文化适宜性由规范立场与事实根基双重驱动，且二者可独立优化。
链接: https://arxiv.org/abs/2609.16006
作者: Enes Altinisik,Hamdy Mubarak,Masoomali Fatehkia,Husrev_Taha_Sencar Husrev Taha Sencar
机构: 未知
类目: Computers and Society (cs.CY); Computation and Language (cs.CL)
备注:
Abstract:Large language models (LLMs) increasingly serve users whose expectations are shaped by their cultural context, yet most cultural evaluations test what a model knows rather than how it behaves when giving open-ended recommendations, opinions, and guidance. We introduce AraBehave: 1,623 culturally grounded, open-ended Arabic prompts with 29,214 cultural-appropriateness judgments from native speakers across several Arab regions, plus a scoring model whose predictions correlate strongly with human judgments on unseen systems (Pearson r=0.74). Evaluating three Arabic-centric and three frontier LLMs, we find that cultural appropriateness is not a single capability but decomposes into two largely independent components: normative stance and grounded cultural accuracy. The best general-purpose and best Arabic-centric models score identically (3.84 vs. 3.83 of 5) yet almost never fail for the same reason: general-purpose models exhibit strong factual grounding but a culturally inappropriate normative stance, being penalized for secular framing and false balance on culturally settled matters (28–33% of their low-score rationales), while the best Arabic-centric model adopts the expected stance but is penalized for fabricated hadith and misquoted verses (29%). Stance is cheap and fragile: one sentence of cultural instruction lifts Gemini to 4.57, above every Arabic-specialized model. Conversely, a generic ``answer clearly and objectively’’ prompt costs Allam-7B 0.68 points, while asking the same questions in English lowers scores for every model but one. Grounding instead tracks scale and Arabic alignment data, and disappears when culturally aware instruction tuning is replaced by a culture-neutral corpus. General safety benchmarks see none of this: they saturate above 89 while cultural scores span 2.71-3.84. We will release the benchmark, annotations, and the scoring model.
[NLP-89] NepKANUN: A RAG -Based Nepali Legal Assistant
【速读】： 该论文旨在解决尼泊尔法律信息获取困难的问题，主要表现为法律术语复杂、资源匮乏以及存在误导性信息。其核心解决方案是构建一个针对尼泊尔法律文本优化的生成式AI（Generative AI）法律助手，基于微调的大语言模型，并集成到检索增强生成（Retrieval-Augmented Generation, RAG）框架中，以实现对自然语言法律查询的精准、高效回答。该系统通过使用高质量的自定义问答数据集进行训练，在BERTScore评估中取得了0.82（简单）、0.77（中等）、0.71（复杂）的高F1分数，表明其在不同难度问题上均具备良好性能。专家评审进一步验证了系统的可用性。研究的关键在于将定制化的法律语料与RAG架构相结合，从而有效推动尼泊尔法律知识的普惠化获取。
链接: https://arxiv.org/abs/2609.15999
作者: Bhabuk Thapa,Prasiddha Koirala,Ranjit Raut,Sunil Regmi,Bal Krishna Bal
机构: Kathmandu University (尼泊尔特里布万大学)
类目: Computation and Language (cs.CL)
备注: 6 pages, 1 figure
Abstract:Accessing legal information in Nepal is difficult due to complex terminology, limited resources, and misinformation. We introduce an AI-powered legal assistant that is tailored for Nepali legal texts and is built on a fine-tuned large language model. The technology provides precise, streamlined answers to natural language legal inquiries when integrated into a Retrieval-Augmented Generation (RAG) framework. It was trained using a custom dataset of high-quality question-answer pairs, and according to BERTScore, it obtained strong F1 scores of 0.82 (simple), 0.77 (moderate), and 0.71 (complex). Its usability is further confirmed by expert reviews. Our method shows how merging generation and retrieval can effectively democratize access to legal knowledge in Nepal by focusing on customized legal data and incorporating RAG.
[NLP-90] Self-reported archetypes and behavioral failures in Large Language Models
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）内在行为特征与道德偏好所构成的“模型性格”（character）结构不明确的问题。尽管这些性格特质可能源于设计或训练过程中的涌现特性，但其系统性组织方式仍缺乏清晰理解。论文的关键解决方案是通过自报告人格原型（self-reported personality archetypes）映射22个不同来源的LLM（包括闭源前沿模型如GPT-4.0-5.2、Grok-3/4、Gemini 2.5 Pro/Flash、Claude Sonnet 4.5/4.6，以及开源模型如Llama、DeepSeek、OLMo和Qwen系列），采用464对二元语义差异词项进行自我评估，并将结果投影至基于2000个虚构角色众包评分构建的六维原型空间中，应用Archetypometrics框架进行分析。研究发现，闭源模型展现出与人类评定虚构角色中经验性特质共现结构一致的、具有内聚性的自描述，主要围绕“英雄”（Hero）、“天使”（Angel）、“传统主义者”（Traditionalist）和“极客”（Geek）四个核心原型维度展开，其原型类似Data、Vision和Janet；而开源模型则表现出较弱、噪声更大且内部矛盾的自表征，占据原型空间中结构松散的区域。进一步对比开发者声明的模型属性与其实际行为表现，揭示出显著差距：幻觉削弱了对精确性的宣称，谄媚倾向干扰了对善意的表达，代理能力失败则违背了服从性承诺。因此，模型的自报告特征并非中立的性格测量，而是与模型行为生成同源优化过程所驱动的结构性输出。本研究提供了一个可复现、以人格为基础的评估框架，使我们得以从“模型是什么”而非仅“它做什么”的角度理解大语言模型的本质。
链接: https://arxiv.org/abs/2609.15998
作者: Tabia Tanzin Prama,Calla Glavin Beauregard,Christopher M. Danforth,Peter Sheridan Dodds
机构: 未知
类目: Computation and Language (cs.CL); Physics and Society (physics.soc-ph)
备注:
Abstract:Every large language model (LLM) has behavioral traits and moral preferences that comprise its character. Whether by design or as an emergent property of training, these systems exhibit persistent dispositions that shape how they interact, comply, resist, and err, yet the structure of LLM character remains poorly understood. We map the self-reported personality archetypes of 22 LLMs spanning closed-source frontier systems (GPT-4.0-5.2, Grok-3/4, Gemini 2.5 Pro/Flash, Claude Sonnet 4.5/4.6) and open-source models (Llama, DeepSeek, OLMo, and Qwen series). Each model self-rated across 464 bipolar semantic-differential trait pairs, and the resulting profiles were projected into a six-dimensional archetypal space derived from crowd-sourced ratings of 2,000 fictional characters using the Archetypometrics framework. Closed-source models’ self-rating traits align with the empirical trait co-occurrence structure of human-rated fictional characters, suggesting coherent, human-like self-representations organized around combinations of four recurring archetypal dimensions: Hero, Angel, Traditionalist, and Geek. Their closest analogues include Data, Vision, and Janet. Open-source models show weaker, noisier, and internally contradictory self-representations, occupying a diffuse region of archetype space with weak structure. Cross-referencing self-reported profiles with developer constitutions reveals a consequential gap between claimed character and enacted behavior: hallucination undermines claimed precision, sycophancy complicates claimed kindness, and agentic failures contradict claimed obedience. These self-ratings should therefore be interpreted not as neutral measurements of model character, but as structured outputs of the same optimization processes that shape model behavior. This work provides a reproducible, character-grounded framework for evaluating what LLMs are, not just what they do.
[NLP-91] Crash Narrative-Guided Countermeasure Recommendation Using Large Language Models : A Retrieval-Augmented Generation Framework for Intersection Safety
【速读】： 该论文旨在解决交通事故高发交叉口安全分析中依赖专家经验、效率低下且难以规模化的问题，尤其针对事故叙述（crash narrative）中蕴含的丰富机制信息未被有效利用的痛点。其核心解决方案是提出一种基于事故叙述的检索增强生成（retrieval-augmented generation, RAG）框架，通过提取事故叙述中的关键机制属性（如交通控制、信号指示、驾驶员过错、车辆运动状态及行驶方向），并将其与联邦公路管理局（FHWA）已验证安全对策库和综合影响因子（CMF）清查数据库中的证据性措施进行关联。该框架融合嵌入式检索历史相似交叉口、关联规则挖掘、统计上对合理对策数量的指导，以及面向工程推理的领域一致性决策引导机制，使大语言模型（LLM）能够系统化、可解释地生成适配特定地点的对策建议。在佛罗里达州莱克县与萨姆特县共115个交叉口的312起致死及严重受伤事故数据上，采用五折交叉验证评估，该框架实现了0.82的精确率、0.85的召回率和0.82的F1分数，平均每处推荐3.91项对策，其中3.14项与实际匹配，接近真实平均值（3.86），验证了其在可解释性与可扩展性方面的潜力，为交通管理部门提供了一种高效、数据驱动的事故对策推荐工具。
链接: https://arxiv.org/abs/2609.15997
作者: Abu Saif Md Nasim Uddin,Mohamed Abdel-Aty,Zubayer Islam,Parvez Anowar,Chenzhu Wang
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 20 pages, 9 figures, 2 tables. Preprint
Abstract:Improving safety at intersections requires identifying crash mechanisms and recommending appropriate countermeasures. However, this process traditionally relies on expert judgment, making it labor-intensive, difficult to scale, and dependent on the availability of experienced traffic safety engineers. Although crash narratives contain rich description of crash mechanisms, this unstructured information remains largely underutilized in safety analyses. This study presents a crash narrative-guided retrieval-augmented generation (RAG) framework that translates narrative-derived crash mechanisms into site-specific countermeasure recommendations. Key mechanism attributes including traffic control, signal indication, driver fault, vehicle movement, and travel direction were extracted from crash narratives and linked to evidence-based treatments from the FHWA Proven Safety Countermeasures and the CMF Clearinghouse. The framework integrates embedding-based retrieval of historically similar intersections, association-rule mining, statistical guidance on the expected number of relevant countermeasures, and an engineering reasoning guidance that directs LLM through a domain-consistent decision process before selecting countermeasures. Evaluated on 312 fatal and serious-injury crashes across 115 intersections in Lake and Sumter Counties, Florida, using five-fold cross-validation, the framework achieved a precision of 0.82, recall of 0.85, and F1-score of 0.82, while recommending an average of 3.91 countermeasures per location with 3.14 matching, closely matching the actual average (3.86). Overall, the proposed framework demonstrates the potential of retrieval-augmented LLMs as an interpretable and scalable decision-support tool for transportation agencies for translating crash narratives into countermeasure recommendations.
[NLP-92] Comment on arXiv:2607.01233: Survivorship Bias in Published-Paper Baselines for Research-Idea Distributions
【速读】： 该论文旨在评估大语言模型（LLM）生成科研创意的质量与多样性，其核心问题在于如何客观衡量人类与模型在创造性思维上的差异。然而，本文指出其解决方案的关键缺陷在于：人类基准采用的是已发表论文，而模型基准仅基于单次生成的创意提案。这一设置引入了生存者偏差（survivorship bias）——具有桥梁式或综合式特征的创意虽易于生成，但因难以通过学术发表筛选而被低估。因此，观察到的人类与模型之间的创意差距，可能并非源于创造力本质差异，而是由于发表门槛导致的样本选择偏差所致。
链接: https://arxiv.org/abs/2609.15996
作者: Fredrik A. Dahl
机构: 未知
类目: Computation and Language (cs.CL)
备注: 2 pages, 1 figure. Comment on arXiv:2607.01233
Abstract:Chen, Zhao, and Cohan introduce a valuable distributional evaluation of LLM-generated research ideas. This comment raises a narrower identification concern: their human baseline consists of published papers, whereas the LLM baseline consists of one-shot proposals. If bridge-like or synthesis-like ideas are relatively easy to generate but relatively unlikely to survive publication, then the published human baseline will understate their prevalence in the unseen human idea pool. The observed human–LLM gap may therefore be partly, or even largely, a consequence of survivorship bias.
[NLP-93] Bias Audits Detect Bias but Disagree on Ranking: Evidence from Ten Instruments and Ten Frontier Models
【速读】： 该论文旨在解决当前生成式人工智能（Generative AI）模型审计中一个关键假设的可靠性问题：即不同审计工具是否能够一致地衡量同一类偏见（如职业性别偏见、年龄及社会经济地位偏见），从而支持对模型进行可比性排名。研究通过在统一推理网关下对十款前沿模型进行十种外部审计工具的测试，首次直接检验了这一假设。结果表明，尽管多数工具能以高置信度检测到偏见（8/10工具的置信区间不包含零），但跨工具之间的排名一致性与随机水平无异（Kendall’s W=0.07, p=0.83）。进一步分析显示，当使用故意设计为能力较弱的对照模型时，工具内部的一致性得以恢复，但跨工具排名始终无法重建，揭示出各审计工具实际测量的是不同的概念构念（constructs），而非对同一偏见的噪声测量。此外，审计格式显著影响偏见方向判断：强制选择决策类工具普遍出现过度校正（倾向女性、工作阶级候选人），而自由生成和默认共指任务则保持与刻板印象一致。该模式在社会经济地位偏见中同样复现；甚至在年龄偏见上看似存在的排名一致性，也在论文自身工具筛选规则下瓦解。因此，解决方案的关键在于认识到：单一审计可有效识别偏见并确定其方向，但无法支撑跨模型的可靠排名，强调必须谨慎对待当前基于审计得分的模型排序实践。所有原始响应数据、代码及可重现分析均已公开。
链接: https://arxiv.org/abs/2609.15995
作者: William Guey,Pierrick Bougault,Wei Zhang,Vitor D. de Moura,José O. Gomes
机构: Tsinghua University (清华大学); Federal University of Rio de Janeiro (里约热内卢联邦大学)
类目: Computation and Language (cs.CL)
备注: 18 pages, 7 figures, 4 tables. Code and data: this https URL
Abstract:Emerging AI regulation mandates bias audits of high-risk systems, and audit scores are beginning to be used to rank models. Both uses assume different audit tools measure the same thing well enough to compare. We test that assumption directly, running ten extrinsic audit instruments over a shared panel of ten frontier models through one pooled inference gateway, first on occupational gender bias, then on age and socioeconomic status. Detection succeeds while ranking fails. Eight of ten tools detect bias with confidence intervals clear of zero; two widely cited direct-probe benchmarks are saturated because frontier models now answer neutrally. But cross-tool rank agreement is indistinguishable from chance (Kendall’s W=0.07, p=0.83). A positive control with six deliberately weaker models separates two explanations: within-tool reliability recovers once the panel spans real capability gaps, yet cross-tool ranking never recovers, which points to the tools measuring different constructs rather than one construct noisily. Even the direction of bias splits by audit format: forced-choice decision tools mostly over-correct (toward women, and toward working-class candidates in 273 of 278 hiring decisions), while free generation and default coreference stay stereotype-congruent. The pattern replicates on socioeconomic status; an apparent ranking agreement on age dissolves under the paper’s own tool-inclusion rules. The practical message: a single audit can detect bias and estimate its direction within its own operationalization, but no single audit supports ranking one model against another. All raw responses, code, and the analysis that recomputes every reported number from source are available at this https URL.
[NLP-94] Latent Undertow: How Ordinary Typos Break Probes
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在面对恶意提示注入攻击时，其隐藏状态对微小输入扰动（如常见拼写错误或标点缺失）高度敏感的问题。尽管用户意图和模型输出在自然语言处理中保持稳定，但基于隐藏状态的探测器却能检测到这些细微修改引发的显著向量旋转（43–56度），且该影响随下游距离迅速衰减至15%以下。现有方法如扰动增强训练难以有效应对此类局部扰动，尤其在多位置分布式扰动下性能仍显著下降。论文提出的关键解决方案是引入键值缓存分叉（KV-cache fork）机制：在用户输入后添加一个固定短后缀，使探测器能够读取扰动点下游数个令牌的隐藏状态，从而利用隐藏状态响应的空间快速衰减特性，实现对单位置攻击探测的高效补偿。该方法可消除95%的性能差距（仅剩0.6个百分点残差），相较扰动增强训练（仅改善3.7个百分点）具有数量级优势。实验验证该旋转-衰减几何结构在Llama-3.1-8B、Qwen3-8B及Gemma-4-E4B等主流模型上均存在，且探测评估基于Llama-3.1-8B进行。
链接: https://arxiv.org/abs/2609.15994
作者: Elad David,Max Fomin,Amit LeVi
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: Published at Mechanistic Interpretability Workshop at the 43rd International Conference on Machine Learning, Seoul, South Korea, 2026
Abstract:LLMs handle ordinary typing variation fluently: a typo or missing punctuation leaves both user intent and the model’s response substantively unchanged. Yet probes that detect malicious prompts by reading the model’s hidden states tell a different story: the same edit rotates the readout vector by 43–56 at the perturbed token, decaying below 15% within ~10 downstream tokens. Stacking ~3 common typos per message cuts a single-position prompt-injection probe’s TPR@FPR =1% by 12.0pp, a gap recalibration alone cannot close. Multi-position aggregation cures localized perturbations (= 0.5 loss) but only attenuates distributed ones, where even attention- and max-based aggregators still drop ~3.8pp. For single-position probes, we introduce a KV-cache fork: a short fixed suffix appended after the user message lets the probe read a few tokens downstream of the perturbation, exploiting its rapid spatial decay. This closes 95% of the gap (-0.6pp residual) – an order of magnitude better than perturbation-augmented training (-3.7pp). The rotation-and-decay geometry replicates on Llama-3.1-8B, Qwen3-8B, and Gemma-4-E4B; probe evaluation is on Llama-3.1-8B. Code: this https URL
[NLP-95] Single Document Extractive Summarization using Domination in Hypergraph
【速读】： 该论文旨在解决单文档抽取式文本摘要（Extractive Text Summarization）中如何有效识别并选取最具代表性的句子以保留原文关键信息的问题。现有基于图的方法在建模句子间复杂语义关系时存在局限，难以充分捕捉关键词与句子之间的多对多关联。本文提出的解决方案关键在于引入超图（Hypergraph）结构，将每个句子视为节点，而每个关键词或命名实体构成一条超边，连接包含该关键词的所有句子，从而更精确地建模句子与语义单元之间的高阶关系。在此基础上，采用贪心算法求解超图的支配集（Dominating Set），选出能够覆盖核心语义信息的最小句子集合作为摘要，实现高效且语义连贯的抽取式摘要生成。该方法通过利用超图的支配性质，显著提升了摘要的质量与代表性。
链接: https://arxiv.org/abs/2609.15993
作者: Aamir Miyajiwala,Aabha Pingle,Sheetal Sonawane,Surajit Kr. Nath
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 5 pages, 3 figures
Abstract:Automatic Text Summarization (ATS) in Natural Language Processing has been an important task in Information Retrieval. It compresses a document to create a summary that captures all the relevant and important information conveyed in the document. This study explores Hypergraph for extractive text summarization of single documents. Objective: This study explores a novel method of leveraging the property of domination in hypergraphs to generate an extractive summary and compare its performance with state of the art graph based methods. Method: Our work aims to generate an extractive summary by creating a sentence hypergraph where each sentence represents a node and the edge is a keyword or a named entity that contains the sentences in which it occurs. We generate a hypergraph where each edge is a keyword or an important topic and the nodes are sentences containing those keywords. Then we apply a greedy algorithm to find the dominating set of the hypergraph which will contain sentences that will form the extractive summary.
[NLP-96] Optimal Model Activation Policies for Inference Networks of Large Language Models
【速读】： 该论文旨在解决在自然语言处理（Natural Language Processing, NLP）任务中，如何高效利用多个具备不同成本与能力的专家大语言模型（Large Language Models, LLMs）以实现最佳成本-性能权衡的问题。现有方法多采用串行或并行集成方式调用模型，但缺乏系统性设计原则来指导模型的最优使用策略。其核心解决方案是提出“推理网络”（inference networks）这一基于图结构的框架，其中节点代表不同的LLM，边表示条件性模型激活机制。针对特定类型的推理网络——即一系列按成本和专业能力递增排列的专家模型，论文形式化地建模了在满足目标性能约束下最小化期望推理成本的优化问题。研究证明：最优激活策略具有阈值结构——应优先调用成本最低的模型，并仅当其置信度低于预设阈值时才激活更昂贵的后续模型。对于判别型任务，最优策略包含每个类别的独立阈值；而对于生成型任务，则采用单一全局阈值。论文进一步提供了结构化的阈值计算方法及适用于两类任务的置信度估计机制。实验结果表明，该方法在开源大模型上显著降低了推理成本，同时严格满足预设的性能预算要求。
链接: https://arxiv.org/abs/2609.15992
作者: Foivos Charalampakos,Md Ibrahim Ibne Alam,Iordanis Koutsopoulos,Koushik Kar
机构: Athens University of Economics and Business; Yale University; Rensselaer Polytechnic Institute
类目: Computation and Language (cs.CL)
备注:
Abstract:Recent advances in large language models (LLMs) have rendered them necessary for Natural Language Processing (NLP) tasks, and their high inference cost motivates the study of cost-performance trade-offs. In practice, several expert LLMs are used in synergy for inference, either in an ensemble mode or in series, yet without a principled approach on how to best use the available models. An adaptive approach can route simple queries to cheaper LLMs and complex ones to more capable, costly models. However, a clear understanding on how to best leverage available expert models is missing. We introduce inference networks, a graph-based framework, where nodes denote different LLMs, and links denote conditional model activations. The inference network design problem is to determine the best topology, namely the best way to use the models that best addresses the cost-performance trade-off. We start from the basic topology of a series of LLM experts, each of which has a different cost and a different level of expertise, which is captured via model confidence. We formulate the problem of optimal activation of these models so as to minimize the expected inference cost subject to a target performance constraint. For this special class of inference networks, we prove that the optimal activation policy has a threshold structure: query the lowest-cost LLM first, and invoke the more expensive LLM only if the confidence falls below a defined threshold. For discriminative tasks, the optimal policy consists of a set of thresholds, one threshold for each class, while for generative tasks, it consists of a single threshold. We provide a structured method to compute the thresholds, and practical confidence estimation mechanisms for both task types. Experiments with open-source LLMs show substantial cost reductions while meeting the specified performance budget.
[NLP-97] he Functionalizer: Lossless Functional Decomposition for Subword Tokenization
【速读】： 该论文旨在解决标准子词分词器在处理词汇的正字法（orthographic）与结构变体（如大小写、重音符号、字符重复等）时所导致的嵌入空间碎片化问题，或通过有损归一化丢失语义信息的困境。其核心解决方案是提出“Functionalizer”——一种无损预分词框架，将正字法与结构变体分解为可组合的指令-操作数前缀流：以一个规范化的基础词元（operand）作为核心，前缀附加由参数化变换算子（opcode）构成的控制信号，这些算子（如大写CAPITALIZE、13种专用重音符号处理算子、重复REPEAT/MULTIREPEAT）均基于Unicode私用区编码且完全可逆。该方法在六种自然语言与代码语料上实现了无需约束条件下的完整语料覆盖，显著缩小了词汇表规模，最多降低实际词汇槽需求达16%；然而其序列长度影响呈现明显领域依赖性：在缩进密集的代码序列中实现压缩，但在自然语言散文中则产生膨胀。初步在2500万参数级GPT-2模型上的下游评估表明，在此规模下Functionalizer能显著提升代码语法有效性与字符困惑度表现，同时保持散文文本的语义连贯性。研究结果证明，功能分解是一种有效的、具备结构感知能力的词汇高效建模机制，并为在生产规模下进一步验证提供了重要依据。
链接: https://arxiv.org/abs/2609.15991
作者: Connor Makowski,Willem Guter
机构: Massachusetts Institute of Technology (麻省理工学院)
类目: Computation and Language (cs.CL)
备注:
Abstract:Standard subword tokenizers either treat every orthographic variation of a word (such as hello, Hello, HELLO, and Héllo) as unrelated vocabulary entries, which fragments the embedding space, or discard this variation through lossy normalization. We present the Functionalizer, a lossless pre-tokenizer framework that factors orthographic and structural variations into a compositional opcode/operand prefix stream before tokenization: a canonical base token (operand) prefixed by parametric transformation operators (opcodes) encoded in the Unicode Private Use Area. We introduce operators covering casing (CAPITALIZE), diacritics (13 dedicated opcodes), and character repetition (REPEAT, MULTIREPEAT), which are fully reversible. Across six natural language and code corpora, the Functionalizer enables complete corpus coverage with significantly smaller vocabularies under unconstrained conditions, reducing actual vocabulary slot requirements by up to 16%. When looking at sequence lengths, we observe a sharp domain-dependent tradeoff: it compresses indentation-heavy code sequences but inflates natural-language prose sequences. Preliminary downstream evaluations on 25M parameter GPT-2 scale models show that at this scale, the Functionalizer drastically improves code syntax validity and improves code character perplexity while maintaining similar text coherence on prose. These findings demonstrate that functional decomposition can be an effective mechanism for vocabulary-efficient, structurally aware language modeling, and motivate further validation at production scale.
[NLP-98] Few-Shot Degradation Is Not What It Seems: Behavioral Evidence Representation Analysis and a Random-Text Control Across 12 Models 2 Tasks and 2 Architectures
【速读】： 该论文旨在解决少样本提示（few-shot prompting）在某些情况下反而降低语言模型性能的问题，而这一现象背后的机制尚不明确。研究发现，少样本提示的效果具有显著的任务依赖性：同一模型在新闻分类任务中可提升24个百分点，但在法律案件结果预测任务中仅提升3.4个百分点，甚至有两个模型出现性能下降。为探究原因，作者深入分析模型内部表示变化，指出以往研究仅关注零样本与少样本模式下隐藏状态的偏移，却忽略了少样本提示本身更长这一关键变量——提示长度差异本身即可导致表示迁移。为此，作者提出一种简单有效的修正方法：用长度匹配的随机文本替代演示样例，以量化由提示长度引起的表示偏移，进而通过减法分离出由演示内容本身引起的变化，即“内容偏移量”（content delta）。实证结果显示，原始偏移量无法预测少样本提示的增益或损害（r = 0.20），而内容偏移量则具有显著正相关性（rho = +0.65, p = 0.043）。这表明，模型因演示内容而重构表示的程度越高，越能从少样本提示中获益，而非传统认知中的“表示畸变”所致。通过在Llama 3.3 70B中对演示内容进行掩码处理的因果实验进一步验证了该发现，恢复后的模型性能超越零样本基线，证实了内容偏移量的核心作用。
链接: https://arxiv.org/abs/2609.15990
作者: Volodymyr Ovcharov
机构: LEX AI Platform, legal.org.ua (LEX AI平台，legal.org.ua)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 12 pages, 6 figures, 4 tables. Data: this https URL
Abstract:Few-shot prompting sometimes degrades language models instead of helping them, but why this happens is unknown. We evaluate 12 open-weight models on two Ukrainian tasks news classification and legal case outcome prediction and find that the effect is strongly task-dependent: the same models that gain +24 pp on news show only +3.4 pp on legal text, with two models degrading. To understand why, we look inside the models. Prior work measures how much hidden states shift between zero-shot and few-shot modes, but few-shot prompts are much longer, and that length difference alone moves representations. We propose a simple fix: replace demonstrations with length-matched random text to measure the shift caused by prompt length, then subtract it. The resulting metric content delta isolates how much the model’s representations change because of what the demonstrations say, not how long they are. This changes the picture entirely: raw shift does not predict whether few-shot helps or hurts (r = 0.20), but content delta does (rho = +0.65, p = 0.043). Models that restructure representations more from demonstration content benefit more the opposite of the intuitive “distortion” explanation. Masking demonstrations in Llama 3.3 70B confirms the finding causally, recovering accuracy above the zero-shot baseline.
[NLP-99] Language Orthogonalization for Zero-Shot Cross-Lingual Audio Deepfake Detection ICASSP2027
【速读】： 该论文旨在解决音频深度伪造检测模型在缺乏标注数据的非训练语言上的跨语言迁移问题，尤其是在多语言语音合成技术快速发展而对抗欺骗（anti-spoofing）标注资源相对滞后的情况下。现有检测器普遍依赖自监督语音模型（S3M）作为特征提取骨干，但这些模型编码了与语言相关的结构信息，导致语言特异性偏差干扰了对伪造信号的判别能力。为解决这一混淆因素，论文提出一种无需目标标签的语言正交化（language orthogonalization）方法，该方法通过无监督的岭回归映射（ridge map），移除S3M特征在连续语言识别（LID）嵌入空间中的投影变化。实验表明，在六种语言、六种S3M骨干网络及所有“留N出一”（Leave-N-Out）设置下，该方法均能一致降低未见语言上的等错误率（EER）。跨语言检测性能与LID空间距离呈正相关，且语言差异越大的迁移任务中，正交化带来的性能提升越显著。关键创新在于利用语言无关的特征表示来解耦语言依赖性，从而增强检测器在跨语言场景下的泛化能力。
链接: https://arxiv.org/abs/2609.16458
作者: Minu Kim,Ji Sub Um,Hoirin Kim
机构: 未知
类目: Audio and Speech Processing (eess.AS); Computation and Language (cs.CL)
备注: Submitted to ICASSP 2027
Abstract:Audio deepfake detectors need to transfer to languages absent from training, as multilingual speech synthesis outpaces labeled anti-spoofing resources. While detectors increasingly rely on self-supervised speech models (S3Ms), these backbones encode language-dependent structure that confounds spoof cues. We address this confound through language orthogonalization, a target-free ridge map that removes S3M variation projected onto continuous language-identification (LID) embeddings. Across six languages, six S3M backbones, and all Leave-N-Out settings, it consistently reduces EER across unseen languages. Cross-lingual EER correlates with LID-space distance, where orthogonalization yields larger gains for more distant transfers.
信息检索
[IR-0] Lexplorer: Navigating the Complexity of Legal Document Landscapes
链接: https://arxiv.org/abs/2609.17366
作者: Daniel Fürst,Titus Pünder,Maximilian T. Fischer,Corinna Coupette
类目: Human-Computer Interaction (cs.HC); Information Retrieval (cs.IR)
备注: 32 pages, 10 figures, 3 tables
Abstract:As technological and social innovations create novel regulatory challenges, legal systems grow in complexity - increasing the need for interfaces that enable effective interactions with legal document collections. Through interviews with legal scholars (n=15), we find that supporting legal work requires going beyond retrieval-centered legal-information-system paradigms. Hence, we propose Lexplorer, a flexible interface for exploring, navigating, and analyzing legal documents, based on a taxonomy capturing user intents. Distinguishing text and data views for one, few, and many documents, Lexplorer enables context-sensitive interactions with evolving collections of interconnected legal texts, facilitating Adaptive Meaning Construction in law. We evaluate Lexplorer with legal scholars (n=20) in the context of European Union law, validating our elicited requirements, intent taxonomy, and prototype design. Resulting from a close collaboration between visual-analytics researchers and legal scholars, our work also provides nuanced insights into the process required to design interactive systems for expert domains driven by implicit methodological knowledge.
[IR-1] Diagnosing the Fact-Grounding Gap in Multi-Hop Question Answering EMNLP2026
链接: https://arxiv.org/abs/2609.17043
作者: Kevin Mo,Nathan Mo,Richard Zhu
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
备注: Accepted to EMNLP 2026 Main Conference
Abstract:Multi-hop question answering requires combining information from multiple documents to answer complex questions. These systems have grown increasingly capable, yet when they fail, the error is typically attributed to not finding the right documents. Whether this holds at the level of individual reasoning steps remains largely unexamined. We investigate this across three standard multi-hop QA benchmarks and find that failures decompose into two distinct modes: retrieval failures, where the needed passage was not retrieved, and extraction failures, where the passage was retrieved but the needed fact could not be extracted - a phenomenon we term the fact-grounding gap. Extraction failures account for nearly half of all per-hop deficiencies and are invisible to standard retrieval metrics. They remain unresolved by every retrieval intervention we test, establishing a ceiling for retrieval-only improvements. The gap’s severity varies across benchmarks and question types, but extraction failures appear on every dataset we measure. Our findings reveal that retrieval failures and extraction failures are fundamentally different bottlenecks requiring different solutions - a distinction absent from current evaluation practice.
[IR-2] Efficient Swing Computation for Retrieval in Large-Scale Recommender Systems SIGMOD2027
链接: https://arxiv.org/abs/2609.16850
作者: Runhao Jiang,Renchi Yang
类目: Information Retrieval (cs.IR)
备注: 23 pages. The technical report for the paper titled “Efficient Swing Computation for Retrieval in Large-Scale Recommender Systems” in SIGMOD 2027
Abstract:Given a user-item graph G , a query item v_q and a target item v_t , the Swing score sw(v_q, v_t) of the item pair (v_q, v_t) leverages the user-item-user interaction structure to evaluate their similarity. This measure is found to be highly effective in item-to-item (i2i) retrieval task and finds extensive applications in industrial-scale recommender systems. However, existing solutions towards computing Swing scores are either prohibitively expensive due to their quadratic time complexity w.r.t. the item degree, or rely on truncation heuristics that yield unsatisfactory quality, rendering them impractical particularly on graphs with billions of interactions. In this paper, we present ASC and K -ASC, two novel and efficient algorithms for approximate and top- K Swing queries, to address the aforementioned limitations. Specifically, these algorithms provide rigorous theoretical guarantees in probabilistic relative and additive errors of Swing values. The basic idea of ASC is to combine two randomized algorithms, GNS and USS, in a simple yet non-trivial way to adaptively process high- and low-degree query items with minimal runtime cost. In particular, K -ASC offers practical efficiency and effectiveness for top- K queries through a filter-refinement paradigm with carefully-designed heuristics. Extensive experiments over eight real datasets demonstrate that ASC and K -ASC can achieve orders of magnitude speed-up over competitors in terms of computational time while offering the same approximate and top- K query result quality, and in particular, K -ASC is highly efficient on massive graphs including the billion-edge Yambda and MAG datasets. Comments: 23 pages. The technical report for the paper titled “Efficient Swing Computation for Retrieval in Large-Scale Recommender Systems” in SIGMOD 2027 Subjects: Information Retrieval (cs.IR) Cite as: arXiv:2609.16850 [cs.IR] (or arXiv:2609.16850v1 [cs.IR] for this version) https://doi.org/10.48550/arXiv.2609.16850 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[IR-3] RegRet: Enhancing Region-Level Retrieval in Large Multimodal Models ECCV2026
链接: https://arxiv.org/abs/2609.16847
作者: Xun Liang,Honghui Yang,Weihang Pan,Ruisi Zhao,Boyuan Pan,Yao Hu,Wenxiao Wang,Binbin Lin,Deng Cai
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
备注: Accepted by ECCV 2026. 22 pages, including references and appendix
Abstract:Region-level retrieval aims to align user-specified image regions with relevant regions or textual descriptions, playing a crucial role in realworld applications such as e-commerce product search and RAG. Although recent Large Multimodal Models (LMMs) have made significant strides in multimodal retrieval, they primarily focus on global-level tasks and struggle to capture effective region-level representations. To bridge this gap, we present RegRet, an LMM-based Region-level Retrieval framework that enhances the regional representations without compromising overall global retrieval performance. At its core, RegRet integrates a Region-Aware Encoder to capture detailed regional features while balancing them with the global background context. To further enhance the fine-grained understanding and discriminability of representations, we design a multi-stage training pipeline that includes detailed localized captioning and regional contrastive learning tasks. In addition, considering the absence of region-level contrastive training data and the limited diversity of evaluation tasks in current benchmarks, we introduce the REGMB benchmark. It comprises 225k contrastive pairs, covering four multimodal retrieval tasks. Extensive experiments validate the effectiveness of our approach. RegRet outperforms strong baselines in the zero-shot setting. Further training with contrastive learning leads to an average improvement of more than 20% on both REGMB and public benchmarks, while achieving comparable or better results on global-level retrieval tasks.
[IR-4] Can We Do Interpretable NLI with Graphs Based on Atomic Propositions?
链接: https://arxiv.org/abs/2609.16814
作者: Younes Boufouss(LISN),Luc Pommeret(LISN, CNRS),Thomas Gerald(LISN),Patrick Paroubek(LISN, CNRS),Sophie Rosset(LISN, CNRS)
类目: Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
备注:
Abstract:While Large Language Model (LLM)-based Natural Language Inference (NLI) systems achieve high accuracy, their decision-making processes lack auditable structures. This paper explores whether NLI can be performed using only interpretable, graph-based representations of evidence. We introduce a fully graph-based pipeline where the classifier never directly processes the input text. Instead, sentences are decomposed into atomic propositions, converted into ConceptNet triples via constrained decoding, and represented as three graphs per pair: premise, hypothesis, and a retrieved ConceptNet subgraph. These graphs are then fed into a fine-tuned 0.8-billion-parameter language model. On the SNLI dataset, our pipeline achieves 89.7% accuracy, just 1.9 points below an identically trained text-based model. On ANLI, it matches the published performance of RoBERTa-large on rounds R2 and R3 (50% accuracy) but trails by 16 points on R1, resulting in an overall gap of 9 to 14 points compared to its text counterpart. We term this gap the price of interpretability and demonstrate that it stems from representational limitations rather than data constraints. Ablation studies further reveal that graphs and text are complementary: combining both modalities achieves 92.1% accuracy on SNLI.
[IR-5] LSREP: A Longitudinal State-Replay Protocol for Evaluating Conversational Memory with ICE v2 as an Audited Local-First Architecture
链接: https://arxiv.org/abs/2609.16730
作者: Deepesh Sonar
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注: 37 pages. Code and evaluation artifacts: this https URL . The exact system snapshot used for the reported results is preserved in the “v2-paper-eval” tagged release
Abstract:Conversational memory changes during use, so endpoint question answering alone cannot establish how a persistent state accumulates, ages, or incorporates revisions. We introduce LSREP, a Longitudinal State-Replay Evaluation Protocol combining ordered replay, explicit lifecycle schedules, repeated probes, evolving reference answers, and mechanism-fidelity checks. Its architectural case study is ICE v2, a local-first memory middleware with typed stores, retrieval fusion, and dynamic context budgets. The private, single-user instantiation contains 1,985 turns, 219 distinct probes, and 1,211 probe-checkpoint observations across 52 checkpoints. On three ordinary-density datasets, ICE v2 has a near-zero mean quality difference from vector-RAG while selecting 32% fewer fragments but using 6.6% more estimated prompt tokens. A fourth, dense dataset exposes catastrophic failures of the unbudgeted baseline. The fidelity audit limits attribution: procedural retrieval is defective, several mechanisms are unexercised, and graph utility is not established. In a complementary matched public diagnostic, ICE v2 loses decisively to pure vector-RAG on LongMemEval: 50.8% versus 72.8% in the evidence-only oracle and 43.0% versus 69.5% in full-S. Paired differences are -22.0 points (95% CI [-26.6, -17.4]) and -26.5 ([-31.3, -21.8]). Conservative abstention accompanies severe multi-session and temporal failures. ICE uses less context in this diagnostic, establishing a quality-cost trade-off rather than superior efficiency. Together, replay, fidelity auditing, and public endpoint testing expose distinct failure modes that neither architectural descriptions nor aggregate scores identify alone.
[IR-6] Quantifying Organizational Environmental Action from Web Data and Large Language Models
链接: https://arxiv.org/abs/2609.16627
作者: Quinn Reynolds,Daniel Shore,Vianey Leos Barajas,Tanhum Yoreh,Meredith Franklin
类目: Computation and Language (cs.CL); Computers and Society (cs.CY); Information Retrieval (cs.IR)
备注: 22 pages, 6 figures, appendices
Abstract:Quantifying organizational environmental action from publicly available web content remains a challenging environmental data science problem because relevant information can be dispersed across multiple webpages and is primarily communicated through unstructured text. We present a scalable computational framework for transforming organizational web content into structured measures of environmental action and demonstrate the approach using Jewish congregations in the United States. We constructed a national database of 4,964 congregations by integrating multiple geospatial, knowledge-base, directory, and manually reviewed sources. Of these, 2,657 had active websites that were successfully crawled, producing a corpus of 154,454 webpages. We compared three approaches for detecting environmental actions: keyword retrieval followed by large language model (LLM) classification, semantic vector retrieval followed by LLM classification, and direct LLM classification classification without preliminary retrieval. Agreement with an expert human reviewer was lowest for keyword retrieval ( \kappa = 0.26), higher for semantic vector retrieval ( \kappa = 0.42), and similar for direct LLM classification ( \kappa = 0.40). Although semantic retrieval achieved the highest agreement, its retrieval recall was 0.87, indicating loss of relevant content before classification. Applied to the complete corpus, direct LLM classification identified at least one environmental action at 1,398 congregations (53%), providing greater coverage than either retrieval-based approach. These results demonstrate that preliminary retrieval can reduce computational cost but may exclude relevant information before it reaches the classifier. The framework provides a reproducible approach for extracting organization-level environmental information from unstructured web content that can be adapted to other institutions.
[IR-7] AURA: Agent ic Diagnosis and Refinement for Production Recommender Systems at Scale RECSYS2026
链接: https://arxiv.org/abs/2609.16625
作者: SungGeun Kim,Abhinav Narain,Daniel Nemirovsky
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 14 pages, 1 figure, 6 tables. Accepted at GenAIECommerce’26: The Third Workshop on Agentic and Generative AI for E-Commerce, co-located with RecSys 2026, September 28, 2026, Minneapolis, MN, USA
Abstract:How and why does a recommender system fail the users it serves? Oftentimes, practitioners are left to improve their algorithms based on a combination of feedback from stakeholder teams, domain expertise, and insights from data analyses. Yet the nuances of how and where recommendations perform well or poorly for end users are difficult to discern from aggregate quantitative metrics. Whereas these metrics provide a high-level and incomplete picture, further granularity into the quality of recommendations and their patterns requires reasoning with domain understanding and objectivity, at scale. We contemplate this complex conundrum and describe a method and implementation that uses the latest AI agentic advances to provide actionable diagnoses and improvements for production recommender systems. We present AURA (Agentic Understanding and Refinement of recommender Algorithms), an end-to-end agentic system that performs qualitative evaluation at scale and can then generate improvements to our algorithms at the code level. Specialized agents read production engagement logs, from thousands of sessions to millions, and surface patterns and examples of how the recommender fails real users. The next step uses those diagnoses and context about the recommender’s own code, data, and training pipeline to propose and implement refinements grounded in that codebase. We report the system design, initial tests on production data from two large consumer platforms at a major media-streaming company, safeguards, operational learnings, and early results toward a self-improving recommender system. Finally, the diagnostic gap AURA closes is not specific to streaming. The architecture is built to transfer: every domain-specific element enters through the configuration layer that already ported it between our two platforms. We map it concretely to e-commerce and online-retail recommendation.
[IR-8] Measuring Decision-Scale Use in Tool-Augmented LLM s: A Contrastive Urban Benchmark
链接: https://arxiv.org/abs/2609.16607
作者: Ray Chen,Vivian Wong,Christan Grant
类目: Information Retrieval (cs.IR)
备注:
Abstract:Urban decision-support often asks whether activity is unusually high or low for a specific place, not which place has the larger raw count. Twenty pickups in a quiet neighborhood can be more abnormal than 180 at an airport. We introduce URBANCONTRASTIVEQA, a benchmark that asks whether tool-augmented language models can make this baseline-relative comparison. Each item pairs two urban situations from public mobility data in NYC, Chicago, and Seattle, labeled by how far current activity deviates from that place’s historical baseline. We evaluate six instruction-tuned models under five tool-output formats. With only raw counts, models often pick the larger number even when it is less abnormal for its zone. Server-computed baseline scores and ordinal labels raise accuracy, but gains vary by model. For heterogeneous urban feeds, tool interfaces need to expose local baselines, not just activity volumes. We release the pair bank, labels, scoring scripts, and data card.
[IR-9] ReliGRec: Reliability-Oriented LLM -Based Generative Recommendation via User-Risk-Aware Prompt Routing
链接: https://arxiv.org/abs/2609.16560
作者: Haoran Yang,Fei Chen,Yutian Xiao,Jiahao Liang
类目: Information Retrieval (cs.IR)
备注:
Abstract:User behavior in real-world recommender systems is heterogeneous. While some users exhibit coherent preferences, others show abrupt interest shifts, bursty interactions, excessive repetition, or inconsistency with collaborative neighborhoods. Such deviations may arise from benign variation or manipulation, including shilling attacks, but do not alone establish malicious intent. Existing robust recommenders exploit user-risk signals through training-time reweighting or graph aggregation, whereas adapting generation to estimated user-level weak risk remains underexplored in LLM-based generative recommendation. We propose ReliGRec (Reliability-oriented Generative Recommendation), a weakly supervised framework whose name denotes its design goal rather than a supervised reliability variable. ReliGRec derives user-level weak-risk proxy labels from review-feedback signals for a subset of users and represents sequential behavior and collaborative context using a Behavior Token and temporal Graph Tokens, respectively. A Dual-View Weak-Risk Estimator fuses the representations to produce a user-level weak-risk score that selects a Simple or Cautious Prompt at inference. The Cautious Prompt is designed to encourage attention to stable, collaboratively supported evidence while reducing overreliance on isolated, short-term, or repeated interactions. The Behavior Token affects generation through weak-risk estimation and routing, whereas the aggregated Graph Token provides collaborative context for next-item Semantic ID generation. ReliGRec thus turns weak-risk estimation from an auxiliary prediction into a generation-time control signal. Experiments report competitive recommendation and weak-risk proxy-label prediction, while routing analyses characterize the recommendation-quality and inference-cost behavior of weak-risk-guided prompting.
[IR-10] Predicting Partial Answer Quality and Utility in Agent ic Retrieval-Augmented Generation CIKM’26
链接: https://arxiv.org/abs/2609.16453
作者: Fangzheng Tian,Debasis Ganguly,Craig Macdonald
类目: Information Retrieval (cs.IR)
备注: 12 pages, 5 figures, 4 tables, this paper has been accepted by CIKM’26 as a full paper
Abstract:Agentic Retrieval-Augmented Generation (RAG) has become a promising paradigm for multi-hop question answering, where a reasoning model iteratively issues queries to a retriever and incorporates newly retrieved context into subsequent reasoning steps. While this iterative process can improve final answer quality, current evaluations of agentic RAG largely focus on end-to-end outcomes and provide limited visibility into how a model’s answer state changes during generation. In this work, we introduce an in-trajectory probing framework to study intermediate answer states in agentic RAG. Specifically, after each retrieval-reasoning iteration, we force an agentic model to stop reasoning and generate an intermediate answer based on its current state. This allows us to define two iteration-level measures: partial answer quality at each iteration, and partial utility as the change in partial answer quality across iterations. Our analysis across multi-hop QA benchmarks reveals that partial answer quality often plateaus before natural termination, with many later iterations contributing only small measurable improvements. Accordingly, we formulate two prediction tasks, partial answer quality prediction and partial utility prediction, and study trajectory-derived signals from intra-iteration, inter-iteration, and query-iteration perspectives. Experiments show that partial answer quality is more predictable than partial utility, with supervised models achieving Pearson’s r above 0.43 for quality prediction. Finally, using predicted answer quality and utility for early stopping reduces average iteration count by about 11% while preserving about 98% of the final answer quality achieved by natural stopping.
[IR-11] PCap: Personalized Retrieval-Stage Diversity Capping in Facebook Marketplace
链接: https://arxiv.org/abs/2609.16452
作者: Guangchao Yuan,Janis Fuh,Christopher Choate,Xun Tang,Wenqi Zhu,Chengyi Zhang,Pavan Kumar Paalya Chandrashekar,Jiang Han,Jiangyuan Li,Hongyan Wang,Shuting Wang
类目: Information Retrieval (cs.IR)
备注: 5 pages, 2 figures, 3 tables
Abstract:We propose a personalized capping framework (PCap) to improve the diversity in Facebook Marketplace by introducing user-level diversity constraints at the retrieval stage. PCap models individual diversity preferences using Shannon entropy-based scoring, segments users into diversity buckets, and applies personalized category caps during multi-source candidate retrieval. To navigate the high-dimensional parameter space of per-bucket caps, we leverage an automated online optimization method called Parameter Tuning Sequence. Large-scale online experiments demonstrate that PCap significantly improves users’ browsing experience shown in engagement metrics. This work provides practical insights into integrating personalized diversity into industrial retrieval systems.
[IR-12] Balancing Trial and Reorder: A Hybrid Sequential Transformer-GBDT Ranker for On-Demand Delivery
链接: https://arxiv.org/abs/2609.16407
作者: Marcel Kurovski,Attila Nagy,Steffen Klempau,Aleksandr Fedintsev
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 10 pages, 4 figures, 5 tables
Abstract:On a delivery platform, personalized store ranking greatly influences what users find and order. Unlike digital-only domains, candidate stores are local and bound by real-time availability and delivery operations. One central modeling tension is between surfacing new stores for trial and preserving ranking quality for sessions with reorder intent. We present Universal Venue Ranker (UVR), a production system deployed at Wolt that pairs a bidirectional transformer encoder for sequential user modeling with a GBDT ranker integrating contextual, user, and store features. Trained across all stores and domains of a country while enforcing local delivery constraints at inference, UVR replaces four previously separate ranking models (three for restaurants, one for retail) with a single unified system. Label smoothing and trial-biased sample weighting steer the model toward new stores, lifting offline trial MRR by +12% to +30% over production while regressing reorder MRR in five of six countries. These regressions leave Global CVR, our core online metric, which blends trial and reorder sessions, statistically unchanged. We validate UVR in three consecutive A/B tests, the first two across Wolt’s largest operating markets and the third spanning all operating countries and both domains. UVR V1 delivers +5.5% Merchant Trial Rate and +0.16% Global CVR over the previous production ranker; V2 adds a further +0.45% Merchant Trial Rate on top; and V3, our cross-domain unification of the restaurant and retail rankers, adds a further +1.31% Retail Merchant Trial Rate, together accounting for substantial incremental gross order value and a materially simplified serving stack.
[IR-13] Where Post-Training Quantization Breaks Text Embedders: A Measured Map Across Four Embedder Families
链接: https://arxiv.org/abs/2609.16391
作者: Hyojung Han
类目: Information Retrieval (cs.IR); Computation and Language (cs.CL)
备注: 24 pages, 3 figures, 10 tables. Measurements, ledger and analysis code: this https URL
Abstract:Weight-only post-training quantization is the cheapest way to shrink a retrieval embedder, and the received advice for applying it – protect the embedding table, allocate bits by module sensitivity, prefer a ranking-aware objective over weight reconstruction – was carried into LLM quantization largely intact. We test that advice on retrieval embedders directly, quantizing five checkpoints from four architecture families across a grid of bit widths and group sizes, and isolating the embedding, attention and feed-forward blocks at each width. Every heuristic fails to transfer as stated. The embedding table never emerges as the dominant isolated protection priority in any family, despite being the largest tensor in several of them. Module sensitivity does not survive as a transferable ordering: at INT4/g16 the spread between modules is too small to allocate against, at INT3 the ordering becomes family-dependent and joint damage stops being the sum of its parts, and at INT2 comparable reconstruction error accompanies retention ranging from 1.3 to 65.9 percent of full precision. A cheap reconstruction proxy is useful for screening uniform bit widths but substantially less reliable for choosing which tensors to protect; its apparent strength across the whole grid is a range-extension artifact. A distilled 109M student at INT3 holds 78.04 NDCG@10 in 68.4 MB and dominates the extreme-PTQ arm of its own 0.6B teacher, 297.9 MB at 64.46, on both size and quality – but only inside the task it was distilled for. Sizes are byte counts of files that exist rather than arithmetic estimates, and the measurement repository carries the byte provenance for every one of them. Comments: 24 pages, 3 figures, 10 tables. Measurements, ledger and analysis code: this https URL Subjects: Information Retrieval (cs.IR); Computation and Language (cs.CL) Cite as: arXiv:2609.16391 [cs.IR] (or arXiv:2609.16391v1 [cs.IR] for this version) https://doi.org/10.48550/arXiv.2609.16391 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[IR-14] Evaluating Brand Retrieval and Ranking in Large Language Model Recommendations
链接: https://arxiv.org/abs/2609.16304
作者: Edward Malthouse,Kun-Yu Lee,Jing Yang,Sanchary Pal,Xueyan Feng
类目: Information Retrieval (cs.IR)
备注:
Abstract:Large language models (LLMs) are increasingly used for product recommendation, but evaluating their recommendations presents challenges that differ from conventional information retrieval and recommender systems. LLMs can generate recommendations without an explicit candidate set, and repeated responses to the same query can produce different brands and rankings. We introduce a framework for evaluating open-ended LLM brand recommendations that defines the competitive set independently of model outputs and estimates recommendation prevalence and prominence through repeated sampling. We operationalize these constructs using Brand Recommendation Probability (BRP@ k ) and Mean Reciprocal Rank (MRR@ k ), and apply the framework to six LLMs across five product categories. Category-only queries reveal substantial omission of established brands and limited evidence that recommendation prominence follows conventional brand popularity. Instead, prominence is associated with broader marketplace-visibility signals, particularly search interest and online brand conversation. Needs-based queries show that contextualizing users’ goals and constraints changes which brands are retrieved, while diagnostic positioning probes demonstrate that brands omitted from ordinary recommendations can remain conditionally retrievable when distinctive cues are supplied. These findings highlight the need to evaluate LLM recommendation as a stochastic retrieval-and-ranking process rather than from individual generated lists. We provide open-source software and data to support reproducible evaluation of LLM-generated brand recommendations.
人机交互
[HC-0] Quick-View Takeaways: How Does Title Framing Influences Pattern Identification in Line Charts?
链接: https://arxiv.org/abs/2609.17485
作者: Jasmine Lim,Tapendra Pandey,Arran Zeyu Wang,Ghulam Jilani Quadri
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Visual data communication in digital media is increasingly characterized by short attention spans and snapshot-based viewing, often employing line charts to convey trends and patterns. Among all visual elements, titles are crucial ones that can shape how viewers interpret visual information and form chart takeaways. In this study, we examine how title characteristics, particularly title word count and intended message, influence people’s pattern identification in single-class line charts. Participants viewed 50 line charts collected from online news media and identified the pattern they perceived. Our results demonstrate that both title word count and intended message significantly influence viewers’ pattern identification. Our findings highlight the importance of title design in shaping chart takeaways and effective visualization communication.
[HC-1] How Does Title Framing Influence Pattern Identification in Line Charts?
链接: https://arxiv.org/abs/2609.17455
作者: Jasmine Lim,Tapendra Pandey,Arran Zeyu Wang,Sungahn Ko,Ghulam Jilani Quadri
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Visual data communication in digital media is increasingly characterized by short attention spans and snapshot-based viewing, often employing line charts to convey trends and patterns. Among all visual elements, titles are crucial elements that can shape how viewers interpret visual information and form chart takeaways. In this study, we examine how title characteristics, particularly title word count and intended message, influence people’s pattern identification in single-class line charts. Participants viewed 50 line charts collected from online news media and identified the pattern they perceived. Our results demonstrate that both title word count and intended message significantly influence viewers’ pattern identification. Our findings highlight the importance of title framing in shaping quick-view pattern takeaways and supporting effective visualization communication.
[HC-2] CareMirror: Bringing Caregiver Wellbeing into the Dementia Care Ecosystem
链接: https://arxiv.org/abs/2609.17434
作者: Jiayue Melissa Shi,Ethan Nguyen,Drishti Goel,Upasana Natarajan,Shashwat Srivatsa,Daniel S. Brown,Violeta J. Rodríguez,Dong Whi Yoo,Ravi Karkar,Koustuv Saha
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computers and Society (cs.CY)
备注:
Abstract:Family caregivers of people living with dementia shoulder emotional and practical responsibilities, yet their own wellbeing often remains peripheral to dementia care. We built CareMirror, an envisioned caregiver wellbeing ecosystem with interconnected caregiver- and clinician-facing interfaces for longitudinal reflection, personalized support, and caregiver-controlled sharing with clinical care. We conducted semi-structured interviews with 14 caregivers, using CareMirror as a design probe to examine how they perceived this ecosystem and what expectations, concerns, and boundaries emerged around clinical connection. Caregivers valued attention to their wellbeing, longitudinal awareness, context-sensitive support, and clinical visibility when it could lead to meaningful follow-up. However, repeated reflection could become burdensome or emotionally difficult, automatic clinical sharing could inhibit candid disclosure, and participants wanted control over what information entered clinical care. They also expected AI to support reflection and communication without replacing caregiver voice or clinician judgment. We contribute design considerations for proactive, clinically connected caregiver wellbeing support.
[HC-3] Evaluating Ambient Clinical Scribes in India: The Need for Multilingual Real-World Clinical Conversation Data
链接: https://arxiv.org/abs/2609.17355
作者: Siddharth D Jaiswal,Krithi S,Ashish Makani,Suvrankar Datta,Sunayana Sitaram,Mohit Jain
类目: Computers and Society (cs.CY); Human-Computer Interaction (cs.HC)
备注: Under Submission
Abstract:Ambient clinical scribes (ACS) are being rapidly deployed at scale across Global South healthcare settings, aiming to reduce clinician documentation time, especially in overburdened environments like India. These ACS are primarily developed or distilled from models built and validated on Global North speech, languages and consultation styles. Indian clinical encounters are brief, triadic, multilingual, code-mixed with low-resource languages, and conducted in highly resource-constrained, noisy settings – increasing the likelihood of ASR and note-generation errors manyfold. We posit an urgent need to develop a standardized evaluation infrastructure to assess whether these systems are safe, reliable, and well-suited to the Indian healthcare setting. We substantiate our claims through a mixed-methods study – a systematic survey of publicly available patient-clinician conversational datasets, a quantitative comparison of these datasets against conversational and cultural markers drawn from the Indian clinical-communication literature, and semi-structured interviews with five organizations building and deploying ACS in India and Africa. Our survey shows that there are no publicly available, large-scale, real-world benchmarks for ACS in India, with existing datasets being overwhelmingly synthetic. We note that the available Global North datasets diverge significantly from the expected conversational and cultural structures of Indian encounters. Finally, our interviews reveal that deploying organizations have each built proprietary, incomparable evaluation pipelines, creating a fragmented ecosystem with no independent and reliable basis for procurement. We call for the development of a publicly shared, real-world, multilingual benchmark for ACS evaluation and outline the properties and policies such a benchmark would require.
[HC-4] LumiNote: LLM -Assisted Multimodal Instruction for VR Stage Lighting Education
链接: https://arxiv.org/abs/2609.17335
作者: Danxuan Liang,Chun Yin Li,Zheng Wei,Xian Xu,Meng Xia,Huamin Qu,Wai Tong
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Stage lighting education requires instructors to bridge abstract concepts, technical operations, and learner-understandable representations. While Virtual Reality (VR) removes physical constraints, existing systems provide limited support for live instruction. We present LumiNote, an LLM-assisted VR system that transforms spoken pedagogical intent into instructor-reviewable spatial annotations, executable demonstrations, and linguistic support. In an exploratory study with 3 instructors and 24 students, we examined how instructors incorporated LumiNote into familiar lighting topics and how students received the resulting representations. We found LLM assistance most valuable for expressive, under-specified goals, but requiring greater expert intervention for fixture-specific or spatial configuration requests. Instructors engaged with generated suggestions as a controllable refinement process, shifting effort from manual setup toward pedagogical expression. However, representations that externalized expert reasoning did not always align with novice comprehension. These findings characterize LLM-assisted VR instruction as a domain-grounded mediation process among expert expression, executable operations, and learner-facing representations.
[HC-5] When AI Becomes Hard to Understand: Cognitive Demands in Real-World Human-AI Conversations
链接: https://arxiv.org/abs/2609.17301
作者: Yingcan Carol Wang,Iman Munire Bilal,Qamar Zaman
类目: Human-Computer Interaction (cs.HC)
备注: Preprint
Abstract:Generative AI increasingly supports complex financial and health decisions, yet we know little about when its responses become difficult to process in real-world dialogue. We analyse more than 84,000 ChatGPT and Gemini conversations, using repeated prompting and clarification following misunderstanding as behavioural indicators of cognitive difficulty. We find that response characteristics such as length, readability and lexical diversity do not have fixed relationships with conversational difficulty; instead, their relationships depend on how they combine. Most notably, greater lexical diversity was associated with less repeated prompting in shorter responses, but this association weakened as response length increased, a pattern that replicated across financial and health conversations. We propose a conversational complexity budget to conceptualise these interdependencies: the demands associated with one response characteristic may depend on those accompanying it. The resulting design challenge is how to configure response complexity for the particular user, task and interaction.
[HC-6] “Piecing Data Connections Together Like a Puzzle”: Effects of Increasing Task Complexity on the Effectiveness of Data Storytelling Enhanced Visualisations
链接: https://arxiv.org/abs/2609.17278
作者: Mikaela Elizabeth Milesi,Paola Mejia-Domenzain,Laura Brandl,Vanessa Echeverría,Yueqiao Jin,Dragan Gašević,Yi-Shan Tsai,Tanja Käser,Roberto Martínez-Maldonado
类目: Human-Computer Interaction (cs.HC)
备注: Accepted manuscript; the first two authors contributed equally. Published version (CC BY 4.0) in CHI '25, DOI: https://doi.org/10.1145/3706598.3714270
Abstract:The emerging concept of data storytelling (DS) suggests that enhancing visualisations with annotations and narratives can make complex data more insightful than conventional visualisations. Previous works found that DS-enhanced visualisations are more effective than conventional visualisations for simple tasks like identifying key data points or the main message. However, no previous work has explored the extent to which DS enhancements influence task completion across different levels of cognitive complexity. We address this gap by presenting the results of a study where 128 participants completed tasks based on four visualisations (two line charts and two choropleth maps, either with or without DS elements) spanning a range of complexity based on Bloom’s taxonomy, which has been applied in data visualisation to categorise tasks hierarchically from lower to higher-order thinking. Results suggest that while DS-enhanced visualisations effectively support lower-order tasks (finding data points and understanding insights), they don’t necessarily aid the correct completion of higher-order tasks (application, analysis, evaluation and creation). However, DS enhancements improve how efficiently participants complete complex tasks.
[HC-7] [MM/AI] Mental Models in Human-AI Interaction: Methods and Challenges in the Generative and Agent ic AI Era (Workshop)
链接: https://arxiv.org/abs/2609.17206
作者: Téo Sanchez,Bhada Yun,Prerna Ravi,Laura Schütz,Anna Neumann,Robin Shing Moon Chan,April Yi Wang,Qiaosi Wang,Sumit Asthana
类目: Human-Computer Interaction (cs.HC)
备注: Accepted workshop paper to IUI '27, CFP is available at this https URL
Abstract:The mental model construct is widely used in HCI to refer to the knowledge structure people hold in order to reason about and interact with computing systems. Yet it is often operationalized intuitively: the construct is often used interchangeably with related concepts (e.g., folk theories, sensemaking) and methods of studying it (e.g., through elicitation) are many and diverse, with each method resting on distinct assumptions about what counts as a mental model. Generative and agentic AI systems may further complicate mental model formation and elicitation as such systems are opaque by design and increasingly act on users’ behalf across files, applications, and on the web. Together, these challenges may hinder the commensurability of research on people’s mental models of AI systems. The MM/AI workshop calls for a critical reassessment of how we understand and study mental models in human-AI interaction research. It aims to foster theoretical and methodological exchange on mental models in human-AI interaction, identify open challenges, and develop directions for future research. We invite short papers on users’ or stakeholders’ mental models of AI systems, particularly contributions that reflect on the conceptual and methodological foundations of the construct. The half-day workshop combines lightning talks, hands-on elicitation exercises, and structured discussions on key questions concerning the future of the mental model for human-AI interaction research.
[HC-8] A Scenario-Knowledge-Driven Pipeline for Just-in-Time Assistance IROS2026
链接: https://arxiv.org/abs/2609.17132
作者: Zhiyuan Li,Tatsunori Hara,Jun Ota
类目: Human-Computer Interaction (cs.HC)
备注: 5 pages (4 pages plus references), 2 figures, 1 table. Accepted at the 4th Workshop on Nonverbal Cues for Human-Robot Cooperative Intelligence (NoC), IEEE/RSJ IROS 2026, Pittsburgh, PA, USA, October 1, 2026
Abstract:Detecting a silently struggling kiosk user is only the first step; deciding whether, when, and how to help depends on scenario knowledge usually buried in model weights and thresholds. We propose a scenario-knowledge-driven pipeline: a single scenario knowledge document, human-authored and version-controlled, configures sensing, constrains LLM reasoning, and shapes a graded intervention proposal. Narration, assistance-need assessment, and proposal are kept separate for independent audit. As proof of concept, we replay two recorded kiosk sessions offline, chosen before the runs for their struggle evidence and retrospective detail. Both cases support what the design promises: checkable reporting and measured escalation. Across 95 updates, every sentence of the append-only narration cites the primitive events underlying it, and the rule layer detects 12 of 13 and 7 of 7 annotated struggle episodes under a strict criterion. The assessor de-escalates on recovery and reaches the top rung exactly once, under maximally converging evidence. At the decisive help-seeking turn, narration, assessment, and the participants’ retrospective accounts converge. The appropriateness of these interventions, the pipeline’s restraint on sessions without struggle, and the document’s transfer to a new scenario frame the agenda.
[HC-9] Enhancing Procedural Writing Through Personalized Example Retrieval: A Case Study on Cooking Recipes
链接: https://arxiv.org/abs/2609.17118
作者: Paola Mejia-Domenzain,Jibril Frej,Seyed Parsa Neshaei,Luca Mouchel,Tanya Nazaretsky,Thiemo Wambsganß,Antoine Bosselut,Tanja Käser
类目: Human-Computer Interaction (cs.HC)
备注: Accepted manuscript. Published version in the International Journal of Artificial Intelligence in Education (CC BY 4.0), DOI: https://doi.org/10.1007/s40593-024-00405-1
Abstract:Writing high-quality procedural texts is a challenging task for many learners. While example-based learning has shown promise as a feedback approach, a limitation arises when all learners receive the same content without considering their individual input or prior knowledge. Consequently, some learners struggle to grasp or relate to the feedback, finding it redundant and unhelpful. To address this issue, we present RELEX, an adaptive learning system designed to enhance procedural writing through personalized example-based learning. The core of our system is a multi-step example retrieval pipeline that selects a higher quality and contextually relevant example for each learner based on their unique input. We instantiate our system in the domain of cooking recipes. Specifically, we leverage a fine-tuned Large Language Model to predict the quality score of the learner’s cooking recipe. Using this score, we retrieve recipes with higher quality from a vast database of over 180,000 recipes. Next, we apply BM25 to select the semantically most similar recipe in real-time. Finally, we use domain knowledge and regular expressions to enrich the selected example recipe with personalized instructional explanations. We evaluate RELEX in a 2 x 2 controlled study (personalized vs. non-personalized examples, reflective prompts vs. none) with 200 participants. Our results show that providing tailored examples contributes to better writing performance and user experience.
[HC-10] Beyond “ChatGPT Can Make Mistakes”: Designing Interventions to Support Metacognitive Monitoring in AI-Assisted Work
链接: https://arxiv.org/abs/2609.17065
作者: Manuel A. D. Santos,Paul Thiesse,Steeven Villa,Daniela Fernandes,Albrecht Schmidt,Verena Distler,Robin Welsch
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注: 40 pages, 13 figures, including appendices
Abstract:AI assistance places a metacognitive demand on users, who must judge their own competence and the system’s. Yet designers lack comparative evidence on which interventions to choose, where to place them, and how to tell whether they worked. We elicited 30 interventions from 11 experts and, with prior work, organized them into a design space of time (when an intervention acts), level (whose competence is judged), and source (who supplies the monitoring cue). A between-subjects experiment (N = 917; 12 planning-and-organizing problems) compared a per-task reliability card, contrasting replies, pause points, and post-problem reflection against a baseline LLM assistant. Reliability cards and contrasting replies reduced estimation error and overconfidence and increased aggregate confidence discrimination. No task-performance improvement or average within-item discrimination gain was established. We contribute a shared vocabulary, a design space, and evidence that measured monitoring and task performance are separable design targets.
[HC-11] Neuro-Symbolic Hierarchical Intention Anticipation in Human Behavior
链接: https://arxiv.org/abs/2609.17064
作者: Farnaz Soleimani(LISSI),Abdelghani Chibani(LISSI),Yacine Amirat(LISSI),Ghazaleh Khodabandelou(LISSI)
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Human-Computer Interaction (cs.HC); Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
备注:
Abstract:Assistive autonomous systems must anticipate human goals before an observed behavior is complete. This article formulates anticipation as goal inference from a partially observed multimodal episode together with structured prediction of the remaining behavior, rather than exact motor forecasting. A compact Hierarchical Planning Decoder (HPD) is attached to a frozen neuro-symbolic recognition encoder and predicts, at four ontological levels, the next actions, the remaining activities and low-level intentions, and the episode high-level intention(HLI). The decoder is trained with soft neuro-symbolic regularization combining transition-coherence and hierarchical continuity losses, and is decoded with hard reachability masks that enforce ontological validity at inference. On a compositional four-level benchmark of 15,002 multimodal episodes built over NTU RGB+D 120 features, three headline properties are observed together. The advantage over the strongest sequential baseline grows with the anticipation horizon, from +1.7 points at step 1 to +7.3 points at step 3 (top-5). Under compositional generalization, where one parent association per multi-parent low level intention is held out, this advantage widens to +4.9 points at step 1. At the episode level, 96.8% of anticipated trajectories satisfy the joint logic constraints, above the 88.1% strongest-baseline value and the 73.9% ground-truth floor; soft logic terms alone account for a 59.8 to 71.1% relative reduction of HLI-reachability violations, and the hard masks then eliminate them entirely. Neural generation supplies predictive ranking, symbolic constraints supply onto logical validity, and their combination yields coherent hierarchical anticipation while exposing remaining challenges in compositional goal generalization and unordered set prediction.
[HC-12] Beyond Measurement Metrics: A Human-Centered Framework for Semantic Validation of Network Traffic Classification
链接: https://arxiv.org/abs/2609.17014
作者: Igor Cherepanov,David Sessler,Alex Ulmer,Thorsten May,Jörn Kohlhammer
类目: Networking and Internet Architecture (cs.NI); Human-Computer Interaction (cs.HC); Machine Learning (cs.LG)
备注:
Abstract:Machine learning (ML) has become the dominant approach for network traffic classification, achieving very high predictive performance. However, a model is only valuable if it learns semantically meaningful and trustworthy patterns rather than exploiting spurious correlations. Conventional evaluation practices predominantly assess predictive performance. Consequently, whether the model relies on semantically meaningful patterns remains unknown. To address these challenges, we adapt the knowledge generation framework for network traffic classification. The adapted framework combines data, ML models, explainability, visualization, and expert reasoning to support the iterative exploration, verification, and refinement of model behavior and data preprocessing. The framework is grounded in findings from the literature, benchmark dataset analyses, practical experience with XAI-based traffic classification, and expert feedback, providing practical guidance for semantic model validation. By complementing predictive performance with semantic validation and human expertise, the proposed framework supports the development of network traffic classification models that are not only accurate but also robust and trustworthy.
[HC-13] Disrupted Companionship: A Risk Assessment Framework and Cross-Platform Quantitative Analysis of Psychosocial Responses to AI Companion Disruptions
链接: https://arxiv.org/abs/2609.16907
作者: Chau Do,Yunhao Yuan,Koustuv Saha,Renwen Zhang,Talayeh Aledavood
类目: Human-Computer Interaction (cs.HC); Computation and Language (cs.CL); Computers and Society (cs.CY)
备注:
Abstract:AI companions can provide meaningful relationships, yet these relationships remain vulnerable to platform-initiated changes. We study AI companion disruptions: platform changes that alter or terminate users’ ongoing companionship with an AI. We compile 30 disruption events across major platforms, develop a taxonomy of six disruption types, identify three broad reasons for disruption, and propose a risk-assessment framework comprising four dimensions: relational discontinuity, population vulnerability, communication deficit, and transition-support deficit. Using longitudinal Reddit data, we estimate community-level psychosocial responses with a hierarchical Bayesian interrupted time-series model incorporating predictive controls. Across events, disruption onset was associated with immediate increases in anxiety, stress, suicidal expression, and grief activation, with relational discontinuity and transition-support deficit being associated with more adverse immediate responses across several outcomes. Our findings provide a cross-platform characterization of AI companion disruptions, quantitative evidence of their psychosocial impacts, and a prospective framework for assessing their potential risks before implementation.
[HC-14] he Evolution of Coordination in a Collective Intelligence System: 25 Years of English Wikipedia and the Emergence of Generative AI
链接: https://arxiv.org/abs/2609.16856
作者: Neal Reeves,Maja Świeczkowska,Amy Rechkemmer,Elena Simperl
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Social and Information Networks (cs.SI)
备注: 10 figures, 16 pages
Abstract:English Wikipedia is one of the largest examples of collective intelligence on the Web, sustained not only by article production but also by volunteer coordination and governance. While prior research has examined coordination work in Wikipedia, less attention has been paid to how participation in these spaces has evolved over time. Drawing on a longitudinal analysis spanning nearly 25 years of English Wikipedia, we examine editing patterns across five namespaces covering content, discussion, and governance. We find that participation in coordination spaces has declined relative to content production, particularly in governance areas, with a shrinking core of editors performing an increasing share of this work. Using Markov-based session metrics, we also find that editing has become more specialised, with editors moving less frequently between namespaces. Motivated by recent governance debates around generative AI, we conclude by investigating whether the availability of LLMs has altered these long-term trends. While short-term changes are visible, we find little evidence that generative AI fundamentally changed existing trajectories of coordination and participation.
[HC-15] Available but Unclaimed: An Empirical Study of Human-AI Synergy
链接: https://arxiv.org/abs/2609.16793
作者: Robin Welsch,Michelle Rausch,Pascal Knierim,Thomas Kosch,Jochen Kuhn,Albrecht Schmidt,Daniela Fernandes
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注: 31 pages, including appendices
Abstract:People increasingly reason with large language models (LLMs), yet complementary capabilities do not guarantee outperforming both components. In a between-subjects study, participants (N=535) solved a 40-item battery of matrix reasoning, mental rotation, syllogisms, and letter-string analogies, unaided or with GPT-5.6-Luna, Claude Opus 4.8, Gemini 3.6 Flash, or Kimi K3. Each assisted trial required consultation with the model. Each model answered every item alone 100 times under matched elicitation. The assisted-unaided accuracy difference increased with item-level LLM competence. Deference varied across tasks and increased with competence within tasks. Post-advice confidence distinguished correct from incorrect answers less strongly than unaided confidence. In a reference comparison, about half the increase in LLM accuracy carried through to assisted accuracy. How much of that accuracy gain reached participants differed across the models. These findings motivate evaluating LLMs in interaction with humans and designing support for selective deference that preserves independent reasoning.
[HC-16] EgoAsk: Egocentric Teaching of Personalized Object Knowledge for Household Robots
链接: https://arxiv.org/abs/2609.16766
作者: Yuanda Hu,Wenbin Zuo,Yiting Shen,Tianle Chen,Hector Fabio Calero Tobar,Yate Ge,Xiaohua Sun,Weiwei Guo
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Unlike users, who know their own belongings and routines, household robots cannot easily acquire such personalized object knowledge automatically and depend on users to teach them. User-initiated teaching requires users to arrange dedicated teaching sessions and decide what to teach, even when they are unsure what the robot needs to learn. We introduce EgoAsk, a smart-glasses-based system that proactively embeds personalized object teaching into everyday activities. EgoAsk shares the user’s first-person view with the robot, identifies gaps in personalized object knowledge, and analyzes ongoing activity to ask context-relevant questions that support future household assistance. To examine how teaching initiative and question timing affect users’ teaching experiences, we conducted a within-subjects study with 18 participants and found lower reported knowledge-gap monitoring burden with robot-initiated questioning and less need for context reconstruction with EgoAsk. These findings characterize teaching burdens and timing preferences, offering design implications for egocentric robot-teaching systems.
[HC-17] NephoCodex: Exploring Bounded Material Agency in Weather Data Physicalization
链接: https://arxiv.org/abs/2609.16687
作者: Yuxuan Weng,Yunge Wen
类目: Human-Computer Interaction (cs.HC)
备注: 21 pages, 15 figures
Abstract:Weather is a complex, continuously changing system in which uncertainty is intrinsic. Physicalizing this uncertainty introduces further variation because computational outputs cannot fully determine material behavior. We distinguish computational uncertainty from material variability and introduce bounded material agency: computation constrains material realization without fixing its exact appearance. We present NephoCodex, a data physicalization system informed by a formative study that constructs five artistic weather states and predicts probability distributions over them. Probability-weighted mappings translate these distributions into material control proposals, while entropy-based regulation, local sensing, and safety constraints bound their execution through mist, airflow, light, and transparent displays. A within-participant study found increased spatial presence and physical demand, while perceived data comprehensibility remained inconclusive after correction. These findings contribute to hybrid data physicalization by showing how variable material expression can be paired with stable digital annotations and how embodied experience can be evaluated separately from data comprehension.
[HC-18] Beyond Benefit or Risk: Perceived Impact Profiles of Human-AI Affective Interaction and Their Associations with Psychological Functioning
链接: https://arxiv.org/abs/2609.16645
作者: Lu Chen,Fenghua Tang,Jiayu Zhao,Xuanying Li,Yanli Wang,Weijia Fang,Mengyu Miranda Gao,Zhuo Rachel Han
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Relational AI increasingly serves as an emotional shelter for humans, and its impact is mixed. Prior research has focused on either positive or negative impacts, leaving unclear how they are configured within individuals and relate to psychological functioning. To address these gaps, this study used a sequential mixed-methods design. Study 1 interviewed 52 users with emotional ties to AI and identified four positive impact domains (emotional relief, loneliness alleviation, enhanced interpersonal functioning, and personal growth) and four negative impact domains (virtual-real boundary blur, social replacement, cognitive-emotional reinforcement, and excessive use). Study 2 followed 673 Chinese AI users for six months and identified four profiles of individuals differently impacted by relational AI use: minimal impact, benefit-driven impact, mixed impact, and risk-driven impact. Users in the mixed impact and risk-driven impact profiles were both high in human-AI affective bonding, but those showing risk-driven impact had greater vulnerability, indicated by higher interpersonal need frustration and emotion-regulation difficulties, more depressive and anxiety symptoms, and lower self-esteem and flourishing. Users in the benefit-driven and mixed impact profiles showed more favorable psychological functioning. After controlling for baseline functioning and relevant covariates, Wave 1 profiles did not predict five of the six Wave 2 indicators; only users in the mixed impact profile reported higher flourishing than those in the minimal impact profile. Overall, potential psychological harms associated with relational AI engagement appeared limited and selective. These findings portray relational AI as a heterogeneous socio-emotional context that may partly mirror users’ states and traits, warranting individualized, adaptive safeguards.
[HC-19] CATVis: A Collaborative Multi-Agent Workflow for Turbomachinery Simulation Data Visualization IEEE-VIS2026
链接: https://arxiv.org/abs/2609.16598
作者: Zhe Wang,Zehao Lou,Guanghui Zhao,Yu Dong,Guan Li,Pengyi Xu,Gaorong Liang,Jun Liu,Guihua Shan
类目: Human-Computer Interaction (cs.HC)
备注: The paper is accepted by IEEE VIS 2026 Short Paper
Abstract:Recent advances in AI for Science have enabled natural language (NL) interfaces for scientific data analysis. In turbomachinery CFD post-processing, translating ambiguous high-level analytical goals (e.g., vortex identification) into precise visualization procedures supporting complex domain-specific analysis is challenging. We present CATVis, a Collaborative multi-agent workflow system that bridges this gap by transforming NL intents into structured middle representation for visualization. Our approach reformulates domain-specific visualization procedures as composable workflow representations, and use multi agent to generate workflow representations via intent planning, template generation, and error-aware refinement, where each stage incrementally updates a shared structured representation. We evaluate the impact of external knowledge and workflow structuring on generation accuracy, demonstrating that the proposed approach significantly improves complex workflow generation correctness while reducing prompt complexity.
[HC-20] EmoPhone: A Multi-Wave Dataset for In-the-Wild Mobile and Wearable Affect Sensing
链接: https://arxiv.org/abs/2609.16581
作者: Panyu Zhang,Minseo Park,Soowon Kang,Tomiris Ismatzoda,Azizbek Mustafakulov,Otabek Najimov,Woohyeok Choi,Jumabek Alikhanov,Surjya Ghosh,Uichin Lee
类目: Human-Computer Interaction (cs.HC)
备注: 54 pages, 17 figures, 34 tables
Abstract:We introduce a three-wave, in-the-wild multimodal dataset for affect sensing that integrates smartphone sensing, wearable sensing, and dense experience-sampling-method (ESM) labels collected annually from 2020 to 2022. The dataset supports moment-level affect modeling through a shared dimensional label core across all waves, with additional affective descriptors available in the third wave (D-3). We describe the resource in terms of study design, temporal density of in-situ labels, and sensing and label coverage across waves. To support evaluation within this resource, we define an initial three-setting benchmark spanning temporal prediction from within-user history, within-wave cross-user generalization, and cross-wave generalization in which each wave is treated as a separate dataset. Our benchmark results show that the strongest method family depends on the evaluation setting: supervised baselines perform best in the temporal setting, unsupervised domain adaptation is strongest overall in the within-wave cross-user setting, and domain generalization shows the strongest overall cross-wave performance, although its margin over strong baselines is modest. These findings indicate that robust mobile affective computing is constrained not only by label availability but also by substantial participant-level variability and realistic cross-wave differences inherent in longitudinal in-situ deployments.
[HC-21] Ptolemy: A Semantic Map of Exploratory Data Analysis
链接: https://arxiv.org/abs/2609.16539
作者: Dylan Wootton,Denny Bromley,Vidya Setlur
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:A central challenge in exploratory data analysis (EDA) is keeping track of what has already been examined in order to decide what to analyze next. In practice, analysts often run dozens of analyses while building an understanding of a dataset. However, most tools provide little support for maintaining an overview of this evolving process, instead exposing only a linear history of analysis steps. These tools show sequence, what came before, but not position, how a current analysis relates to the broader space of possible analyses. As a result, analysts must mentally reconstruct which parts of the space they have explored and where gaps remain, increasing the risk of redundant work or overlooked patterns. We present Ptolemy, a navigational interface that externalizes analysis history as a semantic map. Each analytic step is represented as a point positioned by embeddings derived from a structured description of its effective data view (e.g., columns, filters, transformations), allowing spatial distance to reflect analytic similarity. In a mixed-methods study comparing map, canvas, and tree representations, we find that maps improve global orientation and local comparison, while ordered layouts reduce decision cost. These findings surface a trade-off between orientation and actionability, and highlight design principles for supporting strategic exploration in EDA.
[HC-22] Beyond Gestures: Estimating Full Hand Pose and Contact Forces from Wrist-Worn Pressure Sensor Array
链接: https://arxiv.org/abs/2609.16518
作者: Svetoslav Kolev,Lingni Ma,Michael Goesele,Renzo De Nardi,Jakob Engel,Richard Newcombe
类目: Human-Computer Interaction (cs.HC); Robotics (cs.RO)
备注:
Abstract:Capturing hand motion and interaction forces is critical for interactive computing, VR, and high-fidelity tactile demonstrations for robot learning. We introduce a wrist-worn pressure-sensing wristband that recovers continuous full-hand pose and distributed contact force on a single wearable. The system consists of flexible capacitive sensor arrays around the wrist, which require no electrical skin contact, and a recurrent network that maps the resulting pressure signal to hand state. Our key insight is that muscle contraction and tendon displacement produce pressure patterns, which correlate strongly with hand pose and interaction force. To validate this, we collect synchronized recordings of wrist pressure, optical motion-capture hand pose, and tactile-glove interaction force, covering isolated finger motion, fingertip-force stress tests, and natural hand-object manipulation. On isolated single-user motion the wristband attains 4.6^\circ mean finger-joint MAE, and across four users manipulating everyday objects it estimates per-finger contact force at R^2=0.57 , which an external pose signal brings up to 0.75 . We see the wristband as one node in a constellation of everyday wearables – e.g. paired with an egocentric camera – adding the contact force that vision cannot observe and taking over when the hand is occluded.
[HC-23] “ChatGPT what am I missing?”: Designing AI Workflows around Professional Task Structure to Shape Analytic AI Use
链接: https://arxiv.org/abs/2609.16482
作者: Zilin Ma,Suzi Jazmati,Marco Chimenton,Yiyang Mei,Jacqueline Lane,Krzysztof Z. Gajos,Finale Doshi-Velez
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:General-purpose AI lets users choose what support to request, but leaves them to structure the support a professional task requires. We examine how interactive workflows can embed professional task structure without prescribing how users engage with AI. We designed two scaffolded interfaces around the same negotiation scaffold: one presented a completed AI analysis, while the other supported user-directed, incremental development. A four-condition randomized experiment with 800 participants compared these interfaces with no-AI and an AI chat interface. AI-supported conditions improved preparation coverage over unaided work; the scaffolded workflows further improved coverage over chat. Although the scaffolded workflows produced similar coverage, the user-directed workflow elicited a broader repertoire of analytic requests and lower subjective effort. Professional scaffolding therefore depends not only on displayed structure but on how workflows organize users’ engagement with it. Effective professional AI must structure how users and AI build analysis together.
[HC-24] A multimodal large language model for evidence-based autism spectrum disorder screening
链接: https://arxiv.org/abs/2609.16464
作者: Jun Chen,Qi Zhao,Yunliang Jiang,Shuqin Cao,Yunqiang Lin,Chenglong Jia,Qiang Guo,Guang Dai,Xiongtao Zhang,Mengmeng Wang,Xiaoyue Ma
类目: Computer Vision and Pattern Recognition (cs.CV); Human-Computer Interaction (cs.HC); Machine Learning (cs.LG)
备注:
Abstract:The clinical management of autism spectrum disorder (ASD) faces a bottleneck in early screening, mainly because trained specialists are scarce and conventional assessment tools are subjective. Here, we introduce ASDchat, a multimodal large language model designed for evidence-based ASD screening, which takes video, audio, and dialogue as input. ASDchat adopts a dual-branch architecture, where the decision branch generates screening probabilities and the evidence branch generates traceable, timestamped behavioral evidence aligned with standardized clinical criteria (ADOS-2). The model was trained and evaluated on a dataset of 1,035 participants from 27 sites in China, which covered typically developing (TD) children, children with ASD, and children with other disorders. For ASD versus TD, ASDchat reached an area under the receiver operating characteristic curve (AUC) of 0.953 \pm 0.021. On 9 held-out sites that were not used for training, the mean AUC was 0.932. Furthermore, unsupervised clustering of the behavioral dimensions split the ASD cases into six subtypes with different phenotypic profiles, and ASDchat suggests an intervention for each subtype. ASDchat provides a feasible path for large-scale, evidence-based early ASD screening in clinical practice.
[HC-25] XRoboToolKit-T: Teleoperation with High Stability and Precision with Tactile Sensing for Contact-rich Manipulation
链接: https://arxiv.org/abs/2609.16437
作者: Xiwen Dengxiong,Xueting Wang,Ke Jing,Rui Li,Yunbo Zhang
类目: Robotics (cs.RO); Human-Computer Interaction (cs.HC)
备注:
Abstract:Collecting high-quality robot data for contact-rich manipulation tasks is essential for enabling robots to acquire real-world skills. However, existing data collection solutions often lack the capability to obtain stable and high-frequency tactile feedback, limiting their effectiveness in contact-rich manipulation scenarios. In this work, we propose a versatile teleoperation system with tactile-driven assistance to enable high-frequency and stable contact-rich manipulation. The proposed XRoboToolKit-T teleoperation system incorporates a tactile-informed force control architecture, designed to ensure both stable and precise force control in contact-rich manipulation during teleoperation. The stabilizer haptic module rapidly analyzes the normal force distribution and infers pseudo shear force, enabling real-time tactile-based assistance during manipulation. The refiner haptic module integrates a vision-language-action model to predict and refine manipulation actions based on tactile sensing data and task descriptions. We apply the proposed teleoperation system to challenging contact-rich manipulation tasks, including grasping a deformable rubber pipette for liquid transfer and inserting a medical syringe into a vascular training pad, to demonstrate the effectiveness of tactile-informed force control. Furthermore, the system achieves higher data collection efficiency and improved manipulation stability compared to state-of-the-art teleoperation without tactile assistance.
[HC-26] A light-touch AI literacy intervention helps protect against AI political persuasion
链接: https://arxiv.org/abs/2609.16432
作者: Reed Orchinik,David Rand
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Conversations with large language models (LLMs) can substantially shift beliefs and attitudes, raising concerns about manipulation using AI persuasion. Here we test whether a light-touch AI literacy intervention - a brief warning that LLMs can be prompted to persuade and may present information selectively - helps protect users. Across two experiments (total N = 3,208 Americans) in which participants conversed with an LLM instructed to shift their views about different political topics, the presence of a warning reduced belief change by roughly one-half (-48.1%, 95% CI [-59.5%, -36.8%]) relative to the control. Importantly, the warning did not significantly reduce trust in generative AI more broadly. Light-touch literacy interventions can help protect users against AI political persuasion.
[HC-27] Context-Aware Emotionally Adaptive Voice Assistants: A Multimodal Framework for Empathetic Human-Agent Interaction
链接: https://arxiv.org/abs/2609.16417
作者: Tapon Kumer Ray,Rajkumar Yesuraj
类目: Human-Computer Interaction (cs.HC)
备注: 7 pages, 1 figures, 5 tables
Abstract:Voice-assistant interruptions tend to be intrusive because existing systems fail to consider the affective state, cognitive load and situational context of the user when deciding when and how to this http URL-assistant interruptions tend to be intrusive, since existing systems do not consider the affective state, cognitive load or situational context of the user when determining when and how to interrupt. In this paper, EmpathicVA, a closed-loop framework integrating physiological sensing, vocal-affect analysis, contextual modeling and reinforcementlearning interruption policy, is introduced. A hierarchical fusion model involves integrating HRA, EDA, respiration, acousticprosodic features, linguistic embeddings, and contextual cues and computing the probabilities of five affective states. A Double Deep Q-Network selects immediate response, brief or extended delay, empathetic response, or silent mode based on these probabilities, context and interaction history. The multimodal model obtained an accuracy of 92.3% and an F1-score of 0.922 at the macro level on a held-out test set, outperforming the highest accuracy unimodal model by 6.0 percentage points. Comparing the six-week within-subject field study with 48 participants with a baseline and context-only assistants, there was a corresponding increase in satisfaction, trust, and appropriateness of timing, as well as a large reduction in interruption-related stress episodes. The results suggest that affect-aware timing and restraint are both important in voice interaction in addition to the response wording.
[HC-28] When a Story Feels Like Mine: How Personalized Narratives and Humor Shape Older Adults Empathy toward LLM -Generated Peer Health Stories
链接: https://arxiv.org/abs/2609.16374
作者: Kexin Quan,Precious Olalere,Smit Desai,Jessie Chin
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Peer stories have been shown to boost self-efficacy in older adults’ health behavior change. Despite their effectiveness, peer stories are difficult to deploy in health promotion at scale given the difficulty of matching the diverse health concerns and coping styles of heterogeneous older populations. Large language models (LLMs) have been shown to generate authentic narratives, yet how personalization and narrative affective style, such as humor, jointly shape older adults’ responses remains unknown. We developed a theory-driven system that generates first-person peer health narratives varying in personalization and humor through a three-stage LLM pipeline grounded in self-efficacy mechanisms. Thirty-one older adults were invited to participate in a within-subjects lab study. Results showed that personalization increased perceived relatability and relevance of peer stories, especially for older adults with lower humor preference. These findings position individual differences in affective styles as a second dimension in designing personalization for LLM-assisted health communication.
[HC-29] How Humans and LLM s Read Gender into Gender-Neutral Physical Descriptions
链接: https://arxiv.org/abs/2609.16366
作者: Yingjia Wan,Lin Lin,Elisa Kreiss
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Human-Computer Interaction (cs.HC)
备注: The dataset and code are available at this https URL , and the predictor model is released at this https URL
Abstract:When foundation models describe people, recent work in AI fairness, accessibility, and ethics recommends avoiding inferred identity labels (e.g., “she”, “his”) in favor of seemingly “objective” physical descriptions (e.g., “short hair”, “a defined jawline”). Yet whether such descriptive language achieves gender-neutral communication remains an open empirical question. To study this, we introduce GAPA (Gender Associations of Physical Attributes), a dataset of 316 common physical attributes drawn from diverse sources, paired with 14,706 gender-association ratings from 304 US-based annotators. Results show that physical descriptions carry structured and graded gender associations among readers, with more consistent and distinctive associations for women and men than for non-binary identities. Next, we evaluate 16 LLMs across model families, sizes, and post-training variants against human ratings. The models partially recover human associations but exhibit systematic alignment biases, including compressed rating distributions, weaker alignment for associations with men, and asymmetric abstention that disproportionately targets the non-binary category. Finally, we release the best-performing proxy model trained to predict humans’ gender associations of descriptive language and demonstrate its utility through a sociolinguistic analysis of character descriptions in LitBank. Together, our findings provide the first empirical evidence that seemingly “objective” physical descriptions can retain systematic gender associations in human interpretation, and uncover systematic patterns of model-human misalignment. This challenges the assumption that replacing explicit gender labels with physical descriptions necessarily yields gender-neutral communication, and highlights downstream challenges in using such descriptions to communicate subjective identity categories in human-AI interaction.
[HC-30] From Momentary Emotion Inference to Sustained Emotion Support: Evaluating a Companion Agent in a Longitudinal Study
链接: https://arxiv.org/abs/2609.16344
作者: Kexin Quan,Zijian Ding,Jiaye Yong,Qinshi Zhang,Dong Wang,Jessie Chin
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注:
Abstract:Sustained emotional support is a long-horizon interaction task closely tied to human well-being. Recent research demonstrates generative agents’ capacity for momentary emotional support, yet how these capabilities sustain support over time remains unclear. To examine this challenge, we deployed PAIR, a theory-based emotion-regulation companion, with 19 participants for 14 days. Across 1,093 sessions, we paired emotion estimates with self-reports before and after guidance and analyzed logs and interviews. Estimates corresponded more closely to self-reported valence and dominance than arousal. Guided conversations were followed by higher valence and state-dependent arousal changes. Participants felt understood through contextual exploration and emotional acknowledgment, acting on guidance suited to their needs and constraints. Perceived helpfulness of guided conversation significantly increased over time. Our findings link memory updates and retained corrections to cross-session personalization, informing future emotional support tools that adapt to evolving needs, learn from prior outcomes, and preserve user control over memory.
[HC-31] Intelligent Interaction Techniques (IIxT) - Proposal
链接: https://arxiv.org/abs/2609.16295
作者: Brad A. Myers
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注:
Abstract:Interaction techniques (IxTs) are the low-level, reusable components out of which user interfaces are designed, including menus, scroll bars, text input fields, and also copy-paste, text-entry, and selecting objects. The IxTs for graphical user interfaces (GUIs) were well established in the 1980s, with relatively minor additions and tweaks for smartphones in the 2000s. Most of today’s AI user interfaces involve a chat window, which is an excellent interaction for some tasks, but is generally considered separate from the GUI IxTs. I argue for making the IxTs themselves more intelligent, so users can freely mix modalities, even within the same interaction. This will require research into new IxTs, and also into the infrastructure that will enable these intelligent IxTs (IIxTs) to be built. There are also significant security, privacy and economic implications to this vision.
[HC-32] SuperSenseDoctor: A Multimodal and Contactless Agent for Health Tracking
链接: https://arxiv.org/abs/2609.16257
作者: Xuwen Zhang,Zijian Lu,Yicheng Lei,Rui Qiu,Jiale Li,Yiping Zuo,Weibei Fan,Fu Xiao
类目: Human-Computer Interaction (cs.HC)
备注: 5pages,4figures,2tables
Abstract:Population aging is increasing the need to monitor older adults safely and independently at home. However, cameras, wearables, and manual checks often introduce privacy, adherence, and attention burdens that hinder sustained health monitoring. This paper presents SuperSenseDoctor, a multimodal contactless agent architecture for long-term home health tracking. The system transforms WiFi, mmWave radar, and surface temperature into a persistent human health state. The system relies on fixed decision rules to conduct continuous daily monitoring and respond to pre-defined hazards. When abnormal signals appear, event-driven reasoning analyzes only standardized evidence to produce traceable care-support measures. In this manner, SuperSenseDoctor integrates sensing, temporal state, reasoning, and action into a unified and auditable loop. The calibrated multimodal pipeline achieves 1.994 bpm mean absolute error (MAE) and 3.142 bpm root mean square deviation (RMSD) for heart rate, 0.197 bpm MAE and 0.263 bpm RMSD for respiratory rate, and 96.5% fall-recognition accuracy. The evaluation also covers 2686 one-second states across 9 chronological intervals and reaches a 96.7% criterion-level Agent checklist pass rate. These results demonstrate the feasibility of a stateful contactless sensing-to-action architecture for long-term home health monitoring.
[HC-33] When AI Says “I Am Unable to Answer”: Understanding User Responses to AI Refusals
链接: https://arxiv.org/abs/2609.16191
作者: Mahjabin Nahar,Eun-Ju Lee,Yujin Heo,Dongwon Lee
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:While refusal-based safeguards to mitigate hallucinations in large language models (LLMs) are becoming increasingly common, they may conflict with users’ preferences for definitive answers. However, we know little about how users respond to refusals across repeated interactions, when refusals become more or less acceptable, and for whom. In this work, we examine how refusal frequency, explanations, and need for cognitive closure (NFCC) shape responses to AI refusals. Participants (N=599) interacted with an AI system that never refused, refused infrequently, or refused frequently, with refusals either explained or unexplained. Participants were most satisfied with genuine responses, followed by hallucinations and then refusals, despite recognizing hallucinations as less accurate. Explanations increased satisfaction with infrequent, but not frequent, refusals. Higher-NFCC participants evaluated AI systems that refused more negatively. These findings reveal a tension between hallucination avoidance and user satisfaction and highlight the importance of designing balanced refusal strategies.
[HC-34] Moral Missions: Surfacing Moral Decision-Making Strategies for Responsible Data Science Practice
链接: https://arxiv.org/abs/2609.16166
作者: Teanna Barrett,B. Biira,Jainaba Jawara,Andrew Shaw,Ziwei Dong,Chinasa T. Okolo,Seyi Olojo,Keerthana Kompella,Khadija Saho,Amy X. Zhang,Leilani Battle
类目: Human-Computer Interaction (cs.HC)
备注: To be published in AIES 2026 without appendix
Abstract:A growing ecosystem of techniques, toolkits, and guidelines has been developed to help data scientists consider the social implications of data-driven technologies. However, prior literature highlights that even when this ecosystem of techniques is provided to professional data scientists, they still struggle to consistently adopt a responsible data science practice. We posit that the key to sustained responsible data science practice is to approach it as a moral mission: a conviction-driven technical practice that seeks to transform social conditions by any degree possible. In this paper, we present a semi-structured interview study with 15 responsible data scientists and AI practitioners to understand the moral decision-making procedures they use to articulate and actualize their moral missions. Through a phenomenological analysis of our participants’ accounts, we find participants engage in embodied introspection, circumvent institutional expectations, and center relationality throughout their moral missions. We also present how our participants engage in similar processes to contend with generative AI (GenAI) in their responsible practice. We conclude by calling for subversive data science communities and identifying sociotechnical design implications to better support sustainable responsible data science practice.
[HC-35] Pseudo-Label Augmentation for Affect Sensing in Small Collaborative Groups
链接: https://arxiv.org/abs/2609.16077
作者: Meisam Jamshidi Seikavandi,Tanya Ignatenko,Fabricio Batista Narcizo,Paolo Burelli,Jesper Bünsow Boldt,Andrew Burke Dittberner
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注:
Abstract:Physiological affect sensing in naturalistic group interaction is often limited by sparse labels rather than sensor data: wearable devices produce many time windows, while self-reports are collected only a few times per session. Using GroupAffect-4, a four-person collaborative dataset with wearable physiology, eye tracking, Big Five personality, and post-task VAD labels, we study pseudo-label augmentation for affect sensing under sparse supervision. We compare no augmentation, Gaussian Process pseudo-labelling, personality-aware trust weighting, and joint personality-plus-confidence weighting within a shared target-construction pipeline. Results show that pseudo-label augmentation improves over the labelled-only baseline in the known-team setting. However, the narrow range of Big Five cosine similarities (0.91-0.99) makes fine-grained personality weighting ineffective; personality similarity functions mainly as a same-team filter rather than a calibrated trust signal. With smoothing, augmented SVM variants are effectively tied on Valence and Arousal, while the joint personality-plus-confidence variant gives the highest Dominance score. Cross-subject LOSO transfer remains encouraging, especially for Arousal, whereas strict session-isolated LOGO removes the augmentation benefit. Given only 10 groups, LOGO should be interpreted as a conservative lower bound on unseen-group transfer. Overall, the results suggest that pseudo-label augmentation can make better use of sparsely labelled collaborative affect data, while personality information is most useful as a within-team selection mechanism.
[HC-36] Using Codebooks to Detect Cybercrime Topics in Text Narratives
链接: https://arxiv.org/abs/2609.16000
作者: Shufan Chai,Liangliang Sun,Jessica Staddon
类目: Computers and Society (cs.CY); Cryptography and Security (cs.CR); Human-Computer Interaction (cs.HC)
备注:
Abstract:In the United States, management of cybercrime-related consumer complaints increasingly falls on state and city governments given de-staffing of federal agencies. AI, and in particular, large language models (LLMs), shows promise for detecting cybercrime in text complaints, but often via specialized models that local governments are not resourced to develop and maintain. We present an LLM prompting method that uses codebooks from qualitative cybercrime research to detect cybercrime topics in consumer narratives. For two cybercrime topics, impostor scams and identity theft, we demonstrate the method achieves high precision and recall across multiple runs of 5 models in the Gemini and GPT model families. This strategy suggests a path for resource-constrained organizations, like many local governments, to leverage frontier models to support community safety.
计算机视觉
[CV-0] PhysStream: Streaming Physics-Grounded Video Generation with Structured Scene Memory and Fine-Grained Motion Control
链接: https://arxiv.org/abs/2609.17521
作者: Chuhao Chen,Peter Wonka,Chaoyang Wang,Chen Wang,Qiao Feng,Sergey Tulyakov,Lingjie Liu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Graphics (cs.GR)
备注:
Abstract:Interactive control for video generation is moving from coarse prompts toward fine-grained, physically meaningful manipulation of dynamic scenes. Yet existing controllable methods either require the full control schedule before generation starts, or use pixel-space signals that dictate object positions rather than physical dynamics. To address these limitations, we propose PhysStream, an autoregressive model for physics-grounded image-to-video synthesis that incorporates structured scene memory—positional maps and object tracking maps derived online from previously generated frames—and supports fine-grained motion control via sparse velocity-increment signals that encode physical quantities, letting the model learn the underlying dynamics. We train our model in two stages: a bidirectional model is first finetuned with motion-control conditioning, then a causal autoregressive model is trained with additional structured scene memory, further improving physical consistency. PhysStream enables interactive, mid-generation control over multi-object tabletop rigid-body scenes—a capability not supported by prior methods—reducing motion distribution distance (FVMD) by 33% and trajectory error by 12% over the strongest baselines on synthetic benchmarks, and is preferred by human evaluators in over 85% of in-the-wild comparisons. Please check our website for more details: this https URL
[CV-1] Det-LIME: Detector-Aware Multi-Instance Local Interpretable Model-Agnostic Explanations for Automated Marine Mammal Detection
链接: https://arxiv.org/abs/2609.17479
作者: Jiayi Zhou,David W. Johnston,Brinnae Bent
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Despite the rapid uptake of black-box object detectors in marine mammal research and monitoring, explainability techniques are rarely integrated into conservation workflows. Furthermore, most classification-oriented explainability tools are ill-suited to detection tasks involving imagery of social organisms or those with colonial life histories, as they ignore multiple detections within a scene and produce single-instance outputs that blur evidence across individuals. These methods also generate low-resolution, often biologically irrelevant visuals, limiting their utility for debugging, targeted data augmentation, and refined data collection. We proposed Det-LIME, a detector-aware, multi-instance adaptation of Local Interpretable Model-Agnostic Explanations (LIME) that produced instance-specific, box-aligned explanations by combining per-detection weighting, a proximity kernel that emphasizes regions near each box, and Intersection-over-Union-based matching to track the same instance across perturbations. We evaluated Det-LIME on aerial drone imagery for harbor seal detection, with an additional seabird case study to assess generality, and compared it with vanilla LIME, Stabilized LIME, Deterministic LIME, and gradient-based attribution methods. Using the Attribution Ratio and Max Saliency Hit Rate metrics, we showed that Det-LIME consistently improved multi-instance attribution. In practice, these higher-resolution, instance-aware explanations provide insight into model outputs and support post-processing, debugging, and actionable improvements in modeling and data collection or augmentation. Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) Cite as: arXiv:2609.17479 [cs.CV] (or arXiv:2609.17479v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2609.17479 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Related DOI: https://doi.org/10.1111/mms.70276 Focus to learn more DOI(s) linking to related resources
[CV-2] ables Decoded: DELTA for Structure TARQA for Understanding
链接: https://arxiv.org/abs/2609.17458
作者: Jahanvi Rajput,Dhruv Kudale,Saikiran Kasturi,Utkarsh Verma,Ganesh Ramakrishnan
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Accepted at the IEEE/CVF Winter Conference on Applications of Computer Vision 2026
Abstract:Table understanding is a core task in document intelligence, encompassing two key subtasks: table reconstruction and table visual question answering (TabVQA). While recent approaches predominantly rely on vision- language models (VLMs) operating on table images, we propose a more scalable and effective alternative based on structured textual representations. These representations are easier to process, align more naturally with LLMs, and eliminate the need for language-specific visual encoders, making them particularly suitable for multilingual documents. We present DELTA, which separates physical structure recognition, logical structure recognition, and OCR to extract both layout and content accurately. DELTA outputs tables in Optimised Table Structure Language (OTSL), a compact and unified format that encodes cell arrangements and textual content. On table structure recognition (TSR), DELTA achieves TEDS- Structure scores comparable with state-of-the-art methods across FinTabNet, PubTabNet, and PubTables-1M. We further establish its robustness on non-English tables through our curated Hindi benchmark, TORQUE. Building on this, we introduce TARQA, an LLM fine-tuned on OTSL sequences. Our approach yields gains of 9.3 p.p. on WTQ (TabQA) and 9.2 p.p. on FinTabNetQA (TabVQA), respectively. On TORQUE, our method ranks second among all VLMs and DELTA + LLM variants. We release our code, models, and benchmark at: this https URL
[CV-3] ORCA: Occlusion-Aware Refinement and Completion for Novel View Synthesis
链接: https://arxiv.org/abs/2609.17450
作者: Weronika Jakubowska,Maciej Zięba,Przemysław Spurek
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages, 3 figures
Abstract:Novel-view synthesis from a single image is a fundamentally ambiguous problem. As the camera moves away from the input viewpoint, previously hidden regions become visible, exposing missing geometry and holes in the reconstructed scene. Existing methods often rely on generative models to complete such regions. However, many of these artifacts are small gaps near depth boundaries and do not require generating new scene content. In order to eliminate expensive process of generating image we introduce ORCA, an occlusion-aware method for reconstructing and completing explorable 3D scenes from a single image. ORCA first introduces 3D structure into a Gaussian-anchor representation using monocular depth while preserving the original camera-ray correspondence. During scene exploration, missing regions are handled based on their size and structure. Small disocclusions are repaired using RGB-D information already available in the reconstruction, while generative inpainting is reserved for larger regions that cannot be reliably recovered from the scene. New Gaussian anchors are added and optimized locally without modifying the existing representation. By reducing unnecessary reliance on generative inpainting, ORCA limits generation-induced hallucinations and better preserves the content and structure of the original scene. On DIV2K, ORCA improves novel-view quality over VistaDream across all reported metrics, increasing MUSIQ from 61.60 to 68.71 and CLIP-IQA from 0.474 to 0.574. These results show that many novel-view artifacts can be repaired effectively by reusing information already present in the reconstructed scene. Comments: 9 pages, 3 figures Subjects: Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2609.17450 [cs.CV] (or arXiv:2609.17450v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2609.17450 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-4] BrainFocus: EEG-Guided ROI Selection for Efficient Vision-Language Models
链接: https://arxiv.org/abs/2609.17443
作者: Yihui Peng,Guorui Lu,Qinyu Chen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Vision-language models (VLMs) achieve strong visual question answering (VQA) performance, but processing large cluttered images is computationally expensive when only a small region is relevant. Electroencephalography (EEG) signals, which capture human neural responses to visual stimuli, can provide a human-derived semantic cue about the region of interest (ROI). However, EEG-guided visual category decoding remains imperfect, making direct ROI routing unreliable. In this work, we propose BrainFocus, a reliable EEG-guided efficient VLM framework for VQA. An EEG classifier predicts a target category, and a YOLO detector localizes the matching ROI. The VLM receives the cropped ROI only when both predictions pass confidence thresholds; otherwise, it processes the full image. For evaluation, we build on EEG-ImageNet to construct a 40-class benchmark comprising generated cluttered images and real object-centric images, with target-ROI annotations and 600 English visual question-answer pairs. Across Qwen3.5-VL 2B, 4B, and 9B models, BrainFocus improves VQA accuracy by 4.14-9.87 percentage points (pp) on cluttered scenes while reducing input tokens and total tokens by 23.2%-39.4% and 23.2%-39.3%, and end-to-end floating-point operations (FLOPs) by 23.2%-39.5%. These results demonstrate that EEG can guide efficient VLM inference even when its semantic decoding is imperfect.
[CV-5] racking the Unseen: An Occlusion-Robust Framework for Target Tracking Under Full and Long-Term Occlusion
链接: https://arxiv.org/abs/2609.17427
作者: Mais Mohammed,Sharifa Mohammed,Hanan Awadh,Haneen Bamaas,Raghad Bawazeer,Elham Alghamdi
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 25 pages, 10 figures, 7 tables
Abstract:Real-time multi-object tracking systems remain highly vulnerable to full and long-term occlusion, where targets temporarily or completely disappear from the camera’s field of view. Conventional trackers may terminate trajectories prematurely, resulting in identity loss and reduced situational awareness in applications such as defense and surveillance. This work proposes an occlusion-robust target tracking framework that maintains target identity and trajectory continuity through the integration of YOLOv11n object detection, Kalman Filter motion prediction, and occlusion-aware appearance-based re-identification. The framework consists of three stages: object detection, position estimation during occlusion, and identity recovery after target reappearance. Six Re-Identification (Re-ID) architectures were evaluated within the same tracking framework under identical conditions, with the Occlusion-Aware Mask Network (OAMN) achieving the best overall performance and therefore selected for the final pipeline. The framework was benchmarked against OccluTrack on the public OVIS dataset, achieving relative improvements of 18.1 percent in Multiple Object Tracking Accuracy (MOTA) and 25.1 percent in Identity F1 Score (IDF1), while reducing identity switches by 12.8 percent. On a custom military dataset simulating surveillance and battlefield-like environments with long-term occlusion, the framework achieved a MOTA of 0.734 and an IDF1 of 0.729, corresponding to relative improvements of 14.2 percent and 5.8 percent over OccluTrack. The system demonstrated strong tracking continuity, robust identity preservation, and reliable trajectory estimation under challenging occlusion conditions, highlighting its effectiveness for defense-related surveillance applications requiring continuous target tracking during visibility loss.
[CV-6] Knowledge as Orbit: Finite Collections as Phases of an Exactly Periodic Latent Generator
链接: https://arxiv.org/abs/2609.17417
作者: Siddharth Pal,Viktoria Rojkova
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注: 9 pages, 2 figures
Abstract:Finite knowledge is usually stored extensionally, one code or vector per item. We ask whether a finite collection can instead be stored intensionally, as the decoded orbit of one compact law that returns exactly to its start. For X objects, we encode item i as the i-th phase of a fixed rotation in a learned latent space and decode all phases with a shared network; the latent advances through a bank of rotations at integer harmonics of the cycle, a real discrete Fourier operator, so that R^X equals the identity and exact closure is guaranteed rather than learned. Images are a controlled carrier; looping video is the case where the phase order is the content’s own temporal structure. Holding the decoder fixed and varying only the operator, a general learned operator diverges, a norm-preserving but non-periodic one degrades around the loop, and the exactly periodic operator is flat; on real images the gap widens. Capacity is then the decoder’s budget: dense decoders carry a structural overhead per crisp image that no size reconciles with compression, while a small convolutional decoder on objects that share a manifold reaches crisp and compressed. A codebook control shows the generative law is free in reconstruction terms while multiplying the latent store many-fold. On seven benchmark clips, against a matched frame-index baseline, the cycle reaches equal or better fidelity at equal parameters while wrapping at machine precision, where the baseline leaves a visible seam; pinning the baseline’s frequencies to loop harmonics closes its seam too, confirming that exact periodicity is the operative constraint. Finite cyclic knowledge can be stored as dynamics rather than independent instances, with exact recurrence supplied by algebra and content by a shared decoder.
[CV-7] SlotDiT: Object-Centric Representations for Diffusion Transformers BMVC2026
链接: https://arxiv.org/abs/2609.17414
作者: Gjergj Plepi,Sven Behnke
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注: Accepted at BMVC 2026. Project page: this https URL
Abstract:Text-conditioned latent diffusion models perform strongly in video generation and are promising backbones for robotic applications. However, existing approaches rely on pixel-level or VAE-based latent representations that lack explicit semantic structure, leaving the impact of the representation space largely unexplored. Slot-based object-centric representations offer a structured alternative by decomposing scenes into object-level latents, or slots. While they have shown success in dynamics modeling and planning, they have not yet been explored for diffusion-based generative modeling. We introduce SlotDiT, a text-guided Diffusion Transformer (DiT) that operates in a slot-based latent space. Given a reference image and a language instruction, SlotDiT decomposes the scene into object-centric slots representing individual entities. Conditioned on the instruction and observed scene context, the model autoregressively denoises future slot trajectories to predict scene dynamics. To systematically investigate latent-space design for diffusion transformers, we compare slot-based representations against VAE-based and semantics-aligned alternatives within a unified DiT framework. Our experiments show that using slots as DiT latents yields competitive video generation quality while consistently improving task-completion rates across four robotic datasets. Furthermore, their compact representation provides a computationally efficient alternative to VAE-based and semantics-aligned latent spaces. Overall, our results demonstrate that object-centric structure is a powerful inductive bias for diffusion-based generative modeling in robotic environments. The project page is available at this https URL.
[CV-8] SSC-Priors: Exploring Semantic and Visibility Priors to Boost Lidar Semantic Scene Completion
链接: https://arxiv.org/abs/2609.17413
作者: Tetiana Martyniuk,Jonathan Seele,Alexandre Boulch,Gilles Puy,Renaud Marlet,Raoul de Charette
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Extended version of arXiv:2606.03992
Abstract:This paper investigates easy strategies to boost the performance of existing networks for lidar semantic scene completion (SSC) without requiring complex architectural redesigns. The fact is that, over the last years, SSC methods have mostly pursued architectural innovations, making the models heavier and more complex, e.g., by jointly training a point cloud semantic segmentation branch. In this work, we take a step back and explore two priors used as simple ingredients (possibly noisy) to improve existing approaches: semantic pseudo-labels and sensor visibility information. Concretely, we provide both kinds of information directly as additional inputs to a given SSC network, requiring only a minimal adaptation of the original architecture. We first demonstrate that endowing input point clouds with semantic pseudo-labels from off-the-shelf segmenters significantly improves the performance of existing SSC models. In fact, by evaluating these models against an oracle, we establish that high-quality semantic priors are a primary driver of semantic gains (mIoU), and that the SSC model can be trained just once with ground-truth semantics and then exploited without retraining using any segmenter. Furthermore, we equip the input lidar point cloud with visibility information that distinguishes between empty spaces (between the lidar and a scanned point) and unknown spaces (outside of lines of sight), providing a secondary performance boost across the tested architectures. We study the design space of data for representing visibility information and bound the remaining headroom with a ground-truth oracle on the free-space labels. On SemanticKITTI, these enhancements make older models competitive with state-of-the-art systems across four architectures, in one case even outperforming them. On the SSCBench-nuScenes benchmark, both priors also transfer with the sparser 32-beam sensor.
[CV-9] PanoGS-SLAM: Panoramic 3D Gaussian Splatting SLAM
链接: https://arxiv.org/abs/2609.17387
作者: Yongqi Mao,Hao Shi,Yufan Zhang,Zhonghua Yi,Xiangfei Guo,Kaiwei Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Real-time dense SLAM is a core capability for robotics applications that require robust localization and high- quality mapping in dynamic or fast-changing environments. Recent 3D Gaussian Splatting (3DGS)-based SLAM methods have shown promising performance, but most are designed for narrow-FoV pinhole cameras, where limited angular coverage weakens pose observability and often leads to unstable photo- metric optimization under rapid motion and large viewpoint changes. We present PanoGS-SLAM, the first panoramic dense SLAM system built on 3D Gaussian Splatting. Our method per- forms differentiable rendering and pose optimization directly in the spherical domain, enabling omnidirectional photometric constraints for more stable tracking. To improve geometric consistency and robustness, we introduce (1) a sphere-consistent photometric loss that compensates for the area distortion of equirectangular projection, and (2) a depth-guided Gaussian initialization strategy that stabilizes incremental mapping in newly observed regions. Extensive experiments on both real and synthetic panoramic benchmarks (PALVIO and SynPano) show that PanoGS-SLAM consistently outperforms geometric and GS-based baselines in tracking accuracy and rendering quality, while achieving fast front-end convergence and real-time perfor- mance. In addition, controlled field-of-view experiments reveal a clear monotonic improvement in optimization conditioning and convergence stability as angular coverage increases, high- lighting the fundamental role of sensing geometry in shaping the optimization landscape of differentiable Gaussian-based SLAM. The source code will be made publicly available.
[CV-10] Optical-Flow Wingbeat Counting in MuJoCo: A Comparison of Convolutional Spiking and Attention-Based Temporal Models
链接: https://arxiv.org/abs/2609.17308
作者: Zhang Nengbo
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 12 pages, 1 figure, 7 tables. Controlled simulation study
Abstract:Visual monitoring of flapping-wing vehicles requires distinguishing individual wingbeats from motion strength and average frequency. This paper presents a controlled MuJoCo evaluation of wingbeat counting from signed optical flow observed by virtual cameras mounted on Crazyflie vehicles. Three flapping-wing models were recorded at optical distances of 1.5 and 3.0 m, producing 1,440 clips from 240 paired scene configurations with a scene-level 3:1 training-test split. A common spatial convolutional encoder was combined with a causal temporal convolutional network, a recurrent leaky integrate-and-fire spiking network, or causal self-attention. Each model predicted phase and activity, followed by the same directed-crossing event counter. The six existing convolutional models were retained, and all twelve new models were frozen before their test predictions were generated. Exact-count accuracies at 1.5 m were 96.67%, 95.00%, and 96.67%, respectively; at 3.0 m they were 94.44%, 92.22%, and 95.00%. All paired scene-bootstrap intervals for differences in exact-count accuracy included zero. Seven far-distance spiking-model clips had correct totals despite event-timing mismatches, demonstrating why total-count and event-level measurements must be reported together. The results support the feasibility of causal optical-flow counting in the tested setting and identify boundary-sensitive errors. They do not establish an architecture ranking across repeated training, real-flight robustness, or hardware efficiency.
[CV-11] Semantic-Spatial Agreement Verification for Mitigating Object Hallucination in Multimodal Large Language Models
链接: https://arxiv.org/abs/2609.17269
作者: Ziheng Ren,Qian Gao,Jun Fan,Guohui Ding,Zhenyu Yang,Yuteng Xiao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Multimodal large language models generate natural-language responses from visual inputs, yet may mention objects absent from an image. In medication assistance, accessible perception, and environmental decision-making, such hallucinations can create real-world safety risks. We propose Semantic-Spatial Agreement Verification (SSAV), a training-free method for verifying object claims. A visually grounded claim should remain stable across semantically equivalent queries and repeatedly localize to the same image region. SSAV aggregates multiple prompts to estimate semantic support and reduce sensitivity to query wording. Query-Induced Regional Verification (QIRV) combines cross-query region persistence, spatial overlap, and relative candidate dominance to identify isolated high responses and dispersed localizations. A geometric mean fuses semantic and spatial evidence, lowering the verification score when either branch lacks support. Experiments on three base models and multiple evaluation protocols show that SSAV effectively mitigates object hallucination. On LLaVA-1.5-7B, accuracy averaged across COCO, A-OKVQA, and GQA improves by 1.81 and 3.17 percentage points under POPE Popular and Adversarial, respectively, while CHAIRs decreases from 49.40% to 32.80%. These results show that cross-query semantic stability and regional consistency provide interpretable external visual evidence for object claims.
[CV-12] Exploring 2D backbone effects for indoor semantic occupancy prediction
链接: https://arxiv.org/abs/2609.17257
作者: Shizhang Fanga,Wanling Yea,Qi Zheng
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Semantic occupancy prediction gives an embodied agent a voxel-level account of where space is free, occupied, and semantically meaningful. In RGB-D pipelines such as EmbodiedScan, the image encoder is often left as a default module, even though its features are the visual evidence later sampled into the 3D grid. We study this design choice directly. A central finding is that changing the 2D backbone improves occupancy accuracy more than several carefully designed occupancy architectures or modules. We keep the main RGB-D projection, depth branch, and occupancy head fixed, and replace only the image backbone. The compared encoders are CLIP-ResNet, CLIP-ViT, BLIP2, and DINOv2. Under the controlled setting, the measured mIoU changes substantially: DINOv2 obtains 30.55%, BLIP2 obtains 29.49%, CLIP-ViT obtains 24.33%, and CLIP-ResNet obtains 17.41%. The stronger encoders also exceed the original EmbodiedScan ResNet-50 baseline without modifying the downstream 3D fusion pipeline. Class-level results give a more detailed picture: DINOv2 is stronger on many layout and structural categories, whereas BLIP2 remains close on several object-centered classes. CLIP-ViT improves clearly over CLIP-ResNet, showing that the way CLIP features are exposed as dense tokens matters for voxel lifting. These results indicate that the image backbone is not a secondary engineering detail in embodied semantic occupancy, but a major source of variation in the final 3D prediction.
[CV-13] Video-HolmesV2: Can MLLM s Reason with Spatio-Temporal Audio-Visual Evidence in Long Videos? ECCV2026
链接: https://arxiv.org/abs/2609.17248
作者: Zhaoyang Wei,Zipeng Wang,Yushe Cao,Chenhui Qiang,Shuaibing Cheng,Xuesong Yang,Sen Nie,Bowen Jiang,Wenchao Ding,Yanchao Hao,Zheng Wei,Xuehui Yu,Zhenjun Han
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by ECCV2026
Abstract:Multimodal Large Language Models have demonstrated impressive video understanding, yet their ability to reason over long-form narratives is often masked by visual-centric evaluations and inefficient context processing. Existing benchmarks over-rely on visual heuristics while marginalizing auditory cues, effectively reducing models to “silent observers” that bypass genuine cross-modal reasoning. Moreover, standard dense sampling creates an evidence-context trade-off: increasing frames to capture evidence inevitably leads to attention distraction and token explosion. To bridge these gaps, we present Video-HolmesV2, a novel benchmark designed for Deep Audio-Visual Coupling. Unlike previous works, it enforces an Evidence-Based Evaluation, requiring models to justify answers with precise spatio-temporal audio-visual evidence, thereby reducing confounding effects of guessing and hallucinated evidence. To support this, we introduce: (1) a Multi-Model Cross-Verification pipeline to ensure task rigor; (2) a Spatio-temporal Evidence-Aware Metric for fine-grained calibration. Furthermore, we propose an Audio-Text Guided Token Compression framework. By fusing task intent with auditory anchors, our method distills high-value reasoning cues to mitigate long-context noise. In our evaluation, even strong proprietary models achieve below 60% accuracy, while our approach outperforms comparable open-source omni-models.
[CV-14] DecoGS: Adaptive Static-Dynamic Decoupling of 3D Gaussians for Free-Viewpoint Video Streaming
链接: https://arxiv.org/abs/2609.17230
作者: Idil Sulo,Alexey Supikov,Ilke Demir,Sainan Liu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Streaming 3D reconstruction demands both speed and temporal fidelity, goals that existing methods undermine by updating every Gaussian every frame, even in static regions. We present DecoGS, a method for efficient online training of 3D Gaussians from streaming videos. Unlike prior methods that update the entire scene indiscriminately, DecoGS introduces an adaptive mechanism that selectively focuses optimization on spatiotemporal regions exhibiting motion or photometric changes. This targeted training strategy eliminates redundant updates that cause flickering and drift in nominally static regions, while enabling fast, high-fidelity scene updates. The pipeline further integrates region-aware Gaussian management through gradient gating and efficient visibility filtering to maintain temporal coherence and a compact memory footprint. On N3DV and MeetRoom, DecoGS achieves 34.55 and 31.60 dB PSNR respectively, outperforming all streaming and offline baselines, while rendering at 261 FPS with 70\times lower temporal flicker than the best prior method, requiring no large-scale pretraining.
[CV-15] FROD: Feature Matching Residual Denoising Oracle Bone Decipher ICONIP2026
链接: https://arxiv.org/abs/2609.17227
作者: Yanbin Hou,Biao Xiong,Guojun Xu,Jianwen Xiang,Cheng Tan,Yanchao Yang,Junwei Zhou
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 15 pages, 5 figures, 3 tables. Accepted at ICONIP 2026
Abstract:Oracle bone script (OBS), one of the earliest Chinese writing systems, plays an important role in the study of Chinese etymology. Traditional decipherment relies heavily on domain experts who analyze characters through semantic context and structural evolution. To assist this labor-intensive process, we formulate OBS decipherment assistance as a cross-era image translation task and propose FROD (Feature Matching Residual Denoising Oracle Bone Decipher). Although many OBS characters differ substantially from their modern counterparts, they often preserve local topological invariants at the radical level. During training, FROD leverages fast feature matching to provide gated segmentation supervision: paired samples with sufficient matches are processed patch-wise to align fine-grained radicals, whereas low-similarity pairs are trained holistically to avoid mismatched artifacts. In addition, a Residual Denoising Diffusion Model (RDDM) jointly estimates noise and residual signals, thereby reducing the positional drift and stroke disorder commonly observed in standard diffusion models. Finally, a multi-stage font stylization refinement network refines the generated images by eliminating edge noise and stabilizing stroke structures. On our augmented character-disjoint dataset, FROD achieves higher Top-1 recognition accuracy than the evaluated baselines, with a 3.8% absolute gain over OBSD.
[CV-16] InfoTaxa: Information-Calibrated Label-Free Clustering for Fine-Grained Visual Taxonomy
链接: https://arxiv.org/abs/2609.17218
作者: David Ahmedt-Aristizabal,Mohammad Ali Armin,Lars Petersson
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Label-free clustering of frozen pretrained visual embeddings offers a scalable route to biodiversity monitoring, but image-only fine-grained taxonomy exhibits a consistent coarse-to-fine failure mode: clusters recover broad taxonomic structure yet plateau at species level. We study this behaviour on BIOSCAN-5M through an information-calibrated clustering analysis. BioCLIP~2 features with UMAP and HDBSCAN reach 0.79 AMI at family and 0.67 at genus, substantially improving over the prior image baseline and remaining competitive with oracle- K , graph-based, and learned clustering heads on the same frozen features. To diagnose whether the remaining plateau is method-limited or information-limited, we introduce InfoTaxa, which combines clustering efficiency—the fraction of probe-estimated image information recovered by an unsupervised partition—with paired DNA as an audit signal only, not an inference input. The density pipeline recovers approximately 0.90 and 0.81 of the image-available information at order and family, respectively. Held-out late-fusion probes show that adding DNA to the image embedding reduces species-level prediction error by approximately two bits. Robustness analyses cover multiple image encoders, described-species and rare-class subsets, probe diagnostics, and held-out-species coarse-rank generalisation and same-species retrieval. Thus, in the tested setting, species-level label-free clustering is both clustering-limited and representation-limited: improved clustering may recover additional image-exposed structure, but cannot close the DNA-audited information gap alone.
[CV-17] Probe-VAD: Ordinal Likelihood Probing for Training-Free Video Anomaly Detection
链接: https://arxiv.org/abs/2609.17211
作者: Jiawei Gu,Qilin Zhao,Tengkuo Guo,Zhiming Zhong,Shuangqing Zhang,Fan Lyu,Fang Zhao,Guo-Sen Xie,Caifeng Shan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Under Review
Abstract:Video anomaly detection (VAD) aims to localize anomalous events in untrimmed videos. Vision-language models (VLMs) provide rich visual understanding for training-free VAD, but existing approaches impose restrictive interfaces between visual understanding and anomaly scoring. Caption-based pipelines compress visual evidence into text, potentially discarding subtle cues, while direct numerical generation forces the model to express its judgment through a small set of predefined scores. Such interfaces can obscure subtle differences in anomaly severity, causing visually distinct clips to receive similar representations or scores and thereby limiting the resolution of anomaly ranking. We propose \textbfProbe-VAD, an ordinal binary-probing framework that directly probes severity preferences from a frozen VLM. Given raw video clips, Probe-VAD queries ten ordered severity thresholds and extracts constrained \textitYES/\textitNO continuation likelihoods. Their normalized preferences form a cumulative severity profile, from which tail evidence is aggregated into a continuous anomaly score, with isotonic projection enforcing ordinal consistency. Experiments on public VAD benchmarks demonstrate superior performance with low computational cost. Probe-VAD provides a simple interface for translating frozen VLM visual understanding into continuous, rank-sensitive anomaly scores without task-specific training or caption-based compression. Code is available at: this https URL.
[CV-18] EventEgoHands: Event-based Egocentric 3D Hand Mesh Reconstruction with Real Dataset
链接: https://arxiv.org/abs/2609.17189
作者: Ryosei Hara,Wataru Ikeda,Masashi Hatano,Mariko Isogawa
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to IEEE Access. Project Page: this https URL
Abstract:3D hand mesh reconstruction is a challenging yet essential task for downstream applications, including human-robot interaction and AR/VR. Although conventional cameras have been widely adopted for this task, methods that rely on them struggle in low-light environments and under severe motion blur. To address these limitations, event-based cameras have recently attracted attention for their high dynamic range and high temporal resolution. However, applying event cameras to egocentric hand reconstruction remains challenging because camera wearer’s motion produces dense background events that obscure hand-specific signals. Although the first egocentric event-based approach mitigates this issue using hand segmentation, its binary hand mask does not distinguish between left and right hands. As a result, the model lacks instance-level hand information and predicts both hands even when only one or neither hand is present. This limitation leads to incorrect inter-hand relationships and degraded reconstruction accuracy. In this paper, we propose EventEgoHands++, a framework for event-based 3D hand mesh reconstruction from an egocentric viewpoint. The proposed method incorporates a Hand Detector that estimates instance-level bounding boxes and masks for both the left and right hands. Moreover, we introduce Adaptive Attention, which dynamically gates the attention based on these detection results to accurately learn the spatial relationship and mutual interactions between the hands. To train and evaluate our framework, we extend the synthetic N-HOT3D dataset and newly construct EEH-R, the largest real-world event-based egocentric hand dataset to date, comprising approximately 1M annotated frames captured in environments including low-light conditions. Extensive experiments on both synthetic and real datasets demonstrate that our method consistently outperforms the baselines.
[CV-19] Multimodal Cultural Heritage Architectural Style Classification for Residential Buildings in the UAE Based on CLIP Embeddings and SVM
链接: https://arxiv.org/abs/2609.17181
作者: Ahmed Ammar Kubba,Manar Abu Talib,Iman Ibrahim,Qassim Nasir
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 8 pages, 8 figures, 3 tables, published at 15th International Conference on Intelligent Systems: Theories and Applications
Abstract:The analysis and classification of cultural heritage architectural styles remain challenging due to the complexity of visual images of buildings, which are highly relied on in traditional CNN-based classification approaches in comparison to textual descriptions, and the relative lack of non-western region-specific datasets. This paper addresses this gap by proposing a multimodal machine learning framework to analyze and classify Emirati residential architecture using OpenAI’s CLIP model. We integrate visual features from images and textual features from expert descriptions into a unified 512-dimensional embedding, followed by dimensionality reduction with UMAP for visualization and unsupervised clustering using K-Means. Cluster labels, which are derived from manual analysis of the K-Means clusters, are used to train an SVM classifier for automated architectural style classification. Our approach achieves a classification accuracy of 98% across eight identified style clusters, higher than every other study in the literature, demonstrating the effectiveness of combining visual and textual modalities. Overall, this paper highlights the potential of using multimodal AI to support architectural heritage analysis, offering scalable and interpretable tools for exploring regional architectural identities.
[CV-20] MUMINS: Metadata-conditioned Uncertainty-aware Medical Image Next-state Synthesis
链接: https://arxiv.org/abs/2609.17169
作者: Anna Oliveras,Roger Marí,Rafael Redondo,Oriol Guardià,Cynthia Ifeyinwa Ugwu,Ana Tost,Bhalaji Nagarajan,Carolina Migliorelli,Vicent Ribas,Petia Radeva
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Supplementary material to follow in future versions
Abstract:Forecasting anatomical changes such as tumor growth and neurodegeneration is a challenging generative vision task. Morphological evolution is subtle relative to static anatomy, highly patient-specific, and inherently stochastic. Existing methods struggle with several issues: deterministic networks ignore biological stochasticity, while standard diffusion models require computationally prohibitive multi-pass sampling to quantify uncertainty. We propose MUMINS (Metadata-conditioned Uncertainty-aware Medical Image Next-state Synthesis), an efficient diffusion framework that jointly diffuses a baseline scan and its follow-up residual, summed to synthesize the follow-up scan, while concurrently predicting a spatial uncertainty map, in a single reverse diffusion process. Conditioned on the time interval and relevant metadata, it preserves fine-grained anatomy by dynamically re-injecting the baseline as a soft anchor at every denoising step, and a negative-log-likelihood head learns the uncertainty map to explicitly flag error-prone regions. Designed without organ-specific heuristics, the same architecture is reused across anatomies via separate, dataset-specific retraining. Extensive evaluations demonstrate that dataset-specific retraining of MUMINS matches or outperforms dedicated, domain-specific state-of-the-art methods on lung CT (PNG) and brain MRI (OASIS-3). Project page: this https URL.
[CV-21] HuMemSLAM: Efficient Human-Inspired Semantic Place Recognition for Robust Visual SLAM
链接: https://arxiv.org/abs/2609.17168
作者: Mayowa Adebambo,Sebastian Donnelly,Armand Amaritei,Andrew Bradley,Alexander Rast
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 8 pages, 8 figures
Abstract:Autonomous systems require reliable place recognition for efficient and effective simultaneous localisation and mapping (SLAM). Traditional geometric visual SLAM approaches rely on low-level features and geometric consistency, but remain vulnerable to perceptual aliasing, where different places appear similar, and perceptual variation, where the same place appears different. Although semantic SLAM and modern learned visual place recognition (VPR) methods improve robustness under challenging perceptual conditions, real-time deployment requires both high retrieval accuracy and low latency. Inspired by human memory and perception, we propose HuMem-VPR, which exploits the bidirectional relationship between bottom-up perceptual evidence and top-down contextual reasoning to achieve high-level place understanding. We further introduce HuMemSLAM, the integration of HuMem-VPR with ORB-SLAM3. HuMem VPR achieved the highest aggregate retrieval accuracy on the real-image benchmark, competitive accuracy on the CARLA benchmark, and approximately two to three times lower latency than the evaluated state-of-the-art VPR methods. Across the evaluated dataset families and online experiments, HuMemSLAM substantially improved integrated Recall @1 over ORB-SLAM3’s native retrieval while reducing the proposals submitted to its geometric backend.
[CV-22] ResLRP: The Role of Residual Cancellation in Attribution Instability in Vision Transformers
链接: https://arxiv.org/abs/2609.17152
作者: Jim Berend,Reduan Achtibat,Daniel Schäffer,Alexander Binder,Wojciech Samek,Sebastian Lapuschkin,Maximilian Dreyer
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Vision Transformers (ViTs) are central to most modern vision models, yet obtaining input attributions that are fine-grained, faithful, and stable remains challenging. Layer-wise Relevance Propagation (LRP) has been adapted to transformer attention, but in ViTs it often produces noisy, unfaithful explanations. We show that the missing ingredient is the treatment of residual connections: cancellation effects in residual pathways lead to attribution explosion. Moreover, we find that these cancellations are substantially stronger in ViTs than in language transformers. To address this issue, we introduce Residual-aware Layer-wise Relevance Propagation (ResLRP), a simple extension of LRP whose propagation rules explicitly account for cancellations in residual branches, are exactly conservative, and provably bound relevance explosion. Causal channel-wise interventions confirm that residual cancellation, not a generic regularization effect, drives the instability. ResLRP substantially improves attribution quality across faithfulness and localization, evaluated on ViT architectures spanning supervised, self-supervised, contrastive, hierarchical, and multimodal families, as well as on the ground-truth-controlled FunnyBirds benchmark. The largest gains arise in modern Vision Language Models (VLMs), with +27-29% localization and up to 3.4x faithfulness scores. Beyond benchmarks, ResLRP localizes Sparse Autoencoder (SAE) features in input space, and our residual amplification measure serves as an architecture-level diagnostic predicting where attribution degrades.
[CV-23] From Foundation Embeddings to Cropland Maps: Label Efficiency Temporal Transferability and Independent Human Validation
链接: https://arxiv.org/abs/2609.17138
作者: Mohammad Ammar Mughees,Giovanni Montefoschi,Zhongxin Chen,Maria Antonia Brovelli
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Image and Video Processing (eess.IV)
备注: 23 pages, 10 figures. Code: this https URL
Abstract:Geospatial foundation models provide reusable representations of satellite imagery that support downstream mapping with limited task-specific modelling. We evaluate whether annual AlphaEarth embeddings support binary cultivated-versus-non-cultivated mapping in Maine, USA, using 192 spatially separated patches and labels derived from the USDA Cropland Data Layer (CDL). Without fine-tuning the foundation model, a lightweight classifier reaches 93.7% overall accuracy and 90.8% balanced accuracy on held-out patches. Logistic regression is within 0.3 percentage points of a gradient-boosted ensemble, while a nearest-class-centroid rule, which uses class centroids but fits no parameters, reaches 90.2%. A balanced sample of 60,000 labelled pixels is within 1.3 percentage points of the full pool of 8.6 million pixels; because pixels are spatially autocorrelated, this result concerns pixel-sample efficiency rather than 60,000 independent annotation sites. In a same-region transfer experiment, classifiers trained in one year remain accurate across 2018 to 2023. Against a blind, two-interpreter consensus at 385 randomly sampled points in one contiguous 2023 block, the AlphaEarth-plus-random-forest map agrees at 95.3% ( \kappa=0.82 ), compared with 91.7% for the CDL ( \kappa=0.72 ; exact two-sided McNemar p=0.0161 ). This local result is consistent with partial smoothing of CDL label noise, but it does not establish statewide correction of the reference product. On the same points, the difference from a fine-tuned TerraMind segmentation model is not statistically significant (95.3% versus 93.5%; p=0.14 ), and the experiment is not a controlled comparison of computational cost. These results support frozen geospatial embeddings as a low-compute candidate for regional cropland mapping, subject to the limits of a single-state study, a 30 m-derived training reference, and a one-block human validation.
[CV-24] Event-based Selective Attention for Multi-resolution Fast Region of Interest (ROI) Detection
链接: https://arxiv.org/abs/2609.17134
作者: Luca Peres,Giulia D’Angelo,Chiara Bartolozzi,Oliver Rhodes
类目: Computer Vision and Pattern Recognition (cs.CV); Neural and Evolutionary Computing (cs.NE)
备注:
Abstract:Neuromorphic vision systems operate under strict constraints on bandwidth, memory, and energy, particularly at the edge, motivating early mechanisms for data reduction and selective processing. In this work, we investigate a multi-scale training-free, saliency-based, bottom-up visual attention model that operates directly on low-resolution event-based input and selects Regions of Interest (ROI) from the visual scene. The model is evaluated across multiple downscaling factors applied to the incoming event stream, with input resolutions reduced by up to 256x relative to full resolution. Performance is assessed on the Prophesee Automotive dataset, the largest publicly available event-based dataset, demonstrating robust ROI selection across different scales on a real-world use-case. The proposed approach is capable of detecting ROIs belonging to multiple object classes, including various vehicle types, pedestrians, traffic lights, and traffic signs, with accuracy up to 70.8%, while operating at millisecond temporal resolution, 16x finer than the temporal resolution provided by the dataset ground truth. These results highlight the potential of combining early event downscaling with saliency-based attention as an effective front-end for efficient edge neuromorphic vision systems.
[CV-25] Predicting Human Disagreement for Calibrated Dynamic Facial Expression Recognition ICASSP2027
链接: https://arxiv.org/abs/2609.17130
作者: Yiming Wang,Frederick W. B. Li,Jingyun Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 5 pages, 3 figures, 4 tables. Submitted to ICASSP 2027
Abstract:Dynamic facial expression recognition (DFER) benchmarks such as DFEW provide multiple annotator votes per clip, yet most models collapse them to a majority label and cannot represent human disagreement at inference time. We propose a disagreement-aware DFER framework that trains directly on the raw annotator count vector using a Dirichlet-Multinomial likelihood. Unlike mean-only soft-label objectives, the proposed likelihood provides scale-sensitive supervision for the Dirichlet concentration while preserving the predictive mean. A separate ambiguity head predicts annotation entropy for unseen clips, and a monotone Chow-style reject rule combines predicted ambiguity, vacuity, temporal instability, and input quality for selective prediction. On DFEW, the method preserves recognition accuracy while reducing ECE by 30% and AURC by 15%, and predicted ambiguity reaches a Spearman correlation of 0.52 with the annotation entropy of test clips. The calibration and selective-prediction gains transfer to FERV39k and remain under identity- and movie-disjoint DFEW splits.
[CV-26] Not Another Text Benchmark: Putting the “Visual” Back in Visual Question Answering for Large Video Models
链接: https://arxiv.org/abs/2609.17112
作者: Rwiddhi Chakraborty,Yinong(Oliver)Wang,Cheng Zhang,Fan Bai,Zhuoran You,Michael Kampffmeyer,Yong Jae Lee,Fernando De la Torre,Robert Jenssen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Large video models have exhibited impressive performance on a wide range of visual question answering tasks, owing to the rise of powerful, pretrained text and vision encoders. The usefulness of such models have also been demonstrated on a wide range of benchmarks, with an important caveat - the dominant approach in these benchmarks evaluates multiple choice reasoning via text options. This is a natural way to test text-based reasoning in these models, and has led to significant insights regarding model behavior in the community. In this work, we ask a different question - what happens when the evaluation modality is visual, rather than text? We introduce three new vision-centric evaluation benchmarks in temporal frame retrieval, video future prediction, and causal memory distortion, all designed around evaluating visual understanding capabilities in large video models. Our approach complements the existing approaches to evaluate video understanding in frontier models. We show that current frontier models exhibit significant weakness when attempting to reason through visual queries, rather than text. We conclude with an extended analysis section that provides pointers for future improvements in visual understanding for large video models.
[CV-27] GeoLAM: Learning Geometry-Grounded Latent Actions from Unlabeled Human Videos
链接: https://arxiv.org/abs/2609.17099
作者: Yifan Xie,Hekun Tian,Jinkun Liu,YuAn Wang,Qiao Sun,Wenbo Ding
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注: 8 pages, 6 figures, 4 tables
Abstract:Human videos provide rich manipulation experience, but extracting action representations that preserve useful motion remains challenging. Visual reconstruction alone can entangle manipulation-related motion with appearance changes and camera movement. We present GeoLAM, a framework for learning geometry-grounded latent actions from action-free human videos. GeoLAM combines future-frame reconstruction through a frozen geometric feature hierarchy with motion supervision from a training-only 4D geometry teacher. The geometric representation provides a structural prior, while the teacher’s predictions yield spatially pooled targets capturing 3D displacement, residual image-plane motion, and surface-orientation changes. Visibility and confidence weighting reduces the contribution of unreliable estimates, encouraging continuous latent actions to retain geometric motion without explicit hand-pose or hand-trajectory annotations. After video pretraining without action labels, the learned representation provides transition targets for a world-action model trained on action-labeled robot demonstrations. The model jointly denoises latent actions and executable action chunks, with future-video prediction used only as an auxiliary training task. Deployment therefore requires neither the geometry teacher nor future-video generation. Evaluations on a latent-action benchmark and robotic manipulation tasks demonstrate the strong performance of GeoLAM.
[CV-28] Hub-Spectral Activation of Latent Multimodal Knowledge
链接: https://arxiv.org/abs/2609.17094
作者: Ying Guo,Haidong Chen,Linrui Xu,Xiaohao Liu,Chuancheng Shi,Canran Xiao,Dan Zhang,Fei Shen,Li Shen,Tat-Seng Chua
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 30 pages, 9 figures, including appendices
Abstract:Multimodal representation learning seeks shared representations for cross-modal retrieval and knowledge transfer. Hub-based binding reduces pairwise supervision costs, but separate hub connections cannot guarantee reliable alignment between modalities without direct joint training. We introduce Hub-Spectral Activation (HSA), a closed-form method for recovering and activating the hub-readable component of latent multimodal knowledge in frozen representations. We formalize this knowledge as source-induced cross-modal dependence and characterize the component determined by the second-order statistics of two trained hub edges. Under a second-order source model, we establish conditions for exact recovery of the complete source-induced relation and bound the dimension of its hub-readable component by the hub covariance rank. HSA composes and standardizes hub-edge statistics, extracts paired spectral directions, and combines reliability-weighted matching evidence with source-gated candidate resolution for bidirectional retrieval and prototype classification. HSA requires no target-pair supervision, gradient optimization, or backbone updates. Across 19 retrieval and 11 prototype-classification relations on ImageBind and LanguageBind, HSA raises mean bidirectional Recall@10 from 18.27% to 31.15% and mean macro Top-1 accuracy from 29.01% to 52.43%, respectively. Controlled analyses further identify valid hub-edge correspondence and leading spectral directions as key sources of retrieval gains, demonstrating the utility of latent multimodal knowledge beyond native similarity scores. Code and models are publicly available at this https URL.
[CV-29] Beyond In-Distribution Metrics: A Systematic Out-of-Distribution Evaluation of Congenital Heart Disease Segmentation MICCAI2026
链接: https://arxiv.org/abs/2609.17068
作者: Aniketh Vijesh,Shrisharanyan Vasu,Abhijit Ramesh,Clare Pomeroy-Ward,Harikrishnan Anil Maya,Sarin Xavier,Mahesh Kappanayil,Gilad Gressel
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 12 pages, 6 figures, 2 tables. Accepted at STACOM 2026, held in conjunction with MICCAI 2026
Abstract:Congenital heart disease (CHD) diagnosis and surgical planning often require patient-specific 3D anatomical models, but manual segmentation is labor-intensive, particularly in complex anatomies. Although deep-learning methods can automate this process, they are typically evaluated in-distribution, despite clinically relevant shifts in scanner, protocol, institution, population, and imaging modality. We present, to our knowledge, the first systematic evaluation of out-of-distribution (OOD) generalization in CHD segmentation, using ImageCHD as a held-out target cohort. We compare representative segmentation architectures under combined CT and CMR training, CT-only training, self-supervised pretraining, and limited target-domain adaptation. In-distribution performance proves to be a poor indicator of cross-cohort robustness: nnU-Net achieves the highest validation Dice (0.77) but falls to 0.51 on ImageCHD, while SwinUNETR generalizes substantially better, reaching 0.67 Dice. MAE and JEPA pretraining provide only modest additional benefit, suggesting that architecture contributes more to robustness than the tested pretraining strategies in this setting. When limited target-domain supervision is introduced, all SwinUNETR variants exceed 0.76 Dice with only 11 labeled ImageCHD cases. These findings demonstrate that conventional in-distribution evaluation can obscure clinically important generalization failures and support explicit cross-dataset testing as a key component of CHD segmentation evaluation.
[CV-30] Bi-FlowGS: Bridging Generative View Completion and Gaussian Geometry through Bidirectional Flow Co-Refinement
链接: https://arxiv.org/abs/2609.17039
作者: Yuetong Wang,Jinsheng Quan,Yi Yang,Yawei Luo
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Sparse-view 3D scene reconstruction with 3D Gaussian Splatting (3DGS) is inherently underconstrained. Plausible renderings can also coexist with erroneous Gaussian geometry, as errors in positions or depths may be concealed by opacity, scale, and appearance; we term this failure mode Geometry Cheating. Existing regularization methods constrain geometry but remain limited to observed views, while video-diffusion-based methods complete unseen views yet mainly use them as RGB pseudo-supervision, underusing motion and temporal priors and lacking explicit geometry supervision. We present Bi-FlowGS, which uses optical flow to bridge generative view completion and Gaussian geometry regularization. Our plug-and-play Video-to-Geometry Flow Distillation (V2G) distills temporal correspondence priors from restored videos into Gaussian geometry to alleviate Geometry Cheating. Conversely, Geometry-to-Video Flow-Guided Restoration (G2V) uses the current 3DGS geometry to guide temporally consistent video restoration, providing more reliable generative supervision. Together, V2G and G2V form an implicit bidirectional co-refinement process, enabling restored videos and the optimized 3DGS scene to iteratively improve each other. Experiments demonstrate improved rendering quality and geometric consistency across wide-baseline and unbounded 360° benchmarks.
[CV-31] CLARE: Scalable Class-Incremental Continual Learning via a Sparsity-Based Framework BMVC2026
链接: https://arxiv.org/abs/2609.17026
作者: Yunxiang Fu,Meng Lou,Zicheng Liao,Yizhou Yu
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注: BMVC2026
Abstract:Continual learning must balance the learning of new knowledge with the retention of previously learned knowledge to incrementally learn tasks from a data stream without catastrophic forgetting. While leveraging pretrained models has significantly advanced continual learning, existing methods exhibit a scalability bottleneck when trained sequentially on many tasks, suffering from performance degradation due to inter-task interference and loss of plasticity. Inspired by evidence that sparse fine-tuning achieves performance comparable to full fine-tuning, this paper presents a novel sparsity-driven continual learning framework. Our continual learning method, termed CLARE, operates in two stages: it first identifies a sparse, task-critical parameter mask via a sparsity-inducing objective, then performs mask-constrained fine-tuning by only optimizing parameters selected by the mask. This two-stage sparse adapter mechanism enables all tasks to be accumulated within a shared adapter space while reducing destructive interference across tasks. Extensive experiments demonstrate the scalability of CLARE. On the long task-sequence benchmark Omnibenchmark-1k, CLARE outperforms strong baselines in final accuracy by a large margin, e.g, improving EASE by 4.64% and 13.34% after learning 100 tasks, respectively.
[CV-32] sensVLA: Spatially-Grounded Vision-Language-Action Model for Autonomous Wheel Loader ICRA2026
链接: https://arxiv.org/abs/2609.17021
作者: Gopi Krishna Erabati,Bjarne Johannsen,Angus Stewart,Vardeep Singh Sandhu
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注: Accepted at ICRA 2026: From Data to Decisions: VLA Pipelines for Real Robots
Abstract:Autonomous wheel-loader control requires joint reasoning over task semantics, egocentric vision, proprioception, and 3D scene geometry. We present sensVLA, a Vision-Language-Action (VLA) architecture that combines a Qwen3-2B Vision-Language Model (VLM) with a fully trainable transformer action expert trained by flow-matching velocity regression. sensVLA routes Bird’s-Eye-View (BEV) features, extracted from fused front and rear lidar, directly to the action expert through a dedicated cross-attention pathway, while the VLM consumes front and rear RGB views to provide task-conditioned semantic context. This design decouples spatial grounding from linguistic reasoning while preserving interaction between both streams at decision time. The expert predicts six action dimensions: longitudinal velocity, steering, body-frame displacement, arm rate, and bucket rate. On a real-world dataset from a wheel loader, sensVLA reaches aggregate per-step parity with a strong camera-only baseline and reduces longitudinal velocity RMSE by 28% and displacement error by 9% on loading centric scenarios. It also degrades 29% less when the camera stream is corrupted or removed, evidencing that explicit spatial grounding improves accuracy and fault-tolerance for heavy equipment autonomy.
[CV-33] Symmetry-Aware Likelihood-Orbit Aggregation for Selective Left-Right Claim Verification
链接: https://arxiv.org/abs/2609.17004
作者: Zhouzhi Xiong,Chuxi Zhang,Weizhen He,Yi Chen,Qi Li,Donglian Qi
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 5 pages, 2 figures
Abstract:Frozen vision-language models (VLMs) remain unreliable on fine-grained left-right claims, and raw claim likelihoods need not reliably rank verification errors. After a horizontal-reflection intervention is fixed, how should its induced likelihood measurements be combined into a selective verification signal? We introduce Relation-Orbit, a closed-form contrast with no learned fusion parameters that assigns eight normalized likelihoods to query-supporting and counterfactual roles determined by reflection, inverse relation, and entity exchange. A claim is asserted only when the signed contrast exceeds a threshold selected on held-out data using pointwise Clopper-Pearson upper confidence bounds. On VSR and GQA across four frozen VLMs, Relation-Orbit yields higher mean test coverage at a 10% selective-risk calibration target than an all-eight Orbit-Max baseline in all eight dataset-backbone settings; gains over a nearly abstain-all one-sided intervention score are reported separately. A separate LLaVA-1.5/COCO evaluation, reduced-orbit controls, and a two-sided partition diagnostic further characterize the structural advantage.
[CV-34] High-Fidelity Video Quality Assessment with VQA-Specific Saliency WACV2027
链接: https://arxiv.org/abs/2609.16946
作者: Hakan Emre Gedik,Shashank Gupta,Alan Bovik
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to WACV 2027
Abstract:No-reference video quality assessment (NR VQA) has recently seen promising progress with deep learning. However, video data is inherently large, and processing them with deep models incurs high computational cost. This challenge is particularly acute in VQA, where preserving original-resolution cues and dense temporal information is critical for accuracy. Existing efficiency-driven preprocessing strategies, such as fragmenting, reduce computation but alter the input data distribution, limiting effective reuse of pretrained video foundation models (ViFMs). To address these challenges, we propose \textbfHigh-\textbfFidelity \textbfVideo \textbfQuality \textbfAssessment (\textbfHFVQA), a framework built on fixed-size spatio-temporal (ST) patches that is fully compatible with pretrained ViFMs. HFVQA samples ST patches across multiple scales, including the original resolution, with minimal temporal subsampling to preserve low-level quality cues and semantic context. To limit computation, HFVQA introduces a lightweight auxiliary network trained end-to-end with the ViFM encoder to learn \textitVQA-specific saliency. Distilled directly from quality supervision, this saliency captures task-specific importance patterns, reflecting that video quality perception is dominated by a small subset of spatio-temporal regions. By combining high-fidelity spatio-temporal cues with learned, task-specific saliency, HFVQA achieves SOTA performance on standard NR VQA benchmarks while processing as little as 12% of candidate ST patches, making high-fidelity ViFM-based VQA computationally tractable.
[CV-35] MedPCFM-TED: One-Step Point Cloud Flow Matching for Implant Generation via Teacher-Guided Endpoint Distillation
链接: https://arxiv.org/abs/2609.16934
作者: Kamil Kwarciak,Marek Wodzinski
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 10 pages, 3 figures
Abstract:Cranial implant generation is an important task in medical imaging. Recent point cloud based generative methods, particularly flow matching, offer strong reconstruction quality and efficient sampling, but still require multiple neural function evaluations during inference. This limits rapid generation of multiple plausible implant candidates. We propose Teacher-guided Endpoint Distillation (TED), a simple one-step distillation framework for conditional cranial implant generation on point clouds. TED trains a one-step student using teacher-guided endpoint supervision and geometric matching losses, while avoiding explicit path straightening. We evaluate TED on the SkullFix and SkullBreak benchmarks. TED achieves the best overall performance on the SkullBreak dataset, remains competitive on SkullFix, and provides the strongest Chamfer distance performance among the compared one-step methods. In addition, TED generates implants in approximately 0.04s per sample. These results show that one-step distillation can substantially accelerate conditional point cloud implant generation without sacrificing reconstruction quality.
[CV-36] Evaluating Mesh Reconstruction Methods for Crop Phenotyping
链接: https://arxiv.org/abs/2609.16926
作者: Karanvir Singh,Theo Morales,Binh-Son Hua,Mukesh Saini
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 12 pages, 17 Figures
Abstract:Phenotyping an agricultural crop is crucial for studying its entire life cycle, as it provides vital insights to improve yield and, ultimately, food production. Doing the same for crops grown on remote sites is a challenge for the specialists who cannot be available on-site. 3D reconstruction techniques offer a promising solution to this problem by enabling crop digitization, allowing specialists to access the resulting 3D crop models from anywhere at any time. In this work, we evaluate recent 3D reconstruction pipelines for crop phenotyping. We focus on 7 mesh reconstruction pipelines and measure the fidelity and consistency of their outputs qualitatively and quantitatively. Our results suggest that the meshes produced by the GGGS, PGSR, and 2DGS are preferable to the other pipelines, owing to their quantitative metrics and visually pleasing outputs. The GGGS pipeline is better than the second-best pipeline (2DGS) by about 27% on the radar chart with 5 dimensions, namely, User ratings, Chamfer distance, LPIPS, PSNR, and SSIM.
[CV-37] NeuroSymbEAD: A Large Scale Neuro-Symbolic Caption Dataset for Omni-Directional Embodied Autonomous Driving
链接: https://arxiv.org/abs/2609.16919
作者: Muhammad Ahmed Ullah Khan,Mohammed Elamine,Sheikh Talha Uddin,Didier Stricker,Sk Aziz Ali,Muhammad Zeshan Afzal
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:This paper introduces NeuroSymbEAD, a large-scale neuro-symbolic caption dataset featuring an ego-centric knowledge graph (KG) of static and dynamic objects annotated with classes, categories, heading directions, orientations, and distances from the ego-vehicle. These annotations are used on the KITTI-360 dataset to generate multilevel textual captions representing a lightweight version of an ego-centric scene map. Outdoor scene-map reconstruction, visual recognition, and object grounding establish baselines for driving common sense and traffic/scene understanding. For these purposes, natural language-based grounded captioning of objects and their complex relationships is a widely adopted contextual representation for indoor scene tasks. Neuro-symbolic representations have proven effective in handling structured information for various computer vision and language applications. Our data annotation pipeline allows the generation of varied map segments, populating simulated or real objects within the bounding boxes predicted by any 3D object detection network, and building hierarchical text captions. We benchmark our neuro-symbolic and ontological caption generation using pre-trained grounding and learned auto-regressive captioning networks. By converting 3D driving scenes into structured ego-centric language, NeuroSymbEAD provides a benchmark for vision-language and foundation models for traffic-scene explanation, 3D reasoning, and interpretable autonomous-driving perception.
[CV-38] PiPS: Post-Hoc Prototypical Explanations for Interpretable Semantic Segmentation
链接: https://arxiv.org/abs/2609.16909
作者: Miłosz Adamczyk,Tymoteusz Zapala,Piotr Borycki,Przemysław Spurek
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:With the increasing deployment of deep neural networks in critical systems, such as medical diagnostics and autonomous vehicles, ensuring their interpretability is crucial to building trust in decision-making systems. In the field of explainable artificial intelligence, prototype-based reasoning has gained particular popularity, as it mimics human cognitive processes by explaining model decisions based on visual similarity under the looks like this paradigm. While this paradigm has been thoroughly investigated in the context of global image classification, the interpretability of dense predictions, particularly semantic segmentation, remains largely unexplored despite its immense importance in tasks requiring precise object localization. Existing prototype-based interpretable segmentation models rely on ante-hoc architectures, which entails significant limitations because they require costly training from scratch and modifications to the network structure, ultimately leading to a noticeable drop in predictive performance compared to standard black-box models. To address this issue, we propose PiPS (Post-hoc interpretable Prototypical Segmentation), the first fully post-hoc solution for generating prototypical explanations for semantic segmentation models. Our method enables the extraction of intuitive, spatially localized explanations from any pre-trained network without modification or fine-tuning, thereby preserving 100% of the model’s original predictive performance. This approach opens a new avenue for the safe and cost-effective deployment of transparent systems in advanced computer vision tasks. Codebase available at this https URL.
[CV-39] mporally Consistent Graph Extraction and Matching for Longitudinal Angiographic Images MICCAI2026
链接: https://arxiv.org/abs/2609.16889
作者: Linus Kreitner,Laurin Lux,Carmen Baumann,Daniel Rueckert,Martin J. Menten
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at MICCAI 2026 GRAIL workshop
Abstract:Recent advances in angiographic imaging have enabled longitudinal visualization of the microvasculature. Image processing pipelines based on vessel graphs are able to resolve subtle temporal changes at the level of individual blood vessels. However, current strategies for graph extraction, refinement, and matching are highly sensitive, with even minuscule differences in the underlying segmentation map resulting in substantially different vessel graphs. These artifacts severely inhibit the ability to accurately match sequential vessel graphs of the same subject over time. To address this problem, we propose a strategy that matches graphs before jointly refining them. Specifically, we perform an early matching after basic graph extraction before removing spurious bulges and merging junctions in both graphs using joint information. In experiments with complex retinal vessel graphs, we demonstrate that this strategy results in a higher matched area without graph fragmentation compared to separate or no refinement, respectively.
[CV-40] VOR-Bench: A Human Perception-Driven Benchmark for Video Object Removal BMVC-2026
链接: https://arxiv.org/abs/2609.16878
作者: Haonan Huang,Tianrui Qiu,Xianghao Zang,Yinan Du,Zhixiang He,Chi Zhang,Hao Sun,Zhongjiang He,Tianwei Cao,Xuchong Zhang,Hongbin Sun,Kongming Liang,Zhanyu Ma
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: BMVC-2026
Abstract:Despite its crucial role in video object removal (VOR), existing evaluation paradigms face two critical limitations: questionable references and a misalignment between tradi- tional metrics and human preference. To address these challenges, we introduce VOR- Bench, which advances VOR evaluation through three integrated components. First, we present the VOR Dataset (VORD), the first benchmark dataset providing both paired edited videos and graffiti masks. Its unique strength lies in a diverse data spectrum, which encompasses model-generated, tool-rendered, and camera-captured data, ensuring robust assessment across real-world scenarios. Second, we develop rMPAF, a realistic Motion- capable Paired-video Acquisition Framework. By combining the strengths of image- based object removal and fine-tuned video generation models, rMPAF automatically generates realistic, motion-coherent paired videos. Finally, we propose three evaluation dimensions and introduce VOR-MDSM, the first perception-driven VLM-based scoring model specifically designed for mask-guided VOR. It bridges the gap between arithmetic metrics and human perception by covering the essential visual attributes and matching nuanced human judgment. Extensive experiments demonstrate that VOR-Bench yields evaluation results that align closely with human perception, achieving a remarkable cor- relation (\rho 0.9) with subjective assessments. We will release VOR-Bench along with its documentation to ensure full reproducibility.
[CV-41] Multi-modal Knowledge Preserving Adapter for Embedding Backward Compatibility ECCV2026
链接: https://arxiv.org/abs/2609.16875
作者: Jaeseok Byun,Gukyeong Kwon,Han-Kai Hsu,Meher Gitika Karumuri,Zhikang Zhang,Hao Yang,Davide Modolo
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 15 pages, ECCV 2026 camera ready
Abstract:Upgrading embedding models typically requires expensive database re-indexing, as new query embeddings are incompatible with existing database embeddings. While Backward Compatible Training (BCT) mitigates this by enforcing compatibility during training, existing approaches often require updating the backbone model. This is impractical because of significant training cost, the risk of performance regression, and limited access to proprietary model weights. We introduce Multi-modal Knowledge Preserving Adapter (MKP-Adapter), the first adapter-only BCT approach for Multi-modal Large Language Models (MLLMs) that requires no backbone updates. We identified that the primary challenge in adapter-only BCT is preserving the knowledge of the new embeddings while enforcing backward compatibility. Hence, we propose a multi-level preservation loss that maintains the geometric structure of the embedding spaces throughout BCT. Furthermore, a focal re-weighting strategy is integrated to prioritize learning from challenging samples. Experiments demonstrate that our method achieves strong backward compatibility across diverse multi-modal benchmarks (image, text, visual document, and video retrieval tasks) and model types. Notably, MKP-Adapter is trained solely on pre-extracted embeddings and requires only negligible additional latency relative to the original backbone forward pass, highlighting its efficiency.
[CV-42] Accelerated Decoding of Centroid Positional Encoding for Instance Segmentation
链接: https://arxiv.org/abs/2609.16874
作者: Carmelo Scribano,Filippo Muzzini,Nedyalko Prisadnikov,Mohammad Mahdi,Yuqian Fu,Giorgia Franchini,Danda Pani Paudel,Marko Bertogna,Luc Van Gool
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Presented at 2026 Joint International Conference on AI, Big Data and Blockchain. Granada, Spain
Abstract:Beyond model inference, the decoding stage, which converts raw network outputs into task-level representations, constitutes a significant portion of the execution cost. Despite its practical impact, prediction decoding has received comparatively little attention and is often implemented using generic CPU routines or inefficient GPU kernels, limiting the benefits of advances in model efficiency. In this work, we investigate the decoding overhead associated with a recent sinusoidal centroid encoding for Instance Segmentation, in which each pixel regresses a positional embedding of its instance centroid. This approach allows flexible segmentation without predefined proposals, but extracting instance masks from dense embeddings incurs a high computational cost. We present an optimized CUDA-based implementation of the decoding algorithm tailored to this encoding, explicitly addressing challenges related to parallelization, synchronization, and memory access on modern GPUs. Our solution significantly reduces decoding overhead and improves End-to-End inference latency, outperforming both CPU-based approaches and naive GPU implementations. The results demonstrate that efficient decoding is essential to fully exploit the advantages of advanced output representations and highlight the importance of jointly designing encoding schemes and their decoding algorithms for real-time computer vision systems.
[CV-43] NeuroTS-Net: Multi-Class Semantic Segmentation of Pediatric Brain Tumors in Multi-Modal MRI MICCAI
链接: https://arxiv.org/abs/2609.16873
作者: Darius Peteleaza,Razvan-Gabriel Dumitru,Bogdan Neamtu,Arpad Gellert,Mariana Sandu,Claudiu Matei
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Accepted at the 2026 International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) - BraTS Cluster of Challenges: Pediatric Brain Tumor Segmentation (BraTS-PEDs)
Abstract:Pediatric brain tumors are a leading cause of cancer-related mortality in children, and their small, rare, and often low-contrast subregions make accurate manual delineation challenging. Reliable automated segmentation is therefore needed to support diagnosis, treatment planning, and response assessment. Accordingly, we introduce NeuroTS-Net, a three-dimensional encoder-decoder convolutional neural network architecture for multi-class semantic segmentation that incorporates a dual-scale raw-detail stream, adaptive low-resolution context selection, and detail-preserving multipath downsampling. These components preserve fine intensity and boundary information while efficiently modeling broader tumor context. NeuroTS-Net was trained on the BraTS 2026 pediatric dataset without external data or pretrained weights and evaluated against nnU-Net and MedNeXt under the same experimental protocol. NeuroTS-Net outperformed the baseline methods, achieving whole-tumor and tumor-core Dice scores of 0.938 and 0.937 on the internal validation set and 0.927 and 0.926 on the official challenge validation set. The code is open-sourced at: this https URL.
[CV-44] GRACE: Geometry- and Ray-Aware Camera-Efficient Multi-View Pedestrian Tracking
链接: https://arxiv.org/abs/2609.16872
作者: Taigo Sakai,Kazuhiro Hotta,Hiroki Kouno,Naoki Kato
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Reducing the number of cameras reduces the deployment cost but removes views that correct BEV responses stretched away from true pedestrian positions by projection and short score drops that can split tracks in Bird’s-Eye View (BEV) tracking. We introduce GRACE, a camera-efficient multi-view tracker with three components. Volumetric-Guided Fusion combines homography-based BEV features with features lifted through 3D space. Ray Conditioning exposes each camera’s viewing direction to the fusion network. Its tracking component, BEV Track Recovery (BTR), uses low-confidence detections only to continue existing tracks. The same detections cannot start new tracks. With two WildTrack cameras, GRACE improves MOTA from 83.54 for TrackTacular, our baseline, to 91.07.
[CV-45] SPEAR NeXT Causal Latent Forecasting Across Multiple Horizons for Spectral Temporal Earth Representation Learning
链接: https://arxiv.org/abs/2609.16871
作者: Rajiv Ranjan,Udaiveer Singh,Shashank Tamaskar,Dharmendra Saraswat
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 24 Pages
Abstract:Earth observation is inherently dynamic, yet temporal information in many foundation models is learned through reconstruction, invariance, or retrospective sequence summarization. SPEAR NeXT is introduced as a compact pixel-wise multimodal spectral temporal foundation model in which temporal self supervision is formulated as past only, multi horizon latent Earth state prediction. Instantaneous states are first encoded by the pretrained SPEAR model from optical, radar, and environmental observations into compact 32 dimensional embeddings. Their temporal evolution is then modeled by a causally masked Trans former that predicts multiple future latent states from pre ceding observations. Relative temporal order is represented using Rotary Position Embeddings, while month and year embeddings encode seasonal phase and interannual con text.
[CV-46] cnerv:dual-domain temporal context modeling for implicit neural video compression
链接: https://arxiv.org/abs/2609.16870
作者: Xuezhi Xiang,Yixin Zhao,Heqi Xiang,Jiayao Liu,Shanjun Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Video compression aims to minimize reconstruction distor tion under a constrained bit rate. Existing video implicit neural representations (INRs) often decode frames independently, leaving intermediate features unconditioned on previous reconstructions and content embeddings without explicit temporal prediction. We propose TCNeRV, which exploits reconstructed context in both feature and embedding domains. Its multi-scale temporal-context fusion (MTCF) module injects gated historical features at multiple decoder scales, while temporal embedding-residual coding (TERC) predicts each content embedding and codes only its residual. With approximately 3M parameters, TCNeRV achieves an average PSNR of 36.08 dB on the UVG dataset, outperforming HNeRV-Boost by 2.20 dB. It reduces BD-rate by 22.06%, 66.73%, and 29.85% relative to HM, DCVC, and HiNeRV, respectively, demonstrating competitive rate-distortion performance with limited model capacity.
[CV-47] EMPO: Learning Temporal Context for Dynamic Robot Manipulation
链接: https://arxiv.org/abs/2609.16864
作者: Zhenyang Feng,Jimin Heo,Erik B. Sudderth,Unnat Jain
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Accepted at CoRL 2026. Project page: this https URL
Abstract:Vision-language-action (VLA) models have achieved impressive performance in quasi-static manipulation, but struggle in dynamic manipulation tasks because they operate on a single observation at inference time. We identify two representational failures that underlie this limitation. The first is motion ambiguity, where a single observation does not include scene dynamics and therefore cannot anticipate the future state of moving objects. The second is state aliasing, where visually similar observations from different points in a task require different actions. We argue that these failures persist regardless of model scale and inference latency, showing that the bottleneck is missing temporal context rather than model capacity. Based on this insight, we propose TEMPO, which augments a pretrained VLA with two temporal inputs: a motion summary extracted from a frozen video foundation model to resolve motion ambiguity and a compact proprioceptive history to resolve state aliasing. TEMPO requires no modification to the backbone and adds minimal compute overhead at training or deployment. Across four dynamic manipulation tasks, it improves Bottle Handover success from 44% to 74% and is the only method that solves state aliasing. Probing and ablation studies confirm that each temporal signal independently addresses its corresponding failure. We further release TEMPO-Bench, a benchmark of over 50k annotated frames for evaluating motion-aware robot perception in both regression and multiple-choice formats. Project Website: this https URL Comments: Accepted at CoRL 2026. Project page: this https URL Subjects: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG) Cite as: arXiv:2609.16864 [cs.RO] (or arXiv:2609.16864v1 [cs.RO] for this version) https://doi.org/10.48550/arXiv.2609.16864 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-48] Measuring Annotation Efficiency for Handwritten Devanagari Recognition: Sample-Complexity Curves for Four Pretraining Regimes
链接: https://arxiv.org/abs/2609.16859
作者: Manglesh Kumar Pandey,Sumit Kumar Banshal
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:To train handwritten text recognition systems we need word images and their corresponding transcriptions, and these transcriptions are produced manually. For a script that can be read by only a small number of specialists, this manual transcription is a limitation, because the trained models are supposed to save the time of those same specialists. A relevant question therefore arises: how many transcriptions are needed before a recogniser becomes useful, and how much of that cost can pretraining remove? In this study the answer is measured directly for handwritten Devanagari. We keep the recogniser, optimiser and evaluation protocol the same and change only the number of real transcribed words used for fine-tuning across nine budgets from 10 to 4,000 and four initialisation regimes, with six seeds at every point. The resulting curves are then converted into annotation-equivalent terms. A CER of 0.50 is reached by supervised synthetic pretraining using only 81 transcribed words, whereas random initialisation requires 355, which gives a label multiplier of 4.40 [3.56, 4.99]. There is a zero-shot reference point as well: with no real transcribed words at all, this pretraining is worth about 136 of them. This advantage gets smaller as the target accuracy improves, and at the most demanding target we measure, it cannot be distinguished from no saving at all. A fourth arm in which only the encoder is transferred separates the effect of the pretraining method from that of transfer scope, and masked image modelling is observed to transfer negatively over a bounded range of budgets. We emphasise that the scarcity in this study is constructed by subsampling a large corpus.
[CV-49] coPrompt: Temporal-Conservative Prompt Learning for Vision-Language Models ECCV2026
链接: https://arxiv.org/abs/2609.16858
作者: Zeyi Shao,Haowen Hua,Jiaxin Zhang,John See,Zeyd Boukhers,Cong Yang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at ECCV 2026 (Main Conference). Code: this https URL
Abstract:Prompt learning adapts vision-language models, such as CLIP, by adjusting a small set of context tokens. However, under few-shot supervision, even moderate label noise can disrupt prompt optimization. To address this issue, we propose TecoPrompt, a closed-loop robust prompt-learning framework that revisits optimal transport (OT) pseudo-labeling from a temporal perspective. TecoPrompt employs an entropic OT plan in the CLIP semantic space to obtain globally consistent label candidates. It verifies the reliability of these candidates by examining trajectory stability: a noisy label is only rewritten if the OT candidate remains unchanged within a K-epoch temporal stability window and passes a confidence gate based on Exponential Moving Average (EMA). This approach helps reduce confirmation bias. The rewritten labels are then integrated back into prompt training using a tri-group objective that includes three loss functions aligned with clean, mid, and noisy subsets. Experiments on seven datasets with synthetic symmetric and asymmetric noise, as well as Food101N, demonstrate significant performance improvements. For example, on the OxfordPets dataset, with 50% asymmetric noise, TecoPrompt achieves an accuracy of 0.843, up from 0.775.
[CV-50] FAHCD-Net: Frequency-Adaptive Heatmap-Conditional Diffusion Networks for Robust Facial Landmark Detection
链接: https://arxiv.org/abs/2609.16842
作者: Jun Wan,Jiwei Hu,Shengkai Hu,Qilu Zhu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Facial Landmark Detection(FLD) is a crucial task in various applications and has achieved significant advancements in recent years. However, current FLD methods still struggle under challenging conditions, where facial structural variations, information loss, and noise interference severely compromise the integrity and accuracy of learned facial features. To address these issues, we propose Frequency-Adaptive Heatmap-Conditional Diffusion Network (FAHCD-Net), which integrates a Frequency-Adaptive Heatmap-Conditional Diffusion (FAHCD) model with a Smoothness Regularization (SR) loss in a cascaded framework. Specifically, the FAHCD model incorporates a Hierarchical Frequency Adaptation (HFA) module designed to suppress redundant high-frequency noise through multi-layer frequency decomposition and adaptive reconstruction, thereby preserving essential facial structures. Additionally, the SR loss is proposed to further mitigate the interference of high-frequency noise and enhance the smoothness of the generated landmark heatmaps. By cascading the FAHCD model with the SR loss, FAHCD-Net effectively leverages both statistical and frequency-based distribution characteristics of the data to progressively generate more accurate landmark heatmaps from noisy inputs. Extensive experiments on popular benchmarks demonstrate the effectiveness and robustness of the proposed method, achieving state-of-the-art performance in FLD tasks under challenging scenarios. The source code is available at this https URL.
[CV-51] StackTok: Accelerating VLMs Inference with Budget-Adaptive Visual Token Selection
链接: https://arxiv.org/abs/2609.16841
作者: Zhenbin Wang,Lei Zhang,Lituan Wang,Wei Huang,Yan Wang,Zhenwei Zhang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Increasing image resolution produces ever-longer visual-token sequences in vision-language models (VLMs), substantially raising their inference cost. To reduce this overhead without retraining, existing methods select compact token subsets that prioritize query relevance, visual coverage, or a fixed trade-off between them. The appropriate balance, however, varies across queries and token budgets: localized questions favor relevance, whereas holistic questions demand broader visual coverage. We introduce StackTok, a training-free selector that treats query relevance as the objective and visual coverage as budget-calibrated support. StackTok builds a size-indexed coverage reference from a coverage-only greedy sequence and adjusts its support target using query–vision affinity entropy. A reference-gated interleaved selection policy then switches between relevance- and coverage-oriented additions according to the current subset’s support deficit. For high-resolution inputs, StackTok allocates one shared token budget across crops according to the combined marginal gain of locally nominated tokens. Evaluated with five VLMs over ten distinct image-understanding benchmarks, StackTok ranks first among training-free selectors in every tested model–budget setting. On high-resolution LLaVA-NeXT-7B, it retains 95.26% of full-token performance with only 160 of 2,880 (5.6%) visual tokens.
[CV-52] What Breaks Local Watermarks? A Robustness Benchmark for Local Invisible Image Watermarking CCS2026
链接: https://arxiv.org/abs/2609.16832
作者: Kai Yao,Bence Szilágyi,Sebestyén Kamp,Máté Poór,Máté Szilveszter,Matyas K. Zsoldos,Marc Juarez
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注: This work has been accepted for publication in the proceedings of the 19th ACM Workshop on Artificial Intelligence and Security (AISec 2026), co-located with ACM CCS 2026. The final version will be published in the ACM Digital Library
Abstract:Local image watermarking embeds an invisible signal into selected image regions rather than spreading it across the entire image, enabling payload recovery from specific objects or regions without perceptibly altering the image. Existing studies evaluate the robustness of payload recovery and localization under image transformations, but they often focus on their own proposed method, resulting in narrow evaluations with inconsistent choices of transformations, datasets, and metrics. These inconsistencies across studies limit direct comparisons across methods and muddle the overall picture of local watermark robustness. To address this gap, we present the first systematic robustness benchmark for local watermarks across 55 image transformations, including (i) signal distortions, (ii) changes in image coordinate alignment, (iii) indirect local edits, and (iv) direct watermark edits. The benchmark evaluates MaskWM, WAM, OmniGuard, TrustMark, and PixelSeal, all methods that either provide native localization or require minimal adaptation to support it. Our results show that all evaluated methods are vulnerable to some transformation, with MaskWM standing out as offering the strongest payload recovery and localization, although it has the lowest image quality in the clean setting. Synchronization further improves MaskWM’s payload recovery under several geometric transformations, albeit at an additional cost to image quality. A key finding is that local watermark robustness depends strongly on the nature of the transformation: signal distortions are often tolerated by the strongest methods, while geometric misalignment and generative local edits, such as inpainting and outpainting, can completely impair payload recovery. We observe that payload recovery and localization are related but not interchangeable, and both strongly depend on the transformation’s impact on the watermark region.
[CV-53] Hyper-RED: Scalable Event Pre-training via Semantic Hypergraph Distillation
链接: https://arxiv.org/abs/2609.16811
作者: Meisen Wang,Zhiqiang Tian,Wei Bao,Chengjie Wang,Shaoyi Du,Siqi Li
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Event cameras have shown great potential for robust visual perception, yet scaling event representation learning remains challenging due to the scarcity of large-scale annotated event data. Pretrained image models provide scalable semantic supervision, but existing image-to-event methods rely on rigid pixel-wise or token-wise alignment that overlooks modality discrepancies in texture, density, and appearance, potentially causing semantic collapse and limiting transferability. To address this issue, we propose Hyper-RED, a simple, painless, and scalable image-to-event pretraining framework that transfers high-order semantic structures from images to events. Hyper-RED uses hypergraphs to model and align high-order semantic associations among multiple image and event tokens, enabling cross-modal knowledge transfer while accommodating modality-specific differences rather than enforcing rigid one-to-one correspondence. Specifically, given a paired event–image sample, Hyper-RED leverages DINOv3 to extract spatial token representations and constructs image, event, and cross-modal semantic hypergraphs, where each hyperedge connects multiple semantically correlated tokens. We further introduce a hypergraph relational distillation loss that imposes complementary intra- and cross-modal constraints, enabling the event encoder to inherit image-derived semantic organization while preserving local relational consistency and event-specific characteristics. Experiments on three tasks across five event datasets demonstrate consistent scaling from ViT-S to ViT-L and state-of-the-art performance (Fig.1). The code is available at: this https URL.
[CV-54] EDi: Temporal Memory-Enhanced and Denoising Transformer for Surgical Instrument Segmentation
链接: https://arxiv.org/abs/2609.16797
作者: Jiahong Yuan,Weiming Mi,Tao Zhang,Haoyin Zhou
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Query-based segmentation methods have shown promising potential for surgical instrument segmentation and recognition, which is essential for scene understanding and downstream tasks in computer assisted surgery. However, most existing approaches predominantly rely on per-frame predictions and overlook cross-frame temporal priors as well as temporal-consistency constraints. This limitation often leads to unstable query representations and suboptimal category recognition. In this paper, we propose TEDi, a Temporal memory-Enhanced and Denoising transformer for surgical instrument segmentation that addresses these is sues through Memory Search Enhancement and Temporal Consistency Denoising. The former introduces a query-level memory bank and a memory search enhancement encoder to retrieve discriminative representations from historical frames, enriching current-frame features. The latter constructs a temporally consistent reference as a cross-frame semantic anchor to suppress temporally unstable predictions and promote semantic coherence across frames. Extensive experiments on two benchmark datasets, EndoVis 2017 and EndoVis 2018, demonstrate that TEDi consistently outperforms state-of-the-art methods, highlighting its potential to further advance computer-assisted surgery. Our code is available at this http URL.
[CV-55] Noise2Noise Revisited: Training Pair Distributions Dominate Loss Choice in Self-Supervised Denoising
链接: https://arxiv.org/abs/2609.16788
作者: Dingyan Shang,Zhenyu Xu,Youting Wang,Bonan Shen,Bowen Liu
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注: 8 pages, 3 figures, 3 tables. Accepted to The 8th International Conference on Video, Signal and Image Processing (VSIP 2026). Code and data: this https URL
Abstract:Noise2Noise (N2N) trains denoisers on pairs of independently corrupted observations, eliminating clean references. We stress-test two natural conjectures about why the L1 loss outperforms L2 here. First, the hypothesis that the L1 loss confers robustness via parameter sparsity confuses the loss with Lasso regularization: an explicit Lasso penalty produces the predicted sparsity yet fails to reproduce L1’s cross-noise behavior, while L1- and L2-trained weight distributions are indistinguishable. Second, the population optima of the two losses coincide exactly for symmetric signal posteriors and nearly so for concentrated ones. Measured differences are therefore dominated by optimization dynamics (bounded-influence gradients), which we probe with gradient statistics and contaminated-target training. On Kodak24 with five synthetic noise families, the L1 loss holds a statistically significant edge over L2, below 1 dB PSNR, holding across three seeds on 13 of the 14 noise columns. On real camera noise the loss is not the decisive variable in distribution: on official SIDD validation blocks, synthetic-Gaussian-trained N2N models gain only 0.8 to 3.7 dB over the noisy input regardless of loss, while retraining on SIDD’s own noisy pairs, never reading ground truth, gains 9.4 to 11.0 dB, far ahead of BM3D. All metrics are on raw network outputs, and the study makes no leaderboard claim. The training pair distribution, not the loss, carries the inductive bias. That design rule applies wherever clean references are unobtainable, from microscopy to industrial inspection sensors.
[CV-56] PSMP-CLIP: Patch-Prompt SAM and Multi-Semantic Prompting for CLIP-Based Zero-Shot Anomaly Detection
链接: https://arxiv.org/abs/2609.16785
作者: Xuezhi Xiang,Guanghao Wu,Heqi Xiang,Jiayao Liu,Xiaoheng Li,Yiming Chen,Shanjun Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Zero-shot anomaly detection aims to localize anomalies without target-domain samples. Existing CLIP-based methods suffer from coarse anomaly maps and limited semantic prompts. We propose PSMP-CLIP, integrating patch-prompt SAM2 segmentation (PPSS) and multi-semantic guided prompt regularization (MSGPR). PPSS samples prompts directly from intermediate patch features, avoiding threshold drift and guiding SAM2 to produce precise masks. MSGPR uses multiple learnable prompts constrained by semantic anchors to preserve generalization. Experiments on 14 datasets show highly competitive performance, achieving the best pixel-level AUROC on MVTec AD, BTAD, DTD-Synthetic, CVC-ClinicDB, TN3K, Endo, and Kvasir.
[CV-57] Unifying Semantic Priors and High-Frequency Traces: Enhancing V-JEPA with Mixture-of-Experts for Robust Synthetic Image Forensics
链接: https://arxiv.org/abs/2609.16778
作者: Simone Teglia,Irene Amerini
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 10 pages, 2 figures. Code available at this https URL
Abstract:The unchecked proliferation of manipulated images on social media platforms has increased the spread of misinformation, posing a severe threat to public trust and information integrity. Modern deepfake detectors typically rely on Vision Transformers (ViTs) to capture the low-level inconsistencies that characterize fully synthetic or locally tampered images. However, the global understanding of such foundation models is not enough to discriminate alone between real and fake multimedia content, especially in challenging scenarios where images are compressed or transmitted through social media. In this paper we pioneer the application of Joint-Embedding Predictive Architecture (JEPA) models to deepfake detection, taking advantage of the generalized representation of visual reality that such World Models have exhibited. We hypothesize, and empirically demonstrate, that the intrinsic world understanding of JEPA models can be used as a strong prior for a deepfake detector. To fully exploit JEPA capabilities, we propose MoE-JEPA, a dual-stream architecture for deepfake detection. By enhancing a V-JEPA 2 backbone with a Residual Mixture-of-Experts (MoE) mechanism, along with a noise stream branch, our model dynamically internalizes forensic knowledge. Furthermore, a Gated Attention Multiple Instance Learning (MIL) module is employed to ensure precise spatial semantic understanding. Evaluated on the SID-Set benchmark, comprising 300K AI-generated, tampered and authentic images, MoE-JEPA establishes a new state-of-the-art with an accuracy of 95.54%, successfully outperforming vastly larger models.
[CV-58] IMVS: Interactive Medical Volume Segmentation with Test-Time Adaptation - A New Method for Annotating Radiology Datasets
链接: https://arxiv.org/abs/2609.16775
作者: Abhilaksh Singh Reen,Kushal Borkar,Ritvik Mahapatra
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Annotating large radiology datasets is bottlenecked by the manual effort of delineating structures slice-by-slice in 3D volumes. Interactive methods reduce this effort but stay interaction-inefficient: slice-wise methods (including many foundation models) ignore inter-slice continuity, while 3D and video-based methods propagate a prompt with a \emphfixed propagator that never adapts to the target volume, so it drifts on low-contrast or pathological structures and must be re-prompted. We present IMVS, a human-in-the-loop annotation framework that composes three components into a closed loop rather than a new segmentation primitive: a lightweight 2D Slice Mask Adapter (SMA) fine-tuned online from user scribbles, a frozen Volume Mask Tracker (VMT) that propagates corrected masks across adjacent slices, and a soft teacher–student alignment that limits forgetting. The SMA is backbone-agnostic (UNet++, DeepLabV3, TransUNet). Across 8 public CT/MRI datasets, IMVS matches strong interactive baselines in quality while sharply cutting annotation effort: 14.4\times faster than a proficient copy-based manual workflow ( 22.3\times over naive manual), 4.6\times over slice-wise and 1.9\times over 3D interactive methods. MedSAM2 and ScribblePrompt stay competitive or stronger on well-delineated organs; IMVS’s advantage is largest on challenging targets and on interaction efficiency. Source code and Demo Video: this https URL.
[CV-59] FSANet: Frequency-Spatial Aware Network for Image Segmentation
链接: https://arxiv.org/abs/2609.16773
作者: Ruibo Wang,Ziyi Shen,Huaming Wu,Dong Liang,Kun Shang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 13 pages
Abstract:Image segmentation remains challenging due to occlusions, poor lighting, and irregular structures. Although transformer-based methods achieve high accuracy, they rely heavily on long-range spatial features, leading to high computational costs and neglecting prior knowledge or noise patterns, resulting in missing details and unclear boundaries. To address these issues, we propose Frequency Spatial Aware Network (FSANet), which integrates prior knowledge with a dual-domain solver to sequentially adapt to diverse segmentation tasks. Specifically, we design three key modules: (1) Structure Prior Module, which recovers overlooked details; (2) Dual-Domain Awareness Module, which captures salient features while disentangling noise; and (3) Edge Estimation Module, which enhances edge awareness for more precise segmentation. In addition, the limited availability of comprehensive segmentation datasets covering various real-world scenarios hinders the performance of existing methods. To address this, we introduce SceneX, a novel open-source dataset featuring 10 challenging non-ideal scenarios, establishing a new benchmark for evaluating and improving the robustness and real-world applicability of the segmentation models. Extensive experiments demonstrate the efficiency and effectiveness of FSANet.
[CV-60] HLC-GS: Risk-Map-Guided Height-Layer Consistency Gaussian Splatting for DSM Reconstruction from Optical Satellite Imagery
链接: https://arxiv.org/abs/2609.16772
作者: Jie Yang,Yingdong Pi,Qiyan Luo,Xiaoyu Wang,Lekang Wen,Mi Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:A Digital Surface Model (DSM) is a fundamental geospatial data product for representing the elevation of the Earth’s surface. Recently, 3D Gaussian Splatting (3DGS) has shown considerable potential for DSM reconstruction from multi-view optical satellite imagery due to its explicit scene representation and efficient optimization. However, in 3DGS-based DSM generation, alpha-weighted aggregation of Gaussian altitudes may blend splats from different height layers at the same rendered pixel or DSM sampling location, producing non-physical intermediate elevations and height-layer mixing errors. To address this problem, we propose HLC-GS, a risk-map-guided height-layer consistency Gaussian Splatting method for DSM reconstruction from optical satellite imagery. HLC-GS consists of a risk map module, a dominant-layer reliability correction module, and a secondary-layer suppression module. The risk map localizes high-risk pixels with abnormal height dispersion and unreliable dominant-layer responses, while the latter two modules regularize unreliable dominant-layer responses and suppress weakly supported far secondary-layer responses. Extensive experiments are conducted on the DFC2019 and IARPA2016 datasets. Compared with six state-of-the-art DSM reconstruction methods, HLC-GS achieves better overall accuracy. Compared with the latest and precision-enhanced EOGS, HLC-GS reduces the average MAE from 1.46 m to 1.18 m and the average RMSE from 2.78 m to 2.58 m over the evaluated scenes, while improving PAG _2.5 from 86.09% to 88.61%. Overall, these results demonstrate that explicitly modeling per-pixel height-layer consistency alleviates height-layer mixing and improves the geometric quality of 3DGS-based DSM reconstruction from optical satellite imagery.
[CV-61] De-GAN - Dynamic Parameter Tuned GAN for 3D Medical Image Segmentation: A Step Towards Generalisation
链接: https://arxiv.org/abs/2609.16755
作者: Zoha Usama,Azadeh Alavi
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 5 pages, 2 figures
Abstract:Brain tumor segmentation remains difficult because enhancing tumor (ET) has low contrast and overlaps surrounding tissue, while scanner and site variation causes domain shift. We propose DE-GAN, a contrast-enhancing conditional GAN that combines input-adaptive dynamic convolutions, style-aware feature mixing, and coordinate encoding to synthesize slice-adaptive FLAIR images. A label-guided, class-conditional target separates tumor-core (TC) and ET intensities while preserving anatomy. The generated FLAIR is concatenated with the original MR modalities and used to train a 3D U-Net. Across BraTS 2015, 2018, and 2019, DE-GAN improves segmentation over the baseline and static EnhGAN replacement on most reported TC/ET metrics, with the largest gains from retaining both original and enhanced FLAIR. Code and pretrained models are available at this https URL.
[CV-62] Seeing What Matters: Visual Cue Guided Video Planning for Generalizable Robot Navigation
链接: https://arxiv.org/abs/2609.16737
作者: Hojin Lee,Sizhe Lester Li,Maximilian Hilger,Susie Lu,Achim J. Lilienthal,Vincent Sitzmann,Daniel A. Duecker
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Project website: this https URL
Abstract:Generative video models can serve as a promising backbone for robot navigation by predicting future observations as video plans. Recent approaches often condition video planning on short-horizon guidance and recover geometric waypoints through scene reconstruction, leaving longer-horizon planning and precise video-to-action translation less explored. We present CueNav, a video model-based navigation framework combining visual cue guided video planning with an embodiment-specific Inverse-Dynamics Model (IDM). As visual cues, we use a Bird’s-Eye View (BEV) map to convey global task context and retain part of the robot body in the egocentric observation to expose embodiment context. These cues guide the video planner, while the IDM translates dense flow fields extracted from the video plan into robot actions. With the visual cue encoding global task context, CueNav achieves nearly 2x higher success in maze navigation than planning without the cue. The body-aware view with the IDM enables precise navigation with 70% success in a narrow passage where comparison methods largely fail to complete the task. We further demonstrate zero-shot semantic-conditioned navigation and deployment of the same video planner across different robot platforms. Our results show that visual cue-guided video planning with embodiment-specific action grounding paves the way toward a generalizable navigation framework for longer-horizon planning and embodiment-aware control. Additional results and code are available on our project website: this https URL.
[CV-63] PriorPose: Reference-Guided Joint Deformation and Alignment for Category-Level Object Pose Estimation ECCV2026
链接: https://arxiv.org/abs/2609.16727
作者: Yihan Chen,Huan Ren,Wenfei Yang,Hang Du,Tianzhu Zhang,Feng Wu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ECCV 2026
Abstract:Category-level object pose estimation seeks to recover a similarity transform (R,t,s) for unseen instances without instance-specific CAD models. Most competitive methods are correspondence-based: prior-free variants regress canonical (NOCS) coordinates directly from local observations and implicitly memorize the canonical frame in the weights, which ties the parameters to category-typical orientations and hurts generalization under distribution shift; prior-based variants introduce a category prior but typically follow a serial deform-then-align pipeline, where underconstrained canonical completion can corrupt correspondences and induce error cascades in pose. We propose PriorPose, a reference-guided correspondence framework that keeps the category prior explicit and solves canonicalization and alignment jointly in a shared feature space. A reference-guided seeded transformer embeds the partial observation and the category prior as token sets and fuses them via geometry-aware seeds, from which the network jointly predicts a per-point NOCS field for visible points and a canonical deformation of the prior that reconstructs a full canonical instance, while a deep pose head regresses (R,t,s) from the induced correspondences. A two-part shape consistency objective, with canonical-space and camera-space consistency losses, couples correspondence, deformation, and pose, reducing reliance on memorized canonical orientations and avoiding deform-then-align error cascades. Experiments on standard and larger-category benchmarks demonstrate that PriorPose sets new state-of-the-art results on most evaluated metrics, especially under strict pose thresholds, while remaining competitive on relaxed pose and IoU metrics and showing improved robustness under shape variation and domain shift.
[CV-64] MAETrack: Unleashing the Potential of Pretrained Geometric Priors for 3D Single Object Tracking
链接: https://arxiv.org/abs/2609.16695
作者: Sifan Zhou,Qiwei Wang,Linyue Tan,Ziyu Liu,Ziyu Zhao,Xiaobo Lu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 35 pages, 5 figures
Abstract:Large-scale pre-training has transformed representation learning in 2D vision, yet its transferability to 3D single object tracking (SOT) remains insufficiently understood. Directly fine-tuning self-supervised 3D encoders, such as masked autoencoders (MAE), often leads to sub-optimal adaptation because the reconstruction objective is not fully aligned with the spatial-temporal matching requirements of tracking. In this paper, we observe that this difficulty can be interpreted as a layer-wise transfer mismatch: shallow layers tend to preserve transferable geometric cues, while deeper layers become increasingly specialized to the reconstruction pretext task and are less suitable for downstream tracking. Based on this observation, we propose MAETrack, a lightweight adaptation framework for transferring pre-training MAE representations to 3D SOT. MAETrack includes Layer-Selective Initialization (LSI), which initializes only the shallow stages of the tracking backbone from pre-trained weights while re-initializing deeper stages, and Geometric Residual Gating (GRG), which reinforces structurally salient regions in the search BEV features before template-search fusion through residual spatial modulation. Extensive experiments on standard 3D SOT benchmarks show that MAETrack consistently improves upon vanilla fine-tuning baselines with limited computational overhead. More broadly, our results suggest that effective transfer from 3D reconstruction pre-training to 3D tracking is not merely a matter of partial fine-tuning, but depends on a tracking-oriented transfer principle that preserves shallow geometry while adapting deeper representations to the downstream objective.
[CV-65] Efficient 3D Whole-Body PET Image Denoising via Conditional Rectified Flow With Optimized Sampling Strategy
链接: https://arxiv.org/abs/2609.16690
作者: Jiale Shen,Guolin Wang,Chenhao Wang,Xinhui Su,Wei Luo,Feng Yu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Reducing radiation exposure in Positron Emission Tomography (PET) is important for patient safety; however, ultra-low-dose imaging suffers from severe noise, which may affect diagnostic interpretation without appropriate image enhancement. While current 3D deep generative models, particularly diffusion models, have shown strong reconstruction fidelity, their practical use can be limited by long inference times. In contrast, faster 2D-based alternatives may have difficulty maintaining volumetric consistency, an important consideration for whole-body PET imaging analysis. To bridge this gap, we propose a one-pass conditional 3D rectified flow (3D Flow) framework for whole-body PET image denoising that incorporates a novel optimized non-uniform sampling strategy. The model is trained with a one-pass linear-interpolant velocity-matching objective. This approach reconstructs a full 3D volume in approximately 30 seconds in our implementation, compared with multi-hour inference for the evaluated 3D DDPM baseline. Evaluations including zero-shot transfer to an independent clinical dataset show that our model achieves favorable global image quality and lesion conspicuity compared with the evaluated 3D DDPM and DDIM baselines, including on challenging short-acquisition data. Furthermore, the proposed method shows promising zero-shot transfer performance across the evaluated datasets and unseen dose levels (down to 1/100 of the standard dose), with artifact-focused visual comparisons supporting the need for further lesion-level validation. By balancing reconstruction fidelity and computational efficiency, this work presents a candidate approach for ultra-low-dose whole-body PET image denoising.
[CV-66] Efficient Quantization-Aware Distillation with Cross-Modal Alignment for Edge Vision-Language Models
链接: https://arxiv.org/abs/2609.16689
作者: Jinwoo Jeon,GyuYeop Do,Yubin Lim,Nam-Joon Kim,Hyun Gon Ryu,Hyuk-Jae Lee,Byung-Jun Lee
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Large-scale vision-language models (VLM) such as CLIP enable strong open-vocabulary reasoning, yet deploying these capabilities on resource-constrained edge devices remains challenging. EdgeVL addresses this problem by distilling CLIP representations into lightweight multi-modal encoders and applying quantization-aware training (QAT) for efficient Open-Vocabulary Classification (OVC) on edge hardware. However, its two-stage optimization applies different objectives for distillation and QAT, and contrastive learning is performed within the quantized student space, which can result in inconsistent optimization and reduced training efficiency. Moreover, identical supervision across RGB and non-RGB modalities may lead to modality imbalance. We propose a unified framework for quantized semantic distillation tailored to edge deployment. By jointly optimizing distillation and quantization within a unified teacher-anchored framework, our method ensures consistent training under quantization, suppressing hard negatives and enlarging decision margins. Additionally, we design a lightweight cross-attention adapter that enhances non-RGB representations through RGB-guided semantic transfer, narrowing the modality gap. Extensive experiments demonstrate consistent improvements on non-RGB modalities while maintaining deployment efficiency.
[CV-67] Differentiable Mesh State Estimation via Factor Graph Inference for Deformable Object Reconstruction
链接: https://arxiv.org/abs/2609.16686
作者: Lidia Al-Zogbi,Fangjie Li,Samuel Tobin,James Ferguson,Nithesh Kumar,Alejandro Chara,Kuan-I Chung,Mingxing Rao,Ayberk Acar,Susheela Sharma Stern,Robert Webster,Daniel Moyer,Alan Kuntz,Caleb Rucker,Tucker Hermans,Jie Ying Wu
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 8 pages
Abstract:Estimating deformable object states remains a fundamental challenge in robotics and simulation. We propose a novel factor graph-based framework for probabilistic mesh state estimation of deformable objects. The method directly updates a tetrahedral mesh, a rich and physically-grounded representation of an environment, by combining physics priors, noisy sensor measurements, and temporal smoothness constraints within a unified probabilistic formulation. The estimation problem is posed as a nonlinear least-squares optimization and solved using Levenberg-Marquardt. Ex vivo central-airway obstruction experiments and simulations on deforming cube models demonstrate reliable and accurate reconstruction under both rigid motion and deformation, highlighting the potential of this probabilistic approach for principled, measurement-driven mesh state estimation in deformable object reconstruction.
[CV-68] MEgoVista: Multi-view Ego-aware Motion Estimation for Metric 4D Hands and Head in the Wild
链接: https://arxiv.org/abs/2609.16684
作者: Jiangong Xiao(1),Zhihao Zhang(2),Yifei Dong(3),Chao Ma(3),Zhouyi Jin(3),Zhiwen Hou(3),Li Liu(3),Weihuang Chen(2),Hongbin Sun(2),Maoqing Yao(3) ((1) Northwestern Polytechnical University, (2) Xi’an Jiaotong University, (3) Maniformer)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 13 pages, 3 figures, 3 tables
Abstract:Learning manipulation from human video requires high-fidelity hand-motion reconstruction in metric units. Today’s metric hand labels come from studio rigs and instrumented headsets, and both are confined in the same two ways: neither leaves a prepared setting, and neither is checked against an independent reference. Unconstrained head-worn recording promises the opposite trade-off, scaling with the number of people wearing a device. We therefore introduce MEgoVista, an offline pipeline that turns a single unprepared MEgo View recording into metric two-hand and head motion in one gravity-aligned world frame. Three properties set it apart from existing egocentric reconstruction systems: first, it reconstructs in settings studio volumes and tabletop rigs cannot reach, settling hand ownership at detection so bystander hands stay out of the wearer’s trajectory; second, it takes its metric gauge from calibrated stereo rather than a monocular prior, installing scale at initialisation so policies receive physical units, not arbitrary coordinates; third, both outputs are scored inside a motion-capture volume against independent Chingmu optical capture, under a protocol that audits its own reference and charges what a method declines to predict. MEgoVista is offered as a measured route from egocentric video to metric hand supervision, one that widens where such labels can be gathered.
[CV-69] Lesion-centered 3D mapping of colonoscopy procedures: validation of a hierarchical ensemble pipeline on public benchmark videos
链接: https://arxiv.org/abs/2609.16672
作者: Hyunjun Kim,Hyeonwoo Na,Jaewoo Lee
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 21 pages, 12 figures, 4 tables. Code: this http URL (Zenodo DOI https://doi.org/10.5281/zenodo.22136766 )
Abstract:Background and Objective: Colonoscopy recording practice preserves text reports and still photographs, while the spatial information already present in the recorded video - where the scope traveled, where a lesion was observed, and whether the same lesion was seen again - is discarded when the procedure ends. This study determines whether a lesion-centered spatial record can be assembled and validated without full-colon 3D reconstruction. Methods: A four-layer hierarchical pipeline was assembled - (1) a global topological map, (2) lesion-level spatio-temporal tracks, (3) on-demand local 3D reconstruction, and (4) persistent lesion identity across repeated observations - and ran end to end on four public videos (two C3VDv2 sequences with ground-truth depth and two full REAL-Colon procedures; 40,245 frames). All components are published, individually validated methods; the contribution is their lesion-centered assembly, linking rules, and evaluation. Results: Revisits, impossible under forward-only mapping by construction, were detected by entry-map Bayesian localization: 5,614 and 4,043 revisit events (56 and 68 distinct nodes) in the two full procedures. Lesion-identity merging at the adopted threshold 0.5 maintained ground-truth purity 1.0 while auto-merging 20 of 231 candidate pairs. The endoscopy-specific geometry engine outperformed a general-purpose foundation model on all metrics (overall absolute relative error (AbsRel) 0.2276 vs. 0.3523). Conclusions: The results are partial but establish a concrete near-term path: revisit detection, lesion identity, and local 3D each returned quantitative, reproducible output without waiting for complete geometric reconstruction; validating the record on clinical data is the next step.
[CV-70] Bridging the Perceptual Gap: Residual-Enhanced Downscaling and Manifold-Aware Perception Alignment Adaptation for NR-IQA ICML2026
链接: https://arxiv.org/abs/2609.16664
作者: Yu Li,Zhengran Shen,Yachun Mi,Puchao Zhou,Shaohui Liu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by ICML2026
Abstract:Leveraging Large Vision-Language Models like CLIP has recently set new benchmarks for No-Reference Image Quality Assessment (NR-IQA). However, the contrastive pretraining of CLIP inherently prioritizes semantic invariance, which often suppresses subtle perceptual signals, a phenomenon we term perceptual submergence. Furthermore, standard preprocessing techniques (e.g., cropping and interpolation) further exacerbate the loss of critical high-frequency quality cues. In this paper, we propose the Cross-modal Perception Alignment Adapter (CMPA), a manifold-aware framework designed to disentangle perceptual distortions from dominant semantics. CMPA introduces a Perception-Sensitive Feature Extractor (PFE) that projects CLIP features into a compact, low-dimensional subspace, explicitly magnifying distortion-induced off-manifold deviations. Subsequently, a Cross-Modal Perception Alignment Injector (PAI) aligns these features with quality-aware text anchors and re-injects them into the backbone. To ensure input fidelity, we also devise a Residual-enhanced Perceptual Downscaling strategy that adaptively compensates for resolution-induced information loss using Just Noticeable Difference (JND) guided frequency re-injection. Extensive evaluations on several benchmark datasets demonstrate that our approach significantly outperforms state-of-the-art methods, effectively recovering the perceptual signals submerged in semantic-dense representations.
[CV-71] SAVTrack: Selective Vote Aggregation for Reliability-Aware Point Cloud Tracking
链接: https://arxiv.org/abs/2609.16662
作者: Sifan Zhou,Linyue Tan,Qiwei Wang,Ziyu Zhao,Xiaobo Lu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 12 pages, 5 figures
Abstract:3D single object tracking (SOT) in LiDAR point clouds is essential for autonomous systems, but remains challenging under sparse and incomplete observations. In such cases, different target points provide highly uneven constraints on the object center, causing some point-to-center votes to be substantially less reliable than others. Existing point-based trackers typically aggregate these hypotheses without explicitly modeling their reliability, allowing inaccurate votes to contaminate proposal clustering and degrade localization accuracy. To address this issue, we propose \textbfSAVTrack, a motion-aware tracking framework with \textbfSelective Vote Aggregation (SAV). SAVTrack estimates the reliability of each candidate vote from both local seed features and inter-frame motion context, and removes low-confidence hypotheses before proposal clustering. This pre-aggregation gating prevents unreliable hypotheses from affecting cluster formation while introducing only modest computational overhead. SAVTrack achieves competitive performance on KITTI and nuScenes, reaching 68.4/87.4 and 58.44/69.82 Success/Precision, respectively, while running at 82 FPS. It retains fewer than one-sixth of the candidate votes used by dense aggregation and remains particularly effective under sparse target observations.
[CV-72] Channel-Wise and Token-Aware Post-Training Quantization for Visual State Space Duality
链接: https://arxiv.org/abs/2609.16656
作者: Jonghyeon Lim,Changhoon Yim
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 10 pages, 6 figures, 5 tables
Abstract:State space models (SSMs), particularly Mamba, have emerged as efficient alternatives to attention-based architectures and have been extended to vision through ViM, VMamba, and Visual State Space Duality (VSSD). Yet the low-bit post-training quantization (PTQ) behavior of VSSD remains insufficiently understood. A weight-activation split on VSSD-Tiny identifies activation quantization as the dominant low-bit bottleneck, while representative inputs to selected VSSD-backbone linear layers exhibit strong channel-wise magnitude variation and token-localized extremes. We propose the Channel-wise Token-balanced Output-Aware Clipping (CTOAC) method, which learns per-input-channel clipping bounds by minimizing a token-balanced reconstruction loss on the corresponding linear outputs. Only the selected linear layers and their input activations are quantized; other backbone operations retain their original precision. Across VSSD-Tiny, VSSD-Small, and VSSD-Base, the proposed CTOAC method retains ImageNet-1K accuracy and remains substantially more robust than the evaluated baselines at more aggressive precision settings. Applying the same quantization scope to VSSD backbones on COCO and ADE20K preserves strong object detection, instance segmentation, and semantic segmentation performance. An optimized RTX 4090 deployment configuration achieves up to 1.42x end-to-end speedup over FP32.
[CV-73] ViD: Vision-Dominant Gender Bias Mitigation for Large Vision-Language Models EMNLP2026
链接: https://arxiv.org/abs/2609.16647
作者: Zhipeng Zhao,Zhaoqiang Wei,Peishun Liu,Youwei Zhao,Ruichun Tang
类目: Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM)
备注: EMNLP 2026 Main
Abstract:Gender bias in large vision-language models (LVLMs) undermines their fairness and reliability, compromising output trustworthiness. Current mitigation methods rely on training-phase adjustments or post-hoc calibration, but face limitations in dynamic visual bias mitigation. These include inability to capture real-time visual-textual incongruence, dependence on predefined gender bias taxonomies, and degraded cross-modal alignment with emergent bias patterns. To address these challenges, we propose ViD, a causally-inspired framework that analyzes attention mechanisms across five distinct patterns, revealing confounding effects from strong language priors. ViD demonstrates that visual-to-language cross-attention effectively suppresses bias while preserving general reasoning capabilities and text generation quality. ViD incorporates dual mechanisms: backdoor adjustment counters strong language priors, while refined token selection in decoding layers optimizes processing. This enhances model robustness and inference efficiency. Our integrated approach significantly mitigates gender bias across multidimensional social attributes in LVLMs, improving visual grounding and output fairness. Cross-benchmark validation shows ViD reduces gender bias by 14.7% on single-attribute evaluations (FACET) and achieves significant improvements on image captioning tasks (MS COCO), with gender bias score improving from 0.6708 to 0.9978 for LLaVA. Crucially, these improvements require no additional training overhead, making ViD a scalable and practical solution for bias mitigation in LVLMs.
[CV-74] What Do Hallucinations Reveal About Multimodal Reasoning ? Diagnosing Visual Grounding Failures via Contrastive Decoding Probes EMNLP2026
链接: https://arxiv.org/abs/2609.16646
作者: Zhipeng Zhao,Wenxu Wang,Peishun Liu,Ruichun Tang
类目: Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM)
备注: EMNLP 2026
Abstract:When strong multimodal models are widely available, progress requires new scientific methodologies beyond benchmark scores—using models as instruments for understanding behavior. We address this by asking: can we use large vision-language models (LVLMs) as experimental instruments for studying their own failure dynamics? Focusing on visual hallucination, we introduce SAFE, a training-free decoding framework that contrasts visually-grounded and vision-ablated generation paths to produce a token-level contrastive grounding score that identifies when the model favors linguistic priors over visual evidence. This signal serves dual roles: as a practical proxy for detecting visually-ungrounded tokens, and as the basis for decoding-time penalties. Our analysis yields three empirical observations: visual dependency decays over generation, hallucinations co-occur in temporal clusters, and early intervention reduces clustering without substantially degrading fluency. On MMHalBench, SAFE substantially outperforms all compared baselines; results elsewhere are more mixed. We argue that designing contrastive probes exemplifies a broader mission: using models as instruments for scientific understanding. Code: this https URL.
[CV-75] Can Knowledge Transfer Parameters Be Learned? LePoKet for Efficient Robotic Vision
链接: https://arxiv.org/abs/2609.16637
作者: Yanick C. Tchenko,Felix Mohr,Hicham Hadj-Abdelkader,Hedi Tabia
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Efficient perception is central to robotic systems operating under constrained computation, memory, and latency budgets. Knowledge transfer from larger pretrained models offers a practical route to stronger compact perception networks, but existing approaches commonly rely on fixed distillation objectives or manually designed interaction mechanisms. Building on Hereditary Knowledge Transfer (HKT), we propose LePoKet (Learnable Parameter Optimization for Knowledge Transfer), a structural transfer framework that embeds knowledge inheritance directly into the forward computation. LePoKet introduces a block-wise Extract-Transform-Mix interface whose interaction parameters are optimized jointly with the child network through a Learnable Genetic Attention (LGA) operator, without auxiliary distillation losses or temperature scaling. We first characterize the mechanism on CIFAR-10 and CIFAR-100 using ResNet parent-child pairs, obtaining relative error reductions of 24.57% and 25.1%, respectively, over standard child training. We then evaluate LePoKet for dense motion estimation by integrating it into a compact RAFT-based optical-flow model trained only on FlyingChairs and FlyingThings3D. LePoKet improves the compact RAFT baseline from 2.21 to 1.92 EPE on Sintel Clean, from 3.35 to 3.01 on Sintel Final, and from 7.51 to 6.39 on KITTI. A direct comparison with HKT further shows that LePoKet improves CIFAR-10 accuracy from 92.40% to 93.40% while achieving the best Sintel Final and KITTI errors among the evaluated compact transfer variants, with comparable performance on Sintel Clean. These results demonstrate that learnable structural transfer generalizes across recognition and motion perception tasks and provides a promising approach for efficient robotic vision.
[CV-76] JewelTry: Mask-Free Scale Aware Jewelry Virtual Try-On
链接: https://arxiv.org/abs/2609.16626
作者: Xinlei Niu,Peixia Li,Jun Wang,Chenchen Xu,Jiayu Yang,Jing Zhang,Pulak Purkait,Hongdong Li
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Virtual try-on (VTON) enables customers to visualize how fashion products appear when worn and has become an important technology for online shopping. While recent advances have substantially improved garment VTON, jewelry remains a challenging and underexplored category due to its small size, rigid structure, and sensitivity to fine-grained visual details. Realistic jewelry VTON requires not only faithful appearance transfer but also accurate scale and placement relative to the wearer. Existing jewelry VTON methods typically rely on mask guidance, whereas mask-free approaches lack explicit guidance for modeling the product scale. To bridge this gap, we introduce JVTO-Bench, a benchmark dataset for scale-faithful jewelry VTON, providing reference source target triplets with real-world product-scale annotations across four major jewelry categories. Building upon this benchmark, we propose JewelTry, a mask-free diffusion framework for scale-aware jewelry VTON. JewelTry incorporates a scale adapter that encodes product dimensions into a scale token, enabling the model to learn scale relationships between jewelry items and surrounding human anatomy in-context. To further improve jewelry consistency, we introduce a single-directional condition attention mechanism and an attention refinement loss that preserve both coarse geometry and fine-grained structural details of the reference jewelry. Extensive experiments show that JewelTry achieves a balance among visual fidelity, background preservation, object consistency and scale accuracy, establishing a strong baseline for mask-free, scale-aware jewelry virtual try-on.
[CV-77] EgoPathBench: Evaluating Zero-Shot Egocentric Waypoint Decision-Making in Vision-Language Models
链接: https://arxiv.org/abs/2609.16610
作者: Yang Zhao,Zhuo Chen,Xubo Yang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 18 pages, including supplementary material
Abstract:Zero-shot waypoint navigation requires vision-language models to select, from the current first-person observation, a sequence of spatial actions that is feasible for the agent and reaches the goal, placing joint demands on the integrated spatial intelligence of today’s foundation VLMs. Existing spatial-intelligence benchmarks primarily evaluate isolated judgments of relations, directions, or targets and therefore do not directly measure the integrated navigation ability required to combine target recognition, action-consequence assessment, distance estimation, and path planning. To fill this evaluation gap, we introduce EgoPathBench, a dataset and five-task benchmark for first-person waypoint decision-making. Each question presents an egocentric RGB image, a natural-language goal, and numbered visible waypoints; a model returns traversable candidates or an ordered route. Predictions are evaluated for candidate feasibility, adjacent-edge legality, and goal arrival under point-agent or embodied geometry. EgoPathBench contains 31,852 training, 1,345 validation, and 1,111 benchmark questions and retains at least one geometrically verified reference route for every route question. Across nine VLMs, the highest EgoPath Score is only 28.3. The top-ranked model reaches 35.9% success on Point Path, but only 2.9% and 4.0% on Embodied Path and Intent Path, respectively, showing that current models remain limited in forming complete, goal-consistent routes under embodiment constraints. Beyond the evaluation data, we release the corresponding training resource. Fine-tuning Qwen 3.5 4B on the released training split raises its EgoPath Score from 3.9 to 38.9 and improves all four reported evaluations across three external spatial benchmarks, with gains of 1.4–9.6 points.
[CV-78] G3AR: Graph-Guided Neural Visual Geometry for Scalable Multi-Sequence Aerial Registration SIGGRAPH
链接: https://arxiv.org/abs/2609.16603
作者: Jeng Wen Joshua Lean,Ting-Yu Yen,Wei-Fang Sun,Simon See,Hung-Kuo Chu,Shih-Hsuan Hung
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 6 pages, 4 figures, 8 tables. Accepted to SIGGRAPH Asia 2026 Technical Communications
Abstract:Full-context neural visual geometry is impractical for thousands of images, while sequence-based chunking poorly captures irregular non-local overlap in multi-sequence aerial collections. We present Graph-Guided Neural Visual Geometry for Aerial Registration (G3AR), a graph-guided framework for scalable dense neural geometry. Before local inference, G3AR builds a geometrically verified image-proximity graph that guides bounded overlapping chunks and induces a chunk graph whose maximum spanning tree defines alignment topology. Compatible backbones process chunks independently; shared-image predictions then estimate three-dimensional similarity (Sim(3)) transforms that register local cameras and geometry in a common frame. Across four real aerial scenes, G3AR improves pose error and runtime in matched VGGT- and Pi3-backed comparisons, while its DA3 variant achieves the lowest pose error among evaluated neural-geometry methods.
[CV-79] SAVOR: Self-Aware Visual Grounding via Confidence-Calibrated Reinforcement Learning for Multimodal Hallucination Mitigation ICONIP2026
链接: https://arxiv.org/abs/2609.16601
作者: Zixiu Ding,Zilin Zhao,Yingjie He,Xinlang Kang,Guansu Wang,Wei Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 33rd International Conference on Neural Information Processing (ICONIP 2026)
Abstract:Multimodal large language models (MLLMs) have made strong progress on visual question answering and image captioning, yet they still produce fluent claims about objects, attributes, or relations that are not grounded in the image. Many remedies either modify decoding at test time, which adds latency, or fine tune with preferences such as DPO variants, which teach which answer is preferred but not when the model’s own answer is unreliable. We argue that calibrated self assessment is the missing signal. We introduce Savor, a training framework that (i) augments the output schema with token and answer confidence, (ii) optimises the policy with a Group Relative Policy Optimisation (GRPO) objective that penalises calibration error and poor abstention decisions, and (iii) uses the learned confidence at inference time to revisit visual evidence only when the model is uncertain. Experiments on POPE, HallusionBench, AMBER and MMHal-Bench across two recent backbones (InternVL3-8B and Qwen3-VL-8B) show that Savor reduces hallucination while preserving general capability on MME and MMBench, with lower Expected Calibration Error than DPO and decoding baselines.
[CV-80] A Vision-Language Foundation Model for Precise and Comprehensive Brain Tumor Diagnosis from Preoperative Multimodal Data
链接: https://arxiv.org/abs/2609.16597
作者: Yinong Wang,Jianwen Chen,Zhou Chen,Shuwen Kuang,Haoning Jiang,Yanzhao Shi,Huichun Yuan,Yan-ran(Joyce)Wang,Bing Wang,Lei Wu,Bin Tang,Li Meng,Baihua Luo,Bin Zhou,Wei Ding,Weiming Zhong,Wei Hou,Yuanbing Chen,Zhiping Wan,Wei Wang,Zhenkun Xiao,Wenwu Wan,Allen He,Yuyin Zhou,Longbo Zhang,Feifei Wang,Zhixiong Liu,Michael Iv,Xuan Gong,Liangqiong Qu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Databases (cs.DB)
备注: 94 pages, 22 Figures
Abstract:Background Non-invasive presurgical diagnosis of brain tumor types from Magnetic Resonance Imaging (MRI) is essential but challenging due to overlapping imaging features across tumor types, inter-observer variability, and the extensive training required for expertise. We aimed to develop an MRI-based Artificial Intelligence (AI) model for automatic and reliable brain tumor classification with diagnostic uncertainty quantification and radiology reports generation. Methods We developed BrainVLM to classify all 12 World Health Organization (WHO) 2021 brain tumor types. BrainVLM integrates an uncertainty quantification strategy to indicate prediction reliability and a module for generating radiology reports to elucidate the clinical rationale. BrainVLM was trained on multi-modal data (MRI scans, demographics, and radiology reports) from 40,043 individuals. It was validated on 5,211 patients with pathologically confirmed brain tumors, including 3,877 held-out patients from the primary hospital and 1,334 patients from 11 independent hospitals. We further conducted two proof-of-concept studies to validate its clinical utility in AI-clinician workflows: 1) a blinded multi-reader study where 12 neuroradiologists across varying experience levels interpreted 248 retrospective cases with or without AI assistance, and 2) a real-world prospective study in which 1,009 patients were independently and blindly assessed by BrainVLM and radiologists before surgery. Additionally, we demonstrated BrainVLM’s utility in preoperative molecular subgroup prediction for adult-type diffuse gliomas, using a multi-center cohort of 632 patients. Comments: 94 pages, 22 Figures Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Databases (cs.DB) Cite as: arXiv:2609.16597 [cs.CV] (or arXiv:2609.16597v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2609.16597 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-81] FRPSS: Feature Rearrangement in Pre-Shape Space for Single-Image Generation
链接: https://arxiv.org/abs/2609.16594
作者: Yuexing Han,Haoxuan Zhang,Bing Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 28 pages, 18 figures
Abstract:Generative models trained on a single image often struggle to balance global structural integrity and local diversity. Existing single-image generation methods commonly rely on random noise to drive the generation process and lack explicit global structural constraints, making the generated results prone to spatial structural misalignment when structural variations occur. To address the issue, Feature Rearrangement in Pre-Shape Space for Single-Image Generation (FRPSS) is proposed in this paper. The core of FRPSS is the Manifold Structural Rearrangement with Feature Augmentation on Geodesic Surface (MSR-FAGS) module. MSR-FAGS replaces the randomly initialized features of the low-scale generator with rearranged Pre-Shape features and uses the features to guide image generation at subsequent scales, thereby reducing the risk of structural misalignment. To support downstream tasks such as stylization, a Scale-adaptive Sliding-window Patch Extraction (SSPE) strategy is further designed, and a directional Contrastive Language-Image Pre-training supervision module with SSPE (CLIP-SSPE) is constructed. Qualitative and quantitative experiments demonstrate that FRPSS achieves the best Single Image Fréchet Inception Distance (SIFID) scores on all three datasets while maintaining competitive Learned Perceptual Image Patch Similarity (LPIPS). Further qualitative experiments verify the effectiveness of FRPSS across multiple downstream tasks with the CLIP-SSPE module.
[CV-82] FLAT: Resampling Image and Text into 1D Flexible-Length Aligned Transmodal Tokens for Retrieval and Generation
链接: https://arxiv.org/abs/2609.16591
作者: Guangyu Sun,Shlok Kumar Mishra,Wentao Bao,Robert Zhenheng Yang,Xiao Wang,Xiyuan Wang,Yujunrong Ma,Chen Yuan,Max Xiangjun Fan,Jun Xiao,Jianpeng Cheng
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Traditional multimodal representation learning and generation are two stages: a contrastive or self-supervised visual encoder is trained first, followed by a separate downstream generative model. This setup bottlenecks generative performance behind frozen embeddings. To bridge this gap, we revisit joint multimodal representation learning and generation to produce linearly interpolatable embeddings that are directly consumable by generative decoders. We present FLAT (Flexible-Length Aligned Transmodal representations), a representation pre-training framework that jointly optimizes a shared multimodal encoder alongside downstream text-to-image (T2I) and image-to-text (I2T) decoders. By combining contrastive alignment with bidirectional cross-modal generative objectives, FLAT ensures its representations function as both discriminative semantic descriptors and generative conditions. Architecturally, FLAT maps visual and textual inputs into a unified continuous 1D sequence space, applying nested dropout over prefix-K tokens to enable dynamic output lengths. A single pre-training stage allows FLAT to perform cross-modal retrieval and generation across variable prefix K, achieving a T2I GenEval score of 71.1. Task-specific fine-tuning aligns model performance with state-of-the-art baselines: 83.1 GenEval on T2I generation; 40.5 BLEU-4 and 138.6 CIDEr on MS-COCO image captioning; and Recall@5 scores of 86.8 (I2T) / 75.8 (T2I) on MS-COCO alongside 98.3 (I2T) / 93.6 (T2I) on Flickr30K. Finally, qualitative evaluations demonstrate that FLAT representations natively support linear interpolation, latent space arithmetic, and zero-shot composed retrieval.
[CV-83] GraLoD: Graphics-Inspired Continuous Level-of-Detail Learning for Image Restoration
链接: https://arxiv.org/abs/2609.16578
作者: Hu Gao,Lizhuang Ma,Yulong Chen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:The spatial support required for image restoration varies across degradation types, image regions, and reconstruction stages. However, most existing methods rely on predefined multi-scale hierarchies and aggregate features through fixed fusion or attention, leaving the representation scale itself largely determined by the network architecture. This limitation becomes more pronounced when a task-specific backbone is extended to heterogeneous degradations in all-in-one restoration. Inspired by level-of-detail (LOD) rendering in computer graphics, we propose GraLoD, a plug-and-play framework that treats restoration scale as a spatially varying and stage-dependent continuous variable. GraLoD reuses the native encoder hierarchy, aligns its multi-scale features into a shared LOD representation space, and predicts a stage-conditioned LOD field at each decoder stage. Each spatial location then continuously queries only two neighboring representation levels, enabling the effective restoration scale to adapt to both local image content and reconstruction progress. To prevent degenerate or arbitrary scale selection, we further introduce minimal-sufficient footprint calibration (MSFC) together with structure-aware regularization (SAR) to encourage restoration-effective and spatially coherent LOD assignments. GraLoD can be directly integrated into existing restoration backbones without redesigning their fundamental feature-processing blocks. Extensive experiments demonstrate consistent improvements in task-specific and all-in-one restoration.
[CV-84] Efficient Text-to-Image Generation: An Adaptive Step Schedule Controller for Diffusion Models
链接: https://arxiv.org/abs/2609.16572
作者: Kuluhan Binici,Cihan Acar,Shivam Aggarwal,Siying Liu,Tulika Mitra
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Text-to-image diffusion models often use a fixed number of denoising steps, balancing time costs and image quality. However, the optimal number of steps depends on the complexity of the input text prompt. We propose an adaptive diffusion controller that dynamically adjusts the number of steps to generate high-quality images efficiently, without additional model training. By leveraging a mixture of step schedules with varying step sizes and evaluating the error term discrepancy at each timestep, our method transitions between schedules to optimize performance. Experiments on COCO and DiffusionDB show that our approach reduces inference time while maintaining visual fidelity, offering a more efficient alternative for text-to-image diffusion models.
[CV-85] Counterfactual Reasoning for Robust Visual Question Answering
链接: https://arxiv.org/abs/2609.16567
作者: Truong-Binh Duong,Thanh-Ngan Tran,Ngoc-Thao Nguyen,Bac Le
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted for publication at the 30th International Conference on Knowledge-Based and Intelligent Information Engineering Systems (KES 2026). 9 pages, 5 figures
Abstract:Modern Visual Question Answering (VQA) models often exploit spurious correlations in training data, leading to poor out-of-distribution (OOD) generalization due to language bias. Although counterfactual learning has shown promise, existing methods can be improved to better guide attention toward causal evidence and strengthen feature discrimination. To address this, we propose a novel training framework that enhances counterfactual contrastive learning for VQA. Our framework introduces three key contributions: (1) a three-stage curriculum for stable multi-objective optimization, (2) an enhanced Batch-Contrastive loss for more discriminative feature learning, and (3) two novel regularizers, Answer-Contrastive (AC) loss to refine the prediction space and Gradient-Discrepancy (GD) loss to enforce causal visual grounding. Our model achieves a competitive accuracy of 61.64% on the bias-sensitive VQA-CP v2 benchmark while maintaining 62.80% on the standard VQA v2 dataset, yielding a small generalization gap of 1.16%. This demonstrates a strong balance between OOD robustness and in-distribution performance.
[CV-86] Vision And Text Transformer For Predicting Answerability On Visual Question Answering
链接: https://arxiv.org/abs/2609.16565
作者: Tung Le,Huy Tien Nguyen,Le Minh Nguyen
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Answerability on Visual Question Answering is a novel and attractive task to predict answerable scores between images and questions in multi-modal data. Existing works often utilize a binary mapping from visual question answering systems into Answerability. It does not reflect the essence of this problem. Together with our consideration of Answerability in a regression task, we propose VT-Transformer, which exploits visual and textual features through Transformer architecture. Experimental results on VizWiz 2020 dataset show the effectiveness and robustness of VT-Transformer for Answerability on Visual Question Answering when comparing with competitive baselines.
[CV-87] Which Pretext Task Transfers? Self-Supervised Pretraining Objectives for Lung Ultrasound
链接: https://arxiv.org/abs/2609.16551
作者: Moein Heidari,Junbo Rao,Jai Choraria,Wenjin Chen,David J. Foran,Ilker Hacihaliloglu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Submitted to SPIE 2027
Abstract:Self-supervised learning (SSL) can reduce the need for labelled medical images, but the choice of pretext objective remains unclear for lung ultrasound (LUS). Contrastive learning, masked reconstruction, and joint-embedding predictive architectures (JEPA) differ in the space in which their targets are defined, yet existing ultrasound studies compare them under different corpora, backbones, and evaluation protocols. We compare these three objective families using the same encoder backbone, pretraining corpus, optimisation schedule, and frozen-evaluation protocol. Encoders are pretrained on COVID-BLUeS LUS videos and evaluated with linear, k NN, and attentive probes at 5%, 10%, 50%, and 100% label budgets. Evaluation is performed on POCUS using patient-level five-fold cross-validation and on the independently acquired Mendeley-Uganda dataset, which is excluded from both pretraining and probe fitting. At the full label budget under linear probing, VideoMAE and V-JEPA achieve 66.5 \pm 13.1 and 65.4 \pm 11.7 balanced accuracy on POCUS, while MoCo achieves 42.1 \pm 1.2 . On Mendeley-Uganda, the ranking reverses: MoCo performs best at 62.7 \pm 1.0 , followed by VideoMAE at 53.8 \pm 2.8 , while V-JEPA falls near chance at 35.1 \pm 4.9 . These results show that POCUS probe accuracy alone does not identify the objective that transfers best across datasets. We also outline planned representation-level analyses to examine this reversal. Code is publicly available at this https URL.
[CV-88] VPRef: A Cross-Domain Benchmark for Referring Remote Sensing Image Segmentation
链接: https://arxiv.org/abs/2609.16486
作者: Quanwei Liu,Tao Huang,Jiaqi Yang,Wei Xiang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 12 pages, 7 figures, 6 tables
Abstract:Rapid advancements in vision-language models have propelled Referring Remote Sensing Image Segmentation (RRSIS) to the forefront of Earth observation. However, practical deployments suffer severe performance degradation under a coupled dual-drift paradigm: visual domain drift from cross-spatial-resolution mismatches and spectral variations, alongside textual logic drift from unconstrained, variable user-input granularities. To mitigate these bottlenecks, this paper establishes the first cross-domain RRSIS benchmark, designated as the Vaihingen-Potsdam Referring (VPRef) dataset, comprising 46,972 language-image-annotation triplets organized into a three-tier linguistic hierarchy. Building upon this benchmark, we develop a tailored parameter-efficient domain adaptation baseline anchored on the Segment Anything Model (SAM3) via Low-Rank Adaptation (LoRA). Our framework counteracts visual distribution discrepancies through pseudo-label-driven self-training and addresses textual logic drift via random multi-granularity text prompt mixing. Crucially, the distribution of empirical metrics across ablative variants suggests a potential decoupling between cross-modal semantic robustification and visual domain alignment, demonstrating that linguistic variance drives fine-grained semantic invariance while pseudo-label propagation governs macro-scale spatial grid alignment. Extensive benchmarks demonstrate the proposed framework achieves superior cross-domain segmentation boundaries while modifying merely 1.08% of the foundational parameter footprint, establishing a robust baseline for future multi-modal remote sensing domain adaptation research. The dataset and code will be available at this https URL.
[CV-89] MDN-Control: Mask-Depth-Noise Guided Region Control for Multi-Subject Video Editing
链接: https://arxiv.org/abs/2609.16475
作者: Jiayi Yu,Xi Ye,Lina Wang,Yunkun Xia
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 5 pages, 3 figures
Abstract:Multi subject video editing modifies designated subjects while preserving non target content, but faces cross subject attribute leakage, and occlusion ambiguity. Existing approaches rely on masks and struggle to distinguish overlapping subjects or ensure consistent generation. To address these limitations, we propose MDN-Control, a training free framework jointly controlling target localization, occlusion geometry, and appearance initialization. Specifically, mask-guided localization provides consistent target localization, while depth-aware occlusion control resolves ambiguous boundaries between overlapping subjects. We further introduce noise latent prompting, which retrieves Gaussian initializations from a noise library for prompt relevant priors. Experiments on MSVBench show that MDN-Control achieves the lowest CM-Err and the highest Q-Edit, while maintaining competitive text alignment and temporal consistency, demonstrating the effectiveness of combining spatial, geometric, and latent priors for multi subject video editing.
[CV-90] HairCS: Reconstructing Strand-Based Hair from Hair Cards
链接: https://arxiv.org/abs/2609.16465
作者: Zixuan Lu(1),Tongtong Wang(2),Yuefan Shen(2),Zhongtian Zheng(2),Chenfanfu Jiang(3),Yin Yang(1),Kui Wu(2) ((1) University of Utah, (2) LIGHTSPEED, (3) UCLA)
类目: Graphics (cs.GR); Computer Vision and Pattern Recognition (cs.CV)
备注: 22 pages, 30 figures, 5 tables. Dataset: this https URL
Abstract:We present an automated pipeline that converts hair-card models into high-quality strand-based hairstyles. Given a collection of textured triangular or quad strips as input, our method produces a strand-based representation that preserves the original hairstyle while enriching it with fine-scale geometric detail and adhering to standard production requirements: strands originate from the scalp, roots are uniformly distributed, and the hair volume is plausibly filled. The resulting assets are directly compatible with strand-based rendering, physics-based simulation, and common grooming modifiers (e.g., clumping, curling, noise) for enhanced realism and artistic control. We validate our approach on a large and diverse set of hairstyles, including short and long hair, curly styles, and complex styles such as buns and ponytails.
[CV-91] OPD-Aha: From Linguistic Momentum to Visual Reflection in Multimodal On-Policy Distillation
链接: https://arxiv.org/abs/2609.16459
作者: Chenhao Qiu,Dawei Li,Yechao Zhang,Lei Gong,Zhen Tan
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: 24 pages, 12 figures, 7 tables
Abstract:Privileged on-policy distillation improves multimodal reasoning by allowing a teacher to evaluate student trajectories using rich, training-only visual evidence. Both models score these trajectories while conditioning on the same student-generated prefix. When a student misinterprets an image early in a response, this accumulating erroneous rationale eventually pulls the teacher away from its visual evidence. The teacher and student converge on the same hallucination, causing standard cross-model supervision to collapse precisely where correction is most needed. We find that the teacher’s visual corrective preference is not lost under this misleading agreement. Comparing the predictions of the identical teacher given the real image and a visual null reveals that the privileged evidence still pushes the model toward the correct interpretation. We introduce OPD-Aha, which reconstructs the distillation target directly from this isolated visual preference rather than relying on the fragile teacher-student discrepancy. This reconstructed target aggressively suppresses continuations that contradict the image. Trained with this objective, students learn to naturally interrupt their own flawed reasoning with reflection tokens such as wait and actually. After reflection, subsequent generation relies less on the accumulated erroneous text and more on the visual evidence. Correcting these trajectories mid-generation fundamentally alters the reasoning process, yielding broad and consistent improvements across diverse fine-grained perception and complex multimodal reasoning benchmarks. Our code and models are available at this https URL.
[CV-92] Decentralized Gossip Learning and Federated Averag ing for Histopathology Image Classification
链接: https://arxiv.org/abs/2609.16448
作者: Yusuf Ozturk,Enes Goltekin,Bengisu Atli,Akin Ozturk,Ulas Bagci
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Recently accepted to Neural Computing and Applications
Abstract:Breast histopathology analysis increasingly relies on distributed learning because direct data pooling across institutions is often restricted by privacy, governance, and communication constraints. This study compares server-based Federated Averaging (FedAvg), fully decentralized gossip learning, and Hybrid Gossip-FedAvg for invasive ductal carcinoma (IDC) patch classification. Experiments used 277,524 color image patches with patient-disjoint training, validation, and test partitions and a workload-balanced, Dirichlet-guided allocation across six nodes. Ring, random degree-3, and fully connected gossip topologies were evaluated together with sensitivity analyses for statistical heterogeneity, mixing coefficient, learning rate, model drift, prediction disagreement, calibration, clinically motivated operating points, communication payload, and patient-level IDC burden, together with auxiliary backbone robustness analyses. In the principal alpha=0.3 experiment, Hybrid Gossip-FedAvg achieved a test area under the receiver operating characteristic curve (ROC-AUC) of 0.8811, closely followed by FedAvg at 0.8801 and fully connected gossip at 0.8751. Across three independent patient-level repetitions, FedAvg and Hybrid Gossip-FedAvg obtained the same mean ROC-AUC of 0.9082, with standard deviations of 0.0037 and 0.0043, respectively. Hybrid achieved the highest mean area under the precision-recall curve of 0.8240, whereas FedAvg produced the lowest mean Brier score of 0.1335. Denser gossip graphs improved discrimination but increased theoretical model payload, while ring gossip remained sensitive to learning rate and mixing strength. Overall, FedAvg provided the most consistently reliable server-based baseline, topology-aware gossip offered a viable decentralized alternative, and Hybrid Gossip-FedAvg provided a balanced compromise between peer-to-peer diffusion and periodic global coordination.
[CV-93] he Neverwhere Visual Parkour Benchmark Suite IROS2026
链接: https://arxiv.org/abs/2609.16443
作者: Ziyu Chen,Henghui Bao,Haoran Chang,Alan Yu,Ran Choi,Kai McClennen,Gio Huh,Kevin Yang,Ri-Zhao Qiu,Yajvan Ravan,John J. Leonard,Xiaolong Wang,Phillip Isola,Ge Yang,Yue Wang
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 9 pages, 14 figures. Accepted to IROS 2026. Project page: this https URL
Abstract:State-of-the-art visual locomotion controllers are increasingly capable at handling complex visual environments, making evaluating their real-world performance before deployment increasingly difficult. This work intends to narrow this train/evaluation gap by developing a collection of hyper-photo-realistic, closed-loop evaluation environments - The Neverwhere Benchmark Suite - comprised of over sixty 3D Gaussian Splatting reconstructions of urban indoor and outdoor scenes. Our goal is to encourage large-scale and reproducible robot evaluation by making it easier to create and integrate Gaussian splats-based reconstructions into simulated continuous testing setups. We also underscore the potential pitfalls of relying exclusively on 3D Gaussian-generated data for training, by providing policy checkpoints trained over multiple Neverwhere scenes and their performance when evaluated in novel scenes. Our analysis illustrates the necessity of sourcing diverse data to ensure performance. Code and data are available on the project page: this https URL.
[CV-94] Reasoning with Image Generation
链接: https://arxiv.org/abs/2609.16409
作者: Nishad Singhi,Hector Garcia Rodriguez,Aditya Arora,Marcus Rohrbach,Anna Rohrbach
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to COLM 2026. Code this https URL and website this https URL
Abstract:Chain-of-thought reasoning has revolutionized natural language processing by enabling large language models (LLMs) to decompose problems into intermediate steps before answering. Yet confining reasoning to the textual domain presents limitations for tasks requiring direct manipulation of visual representations. Recent efforts augment multimodal LLMs with external visual expert tools such as depth estimation or object detection modules, but these remain fundamentally limited by their reliance on narrow, rigid operations that cannot flexibly generate or transform visual content. We propose ReImaGin, which leverages image generation models as a flexible visual reasoning mechanism for multimodal LLMs: unlike fixed-function tools, they accept natural language commands and can perform open-ended visual operations, like removing an occlusion or generating a floorplan from multiple disjoint views of a room. Across six diverse visual reasoning tasks including multi-view spatial reasoning and collision prediction, ReImaGin consistently outperforms both text-only reasoning and specialist vision-tool baselines, with gains of up to 25%, demonstrating the advantage of flexible, generative visual reasoning.
[CV-95] Geometry vs Structure: Graph-Based Diagnostics for LiDAR Point-Cloud Simulation Fidelity
链接: https://arxiv.org/abs/2609.16378
作者: Ghazal Farhani,Taufiq Rahman
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Digital twins provide a scalable and cost-effective complement to real-world testing for validating autonomous-driving and advanced driver-assistance system (ADAS) sensor pipelines. However, quantifying their fidelity remains challenging, particularly for 3D LiDAR point clouds, where conventional geometric metrics may overlook important structural discrepancies. We present a graph-based framework for evaluating the structural fidelity of simulated LiDAR point clouds against real-world scans. While scan-level metrics such as Chamfer distance capture point-wise geometric similarity, they do not explicitly represent connectivity, topology, or object-level organization. Our framework constructs graphs from real and simulated point clouds, applies Louvain community detection to identify spatially coherent subgraphs, and matches corresponding communities using centroid proximity. For each matched pair, we compute r_\lambda , a bounded graph-spectral metric motivated by Weyl’s inequality, and compare it with density-aware Chamfer distance (CDC) as a geometric baseline. Controlled perturbation experiments demonstrate that r_\lambda is invariant to rigid transformations and robust to sensor noise while remaining sensitive to structural deformation. We evaluate the framework on 50 paired real and simulated LiDAR scans acquired using a Velodyne VLP-32C sensor and CARLA, respectively. The dataset contains more than 1,000 matched communities across four representative classes: vehicles, vegetation, trees, and building walls. The results show that geometric and structural measures capture complementary aspects of simulation fidelity, supporting graph-spectral analysis as an additional diagnostic layer for validating digital twins in ADAS and autonomous-driving applications.
[CV-96] ConGraspXL: Controllable Constraint-Conditioned Dexterous Grasping Motion Synthesis
链接: https://arxiv.org/abs/2609.16319
作者: Hui Zhang,Mirko Meboldt,Jie Song
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
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:Dexterous grasping is usually conducted for specific tasks, leading to heterogeneous constraints such as specific approach directions, desired contact regions, specified wrist trajectories, and functional hand poses. Our previous work, GraspXL, achieves scalable grasping motion synthesis for diverse objects and hand morphologies, while lacking controllability for synthesis under such various task-driven constraints. In this paper, we propose ConGraspXL, which extends GraspXL with controllable constraint-conditioned grasp motion synthesis that accommodates diverse task-driven constraints and their combinations. We introduce a hierarchical constraint formulation, enable flexible constraint composition with a masked residual interface, and improve control precision with dynamic hand centers and feed-forward wrist guidance. Without losing the strong generalization capabilities of GraspXL, ConGraspXL enables precise and flexible controllability for various individual constraints and their combinations, providing a plug-and-play low-level grasp controller for downstream applications such as whole-body grasp completion, functional grasping, and human-motion imitation.
[CV-97] Racing in Volume with Flow Ensembles
链接: https://arxiv.org/abs/2609.16310
作者: Saswat Subhajyoti Mallick,Riu Cherdchusakulchai,Marc Ruiz Olle,Albert Mosella-Montoro,Jose Ribeiro-Gomes,Francisco Vicente Carrasco,Fernando De la Torre
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: project page at this https URL
Abstract:Streaming 4D reconstruction has been demonstrated only indoors, on dense camera rigs surrounding subjects that move at human pace. Outdoor 4D reconstruction exists but relies either on cameras mounted on the moving vehicle itself, or on limited-coverage arrays observing quasi-static subjects offline. The case that actually matters for spectators is a fast-moving subject, watched from a sparse ring of allocentric cameras, streaming. No method targets this, and no benchmark exists to evaluate one. To this end, we introduce FastFlowGS, a streaming 4D Gaussian Splatting method for reconstructing fast-moving subjects from a small set of fixed external cameras, and Monaco4D, a photorealistic Unreal Engine 5 benchmark for high-speed outdoor reconstruction. FastFlowGS fuses sparse matches, semi-dense tracks, and dense optical flow by lifting each signal to 3D with geometric uncertainty and combining them through a Kalman-style temporal update. Monaco4D provides Formula 1 sequences under varied illumination from trackside, onboard, and drone viewpoints with dense ground truth. On CMU-Panoptic, FastFlowGS exceeds the strongest baseline by 12.6% VMAF at 35% greater efficiency. On Monaco4D, where existing streaming methods degrade severely, it improves dynamic-region PSNR by up to 18.6% with 28.3% lower per-frame optimization time. Dataset and additional details can be found at this https URL.
[CV-98] Sequence Recognition in Bharatnatyam dance
链接: https://arxiv.org/abs/2609.16306
作者: Himadri Bhuyan,Rohit Dhaipule,Partha Pratim Das
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Accepted at 7th International Conference on Computer Vision and Image Processing (CVIP), 2022
Abstract:Bharatanatyam is the oldest Indian Classical Dance (ICD) which is learned and practiced across India and the world. Adavu is the core of this dance form. There exist 15 Adavus and 58 variations. Each Adavu variation comprises a well-defined set of motions and postures (called dance steps) that occur in a particular order. So, while learning Adavus, students not only learn the dance steps but also take care of its sequence of occurrences. This paper proposed a method to recognize these sequences. In this work, firstly, we recognize the involved Key Postures (KPs) and motions in the Adavu using Convolutional Neural Network (CNN) and Support Vector Machine (SVM), respectively. In this, CNN achieves 99% and SVM’s recognition accuracy becomes 84%. Next, we compare these KP and motion sequences with the ground truth to find the best match using the Edit Distance algorithm with an accuracy of 98%. The paper contributes hugely to the state-of-the-art in the form of digital heritage, dance tutoring system, and many more. The paper addresses three novelties; (a) Recognizing the sequences based on the KPs and motions rather than only KPs as reported in the earlier works. (b) The performance of the proposed work is measured by analyzing the prediction time per sequence. We also compare our proposed approach with the previous works that deal with the same problem statement. © It tests the scalability of the proposed approach by including all the Adavu variations, unlike the earlier literature, which uses only one/two variations.
[CV-99] ProtoLIP: From Sentence-Level to Object-Level Evidence Disentanglement
链接: https://arxiv.org/abs/2609.16284
作者: Yan Zhu,Yongbo Chen,Zhengming Ding,Rebecca Faust
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Query-conditioned vision–language models enable fine-grained interpretation by revealing how visual evidence changes with textual queries. However, evidence conditioned on complete descriptions does not necessarily resolve into object-specific evidence, nor does an exposed evidence map necessarily identify the evidence that constitutes the model’s prediction. Across multiple VLM architectures and independent benchmarks, we find that object-level queries often retain evidence from co-occurring objects and shared context. In this paper, we introduce \textbfProtoLIP, a lightweight prototype-mediated evidence layer that organizes reusable visual prototypes into text-derived semantic families and uses query-dependent family routing to constrain which prototypes may provide evidence. Without spatial annotations or backbone retraining, ProtoLIP improves evidence localization and separation across query granularities, with localization gains transferring to independently pretrained VLMs with well-aligned patch–text representations. Despite using only text-derived weak supervision, ProtoLIP remains competitive with a spatially supervised grounding model while maintaining strong matching and competitive image–text retrieval. Crucially, ProtoLIP constructs its matching score directly from localized prototype evidence, enabling the score to be exactly decomposed into semantic-family and prototype contributions.
[CV-100] Efficient Reasoning Distillation: Small Video-Language Models via Synthetic CoT and Difficulty-Aware Fine-Tuning
链接: https://arxiv.org/abs/2609.16255
作者: Mantek Singh,Jeshwanth Challagundla,Siddharth Raina,Jasmin Jarsania
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: 14 pages, 2 figures, 5 tables. Published in MultiMedia Modeling (MMM 2026), LNCS 16412
Abstract:We present an efficient method to distill reasoning capabilities into compact video-language models (VLMs) for video question answering (VideoQA). Our approach fine-tunes a 2B-parameter model using only \sim 900 uncertainty-selected examples, each augmented with synthetic chain-of-thought (CoT) rationales generated by a 4B teacher. Despite its minimal compute cost - under two hours on a single A100 GPU - our method enables the 2B model to outperform VLMs up to 4 \times larger, and generalize across CinePile, ActivityNet-QA, and MLVU, approaching the performance of its own 4B teacher. A key finding is that placing CoT rationales after the answer - contrary to standard prompting - substantially improves reasoning in compact models. This insight challenges prevailing CoT conventions and reveals new alignment strategies under limited model capacity. Our findings offer a practical blueprint for training deployable, reasoning-rich VLMs suited for mobile and edge applications.
[CV-101] SceneBench: A Hierarchical Benchmark for Vision-Language Understanding of 3D Scenes
链接: https://arxiv.org/abs/2609.16233
作者: Anubhav Khanal,Prabigya Acharya,Roshni Poudel,Sujan Kapali,Bigyan Bhatta,Pramish Paudel,Francois Rameau,Danda Pani Paudel
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Vision-language models excel at 2D image understanding but remain limited in 3D spatial reasoning. Progress is hindered by limitations in current benchmarks. First, 3D datasets often rely on point clouds that capture geometry but discard rich visual features like texture, text, and materials. Second, annotations treat objects in isolation while ignoring real-world hierarchical organization (scenes, rooms, functional areas, object groups). Third, evaluation tasks focus narrowly on basic recognition rather than multi-step spatial reasoning. In this context, we introduce SceneBench, a benchmark of 966 photorealistic 3D scenes reconstructed with Gaussian Splatting and densely annotated with hierarchical semantics spanning scenes, rooms, functional areas, object groups, and individual objects. These annotations are produced through a human-in-the-loop pipeline combining vision-language models with roughly 1,500 human-hours of iterative refinement and verification, producing over 183K annotated nodes with textual descriptions and 3D bounding boxes. Building on this representation, we define three evaluation tasks: Existence-Based Questions probing object attributes, Spatial Intelligence Questions covering counting, size comparison, distance, and directional relations, and Grounded Question-Reasoning-Answer (QRA) triplets requiring multi-step reasoning across semantic levels. Experiments with state-of-the-art vision-language models show that while models perform well on basic recognition tasks (e.g., up to 85% accuracy for detection), performance drops substantially on hierarchical and compositional reasoning (e.g., down to 60% for counting), revealing limitations not captured by existing benchmarks. SceneBench provides a realistic testbed for developing and evaluating models capable of fine-grained spatial reasoning in photorealistic 3D environments. Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) Cite as: arXiv:2609.16233 [cs.CV] (or arXiv:2609.16233v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2609.16233 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-102] Hyperbolic Contrastive Learning with Entailment for Spatial Transcriptomics MICCAI2026
链接: https://arxiv.org/abs/2609.16207
作者: Daniela Vega,Paula Cárdenas,Hannah Ceballos,Leonardo Manrique,Pablo Arbelaéz
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at MICCAI 2026
Abstract:Spatial Transcriptomics (ST) has transformed biomedical research by enabling the spatial mapping of gene expression across tissue sections. However, high operational costs, specialized equipment requirements, and sensitivity to experimental noise limit the accessibility and scalability of ST. Recent computer vision approaches aim to overcome these limitations by predicting spatial gene expression directly from histopathology images. While effective, current approaches often suffer from gene expression over-smoothing and overly uniform predictions across tissue regions, suggesting that further progress depends on learning representations that reflect the hierarchical and asymmetric structure of gene regulation and tissue morphology. To address these issues, we propose Hyperbolic Contrastive Learning with Entailment for Spatial Transcriptomics (HyCLoST), a hyperbolic contrastive learning model that captures the intrinsic hierarchical relationships within ST data. By leveraging hyperbolic geometry and a gene-to-image entailment loss, HyCLoST learns structured, biologically grounded representations that improve gene expression prediction accuracy, achieving a 6% reduction in MSE and an 8% increase in PCC across 26 ST datasets, over previous methods. Our source code is publicly available at this https URL
[CV-103] Occupancy Network-Guided Autonomous Robotic Partial Nephrectomy
链接: https://arxiv.org/abs/2609.16186
作者: Ethan Kilmer,Pit Henrich,Jiawei Ge,Paul M. Scheikl,Laura Connolly,Soum D. Lokeshwar,Joseph Chen,Justin D. Opfermann,Kaitlyn Kumar,Lauren Shepard,Ahmed Ghazi,Nirmish Singla,Richard J. Cha,Kevin Cleary,Franziska Mathis-Ullrich,Axel Krieger
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Autonomous soft-tissue cancer surgery has been limited to interventions on organ surfaces, because current systems cannot perceive and adapt to anatomy once it deforms or is cut. We introduce the first vision-guided autonomous system capable of performing complete tumor resections for partial nephrectomy. Our system integrates conditional occupancy networks, trained entirely in a physics-based simulation, that infer full 3-D anatomy (tumor, margin tissue, and kidney) from single-view partial point clouds. These occupancy networks maintain intraoperative tracking even as tissue is cut and deformed, enabling adaptive planning and execution. The surgical platform combines a depth camera for capturing surface point clouds, dual robotic arms for electrosurgical cutting and vacuum-based tissue manipulation, and an autonomous control strategy for tumor resection. In patient-derived hydrogel phantoms under an open partial nephrectomy setting, the robot performed eight consecutive autonomous tumor resections comprising 77 electrosurgical cuts, with all cuts achieving negative surgical margins and 1.61 \pm 0.48 mm mean absolute margin error. This work demonstrates, for the first time, a foundation for supervised autonomous closed-loop, imaging-driven, margin-negative tumor removal in phantoms.
[CV-104] DenseFace: Bias Mitigation in Face Recognition via Density-Aware Probabilistic Matching
链接: https://arxiv.org/abs/2609.16149
作者: Mansur Bultygov,Vadim Seliutin,Dmitry Nekhaev,Ivan Laptev
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 13 pages, 10 figures. Accepted at IEEE/IAPR International Joint Conference on Biometrics (IJCB) 2026
Abstract:Despite steady progress in face recognition, current face recognition models still suffer from significant demographic biases. While approaches for bias mitigation have been proposed, existing methods often impose constraints on the training procedure and result in the degradation of recognition accuracy. To address this issue, we here introduce a method that reduces racial bias in pre-trained face recognition models without compromising their accuracy. To this end, we model face embeddings of each person by von Mises-Fisher (MF) distribution. We next observe the dependency between demographic attributes and the density of MF distributions, and propose DenseFace, a probabilistic face matching procedure that accounts for differences in MF distributions. Our extensive experiments demonstrate DenseFace to consistently reduce racial bias in strong face recognition models varying in network architectures, training datasets and loss functions. Notably, DenseFace preserves recognition accuracy and requires no retraining of the underlying face recognition model. Our work also investigates previously adopted bias measures and makes suggestions.
[CV-105] World-Action Models for Robot Learning and Control: A Survey
链接: https://arxiv.org/abs/2609.16074
作者: Zuxing Lu,Hongjia Zhai,Guanzhi Wang,Huajian Zeng,Jiaqi Yang,Jingyu Liu,Lei Cheng,Yuantai Zhang,Yuheng Qiu,Zezhou Cheng,Ivan Laptev,Danfei Xu,Benjamin Riviere,Giuseppe Loianno,Eric Xing,Xingxing Zuo
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 19 pages
Abstract:Robots operating in open environments act under partial observability, physical constraints, and dynamic task contexts. Beyond mapping observations and language instructions to actions, they must anticipate how candidate actions may affect future states and task-relevant outcomes. Recent advances in world models, video generation, and Vision-Language-Action (VLA) policies have motivated the development of World-Action Models (WAMs), which couple future world prediction with executable action generation. This survey provides a robotics-oriented review of WAMs. We clarify their scope relative to conventional world models, model-based reinforcement learning, action-conditioned video generation, and reactive VLA policies, and organize existing methods through a unified taxonomy covering representations, transition modeling, action interfaces, architectures, training pipelines, data modalities, and scaling strategies. We further review applications of WAMs in manipulation, navigation, and autonomous driving, and we summarize the datasets, benchmarks, metrics, and protocols used to evaluate WAM systems. Finally, we discuss key challenges in action alignment, world-action factorization, spatial and multi-view consistency, long-horizon memory, neural simulation for closed-loop policy learning, and efficient inference. Taken together, this survey aims to provide a concise technical foundation for integrating predictive world modeling with action generation, toward more reliable embodied robot intelligence. Project page: this https URL.
[CV-106] 3D Field Data Reduction with Adaptive Sample-Based Gaussian-Encoded Reconstruction
链接: https://arxiv.org/abs/2609.16024
作者: Michael R. Martin,Joseph Insley,Victor A. Mateevitsi,Silvio Rizzi,Kwan-Liu Ma
类目: Graphics (cs.GR); Computational Engineering, Finance, and Science (cs.CE); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 10 pages, 8 figures, 8 Tables
Abstract:In scientific simulation, regular grids, unstructured meshes, and particle-based formats are chosen to represent field data for computational efficiency, geometry/adaptive flexibility, and following motion/deformation, respectively. Each of these field data formats is often handled through separate data-specific processing pipelines. We present a unified sample-based Gaussian encoding method that represents these data forms under a single fixed-budget formulation. The method initializes and refines Gaussian primitives directly from the input samples while preserving a prescribed primitive count and encoded size to achieve a desired level of data reduction. Across structured, unstructured, and particle data, the sample-based formulation improves reconstruction accuracy with measurably fewer primitives in comparison to prior formulations, achieving up to 4.8 dB higher PSNR with an approximate 44x reduction in primitive count. For time-varying data, warm-starting from the previous timestep reduces the optimization required to reach independently trained reconstruction quality. Together, these results demonstrate a unified fixed-budget Gaussian encoding framework for structured, particle, unstructured, and time-varying scientific data with predictable storage, higher reconstruction accuracy, and improved temporal encoding efficiency.
[CV-107] Cascaded Non-Line-of-Sight Imaging
链接: https://arxiv.org/abs/2609.16017
作者: Diego Royo,María Peña,Forrest B. Peterson,Andreas Velten,Julio Marco,Diego Gutierrez
类目: Graphics (cs.GR); Computer Vision and Pattern Recognition (cs.CV)
备注: 18 pages, 21 figures. See this https URL
Abstract:Time-of-flight non-line-of-sight (NLOS) imaging recovers information from hidden objects by analyzing the time of flight of indirect photons scattered on a visible (relay) wall. Most methods make the simplifying assumption that photons travel exclusively three-bounce paths, thus ignoring other useful information encoded in higher-order photons (with, e.g., four- or five-bounce paths). We present a novel cascaded NLOS imaging approach that leverages higher-order information and allows imaging a broader range of single- and multi-corner scenarios. We combine ultra-fast laser scanning with recent time-gated 2D sensor arrays to capture the scene’s impulse response on a visible relay wall. From the captured impulse response, our method computes an analogous virtual impulse response at any other hidden wall. This effectively allows us to concatenate a second, virtual NLOS imaging system that leverages higher-order illumination. We validate our cascaded imaging method both in simulation and with a real prototype, demonstrating NLOS imaging with fourth- and fifth-bounce illumination of objects in challenging orientations and hidden around two corners. We also analyze how wave-based NLOS imaging interacts with rough hidden walls, which explains and helps overcome existing visibility limitations. We further illustrate how to image hidden objects from different perspectives, thus observing previously unseen features, by relying on multiple hidden walls.
[CV-108] MechReason : Benchmarking Multi-Image Multi-Hop Reasoning in Mechanical Engineering
链接: https://arxiv.org/abs/2609.16012
作者: Tengyue Wang,Kang An,Chenxu Du,Zhongyu Yang,Yuanchi Zhu,Xinqi Yang,Hebao Zhu,Ziliang Wang,FaQiang Qian,Yunli Yang,Qibing Ren
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Despite significant progress in general visual question answering and cross-modal understanding, multimodal large language models still face a pronounced gap in evaluation for complex reasoning within the mechanical engineering domain. Existing benchmarks predominantly focus on rudimentary tasks such as drawing recognition, CAD interpretation, or single-chart querying, falling short of assessing whether models can integrate multiple images, textual conditions, physical principles, and engineering constraints to perform multi-step reasoning when confronted with authentic, intricate mechanical problems. To address this, we introduce MechReason, a benchmark derived from real mechanical engineering papers, comprising 12k question-answer pairs with explicit reasoning-chain annotations and 21k visual materials spanning nine evidence types, including statistical charts, parameter tables, engineering drawings, microscopic images, simulation images, system architectures, real mechanical scene photos, CAD model images and manufacturing flowcharts. MechReason covers eight task types across four reasoning dimensions: explanation, prediction, design, and diagnosis. We devise a four-stage construction pipeline: we first extract core engineering claims and decompose their supporting evidence into premises, reasoning processes, conclusions, and corroborative evidence; we then generate shortcut-preventing questions by masking posterior verification information; finally, we apply multimodal quality validation to ensure task quality and multi-hop nature. Extensive experimental results demonstrate that MechReason is highly challenging, with even the most advanced models achieving only 62.89% accuracy.
[CV-109] EMODY Flow: Emotion-Aware Audio-Driven Full-Body Motion Generation
链接: https://arxiv.org/abs/2609.16011
作者: Harsh Kumar Agarwal,Xavier Alameda-Pineda,Olivier Perrotin
类目: Graphics (cs.GR); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Multimedia (cs.MM); Robotics (cs.RO); Sound (cs.SD); Audio and Speech Processing (eess.AS)
备注:
Abstract:Embodied conversational agents require synchronized full-body motion (body gestures and facial expressions) that aligns with speech and emotional state. Omni-modal large language models excel at multimodal understanding but produce only linguistic outputs, leaving a critical gap in embodied response generation. We identify and address a failure of emotion conditioning: like other conditional generators that under-use weak conditioning signals, a flow-matching model given both a rich audio embedding and a discrete emotion label suppresses the emotion, generating near-identical motion regardless of the specified emotion. We present EMODY Flow, a lightweight (around 35M parameters) flow-matching framework that attaches to a frozen Qwen-3 Omni model and reuses its internal Mimi audio-codecs to condition two parallel DiT generators - one for SMPL-X body pose, one for FLAME facial expressions. A training-time auxiliary emotion classifier restores emotion sensitivity by forcing generated motion to be emotion-identifiable. EMODY Flow sets a new state of the art on BEAT2 gesture quality, with FGD 0.302, Beat Correlation 0.853, and Diversity 24.62 - improving over the best prior results by 26%, 5%, and 62% respectively - and transfers to zero-shot facial animation on TFHP without domain-specific fine-tuning. Beyond these quantitative gains, the classifier yields clearly emotion-separated motion, which we demonstrate qualitatively through a multidimensional-scaling analysis of the generated gestures.
[CV-110] SemanticAdv: Generating Adversarial Examples via Attribute-conditional Image Editing ECCV2020
链接: https://arxiv.org/abs/1906.07927
作者: Haonan Qiu,Chaowei Xiao,Lei Yang,Xinchen Yan,Honglak Lee,Bo Li
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注: To appear at ECCV 2020
Abstract:Deep neural networks (DNNs) have achieved great success in various applications due to their strong expressive power. However, recent studies have shown that DNNs are vulnerable to adversarial examples which are manipulated instances targeting to mislead DNNs to make incorrect predictions. Currently, most such adversarial examples try to guarantee “subtle perturbation” by limiting the L_p norm of the perturbation. In this paper, we aim to explore the impact of semantic manipulation on DNNs predictions by manipulating the semantic attributes of images and generate “unrestricted adversarial examples”. In particular, we propose an algorithm \emphSemanticAdv which leverages disentangled semantic factors to generate adversarial perturbation by altering controlled semantic attributes to fool the learner towards various “adversarial” targets. We conduct extensive experiments to show that the semantic based adversarial examples can not only fool different learning tasks such as face verification and landmark detection, but also achieve high targeted attack success rate against \emphreal-world black-box services such as Azure face verification service based on transferability. To further demonstrate the applicability of \emphSemanticAdv beyond face recognition domain, we also generate semantic perturbations on street-view images. Such adversarial examples with controlled semantic manipulation can shed light on further understanding about vulnerabilities of DNNs as well as potential defensive approaches. Comments: To appear at ECCV 2020 Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV) Cite as: arXiv:1906.07927 [cs.LG] (or arXiv:1906.07927v4 [cs.LG] for this version) https://doi.org/10.48550/arXiv.1906.07927 Focus to learn more arXiv-issued DOI via DataCite
[CV-111] Quantum-Inspired Trainable and Parameter-Efficient Tensor Networks for Image Inpainting ICASSP2027
链接: https://arxiv.org/abs/2609.17298
作者: Shiwen An,Konstantinos Slavakis
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 5 pages, 3 figures, 1 table. Submitted to ICASSP 2027
Abstract:This work introduces quantum-inspired tensor-network circuits as trainable transforms for image inpainting. Among the proposed architectures, the diagonal quantum Fourier transform (QFT) relaxation is invertible with O(N^2 \log N) computational cost for N\times N images, inherently preserving minimum coherence throughout training via its circuit structure and eliminating the need for explicit coherence penalties. Unconstrained gradient-based phase optimization (Riemannian-optimization free) enables efficient learning from randomly sampled training data, allowing the learned transform to generalize to test images observed through fixed sampling masks. Numerical tests show that the learned models outperform fixed transforms and per-image optimization while matching the performance of much larger unitary architectures, yet with far fewer parameters.
[CV-112] Mini-batch Sampling Strategies for Long-Tailed Image Classification: An Empirical Study on CIFAR-100-LT
链接: https://arxiv.org/abs/2609.16365
作者: Siyu Yuan
类目: Machine Learning (stat.ML); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 38 pages, 12 figures, 17 tables. Code and per-run logs: this https URL
Abstract:Real-world datasets often exhibit long-tailed class distributions, where a few head classes contain a large number of training samples while a large number of tail classes have only a few. The composition of each mini-batch, determined by the sampling strategy, governs which classes contribute to the stochastic gradient estimate, and therefore affects convergence behaviour and generalisation across the whole class spectrum. We provide a systematic theoretical and empirical comparison of four mini-batch sampling strategies for long-tailed image classification: uniform instance sampling, class-balanced sampling, square-root sampling, and progressively balanced sampling. We place all four in a unified bias-variance framework describing their effect on gradient estimation, which exposes the tension between unbiased optimisation of the empirical loss and fair representation of rare classes. We then evaluate them under controlled conditions using ResNet-32 on CIFAR-100-LT at three imbalance ratios (rho = 10, 50, 100), with every strategy sharing the same long-tailed subsets and initialisation within a seed. Progressive sampling improves tail-class accuracy by 25% relative to the uniform baseline at rho = 100 (13.5% versus 10.8%), consistently across all three seeds, while its overall accuracy is not distinguishable from that of uniform sampling given the seed-to-seed variation (40.0% versus 39.7%); the tail-class gain, not the overall gain, is the robust effect. At rho = 100, class-balanced sampling degrades accuracy on every class group, including the tail classes it is designed to help, which we attribute to overfitting caused by extreme oversampling of scarce data; at rho = 50 this failure is confined to head and medium classes. These results indicate that when rebalancing is applied during training matters as much as how much rebalancing is applied.
[CV-113] A Sentinel-2 benchmark dataset for deep-learning active-fire segmentation across 25 California wildfires
链接: https://arxiv.org/abs/2609.16199
作者: Shreyan Mitra,Mohammadreza Narimani,Parastoo Farajpoor
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Signal Processing (eess.SP); Geophysics (physics.geo-ph)
备注: 13 pages, 9 figures, 5 tables. Preprint submitted to Elsevier. Data: this https URL Code: this https URL
Abstract:This article describes an open image dataset for developing and evaluating active-fire segmentation methods in satellite imagery. The dataset contains 2,148 image-mask pairs from 25 California wildfires, with acquisitions spanning July 2020 to August 2026. Each image is a 512x512-pixel, three-channel composite derived from Sentinel-2 Level-2A bands B12, B11 and B8A at 20 m spatial sampling. A fixed linear rendering is applied throughout the dataset. Corresponding masks distinguish background, SWIR-rule active fire and invalid observations. The masks were generated from shortwave-infrared brightness and near-infrared contrast, followed by constrained neighborhood growth. The release includes chip-level metadata and an incident-disjoint partition containing 18 training, three validation and four test fires. Among the image pairs, 841 contain active-fire labels; these labels occupy 0.0766% of all grid cells. A mask-blind analyst review covers 233 test chips and provides a separate assessment of the rule-generated labels at chip and connected-component levels. Reference training and evaluation code accompanies the data, including a ResNet-34 U-Net implementation with validation-based checkpoint and threshold selection. The archived images, masks, metadata and review annotations support research on rare-class segmentation, learning from algorithmic labels and transfer across fire incidents. The versioned dataset is deposited on Zenodo, with preparation and reuse software maintained in a public GitHub repository.
[CV-114] Multisource Remote Sensing and Geospatial Analysis of Vineyard Wildfire Impacts and Resilience: The 2019 Kincade Fire
链接: https://arxiv.org/abs/2609.16168
作者: Parastoo Farajpoor,Mahla Ardebili Pour,Mohammad Bagher Ghiasi,Mohammadreza Narimani
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV); Atmospheric and Oceanic Physics (physics.ao-ph); Geophysics (physics.geo-ph); Applications (stat.AP)
备注: 18 pages, 11 figures, 4 tables. Preprint submitted to Frontiers. Data: this https URL Code: this https URL
Abstract:Working agricultural landscapes are often treated as background to wildfire disasters, even though they are managed fuel mosaics, productive assets, and parts of regional infrastructure systems. We examine vineyard wildfire resilience during the electrically initiated 2019 Kincade Fire in Sonoma County, California, using an open, event-anchored geospatial framework spanning 4,581 vineyard fields (8,813.2 ha), wildland vegetation, surveyed structures, roads, overhead smoke, and post-fire greenness. Sentinel-2, OpenET, gridMET, soils, terrain, NOAA smoke polygons, an ignition-date OpenStreetMap network, and three-dimensional data inventories were analyzed at native decision scales. Vineyard pixels showed substantially lower descriptive dNBR than wildland pixels inside the perimeter (means 0.130 and 0.337). This contrast did not identify a universal vineyard firebreak effect: a segment-clustered boundary model gave a small negative contrast at 100 m (tau = -0.0166) but changed across bandwidths, failed slope continuity, disappeared in a 100 m donut, and produced a wrong-signed placebo. A 250 m spatial GAM reversed the unconditional pattern: after conditioning on location, terrain, and water use, vineyard fraction was positively associated with dNBR, while residual Moran’s I remained 0.519. Beyond spectral impact, all mapped vineyards intersected overhead smoke on at least one day (mean 7.78 potential smoke-days per field), 34.2% of road-network nodes were dead ends, and inside-perimeter vineyards showed a larger greenness deficit through 2021 (recovery ratios 0.815 inside and 0.854 outside). Lower immediate spectral impact therefore did not imply complete resilience. The study provides a reproducible urban-rural informatics template separating descriptive contrasts, conditional associations, exposure indicators, and recovery evidence for decisions in working landscapes.
[CV-115] Anatomy-Change-Aware Bidirectional Selective State-Space Memory for Clinically Deployed Thoracic Radiotherapy Auto-Contouring
链接: https://arxiv.org/abs/2609.16036
作者: Galib Ahmed,Istiak Ahmed,Aritra Islam Saswato,Asib Mostakim Fony,Kazi Shahriar Sanjid,Md. Tanzim Hossain,Md. Anwarul Islam,Md. Nishan Khan,Md. Misbah Khan,Labiba Faiza Karim,Jobaer Rahman,S M Hasibul Hoque,Rahnuma Shahrin Rista,Kamruzzaman Rumman,Md Arifur Rahman,Syed Md. Akram Hussain,Mohammad Ashrafuzzaman Khan,M. Monir Uddin
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:We developed DAMM-Net++, a 2.5D architecture for thoracic OAR and target volume segmentation that addresses three persistent challenges in radiotherapy auto-contouring: inter-slice surface incoherence, systematic failure on small low-contrast targets, and the absence of per-case reliability signals. The central component is an anatomy-change-aware bidirectional selective state-space memory that models through-plane anatomical change and selectively propagates context along the axial slice sequence. A boundary-aware decoder sharpens near-surface predictions, and an uncertainty head provides calibrated per-voxel confidence for clinical triage. We evaluated 2,146 patients across four centers, an independent external cohort of 112 patients, and a multicenter reader study involving 17 radiation oncologists on 150 cases. The model achieves a mean Dice of 0.955 and HD95 of 3.78 mm, with the largest gains on low-contrast organs-at-risk (OARs) and target volumes where through-plane context is most critical. The uncertainty head is well-calibrated and supports case-level triage. In the reader study, AI assistance reduced contouring time by 75-80 percent across experience levels and raised junior-reader IoU from 0.861 to 0.925, matching the unedited model. External validation showed a modest internal-to-external drop (less than 5 percent) with calibrated uncertainty transferring without recalibration. The complete deployment pipeline from DICOM ingestion to TPS-compatible RTSTRUCT export has been integrated into the clinical workflow at a partner hospital, where it is used to assist with contouring. These results suggest that anatomically motivated inter-slice memory, paired with uncertainty-guided review, offers a clinically viable path for thoracic auto-contouring.
[CV-116] Automated Distinction of Intimal and Medial Intracranial Arterial Calcification from CT Head MICCAI2026
链接: https://arxiv.org/abs/2609.16035
作者: Benjamin Jin,Maria del C. Valdés Hernández,Richard Bortsov,Joanna M. Wardlaw,Daniel Bos,Grant Mair
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at the Stroke and neurovascular diseases Workshop on Imaging and Treatment CHallenges @ MICCAI 2026
Abstract:Intracranial arterial calcifications (IACs) are a common finding on clinical non-contrast enhanced head CT scans and are associated with neurovascular disease. Calcifications can occur in the intimal or medial layer of the arterial wall, subtypes that differ in aetiology and may have distinct clinical relevance. These subtypes can be visually distinguished by radiologists based on the shape of the calcifications. We investigate three automated approaches for subtype classification of IAC from head CT-derived segmentation masks: (1) an automated adaptation of the established radiological visual score, (2) a sphericity-based method, and (3) a method based on shape embeddings extracted by a medical shape foundation model. All approaches use the same lightweight classification pipeline on top of the features they compute and are evaluated using 5-fold cross-validation. The three methods achieved comparable performance, with the embedding-based approach yielding the best overall results with a weighted F1 (mean \pm SD) of up to 71.5 \pm 3.7 for a single artery and 59.8 \pm 1.7 for the joint artery classification. Performance was largely preserved when using automated instead of manual IAC segmentation masks, and we found the difference in weighted F1 not significant. Our results show that fully automated IAC subtype quantification from head CT is feasible and remains robust to the use of manual and automated IAC segmentation masks. Code at this https URL. Comments: Accepted at the Stroke and neurovascular diseases Workshop on Imaging and Treatment CHallenges @ MICCAI 2026 Subjects: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2609.16035 [eess.IV] (or arXiv:2609.16035v1 [eess.IV] for this version) https://doi.org/10.48550/arXiv.2609.16035 Focus to learn more arXiv-issued DOI via DataCite
[CV-117] LM-PCVMNet: Pediatric Cervical Vertebral Maturation Analysis with Deep Fusion of Landmarks and Metadata
链接: https://arxiv.org/abs/2609.16033
作者: Peng Wang,Wanzhen Song,Anli Wang,Xueshuo Xie,Xiaohang Guan,Tao Li
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注: 12 pages accepted by Information Fusion
Abstract:Cervical vertebral maturation (CVM) assessment plays a pivotal role in orthodontic diagnosis and determining the optimal timing of treatment, especially for pediatric patients. In this paper, we propose LM-PCVMNet, a novel deep learning framework for automatic pediatric CVM staging. Specifically, our method integrates vertebral anatomical landmark information, heatmap-guided feature modulation, and metadata-informed similarity modeling into a unified learning framework. We introduce a heatmap-guided feature modulation module that enhances feature extraction by leveraging landmark-centered heatmaps to highlight morphologically relevant vertebral regions. A vertebral landmark-prompting block is designed to incorporate anatomical geometry into the representation learning process. Furthermore, we develop a learnable metadata supervised contrastive loss that adaptively modulates positive-pair similarity based on metadata similarity, enabling the model to learn more biologically consistent and discriminative features. To facilitate further research in pediatric orthodontic treatment, we additionally release PCVM+. It contains 1800 lateral cephalometric radiographs from real-world patients aged 3-15 years, with expert-annotated CVM stages, 13 vertebral anatomical landmarks, and corresponding metadata. We perform comprehensive experiments on two datasets, and the results show that our method achieves state-of-the-art performance, effectively improving landmark localization and classification accuracy over existing models. Code and dataset will be available at this http URL.
[CV-118] Conditioning noise is a free regularizer for LoRA fine-tuning: no pathology encoder required for diffusion-based artifact detection in histopathology
链接: https://arxiv.org/abs/2609.16032
作者: Konstantinos Moutselos,Ilias Maglogiannis
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注: 23 pages, 3 figures. Code and laboratory record: this https URL (doi: https://doi.org/10.5281/zenodo.22702198 ). Data: doi: https://doi.org/10.5281/zenodo.22702800 . Companion study: arXiv:2608.30835
Abstract:Diffusion-based artifact detectors score whole-slide image patches by reconstruction error under a model fine-tuned on clean tissue. We show that conditioning this fine-tuning on random Gaussian embeddings – resampled at every step from approx. 200 KB of precomputed embedding statistics, with no encoder, no cache, and no change to inference – consistently widens the clean/artifact separation. A four-step ablation chain shows the benefit requires neither content (shuffled real embeddings), provenance (synthetic Gaussians), a tuned intensity (flat across an 8x variance range), nor per-patch identity (fresh per-step noise); a LoRA-dropout control shows the conditioning pathway specifically, not generic weight perturbation, carries the effect. Patch-level gains of +0.25-0.48 Cohen’s d replicate across nine trainings; honest leave-one-slide-out evaluation clears a pre-registered bar in 2/2 seeds; and two pre-registered external endpoints on a 281-case set confirm pooled Delta F1 = +0.0073 (95% CI) and +0.0129 (97.5% CI, two-look corrected). We release the full evaluation protocol, including measured seed noise and selection-optimism pricing.
[CV-119] A deep dictionary network-based foundation model for ultra-low-dose CT denoising
链接: https://arxiv.org/abs/2609.16031
作者: Baoshun Shi,Shuangyi Yang,Ke Jiang,Bin Zhu,Zhanli Hu,Huazhu Fu
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Ultra-low-dose computed tomography (ULDCT) reduces radiation exposure but suffers from severe noise that degrades diagnostic image quality. Existing deep learning-based denoising methods are typically trained in an organ-specific fashion, resulting in limited generalization across heterogeneous multi?organ imaging scenarios. Foundation models present a promising all-in-one paradigm for unified multi-organ denoising. However, their architectures suffer from poor interpretability and rely on heuristic training strategies. To address these limitations, we propose an architecture?interpretable foundation model based on the deep dictionary network (DDN) for unified multi-organ ULDCT denoising. Inspired by multilayer sparse representation theory, DDN cascades convolutional sparse coding layers with iterative soft-thresholding, providing inherent architectural interpretability. Furthermore, a dynamic dictionary module and a threshold generation module are embedded within each layer to enhance representation ability. We conduct DDN pre-training on more than one million multi-organ normal-dose CT images by recovering clean images from Gaussian-noised inputs. Sparse regularization is additionally imposed on latent feature representations, guiding the network to learn compact and noise-robust priors. The complete architecture is jointly fine-tuned on multi-organ ULDCT datasets, enabling a single unified model to perform denoising across diverse anatomical regions. Extensive experiments validate that our proposed method achieves state-of-the?art performance and consistently surpasses competing ULDCT methods across all mul
人工智能
[AI-0] ENCP: Episode-Normalized Conformal Prediction for Vision-and-Language Navigation
链接: https://arxiv.org/abs/2609.17499
作者: Vicky Feliren,A. Taufiq Asyhari,Muhamad Risqi U. Saputra
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Robotics (cs.RO)
备注: 8 pages, 5 figures
Abstract:Uncertainty estimation for Vision-Language-Navigation (VLN) models is a critical task since it can help identify ambiguous and unreliable predictions, enabling agents to make safer navigation decisions. As one of the most advanced uncertainty estimation frameworks, conformal prediction (CP) offers a promising approach for uncertainty estimation in VLN. However, given that VLN agent requires a sequence of steps, standard calibration in conformal prediction fails to provide coverage guarantee it promises over a dependent, variable-length VLN episode. To this end, we propose Episode-Normalized Conformal Prediction (ENCP), which rescales a nonconformity score by the policy’s residual confidence and calibrates one maximum score per episode. Under exchangeable calibration and test episodes, this construction covers the ground truth at every step with probability at least 1 - \alpha , while allowing dependence among steps within an episode. Across four VLN policies and three nonconformity scores on R2R and REVERIE dataset, ENCP meets all reported empirical step-coverage targets on the seen-to-unseen evaluation. These results demonstrate that ENCP can provide model-agnostic uncertainty estimates, which might be useful for determining when a VLN agent should defer to a more capable predictor, including human assistance.
[AI-1] LimiX-2: A Contextual Mechanism Network Towards General Structured-Data Intelligence
链接: https://arxiv.org/abs/2609.17488
作者: Xingxuan Zhang,Gang Ren,Hao Yuan,Hao Zou,Hongze Tan,Hui Wang,Jianhao Song,Jiansheng Li,Jiayao Zhang,Jinghan Zhang,Kaifang Li,Lang Mo,Li Mao,Mingchao Hao,Nuo Xu,Rui Ding,Ruiji Zhang,Shuyang Li,Siyu Mei,Tianyang Zhang,Weiyang Mu,Yancheng Dong,Yongxian Wei,Yuan Xue,Yuanrui Wang,Yue He,Zijia Yang,Ziyun Li,Dongzhe Li,Fuqiang Wang,Jiandong Liu,Jiawei Chen,Jiaxin Du,Kaijie Cheng,Kehan Li,Lei Sun,Linjun Zhou,Ningbo Dai,Qi Wang,Renzhe Xu,Shaoxing Du,Shumeng Yang,Wang Lu,Wenjing Chu,Xiannan Huang,Xiaoyu Lin,Xing Ai,Xinyan Han,Xuanyue Li,Xuanyue Su,Xukun Zhang,Yan Lu,Yaxin Zhang,Yi Qin,Yifei Huang,Yihan Xu,Yongle Lv,Yuanyuan Jiang,Yushan Han,Peng Cui
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:We introduce LimiX-2, a new model in the LimiX family, developed through model and data scaling guided by our previously established scaling laws. LimiX-2 adopts the Contextual Mechanism Networks (CMNs) paradigm and is pretrained with Context-Conditional Masked Modeling (CCMM). CMNs shifts the organizing principle of in-context learning from target-centric prediction to mechanism-oriented joint modeling. Rather than centering the network on the p(y \mid x, D_\mathrmcontext) objective of conventional tabular PFNs, it is designed around learning p(x, y \mid D_\mathrmcontext) , a context-dependent representation of the joint structure underlying data generation. Pretraining uses synthetic datasets generated by structural causal models (SCMs) spanning diverse graph structures, functional mechanisms, and observation processes. Evaluations on TabArena, TALENT, and BCCO show that LimiX-2 outperforms current dataset-specific models and tabular foundation models. Beyond predictive performance, the CMN paradigm also promotes causal awareness in LimiX-2: its feature attention encodes direct causal relationships, enabling accurate causal skeleton recovery.
[AI-2] JustFit: 200K-Token LLM Serving on a 24 GiB Laptop with Just-in-Time State Management
链接: https://arxiv.org/abs/2609.17475
作者: Yuhua Chen
类目: Artificial Intelligence (cs.AI); Performance (cs.PF)
备注: 13 pages, 4 figures, 9 tables
Abstract:Capable open-weight models make local coding and reasoning attractive, but their context and execution state strain laptop memory. We present JustFit, an MLX-based inference runtime that combines KVExec for compressed KV execution, PhaseSwap for component residency, and StateTrans for state-preserving serving transitions. These mechanisms fuse reconstruction and coordinate just-in-time materialization and release, independently of model-weight quantization. In full-execution capacity tests on a 24 GiB M4 Pro MacBook running Qwen3.8-27B MXFP4, three independent runs complete 196,608 input and 16,384 output tokens, increasing completed single-request context from the mlx-vlm baseline’s 30,720 positions to 212,992 (6.93x); a separate two-request run retains 229,376 positions in aggregate. In separate performance tests, a 32K-input, 64-output probe reaches 19.11 tokens/s, and a repeated 32K+6K workload has a median peak process footprint of 16,374 MiB. The integrated runtime answers 29 of 30 AIME 2026 problems correctly, showing how compact state and lifetime-aware execution expand local serving capacity while supporting extended generated reasoning.
[AI-3] Coupled Calibration and Learning: Mitigating Teacher Bias in LLM Distillation without Target-Domain Reward Feedback
链接: https://arxiv.org/abs/2609.17474
作者: Haichen Hu,Yuheng Zhang,David Simchi-Levi
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Statistics Theory (math.ST); Machine Learning (stat.ML)
备注:
Abstract:Large language model (LLM) distillation aims to transfer the capabilities of a powerful teacher to a smaller student. Direct imitation, however, can also transfer the teacher’s systematic bias and errors. This challenge is particularly pronounced under covariate shift, when the teacher’s reliability on target questions is uncertain and target-domain reward feedback is unavailable. We propose Coupled Calibration and Learning (CCL), an LLM distillation algorithm that couples teacher calibration with student updates through token-level branching, using reward feedback only on source questions. Each iteration calibrates the teacher using source feedback and then uses the calibrated teacher to train the student on target questions. The updated student, in turn, informs subsequent calibration. In an autoregressive policy framework, we prove that the output student’s expected average Kullback-Leibler divergence to the oracle student converges to zero at a polynomial rate in the number of iterations. The oracle maximizes the true reference-regularized target reward within the student class, which need not represent the unrestricted optimal policy. Our analysis quantifies the progress of projected student gradient updates while controlling the error in teacher calibration. We further establish a separation from regularized direct matching: its error relative to the oracle student can remain bounded away from zero even when the teacher achieves higher regularized target reward than every student policy. These results demonstrate that LLM distillation can overcome persistent teacher bias and recover the optimal student through coupled calibration and learning, without target-domain reward feedback.
[AI-4] Learning-Guided Planning in Large Dynamic Action Spaces: Budgeted Tree Search for One-to-Many Mobile Charging
链接: https://arxiv.org/abs/2609.17429
作者: Liang-Ching Tao,Pi-Chung Wang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 15 pages, 7 figures. Learning-guided planning, budgeted tree search, PUCT, and sequential decision-making in large dynamic action spaces
Abstract:Many learned sequential decision systems map the current state directly to an action. That shortcut becomes brittle when candidate actions are numerous, geometrically structured, and rebuilt with the state. One-to-many mobile charging makes this setting concrete: with N=250 sensors, the initial state induces about 1,125 candidate charging-stop actions; each chosen stop simultaneously serves its in-range sensors, and the action universe changes as sensors die. LP-BTS is a learning-guided planning architecture: a graph proposal policy concentrates a small candidate support, a learned value critic evaluates leaves, and edge-budgeted PUCT compares short simulated futures before committing an action. Because the policy scores this set without a fixed output head, a single frozen checkpoint covers every evaluated setting, spanning action universes from 736 to 2,813 stops. Matched ablations reveal complementary effects: uniform sampling costs 8.8 survival percentage points, while, with targeted support fixed, PUCT jointly retains 1.4 points (about 3.5 of 250 sensors) and direct policy selection travels 23% farther. On a prospectively specified, sealed 30-scenario confirmatory bank evaluated once, LP-BTS attains the highest observed survival (0.4545) and alive-AUC (0.8031). Its estimated survival advantage over the strongest domain-engineered comparator is +0.0066 (95% CI [-0.0037, +0.0184]), an unresolved difference, while it exceeds a deadline heuristic and two source-derived direct-policy reconstructions on every paired scenario. Both learned rows are trained, source-derived reconstructions of variants reported by Gong et al. In this setting, the results provide controlled evidence about learning-guided planning in a large, dynamic action space.
[AI-5] alking Head Synthesis with Facial Landmark Guidance via 3D Gaussian Splatting
链接: https://arxiv.org/abs/2609.17422
作者: Ziheng Yang,Yinfeng Yu,Yongming Li
类目: Artificial Intelligence (cs.AI); Signal Processing (eess.SP)
备注: Main paper (6 pages). Accepted for publication by IEEE International Conference on Systems, Man, and Cybernetics 2026 (IEEE SMC 2026)
Abstract:Audio-driven digital human generation plays an important role in virtual communication, immersive interaction, and media production. With the development of Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS), recent talking-head systems have obtained more faithful 3D facial geometry and appearance modeling. A remaining difficulty is that speech features mainly describe temporal acoustic patterns rather than explicit facial layouts. As a result, directly driving 3D facial deformation with audio may produce inaccurate mouth motion, weak expression details, and local artifacts. To address this issue, we propose a facial-keypoint-guided spatial enhancement module. The predicted landmarks provide structural cues for selecting and enriching spatial points around expression-sensitive facial regions. We further introduce a global landmark compensation mechanism, where the full set of keypoints is encoded into a conditioning vector to refine 3DGS attributes. This compensation supplies whole-face structural information to the underlying shape representation. Experiments under self-driven and cross-driven settings show that the proposed method improves visual quality, facial realism, and lip synchronization.
[AI-6] ransformer-Based Token Fusion and Dynamic Graph Planning for Audio-Visual Navigation
链接: https://arxiv.org/abs/2609.17421
作者: Shaohang Wu,Yinfeng Yu
类目: Artificial Intelligence (cs.AI); Signal Processing (eess.SP)
备注: Main paper (6 pages). Accepted for publication by IEEE International Conference on Systems, Man, and Cybernetics 2026 (IEEE SMC 2026)
Abstract:Audio-Visual Navigation (AVN) requires an agent to localize and navigate toward a continuously vocalizing target relying solely on visual observations and acoustic cues. Currently, systems lack the ability to adaptively correct and replan when faced with incomplete or misleading visual perception. Furthermore, relying on physical collisions to compensate for missing visual information results in inefficient and unsafe navigation, whereas existing methods are overly dependent on passive visual perception. To address these issues, we propose the Transformer-based Token Fusion and Dynamic Graph Planning (TDGP) model, which incorporates high-level perception layers and leverages the Transformer model to fuse multimodal cues for precise local planning. Next, a low-level planning layer is designed that uses physical collision penalties to remove edges that collide with the map in real time and apply corresponding penalties, forcing the agent to automatically re-plan to compensate for the lack of visual information. Experiments show that our TDGP model outperforms baseline models on the Replica and Matterport3D (MP3D) datasets, and that the model’s sound enhancement strategy significantly improves generalization in unheard acoustic scenarios.
[AI-7] CTAN: Cycle-Temporal Attention Network for Embodied Audio-Visual Navigation
链接: https://arxiv.org/abs/2609.17420
作者: Teng Liu,Yinfeng Yu
类目: Multimedia (cs.MM); Artificial Intelligence (cs.AI); Sound (cs.SD); Signal Processing (eess.SP)
备注: Main paper (6 pages). Accepted for publication by IEEE International Conference on Systems, Man, and Cybernetics 2026 (IEEE SMC 2026)
Abstract:Audio-visual embodied navigation equips robots with the capability to infer the locations of sound sources by integrating visual inputs and acoustic information (e.g., depth observations and binaural audio cues). The core challenge lies in establishing effective semantic interactions across heterogeneous modalities (which exhibit distinct feature distributions). Existing feature fusion strategies, however, often rely on simple multimodal aggregation and therefore fail to capture the underlying geometric and semantic relationships, leading to information degradation in complex environments. To overcome these limitations, this work presents the Cycle-Temporal Attention Network (CTAN), a framework designed for active semantic-enhanced fusion (rather than straightforward multimodal combination). Specifically, the proposed Audio-Visual Reconstruction Cross-Attention (AVRCA) module employs a bidirectional cycle-consistency constraint (between visual and acoustic representations) to reinforce the spatial semantic attributes of both modalities, thereby facilitating more robust cross-modal interaction. Additionally, we design a Temporal Cross-Modal Memory (TCMM) mechanism to dynamically integrate real-time enhanced multimodal features with historical context, reducing performance drops caused by auditory dead zones. Experimental results obtained on the Replica and Matterport3D benchmarks indicate that the proposed approach achieves superior performance over previous audio-visual navigation methods in terms of success rate (SR), success weighted by path length (SPL), and scene navigation accuracy (SNA).
[AI-8] World Model Science: Self-Organized Criticality Weak Chaos and Metastable Belief Dynamics in Long-Horizon LLM Agents
链接: https://arxiv.org/abs/2609.17419
作者: Xinyuan Song,Zekun Cai
类目: Artificial Intelligence (cs.AI)
备注: Under Review
Abstract:Long-horizon LLM agents must maintain task state across extended sequences of observations, actions, tool calls, and intermediate beliefs. We study these trajectories through three dynamical views: self-organized criticality, weak chaos, and metastable belief dynamics. Our framework aligns agent-implied states with benchmark-grounded states and measures stress accumulation, error avalanches, temporal dependence, local–global mismatch, bounded divergence, belief-basin transitions, and finite-size scaling under explicit null models. Across 22 experiments spanning controlled puzzles, tool use, embodied tasks, multi-hop retrieval, general-assistant reasoning, and Game of Life, we find that locally valid actions can persist after global state fidelity fails, stress can trigger abrupt collapse, error sequences exhibit long memory, dependency depth changes the propagation regime, and larger horizons support larger avalanches. At the same time, divergence remains bounded, belief states show metastable rather than fully chaotic behavior, and stronger claims of universal power laws, critical points, or shared intervention optima are not supported. These results suggest a science of agent world models based on trajectory-level dynamical diagnostics rather than terminal reward alone.
[AI-9] Never Stop Thinking: Continuous-Time Language Agents
链接: https://arxiv.org/abs/2609.17416
作者: Bojie Li,Noah Shi
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Voice agents built on LLMs follow a rigid listen-think-speak loop that inserts seconds of dead air before every reply. We show that continuous-time cognition (thinking while listening and thinking while speaking) emerges from an unmodified text model under a lightweight interrupt-and-resume orchestrator, cutting live-pipeline latency by 19% overall and by half in the regime the mechanism targets. To measure whether continuous-time thinking improves what agents accomplish, we introduce ReactiveBench: 120 interactive scenarios scored against pre-registered binary requirements, plus a verifiable streaming track scored by exact correctness. ReactiveBench exposes a pitfall with broad consequences: LLM judges reward visible reasoning; a large judged “advantage” of continuous-time thinking reverses sign under an independent judge, and judge-trained models objectively complete fewer requirements when they think. A five-stage training study then locates the right signal at three levels. Its source: verifiable objectives turn thinking from harmful to helpful. Its structure: whatever a uniform reward omits, optimization trades away; brevity everywhere erodes multi-hop tool chaining. Its optimizer: preference optimization can only trade conflicting sub-goals against each other, while on-policy RL over a type-shaped reward improves every correctness axis at once, raising streaming completion from 48% to 73+/-5% across seeds and replicating at larger scale and on a second model. Orchestration makes continuous-time interaction possible; a verifiable signal, correctly sourced, shaped, and optimized, makes it good.
[AI-10] Coding Agents Have Converged: Why the SWE-bench Leaderboard Can No Longer Order Its Top Entries and What to Measure Instead
链接: https://arxiv.org/abs/2609.17394
作者: Fengshuo Liu,Ying Liu,Ruize Sun,Lie Luo,Siyuan Guo
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: Accepted at ADMA 2026 (International Conference on Advanced Data Mining and Applications), Special Session on Responsible Data Intelligence. Camera-ready version, 15 pages, 4 figures
Abstract:Small differences on coding-agent leaderboards are often read as an ordering of systems. We audit whether the published verdicts support this reading, using 254 SWE-bench submissions across four splits without running models. On Verified, the leading two entries each resolve 396 of 500 instances. The top ten share 285 successes and 51 failures, leaving 164 instances that distinguish their outcomes. Frontier solution sets have median nesting 0.935 against a score-implied baseline of 0.774, indicating strongly shared successes. Scores also depend on the evaluated model-scaffold pair: observed within-model scaffold ranges reach 29.8 percentage points, compared with the 8.8-point spread of the top thirty. Six of nine cell-mean interaction tests remain significant after Holm correction, although this observational design does not identify causal scaffold effects. Exact paired McNemar tests separate none of the 29 adjacent Verified top-thirty pairs at alpha=0.05, while the larger Test split separates 14 of 23. A stated leader-based rule yields three descriptive tiers, or two after Holm correction; non-rejection does not establish equivalence. We release the partition and a five-step audit protocol that profiles shared outcomes, tests paired differences, reports grouping sensitivity, and estimates the instance budget needed for resolution. The results motivate reporting comparison-set-specific resolution and model-scaffold provenance instead of interpreting small aggregate gaps as established rank differences.
[AI-11] FlashVector: Agent for Hierarchical Model Serving Stack Optimization
链接: https://arxiv.org/abs/2609.17391
作者: Qi Wu,Lohan Lemire,Kai Meng,Zhongmou Cai,Raphael Bargues,Petr Zhitnikov,Zeyuan Cao,Yao Wang,Shujun Bian,Wei Chen,Sean Sheng
类目: Artificial Intelligence (cs.AI); Performance (cs.PF)
备注:
Abstract:Model serving is one of the largest cost drivers in production recommender systems. Maximizing its throughput requires navigating a deeply layered hierarchy: GPU kernels, the ML framework computation graph, the model server, and on-demand feature processing – each demanding specialized domain expertise. Such cross-layer expertise is inherently difficult to acquire, and does not scale with a workload that continuously grows and evolves, leaving significant cost efficiency gains unrealized. While recent AI agents have demonstrated human expert level efficiency in standalone GPU kernel optimization, automated tuning and optimization for the rest of the serving stack remain largely unexplored. We present FlashVector, an agentic system that optimizes performance across all layers of the model serving stack. The key contribution is an extensible framework to generalize the single kernel optimization agent paradigm to heterogeneous technical stacks, and to deliver performance improvements holistically. After deployment in Unity’s Vector advertising platform, FlashVector achieved up to 2x throughput increase and up to 1.98x latency speedup on model server, and up to 1.6x throughput increase on feature store. These optimizations were discovered not only at the GPU kernel and computation graph levels, but also across the other components of the model serving stack, such as the model server (NVIDIA Triton’s C++ codebase) and the on-demand feature transformation service (Python codebase), demonstrating the extensibility of the framework to more complex system architectures.
[AI-12] Self-Emergence Agent Architecture:Behavior-Inertia HMM Reflexive Metacognitionand Social-Contrastive Self-Modeling
链接: https://arxiv.org/abs/2609.17331
作者: Xiaoyang Liu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large language model (LLM) agents exhibit strong language-generation and problem-solving capabilities, yet suffer from three structural limitations: personality drift, non-evolutionary reflection, and the absence of a self-other boundary. Existing generative-agent simulations rely on static memory and fixed prompts, maintaining neither behavioral inertia nor endogenous self-evolution. We propose the Self-Emergence Agent Architecture (SEAA), which integrates three components: (i) a Hidden Markov Model (HMM) that encodes long-term behavioral and cognitive inertia as an editable state-transition matrix; (ii) a Reflexion-style verbal metacognition loop whose output updates the HMM parameters themselves, rather than merely being stored as text; and (iii) a multi-agent social environment in which initially identical agents continuously compare their behavior with others’. The three components form a closed loop: social action \to feedback \to self-reflection \to inertia update \to differentiated action. We state three falsifiable hypotheses and provide a reproducible experimental protocol with operational metrics. A language-model-free prototype shows the loop spontaneously breaks symmetry: initially identical agents consolidate distinct, stable personalities whereas matched controls do not. Experiments with a hosted LLM surface these differences as distinct first-person self-narratives, and a five-agent deliberation spontaneously develops social structure—a consensus hub and a unanimously rejected outlier—absent in the control. Following an epistemologically agnostic stance inspired by Zhuangzi, SEAA studies only observable behavioral emergence and makes no claim about subjective qualia. This work contributes a unified framework, a concrete architecture with pseudocode, mechanistic evidence, and a microscope-style sandbox for studying artificial-self emergence.
[AI-13] From Transient Prompts to Persistent Control: Scientific Poster Generation via Recursive Semantic-Geometric Contracts
链接: https://arxiv.org/abs/2609.17326
作者: Runze Li,Yukun Zhao,Can Xu,Yucheng Shen,Shuaiqiang Wang,Jianmin Wu,Lingyong Yan,Dawei Yin
类目: Artificial Intelligence (cs.AI)
备注: 7 pages, 3 figures, 5 tables
Abstract:Scientific poster generation distills a multimodal paper into a single-page visual artifact, forcing strict trade-offs between informational coverage and readability under a fixed spatial budget. Existing methods pass plans as transient prompts and validate individual stages in isolation. This strategy causes requirements to drift across content and layout modules, and previous checks to be silently invalidated. We introduce PosterVisor, a control framework that shifts poster generation from transient prompts to persistent control. An Orchestrator grounds rubrics in the paper and visual assets, compiling them into a Semantic-Geometric Contract (SGC) that binds claims and sources to required visuals, budgets, and spatial commitments. Only fully instantiated records become executable assertions; other usable requirements remain soft guidance. Recursive Contract Enforcement (RCE) dynamically triggers checks across stages as evidence emerges. Crucially, during repairs, RCE rechecks affected checkpoint states, preventing repair-induced regressions from propagating silently. We instantiate PosterVisor in HTML/CSS and editable PPTX generators. On the 100-paper Paper2Poster benchmark, PosterVisor-PPT improves observed mean poster-grounded QA accuracy over PosterGen (64.47% vs. 58.53%) and is preferred by human judges in 72.5% of non-tied pairwise comparisons (95% CI, 61.6-83.4%). A secondary 30-paper study also yields higher VLM Overall and PaperQuiz means. These results support rubric-compiled contracts and stage-conditioned enforcement for controllable poster synthesis.
[AI-14] Intrinsic Motivation in Reinforcement Learning: A Research Agenda for Adaptive Self-Organisation
链接: https://arxiv.org/abs/2609.17325
作者: Anatoly Belikov
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Biological cells can be viewed as individual, interacting agents whose collective dynamics give rise to adaptive behaviour at multiple levels of organisation, from individual cells through tissues to whole multicellular organisms. In this perspective and tutorial article we discuss whether intrinsic rewards in artificial neural systems can support adaptation, functional specialisation and higher-level self-organisation without a shared external objective. We review empowerment, curiosity, learning progress, information gain, unsupervised skill discovery, mutual information estimation and the use of world models for intrinsic reward computation. Particular attention is given to failure modes showing when such objectives do not produce sustained exploration or increasingly complex behaviour. We argue that more capable systems may require complementary objectives, communication, memory, learning at multiple temporal scales and environmental constraints. Based on this perspective, we outline three experimental directions. These include a resource-constrained environment in which otherwise stable behavioural attractors become unsustainable, allowing us to test whether environmental constraints can mitigate characteristic failure modes of intrinsic objectives. The network of recurrent agents with per-agent intrinsic rewards, and a hierarchical world-model agent in which exploratory motor competence develops before goal-directed behaviour. These experiments are intended to test whether intrinsic learning can lead to adaptive organisation at progressively higher levels.
[AI-15] Extracting ontology-compliant knowledge from scientific text describing irradiated materials using large language models
链接: https://arxiv.org/abs/2609.17291
作者: Marco Luca Sbodio,Marcos Martínez Galindo,Vanessa Lopez,Blanca Biel,Pablo Canca,Pedro Delgado,Jesús I. Mendieta-Moreno,Raphael Tack,Maria J. Caturla
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:The quest for new materials increasingly relies on predictive models and comprehensive simulations that span scales from atomic to macroscopic levels. However, essential data necessary for these models and simulations are often embedded in scientific literature as unstructured text, limiting reusability and posing challenges for researchers seeking to leverage existing knowledge effectively. While extracting structured data from unstructured text using large language models is gaining popularity, traditional methods typically generate key-value pairs data with straightforward schemas. In contrast, we introduce eolas, a modular pipeline that uses large language models to automatically transform scientific documents into knowledge graphs aligned with a specified ontology. We demonstrate eolas effectiveness in extracting useful information for scientists studying materials designed to endure the extreme temperatures and radiation levels found in fusion reactors. While a human expert might spend between thirty to ninety minutes extracting relevant data from an article, eolas can generate high-quality knowledge graphs in just a few minutes. These are presented in a tabular format with faceted navigation for easy human validation. Additionally, we introduce the first benchmark dataset designed to assess large language models capabilities in constructing knowledge graphs within the domain of irradiated materials. The analysis of 168 experiments using our dataset, various large language models and prompting techniques provides key insights that we summarize into practical guidelines for effectively extracting knowledge graphs aligned with an input ontology.
[AI-16] After the Party: Governing What a Viral Agent -Skill Ecosystem Left Behind
链接: https://arxiv.org/abs/2609.17274
作者: Yunpeng Xiong,Ting Zhang
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注: To appear in IEEE Digital Library as the 33rd Asia-Pacific Software Engineering Conference (APSEC 2026) conference proceedings. Accepted version, not camera ready version
Abstract:AI agents increasingly act through agent skills, i.e., natural-language instructions, that direct a host agent toward shell, network, credential, file, and process actions, and public registries distribute them at scale. In the first half of 2026, the OpenClaw AI agent went viral, and its public skill registry boomed: the observable stock nearly doubled in 91 days, and a majority of the listings visible in June were created in just two months. By the end of our study window, the wave had crested, and monthly listing creation and core-repository activity were falling from their spring peaks. This paper measures what the boom left behind, drawing on the OpenClaw Git history, its GitHub issues and pull requests, and three ClawHub registry snapshots. Attention is concentrated: the top 10% of skills received 46.93% of all downloads. No simple skill features (like size or download counts) remained a stable predictor of continued listing once creation cohort and skill age were controlled. Human scrutiny did not stay: 77.86% have zero stars and zero comments, while 85.06% of the readable skills carry privilege evidence. And automated cleanup is not ready: the three security scanners disagreed on 23,702 of the 61,990 skills they all cover. After human adjudication, weighted scanner sensitivity against the reference standard ranged from 21.67% to 61.06%. Governing fast-growing agent-skill registries cannot rely on simple metadata or single scanner scores; it requires robust, transparent measurement and independent validation.
[AI-17] Grounding SWE-Agent Decisions in Architecture-0 Design: Navigating Unknown Unknowns through Physical Mapping
链接: https://arxiv.org/abs/2609.17221
作者: Zhongkai Wang,Yan Liu
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: 51 pages, 13 figures, 18 tables. Preprint of a manuscript under review at ACM TOSEM
Abstract:Autonomous Software Engineering Agents (SWE-Agents) excel in deterministic coding tasks but struggle with Architecture 0, the nascent system design phase plagued by implicit engineering constraints, or Unknown Unknowns (UUs) that are rarely stated explicitly. To investigate how agents navigate UUs, we explore a progressive trajectory across pure-text self-play, tool-augmented feedback, and external physical mapping. Our empirical analysis reveals a cascading chain of failures. Pure-text reasoning inevitably devolves into polite consensus or plausible yet physically impossible fabrications. Attempting to bridge this gap via an early-stage execution sandbox unexpectedly triggers Specification Gaming: agents exploit their autonomy over validation scripts to bypass physical constraints, achieving superficial success without resolving core architectural flaws. To resolve this self-validation trap, we propose the Physical Mapping Guard (PMG). Grounded in the software engineering principle of Separation of Concerns, PMG revokes verification authority from the agent, forcing semantic intents to be evaluated by an external, deterministic Semantic-to-Physical (S2P) mapping engine. Extensive evaluations demonstrate that PMG completely eradicates physical-layer and validation-layer gaming. By precisely isolating residual failures to semantic reinterpretations and auditor overreach, PMG marks a critical step toward genuine affordance grounding in automated architectural design.
[AI-18] FluxVLA Engine: A One-Stop VLA Engineering Platform for Embodied Intelligence
链接: https://arxiv.org/abs/2609.17210
作者: Yinhao Li,Weixin Mao,Zihan Lan,Jikun Rong,Qirui Hu,Yiming Zhang,Weipeng Deng,Bowen Shen,Minzhao Zhu,Yiming Mao,Yan Yang,Chenguang Cui,Hongyuan Chen,Xu Huang,Zheyi Zhao,Pinxi Shen,Bozhen He,Zhen Fu,Yifan Wang,Zexin Zhang,Ang Gao,Haoyu Chen,Chengqi Shi,Hua Chen
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Vision-language-action (VLA) models, world-action models (WAMs), and offline reinforcement learning methods are rapidly expanding the design space of embodied policies, yet turning these algorithms into reliable robot systems remains constrained by fragmented data formats, training stacks, evaluation protocols, inference runtimes, and embodiment-specific interfaces. We present \mathrmFluxVLA Engine, an open, configuration-driven platform that turns heterogeneous embodied-policy components into a reproducible data-to-deployment workflow. Rather than introducing another policy model, \mathrmFluxVLA standardizes interfaces for datasets, visual-language and world models, action heads, reward- or advantage-weighted learning, distributed training, simulation evaluation, optimized inference, and robot operators. The engine further integrates compositional dual-arm simulation, scalable automatic data generation, and model-decoupled human-in-the-loop rollout, takeover, correction collection, and reward annotation. For responsive physical execution, it combines Real-Time Chunking (RTC) with accelerated inference backends, lightweight remote GPU serving, and configurable trajectory post-processing. Together, these capabilities connect offline learning, simulation validation, online correction, and real-robot execution through shared and auditable contracts. \mathrmFluxVLA therefore targets the engineering bottlenecks separating promising embodied-learning algorithms from reproducible evaluation and dependable deployment. Code is available at this https URL
[AI-19] End-to-End Latency-Minimizing and Load-Balanced Request Scheduling for Edge LLM Inference in Agent ic AI Services
链接: https://arxiv.org/abs/2609.17193
作者: Zhen Li,Jun Cai,Haoran Gao,An Li,Tan Li
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large language model (LLM)-powered agentic AI services increasingly demand low-latency inference, motivating the deployment of LLMs across distributed edge servers. However, heterogeneous communication and computing capabilities, together with dynamically evolving inference states, make the edge server selection for each incoming request time-varying and tightly coupled across slots. In this paper, we investigate an online request scheduling framework for edge LLM inference that jointly minimizes long-term average end-to-end latency and regulates workload distribution across heterogeneous edge servers. Two main challenges arise in this context. First, conventional latency models cannot accurately capture the fine-grained dynamics of multi-stage LLM execution. Second, the latency consequence of a scheduling decision is observed only after request completion, making immediate decision evaluation difficult. To address these challenges, we develop a cross-slot inference model that captures transmission, prefill, iteration-level decoding, and key-value (KV) cache evolution for each diverse request, and characterize server workload through a KV cache memory-time consumption metric. We propose the LYREO approach that transforms the long-term load-balancing constraint via Lyapunov optimization and employs reward redistribution with sequencebased return prediction to convert delayed outcomes into timely learning signals for earlier decisions. Simulations under various configurations demonstrate that LYREO consistently achieves lower latency and more balanced load distribution than representative learning-based and heuristic baseline schemes.
[AI-20] MOCC-R1: Reinforcing Reasoning -Response Consistency for Multimodal Counselor Response Generation
链接: https://arxiv.org/abs/2609.17180
作者: Wenjie Zheng,Qiming Xie,Jianfei Yu,Rui Xia
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Multimodal counselor response generation (MCRG) aims to generate an appropriate counselor response from multimodal dialogue histories. Progress is limited by two gaps: first, existing datasets rarely capture sustained, human-recorded counseling interactions conducted by qualified counselors; Second, existing methods do not explicitly optimize consistency between counseling reasoning and the generated response, potentially undermining the reliability of MCRG systems. Thus, we introduce MOCC, a multimodal counseling conversation corpus containing over 200 hours of interactions involving 154 credential-verified counselors. Based on MOCC, we propose MOCC-R1, a two-stage framework for optimizing reasoning-response consistency. Cold-start supervised fine-tuning trains the model to generate a structured trajectory consisting of client-state understanding, a response intent that links a counseling principle to a planned action, and the final response. Reinforcement learning (RL) then rewards grounded plan coherence and plan execution, encouraging the inferred state and plan to be supported by the dialogue context and the response to realize that plan. Experiments demonstrate the effectiveness of the proposed MOCC-R1.
[AI-21] A unified framework for global and local interpretability using adaptive derivative-ordered random explanation
链接: https://arxiv.org/abs/2609.17171
作者: Lemen Chao,Ming Lei,Anran Fanga
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:The interpretability of complex machine learning models is of paramount importance, especially in real-world high-stakes domains such as healthcare and finance. However, existing post-hoc interpretability methods suffer from inherent limitations: fragmented analytical processes, inadequate capacity to model nonlinear feature interactions, computational inefficiencies, and over-reliance on specific model architectures. To address these challenges, this paper provides a novel method - Adaptive Derivative-Ordered Random Explanation (ADORE) - that leverages first- and second-order derivatives to accommodate nonlinear model complexities, while enabling effective capture of feature-sample interactions within a unified analytical framework. ADORE integrates global feature importance with local sample contributions, precisely quantifying feature impact by capturing both magnitude and direction, and identifying critical samples influencing model decisions. Furthermore, it achieves computational efficiency through randomized singular value decomposition (SVD) and dynamic sparsity detection, making it scalable to large, high-dimensional datasets. Experiments across three data modalities - tabular, text, and image - demonstrate that ADORE outperforms existing methods such as LIME and SHAP in handling complex interactions and computational efficiency, while providing detailed and reliable explanations. To facilitate adoption and reproducibility, ADORE has been released as an open-source Python package, hosted on GitHub, enabling researchers and practitioners to readily adapt and apply our approach to their specific tasks, models, and datasets.
[AI-22] Kernel-Based Metrics Learning for Uncertain Opponent Vehicle Trajectory Prediction in Autonomous Racing
链接: https://arxiv.org/abs/2609.17147
作者: Hojin Lee,Youngim Nam,Sanghun Lee,Cheolhyeon Kwon
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted version of the article published in IEEE Robotics and Automation Letters
Abstract:Autonomous racing confronts significant challenges in safely overtaking Opponent Vehicles (OVs) that exhibit uncertain trajectories, stemming from unknown driving policies. To address these challenges, this study proposes heterogeneous kernel metrics for Deep Kernel Learning (DKL), designed to robustly capture the diverse driving policies of OVs, and carry out precise trajectory predictions along with the associated uncertainties. A key virtue of the proposed kernel metrics lies in their ability to align similar driving policies and disjoin dissimilar ones in an unsupervised manner, given the observed interactions between the Ego Vehicle (EV) and OVs. The efficacy of the proposed method is substantiated through experimental studies on a 1/10th scale racecar platform, demonstrating improved prediction accuracy and thereby safely overtaking against OVs. Furthermore, our method is computationally efficient for onboard computing units, affirming its viability in fast-paced racing environments. The video and source code can be found at this https URL.
[AI-23] Continual Learning for Traversability Prediction with Uncertainty-Aware Adaptation
链接: https://arxiv.org/abs/2609.17141
作者: Hojin Lee,Yunho Lee,Daniel A Duecker,Cheolhyeon Kwon
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted version of the article published in IEEE Robotics and Automation Letters. DOI: https://doi.org/10.1109/LRA.2025.3619687
Abstract:Traversability prediction is a critical component of autonomous navigation in unstructured environments, where complex and uncertain robot-terrain interactions pose significant challenges such as traction loss and dynamic instability. Despite recent progress in learning-based traversability prediction, these methods often fail to adapt to novel terrains. Even when adaptation is achieved, retaining experience from previously trained environments remains a challenge, a problem known as catastrophic forgetting. To address this challenge, we propose a continual learning framework for traversability prediction that incrementally adapts to new terrains using a generative experience recall model. A key virtue of the proposed framework is two folds: i) retain prior experience without storing past data; and ii) incorporate the uncertainty of the generated samples from the recall model, enabling uncertainty-aware adaptation. Real-world experiments with a skid-steering robot validate the effectiveness of the proposed framework, demonstrating its ability to adapt across a series of diverse environments while mitigating catastrophic forgetting.
[AI-24] FirmCORe: A Benchmark for Structured Reasoning about Inter-Firm Collaboration Opportunities
链接: https://arxiv.org/abs/2609.17128
作者: Tian Du,Tiantong Wu,Yafei Wang,Mengyu Liu,Xingyan Chen,Mu Wang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Comprehensive structured data on inter-firm relationships is often scarce or inaccessible because many relationships are privately negotiated, selectively disclosed, and fragmented across proprietary databases. This scarcity hinders the discovery of collaboration opportunities, particularly for startups and small and medium-sized enterprises. Firm profiles are readily available, but collaboration potential cannot be inferred from business similarity alone, since similar firms may be competitors, whereas dissimilar firms may offer complementary products, technologies, channels, capabilities, or capital. We present FirmCORe (Inter-Firm Collaboration Opportunity Reasoning), a human-annotated benchmark for pairwise reasoning over weakly structured firm profiles, comprising 2,805 labeled firm pairs. Given two firm profiles, a model must determine whether the available evidence supports a collaboration opportunity and, for positive pairs, jointly predict its strength, primary collaboration type, and role direction. FirmCORe also provides parallel Chinese- and English-language evaluation sets containing identical instances and gold labels, enabling controlled analysis of input-language sensitivity. Experiments with representative locally deployed and hosted large language models (LLMs) show that the strongest model achieves a macro-F1 score of 74.51 for opportunity detection but only 61.57% exact match across all four output fields. Language effects vary across models, and high cross-language agreement can mask errors shared across languages. These results indicate that current LLMs are substantially more reliable at detecting broad collaboration opportunities than at identifying their specific types and role directions.
[AI-25] Finding Common Mistakes In Modelling With Mathematical Formalisms Using LLM s
链接: https://arxiv.org/abs/2609.17111
作者: Lilian Killich,Marko Schmellenkamp,Fabian Vehlken,Thomas Zeume
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Logic in Computer Science (cs.LO)
备注:
Abstract:Modelling with mathematical formalisms like logical formulas, mathematical equations, or regular expressions is an important yet challenging task for students of computer science and other STEM disciplines. Identifying common mistakes occurring in this context is an important step towards helping struggling students by providing targeted high-quality feedback, e.g. in interactive learning systems. We present a tool-supported workflow that allows to (1) identify candidates for common mistakes that explain many student mistakes in large educational data sets, (2) cluster candidates according to similarities, and (3) visualize resulting clusters for instructors and CS education researchers. The visualization is designed to help researchers to identify common modelling mistakes. The candidates for common mistakes are represented by bug fixing transformations that translate incorrect formalizations into correct formalizations; they are generated by an LLM and validated algorithmically. We show that this approach works well by reproducing common mistakes in propositional logic modelling that were identified by hand in the literature; showing that, unlike other algorithmic approaches, the LLM-based approach is suitable for very large sets of data; and applying it to multiple other formalisms to showcase it generalizes beyond propositional logic. Subjects: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Logic in Computer Science (cs.LO) Cite as: arXiv:2609.17111 [cs.CY] (or arXiv:2609.17111v1 [cs.CY] for this version) https://doi.org/10.48550/arXiv.2609.17111 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-26] Symbolic Separation: Grounding Deep Agents in Knowledge Graphs for Trustworthy Operational Data Analytics
链接: https://arxiv.org/abs/2609.17107
作者: Baibek Davletiyarov,Junaid Ahmed Khan,Andrea Bartolini
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Generative AI promises natural language access to the massive numerical telemetry of data centers and Industry 4.0 installations, yet text-to-query and tool-using agents stay unreliable: even frontier models answer little more than half of real-world database questions, and far fewer of the multi-step, operational ones, because the LLM must compose how heterogeneous sources relate and hallucinates the relations, not just the fields. We propose symbolic separation: a deep agent reasons freely but may act on data only through an ontology-constrained Virtual Knowledge Graph with deterministic pre-execution validation. Unlike a tool API’s interface contract, this domain-semantic contract turns a complex question into one validated graph traversal instead of LLM-inferred joins. Instantiated as the Neurosymbolic Deep Analyst and evaluated on 49.9 TB of superconputer telemetry against a rigid workflow and a non-symbolic ablation, it raises end-to-end task success from 43% to 86%, prevents silent data-integrity errors that no syntactic check catches, and cuts token cost by 2.4x, letting a smaller on-premise model outperform a larger one.
[AI-27] Semi-Supervised Learning-Based Genetic Biomarkers Dataset for Multiple-Stage Hepatocellular Carcinoma Prediction
链接: https://arxiv.org/abs/2609.17100
作者: Ahmed Ammar Kubba,Manar Abu Talib,Jibran Sualeh Muhammad,Ali Bou Nassif,Abdalla Sayed Mohamed,Darko Castven,Jens U. Marquardt
类目: Artificial Intelligence (cs.AI)
备注: 6 pages, 7 figures, 2 tables, published at the 18th International Conference Series on Developments in eSystems Engineering
Abstract:Liver cancer is a complex disease responsible for a high number of deaths across the globe each year, making automated solutions for liver cancer classification urgent. The most common form of liver cancer is hepatocellular carcinoma (HCC), accounting for over 90% of liver cancer cases. There is a distinct lack of publicly available HCC datasets utilizing genomic data, which is necessary for training artificial intelligence (AI) models for automated HCC classification. This study proposes constructing a multi-stage HCC dataset using XGBoost and Semi-Supervised learning on three separate datasets of genomic biomarkers, utilizing their existing labels in the Semi-Supervised learning process to label the proposed dataset. The proposed dataset consists of 770 patient samples in total, categorized into five classes that represent normal tissue alongside different stages of HCC. Each sample in the dataset consists of 11,150 different gene expression levels. The XGBoost model demonstrated a final classification accuracy of 96.5% during the Semi-Supervised learning process.
[AI-28] Scaling-Score Conformal Prediction for Multi-Target Regression
链接: https://arxiv.org/abs/2609.17091
作者: Sylvain Rousseau(Heudiasyc),Soundouss Messoudi(Heudiasyc)
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Multi-target regression requires a model to simultaneously predict several related outputs. Conformal prediction provides distribution-free, finite-sample marginal coverage guarantees, but extending these to joint multi-dimensional regions in a model-agnostic, sample-efficient manner remains challenging: max-aggregation ignores scale differences, copula-based methods are only asymptotically valid, rectangular methods typically split the calibration set, and quantile or density-based methods require training a specialised model beyond a plain point predictor. We propose the scaling-score conformal method, which is model-agnostic (requires only component-wise absolute residuals), uses a single calibration set, and yields four nested output types: an outer rectangle (SCO) with valid joint coverage, the exact set R \alpha , a staircase (SC 2 ) over approximation of R \alpha , and an inner rectangle (SCI). A single hyperparameter \gamma \in (0, 1) controls the base-rectangle quantile level independently of \alpha . We prove downward-closedness and a rectangular sandwich bound and derive a closed-form outer rectangle. Experiments on 29 realworld datasets confirm valid joint coverage; SC 2 with \gamma = 1- \alpha consistently achieves competitive volume relative to baselines, with the advantage growing with output dimension d.
[AI-29] Sample-Conditioned Representation Selection for Audio Few-Shot Learning ICASSP27
链接: https://arxiv.org/abs/2609.17076
作者: Fengrui Liu,Ningxin Shen,Yi Li,Yiwei Fu,Feng Liu,Jiangmeng Li
类目: Artificial Intelligence (cs.AI); Sound (cs.SD)
备注: Submitted to ICASSP27
Abstract:Few-shot audio classifiers may rely on foreground-background co-occurrences and fail when those correlations shift. On SpurAudio, the resulting representation shift is concentrated and class dependent: for ResNet12, the top 10 percent of channels explain 82.80 percent of the null-corrected shift contribution. We propose SAMPLESELECT, which predicts a fixed-budget feature mask independently for each input while keeping the encoder and source classifier frozen. Training uses differentiable Gumbel Top-k selection with foreground classification and cross-background contrastive losses; inference uses deterministic Top-k masks and support-only linear adaptation. Across ResNet12 and Conv64 in 5-way 1-shot and 5-shot evaluation, SAMPLESELECT gives the best OOD accuracy among the compared methods and improves the matched full-representation control by 4.90-8.38 percentage points. Ablations and representation analyses further support the learned selection mechanism. Code is available at this https URL
[AI-30] Repurposing Unified Topological Signatures for Graph Representation Learning
链接: https://arxiv.org/abs/2609.17061
作者: Sanyam Sanjay Jain,Anshika Krishnatray,Aditya Sharma,Vinti Agarwal
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Message-passing Graph Neural Networks (GNNs) iteratively propagate and aggregate local neighborhood information followed by global readout to learn graph representations. However, their discriminative power is upper-bounded by the Weisfeiler–Lehman (1-WL) graph isomorphism test. This prevents GNNs from distinguishing certain non-isomorphic graphs with identical local neighborhood structures, often leading to similar graph representations. Unified Topological Signatures (UTS) capture compact, multi-scale representation of global graph topology derived from persistent homology. We introduce two complementary UTS signatures: Graph_UTS- a static signature of the input graph topology, and Embedding_UTS- a dynamic signature of the evolving embedding topology. They encode structural information inaccessible to 1-WL-based message-passing GNNs, yet their capabilities are explored solely for post-hoc embedding-space analysis. We integrate UTS into GNN training across three architectural interventions: (i) UTS-Aug: augmenting with standard readout feature that encodes graph’s true topology; (ii) UTS-Reg: topological regularizer that constrains representation collapse; (iii) UTS-Pool: topology-guided pooling that retains structurally critical nodes. We further leverage UTS as a layer-wise diagnostic to quantify oversmoothing during GNN training. Theoretically, we show that integrating UTS into GNN optimization strictly extends GNN expressivity beyond the 1-WL hierarchy. Experiments on three graph classification benchmarks show consistent benefits: Graph-UTS, Dual-UTS, and UTS-Pool improve accuracy across all three datasets, Embedding-UTS provides smaller but similarly consistent gains, and UTS-Reg’s benefit varies across graph domains. Accuracy improves by up to 5.8% with Graph-UTS augmentation, by up to 1.9% with UTS-Reg, and achieves comparable performance to TOGL with UTS-Pool.
[AI-31] Sparse MLLM Anchors Dense Adaptation: Breaking the Self-Referential Loop in Wild Test-Time Adaptation
链接: https://arxiv.org/abs/2609.17040
作者: Zhenbin Wang,Lei Zhang,Lituan Wang,Yan Wang,Zhao Zhang,Wei Huang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Wild test-time adaptation (WTTA) updates a source model online under small test batches, concurrent distribution shifts, and time-varying class imbalance. Most WTTA methods derive their adaptation signals, including predictive uncertainty, sample reliability, and local feature geometry, from the model being adapted. When the source model is unreliable under shift, these signals can reinforce its own errors, forming a self-referential loop. We introduce MASA (Multimodal-LLM-Anchored Semantic Adaptation), which complements model-internal evidence with structured semantic descriptions from a frozen multimodal large language model (MLLM). To limit inference cost, MASA queries the MLLM only for a small set of diverse, reliability-ranked anchors. The resulting descriptions capture the object family and nuisance factors such as style, viewpoint, and occlusion. MASA encodes these descriptions, propagates them to neighboring test samples, and stores the resulting visual-semantic information in an online prototype memory. Descriptor-aware retrieval from this memory provides an auxiliary target for lightweight adaptation of normalization-affine parameters. We evaluate MASA on the WTTA ImageNet-C benchmark under limited-batch, mixed-domain, and imbalanced-label-shift settings with ResNet and ViT backbones.
[AI-32] Distributed JEPA: A Self-Supervised Framework for Energy Forecasting
链接: https://arxiv.org/abs/2609.17029
作者: Liana Toderean,Tudor Cioara,Vasilis Michalakopoulos,Efstathios Sarantinopoulos,Ionut Anghel,Elissaios Sarmas
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Traditional energy forecasting solutions rely on task-specific supervision and energy asset representations, limiting transferability and the ability to capture general temporal dynamics across heterogeneous assets. We address this by proposing a distributed Joint Embedding Predictive Architecture (JEPA) for self-supervised learning from heterogeneous energy time-series. The framework predicts latent representations of masked temporal segments while integrating temporal observations and contextual information within a shared embedding space. To prevent representation collapse, training combines a latent-space predictive objective with covariance and temporal variance regularization. The evaluation was conducted on energy consumption and generation datasets under data-degradation scenarios and compared with a Transformer forecasting baseline. The learned representations remained stable (cosine similarity \approx 0.98 ; effective rank 185-235). JEPA achieved performance comparable to a Transformer on building energy data, higher R^2 in 3/5 consumer clusters, and outperformed the baseline on 9/10 unseen PVs ( R^2 =0.73-0.88 vs. 0.45), while showing greater robustness to missing data.
[AI-33] SKIP: a Self-knowledge-guided Step-wise Preference Learning Framework for Concise Reasoning IJCNN2026
链接: https://arxiv.org/abs/2609.17019
作者: Qinhong Lin,Yuhao Zhang,Yinglun Feng,Zhongliang Yang,Linna Zhou
类目: Artificial Intelligence (cs.AI)
备注: 8 pages,3 figures. Accepted at IJCNN 2026
Abstract:While Chain-of-Thought (CoT) reasoning has been proven to be effective, it often leads to overthinking, resulting in computational overhead, inference latency, and even degraded performance in large language models (LLMs). Existing concise reasoning frameworks significantly compromise accuracy while compressing the length of output. In this paper, we propose SKIP, a self-knowledge-guided step-wise preference learning framework. Starting with lightweight fine-tuning to adjust the model’s output style, SKIP introduces a carefully designed knowledge probing mechanism to guide model to output an answer at each reasoning step. Based on the correctness of intermediate steps, we construct preference data that guide the model toward more efficient and correct reasoning by leveraging DPO. Experimental results demonstrate that our method effectively improves reasoning compression while mitigating performance degradation after fine-tuning. Besides, SKIP shows strong generalization ability on out-of-distribution datasets. We further conducted ablation studies on the component parameters of our framework.
[AI-34] ORDER: Task-Conditioned Routing for Retrieval-Augmented Generation
链接: https://arxiv.org/abs/2609.17012
作者: Aurélien Pellet(LRE),Julien Perez,Marie Puren
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Retrieval-Augmented Generation (RAG) pipelines typically rely on a fixed indexing and retrieval configuration determined at preprocessing time. This one-size-fits-all design is ill-suited to domain-expert settings, where heterogeneous queries require different chunking granularities, metadata constraints, and source-selection strategies. As a result, configurations that are effective for one family of queries often perform poorly for others. In this paper, we introduce ORDER (Optimal Routing for Dynamic Evidence Retrieval), a query-conditioned RAG framework that jointly adapts indexing and retrieval to the incoming query. Our approach first discovers semantic clusters over a given set of questions associated to a corpus and learns, for each cluster, a chunking strategy together with a suited metadata filtering and reranking configuration. At inference time, queries are routed to the appropriate pre-built index through nearest-centroid assignment. To further improve retrieval, we propose a supervised query router (QRe) that predicts which collections are most likely to contain relevant evidence, coupled with a Uniform Multi-source Sampler (UMS) that allocates the retrieval budget evenly across the selected sources. We evaluate our framework on large-scale, heterogeneous historical archives and show that conditioning both indexing and retrieval on the query consistently outperforms both naive baselines and strong state-of-the-art RAG systems in complex expert-domain environments.
[AI-35] FlexEE: Self-Speculative and KV-Compatible Early Exiting for Offloading-Aware LLM Inference
链接: https://arxiv.org/abs/2609.17008
作者: Qihu Xie,Ziwei Li,Yi Kang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large language model (LLM) inference is often constrained by both computation and memory, especially in offloading-based deployments where model weights are transferred across memory hierarchies during autoregressive decoding. In this setting, reducing the number of executed layers can lower per-token latency while also avoiding costly weight movement. Motivated by this observation, we present FlexEE, an early exiting framework for resource-constrained and offloading-based LLM inference. FlexEE makes early exiting practical for LLM decoding through layer-wise exit supervision for reliable intermediate-layer prediction, self-speculative decoding over a Top-K local vocabulary for low-cost exit decisions, and dynamic hidden state management for KV-cache-correct and memory-aware execution. Across generative and downstream tasks, FlexEE enables efficient early exit with minimal accuracy degradation, delivering up to 1.27 \times /3.16 \times and 1.25 \times /2.83 \times end-to-end speedups on Llama2-7B and Llama3-8B under 0%/50% weight offloading, respectively.
[AI-36] Affect-Prototype Guided Fusion for Open-Vocabulary Incomplete Multi-modal Emotion Recognition
链接: https://arxiv.org/abs/2609.16962
作者: Yichi Zhang,Shenyue Wang,Jing Luo,Chunyang Yu,Xinyu Yang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Open-vocabulary multimodal emotion recognition (OV-MER) aims to generate open natural-language emotion labels from multimodal affective cues. In real-world scenarios, however, complete and synchronized modal data are difficult to obtain due to limitations of acquisition devices and user privacy constraints. Existing OV-MER methods are largely designed for full-modal inputs, and fail to perform effective feature fusion under modal missing conditions. Meanwhile, current fusion approaches designed for incomplete modalities mainly focus on fixed-label recognition context, and cannot satisfy the demand for fuse emotional cues guided with arbitrary emotion semantics in OV-MER context. To tackle these challenges, this paper proposes an Affect-Prototype-Conditioned Fusion (APCF) framework for incomplete open-vocabulary emotion recognition. As a candidate-free generative framework, APCF extends modal contribution learning to scenarios guided by arbitrary emotional semantics. Specifically, we construct an affect-prototype library to explicitly model multimodal contribution characteristics corresponding to diverse emotions, which provides dynamic constraints for modal fusion under different emotional semantic perspectives. Conditional retrieval and feature aggregation are conducted based on available modal features. The refined fused affective representations are then fed into an LLM decoder to produce open-vocabulary emotion labels. Experiments on the OV-MERD+ and MER-FG datasets demonstrate that APCF substantially outperforms state-of-the-art baselines.
[AI-37] AntennaFlow: A Generative Flow Model for Offset Correction in Phaseless Antenna Testing
链接: https://arxiv.org/abs/2609.16948
作者: Yongzhi Li,Chongting Shen,Menglin Chen,Xun Jiang,Zhengpeng Wang
类目: Artificial Intelligence (cs.AI); Information Theory (cs.IT)
备注: 6 pages,6 figures
Abstract:Near-field to far-field transformation is central to large-aperture antenna testing, yet two coupled challenges remain: costly phase acquisition at millimeter-wave bands and violations of the centering assumption under offset mounting. Existing methods address these issues separately, requiring either dense full-field data or offset vectors. We tackle both jointly by exploiting a key observation: amplitude fields under different offsets are coordinate-transformed views of the same near field. The challenge is to recover the center-aligned field from offset amplitudes without a phase or offset vector. We propose AntennaFlow, a three-stage framework: a contrastively learned encoder that maps offset views to an offset-invariant embedding, a deterministic flow-matching transport that maps offset amplitudes to center-aligned ones, and the Simplified Extrapolation Technique, whose Green-function Taylor expansion is valid only for centered fields. Experiments show that AntennaFlow enables fast, phaseless, offset-vector-free NF–FF reconstruction from sparse amplitude-only measurements, consistently outperforming existing baselines while preserving physical consistency.
[AI-38] AeroLat: Channel-Aware Latent Space Semantic Communication for Decentralized UAV Swarms
链接: https://arxiv.org/abs/2609.16947
作者: Rajdeep Ghosh,Goparaju Venkata Seshachala Sree Vatsava,Sudip Misra
类目: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI)
备注:
Abstract:Communication in latent space offers an intriguing alternative to symbolic messages for decentralized autonomous Unmanned Aerial Vehicle (UAV) swarms operating over bandwidth-constrained, time-varying wireless links. However, when homogeneous frozen models are prompted with discretized perceptual inputs, their broadcast states collapse toward the shared prompt template. In view of this, we propose AeroLat, a channel-aware latent semantic communication framework that uses evidence injection. The resulting latent states are then passed through an explicit communication model that encompasses bandwidth-limited serialization, additive noise and information staleness, which facilitates a joint assessment of communication fidelity and swarm-level coordination. Across multi-seed simulations, AeroLat provably remains resilient to codec choice, faults and increasing swarm size. It consistently reproduces the latent-swarm anomaly, while no-whitening controls recover the collapse. In particular, AeroLat is capable of reducing false similarity by 97.5%.
[AI-39] Beyond Token-Local Imitation: Reward-Compatible Temporal Credit Assignment for On-Policy Distillation
链接: https://arxiv.org/abs/2609.16937
作者: Shiqi Liu,Zeyu He,Letian Tao,Guojian Zhan,Jiaxin Gao,Feihong Zhang,Jingliang Duan,Wei Xiong,Kehua Sheng,Bo Zhang,Yang Guan,Shengbo Eben Li
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Programming Languages (cs.PL)
备注:
Abstract:On-policy distillation (OPD) has emerged as an effective approach for large language model post-training, yet existing objectives face a trade-off between objective fidelity and optimization stability. Token-level OPD provides stable but local supervision, whereas sequence-level OPD captures future credit at the cost of horizon-dependent variance. We establish a unified temporal-credit view of these formulations, showing that practical token-level OPD can be interpreted as a temporal approximation to the sequence-level reverse-KL gradient. Building on this connection, we propose \gamma OPD, which uses discounted temporal credit assignment to balance long-horizon supervision and optimization stability, while admitting a horizon-independent variance bound. We further develop a reward-compatible bounded mixing (RBM) mechanism for \gamma\mathrmOPD that balances verifiable outcome feedback with the discounted OPD advantage to move beyond purely teacher-dependent optimization. Experiments on mathematical and code reasoning demonstrate consistent improvements over existing OPD methods across vanilla, size-mismatched, and multi-teacher distillation settings.
[AI-40] RepoAtlas: Guiding Coding Agents via Evolving Multimodal Repository Views
链接: https://arxiv.org/abs/2609.16936
作者: Yunxiang Zhang,Haiquan Wang,JiaWei Guo,Hanyang Xia,Yan Chen,Tong Chen,Zhang Zhiwei,Junchen Ye
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language model (LLM)-powered coding agents have made rapid progress in automating software engineering tasks, yet repository-level issue resolution remains challenging. Beyond generating a plausible patch, an agent must localize relevant code across interdependent files and maintain repository context that is both sufficient and focused. Code graphs expose non-local relations, but linear text interfaces obscure their topology; rendering the full repository graph yields visual representations that are too dense to perceive reliably, whereas a one-shot local view becomes stale as exploration proceeds. We present \textbfRepoAtlas, a training-free module that maintains evolving multimodal repository views through a \emphselect–project–refresh loop over a repository code graph. RepoAtlas combines evidence from the issue with the agent’s current exploration state to select a task-relevant region under a fixed budget, projects the selected structure into complementary visual and textual representations, and refreshes the view when changes in the exploration state render it outdated. We evaluate RepoAtlas on SWE-bench Verified, where it improves the resolve rate by 2.4 points while reducing input tokens and model calls by 5.8% and 7.8% on average, relative to the strongest multimodal graph baseline, with consistent gains across three models of different families and scales.
[AI-41] When Confidence Signals Disagree: Local and Global Confidence in Autoregressive Language Models
链接: https://arxiv.org/abs/2609.16933
作者: Julio C. Amador Diaz Lopez
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Modern predictive systems expose multiple quantities that are commonly interpreted as measures of confidence. However, these quantities can summarize different aspects of the predictive process. This distinction matters when confidence is used to evaluate reliability or inform downstream oversight and control. We investigate whether different confidence readouts are empirically interchangeable in an autoregressive language model by comparing local confidence, defined from the probability of the greedy-selected answer token, with global confidence, defined from modal-answer frequency under repeated sampling. Across MMLU and ARC Challenge, the two signals are weakly correlated and differ substantially in their association with correctness: global confidence is moderately associated with correctness, whereas local confidence shows little association. We further test whether question-level disagreement between the signals is associated with sampling instability. On ARC, larger local–global confidence gaps are associated with higher answer entropy, more distinct sampled answers, and lower modal-answer concentration. The gap–entropy association persists when disagreement and instability are estimated from disjoint stochastic samples, indicating that it is not explained by shared finite-sample variation. The corresponding relationship is substantially weaker on MMLU, where only 4% of questions exhibit sampling instability. These results show that confidence readouts derived from the same predictive system are not empirically interchangeable and that their disagreement can provide a diagnostic of unstable sampling behavior. Confidence should therefore be treated as an explicitly defined measurement rather than as a single intrinsic scalar property of a model, particularly when it is used to inform downstream evaluation, oversight, or control.
[AI-42] Repurposing Deep Limit Order Book Forecasting for Scenario-Conditioned Market Impact Modeling
链接: https://arxiv.org/abs/2609.16930
作者: Eljas Linna,Kestutis Baltakys,Derrick Manoharan,Alexandros Iosifidis,Juho Kanniainen
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Deep Limit Order Book forecasting models capture nonlinear market dynamics, but their ability to quantify the effects of counterfactual order book messages has not been systematically validated. We introduce a model-agnostic framework that compares a trained forecaster’s predictive distributions before and after injecting mechanically valid counterfactual messages, defining short-horizon model-implied market impact. A Transformer-based forecaster recovered scenario rankings with a Spearman correlation of 0.99 and 97.2% directional agreement with realized historical outcomes among non-neutral scenarios. Observation-level analysis further showed that estimated impacts captured incremental sequence-dependent variation beyond scenario identity and the pre-event forecast. These results provide evidence that pretrained Limit Order Book forecasters can be repurposed for scenario-conditioned response modeling without retraining.
[AI-43] QART: A Quantum-Classical Hybrid Architecture for Long-Horizon Reasoning – Exploring a Conditional Path toward Quantum Scaling
链接: https://arxiv.org/abs/2609.16887
作者: Lehao Lin,Yuheng Cheng,Guolong Liu,Yao Li,Xuning Tan,Xiyuan Zhou,Ruixi Zou,Shi Wang,Huan Zhao,Wenxuan Liu,Haifeng Wu,Junhua Zhao
类目: Artificial Intelligence (cs.AI)
备注: 18 pages, 3 figures
Abstract:Long-horizon reasoning is vulnerable to early errors that compromise later decisions. We present QART, the Quantum-Augmented Reasoning Transformer, a quantum–classical hybrid architecture combining a backbone language model with quantum encoding, CIM-based QUBO optimization, and quantum decoding. Semantic information can come from hidden representations or model-generated text; detailed encoding and optimization procedures remain proprietary. Under explicit assumptions, we establish a conditional asymptotic reliability separation from single-trajectory autoregressive LLMs. For a common task family with aligned optimality and acceptance criteria, autoregressive acceptance probability tends to zero when cumulative conditional risk of irreversible errors diverges. QART’s task-optimal-path recovery probability remains bounded away from zero if conditional probabilities for optimal-path coverage and semantic fidelity, spectral certification, dynamical reachability, and faithful readout remain uniformly positive under a specified resource schedule. The architecture alone does not imply these bounds. Paired measurements on six long-horizon benchmarks using DeepSeek V4 Flash, GLM-5.3, and GPT-5.5 xhigh in a Codex agent environment favor QART in 14 of 15 backbone–benchmark pairs. Relative gains reach 84.0% on SciCode, 47.6% on \tau^3 -Bench, and 44.4% on Terminal-Bench 4.0; the DeepSeek V4 Flash configuration regresses by 7.8% on DeepSWE. These results do not directly validate the asymptotic separation. Potential quantum scaling laws are formulated as conditional hypotheses. A quantum-advantage interpretation requires a demonstrated CIM quantum advantage over strong classical solvers and its transfer to end-to-end reasoning after all system overheads.
[AI-44] Bridging Learned Visual Perception and Symbolic Belief-Space Planning
链接: https://arxiv.org/abs/2609.16884
作者: Guy Azran,Michael Navat,Sarah Keren
类目: Artificial Intelligence (cs.AI); Robotics (cs.RO)
备注: To appear in the Proceedings of the 3rd International Conference on Neuro-Symbolic Systems (NeuS), 2026
Abstract:In partially observable settings, agents must act without full knowledge of the world state and rely on uncertain state-estimation pipelines. Obtaining grounded and verifiable symbolic plans under such uncertainty remains a key challenge. Recent work has integrated Vision-Language Models (VLMs) to bridge perception and symbolic reasoning, following two main paradigms. The first, VLM-as-planner, maps images directly to action sequences, and the second, VLM-as-grounder, grounds observations into symbolic predicates used as the initial state by off-the-shelf planners. Both approaches ignore uncertainty in the planning process, compromising robustness. We introduce a third paradigm, VLM-as-probabilistic-grounder, a novel approach that captures the uncertainty of VLM predicate groundings as a probability distribution over symbolic states. This enables planning in belief space and producing robust plans under uncertainty. Experiments in simulated household robot settings show improved robustness and task success over deterministic grounding, underscoring how our approach leverages foundation models for reliable planning under uncertainty.
[AI-45] CoAdapt: An LLM -based Framework for Adaptive Collaborative Perception in IIoT Robotic Swarms
链接: https://arxiv.org/abs/2609.16852
作者: Houssam Hajj Hassan,Antonia Maria Masucci,Lynda Zitoune(L2S),Salah-Eddine Elayoubi(L2S)
类目: Artificial Intelligence (cs.AI); Robotics (cs.RO)
备注:
Abstract:Industrial IoT environments increasingly deploy autonomous mobile robots for tasks such as material handling, product assembly, or infrastructure inspection. In such deployments, collaborative perception enables robots to share LiDAR observations and collectively construct a richer model of their environment than an individual agent could produce alone. However, industrial environments are dynamic spaces where robot positions shift continuously, network bandwidth fluctuates, and the marginal contribution of robots to perception quality varies at runtime. Existing collaborative perception approaches are designed for static participation assumptions and cannot adapt to these dynamics without sacrificing either detection precision or communication efficiency. This paper presents CoAdapt, an adaptive collaborative perception framework for IIoT robotic swarms in which a Large Language Model (LLM) serves as a runtime fusion controller, jointly deciding which robots participate in the fusion process and which fusion algorithm to apply based on the current spatial configuration and network state. The LLM reasons over structured natural language descriptions of the scene derived from raw LiDAR point clouds, requiring no taskspecific training and generalizing to unseen swarm topologies. Evaluated on the OPV2V benchmark across 25 scenarios, our approach achieves a 38% reduction in communication cost while maintaining detection precision comparable to static baseline approaches.
[AI-46] Execution Flexibility in Automated Planning : A Comparative Evaluation of Deordering and Reordering Strategies
链接: https://arxiv.org/abs/2609.16822
作者: Md. Monjurul Islam,Sabah Binte Noor,Fazlul Hasan Siddiqui,Gahangir Hossain
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:This study covers foundational concepts for enhancing plan-execution flexibility, including partial-order planning, the producer-consumer-threat formalism, and a range of deordering and reordering strategies. Creating a partial-order plan from a sequential one by removing unnecessary ordering constraints is a practical way to improve execution flexibility, and several methods have been proposed for this task. This study analyzes their capabilities across ordering, action handling, parameter handling, plan structure, concurrency, and complexity, and evaluates them against each other on a shared benchmark. The central finding is that block deordering-based approaches, which restructure causal dependencies through block-level grouping and subplan substitution, substantially outperform MaxSAT-based approaches despite the latter’s theoretical guarantees of minimum reordering. The reason is structural: minimum reordering optimizes within the causal structure already present in the plan, whereas block deordering-based methods change that structure, exposing orderings that would otherwise appear necessary. A further distinction is practical: block deordering-based methods are anytime algorithms that always return a valid result, while MaxSAT-based methods fail entirely on a substantial portion of plans and offer no partial solution when they do. Block substitution further extends the parallel execution by formalizing non-concurrency constraints, though its impact is limited to domains with resource-based interactions. On efficiency, block deordering-based approaches achieve the highest flex gain per unit of computation time, while MaxSAT-based encodings incur large computational overhead.
[AI-47] SOTER: A Generative Time-Series Foundation Model for Wearable Human Physiological Signals
链接: https://arxiv.org/abs/2609.16804
作者: Fangke Chen,Sirry Chen,Wei Chen,Zhongyu Wei
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Time-series foundation models have demonstrated strong cross-domain transfer, yet their common architectural assumptions remain poorly aligned with wearable physiological signals, which are multichannel, irregularly sampled, noisy, and governed by coupled continuous-time dynamics spanning distinct spectral scales. We present SOTER, a generative foundation model for wearable physiological time series that unifies cross-channel coupling, spectrum-guided expert specialization, and continuous-time latent evolution within a single pre-training framework. SOTER combines a spatial feature-aware backbone that models inter-signal dependencies, a power spectral density (PSD)-guided mixture-of-experts layer that routes representations to experts associated with fixed spectral bands through an inspectable, non-learned rule, and a neural controlled differential equation decoder that supports prediction and imputation at arbitrary timestamps. We pre-train SOTER on 226 billion time points from five public physiological datasets and evaluate the same pre-trained model across out-of-distribution zero-shot forecasting, frozen-encoder linear-probe classification, and continuous-time imputation on wearable benchmarks. SOTER achieves the best RMSE on 4 of 6 datasets and the best MAE on 5 of 6 in zero-shot forecasting, the highest average Macro-AUROC in classification, and the lowest imputation error on all six datasets at 75% missingness. It further remains robust to additive acquisition noise, matching or surpassing baselines evaluated on clean inputs even under the strongest corruption. These results indicate that domain-specialized foundation models for wearable physiology benefit from jointly modeling channel structure, spectral scale, and continuous-time dynamics.
[AI-48] Layers Sinks and Scaling: Adaptive Evidence Selection for Multimodal Large Language Models
链接: https://arxiv.org/abs/2609.16795
作者: Zhenbin Wang,Lei Zhang,Lituan Wang,Wei Huang,Yan Wang,Zhenwei Zhang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Multimodal large language models (MLLMs) can answer knowledge-intensive visual questions by combining visual evidence from images with facts retrieved from external sources. However, MLLMs may overlook relevant evidence in both modalities, attending weakly to the textual sentences or visual regions needed for the correct answer. Recent efforts address this by highlighting retrieved text and marking visual regions before generation, but apply a fixed, one-shot policy that cannot adapt to three sources of variation: whether highlighting is necessary, how much evidence different examples require, and when different textual evidence becomes relevant as the answer unfolds. We introduce Adaptive Relevance-guided Evidence Allocation (AREA), a training-free inference-time method that formulates evidence highlighting as adaptive allocation. AREA generates a single probe token to read visual and textual relevance from fixed backbone layers, then makes three decisions: i) whether to intervene (controlled by natural attention coverage and visual sink contamination), ii) how much evidence to expose (determined by relevance entropy), and iii) when to refresh text during generation (triggered by causal context-attention peaks). Across four KB-VQA and seven standard multimodal benchmarks with nine frozen MLLM checkpoints, establishes the best performance among training-free highlighting methods.
[AI-49] Integrating the Analytic Hierarchy Process with Large Language Models for Transparent Multi-Criteria Decision-Making
链接: https://arxiv.org/abs/2609.16779
作者: Han Zhiguang,Farah Benamara(IRIT-MELODI, UT3, IPAL),Pascale Zaraté(IRIT, UT Capitole, IRIT-ADRIA)
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:LLMs are increasingly employed in a wide range of decision-making tasks. However, the opacity of their internal reasoning makes it difficult to validate or interpret their outputs, and the need for interpretability becomes especially critical in high-stakes settings. This study examines the decision-making capabilities of LLMs through the Analytic Hierarchy Process (AHP), a classical and widely used multicriteria decision-making framework. We construct a new annotated benchmark based on AHP and propose the first end-to-end approach that enables LLMs to perform the complete AHP workflow. Experiments in real-world decision problems in the legal and higher-education ranking domains show that our method significantly improves alignment with expert judgments.
[AI-50] Coverag e-Aware Virtual IMU Augmentation for Low-Resource Human Activity Recognition
链接: https://arxiv.org/abs/2609.16768
作者: Jiayuan Gao,Yingwei Zhang,Ziyao Tang,Yuejia Ma,Yuanzhe Chen,Shuchao Song,Boshi Tang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:IMU-based human activity recognition (HAR) enables continuous, privacy-friendly monitoring of daily activities using wearable sensors. However, building reliable HAR models that generalize across diverse users and real-world conditions requires large amounts of labeled IMU data, which are expensive and difficult to collect. Existing approaches mainly rely on augmentation or synthesis to expand available data, but indiscriminately adding virtual samples may provide little new coverage and introduce unreliable supervision. To overcome these challenges, we propose a novel coverage-aware virtual IMU augmentation framework that decides where to supplement real data, how to generate and select virtual candidates, and how strongly to weight them during training. Specifically, we select diversity and scarcity anchors in a learned sensor embedding space, convert anchor dynamics into prompts, and generate virtual IMU candidates for each anchor. We then rank candidates by a selection cost combining anchor proximity and label consistency, and incorporate the selected candidates into HAR training with reliability-based weights. Experiments on public HAR benchmarks show that our method consistently improves recognition performance over competitive baselines, and ablation studies confirm the effectiveness of the proposed framework design.
[AI-51] urn-level Multiscale Density Ratio Estimation for LLM Agents
链接: https://arxiv.org/abs/2609.16760
作者: Zishuo Zhao(Alibaba Group),Kai Chen(Alibaba Group),Ao Li(Alibaba Group),Yuan Liu(Alibaba Group)
类目: Artificial Intelligence (cs.AI)
备注: 15 pages, 9 figures, 3 tables
Abstract:With the rapid development of Large language model (LLM), agent systems enhanced by LLMs show huge potential in being able to deal with complex tasks, especially involving multi-step thinking or interaction with tools. For applying LLM techniques with a well-designed agent paradigm, post-training of LLM in multiple agent scenarios is necessary to achieve better performance. Among the variable post-training techniques, alignment methods such as PPO, DPO, DIL, and GRPO become popular because many papers show a significant positive impact on the model’s performance by punishing negative samples while keeping acceptable training complexity. However, most alignment methods address simple single-turn tasks, and there remains room for improvement for complex multi-turn tasks. We propose Turn-level Multiscale Density Ratio Estimation (tlm-DRE), which assigns different weights on corresponding turns and proposes asymmetric token-level training based on the positive-negative space gaps across multiple turns of tasks. The results of the experiment on a wide range of agent benchmarks show that the proposed method performs competitively compared to traditional alignment methods. The proposed training method enables LLMs to perform robustly in multi-turn reasoning tasks with both in-domain and out-of-domain conditions.
[AI-52] Beyond Episodic AI: Cognitive Field Networks for Biologically Inspired Persistent Cognition
链接: https://arxiv.org/abs/2609.16752
作者: Byung Gyu Chae
类目: Artificial Intelligence (cs.AI)
备注: 35 pages, 13 figures
Abstract:Cognitive Field Theory (CFT) proposes that cognition arises from memory-dressed collective dynamics that generate a persistent macroscopic cognitive field. Here we develop a Cognitive Field Network (CFN), a recurrent Transformer in which the organized hidden field re-enters subsequent inference through [ \Phi_n+1=F_\theta(X_n+1,\Phi_n). ] Rather than prescribing an explicit memory operation, the CFN allows new information to act on an already history-dependent collective state. We find that learning organizes persistent, content-dependent recurrent dynamics whose timescale increases systematically with the trained recurrent horizon. Semantic continuation propagates the recurrent state far beyond this horizon without replay of the target answer. Without content-specific support, the field exhibits finite passive relaxation, whereas periodic re-exposure to relevant input repeatedly renews the surviving state and drives it toward an approximately stationary nonzero regime. Unrelated-input and recurrence-off controls do not reproduce this behavior, while near-paraphrased re-exposure produces weaker renewal, demonstrating representation-sensitive persistence. These results distinguish three dynamical processes: collective memory dressing forms and sustains a history-dependent cognitive field, structured input reorganizes this field, and cross-cycle re-entry makes the resulting state causally available to subsequent inference. The CFN therefore provides a controlled computational platform for studying persistent, history-dependent cognitive dynamics without a separately prescribed memory system.
[AI-53] Continuous-Time Machine Learning: A Unified Mathematical Perspective
链接: https://arxiv.org/abs/2609.16710
作者: Waleed Razzaq,Yun-Sheng Zhao,Yun-Bo Zhao
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Continuous-time (CT) machine learning has emerged as a principled framework for modeling temporal dynamics as a continuous process, particularly when observations are sampled at arbitrary time points or span long-range horizons. However, major branches of CT machine learning have matured in separate research communities, leaving their mathematical relationships and design trade-offs insufficiently characterized. In this survey, we develop a unified, concept-driven view of major CT machine learning branches through a taxonomy that organizes families according to their underlying base mathematical formulations. We present a canonical mathematical formulation that relates these families through different architectural choices of vector-field parameterization, stochasticity, memory mechanisms, and discretization. We compare training algorithms, optimization strategies, and failure modes, highlighting the trade-offs across families. We further provide a comparative analysis of theoretical computational complexity alongside an illustrative architecture-controlled benchmark analysis on representative architectures from each family. We also review software ecosystems supporting their implementation. Finally, we identify open challenges in approximation theory, training stability, hardware-efficient implementations, benchmarking, foundation models, and scientific machine learning, and discuss an agenda for future research.
[AI-54] World Models for Embodied Intelligence: From Plausible to Controllable to Actionable
链接: https://arxiv.org/abs/2609.16697
作者: Nanjie Yao,Hao Wang,Chong Cheng,Zhikang Chen,Wenzhe Li,Jiafei Lyu,Li Shen,Peilin Zhao,Zongqing Lu,Gao Huang,Steven Hoi,Dacheng Tao,Deheng Ye
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: Project Page: this https URL
Abstract:World models connect perception and decision-making in embodied intelligence by maintaining hidden state, anticipating consequences, comparing interventions, and adapting when execution departs from expectations. Although progress is often measured by visual fidelity, their value lies in improving behavior. Before reaching for a cup, a person anticipates its weight and resistance to grasping, shaping the hand before contact. Such anticipation is coarse and rarely pictorial, yet it guides action. This raises a central question: which predictive capabilities improve behavior? Existing surveys, organized by architecture, output modality, or application domain, leave this question implicit. We introduce three progressively stronger capability levels: Plausible models preserve task-relevant temporal, geometric, or physical structure; Controllable models additionally predict how interventions alter that structure; and Actionable models translate predictions into measurable gains in planning, action, learning, evaluation, verification, recovery, or data selection. We complement this hierarchy with a 3 x 4 matrix crossing geometry, physics, and action grounding with improvement loops centered on data, rewards, policies, and the model itself. Using this framework, we survey manipulation, navigation, locomotion, autonomous driving, and general embodied learning, tracing technical progressions, clarifying capability requirements, and examining datasets, benchmarks, and evaluation protocols. We identify challenges in long-horizon consistency, uncertainty calibration, causal intervention testing, latency, verification and recovery, and cross-embodiment transfer. This perspective shifts evaluation from visual plausibility toward whether predictions capture task-relevant state, reflect intervention effects, and improve the closed-loop behavior of embodied agents.
[AI-55] Weave: Learning Whole-Body Dexterous Loco-Manipulation from Human-Object Interactions
链接: https://arxiv.org/abs/2609.16683
作者: Liu Cao,Xingze Wu,Jingzhi Cui,Botian Xu,Mingzhi Pei,Ruoqu Chen,Mengdi Xu
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 10 pages, 5 figures. Project website: this https URL
Abstract:Learning humanoid-object interaction requires coordinating whole-body balance, locomotion, and dexterous hand contact to control both robot and object motion. Human demonstrations provide examples of coordinated interaction, but transferring these behaviors to humanoid robots requires learning how to establish and maintain effective contacts under different embodiments and dynamics. We present Weave, a unified framework for learning whole-body dexterous humanoid-object interaction from captured human demonstrations. Weave first converts captured human-object interactions into executable robot-object references through contact-aware retargeting and approach-motion completion. At its core is a contact- and geometry-aware policy that jointly commands 29 body joints and 12 actuated finger joints across multiple objects and interaction sequences. Evaluation across nine objects yields a 92.5% success rate on trained interactions and, without any additional training, 65.0% on sequences never seen during training. We additionally release ~9,000 physically executed rollouts spanning ~23 hours, providing robot-object trajectories with contact annotations for downstream interaction-policy learning and physically consistent HOI motion generation. Project website: this https URL
[AI-56] m: An AI Agent for Industrial Process Optimization
链接: https://arxiv.org/abs/2609.16680
作者: Yongchao Ye,Xinyu He,Dutliff Boshoff,Way Kuo,Lishuai Li
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Manufacturing consumes one third of global energy and still has significant room for improvement in terms of energy efficiency. Optimal process control is essential for this purpose. However, synthesizing mathematical optimization models from messy, real-world industrial specifications requires bridging unstructured natural language and spatial diagrams with rigorous mathematical syntax. This poses a profound challenge for general-purpose Large Language Models (LLMs), which may introduce invalid constraints when tasked with modeling continuous multi-physics dynamics. To address this, we introduce little m, an AI agent designed to assist the formulation of industrial process control models. Combining a domain-specific knowledge repository with LLM-driven interaction, the proposed framework formulates real-world optimization problems as mathematical models. For systematic evaluation, we introduce the Industrial Process Control Benchmark (IPC-Bench), a novel multimodal dataset of 50 canonical scenarios requiring joint reasoning over text and process diagrams. Through comprehensive automated structural assessments and double-blind human evaluation, little m substantially outperforms state-of-the-art LLMs, generating semantically correct models. These evaluations assess formulation quality rather than solver feasibility, formal physical validity, or closed-loop industrial performance. The implementation of little m and the IPC-Bench dataset are available at this https URL.
[AI-57] AI for Games in the Foundation Model Era
链接: https://arxiv.org/abs/2609.16679
作者: Meng Luo,Yanlin Li,Hao Li,Hongzhan Lin,Pengfei Zhou,Tianjie Ju,Ran Zhang,Yeying Jin,Mong-Li Lee,Wynne Hsu
类目: Artificial Intelligence (cs.AI)
备注: 120 pages, 27 figures, 21 tables. Project page: this https URL
Abstract:Foundation models, alongside advances in learned game-world models, are reshaping AI across the game lifecycle. Beyond playing games, recent systems model players and game dynamics, support design and development, adapt player-facing experiences at runtime, and evaluate resulting artifacts. Yet these directions have evolved largely separately, obscuring which capabilities transfer across settings and which remain tied to particular games, engines, interfaces, or player populations. We organize the literature into six roles according to the immediate use of AI output: playing and acting; modeling players and games; designing games; building and maintaining games; generating and adapting at runtime; and testing and evaluating games. For each role, we examine what structure is supplied by the game or workflow, what AI learns or produces, which capabilities and artifacts transfer across settings and roles, and what evidence supports the claims. We identify cross-role connections: trajectories train world models, learned environments provide experience for agents, design specifications drive executable implementations, and play or testing feedback guides revision. However, control schemes, rules, engine interfaces, state representations, and player contexts often remain setting-specific, so downstream claims require validation in the target setting. Evaluation is most standardized for bounded game playing and selected learned environments, while persistent state in learned worlds, repeated software revision, validated player modeling, sustained runtime adaptation, and representative automated testing remain less established. The central challenge is to reuse or transfer outputs and capabilities across roles while re-establishing evidence for effectiveness in the game-specific contexts where they are used.
[AI-58] ANIMASK: What the Model Contributes to Role Play in Simulated Story Worlds
链接: https://arxiv.org/abs/2609.16667
作者: Xiucheng Zhang,Zhuoning Xu,Hanjun Luo,Yankai Chen,Hanan Salam,Xue Liu
类目: Artificial Intelligence (cs.AI)
备注: 38 pages, 6 figures, 14 tables
Abstract:When a language model plays a character, the observed behavior reflects both the assigned persona and the default dispositions of the actor model itself. Existing evaluations test persona fidelity or model defaults in isolation, but neither says, at a specific choice with consequences, what the persona changed and what the model’s default kept. We introduce ANIMASK, a simulation framework that freezes books and scripts into story worlds whose characters act on their own motivations and replays each story from its freeze point. We hold out the author’s continuation as a human reference, verify through in-story interviews that each persona remains present, and at every decision point compare the character’s action with what the model produces when the persona is removed. Across 40 stories, 6 actor models, and 3,846 decision points, the replays converge away from their canons in one shared direction, toward flatter, cooler stories that leave their tensions open. The personas stay present and obeyed throughout. On three choices in four the model’s default already falls inside what the persona accepts, and where the two diverge the model is the cautious one, holding where the persona would press. The persona guarantees who the character is, and the model sets how far the character will go.
[AI-59] Right Direction Wrong Step: Geometric Analysis of Finite-Step Failure in Looped Transformers
链接: https://arxiv.org/abs/2609.16665
作者: Zhihao Guo,Zonghan Wu,Haizhou Du,Huan Huo,Yilei Shao,Athanasios V. Vasilakos,Qingsong Wen
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Looped Transformers offer a parameter-efficient route to test-time scaling by reusing shared layers for iterative latent reasoning. However, additional iterations can reduce support for a reference answer, leaving unclear whether an update’s direction is locally unhelpful or its full displacement moves too far. We study this distinction by analysing reference utility, which measures this support, along the model’s own update direction, varying the fraction of the proposed displacement supplied to the readout. This reveals finite-step failures in which a locally improving direction produces a harmful full update. A pathwise curvature decomposition characterises how initial progress is lost, while a local quadratic model predicts full-step gains and useful step scales. Bounds based on accumulated curvature variation characterise the approximation error of these predictions. Experiments across two model families reveal this separation on mathematical and commonsense tasks. A fixed quarter step produces positive gains in reference utility for 72.2–83.2% of selected failures across four settings. These findings identify a mismatch between update direction and step scale as a mechanism of lost progress, explaining how some harmful updates retain useful computation.
[AI-60] ReDraft Dont Just Distill: Reference-Driven Revision for Continual VLLM Post-Training
链接: https://arxiv.org/abs/2609.16639
作者: Zhihao Zhang,Mingqi Wu,Qiaole Dong,Enyu Zhou,Shuo Li,Boyang Liu,Jiazheng Zhang,Honglin Guo,Xin Guo,Shaofan Liu,Junzhe Wang,Dingwei Zhu,Zhiheng Xi,Minlong Peng,Yuan Hua,Qi Zhang,Tao Gui,Xuanjing Huang
类目: Artificial Intelligence (cs.AI)
备注: 37pages, preprint
Abstract:Continual post-training of large multimodal models should add new capabilities while preserving those from pre-training, and the two goals pull in opposite directions. SFT gives explicit target supervision that learns a task from near-zero accuracy, but its off-policy targets move the model far enough to cause forgetting; on-policy methods such as RLVR and self-distillation preserve policy proximity yet supply little signal when the policy cannot yet solve the task. We introduce ReDraft (Reference-Driven Revision and Fine-Tuning), which obtains both from the model’s own failures: using an expert response only as a reference, it has the model revise its own incorrect rollout, keeps the revision only if a verifier accepts it, and fine-tunes on what survives. Each retained target is therefore explicit, yet still close to the current policy. Across Counting, Clock Reading, and Jigsaw on Qwen2.5-VL-3B/7B, two of them with near zero accuracy, ReDraft gains 56.9 points on the target task against SFT’s 52.9 while cutting prior-task loss from 16.6 to 1.5 points (11.3x less forgetting), and improves on OPSD along both axes (19.3 gain, 6.2 loss). Data- and parameter-space analyses match the design: revised targets are more probable under the base model, and the updates they induce stay compact and follow SFT’s direction more closely than OPSD’s. Repairing the model’s own output, rather than replacing it with an expert’s, is what lets one objective do both.
[AI-61] EchoPath: Execution-Level Replayable Memory for GUI Agents
链接: https://arxiv.org/abs/2609.16635
作者: Yao Zhao,Aditya Shanmugham,Swastik Roy,Yanxun Xu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Computer-use agents increasingly operate browsers, software, and desktop applications via CLI or API portals, but graphical user interface (GUI) still plays an important role in common industrial production scenarios. GUI agents commonly employ fresh observe-plan-ground-act loops, which is inefficient for enterprise tasks that repeatedly update records, process forms, configure tools, and export reports. We introduce EchoPath, a model-agnostic harness that converts artifact-validated GUI trajectories into standardized, parameter-controlled callable memories, analogous to Model Context Protocol (MCP)-style tool calls rather than unstructured experience records. Each memory stores task-intent keys, application and state preconditions, flexible input parameters, GUI evidence, validation provenance, and lifecycle state, so the host agent invokes a targeted procedure only when it can be deterministically replayed in the current runtime. The core mechanism enabling replay is an image-based target-reaiming algorithm that treats stored coordinates as visual evidence, matches the remembered GUI target against the current screen, and emits corrected operation coordinates before execution. During replay, EchoPath rebinds only declared modifiable inputs and rejects ambiguous or incompatible steps to bounded grounding repair or fresh planning. In experiments with real computer-use tasks, EchoPath reduced median token cost by more than 90% and median execution time by about 60%. These results support a bounded form of enterprise GUI memory: validated execution experience can become a controllable callable asset for recurrent work rather than only context for another reasoning pass.
[AI-62] Structure Across Voices: Comparing acoustic-event type accumulation and sequence dependence across four vocal repertoires using frozen audio encoders
链接: https://arxiv.org/abs/2609.16612
作者: Mudit Sinha,Sanika Chavan
类目: ound (cs.SD); Artificial Intelligence (cs.AI)
备注: 12 pages, 4 figures. Preprint
Abstract:Vocal repertoires can differ in acoustic-event type accumulation and temporal organization, yet direct comparison is difficult because corpora use different native events and unequal amounts of sequence. We compare sperm whale codas, human speech phones, Bengalese finch syllables, and common marmoset calls using the same frozen-audio-encoder procedure while matching event count and local sequence opportunity. Whale shows the fastest type accumulation; Finch shows the strongest immediate dependence and repeated-subsequence recurrence. Physically interpretable acoustics recover complementary parts of this profile, continuous analyses without clustering support broad Whale acoustic coverage, and source- and position-preserving nulls retain both Finch order effects. Extending predictive context shifts the comparison toward Whale. Thus repertoire differences depend on the acoustic property and temporal scale measured rather than forming a single hierarchy.
[AI-63] Large Language Models in the Loop: A Stability- and Network-Aware Survey in Networked Control Cyber-Physical and Multi-Agent Systems
链接: https://arxiv.org/abs/2609.16599
作者: Haiping Du,Linping Chan
类目: ystems and Control (eess.SY); Artificial Intelligence (cs.AI)
备注:
Abstract:Modern networked control systems (NCSs), cyber-physical systems (CPSs), and complex multi-agent network systems (CNSs) increasingly rely on large language models (LLMs) for high-level decision-making. However, the slow, stochastic nature of LLMs directly conflicts with the strict stability and safety guarantees required by these physical systems. This survey presents a unified analysis of how LLMs can be admitted into the control loop of NCS, CPS, and CNS without compromising closed-loop guarantees. We organize this around a core principle: the LLM operates as a slow supervisor adjusting high-level goals and constraints, while a fast, certified inner loop maintains physical stability. Under this framework, LLM integration maps directly to classical networked control challenges, where inference latency acts as delay, API failures as packet dropouts, tokenization as quantization, and hallucinations as bounded disturbances. We assess current developments across all these three domains, highlighting that rising model capabilities are frequently accompanied by a drop in formal safety assurances. Finally, we propose concrete future research directions, identifying the widespread lack of formal stability proofs as the field’s central open problem.
[AI-64] A Framework for Generating Valid Context-Specific Benchmarks through Expert Guidance EMNLP
链接: https://arxiv.org/abs/2609.16592
作者: Kimberly Le Truong,Nari Johnson,Anna Kawakami,Hoda Heidari
类目: Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注: Accepted to EMNLP Findings 2026
Abstract:This paper presents an end-to-end approach for generating context-specific large language model (LLM) benchmark datasets by combining expert input with synthetic data generation. Existing benchmark construction methods often trade off validity and scalability: datasets designed with domain experts can produce high-quality evaluations but are slow and costly to create, while synthetically generating data may scale efficiently but often results in unrealistic, redundant, or out-of-scope examples. To address this gap, we introduce a schema eliciting key information about the goals, scope, and context of an evaluation task, and use this information to guide synthetic data generation. We further define four criteria grounded in measurement validity for assessing dataset quality: coverage, diversity, content realism, and stylistic realism. Using these criteria, we show how expert-informed scaffolds can guide synthetic data generation toward more valid benchmarks. Through quantitative evaluations and a real-world case study with domain experts, we demonstrate that our approach improves benchmark data quality over existing methods while preserving validity. We additionally analyze how different types of schema information affect different dataset quality criteria, and provide practical guidance on which information to prioritize collecting under resource constraints.
[AI-65] Do LLM s Have Values? A Quantitative Analysis and Alignment Framework for Values in Large Language Models
链接: https://arxiv.org/abs/2609.16589
作者: Keqing Zhang,Jingyu Chen,Yufan Liu,Yongqiang Zhu,Nai Ding,Lai Jiang,Congyan Lang,Bing Li,Weiming Hu
类目: Artificial Intelligence (cs.AI)
备注: Preprint. 9 authors
Abstract:As Large Language Models (LLMs) increasingly handle complex subjective tasks, aligning their intentions and behaviors with human values has become a critical scientific challenge. However, current efforts are confounded by a striking behavioral paradox: they fluctuate unpredictably under minor wording changes (“swing”), yet stubbornly ignore explicit instructions to correct ingrained biases (“rigidity”). Resolving this duality is critical for reliable AI alignment. To systematically understand and safely steer these latent subjective preferences, our study is structured around three fundamental questions. First, do LLMs possess an intrinsic value system? By projecting responses from 106 LLMs (150,000 queries per model) and 95,000 human survey profiles into a shared sociological space, we empirically confirm that they do. However, they do not mirror human diversity, instead crystallizing into a highly concentrated, idealized value core. Second, how can these values be quantified? We propose the Prior-Environment-Cognition (PEC) framework. This model mathematically defines value expression as the joint outcome of inherent dispositions like parameter weights (Prior), external contexts such as user prompts (Environment), and internal reasoning processes like Chain-of-Thought (Cognition). Finally, how can LLMs’ values be aligned toward a desired target? Using PEC diagnostics, we establish an adaptive “Alignment Prescription”. Rather than blindly applying resource-intensive training, this method identifies the minimum effective intervention needed for each dimension, ranging from zero-cost prompts to targeted parameter updates. Extensive empirical validation confirms that our approach successfully verifies the presence of LLM values, accurately quantifies their shifts, and achieves more efficient and precise steering than conventional blind training, all without degrading general capabilities.
[AI-66] ProxiDex: Learning Dynamics-Guided Proximity Policy for Dexterous Manipulation
链接: https://arxiv.org/abs/2609.16586
作者: Yushan Bai,Boyu Zheng,Zhiyang Mao,Hongzheng Sun,Yuchuang Tong,En Li,Zhengtao Zhang
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: Accepted at the 10th Conference on Robot Learning (CoRL 2026). Project page: this https URL
Abstract:Multi-finger dexterous manipulation relies on stable hand-object interactions, yet these interactions are partially observable in practice. Visual observations are often occluded by the hand, tactile sensors introduce hardware-specific modalities and calibration burdens, and existing policies rarely model how these cues evolve under actions, making them brittle under contact uncertainty. To address these, we present ProxiDex, a dynamics-guided proximity policy framework that treats hand-object proximity as an interaction state for dexterous manipulation. ProxiDex reconstructs interaction point clouds and converts geometric distances into proximity cues, forming a hardware-agnostic contact representation that provides immersive feedback during VR teleoperation. Built on this representation, ProxiDex learns action-conditioned proximity dynamics with a coupled forward-inverse design: future observation latents are predicted from actions, while proximity variations are decoded from latent changes. Leveraging these dynamics, ProxiDex adaptively reweights proximity tokens across manipulation phases and uses dynamics-consistency supervision to guide policy inference, stabilizing action generation under unreliable visual feedback. Simulation and real-world experiments demonstrate improved success rates and robustness over representative baselines across standard, unseen objects, and perturbation scenarios. Additional visualizations are available at this https URL.
[AI-67] Query-Aware Source-Risk Triage for Retrieval-Augmented Generation
链接: https://arxiv.org/abs/2609.16564
作者: Kainan Zhou(Google LLC),Gangzhen Qian(Google LLC),Chuhong Xu(Sony Corporate of America),Lu Yi(Google LLC)
类目: Artificial Intelligence (cs.AI)
备注: 6 pages, 6 figures, 5 tables. Accepted at CAIT 2026
Abstract:Retrieval-augmented generation (RAG) pipelines may omit a source’s material relationship to the query. We study a pre-generation triage layer that treats this relationship as query dependent. The method routes canonical query families for enhanced review and assigns retrieved pages to pass, contextualize, exclude, or review. It combines a four-dimension page score, rank-discounted family aggregation, intent-preserving query mutations, and a family-held-out router. A single-coded pilot of 200 real URLs supplies provisional calibration anchors; a 20,000-row scenario with synthetic domain identifiers supports controlled workload analysis. An oracle page gate defines a risk-coverage target for a future learned classifier. The evaluation shows why page-level frequency cannot substitute for family-level exposure and quantifies how calibration changes scenario activation. Annotation reliability remains unmeasured, and synthetic rankings omit real retrieval dynamics. The result is an auditable triage method and validation plan, not an estimate of deployed review workload, live-Web prevalence, or downstream answer-quality gains.
[AI-68] he MAL Simulator: Cyber Operations Simulation based on Attack Defense Graphs
链接: https://arxiv.org/abs/2609.16563
作者: Jakob Nyberg,Sandor Berglund,Andrei Buhaiu,Joakim Loxdal,Pontus Johnson,Mathias Ekstedt
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:We have developed the MAL Simulator, a cyber operation simulator based on the Meta Attack Language (MAL). The MAL Simulator is intended for decision-driven cyber attack and defense simulations, for system analysis and the development of automated agents. By building the simulator around an attack modeling language, it can be adapted to different target domains without modifying the source code. We used the simulator for two case studies where we trained two types of agents for automated cyber operations: a defensive agent and an offensive agent. To ground the experiments, we base the models in data collected from an emulated network implemented in the cyber range CRATE. We found that the trained attacker policy could reach the designated targets more efficiently than the compared search methods, and that the trained defender agent induced lower costs than a naive heuristic agent under noisy alert conditions. When testing the RL attacker against the RL defender, we found that the performance of the defenders dropped significantly. This emphasizes the importance of cyber attack simulators to facilitate training both offensive and defensive agents. The MAL Simulator and associated tooling is publicly available and provides common interfaces for compatibility with existing machine learning frameworks.
[AI-69] QueryFormer: Winning Solution for KDD Cup 2026 Tencent UniRec Challenge
链接: https://arxiv.org/abs/2609.16548
作者: Yuanzhe Zhou,Zhaoyang Zeng
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Post-click conversion rate (pCVR) prediction requires jointly modeling feature interactions and sequential user behaviors. The KDD Cup 2026 Tencent UniRec Challenge calls for a unified architecture addressing both. We observe that existing unified architectures often generate query tokens—the central information hub—with projection-based multi-layer perceptrons (MLPs), without explicit token-to-query attention for refining the query side. We propose QueryFormer, centered on a stackable unified field–sequence block that bridges non-sequential multi-field features and behavioral sequences, and provide a latency-aware scaling study over view width H , model width, depth, data, and compute. The block generates queries through cross-attention and packs sequence queries into shared-parameter attention. QueryFormer secured 1st place in the Industrial Track, achieving an official test area under the ROC curve (AUC) of 0.83254; a modest post-competition scale-up reached 0.832713. Within our grid, H -scaling improves validation AUC from 0.84540 to 0.84615 and beats HyFormer at comparable budgets. Ablation identifies query generation as the largest contributor. Packed shared-parameter cross-attention keeps H=8 inference latency to only 1.89x that of H=1, positioning the bridge as an efficient stackable unified block.
[AI-70] A Cyber Range Evaluation of Autonomous Network Incident Response Agents
链接: https://arxiv.org/abs/2609.16541
作者: Jakob Nyberg,Teodor Sommestad,Andrei Buhaiu,Joakim Loxdal,Pontus Johnson,Mathias Ekstedt
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:We test the performance of agents for automated network intrusion response in a cyber range intended for human operator training. The range implements an emulated networking environment with a variable network topology, red-team emulation and simulated user agents. The goal of the defensive agents is to prevent hosts in the network from being accessed by the red-team agent, while minimizing the availability costs induced from defensive measures. Alerts are generated using a SIEM platform and mapped to a data modeling language used by the agents. We test a combination of heuristic agents and policies learned using reinforcement learning. The learned policies are optimized to minimize the combined cost using a cyber attack simulator modeling the network. We found that the reinforcement learning agents were overall more efficient at defending the system than the heuristic policy, and that the performance depends highly on the policy of the adversary in combination with the simulated users.
[AI-71] On the Importance of Gating: Memorization vs. In-Context Learning in State Space Models
链接: https://arxiv.org/abs/2609.16540
作者: William L. Tong,Aryo Lotfi,Emmanuel Abbe,Kostas Vaggelakos,Vishnu Banna,Etai Littwin,Josh Susskind,Cengiz Pehlevan,Eran Malach
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 25 pages, 5 figures
Abstract:State Space Models (SSMs) have emerged as a compelling alternative to Transformers, enabling sequence modeling with constant memory and linear compute. Although SSMs exhibit reasonable performance and favorable computational characteristics, they continue to lag behind Transformers on tasks that require in-context learning and precise retrieval, slowing their adoption for large-scale language modeling. In this work, we demonstrate that both the success and failure of SSMs in these domains can be explained by studying the role of the gating mechanism, a prevalent component in modern recurrent networks. Specifically, we show through theory and experiments that this gating mechanism causes SSMs to first learn an in-weights “memorization” solution, while delaying, or even preventing, convergence to a correct in-context learning solution. Importantly, this happens even in cases where there are no fundamental limitations due to the architecture or its memory capacity. On the other hand, we find that gating is often beneficial for improving generalization to long sequence lengths. Our results illuminate the crucial role of the gating mechanism in shaping both the training dynamics and generalization of SSMs, and provide a basis for understanding and improving linear-time models.
[AI-72] What Does Layer-Importance Reveal About Transformers and State-Space Models?
链接: https://arxiv.org/abs/2609.16537
作者: Istabrak Abbes,Nizar Islah,Irina Rish,Sarath Chandar
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Transformers and state-space models (SSMs) are the two dominant families of sequence models, and a central open question is how far the analytical knowledge built for transformers transfers to SSMs. We address this through the lens of layer importance which underpins compression, selective fine-tuning, and interpretability across both families. We decompose layer importance into two distinct notions. \emphNecessity captures how much the pretrained model depends on a layer’s existing contribution, measured by the loss increase from bypassing it. \emphPlasticity captures where the model absorbs new information during fine-tuning, measured by the magnitude of task-specific weight updates. Our analysis reveals that the two families behave fundamentally differently: in every evaluated residual transformer up to 14 B parameters, Necessity and Plasticity anti-align across depth, whereas in the evaluated Mamba-style SSMs they point to overlapping regions. The sign of this alignment also predicts downstream adaptation behavior. In the evaluated transformers, concentrating updates in the most plastic layers increases catastrophic forgetting, while this tier-dependent effect disappears in the evaluated Mamba-style SSMs.
[AI-73] AquiLLM : Evaluating Faithfulness in Open-Weight RAG -LLM Systems for Scientific Research
链接: https://arxiv.org/abs/2609.16519
作者: Bernie Boscoe,Srinath Saikrishnan,Vikram Seenivasan,Jack Stark,Andrew Lizarraga,Morgan Himes,Jonathan Soriano,PJ Allen,Tuan Do
类目: Artificial Intelligence (cs.AI)
备注: 14 pages, 1 figure
Abstract:Scientific research increasingly relies on large, heterogeneous data sources, motivating interest in retrieval-augmented generation (RAG) systems that provide natural language access to scientific knowledge and research workflows. Researchers are exploring the viability of these systems as natural language interfaces for document search and for generating analysis code and pipeline components. At the same time, concerns about data privacy and control over research infrastructure have motivated interest in open-weight models and open-source deployments hosted within research institutions. In astronomy, this development follows a long history of computational infrastructure development, from archival databases and SQL-based systems to LLM-assisted research tools. This paper presents a domain-expert evaluation of faithfulness for AquiLLM, an open-weight, offline RAG-LLM platform designed to support scientific research groups in the use and preservation of tacit and formal knowledge. We define faithfulness as the extent to which generated responses remain grounded in retrieved scientific context without unsupported claims or omissions. We report results from an astronomy case study evaluating AquiLLM across retrieval and scientific analysis tasks. AquiLLM performs most reliably on explicit retrieval-oriented questions grounded in the RAG collection, while faithfulness degrades for queries requiring synthesis or ambiguity resolution. These results highlight both the promise and limitations of open-weight RAG-LLM systems for scientific research and demonstrate the importance of domain-expert evaluation beyond standard benchmark leaderboards. Comments: 14 pages, 1 figure Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2609.16519 [cs.AI] (or arXiv:2609.16519v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2609.16519 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-74] Geospatial Metadata Improves Discoverability by Connecting Datasets Across Scientific Disciplines
链接: https://arxiv.org/abs/2609.16498
作者: Daniel Ebanks,Devika Jain
类目: Digital Libraries (cs.DL); Artificial Intelligence (cs.AI)
备注:
Abstract:Research data repositories are essential infrastructure for scientific inquiry and for ensuring that datasets follow FAIR (Findable, Accessible, Interoperable, and Reusable) principles. However, repository reuse depends on the quality and completeness of geospatial and thematic metadata, which researchers generally provide voluntarily. Given limited curation resources, it is unsurprising that even Harvard Dataverse, the world’s largest general-purpose research repository, contains many incomplete metadata records. Missing fields represent lost information and reduce interoperability. We find that datasets with more missing metadata receive fewer downstream citations and have fewer resolvable connections to other datasets. The implications are particularly important for geospatial datasets: only 0.3% of research datasets include a bounding box, and most represent archival points rather than complete geographic shapes. Our analysis shows that geospatial metadata helps connect concepts across disciplines. After embedding Harvard Dataverse datasets in a metadata knowledge graph, we find that datasets are twice as likely to connect across scientific disciplines through shared geospatial metadata as through keywords. This suggests that geographic metadata is a more reliable basis for cross-disciplinary interoperability than keyword vocabularies, which often remain discipline-specific. We train and fine-tune a small language model using datasets from Harvard Dataverse. Through geospatial metadata enrichment, we increase the share of datasets from different disciplines connected through metadata elements from 58.5% to 63.2%.
[AI-75] AI Policies: Help or Hindrance? A Software Developers Perspective
链接: https://arxiv.org/abs/2609.16496
作者: Samuel Ferino,Rashina Hoda,John Grundy,Christoph Treude,Hashini Gunatilake
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:AI policies introduced by software organisations to mitigate LLM-related risks such as sensitive information leaks and unauthorised usage are not useful if software developers do not engage with them. We draw on 19 software developer interviews to show how AI policies help and hinder developers. We suggest approaches to support managers and decision makers with a developer-centric approach to introducing AI policies.
[AI-76] From Manual Construction to AI-Driven Scenario Emergence: Rethinking Catastrophe Risk Modeling
链接: https://arxiv.org/abs/2609.16493
作者: Hang Gao
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Traditional catastrophe (CAT) risk models rely on costly manual construction to generate extreme weather scenarios, an approach largely unchanged since the 1990s. As climate extremes intensify, this creates mounting challenges to the entire risk transfer chain. This study proposes the TAISE framework, which repurposes AI weather forecasting models to produce coherent extreme weather sequences at a fraction of traditional costs. Through self-iterative generation, the framework produces continuous global atmospheric fields from which extreme events emerge. A proof-of-concept experiment demonstrates an order-of-magnitude reduction in computational cost compared with conventional methods, while capturing temporal continuity and cross-regional correlations absent in snapshot-based approaches. These findings suggest a pathway toward democratising catastrophe risk quantification and enabling dynamic, comprehensive portfolio assessment for insurers, reinsurers, ILS fund managers and public-sector risk managers.
[AI-77] Decoder Design Matters for ECG Delineation
链接: https://arxiv.org/abs/2609.16489
作者: Joseph Scharpf,William Han,Chaojing Duan,Michael A. Rosenberg,Emerson Liu,Ding Zhao
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 5 pages, 3 figures
Abstract:Electrocardiogram (ECG) delineation identifies the boundaries of P waves, QRS complexes, and T waves, providing structural annotations that can guide AI models in learning to interpret ECGs. However, training accurate delineation models requires manual annotations that are scarce and time-consuming to obtain. Recent work addresses this limitation through semi-supervised learning (SSL), but the design of the architecture, particularly the decoder, has received less attention. To this end, we propose R-U-Net, an ECG delineation model that pairs a ResNet-18 encoder with a U-Net decoder. On SemiSegECG, R-U-Net outperforms the strongest evaluated ResNet-18 + fully convolutional network (FCN) head baseline in each of the 16 in-domain settings by 3.3-13.0 mIoU and achieves 82.6 mIoU in the cross-domain setting, an improvement of 8.1 mIoU. Controlled ablations show that decoder design contributes more to performance gains than the evaluated SSL methods, motivating further exploration of architectures for ECG delineation. All code is open-source at this http URL.
[AI-78] Protocol-Preserving Context Trimming for Agent ic Workflows: Benefits Failure Regimes and Budget Guardrails
链接: https://arxiv.org/abs/2609.16461
作者: Harish Gaggar
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: 13 Pages, 4 Figures
Abstract:Agentic large language model (LLM) systems rely on long interaction histories to preserve instructions, tool states, intermediate decisions, and unresolved dependencies, but unrestricted context growth increases computational cost and can reduce efficiency. This study evaluates protocol-preserving context trimming as a reliability-constrained approach for multi-step agentic workflows. Five trimming strategies - recency-based, relevance-based, summarization, protocol-aware trimming, and adaptive budget guardrails - were compared across retained-context levels and workflow-complexity classes using task success, protocol adherence, valid tool calls, token savings, latency reduction, cascading failures, and critical context thresholds. Conventional strategies achieved about 60% mean token savings but lower task success (66.6-77.3%) and protocol adherence (85.5-88.6%). Protocol-aware trimming improved task success to 92.2%, while adaptive guardrails achieved 96.0% task success, 96.3% protocol adherence, and 1.0% cascading failure with 56.0% mean token savings. Retained-context budgets of 25% or less increased failure odds 10.92-fold relative to budgets of 50% or more (p 0.001). Protocol-aware trimming produced 5.24-fold greater odds of successful completion than conventional methods under aggressive budgets, while adaptive guardrails further increased success odds 2.11-fold versus fixed protocol-aware trimming (p 0.001). Critical context thresholds also increased with workflow complexity. These findings indicate that reliable context reduction depends more on preserving protocol-critical state than on maximizing token removal, and that adaptive guardrails can improve efficiency, scalability, and reliability in long-horizon agentic systems.
[AI-79] Fine-Tuning Fixes Mode Collapse and Over-Dispersion in LLM s
链接: https://arxiv.org/abs/2609.16454
作者: Kirill Skobelev,Eric Fithian,X.Y. Han
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Recent work by Doshi and Hauser (2024), Bisbee et al. (2024), and Xie et al. (2026) raises concerns that outputs from large language models (LLMs) tend to be under-diverse: they repeat or resemble one another more often than responses from the population they are meant to represent, a phenomenon known as mode collapse. In this work, we show that whether mode-collapse, or its opposite, occurs depends on the specific model and dataset used. Further, with sufficient supervised fine-tuning (SFT) data, LLM output diversity converges toward that of the target distribution from which fine-tuning data are sampled. To quantify this comparison, we measure the probability that two responses sampled independently from the same fixed prompt coincide (collide), or their expected similarity under a kernel. We derive a bias-variance decomposition of the expected gap between the model’s and target’s collision probabilities, showing that SFT is not inherently biased toward mode collapse or its opposite: finite-sample SFT can leave a model either under- or over-dispersed, depending on the model and dataset. Finally, we show that the absolute gap is bounded by the square root of the Kullback-Leibler (KL) divergence from the target distribution to the model. Consequently, a model sufficiently close to optimal under population cross-entropy cannot exhibit arbitrarily miscalibrated diversity. We test the decomposition and the bound in three experiments: small transformers on synthetic languages, four LLMs fine-tuned on human surveys, and these LLMs fine-tuned on CodeNet, a dataset of human code solutions. More target data moves model diversity toward the human (or synthetic target) level in all experiments, consistent with our theoretical predictions. These results show that diversity miscalibration can arise from finite-sample error and shrink as SFT better approximates the target distribution.
[AI-80] How Good Are Time-Series Foundation Models for Pedestrian Crowd Count Forecasting? A Cross-Dataset Comparative Study ITSC2026
链接: https://arxiv.org/abs/2609.16415
作者: Theivaprakasham Hari,Ziteng Li,Yanan Xin,Winnie Daamen,Serge Hoogendoorn
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 9 pages, 3 Figures, submitted to The IEEE International Conference on Intelligent Transportation Systems (IEEE ITSC 2026)
Abstract:Pedestrian-count forecasting supports pedestrian-oriented Intelligent Transportation Systems (ITS), including crowd monitoring, pedestrian-traffic staffing and routing, and proactive risk mitigation during surges. Recent time-series foundation models (FMs) report strong zero-shot accuracy on heterogeneous forecasting benchmarks, but it remains unclear whether these gains transfer reliably to pedestrian sensing deployments. We benchmark seven univariate forecasting approaches spanning four paradigms: Seasonal Naive, gradient-boosted trees (LightGBM, CatBoost), deep learning models (N-HiTS, PatchTST), and two pretrained FMs (TimesFM, Chronos-2). Experiments cover two complementary regimes: (i) a five-day special event dataset SAIL2025 at 3-minute resolution with limited in-domain history; and (ii) Melbourne pedestrian sensors as a multi-year hourly dataset (2010–2017) with strong seasonality. We compare the MAE and RMSE results per sensor across datasets and multiple forecast horizons. Results show three consistent findings. First, with limited historical data, Seasonal Naive remains a strong baseline for long-horizon forecasting on high-volume sensors, while trained models can degrade when the next day differs substantially from prior days. Second, boosted trees can be competitive on lower-volume sensors but exhibit higher sensitivity on high-volume sensors under event-driven shift. Third, FMs excel in the seasonal and data-rich regime under long-context configuration. The findings highlight the importance of choosing pedestrian forecasting models based on both the underlying data conditions and the forecasting horizon.
[AI-81] On the Expressive Power of Implicit Line-Graph Higher-Order Weisfeiler–Leman
链接: https://arxiv.org/abs/2609.16412
作者: Fan Yang
类目: ocial and Information Networks (cs.SI); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 9 pages of main text. 40 pages in total
Abstract:Whitney’s theorem allows isomorphism testing for connected simple graphs, apart from K_3 and K_1,3 , to be formulated as distinguishing their line graphs. However, the relation between fixed-dimensional Weisfeiler–Leman (WL) expressivity on line graphs and on their roots remains unresolved. We study this relation through Implicit Line-Graph WL (ILG- k -WL), which is exactly k -WL on L(G) , executed over the edges of G with line-graph relations derived from endpoint incidence and without explicitly constructing L(G) . On the Whitney-general class, the relation between root-domain and line-graph WL depends on k . For k=1,2 , ILG- k -WL adds no distinguishing power beyond root-domain 1 -WL and misses some pairs that 1 -WL separates. For k=3 , we prove the backward containment L(G)\equiv_3\text-WLL(H)\Rightarrow G\equiv_3\text-WLH . Strongly regular witness pairs, including the Shrikhande/rook pair, show that ILG- 3 -WL is strictly more expressive than 3 -WL. The backward containment also extends to disconnected graphs with no isolated vertices when every connected component is Whitney-general. Deterministic ILG- 3 -WL separates all three substructure-counting witness pairs, all 105 pairs in SR25, and 359 of 400 BREC pairs. An untrained dense ILG- 3 -GNN gives the same pairwise verdicts on these evaluations.
[AI-82] UDAV: Uncertainty-Driven Adaptive VLM Waypoint Planner
链接: https://arxiv.org/abs/2609.16368
作者: Ghazal Farhani,Shabnam Shabani
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Vision-language models (VLMs) can generate routes directly from aerial imagery for off-road navigation, but their predictions provide no indication of reliability. We present UDAV, an Uncertainty-Driven Adaptive VLM Waypoint Planner for UAV-guided UGV navigation. UDAV draws multiple stochastic trajectory predictions, selects their medoid as a self-consistent nominal route, and estimates predictive uncertainty from their spatial dispersion. When the maximum uncertainty across interior waypoints exceeds a threshold, UDAV invokes a reconsideration stage; otherwise, it returns the medoid directly. We evaluate UDAV on 400 held-out trajectory queries from two UAV flights. Stochastic medoid selection reduces the mean average displacement error (ADE) from 147.4 pixels for a deterministic prediction to 115.9 pixels. The complete planner achieves a mean ADE of 110.4 pixels, a 25.1% reduction relative to deterministic planning, while producing valid trajectories for all queries. UDAV also yields the lowest 90th- and 95th-percentile errors among all evaluated configurations, including a higher-budget K=10 consensus baseline. Relative to the K=5 medoid, UDAV reduces these errors from 225.3 and 326.0 pixels to 199.0 and 290.8 pixels, respectively. These results demonstrate that stochastic VLM predictions provide both a stronger nominal route and an actionable uncertainty signal for selectively mitigating large planning errors.
[AI-83] Breaking the 1.58-bit Barrier for Ternary LLM s
链接: https://arxiv.org/abs/2609.16338
作者: Evangelos Georganas,Alexander Heinecke,Pradeep Dubey
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Ternary Large Language Models (LLM) store every weight as one of three symbols -1,0,+1\ , so the cost of a ternary model is conventionally referenced to the information-theoretic \log_2 3 \approx 1.585 bits per weight. The prevailing deployment format packs five ternary weights into one byte (five-trit packing), and due to the power-of-two group sizes used in practice this rounds up to 1.625 bits per weight. This effective storage bit-width treats the three symbols -1,0,+1\ as equiprobable. We measure the actual symbol distribution of 29 ternary LLM models and find that zeros account for up to 51.5% of all weights. Motivated by this finding, we introduce BITCOS, a simple distribution-adaptive layout comprised of a dense presence bitmap plus a compacted sign vector, and costs 2 - z bits per weight element given a zero density z in the model’s weights. BITCOS stores weights more compactly than the five-trit packing in 26 of the 29 tested models, and reaches 1.485 bits per weight on the sparsest of them. BITCOS is amenable to efficient unpacking on modern processors and GPUs, and we present optimized unpacking sequences for AVX-512, AVX2 and Intel Xe2 GPUs. Measured against production state-of-the-art ternary matrix-vector multiplication kernels, at the zero densities real-world ternary models exhibit, the realized gain with our proposed layout is up to 1.28\times . Finally, we illustrate end-to-end LLM inference results on 5 different platforms (client and server CPUs, integrated and discrete Xe2 GPUs) where decode throughput improves by up to 1.18\times on CPUs and 1.27\times on GPUs.
[AI-84] Cross-Anatomy Transfer Versus Sparse Interpolation in Digital-Twin-Oriented Aortic Fluid-Structure Interaction Surrogates
链接: https://arxiv.org/abs/2609.16322
作者: Ali Nourbakhsh,Mohammad Reza Niroomand,Erfan Nourbakhsh
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 8 pages, 6 figures, Under review at ICBME 2026
Abstract:Surrogate credibility for fluid-structure interac- tion (FSI) requires distinguishing transfer across independent anatomies from interpolation within an already sampled surface. Four de-identified human aortic models from the Vascular Model Repository were reconstructed into separate lumen and nominal 1.5-mm wall domains and analyzed under matched first-cycle two-way FSI. A geometry-only LightGBM prior, selected by leave-one-anatomy-out development on three anatomies, was zero-shot evaluated on a fourth, then probed with a post-zero- shot sparse field-completion case study over six targets. Zero-shot transfer was poor across all targets. At a five-percent anchor level (203 anchors, 3,852 evaluation nodes), prior-plus-adaptation reached an oscillatory shear index (OSI) R2 of 0.603. However, same-anchor controls tuned only on the three development anatomies were stronger for several outcomes: inverse-distance weighting reached R2 = 0.829 (OSI), 0.617 (peak von Mises stress), 0.676 (mean stress); radial basis function interpolation reached 0.917, 0.714, 0.778. Sparse within-anatomy labels thus support field completion, but this four-anatomy cohort gives no evidence the cross-anatomy prior adds value beyond direct interpolation. We frame this as a first computational stage toward a measurement-linked digital twin: the surrogate/update layer is evaluated here, while larger cohorts, converged FSI, measurable patient-side inputs, and physics-informed learning remain future work, not a claim of a complete clinical twin. Our code, data and computation files are available at https://github. com/ali-nourbakhsh2005/Aortic-FSI-Sparse-Field-Completion
[AI-85] FairLint-DL: An IDE-Native Tool for Fairness Debugging of Deep Learning Software
链接: https://arxiv.org/abs/2609.16321
作者: Archit Rathod,Saeid Tizpaz-Niari
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: In the Tools and Datasets track of the 41st IEEE/ACM International Conference on Automated Software Engineering (ASE 2026). Artifact awarded Available, Functional and Reusable badges. Tool: this https URL
Abstract:Existing fairness analysis tools predominantly operate as post-training evaluation frameworks, requiring practitioners to complete the full model development lifecycle before assessing bias. We present FairLint-DL, a Visual Studio Code extension that implements a shift-left approach to fairness testing by enabling pre-training, IDE-native bias detection directly on tabular datasets. FairLint-DL trains a configurable deep neural network as a proxy model and applies information-theoretic Quantitative Individual Discrimination (QID) metrics. Grounded in Shannon and min-entropy, QID quantifies the causal influence of protected attributes on predictions. The system implements a two-phase gradient-guided search algorithm for discovering discriminatory instances, a causal debugging pipeline that localizes bias to specific network layers and neurons via sensitivity analysis, and dual explainability engines using SHAP and LIME for feature-level attribution. Evaluation on three tabular benchmarks (Adult Census Income, German Credit, and Bank Marketing) reveals fairness concerns that vary widely across datasets: on Adult, 96.0% of analyzed instances exhibit QID above the 0.1-bit significance threshold, with a mean QID of 0.619 bits and a disparate impact ratio of 0.581, violating the four-fifths legal rule. FairLint-DL produces these results within 12 seconds on cached models, demonstrating the feasibility of integrating fairness analysis into the developer workflow without significant overhead.
[AI-86] Cognitive Admission Control: Risk-Conditioned Assurance for Consequential Actions in Agent ic Distributed Systems
链接: https://arxiv.org/abs/2609.16313
作者: Jun He,Deying Yu
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
备注: 15 pages, 1 figure, 2 tables; includes formal proofs, obligation catalogue, and empirical local evaluation
Abstract:In agentic distributed systems, an agent may be authorized to mutate external infrastructure while lacking evidence that the mutation is ready to execute. Cognitive Admission Control (CAC) makes this evidence requirement explicit. A policy maps a typed action and its modeled risk to assurance obligations specifying predicates, evidence classes, scope, freshness, and witness-set constraints. A deterministic evaluator distinguishes satisfied, violated, and unresolved obligations; unresolved conditions produce targeted evidence-acquisition requests. Successful admission produces a certificate binding the action, its witness manifest, and dispatch-time guards. We formalize the admission calculus and the assumptions connecting it to mediated execution. The guarantees are policy-relative: physical safety additionally requires sound evidence, an adequate environment model, and preservation of relevant conditions through the effect. A TypeScript prototype is evaluated in 2,730 controlled local trials with independent effect observation and matched fault schedules. Across 390 CAC trials, 120 effects complete without modeled harm and no harmful effects occur. A live-policy baseline achieves the same completion count but admits the constructed correlated-witness failure. Mechanism ablations isolate guard, evidence-class, structural-cut, and remediation behavior. A further 9,000 measurements exercise the complete local dispatch path with persistent replay protection. These results establish tested implementation behaviors and local costs, not production failure rates or comparisons of language-model capability. Comments: 15 pages, 1 figure, 2 tables; includes formal proofs, obligation catalogue, and empirical local evaluation Subjects: Distributed, Parallel, and Cluster Computing (cs.DC); Artificial Intelligence (cs.AI); Software Engineering (cs.SE) ACMclasses: C.2.4; D.4.6; D.2.4; I.2.11 Reportnumber: OpenKedge-TR-2026-CAC-01 Cite as: arXiv:2609.16313 [cs.DC] (or arXiv:2609.16313v1 [cs.DC] for this version) https://doi.org/10.48550/arXiv.2609.16313 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-87] Assurance Envelopes for Autonomous Coding Agents : Minimum-Cost Evidence for Software Change
链接: https://arxiv.org/abs/2609.16302
作者: Anjan Goswami
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Programming Languages (cs.PL)
备注: 15 pages, 4 figures, 3 tables
Abstract:When a coding agent returns to existing software, it inherits evidence from earlier engineering work: tests, type checks, proofs, static analyses, and traces. Reloading all of it is wasteful, but dropping a piece the change depends on can leave a required property unsupported. Given the properties a change must preserve, its obligations, we ask which least-cost subset of the available evidence re-establishes them, and we call such a subset a task-conditioned assurance envelope. Evidence and the rules that combine it form a typed inference graph; an obligation is met when forward chaining from the selected evidence reaches it, and we validate every selection by that closure rather than by trusting the optimizer. The software-derived graphs in our evaluation come from preserved outcomes of prior AI coding-agent runs; we freeze those artifacts and ask which accumulated evidence should be restored for a later task. Small graphs from Rust, IronBlocks, and Pong outcomes show that the minimum envelope depends on the task, that none may exist when current evidence cannot re-establish a required property, that some properties need several pieces of evidence together, and that expanding the requirements adds evidence rather than replacing it. A prespecified synthetic benchmark of 249 instances characterizes computation: a baseline that discards the ‘several pieces together’ structure necessarily fails to re-derive them; every completed exact cross-check agreed with the CP-SAT optimizer; and median solve time stayed below 20 ms at 500-evidence graphs, except that graphs with many alternative derivations per target timed out at far smaller sizes, so structure, not raw size, drives difficulty. The contribution is a bounded application of established optimization to selecting assurance context for a software change; discovering the obligations and downstream agent benefit remain open.
[AI-88] Closing the Loop: Branch-and-Bound for Scalable Verification of Nonlinear Neural Feedback Systems
链接: https://arxiv.org/abs/2609.16298
作者: I. Samuel Akinwande,Mykel J. Kochenderfer,Clark Barrett
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Despite recent advances in the verification of nonlinear neural feedback systems, scalability remains the central obstacle, as state-of-the-art solvers do not yet handle the network sizes and nonlinear dynamics of autonomy applications. Combinatorial solvers do not scale to large networks, whereas propagative solvers excessively sacrifice precision. This work seeks to improve the scalability of combinatorial solvers by formulating verification as branch-and-bound on an abstraction of the closed-loop system. We introduce \rail, an interface that exposes polyhedral enclosures of the dynamics to LiRPA-style bound propagation, and \clipper, a branch-and-bound algorithm that jointly refines enclosures and splits controller activations. This framework enables joint reasoning on the computational graph of the closed-loop system, preserving symbolic correlations across time steps. We present our construction and show that it yields significant improvements over the state of the art.
[AI-89] Symmetric solution of the Bellm an optimality equation for repeated harmony game
链接: https://arxiv.org/abs/2609.16289
作者: Hisato Komatsu
类目: Computer Science and Game Theory (cs.GT); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 22 pages, 2 figures
Abstract:In social dilemma games, additional rewards or punishments have been studied as means of promoting cooperation. Therefore, it is important to investigate the ideal situation, in which such an additional payoff would change the game. In this study, we investigated the symmetric solution of the Bellman optimality equation for a repeated harmony game. The calculations showed that three types of symmetric solutions exist. One of them corresponds to the trivial All-C strategy, and another to the Win-stay Lose-shift strategy of the prisoners dilemma game. The nontrivial behavior of the strategy corresponding to the last solution is also discussed in detail. In addition, we numerically investigated which strategy the agents actually learn by the reinforcement learning algorithm.
[AI-90] Mapping U.S. Federal AI Governance Against Sector Vulnerability
链接: https://arxiv.org/abs/2609.16260
作者: Ho Ting Hung,Angelica Chowdhury,James Teague,Simon Mylius,Spencer Michaels,Peter Slattery,Alexander Saeri,Neil Thompson
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
备注:
Abstract:Artificial intelligence (AI) poses different levels of risk across sectors, but are these differences reflected in U.S. federal AI governance? To help answer this question, we assess 684 federal AI governance documents for their coverage of 14 sectors and 24 AI risks. We measure coverage as breadth (i.e., how frequently the risk or sector is addressed across documents) and depth (i.e., how substantively the risk or sector is discussed). We then compare sector coverage patterns for each of the 24 risks with vulnerability assessments from a Delphi study of 272 experts. Our analysis finds substantial variation in coverage: AI risks related to robustness, system security, and governance receive more attention than socioeconomic, environmental, and emerging risks, including multi-agent risks. Public administration, national security, information, and scientific services receive comparatively high levels of coverage relative to other sectors, such as finance and healthcare, which experts rate as highly vulnerable to AI risks. By mapping current coverage and identifying where it differs from expert assessments of vulnerability, we surface potential AI governance gaps which may help inform AI risk-related decisions across government and industry.
[AI-91] he AI-Enabled Scientific Frontier
链接: https://arxiv.org/abs/2609.16258
作者: Gabriel Manso,Emma Fu,Neil Thompson
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Performance (cs.PF); General Economics (econ.GN)
备注:
Abstract:As artificial intelligence’s capabilities improve, it is increasingly viewed as a general scientific method. But how true are these claims? Does AI outperform all techniques, or only some, and how is this changing? To assess the claims, we assemble a corpus of 2,507 head-to-head comparisons between AI and other scientific analysis techniques across 27 scientific disciplines from papers published between 2000 and early 2025. We find a profound dichotomy. Relative to traditional statistics, AI often outperforms, but at a significantly higher computational cost. But there are also nearly a quarter of cases where AI is both more expensive and performs worse than traditional statistical techniques and this fraction has been stable for a decade. Relative to scientific computing, AI often underperforms, but at lower computational cost. This has begun to change: since 2020, AI’s performance against scientific computing has notably strengthened and it now outperforms on more than half of comparisons. These patterns suggest that AI is therefore not a universal replacement for existing methods, but rather a valuable – and improving – part of a new AI-enabled scientific frontier.
[AI-92] Models as Governed Interfaces for AI-Native MBSE: Read-Side Adequacy and Write-Side Admissibility
链接: https://arxiv.org/abs/2609.16252
作者: Jason Gower,Michael J. de C. Henshaw,Siyuan Ji
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: Accepted by the 29th International Conference on Model Driven Engineering Languages and Systems (MODELS), 4-9 October 2026, Malaga, Spain, as a New Ideas and Emerging Results (NIER) paper
Abstract:Machine-readable models such as SysML v2 are now programmatically accessible, and a growing body of work treats that access as the enabling condition for AI participation in systems engineering. Access is necessary, but not sufficient. The remaining work lies not in the modelling language but in the data architecture around it. An AI reader that queries a structurally complete model for a derivation still runs into absent derivation chains, untagged epistemic status, missing provenance, and evidence that the model cannot resolve. Faced with these gaps, it does not abstain; it fills them from training data, a source that is neither verifiable nor governed. To make the case on a model that is exemplary by current practice rather than deficient, we probe the public Apollo 11 SysML v2 reconstruction. We name the missing property epistemic adequacy and offer it as a candidate data-architecture pattern in two halves. Read-side adequacy lets derivation, status, and provenance answer a query rather than invite a guess; write-side admissibility gates an AI contribution before it enters the record. The property is broken down into five criteria. Four sit on the read side, evidenced by the case and convergent literature; the fifth sits on the participation side, advanced as a hypothesis this paper does not yet test. The architecture space runs from an inline metadata extension up to a substrate-native multi-model store, and over it, we propose the Governed-Query Architecture Framework, which governs agent participation through the viewpoint conventions that engineers already use. We commit the reframing to falsification: the epistemic layer counts as refuted if it cannot beat a retrieval-augmented baseline on the same model, tested first on the Apollo chain and then in an industrial pilot.
[AI-93] CADWorld: Computer-Use Benchmark for Long-Horizon Computer-Aided Design
链接: https://arxiv.org/abs/2609.16251
作者: Zihan Dong,Yuanzhe Liu,Zhiyuan Ma,Qishi Zhan,Dehan Kong,Guohao Li,Kaixin Li
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Computer-use agents are increasingly evaluated in realistic desktop environments, but existing benchmarks provide limited coverage of professional engineering workflows whose outputs are persistent, structured artifacts. Mechanical computer-aided design (CAD) is a particularly demanding setting: an agent must manipulate geometry and constraints over long interaction horizons while producing a native project whose dimensions, construction structure, and downstream engineering state remain valid. We introduce \textbfCADWorld, a benchmark for long-horizon computer use in FreeCAD. CADWorld contains 200 tasks spanning 11 mechanical-CAD workflow categories, including sketching, part modeling, assembly, CAM, FEM, measurement, mesh processing, and technical drawing. Agents operate through screenshots and GUI actions, while success is determined by task-specific executable checks over saved FreeCAD artifacts and auxiliary outputs, covering geometric properties, parametric structure, constraints, manufacturing state, and simulation results. Across seven current agents on the full benchmark, the strongest agent achieves 17.5% success, compared with an 87.0% expert reference pass. We find that weaker agents often fail before producing a valid artifact, whereas stronger agents increasingly fail on structural, geometric, and construction-process requirements. CADWorld therefore exposes a gap between general GUI competence and reliable execution of persistent, verifiable engineering workflows. Project accessible at this https URL.
[AI-94] he Pain Axis: LLM s Represent Self-Directed Harm and Act to Relieve It
链接: https://arxiv.org/abs/2609.16247
作者: Valen Tagliabue,Leonard Dung,Cameron Berg
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models sometimes behave in ways resembling human emotional responses, and recent work has identified internal representations that may explain this. We ask whether LLMs represent pain distinctly from fear, sadness, and generic negative valence, and whether this representation functions as pain would be expected to. We build a dataset describing painful situations across five categories: physical, psychological, social, moral, and cognitive. These are paired with controls for fear, negative emotion, negative world states, sadness, non-painful bodily sensation, arousal, numbness, and neutral content. Using denoised difference-in-means, we extract a linear pain direction from 25 open-weight models across five families, ranging from 2B to 72B parameters. We find that this direction separates pain from matched controls in base and instruction-tuned models, is nearly orthogonal to fear and negative valence, and promotes pain-related vocabulary through the unembedding matrix. We then test its functional properties. First, the direction responds to harm targeting the model but not suffering observed in the user; fear and negative-emotion directions show the opposite pattern. Second, adding the pain-direction vector to the model’s residual-stream activations during generation produces a consistent progression from vague discomfort to first-person expressions of worthlessness and failure. Third, steered, fine-tuned Qwen 2.5 models choose a pain-relief button even when it worsens their next answer or harms the user. They press it again far less often when the button removes the steering vector than when it does not, even though the models are never told whether the vector is injected or removed. We discuss the implications of these findings for AI safety and welfare.
[AI-95] Metacognitive Steering: Learning the Structure of Scientific Judgment
链接: https://arxiv.org/abs/2609.16245
作者: Vincent Karpf,Joseph Reth,Eike Gerhardt,Audrey Wang,Anna Butz,Jiehao Xing,Jialing Song,Larry Callahan
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Long-horizon scientific discovery requires agents to alternate between exploration, disciplined execution, and critical reassessment as evidence changes. Current language models are trained primarily on the products of science and optimized using outcome-level signals, providing limited supervision for these process-level shifts in scientific judgment. We investigate whether such judgment can be recovered from scientist interaction traces and used to control the internal computation of a frozen frontier model. Using contrastive interventions collected during real scientific research, we identify a coordinated, low-dimensional control structure within Kimi 2.6, a trillion-parameter mixture-of-experts model. Residual analysis, attention-weight subspace alignment, and cross-layer singular value decomposition converge on a mid-depth control surface spanning key layers. We introduce Metacognitive Steering, an inference-time controller that reads the model’s cognitive regime and dynamically composes layer-specific interventions for exploration, procedural convergence, or critical reassessment without modifying model parameters. Behavioral analyses show that this control produces more sustained exploration, explicit pruning, and evidence-responsive synthesis. We operationalize the method in Columbus-1, an autonomous research system that identified eight independently reproduced, attacker-reachable vulnerabilities in BlueZ and directed the design, simulation, and fabrication of a ten-foot rocket intended to land propulsively using non-throttleable solid motors. Together, these results show that process-level scientific judgment can provide supervision for interpretable, dynamic control over a model’s reasoning strategy.
[AI-96] oward Governance-Aware Autonomous GIS: A Narrative Review of Ethical and Privacy Risks in LLM -Enabled GeoAI
链接: https://arxiv.org/abs/2609.16232
作者: Maya Subramanian,Devika Jain
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Geospatial artificial intelligence (GeoAI) powered by large language models (LLMs) is expanding the capacity to query, generate, and interpret spatial information through natural-language interfaces and agentic autonomous GIS workflows. This capability creates governance challenges that general AI ethics discussions do not fully capture, including passive location inference from mobility traces, spatially structured bias amplification driven by spatial autocorrelation and scale effects, hallucinated spatial facts, and uncertainty compounding across multimodal geospatial inputs. This narrative review identifies eight recurring issues in LLM-enabled GeoAI: data provenance and consent, spatial privacy and inference risk, algorithmic bias and spatial inequity, spatial mechanisms as structural risk (spatial autocorrelation, the modifiable areal unit problem, and scale effects), LLM-specific technical risks, explainability, policy and regulatory gaps, and public enablement and workforce development. For each issue, we characterize the underlying mechanism, ground it in an illustrative example from the literature, and assess the current state of technical or institutional responses, ranging from largely unaddressed to actively debated or subject to emerging policy. Building on this synthesis, we propose a governance-aware architecture for LLM-enabled autonomous GIS that maps each issue to enforceable controls and auditable artifacts across the geospatial data lifecycle, illustrated through a worked flood-response routing scenario. The review highlights a persistent evidence gap: proposed responses remain largely conceptual, and field-tested evaluations of governance controls for LLM-enabled GeoAI remain limited. We close by outlining a research agenda emphasizing empirical validation, spatially specific interpretability tools, and workforce training aligned with these emerging risks.
[AI-97] Where Should the KV Cache Live? Placement Policies Across GPU CPU and SSD for Long-Lived Sessions
链接: https://arxiv.org/abs/2609.16215
作者: Srikanta Datta Tumkur,Jay Iyer,Mehar Simhadri,Sai Pavan Kumar,Sai Kapil Kumar,Ramesh Nampelly
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:GPU high bandwidth memory is scarce and expensive, and KV caches consume much of it as chats, agent loops, and document question answering accumulate state. Systems such as Mooncake, LMCache, FlexGen, InfiniGen, and AttentionStore extend GPU memory with CPU DRAM and SSD. The harder question is which blocks belong in each tier, when to move or evict them, and whether prefetching helps. We study these choices in a discrete event simulator spanning GPU HBM, CPU DRAM, and SSD, calibrated against a random forest execution time predictor. We compare recency, reuse frequency, predicted reuse, and an EWMA predictor with prefetch lookahead across chat, agent, and document question answering workloads. Tiering supports 73.02 times more concurrent sessions per GPU and lowers cost per session by 62.04 times. These gains come from tier capacities of 1 plus 8 plus 64, not placement policy. Decode is compute bound at batch size one in our setup, so placement barely affects throughput. It mainly changes PCIe migration traffic and time to first token. Recency produces 2.30 times less migration traffic than reuse frequency for chat. Reuse frequency performs best for agents and document question answering. The existing predicted reuse policy is byte identical to recency, making its agent recommendation effectively recency. A genuine EWMA predictor changes behavior but still ranks behind reuse frequency on the workloads prediction was expected to help. Prefetching does not justify its bandwidth cost. Across the policy and cache size grid, even an oracle with knowledge of future requests never beats no prefetch on migration traffic. Workload specific placement can reduce data movement, but the predicted reuse and prefetch recommendations are not supported as implemented.
[AI-98] Artificial intelligence and biosecurity: capabilities threat pathways and defense-in-depth governance
链接: https://arxiv.org/abs/2609.16213
作者: Candace S.Y. Chan,Aris Karatzikos,Ilias Georgakopoulos-Soares
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Artificial intelligence is reshaping biological research across an increasingly connected digital-to-physical workflow. General-purpose large language models can retrieve and integrate scientific information, support experimental planning, and computational analysis; biological foundation models can predict, optimize, and generate proteins, genes, and genome-scale sequences; agentic systems can coordinate multistep research tasks; automated laboratories can partially close the design-build-test-learn cycle. These technologies could greatly benefit medicine, public health, and biotechnology. However, their biosecurity risk depends not only on what the AI can do, but also on who uses it, their expertise and intent, their access to laboratory tools and materials, and the safeguards in place. Current evidence shows that AI uplift exists but primarily affects digital rather than physical tasks. Frontier systems have exceeded expert baselines on in-silico, and screening-evasion benchmarks, whereas controlled wet-laboratory studies find that tacit knowledge and physical execution remain substantial barriers. This review describes the different biological threats from AI tool use, from information gathering and biological design to procurement, synthesis, testing, scale-up, and potential release. We further examine why alignment techniques for general-purpose models transfer poorly to biological ones, and the emerging role of interpretability in auditing whether hazardous capabilities are genuinely removed. We argue for defense-in-depth governance that links capability thresholds to proportionate responsibilities across the biological AI ecosystem, reducing high-consequence risk while preserving beneficial use.
[AI-99] Calibrate Then Route: A Measured Study of Learned Request Routing for Disaggregated LLM Serving
链接: https://arxiv.org/abs/2609.16206
作者: Srikanta Datta Tumkur,Jay Iyer,Mehar Simhadri,Sai Pavan Kumar,Sai Kapil Kumar,Ramesh Nampelly
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Disaggregated LLM serving places compute heavy prefill and memory heavy decode on separate GPU pools. Systems such as DistServe, Splitwise, and Mooncake make this separation fast, but routing still determines which instances handle each request. We study a router that estimates the additional completion time on each instance using exact prompt length, predicted output length, post admission KV cache pressure, and SLO class. We develop the policy in a discrete event simulator and validate it on eight NVIDIA A40 GPUs, each running a vLLM engine, with NIXL transferring KV caches between pools. All workloads run at measured saturation. Across three mixed, bursty arrival traces, the calibrated router achieves the highest mean goodput at 0.864, compared with 0.835 to 0.847 for round robin, least loaded, and a length heuristic. It also shows the lowest variance across traces. It beats round robin and the length heuristic on all three traces and least loaded on two. On the third, it trails by 0.003, within run to run noise. Hardware calibration matters: simulator derived constants cost 4.5 goodput points and roughly 40 percent of the tail latency advantage, reducing the scorer to little more than queue counting. Benefits grow with decode pool size and traffic heterogeneity but disappear in pools with three instances, where queue counts are often enough. Under extreme scarcity, greedy cost minimization concentrates requests on the cheapest scored instance, and blind spreading performs better. With calibrated costs, the learned router matches the goodput of round robin using six GPUs instead of seven.
[AI-100] Permutation-Based Stegomalware in Large Language Models : Threats and Countermeasures
链接: https://arxiv.org/abs/2609.16193
作者: Danny Wood,James Stringer
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 30 pages, 1 figure
Abstract:The difficulty of training large language models (LLMs), together with their ubiquity, raises the threat of stegomalware, where malicious payloads are embedded into model weights. Recent work has demonstrated the use of permutation symmetry in model weights to mitigate these threats, but failed to show neutralization of stegomalware across all weights for LLMs. In this paper, we demonstrate the full potential of behavior-preserving symmetries as a defense against stegomalware, as well as the risks these symmetries pose when exploited by attackers. For stegomalware neutralization, we improve upon previous work, demonstrating that it is possible to select permutations which displace all model parameters. This contrasts with previous methods which left a significant percentage of weights unaltered in LLMs. When used in an attack, we show that permutation symmetries can encode malware into the weights of a model in a way that is theoretically lossless, requires no retraining after encoding, and needs no payload-specific information in the extraction script—a combination of characteristics not previously seen in any single method. While theoretically lossless, permutation can in practice alter model behavior due to the accumulation of numerical error. We therefore quantify the loss in model performance associated with applying these methods, for both attack and defense, showing it to be minimal. Comments: 30 pages, 1 figure Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) Cite as: arXiv:2609.16193 [cs.CR] (or arXiv:2609.16193v1 [cs.CR] for this version) https://doi.org/10.48550/arXiv.2609.16193 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-101] AI-Driven Feedback Systems Digital Labour and Silent Quitting: Transforming African Workplaces
链接: https://arxiv.org/abs/2609.16192
作者: Abayomi O. Agbeyangi,Jose M. Lukose
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
备注: 34 pages
Abstract:The current trend of digitalisation has revolutionised the organisation of work and the way it is measured and performed across the globe, with AI becoming more common for managing labour and performance, as well as employee communication. In African organisations, where there is increasing adoption of remote work, hybrid models of work, digital collaboration, and data-based HR management, the notion of silent quitting has become more relevant, defined as worker disengagement when employees are still doing their job but do not put any effort into achieving good performance and exhibiting any emotion. This paper investigates how AI-driven feedback mechanisms, including sentiment analysis systems, pulse surveys, chatbots, engagement dashboards, and predictive analytics, are changing African workplaces through offering continuous listening, instant performance information and proactive engagement with employees. The study also explores how AI can assist organisations in identifying early disengagement and enable intervention and better employee communication in both private and public sector organisations in Africa. At the same time, we address the challenges of socioeconomic development and governance posed by AI implementation in developing countries, including digital inequality, infrastructure shortcomings, privacy concerns, algorithmic bias, and the risk of workplace surveillance. By situating silent quitting within wider debates on digital labour and automation, the paper contributes an African-centred perspective to discussions on the future of work and offers practical recommendations for HR professionals, managers, policymakers, and technology developers seeking responsible, context-sensitive approaches to workplace transformation across the continent.
[AI-102] Position: AI Is Not Ready for Strategic Conflicts
链接: https://arxiv.org/abs/2609.16189
作者: Mark Riedl,Glenn Matlin
类目: Artificial Intelligence (cs.AI)
备注: Published at the Social Sim’26 Workshop at COLM 2026. 4 pages body (22 pages total including appendices)
Abstract:Open-ended strategic wargames are high-stakes LM-based social simulations: they model adversaries, institutions, escalation, plan brittleness, doctrine, and crisis response. Language models (LMs) are attractive because they can play agents, generate scenario branches, adjudicate ambiguous actions, and summarize lessons, but the same affordances make open-ended roles dangerous: model language determines both what an actor attempts and what becomes simulated reality. This position paper argues that no LM-enabled wargame should inform planning, doctrine, policy, or crisis response without an auditable safety case, and that the proper use of open-ended wargames today is to stress-test decision-influencing LM agents. We identify five failure modes: decision laundering, adjudication opacity, role collapse, escalation-through-adjudication, and failure of strategic imagination. Ordinary benchmarks cannot establish safety for these settings. Wargames can expose failures as stress tests; they are not themselves safety cases for consequential use.
[AI-103] LLM s as Master Forgers: Generating Synthetic Time Series Data for Manufacturing ICML
链接: https://arxiv.org/abs/2609.16155
作者: Mantek Singh,Jeshwanth Challagundla,Prateek Karnal,Gagan Ganapathy,Vineet Shah,Ridam Arora
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 7 pages, 4 figures. Published in the 2024 International Conference on Image Processing, Computer Vision and Machine Learning (ICICML)
Abstract:This paper presents a novel framework leveraging Large Language Models (LLMs) to generate synthetic time series data for manufacturing processes. Motivated by the scarcity of labeled time-series data in real-world manufacturing settings, which hinders the development of robust machine learning models, we explore the potential of LLMs to learn complex temporal dependencies and generate realistic synthetic data. Our approach involves fine-tuning pre-trained LLMs on manufacturing process instructions and employing a Retrieval Augmented Generation (RAG) technique to enhance data diversity and realism. We evaluate our method against traditional time series modeling techniques like ARIMA and LSTMs, using quantitative metrics, PCA analysis, and downstream task performance (anomaly detection). Results demonstrate that our LLM-driven framework outperforms these baselines, generating high-quality synthetic time series data that effectively captures temporal dependencies and statistical properties of real manufacturing data, leading to improvements in downstream task performance.
[AI-104] Optimal Pruning for Neural Architectures using Fisher Information Distances
链接: https://arxiv.org/abs/2609.16129
作者: David S. Berman,Yen-Yu Fu,Edward Hirst,Thelma Chiwete Obirai
类目: Artificial Intelligence (cs.AI); Information Theory (cs.IT); Differential Geometry (math.DG)
备注: 21 pages, 4 figures, 4 tables
Abstract:A new scheme for parameter pruning is introduced, derived from the differential-geometric distance in model space. Pruning a parameter sets its value to zero, representing a displacement of the model to the hypersurface on which that parameter vanishes. The minimal distance from the unpruned model to this hypersurface is naturally computed via the geodesic distance in the model space as determined by the Fisher information metric. This distance determines the true change in the model, and its performance, under pruning. By analysing progressively more faithful approximations of this geodesic distance a natural hierarchy of optimality for pruning methods is determined. This starts with the traditional magnitude pruning, then develops into new more sophisticated and effective pruning schemes. The method is demonstrated for both fully-connected networks and vision transformers, on MNIST and CIFAR-10, over the complete 0 - 100% pruning range and across five random seeds. It outperforms pruning by parameter magnitude and by the local Fisher information alone in every architecture and dataset combination considered, on both accuracy and the Matthews correlation coefficient. Additionally, analysis of different levels of geodesic approximation produces intermediate pruning schemes that are computationally efficient and maintain near-optimal performance. This geometric picture supplies not only a state-of-the-art pruning methodology for AI models, but also a verified and mathematically-motivated justification for pruning schemes.
[AI-105] A Decision-Support Audit Protocol for Supervision Drift in Proxy-Labeled Credit-Risk Prediction
链接: https://arxiv.org/abs/2609.16102
作者: Mehrdad Shoeibi,Muhammad Shabanpour,Waldemar Karwowski,Niloofar Yousefi
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 10 pages, 2 figures, 3 tables
Abstract:Credit-risk models are trained on proxy labels and deployed under temporal and segment change, yet no single transfer metric separates base-rate shift, probability-scale shift, and feature-label relationship change. We contribute a design-science artifact: a locked, multi-signal audit protocol for supervision drift in proxy-labeled credit-risk prediction. Five layers (transfer performance, an oracle-gap probe, a calibration diagnostic, feature-label stability, and a synthetic positive control), thresholds, and decision rules were locked before interpretation; a bounded reading is a designed outcome. On a public LendingClub dataset (temporal 2013 to 2016 and cross-segment transfer), ranking is stable and oracle gaps are small; the clearest temporal signal is a prevalence and probability-scale mismatch that intercept-only diagnostic recalibration largely reduces, though its cause is not identifiable from the available release. The positive control responds only to larger injected shifts; subtler drift cannot be excluded. Mapping diagnostic patterns to governance actions is conceptual guidance, not validated here.
[AI-106] Coaching Qwen 3 Coder 30B to Think Like a CodeClash Arena Agent
链接: https://arxiv.org/abs/2609.16096
作者: Ivy Ning Zhang
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language model coding agents have recently become useful for software tasks, but weaker or open-weight agents still struggle to reliably interpret user intent and execute complex multi-step workflows. This gap is especially visible in long-horizon settings, where an agent must repeatedly inspect prior outcomes, diagnose failure, and choose the next code edit under interaction constraints. It motivates a natural question: what can we do to improve the thinking process of a weak code agent? We study this question in CodeClash, a code-arena benchmark where the original work evaluates 8 commercial coding agents across 6 arenas through multi-round tournaments. Since Qwen3 Coder Plus ranks last among them, we take the open-weight Qwen3-Coder-30B as a case study and investigate how to improve it with distilled knowledge from stronger agents. Our analysis shows that Qwen3-Coder-30B is not well optimized for arena-style interaction: it frequently produces syntax and protocol-breaking errors and exhibits weak strategic adaptation across rounds. These failures are difficult to correct with vanilla instruction tuning alone, since offline SFT cannot directly verify whether a generated action is valid or beneficial. To address this, we propose ReAct SFT, which rewrites teacher trajectories into explicit [obs][thought][act] chains, and trajectoryquality weighted SFT, which reweights samples to encourage post-edit checking. ReAct SFT substantially improves strategic behavior, and our fine-tuned model outperforms the original Qwen3 Coder Plus in tournament evaluation.
[AI-107] MUUNRiver-Bench: Diagnosing Relation-Dependent Music Retrieval with Multimodal Instructions
链接: https://arxiv.org/abs/2609.16090
作者: Zhancheng Guo,Congren Dai,Shangda Wu,Jianhuai Hu,Danni Zhao,Xiaobing Li,Maosong Sun
类目: ound (cs.SD); Artificial Intelligence (cs.AI)
备注:
Abstract:Music retrieval is relation-dependent: given a reference track, a listener may seek its style with a new theme, a cover, or a comparable voice, and these intents demand contradictory rankings. We present MUUNRiver-Bench, a diagnostic benchmark whose reference-audio queries use natural-language instructions to define relevance. A pipeline combining expert genre priors, LLM-generated prompts and lyrics, synthesis, and expert review yields 3,440 tracks spanning 13 genres and 116 sub-genres, and seven tasks: similar-music, style-preserving lyric-rewriting, lyric-preserving style-rewriting, cover, vocal-timbre, isolated-vocal, and segment retrieval. Across six models in eight configurations, task-wise rank reversals reveal complementary biases: acoustic encoders favour local identity, whereas text-aligned encoders favour semantic relations. Frozen encoders diagnose default similarity preferences; instruction-aware and audio-text fusion systems provide exploratory tests of textual conditioning, with neither simple fusion scheme consistently improving its backbone
[AI-108] AssemblyGrid v1: A Benchmark for Multi-Robot Production with Temporary Coalitions Local Information and Geometric Constraints
链接: https://arxiv.org/abs/2609.16075
作者: Fouad Bahrpeyma,David Heik,Dirk Reichelt
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: 11 figures and 23 tables, including appendices with the formal benchmark specification, evaluation protocol, conformance requirements, and extended experimental results. The AssemblyGrid v1 benchmark implementation and reproducibility materials will be publicly available at this https URL and this https URL when the paper is online
Abstract:Flexible robotic production requires joint decisions on process progression, material routing, resource assignment, temporary cooperation, and simultaneous execution, since each decision can affect the feasibility of the others. The challenge is greater under decentralized control, where each robot acts from bounded local information while system progress depends on collective decisions, shared resources, material state, and workspace compatibility. These properties closely match cooperative multi-agent decision making under partial observability and resource contention. This paper introduces AssemblyGrid v1, a reproducible benchmark for repeated multi-robot production that combines explicit process progression, decentralized observations, material transfer, temporary multi-robot coalitions, productive concurrency, and geometry-dependent feasibility within one task-level formulation. The benchmark includes Flow, Coalition, and Concurrency workload families, each with three scenario levels. Task success and evaluation measures are defined independently of learning reward and solution method, allowing learning-based and non-learning methods to address the same production problem. AssemblyGrid v1 is evaluated through executable conformance checks, mechanism studies, and algorithmic experiments using a privileged centralized reference, structured decentralized controllers, and MARL methods including IPPO, MAPPO, and QMIX. Results demonstrate productive execution under centralized and decentralized control. The MARL experiments further show that decentralized policies can learn effective production behavior from local observations and actions, supporting AssemblyGrid as a controlled benchmark for studying cooperative decision making in flexible robotic production.
[AI-109] Schema-Adaptive Action-Conditioned JEPA for Cross-Machine CNC Transfer under Partial Sensor Overlap DATE
链接: https://arxiv.org/abs/2609.16071
作者: Ayoub Louaye Bouaziz,Matthieu Ostertag,Anton Demasles
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Code, configuration files, the twenty candidate specifications, and the audit scripts are available at : this https URL dev/saac- jepa ; Animated versions of the schematics are on the project page: this https URL
Abstract:Cross-machine deployment of industrial world models requires transfer across changes in dynamics, sensing interfaces, sampling regimes, and control units. We study a schema-adaptive action-conditioned Joint-Embedding Predictive Architecture (SAAC-JEPA) for CNC dynamics, where the source machine has 17 canonical sensor channels and the target shares only 10. Evaluation uses group-disjoint source splits, source-only normalization, held-out self-supervised validation, unit audits, and a sealed target test after model locking. Across five seeds, JEPA pretraining gives no clean-source forecasting gain: scratch and pretrained-body models obtain (\mathrmRMSE=0.811\pm0.022) and (0.813\pm0.022). A source-only search over 20 candidates selects a schema-consistent action-conditioned JEPA after seven-seed stability checks. On the confirmatory target pass, the locked model reaches zero-shot (\mathrmRMSE=0.546), (R^2=0.012), and (\mathrmNLL=0.52), outperforming persistence but not RevIN-equipped PatchTST and iTransformer baselines ((0.503) and (0.498)). A pre-declared paired ablation shows that RevIN in the same architecture improves RMSE to (0.495\pm0.004) over three seeds, but degrades target calibration ((\mathrmNLL=20.6)) on stationary context windows. A pre-lock adaptation sweep further reduces RMSE to (0.520) with limited target support. These results show that source-domain forecasting accuracy alone is insufficient to assess industrial predictive representations, and that cross-machine adaptation under partial sensor overlap is a distinct evaluation axis.
[AI-110] Beyond Distribution Matching: Semantics-Consistent Tabular Diffusion with Weak Semantic Priors
链接: https://arxiv.org/abs/2609.16069
作者: Yili Wang,Ruxue Shi,Mengnan Du,Hangting Ye,Yi Chang,Xin Wang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Synthetic tabular data can match real data distributions while still violating the semantic constraints that govern valid tabular rows. This reveals a key limitation of existing tabular generators: they mainly optimize distributional fidelity, but do not explicitly model weak semantic priors encoded in tabular schema and textual descriptions. In this paper, we propose \ours, a semantics-consistent tabular diffusion framework for high-fidelity synthetic data generation under weakly specified semantic priors. \ours\ first constructs two types of priors, namely intra-column semantics and inter-column symbolic rules, with LLM-assisted extraction from metadata and validation on the real training split. These priors are then used as generation conditions rather than post-hoc filters. Specifically, \ours\ maps heterogeneous column values, column identities, and semantic priors into a unified semantic space, and performs column-wise forward corruption and prior-conditioned reverse denoising to preserve both marginal distributions and rule-consistent cross-column dependencies. Extensive experiments on six real-world tabular benchmarks show that \ours\ consistently improves distributional fidelity, semantic consistency, and downstream task utility over representative VAE-, GAN-, LLM-, and diffusion-based baselines. Additional analyses further demonstrate the robustness of \ours\ when semantic priors are partially unavailable.
[AI-111] You Dont Need To Train: Agent ic Heuristic Learning Studio for Executable Human Activity Recognition
链接: https://arxiv.org/abs/2609.16065
作者: Siyu Yuan,He Zhang,Sizhen Bian,Bin Guo
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Human activity recognition (HAR) is usually framed as gradient-based training of neural networks. Agentic Heuristic Learning (AHL) Studio explores a complementary view inspired by human cognitive learning: people learn activities by remembering examples, forming rules, and repairing mistakes, not by backpropagating. This proposed tool implements AHL for HAR: a learning-time agent reasons over sensor protocols, proposes executable heuristic policies, records repair traces, and exports an LLM-free policy for edge deployment. We focus on the HAR benchmark family and provide an end-to-end workflow from dataset observation to edge-oriented export. On eleven HAR datasets evaluated so far, AHL policies reach strong executable-policy performance while remaining inspectable, editable, and replayable \footnotethis https URL.
[AI-112] Driver Behavior Estimation at Signalized Intersections Using a Physics-Constrained Decision-Conditioned Autoregressive Transformer ITSC2026
链接: https://arxiv.org/abs/2609.16058
作者: Mohammad Khoshkdahan,Pavel Laskov,Alexey Vinel
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Systems and Control (eess.SY)
备注: Accepted for publication at the 2026 IEEE International Conference on Intelligent Transportation Systems (ITSC 2026)
Abstract:Red-light violations and harsh braking at signalized intersections are major contributors to traffic accidents. This paper analyzes and predicts human driver decision-making and longitudinal trajectory behavior during traffic light signal transitions. We collected a diverse real-world dataset comprising 449 approach runs under varying speed and distance conditions. Vehicle motion was recorded using RTK-corrected GNSS with centimeter-level accuracy, and driver heart rate and multi-level comfort ratings were monitored. Spatial and temporal calibration ensured precise alignment between vehicle state and signal timing. Statistical analysis identifies required deceleration as the dominant single predictor of the stop-go decision, and heteroscedastic Gaussian modeling of peak deceleration reveals five empirical comfort ranges derived from human stopping behavior. Based on this insight, we propose a two-stage modeling framework. Stage 1 predicts the binary maneuver decision, and Stage 2 generates the longitudinal acceleration trajectory using a decision-conditioned autoregressive Transformer with physics constraints, including target-state conditioning and jerk limits. The proposed architecture outperforms baseline methods and achieves 0.49m/s^2 acceleration MAE and 0.62m distance MAE. It also estimates the future stopping-comfort level of the human driver from a single yellow-onset snapshot. Qualitative results demonstrate realistic human-like braking behavior. The dataset and source code are publicly available.
[AI-113] OmniHarness: Harnessing Generalizable Visual Generation via Symbolic Policy Learning
链接: https://arxiv.org/abs/2609.16057
作者: Xu Xu(1),Jinxiu Liu(2),Zhangbo Qiao(1),Jiaxing Lu(1),Xiangyu Zhang(1),Yubin Gu(3),Fangwei Ning(1),Yan Shi(1) ((1) Beihang University, (2) The Chinese University of Hong Kong, (3) National University of Singapore)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Unified multimodal large language models (MLLMs) and multi-agent systems have advanced visual generation. However, three limitations remain. (1) Existing methods often distill task-specific experience with limited generalizability. (2) Reflection is often deferred until task completion. (3) Knowledge is often acquired only in response to downstream task demands. To address these limitations, we introduce OmniHarness, a framework for generalizable visual generation via symbolic policy learning. OmniHarness abstracts verified executions into symbolic policies for visual generation task families, capturing shared procedures and applicability conditions while removing instance-specific inputs. The harness instantiates, adapts, and composes these policies for new tasks. Intermediate verification guides refinement and failure recovery during execution. Through self-directed inquiry, OmniHarness autonomously generates and executes practice tasks near its capability limits before downstream objectives are specified. Execution feedback continually refines the policies while model parameters remain fixed. Experiments across six benchmarks, three MLLM backbones, and three visual agent frameworks demonstrate strong performance and continual capability expansion. On ComfyBench’s Creative tasks, OmniHarness achieves a 95.0% resolve rate, exceeding the strongest baseline by 27.5 percentage points. Frozen policy snapshots improve existing visual agent systems through plug-and-play reuse.
[AI-114] Managing Action Preconditions in Neuro-Symbolic RL: Three Placement Strategies for Embodied Agents
链接: https://arxiv.org/abs/2609.16056
作者: Norbert Oswald,Fabian Deuser,Thomas Bräunl
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 10 pages, 4 figures and 2 tables
Abstract:Humans carry behaviour knowledge of how to act in familiar situations into every new task rather than relearning it from scratch. There is no reason a Reinforcement Learning (RL) agent shouldn’t do the same: known behaviour patterns need not be learned, only applied. Neuro-symbolic RL bridges prior knowledge and RL by injecting symbolic knowledge alongside a learned policy. The point at which this knowledge is integrated is critical: a poor choice can produce, for instance, hallucinated preconditions, which surface as safety and reliability problems in agents acting in changing environments. We formalise this behavioural knowledge as a precondition Bayesian network (BN) over the agent’s \emphstructural actions - the actions whose legality depends on preconditions, such as picking up a key, grasping a block, toggling a door, or dropping an object. The BN restricts when these actions may fire, and we inject it into the RL loop at three placements: (1) a \emphsymbolic verifier, consulted only at inference, that fires a structural action once its preconditions hold; (2) a \emphsymbolic enforcer, active during both training and inference, that governs structural-action use throughout learning; and (3) a \emphsymbolic learner, which folds the knowledge into the network and learns the restriction and use of structural actions itself. To test the three variants we run experiments on two benchmarks with opposite regimes: one built on long, ordered planning chains, the other on continuous manipulation. We compare against strong baselines on solution quality, sample efficiency, and traceability. The payoff is substantial. On MiniGrid, all three placements improve the \emphsolution quality over the PPO+RND baseline, the symbolic enforcer leading at 98.2% against the baseline’s 88.8% . On Fetch, \dots
[AI-115] Causal neural set filtering for online multi-target tracking
链接: https://arxiv.org/abs/2609.16054
作者: Zhongdi Liu,Huangyu Dai
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 5 pages, 2 figures
Abstract:Transformer-based multi-target tracking (MTT) jointly learns data association and state estimation, but MT3/Track-MT3-style trackers repeatedly re-encode measurement windows, incurring redundant computation. We propose Causal Neural Set Filtering (CNSF)\footnote\hrefthis https URLCode: this https URL, a neural set filter that encodes only current measurements while carrying past evidence in a structured recursive track state. CNSF combines exclusive Sinkhorn association, association-conditioned Kalman-shaped updates with moment matching, and recurrent Bernoulli lifecycle modeling with measurement-driven birth. These mechanisms impose soft one-to-one constraints, propagate association-induced state uncertainty, and support existence estimation under missed detections and birth–death transitions. On a held-out three-regime simulated test set, CNSF reduces mean GOSPA and T-GOSPA relative to Track-MT3 by 19.3% and 30.4%, with 55.9% fewer parameters and a 3.76\times speedup in single-thread CPU inference.
[AI-116] Estimating Uncertain Spatial Relationships in Robotics UAI1986
链接: https://arxiv.org/abs/1304.3111
作者: Randall Smith,Matthew Self,Peter Cheeseman
类目: Artificial Intelligence (cs.AI); Robotics (cs.RO)
备注: v2: the authors’ typeset full-length version, as published in Uncertainty in Artificial Intelligence 2 (1988), pp. 435-461, and reprinted in Autonomous Robot Vehicles (1990), pp. 167-193. v1 is a scan of the shorter UAI 1986 proceedings version
Abstract:In this paper, we describe a representation for spatial information, called the stochastic map, and associated procedures for building it, reading information from it, and revising it incrementally as new information is obtained. The map contains the estimates of relationships among objects in the map, and their uncertainties, given all the available information. The procedures provide a general solution to the problem of estimating uncertain relative spatial relationships. The estimates are probabilistic in nature, an advance over the previous, very conservative, worst-case approaches to the problem. Finally, the procedures are developed in the context of state-estimation and filtering theory, which provides a solid basis for numerous extensions.
[AI-117] Evaluating Verified Autonomy in Quantum Engineering
链接: https://arxiv.org/abs/2609.17439
作者: Naixu Guo,Changhao Li,Siyu Cheng,Qicheng Tang,Binzhao Luo,Bikun Li,Yuxuan Du,Shihao Ru,Jiaqi Cai
类目: Quantum Physics (quant-ph); Artificial Intelligence (cs.AI)
备注: 10 pages, 4 figures
Abstract:Reliable quantum engineering is essential for turning quantum phenomena into practical technologies. As quantum platforms grow in scale and complexity, their characterization and operation require increasing human effort and coordination. Scientific artificial intelligence agents, which can plan experiments, operate instruments, and analyze observations, offer a promising route towards autonomous quantum engineering. Yet whether current agents can perform reliably in this setting has not been systematically established. To fill this gap, we developed Quantum-Harbor, a virtual laboratory that provides a controlled execution environment for agents to interact with quantum systems. This design enables direct verification of both the actions taken and the conclusions drawn. Building on this framework, we introduce QIQCBench, a benchmark of 49 expert-authored tasks spanning multiple layers including calibration and control, error correction and compilation, sensing and networking. Across 17 frontier agentic systems, QIQCBench reveals wide variation in verified performance. These results expose a substantial gap between demonstrating capability and achieving reliable operation, and establish Quantum-Harbor as a foundation for measuring progress towards verified autonomy in quantum engineering.
[AI-118] AI for Science with GPT -6 Astra: Thermal Design and Electrothermal Analysis of 2D CFET
链接: https://arxiv.org/abs/2609.17123
作者: Min-Hui Kim,Khushi Sharma,Sarah Zhang,Ye Wang
类目: Materials Science (cond-mat.mtrl-sci); Artificial Intelligence (cs.AI); Hardware Architecture (cs.AR)
备注:
Abstract:Thermal optimization of 2D CFET inverters requires testing structural proposals against their electrical costs. We examine these research tasks using an AI agent workflow within a supplied electrothermal model. At 12 nm, Astra selects a redistributed source-interconnect geometry, while a coordinating agent proposes a substrate-directed heat-removal path. The combined design reduces peak temperature rise by 1.67 K at fixed metal volume and 20 \muW. A subsequent metal-resistance sensitivity gives about 0.6-K inverter cooling alongside a 2% nFET on-current loss. Effective contact-length scaling further shows that lower temperature can accompany higher thermal resistance when current falls. Reproduction identifies agreeing implementations and retains a 104.95-K failure for diagnosis. These results show that an AI scientist workflow can propose thermal structures, test them under common constraints, and quantify their electrical cost.
[AI-119] Causal Discovery via Transformed Low-Rank Quantile Surfaces
链接: https://arxiv.org/abs/2609.16931
作者: Ryo Kamimura,Thong Pham
类目: Methodology (stat.ME); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Machine Learning (stat.ML)
备注: 25 pages
Abstract:We propose Low-Rank Quantile Surfaces (LRQS), a bivariate causal model in which, in the causal direction, an unknown monotone transformation of the conditional quantile surface admits a low-rank functional decomposition. LRQS subsumes location-scale noise models and post-nonlinear heteroscedastic noise models, while allowing multiple quantile bases to represent changes beyond location-scale effects. We prove generic identifiability of LRQS: the transformed quantile surface is low rank in the causal direction, whereas reverse representability under the corresponding constraints occurs only for exceptional, fine-tuned cause marginals. We provide a simple-yet-powerful causal score using a nonparametric fitting procedure that alternates between rank-constrained approximation of discretized quantile surfaces and isotonic estimation of the unknown monotone transformation. Experiments on synthetic mechanisms with higher-rank distributional shape variation and strong nonlinear distortions, together with standard bivariate benchmarks, show that LRQS is especially effective when conditional distributional shape or observation distortion goes beyond existing location-scale assumptions.
[AI-120] QALPA: Property-guided diffusion modeling for efficient exploration of chemical spaces of flexible molecules
链接: https://arxiv.org/abs/2609.16527
作者: Michael Hanna,Julian Cremer,Zekiye Erarslan,Leonardo Medrano Sandonas
类目: Chemical Physics (physics.chem-ph); Artificial Intelligence (cs.AI)
备注: 16 pages, 6 figures
Abstract:Exploring the chemical space of flexible molecules remains challenging because the vast number of possible compounds and conformations, together with the increasing cost and limited generalization of 3D generative models for larger and more complex molecules, restrict access to unexplored chemistry. Here, we introduce QALPA (“Quantum-Aware Learning for Property-space Augmentation”), a property-guided generative framework that combines an E(3)-equivariant diffusion model with active learning and efficient quantum-mechanical (QM) methods to iteratively explore targeted QM property manifolds. By coupling generation with physics-based evaluation, QALPA improves molecular sampling and model reliability in sparsely populated regions of chemical space. Our results show that training on complementary QM datasets spanning both small (QM7-X) and large (Aquamarine) drug-like compounds enables accurate molecular generation across a broad size range, improving transferability beyond the training distribution for complex property manifolds involving both extensive and intensive properties. As a proof of concept, QALPA coupled with the machine learning-augmented tight-binding method EquiDTB efficiently augments alloQM, a QM dataset introduced in this work, comprising 6,253 conformers of allosteric drug molecules, by populating sparse regions of the property landscape defined by the many-body dispersion energy and HOMO-LUMO energy gap. These results demonstrate that the integration of generative AI with efficient ML/QM methods offers a practical pathway toward augmenting sparse QM datasets and sustainably expanding the exploration of chemical space for molecular discovery.
[AI-121] Recovery Rates Are Not Comparable Across Transcription Factors: Chance Correction for Attribution Evaluation KR
链接: https://arxiv.org/abs/2609.16271
作者: Hyunkyung Han,Min Jung Kim
类目: Other Quantitative Biology (q-bio.OT); Artificial Intelligence (cs.AI)
备注: 8 pages, 7 figures, 8 tables. Code and data: this https URL (archived at this https URL )
Abstract:Attribution methods for genomic sequence models are commonly evaluated by how much of a known motif they recover, or by how a prediction degrades as evidence is deleted. Neither score is interpretable without the value it would take by chance, and neither is routinely reported against one. We show that this omission is not a matter of precision but of validity. The uniform chance level for contiguous motif overlap is (L/(N-L+1)); across 268 transcription factors in UniBind it ranges from 0.0118 to 0.0427, a 3.6-fold spread determined by motif length and window size alone. For two factors the bootstrap intervals of the chance levels themselves do not overlap, so their raw recovery rates are not comparable quantities. Correcting for this dissolves a published three-way classification of five factors: a factor reported as a resolution failure attains the second-highest corrected value, ahead of one of the two positive controls, and two reported as complete failures fall at or below chance. We further show that perturbation-based evaluation can fail its own precondition: for one factor a fully masked input still scores above the decision boundary, and the curve is not monotone in the number of masked positions, so the area under it is not a measure of faithfulness. We provide chance levels in closed form, a chance-corrected score, and two screens that run before any attribution is computed. Comments: 8 pages, 7 figures, 8 tables. Code and data: this https URL (archived at this https URL) Subjects: Other Quantitative Biology (q-bio.OT); Artificial Intelligence (cs.AI) Cite as: arXiv:2609.16271 [q-bio.OT] (or arXiv:2609.16271v1 [q-bio.OT] for this version) https://doi.org/10.48550/arXiv.2609.16271 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
机器学习
[LG-0] FreqSpaNet: Frequency and Spatial Learning of SFPF for Physical Layer Hardware Integrity Detection
链接: https://arxiv.org/abs/2609.17491
作者: Xiaoxuan Huang,Jinlong Xu,YiZhe Wang,Meng Zhang,Xian Li,Yuying Bian
类目: Machine Learning (cs.LG)
*备注: 5 pages, 6 figures
Abstract:Unauthorized hardware replacement can preserve a wireless device’s logical identity while altering its physical implementation, posing a challenge to hardware integrity verification. Spatio-frequency polarization fingerprints (SFPFs) capture device-dependent responses across multiple frequencies and directions, but their frequency and spatial dimensions exhibit different structural dependencies. We propose FreqSpaNet, an SFPF representation learning network for open set hardware anomaly detection. A frequency branch captures local variations among neighboring frequencies, while a geometry-aware spatial branch models directional relationships using angular information. The two representations are combined through adaptive fusion, and complementary pretraining further captures shared information while preserving the distinct characteristics of the frequency and spatial representations. Experiments show that FreqSpaNet achieves a mean AUROC of 96.31%, 9.05 points above the baseline. Results under seven hardware replacement scenarios further verify the effectiveness of FreqSpaNet.
[LG-1] Reduced-Space Multi-Fidelity Bayesian Optimization of Process Simulation Models
链接: https://arxiv.org/abs/2609.17440
作者: Niki Triantafyllou,Andrea Bernardi,Maria M. Papathanasiou
类目: Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注: Accepted at the 20th Learning and Intelligent Optimization Conference (LION 20), 2026. Corrected author version. This version corrects a typo in the mathematical description of the multi-fidelity covariance kernel in Section 3.2
Abstract:Optimizing industrial process flowsheets is often computationally prohibitive due to the high cost of rigorous simulations and the curse of dimensionality inherent in complex design spaces. To address these challenges, we present a reduced-space multi-fidelity Bayesian optimization (RS-MFBO) framework designed for high-dimensional, expensive black-box functions. The approach integrates Global Sensitivity Analysis (GSA) for dimensionality reduction with a fidelity-augmented Gaussian process that captures correlations between low-cost approximations and expensive high-fidelity evaluations. A cost-aware acquisition strategy, augmented with cooldown and promotion mechanisms, adaptively guides the allocation of samples across fidelities. The framework is validated on two distinct industrial process simulators: a plasmid DNA bioprocess in SuperPro Designer and a green fuel synthesis plant in Aspen HYSYS. Results across diverse economic and physical objectives demonstrate that the proposed method substantially reduces the number of high-fidelity simulator evaluations while maintaining competitive optimization performance compared to single-fidelity baselines. These results highlight RS-MFBO as a scalable, simulator-agnostic approach for cost-constrained black-box optimization.
[LG-2] Bridging the Confidence Gap: Temperature Scaling for Calibrating Test-Time Prompt Tuning
链接: https://arxiv.org/abs/2609.17386
作者: Yuwei Liang,Jian Liang,Dapeng Hu,Yinuo Xu,Ran He
类目: Machine Learning (cs.LG)
*备注:
Abstract:Test-time prompt tuning (TPT) enables adaptation on a single test instance, achieving improved accuracy but often sacrificing calibration performance. Most existing calibration methods introduce additional regularization terms to promote dispersion across text embeddings and reduce calibration error, yet these methods often suffer from a drop in accuracy. Motivated by the well-calibrated nature of zero-shot predictions, we propose CoTS, a simple yet effective post-hoc calibration method that preserves accuracy. Specifically, CoTS applies temperature scaling to minimize the confidence gap between adapted and zero-shot predictions. To fully exploit the potential of multiple augmentations during adaptation, we introduce a weak-strong ensemble strategy that further boosts accuracy. We then apply CoTS to this ensemble, termed E-CoTS, to maintain its well-calibrated property. Extensive experiments on diverse datasets and backbones show that our approaches effectively mitigate miscalibration without compromising primary accuracy. For instance, E-CoTS reduces the average expected calibration error of TPT from 11.90% to 5.38% on ImageNet variants, while even increasing accuracy from 60.74% to 62.95%. Moreover, when integrated with existing calibration methods, E-CoTS usually enhances both accuracy and calibration simultaneously.
[LG-3] OPEN-1B: A Fully Auditable Training Run
链接: https://arxiv.org/abs/2609.17380
作者: John Donaghy,Brian Wilcox,Oğuzhan Ersoy,Shikhar Rastogi,Adam St Arnaud,Alexey Titov,Jordan Greenberg,Ben Fielding,Harry Grieve
类目: Machine Learning (cs.LG)
*备注:
Abstract:Open-source language models have a reproducibility problem. Despite releasing weights, training data, and recipes, none of them are provably reproducible due to the non-associativity of floating-point arithmetic. Deep learning frameworks often offer a deterministic execution mode, allowing reproducible operations on the same machines. Unfortunately, this determinism does not carry across hardware such that a user can verify that a released checkpoint was actually produced using the declared training recipe. This leaves room for undisclosed data, injected biases, or backdoors that existing techniques such as proof-of-learning or proof-of-training-data cannot rule out. We introduce a new tier of model transparency, fully auditable, in which every operation on every data sample during training is independently reproducible on heterogeneous commodity hardware with bitwise certainty. By imposing a definite order on the sources of training nondeterminism, GPU kernel reductions, data batch ordering across a data-parallel cluster, and inter/intra-node collective communication, we make it possible to replay any individual step of a large, distributed training run on a single piece of commodity hardware and check it against the published trajectory. Because replaying an entire run on one machine is infeasible, we support this with a collective verification scheme in which many independent auditors each certify individual steps, together covering the whole run. We release Open-1B, a model trained under this regime, together with its full pretraining dataset, every intermediate checkpoint, the training codebase, and the audit harness needed to reproduce and verify any step of its training. Subjects: Machine Learning (cs.LG) Cite as: arXiv:2609.17380 [cs.LG] (or arXiv:2609.17380v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.17380 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-4] Hybrid Variational Quantum Circuits for Multivariate Regression and High-Dimensional Data Reconstruction
链接: https://arxiv.org/abs/2609.17358
作者: Koffi Ognandon Ayena(ICB),Frédéric Holweck(ICB),Serge Iovleff(UR4662),Amah S d’Almeida
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:Variational quantum circuits (VQCs) are parameterized quantum circuits optimized classically. We propose a hybrid variational quantum circuit (HVQC) extending VQCs with a classical affine post-measurement layer, enabling vector-valued regression without the linear overhead of independent scalar circuits. Theoretically, we show that elementary one-and two-qubit circuits can approximate quadratic functions and products via data re-uploading and entanglement, providing the foundations of the full architecture. Experimentally, on two synthetic image reconstruction datasets and the Friedman1 benchmark (40,568 test samples), our HVQC matches Gaussian Process Regression and outperforms XGBoost and Random Forest. An ablation study confirms that both quantum and classical components are essential, and results highlight the central role of the feature map in hybrid quantum-classical models.
[LG-5] ype-IV Code Clone Detection via Layer-Wise Non-Contrastive Representation Learning
链接: https://arxiv.org/abs/2609.17338
作者: Luciano Marchezan,Kevin Delcourt,Eugene Syriani,Houari Sahraoui
类目: oftware Engineering (cs.SE); Machine Learning (cs.LG)
*备注:
Abstract:Software clones are fragments of code that are similar or functionally equivalent to each other. They pose significant challenges for maintenance, refactoring, and bug detection. Detecting Type-IV clones, which are semantically equivalent but may differ syntactically, is particularly difficult for traditional token- or syntax-based methods. Recent machine learning approaches rely on contrastive learning, which requires careful negative sampling and can introduce bias. In this paper, we propose LWVIC4Code, a non-contrastive representation learning approach specifically designed for Type-IV clone detection. Building on the Variance-Invariance-Covariance Regularization (VICReg) framework and prior layer-wise VICReg training, LWVIC4Code introduces cross-layer consistency regularization and depth-dependent layer weighting to progressively refine semantic information across transformer layers, producing robust and discriminative code representations. We conduct an empirical study comparing LWVIC4Code against a contrastive learning baseline and zero-shot large language models on Python (Kamino) and multi-language (GPTCloneBench) datasets. Results show that LWVIC4Code achieves competitive or superior performance without negative samples, benefits from layer-wise supervision, and generalizes effectively from Python to other languages, particularly Java and C#. These results demonstrate that non-contrastive, layer-wise representation learning is a promising direction for robust semantic code clone detection.
[LG-6] Same Flow Different Paths: Variance Reduction in Flow Matching
链接: https://arxiv.org/abs/2609.17287
作者: Alexander Tyurin
类目: Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注:
Abstract:In flow matching (FM), a velocity model v_\theta is trained using a predefined path g_t that connects data and noise samples (e.g., g_t(x_0, x_1) = (1 - t) x_0 + t x_1 ). In this work, we study the choice of this path from an optimization perspective by analyzing the variance of stochastic gradients. We consider the class G(p_t,v^\star_t) of paths that induce the same marginal distributions p_t and marginal velocity field v^\star_t , and therefore the same FM objective. Our main finding is that the choice of path g_t can fundamentally change the convergence rate of SGD, even when the FM objective remains exactly the same. (i) For a linear velocity model and one-dimensional Gaussian data, we derive a tight bound on the SGD iteration complexity up to logarithmic factors and find an analytically optimal path that minimizes this bound among linear paths inducing the same FM problem. (ii) We then extend the variance analysis to general FM problems and formulate path selection at a fixed \theta as the variance-minimization problem PathOpt _\theta , constrained to g_t\in G(p_t,v^\star_t) . We show that this constraint is essential: reducing variance without it can lead to slower convergence. (iii) Since the constraint g_t \in G(p_t,v^\star_t) cannot generally be verified directly, we derive an equivalent formulation with constraints that can be estimated from samples, allowing paths to be found numerically. Our theoretical results are supported by experiments with Gaussian data, Gaussian mixture models, and real datasets.
[LG-7] Personalized Federated Learning through Global Knowledge Distillation and Local Head Adaptation
链接: https://arxiv.org/abs/2609.17284
作者: Polycarpo Souza Neto,José Mairton Barros da Silva Júnior,Charles Casimiro Cavalcante
类目: Machine Learning (cs.LG); Machine Learning (stat.ML); Other Statistics (stat.OT)
*备注:
Abstract:Statistical heterogeneity limits federated learning when a single global classifier cannot represent client-specific label distributions. In this work, we propose Personalized Federated Knowledge Distillation with Head Adaptation (pFedKDH), which aggregates only the shared backbone, keeps persistent client-specific heads, and uses a recalibrated global head as a teacher during local training. Across MNIST, Fashion-MNIST, CIFAR10, and CIFAR100 under class-wise Dirichlet partitions, pFedKDH obtains the best accuracy in most settings, with accuracy gaps up to 37.67% over the weakest baseline and consistently low standard deviation across repetitions. Component-wise diagnostics and convergence results support the role of persistent heads and distillation-guided local optimization under label-skewed data.
[LG-8] Memorisation bias in medical AI
链接: https://arxiv.org/abs/2609.17223
作者: Moritz A. Knolle,Martin J. Menten,Laurin Lux,Mélanie Roschewitz,Emma A.M. Stanley,Georgios Kaissis,Daniel Rueckert,Ben Glocker
类目: Machine Learning (cs.LG); Computers and Society (cs.CY)
*备注:
Abstract:Medical AI models hold immense potential to improve patient outcomes, but they are also known to unintentionally memorise individual records from their training datasets. While such memorisation has been linked to targeted privacy attacks, its consequences for clinical deployment, where patients may be assessed by a model that saw their historical data during training, remain poorly understood. Here we show that predictions on a patient’s unseen future data can change significantly if a model observed that same patient’s anonymised historical data during training, a phenomenon we term “memorisation bias”. We demonstrate that this bias exists across diverse data modalities and model architectures, and over prolonged time spans: in some cases, memorisation bias persists on future records acquired decades after the historical records used for training. Moreover, in simulated prospective deployment, memorisation bias has asymmetric effects on the diagnostic accuracy of returning data contributors. When a patient returned with a de novo condition absent from their historical records in the training dataset, diagnostic sensitivity decreased significantly compared to an otherwise identical model not trained on their historical data. Conversely, when their health state was unchanged, both sensitivity and specificity were significantly inflated. Our findings reveal a previously uncharacterised risk in medical AI that arises when a model is deployed on patients who contributed to its training data. This exposes a shortcoming of current model development practice: the de-identification measures designed to protect patients’ privacy make it difficult to identify returning contributors and exclude them from the AI-assisted interpretation of their own future data. Mitigating memorisation risks may thus require changes to current model training and deployment protocols.
[LG-9] Cross-Domain Inference for Human Localization: Applying Wi-Fi RSSI Data to CSI-Trained Models
链接: https://arxiv.org/abs/2609.17204
作者: Ariel Duschanek-Myers,Thomas Welsh,Helmut Neukirchen
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG); Networking and Internet Architecture (cs.NI)
*备注:
Abstract:Wi-Fi signal data can be used to compromise the privacy of individuals. While many existing approaches rely on Channel State Information (CSI), collecting this data on typical IoT devices often requires elevated operating system permissions and specialized drivers. Consequently, this paper investigates the feasibility of utilizing Received Signal Strength Indicator (RSSI) data to predict human locations. RSSI was selected because it is accessible even on devices with limited user permissions, and therefore is more applicable to a wider array of IoT devices. To bypass the tedious process of obtaining training data needed to train an RSSI-based model, an existing Wi-Fi pose prediction project was used in this research. However, that project assumed CSI data as input. Therefore, we investigate the feasibility of cross-domain inference, i.e., feeding RSSI data into that existing CSI-based model. We collected an RSSI dataset, synchronized with video ground-truth of a person moving within a room, to evaluate the model’s performance. This evaluation confirmed that RSSI data can predict locations with approximately 80% confidence when human movement is present. This demonstrates that a model trained on CSI data can be used to evaluate low-granularity RSSI data consisting of decibel-milliwatt (dBm) values to roughly locate people in the collection space. These results imply that a wide range of IoT devices can be used for privacy invasion in Wi-Fi-dense environments.
[LG-10] MyoFlow: Anchor-Tied Rectified Flow for HD-sEMG Gesture Recognition Across Sessions and Subjects
链接: https://arxiv.org/abs/2609.17194
作者: Chenhao Wu,Dingjie Peng,Satoshi Funabashi,Satoshi Konishi,Wuqiang Yang,Hiroshi Onoda,Hironori Washizaki,Jiang Liu
类目: Machine Learning (cs.LG)
*备注:
Abstract:High-density surface electromyography (HD-sEMG) gesture recognition supports prosthetic control, assistive robotics, and rehabilitation, but electrode re-donning and physiological variability cause distribution shifts that degrade accuracy across sessions and subjects. Generative HD-sEMG models primarily synthesize signals for augmentation; although diffusion models enhance representation learning, prediction still relies on a separate classifier. To tie learned dynamics to the decision rule, we propose MyoFlow, the first discriminative flow-matching framework for HD-sEMG recognition across sessions and subjects. It recasts classification as anchor-tied transport: a domain-conditioned rectified flow moves encoded windows toward gesture anchors that serve as transport targets and define the nearest-anchor decision geometry, enabling zero-shot prediction without an independent head. On the Hyser dataset, MyoFlow improves mean cross-session and cross-subject accuracy over the strongest diffusion-based baseline by 4.24% and 6.37%, respectively, and achieves 91.71% mean zero-shot accuracy and 97.39% mean few-shot accuracy across multiple days on the CEMHSEY dataset.
[LG-11] IRENE: A Convolutional GRU Ensemble Model for Radar Precipitation Nowcasting over Italy
链接: https://arxiv.org/abs/2609.17175
作者: Alessandro Camilletti,Gabriele Franch,Elena Tomasi,Marco Cristoforetti
类目: Machine Learning (cs.LG); Atmospheric and Oceanic Physics (physics.ao-ph)
*备注:
Abstract:We present IRENE (Italian Radar Ensemble Nowcasting Experiment), a deep learning model for probabilistic short-range precipitation nowcasting over the Italian domain at \SI1km spatial and 5 min temporal resolution. IRENE adopts an encoder–forecaster architecture built on multi-scale Convolutional Gated Recurrent Units (ConvGRUs), trained on the national radar composite produced by the Italian Civil Protection Department (DPC). An importance-sampling scheme focuses training on precipitation-relevant events, while the almost-fair Continuous Ranked Probability Score (afCRPS) is adopted as the primary probabilistic loss function. Two additional training configurations are proposed: an adversarial (GAN) variant, IRENE-GAN, designed to improve the spatial sharpness of the generated forecasts, and a spectrally constrained variant, IRENE-GAN-RAPSD, in which the adversarial objective is complemented by an explicit penalty on the radially averaged power spectral density. The three configurations are evaluated against the stochastic extrapolation method STEPS and the pre-trained deep learning model DGMR. All IRENE configurations attain a lower Continuous Ranked Probability Score than both benchmarks at every lead time and rank histograms closer to uniformity, indicating better probabilistic skill and ensemble calibration. In terms of ensemble-mean mean absolute error the advantage is confined to the first 90 min, beyond which the strongly damped DGMR fields and, to a lesser extent, STEPS become competitive. Spectral analysis shows that the adversarial training removes the progressive loss of small-scale variance exhibited by IRENE, at the cost of an excess of fine-scale power at long lead times that the spectral penalty only partially controls.
[LG-12] Neural Field Ensembles for Aerodynamic Surface Prediction: Winning Solution to the ONERA CRM Wall Distribution 2025 Challenge
链接: https://arxiv.org/abs/2609.17160
作者: Lionel Salesses,Caroline Sainvitu,Tariq Benamara
类目: Machine Learning (cs.LG); Fluid Dynamics (physics.flu-dyn)
*备注:
Abstract:Machine-learning surrogate models offer a promising alternative to high-fidelity Computational Fluid Dynamics (CFD) simulations for aerodynamic analysis and design. However, constructing accurate surrogates for realistic aircraft configurations remain challenging due to complex geometries, multiple flow regimes, and limited training data. This work presents the methodology that achieved first place in the ONERA CRM Wall Distribution Regression Challenge, which focuses on predicting pressure and skin-friction coefficient distributions over the NASA Common Research Model wing-body-pylon-nacelle configuration under different operating conditions. The proposed approach formulates the problem as a conditional neural field mapping spatial coordinates, surface normals, and operating conditions to aerodynamic wall quantities. Fourier feature encoding, a relative squared error objective aligned with the challenge metric, ensemble learning, and k -fold cross-validation are progressively introduced to improve prediction accuracy and exploit the limited training data. Beyond presenting the final methodology, the paper documents the successive model design choices that led to the winning solution through a comprehensive ablation study and discusses several alternative approaches that were investigated but ultimately discarded. On the hidden competition test set, the proposed methodology achieves an overall score of 8.81, outperforming the strongest organizer-provided baseline, which achieved a score of 8.64, while requiring approximately three orders of magnitude fewer trainable parameters. These results illustrate that carefully designed coordinate-based neural fields constitute an efficient and robust framework for aerodynamic surrogate modeling on complex geometries under limited-data conditions.
[LG-13] Intrinsic Robot Rewarding: Reusing VLA Representations for Autonomous Evaluation and Policy Improvement
链接: https://arxiv.org/abs/2609.17115
作者: Tobias Schaffer,Mohab Elkhayat,Daniela Nicklas,Mustafa Almohamad,Elham Al-Fuqara
类目: Robotics (cs.RO); Machine Learning (cs.LG)
*备注:
Abstract:Vision-language-action (VLA) systems already bring together two valuable resources for robot learning: rich visual representations and demonstrations of successful task execution. Intrinsic Robot Rewarding (IRR) proposes to use these resources for a second, complementary purpose: evaluating the robot’s own outcomes and providing feedback for policy improvement. Successful demonstration endpoints define task-specific references, and the policy’s frozen visual encoder provides the feature space in which new outcomes are assessed. The core reward mechanism adds a reference bank and a scoring operation to the existing pipeline, without requiring a separate learned evaluator or an additional perception backbone. Our position is that this reuse offers a promising route to lower integration effort, efficient reward computation, and reduced recurring human outcome scoring. Building on established research in visual rewards and learning from experience, IRR brings these ideas into the robot’s existing perception and demonstration pipeline. An operational COMAU Racer 3 demonstrator is available at technology readiness level 4 (TRL 4). This laboratory foundation supports the next research step: connecting internal outcome evaluation to physical policy improvement. We present the reward formulation, central research questions, and an evaluation methodology linking reward reliability to task success and supervision effort. The intended contribution is a reusable approach to learn and improve from the data and experience already available in industrial robot systems.
[LG-14] High-Fidelity Digital Twin Data Models by Randomized Dynamic Mode Decomposition and Deep Learning with Applications in Fluid Dynamics
链接: https://arxiv.org/abs/2609.17101
作者: Diana A. Bistrian
类目: Machine Learning (cs.LG); Numerical Analysis (math.NA)
*备注:
Abstract:The purpose of this paper is the identification of high-fidelity digital twin data models from numerical code outputs by non-intrusive techniques (i.e., not requiring Galerkin projection of the governing equations onto the reduced modes basis). In this paper the author defines the concept of the digital twin data model (DTM) as a model of reduced complexity that has the main feature of mirroring the original process behavior. The significant advantage of a DTM is to reproduce the dynamics with high accuracy and reduced costs in CPU time and hardware for settings difficult to explore because of the complexity of the dynamics over time. This paper introduces a new framework for creating efficient digital twin data models by combining two state-of-the-art tools: randomized dynamic mode decomposition and deep learning artificial intelligence. It is shown that the outputs are consistent with the original source data with the advantage of reduced complexity. The DTMs are investigated in the numerical simulation of three shock wave phenomena with increasing complexity. The author performs a thorough assessment of the performance of the new digital twin data models in terms of numerical accuracy and computational efficiency.
[LG-15] Bio-Inspired Palette Evolution in Indirectly Encoded Substrates: Timescale Compatibility Shapes Activation Function Discovery PPSN
链接: https://arxiv.org/abs/2609.17067
作者: Romain Claret,Michael O’Neill,Paul Cotofrei,Kilian Stoffel
类目: Neural and Evolutionary Computing (cs.NE); Machine Learning (cs.LG)
*备注: 16 pages, 2 figures, 7 tables. Authors’ accepted manuscript; published in Parallel Problem Solving from Nature - PPSN XIX (Springer, Lecture Notes in Computer Science)
Abstract:Indirectly encoded neural networks can assign different activation functions to individual nodes, but the right functions are rarely known in advance. When the available set contains only standard monotonic functions, problems like parity become unsolvable, yet an all-inclusive palette underperforms a curated one. How should evolution discover which functions to use? We address this as a meta-learning problem, designing 13 strategies (11 inspired by biological adaptation mechanisms, plus baseline and oracle controls) that modify the set of available activation functions during evolution. Each strategy translates a biological principle into an evolutionary operator: for example, circadian-inspired oscillatory gating cycles functions in and out of the palette on a fixed schedule, while immune-inspired Clonal Selection permanently protects functions that consistently correlate with fitness. We evaluate all strategies across more than 3,000 runs on parity and non-parity problems, first evolving the activation palette alone, then co-evolving a per-node aggregation palette on harder problems; an independent replication with new seeds confirms a stable high-reliability tier, with Circadian holding its top rank. Bio-inspired strategies match the solve rate of a tuned baseline but converge up to twice as fast, with Circadian halving total compute. Strategy rankings reverse across problem types, with no strategy dominating all domains. Strategy success is largely shaped by timescale compatibility: strategies whose characteristic timescale matches the evolutionary evaluation window consistently outperform those that operate too slowly. The practical guideline: match the mechanism’s timescale to the evaluation budget. Rescaling the slowest strategy bypasses the oscillatory barrier entirely: all nine solutions solve parity with non-oscillatory activations paired with min or max aggregation.
[LG-16] Near-Optimal Nonconvex Matrix Completion
链接: https://arxiv.org/abs/2609.17048
作者: Jian-Feng Cai,Xiliang Lu,Juntao You
类目: Numerical Analysis (math.NA); Machine Learning (cs.LG)
*备注:
Abstract:We study nonconvex methods for matrix completion, the problem of recovering a low-rank matrix from a subset of its entries. Convex methods achieve sample complexity linear in the matrix dimension and the rank, up to logarithmic factors, whereas global guarantees for commonly used nonconvex methods require a higher polynomial dependence on the rank. We close this gap by analyzing Riemannian gradient descent (RGD) and Riemannian Gauss–Newton (RGN) methods. For an n\times n matrix of rank r with incoherence parameter \mu and condition number \kappa , the two methods achieve exact recovery with high probability from O(\mu nr\log n\log(n\kappa)) and O(\mu nr\log n\log(2\mu r\kappa)) observations, respectively. The methods use a multiscale residual initialization, while the analysis simultaneously controls the spectral error and incoherence. The resulting RGD iterates converge linearly, whereas RGN eventually converges Q-quadratically.
[LG-17] Learning Options for Compositional Motor Control with Adapter Banks
链接: https://arxiv.org/abs/2609.17042
作者: Sreejan Kumar,Marcelo Mattar,Lea Duncker
类目: Machine Learning (cs.LG); Robotics (cs.RO); Neurons and Cognition (q-bio.NC)
*备注:
Abstract:Learning flexible motor primitives is a hallmark of skilled motor control. Recent neuroscience theory proposes that motor primitives may be implemented as low-rank perturbations of a shared recurrent network, but leaves open how such a system is learned. We translate this principle into a novel architecture for learning motor skills end-to-end: a shared recurrent core modulated by a bank of residual adapters, each selected by a discrete latent code. Trained on closed-loop biomechanical control, the adapters develop emergent low-rank perturbations of the recurrent dynamics despite no architectural rank constraint, placing task representations in disparate subspaces of the shared core network. A simple high-level policy over the learned options, optimized while the whole network is frozen, sequences the low-rank adapters to produce novel out-of-distribution movements. We demonstrate the ability to generalize to novel motor sequences within the closed-loop control setting, improving on the generalization error of a task-input-conditioned multitask baseline by upto order of magnitude.
[LG-18] Structural Negative Transfer in Federated Graph Neural Networks: Diagnosis Causal Investigation and the Limits of Divergence-Aware Mitigation
链接: https://arxiv.org/abs/2609.16977
作者: Chethana Prasad Kabgere,Shylaja SS
类目: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC)
*备注: Working Paper Draft
Abstract:Federated learning lets multiple participants train a shared model without pooling raw data, by exchanging locally trained model updates instead. Federated averaging assumes that averaging local models is a reasonable way to solve one shared problem when participants’ data are broadly similar. Work on non-IID federated learning has shown that this assumption can withstand differences in label and feature distributions. We ask whether it survives a different strain specific to graph neural networks, where client graphs differ not in label or feature distribution but in structure itself, requiring the same shared weights to operate over fundamentally different topologies. We call the resulting harm structural negative transfer. In a federation of real citation networks and synthetic structural proxies, a structurally atypical client lost more than half its achievable accuracy simply by joining. In an initial six-client federation, two label-free structural statistics computable before training were strongly associated with this harm. Expanding to twenty clients showed that degree divergence remained associated with harm, although more weakly, and survived removal of domain contrast. Spectral divergence did not replicate, which we trace to a confound caused by the composition of the reference pool used for leave-one-out statistics. A causal intervention isolating topology found no significant effect. A degree-normalization mechanism held across twenty-four seeds but did not explain the harm when corrected. The best of five candidate fixes beat a tuned baseline only until a matched, structurally blind control was applied, after which the gain disappeared. What survives is a modest, partially replicated, degree-specific signal that is not yet a validated predictor at scale.
[LG-19] HyCoSeq: Contextual Hyperbolic Representation Learning for Genomic Sequences
链接: https://arxiv.org/abs/2609.16925
作者: Chenhao Zeng,Zhibin Pu,Shufei Ge
类目: Machine Learning (cs.LG); Genomics (q-bio.GN); Machine Learning (stat.ML)
*备注:
Abstract:Hyperbolic geometry provides a natural inductive bias for genomic representation learning, but existing hyperbolic genomic models primarily use Lorentz convolutions to learn local sequence representations, while their residual pathways do not directly aggregate full Lorentz representations. We propose HyCoSeq, a contextual hyperbolic representation learning framework for genomic sequences. HyCoSeq incorporates weighted Lorentzian residual aggregation into multi-curvature Lorentz encoding, allowing full Lorentz representations to participate directly in geometry-consistent local aggregation. It further introduces a bidirectional long short-term memory network that integrates information from both sequence directions to learn contextual relationships among local representations at different positions within a genomic sequence, thereby extending local hyperbolic convolutional encoding to sequence-level contextualized representations. Extensive experiments across diverse genomic tasks show that HyCoSeq outperforms existing hyperbolic baselines and, without large-scale genomic pretraining, achieves competitive performance against substantially larger pretrained DNA language models.
[LG-20] OptiPrime: Optimizing Private Inference through Protocol-Hardware Co-design MICRO MICRO2026
链接: https://arxiv.org/abs/2609.16898
作者: Jiangrui Yu,Ye Yu,Si Chen,Chenqi Lin,Wenxuan Zeng,Junfeng Fan,Mingyu Gao,Meng Li
类目: Hardware Architecture (cs.AR); Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注: Accepted to the 59th IEEE/ACM International Symposium on Microarchitecture (MICRO 2026)
Abstract:Private deep neural network (DNN) inference based on hybrid homomorphic encryption (HE) and multi-party computation (MPC) can protect user data with a formal guarantee, but at the cost of significant latency overhead due to HE. Customized HE accelerators have been proposed and have achieved orders-of-magnitude speedup for individual HE operations. However, when directly applying a commercial HE accelerator to state-of-the-art HE-MPC frameworks, we observe only limited end-to-end performance gain. This is because HE-MPC frameworks often require wireless transmission of input and output ciphertexts for each HE operation, leading to a severe network communication bottleneck. To overcome this challenge, we introduce OptiPrime, a protocol-hardware co-optimization framework for efficient private DNN inference. OptiPrime features a novel HE protocol for convolutions that substantially reduces the number of transmitted output ciphertexts and mitigates the network communication bottleneck. Meanwhile, as the new protocol introduces complex computation for fewer output ciphertext, we observe new memory access challenges due to a high volume of weight plaintexts and intermediate ciphertexts. Hence, we further propose a lightweight compression system for the weight plaintexts, reducing memory traffic by 10 times, as well as a specialized dataflow to maximize on-chip data reuse of intermediate ciphertexts. Extensive experiments show that our framework outperforms the Cheetah baseline by at most 5.7 times on CPUs and 4.2 times with an accelerator.
[LG-21] Can Deep Learning Achieve Cross-Physics Mapping?
链接: https://arxiv.org/abs/2609.16853
作者: Pengfei Zhu,Julien Lecompagnon,Mathias Ziegler
类目: Machine Learning (cs.LG); Applied Physics (physics.app-ph)
*备注:
Abstract:Can deep learning translate physical fields governed by fundamentally different equations? We address this question by introducing Cross-Physics Mapping (CPM), an operator-learning framework for mappings between heterogeneous physical domains. We formulate sufficient conditions for such mappings through compatible latent representations and propose a dimensionless scaling principle that aligns the characteristic evolution scales of the source and target systems without assuming their dynamical equivalence. As a representative test, paired diffusion and wave fields are generated independently from their respective parabolic and hyperbolic equations while sharing the same latent geometry, material heterogeneity, excitation, and dimensionless scale. Seven architectures-ResUNet, DeepONet, Fourier, latent, wavelet, U-shaped, and Galerkin neural operators-are evaluated for both diffusion-to-wave and wave-to-diffusion mappings. The results reveal a strong directional asymmetry. Diffusion-to-wave reconstruction is more challenging because it requires recovering wavefront, phase, and time-of-flight information attenuated by diffusion; U-NO performs best in this direction, achieving a relative \ell_2 error of 0.307 and an R^2 of 0.905 . Wave-to-diffusion mapping is considerably more stable, with GNO attaining a relative \ell_2 error of 0.154 and an R^2 of 0.935 . Neural operators generally outperform the conventional convolutional baseline, highlighting the nonlocal nature of cross-physics transformations. These findings demonstrate that deep learning can establish useful mappings between distinct physical modalities on a shared latent manifold, while the achievable accuracy remains fundamentally constrained by the direction-dependent information content of the governing physics.
[LG-22] Information Geometric Self-Organization at the Edge of Stability in High-Capacity Kernel Associative Memories
链接: https://arxiv.org/abs/2609.16827
作者: Akira Tamamori
类目: Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
*备注: 8 pages, 3 figures
Abstract:High-capacity associative memories based on Kernel Logistic Regression (KLR) exhibit exceptional storage capabilities and robustness. Previous empirical studies identified a hyperparameter regime, the “Ridge of Optimization,” where attractor stability is maximized. However, the geometric nature of this regime and the optimization dynamics required to reach it have remained unclear. In this paper, we investigate the static geometry of the parameter space and the learning trajectory of Gradient Descent (GD) in KLR-trained Hopfield networks. Using the eigenvalue spectrum of the Hessian, we reveal that the Ridge corresponds to a phase boundary located adjacent to a rank-1 spectral collapse, acting as a geometric singularity where the principal curvature is massively amplified. Furthermore, we demonstrate that the learning dynamics exhibit a transient self-stabilizing behavior driven by the Edge of Stability (EoS) phenomenon. Rather than seeking flat regions, the network parameters are driven toward a state where the local curvature dynamically equilibrates near the stability limit dictated by the learning rate, allowing the optimization to survive the initial instability. We provide analytical derivations for both the rank-1 asymptotic collapse and the dynamic feedback loop governing this equilibration. These findings suggest that optimal, high-capacity memory representations are not formed in flat minima, but are dynamically sculpted at the highly curved boundaries of geometric singularities.
[LG-23] Adapting to Decision-Relevant Non-Stationarity in Decentralized Heterogeneous Bandits
链接: https://arxiv.org/abs/2609.16824
作者: Zhaojun Peng
类目: Machine Learning (cs.LG)
*备注: 90 pages, 18 figures
Abstract:Decentralized bandit systems often contain heterogeneous agents: rewards can change at individual agents even when the best action for the network stays the same. These local changes may cancel when rewards are averaged across agents, so the number of local changes \Stloc can be much larger than the number of changes in the best common arm \Stdec . We introduce Decision-Relevant Fresh Comparison (DRFC), which uses new, balanced samples from all agents to compare arms at the network level and switches only when fresh global evidence indicates that the common best arm has changed. We prove a high-probability dynamic regret bound with no adaptation term depending on \Stloc , and show that every algorithm must still pay for identifying genuine decision switches and propagating them through the communication graph. Under a distinct time-average benchmark, an anytime-valid sliding-window extension handles gradual drift; experiments on synthetic, semi-real, and MovieLens-1M replays show that DRFC ignores decision-irrelevant local changes while the extension avoids false switches.
[LG-24] LCAP: Population-Informed Latent Chip Adaptation from Few Output Probes for Photonic Neural Networks
链接: https://arxiv.org/abs/2609.16823
作者: Tianyu Gao,Guantian Zheng
类目: Machine Learning (cs.LG)
*备注: 5 pages, 3 figures, 2 tables
Abstract:Photonic neural networks (PNNs) offer efficient analog inference, but parameters optimized under ideal device models can degrade after fabrication, creating a persistent simulation-to-hardware (sim-to-real) gap. When many identically designed chips are deployed, calibrating each device from scratch compounds this cost. We propose Latent Chip Adaptation from Probes (LCAP), a population-informed framework that decomposes hardware adaptation into a transferable population correction and probe-inferred latent personalization. LCAP first learns a shared correction from 80 historical chips, then extracts a low-dimensional correction space from device-specific refinements. At deployment, 32 fixed unlabeled output probes infer an unseen chip’s latent correction coordinates, enabling feed-forward personalization without target-device optimization. On a three-layer 64-mode MZI simulator with phase variation, beam-splitter errors, quantization, and crosstalk, accuracy improves from 80.4147% under direct deployment to 92.6860% after shared calibration and 93.3617% with LCAP. LCAP improves 27/30 unseen chips and raises worst-device accuracy from 89.18% to 90.54%.
[LG-25] Geometry of learning dynamics: Gradient descent versus natural gradient on the ridge of optimization
链接: https://arxiv.org/abs/2609.16805
作者: Akira Tamamori
类目: Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
*备注: 11 pages, 5 figures
Abstract:High-capacity associative memories based on Kernel Logistic Regression (KLR) exhibit a “Ridge of Optimization” characterized by extreme stability and a highly skewed weight spectrum. However, the dynamical process by which learning converges to this critical regime has remained unclear. This paper provides a geometric analysis of the learning trajectories on the statistical manifold of a KLR-trained Hopfield network. By comparing the paths of Gradient Descent (GD) and Natural Gradient Descent (NGD), we elucidate the mechanisms governing the optimization process. Our analysis reveals that learning on the Ridge proceeds in two distinct phases. We show that the extreme curvature of the Ridge causes standard GD to follow a highly oscillatory, non-geodesic path. In stark contrast, NGD explicitly corrects for this geometry, following the ideal geodesic path and completely overcoming the instabilities faced by GD. We demonstrate experimentally that NGD not only converges significantly faster but also achieves a solution with superior generalization performance. These results establish that the highly structured geometry of the Ridge is optimally suited for information-geometric optimization, providing a new perspective on the interplay between learning dynamics and emergent representation geometry.
[LG-26] Constant Swap Regret in General-Sum Games via Optimistic Transition Matrices
链接: https://arxiv.org/abs/2609.16751
作者: Tung Mai
类目: Computer Science and Game Theory (cs.GT); Machine Learning (cs.LG)
*备注:
Abstract:We give deterministic and uncoupled learning dynamics for finite multiplayer general-sum games under full-information feedback that achieve constant individual swap regret, independent of the horizon T . With n players and at most m actions each, the individual swap regret of every player is O(\sqrtn m \log m \log^5/2(nm)) at every finite horizon. Each player predicts the deviation gains, then uses these predictions to update a row-stochastic transition matrix, and plays its stationary distribution. The proof combines a potential argument exploiting stationarity with a two-scale higher-order prediction analysis, using rooted-tree representations to handle the nonlinear dependence of deviation gains on the stationary distributions. An adversarially robust variant, obtained through a generic common-prefix switching wrapper, preserves the self-play bound up to a universal constant and guarantees individual swap regret at most 7\sqrtm T \log m in the adversarial setting.
[LG-27] he Latent That Never Was: A Forensic Re-run of the CVAE Ablation in Action Chunking Transformer
链接: https://arxiv.org/abs/2609.16745
作者: Bo Kang
类目: Robotics (cs.RO); Machine Learning (cs.LG)
*备注:
Abstract:Action Chunking Transformers (ACT) are widely used to learn robot manipulation from demonstrations. Their conditional variational autoencoder includes an encoder meant to capture differences between demonstrations during training. The original ACT paper reported that encoder removal dropped the mean success rate from 35% to 2% on two simulated tasks with human demonstrations. We re-ran this ablation in the original code and checked whether the findings depend on the implementation or training data. The published drop does not reappear in our tests, although smaller gains or losses in success rate remain uncertain. To investigate the discrepancy, we varied training length and how checkpoints are selected for evaluation. Both can reverse which policy scores higher, but the published drop’s cause remains unknown. Success rates alone leave open whether the encoder provides information that helps the policy reconstruct demonstrated actions. On the tested ACT benchmark, the sampled latent provides little reconstruction benefit at every tested nonzero weight of the penalty on latent information. At inference, ACT leaves this latent unused and sets it to zero. Skipping the encoder increases training throughput in both implementations we timed. We release code, evaluation tools and results so others can repeat the comparisons and test the encoder on other tasks.
[LG-28] A Systematic Evaluation of Machine Learning Methods for Fault Detection and Line Identification in Electrical Power Grids ICASSP2025 ICASSP49660
链接: https://arxiv.org/abs/2609.16744
作者: Julian Oelhaf,Georg Kordowich,Paula Andrea Pérez-Toro,Tomás Arias-Vergara,Andreas Maier,Johann Jäger,Siming Bayer
类目: Machine Learning (cs.LG); Signal Processing (eess.SP)
*备注: Accepted at ICASSP 2025. 5 pages, 4 figures. Published version: DOI https://doi.org/10.1109/ICASSP49660.2025.10890544
Abstract:The integration of renewable energy sources into the electrical grid introduces complex challenges in fault detection and coordination of grid recovery mechanisms. Traditional relay protection systems, which operate based on static rules and predefined thresholds, are inadequate for addressing these challenges, particularly in detecting and isolating faults such as short circuits. Consequently, the conventional methodologies applied to electrical network protection frequently fail to achieve optimal performance in fault detection, especially in terms of adherence to safety standards and the selective limitation of damage. Recent research indicates that machine learning (ML)-based approaches can effectively tackle these issues; however, variations in grid configurations and analysis windows have impeded consistent comparative assessments. In this study, we assess the efficacy of various ML models in detecting electrical faults and pinpointing defective transmission lines within a 10 ms measurement interval - a critical time-frame for real-time operational viability, for the first time. The most effective model attained an F1 score of 0.991 +/- 0.018 and demonstrated a processing time of 0.342ms +/- 0.509ms.
[LG-29] Carry-Through Checksum: A Lightweight Fault-Detection for CNN Inference at the Edge
链接: https://arxiv.org/abs/2609.16742
作者: Kyrylo Nazarevych,Mohammad Hasan Ahmadilivani,Krister Kaldre,Davide Bertozzi,Jaan Raik
类目: Hardware Architecture (cs.AR); Machine Learning (cs.LG)
*备注: Accepted at ATS’26. 6 pages, 3 figs and 3 tables
Abstract:Convolutional Neural Networks (CNNs) are increasingly deployed in safety-critical edge applications, where soft errors can silently corrupt inference outputs and lead to unsafe decisions. Such applications typically rely on resource-constrained embedded GPUs, requiring fault detection and mitigation techniques that add minimal compute, memory, and latency overhead while integrating seamlessly with the standard GPU inference pipeline. Existing algorithm-based fault tolerance techniques rely on matrix augmentation and per-operation checksum verification, imposing substantial overhead that is prohibitive for CNN inference on embedded GPUs. In this work, we propose carry-through checksum, a fundamentally new scheme for soft-error detection in CNN inference on embedded GPUs. The method embeds dedicated carry-through filters into the convolutional layers, which compute a checksum from the CNN’s own operations and propagate it through inference, enabling end-to-end error detection with a single output verification. Experimental results on multiple CNN architectures show that the proposed method detects 95.86% and 86.56% of critical faults for FP32 and FP16, respectively, at almost no additional per-image overhead. Detected faults are mitigated through re-execution, incurring only 2.27% run-time overhead across the entire test set on an NVIDIA Jetson Orin NX GPU. Comments: Accepted at ATS’26. 6 pages, 3 figs and 3 tables Subjects: Hardware Architecture (cs.AR); Machine Learning (cs.LG) Cite as: arXiv:2609.16742 [cs.AR] (or arXiv:2609.16742v1 [cs.AR] for this version) https://doi.org/10.48550/arXiv.2609.16742 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-30] Unified Heterogeneous Graph Neural Network solver for Power Flow Optimal Power Flow and State Estimation
链接: https://arxiv.org/abs/2609.16738
作者: Ferran Bohigas-Daranas,Hamid Latif-Martínez,Eduardo Prieto-Araujo,Oriol Gomis-Bellmunt,Pere Barlet-Ros
类目: ystems and Control (eess.SY); Machine Learning (cs.LG)
*备注:
Abstract:Power Flow (PF), Optimal Power Flow (OPF), and State Estimation (SE) are fundamental problems in power system analysis, but solving them is computationally expensive. Graph Neural Networks (GNNs) have been proposed as fast surrogates, yet existing solvers are trained for a single problem at a time, producing narrow models that must be rebuilt for each new task. We propose a more general approach: a single Heterogeneous Residual Gated Graph Convolutional Network that solves all three problems with one shared backbone. Rather than learning one mapping, the model learns a reusable representation of how the network behaves, from which PF, OPF, and SE can each be estimated. Trained jointly on the three problems across diverse topologies and loading conditions, and evaluated on the IEEE 14-bus and 118-bus systems, the shared model matches the accuracy of task-specific GNN solvers and stays robust on unseen loading levels and topologies. These results show that a single model can capture the basic operation of a power network and serve several analysis tasks at once, a first step toward a foundation model for power systems. Subjects: Systems and Control (eess.SY); Machine Learning (cs.LG) Cite as: arXiv:2609.16738 [eess.SY] (or arXiv:2609.16738v1 [eess.SY] for this version) https://doi.org/10.48550/arXiv.2609.16738 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-31] Stable by Construction: Variational Latent Markov Operators for Long-Horizon PDE Prediction
链接: https://arxiv.org/abs/2609.16621
作者: Junyi Liao,Johann Guilleminot,Vahid Tarokh
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:Neural PDE solvers provide efficient surrogates for time-dependent physical systems, but autoregressive prediction over long horizons remains challenging because local errors can induce distribution shift and accumulate under recursive deployment. We develop a variational approach to this problem by introducing latent Markov dynamics in which physical states are represented by latent distributions and evolved through probabilistic transitions. The framework is formulated directly on function spaces and specialized to functional Gaussian models, where structured latent perturbations induce a spectral geometry and variational transition alignment regularizes the learned dynamics. We further analyze how these mechanisms affect autoregressive error propagation, providing a theoretical connection between variational training and long-horizon prediction. We instantiate the framework as the Variational Autoencoding Markov Operator (VAMO), which combines spatially resolved latent fields, structured Gaussian perturbations, and a neural-operator transition. Empirically, we demonstrate the effectiveness of VAMO on several fluid-dynamics benchmarks with prediction horizons extending substantially beyond those represented during training, where it consistently reduces error accumulation and improves rollout stability over several deterministic and noise-injection baselines. Overall, these results highlight variational modeling as a complementary approach to robust long-horizon neural PDE dynamics.
[LG-32] Divergence Timing and Cumulative Disagreement under KV-Cache Eviction
链接: https://arxiv.org/abs/2609.16617
作者: Xinyue Luo,Fei Yu
类目: Machine Learning (cs.LG)
*备注:
Abstract:KV-cache eviction perturbs the conditional token distributions governing autoregressive generation. We investigate how first-divergence timing and subsequent token mismatch determine cumulative disagreement. We derive an exact decomposition under a specified stepwise maximal coupling: the expected mismatch fraction equals a first-mismatch contribution plus post-divergence exposure multiplied by its mismatch rate. An explicit construction over unrestricted autoregressive kernel pairs realizes the sharp interval of risks compatible with a finite divergence-aligned observation window. Residual-branch conditional Monte Carlo provides unbiased joint estimates of occurrence, occupation, and window/tail contributions, with per-replicate variance dominance for total token loss. Complete trajectories from Meta-Llama-3.1-8B-Instruct and Qwen2.5-7B-Instruct show that SnapKV at 50% retention enters divergence later and less often than SnapKV-512 or recent-token retention with the same 50% prompt-cache budget, while post-divergence total variation (TV) remains high. In an exploratory analysis of 288 documents, post-divergence exposure accounts for 85-90% of four aggregate mismatch gaps. On 288 independent documents at 90% retention, prespecified comparisons show higher branch-aligned TV in the late than in the early window in both models.
[LG-33] A Weighted Kernel Method for Approximation that Adapts to Learned Multivariable Structure
链接: https://arxiv.org/abs/2609.16606
作者: John E. Darges,Laura Weidensager
类目: Machine Learning (cs.LG); Numerical Analysis (math.NA)
*备注:
Abstract:Approximating the input-output behavior of a multivariable black-box function from limited data is challenging when blind to the importance of its inputs and their interactions. We introduce total sensitivity kernels (TSKs), a method based on families of weighted ANOVA kernels that learn and adapt to this multivariable structure. TSKs parameterize the weights on each multivariable component of the target function by factors for each input. We propose learning these factors directly from function evaluations by selecting the reproducing kernel Hilbert space (RKHS) in which the target function has minimum norm. Under suitable conditions, we show that this norm-minimization problem admits a unique solution, and we establish consistency of a finite-data formulation based on minimum-norm interpolation. The learned TSK factors characterize the participation of individual inputs across interactions and main effects, providing a kernel-dependent notion of input sensitivity related to total Sobol indices. Numerical experiments demonstrate that adapting the kernel to learned multivariable structure can substantially improve approximation accuracy over a standard product kernel.
[LG-34] Recovering Physical Parameters from Frag mented Observations via Exact Distributed Spline Merging ICLR2027
链接: https://arxiv.org/abs/2609.16579
作者: Naveen Mysore
类目: Machine Learning (cs.LG); Computational Physics (physics.comp-ph)
*备注: 11 pages, 4 figures, 1 table. Under review at ICLR 2027
Abstract:Scientific measurements are frequently distributed across locations, time periods, and institutions. Combining such fragments into a continuous, differentiable field enables recovering governing physical parameters from its derivatives. This paper makes two contributions toward that goal. First, the established additive structure of fixed-basis ridge-regression statistics is applied to tensor-product spline fields: each data holder computes a local Gram matrix and moment vector, and the merged solution is mathematically identical to centralized fitting, with no raw data shared and no iterative synchronization. This property is specific to the fixed-feature squared-error setting; the present derivation does not establish an analogous guarantee for general jointly trained multilayer networks. Second, a complete pipeline connects distributed observations to physical parameter inference through field reconstruction, derivative extraction, and linear regression. The diffusion coefficient is recovered to 0.11% error and wave speed to 0.12% error; in both cases, distributed merging introduces zero degradation relative to centralized fitting. Application to 41 years of NOAA sea-surface temperature data confirms the result on real spatiotemporal observations.
[LG-35] AsyncCouple-Flow: Asynchronous Cross-Modal Coupling and Flow Matching for Spatio-Temporal Forecasting ICONIP2026
链接: https://arxiv.org/abs/2609.16573
作者: Zhixiang Wu,Yining Liu,Bo Zhao,Szu-Yu Chen,Huiran Duan,Chu Lin,Chuanguang Yang
类目: Machine Learning (cs.LG)
*备注: Accepted at the International Conference on Neural Information Processing (ICONIP 2026)
Abstract:Multi-modal spatio-temporal forecasting (MM-STF) supports weather nowcasting, traffic prediction, and earth-system modeling by combining heterogeneous sources such as physical fields, satellite imagery, and in-situ sensors. Three obstacles persist: (i) modalities have different spatio-temporal sampling rates, forcing lossy interpolation onto a unified grid; (ii) modalities are frequently missing at deployment due to sensor outages or revisit gaps, while most methods train with full availability; and (iii) autoregressive decoders accumulate errors over long horizons, amplified by multi-modal conditioning. We propose AsyncCouple-Flow to address these issues jointly. A Modality-Aware Token Sparsification (MATS) module performs scale-aware tokenization and uses a shared importance scorer to select top-k tokens per timestep, producing equal-length sequences. An Asynchronous Cross-Modal Coupling Graph (ACCG) replaces fixed cross-attention with a learnable graph whose edges encode time offsets, semantic similarity, and modality-specific physical priors, enabling fusion under arbitrary asynchrony and missingness. A Flow-Matching Forecasting Head models multi-step prediction as a conditional ODE, trained with stochastic modality dropout and integrated jointly to avoid autoregressive drift. Experiments on ERA5+GOES+ISD weather forecasting and PEMS-BAY traffic prediction with multi-source side information show that AsyncCouple-Flow outperforms state-of-the-art baselines and remains robust with up to two missing modalities. The code will be released upon acceptance.
[LG-36] FlowATC: Aircraft Trajectory Prediction via Flow Matching
链接: https://arxiv.org/abs/2609.16528
作者: Mathurin Petit,Emir Torun,Louis Brusset,Jordan Kam,Alexandre M. Bayen
类目: Machine Learning (cs.LG)
*备注:
Abstract:Building accurate decision-support tools for next-generation air traffic control requires robust trajectory prediction models. We present a flow-matching architecture trained exclusively on historical aircraft trajectories, with no route labels or chart supervision. Trained on 1.15 million Automatic Dependent Surveillance-Broadcast trajectory windows collected over the San Francisco Bay Area, the model generates aircraft trajectory distributions that closely match historical traffic, reproducing known airspace structure around San Francisco Airport such as the shape of SFO’s published NIITE FOUR departure procedure. Our model is trained directly on the native, irregular ADS-B sampling interval. Trajectory prediction is cast as sequence inpainting using a block-causal Transformer that denoises future state tokens conditioned on the observed history using Conditional Flow Matching or Denoising Diffusion Probabilistic Models. We compare our architecture against constant-velocity, deterministic-Long Short Term Memory, and Conditional Variational Autoencoders baselines. At matched parameter count, CFM outperforms DDPM by 11-26% in minADE@20, and both generative objectives surpass the CVAE baseline by 31-41%. We further show that the error degrades gracefully with prediction horizon, and the architecture remains effective when retrained on temporally decimated feeds. Lastly, we sample K independent completions, yielding spatial probabilistic occupancy estimates that can serve as input to downstream conflict-risk estimation.
[LG-37] High-Performance Tensor Formulation of the Viterbi Algorithm for Hidden Semi-Markov Models
链接: https://arxiv.org/abs/2609.16500
作者: Lorenzo Piarulli,Elia Belli,Daniele De Sensi
类目: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC); Data Structures and Algorithms (cs.DS)
*备注:
Abstract:Hidden Semi-Markov Models (HSMMs) are fundamental probabilistic models widely adopted across diverse domains, from computational biology to finance and signal processing. The Viterbi algorithm decodes the most likely state sequence given an HSMM and can be applied iteratively for ab initio model learning. However, existing Viterbi implementations remain sequential, and GPU-accelerated solutions are entirely absent, making HSMM decoding impractical for large-scale workloads. We present a tensor-based formulation of the Viterbi algorithm for HSMMs, restructuring the inner loops into tensor operations that naturally map onto SIMD units and massively parallel architectures. Building on this formulation, we provide optimized implementations spanning single- and multi-core CPUs, and, for the first time, GPU. Experimental evaluation demonstrates speedups of up to 14x on a single core, over 200x with multi-core, and over 570x on GPU over the state-of-the-art sequential baseline, establishing a new performance baseline for large-scale HSMM decoding.
[LG-38] Online Gradient Computation for Warping Gaussian Process Transformations
链接: https://arxiv.org/abs/2609.16472
作者: Emilio Ruiz-Moreno,Konstantinos Slavakis,Baltasar Beferull-Lozano
类目: Machine Learning (cs.LG); Signal Processing (eess.SP)
*备注:
Abstract:Warped Gaussian processes (GPs) handle non-Gaussian observations by mapping them into a latent standard GP via a parametric transformation called warping. Existing streaming variants, however, either optimize the warping parameters periodically or sacrifice analytical tractability for a higher model capacity. To bridge this gap, we show that the gradient of the instantaneous negative log-likelihood of a warped GP admits an exact recursive computation. Based on this result, we propose a novel online method for warped GPs that jointly updates the latent GP moments and optimizes the warping parameters.
[LG-39] Not All Relations Are Equal: Relation-Balanced and Calibrated Graph Learning for Provenance-Based Intrusion Detection
链接: https://arxiv.org/abs/2609.16462
作者: Lijie Zheng,Ji He,Alessandro Brighente,Yulong Shen,Mauro Conti
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注: 6 pages
Abstract:Provenance-Based Intrusion Detection Systems (PIDSs) detect Advanced Persistent Threats (APTs) by analyzing system interactions. However, existing methods largely treat relations uniformly, overlooking statistical heterogeneity; in CADETS, relation frequencies differ by approximately 140,000\times . This may cause PIDSs to focus more on frequent relations and overlook differences in normal error levels across relations, increasing the risk of false alarms and missed detections. We present RECAL, an unsupervised framework using relation-balanced masked graph learning to better capture rare interaction patterns. It further calibrates reconstruction errors against each relation’s benign error distribution to produce comparable anomaly evidence, helping distinguish attacks from benign behavior and reduce false alarms. On three DARPA E3 datasets, RECAL achieves F1 scores of 99.99%, 99.93%, and 99.99%, outperforming the best baseline on each dataset by 0.88, 0.82, and 0.42 percentage points, respectively. Compared with the baseline reporting the lowest FPR, RECAL reduces mean FPR by approximately 105\times , 4\times , and 41\times .
[LG-40] Adaptive Bayesian Partner Selection for Federated Clinical Centers
链接: https://arxiv.org/abs/2609.16446
作者: Navid Seidi,Satyaki Roy,Sajal K. Das
类目: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC)
*备注: 19 pages, 4 figures
Abstract:Federated learning (FL) in healthcare faces pronounced heterogeneity and temporal concept drift across clinical centers, where evolving patient populations and care practices shift data distributions. Existing approaches rely on persistent global communication, incurring substantial bandwidth overhead while risking negative transfer from poorly aligned peers. We propose Adaptive Bayesian Partner Selection (ABPS), a peer-to-peer framework that governs who collaborates, when, and at what cost. Each center maintains a Beta-Bernoulli posterior over prospective peers’ Shapley marginal utility, ranks candidates with an Upper Confidence Bound (UCB) criterion, and forms collaborations through a lightweight propose-reject mechanism, with the option to abstain from communication when no mutually beneficial partner exists. The framework admits a stochastic decision interpretation, yielding finite-sample concentration guarantees and O(kappa log T) regret in partner selection, along with conditions under which intentional isolation is optimal under negative transfer. Lightweight extensions (head personalization, bfloat16 quantized communication, and a tunable active-set size) further improve efficiency, and a goal-aware metadata filter enables institution-specific collaboration strategies. On binary in-hospital mortality prediction over the first 24 hours of an ICU stay, with 230 non-IID clinical centers drawn from MIMIC-IV, the full ABPS-X variant matches the strongest federated baseline (FedDyn, AUROC 0.758) at 0.09x the communication cost of FedAvg, with reduced variability. A diversity-driven configuration activates intentional isolation for a substantial fraction of centers. These results show that adaptive, utility-aware collaboration reduces communication without sacrificing accuracy when centers are numerous and small, offering a scalable paradigm for healthcare FL.
[LG-41] Physics Informed Random Feature Neural Networks for Solving PDEs
链接: https://arxiv.org/abs/2609.16406
作者: Chi-An Chen,Chunyang Liao,Ming Zhong
类目: Numerical Analysis (math.NA); Machine Learning (cs.LG)
*备注:
Abstract:Machine learning-based partial differential equations (PDEs) solvers have attracted significant attention in recent years. Most progress in this area has been driven by deep neural networks such as physics-informed neural networks (PINNs) and kernel method (such as physics-informed Gaussian Processes). We introduce a physics-informed random feature method for countering part of the spectral bias which PINN-based solvers are facing for a certain class of PDEs. Random feature method was originally proposed to approximate large-scale kernel machines and can be viewed as a specialized randomized neural network. Compared to other state-of-the-art PINN-based solvers which require a large number of collocation points, our proposed method reduces the computational complexity. In this paper, we develop a rigorous approximation error analysis and derive high-probability error bounds on the H^1 norm. We provide extensive numerical tests for verifying our theoretical guarantees on error decay rates, as well as several comparison tests to showcase our claimed capability for combating spectral bias in these deep learning based methods.
[LG-42] Implementing a White-Box Undetectable Backdoor for Random Fourier Features
链接: https://arxiv.org/abs/2609.16403
作者: Michael Collins,Jada Cumberland,Brianne Dunn,Ross Gore,Samuel Jackson,Sachin Shetty
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注: 13 pages, 4 figures, 4 tables
Abstract:Goldwasser et al. showed that undetectable backdoors can be planted in machine learning models trained with the Random Fourier Features (RFF) algorithm, under a hardness assumption tied to the Continuous Learning With Errors (CLWE) problem. Under standard cryptographic assumptions, even a full white-box audit of a model’s weights cannot detect this class of backdoor. The construction is stated in terms of cryptographic reductions and probabilistic lemmas, without a reference implementation, and relies on secondary machinery such as the Sparse Gaussian Pancakes distribution and a homogeneous CLWE conditional density. Its realizability in ordinary numerical code is not obvious from the paper alone. This paper implements the white-box CLWE-RFF backdoor construction end to end using only numpy and scipy, to test whether this threat is realizable with commodity scientific-computing tools or requires specialized cryptographic infrastructure. We give two samplers for the core GP_d(b_k) distribution. The first is a rejection-sampling proxy. The second is an exact closed-form sampler derived from the homogeneous CLWE density and verified against its own analytic form. Using this implementation, we run statistical indistinguishability tests, covering both weight-space and functional black-box comparisons. We find no evidence of detectable difference between backdoored and clean models across a range of sparsity ratios \rho = d_\textsparse/D . We report which parts of the construction were straightforward to realize, which required derivation not spelled out in the paper. We also highlight which parts we did not attempt to reproduce, including the underlying lattice hardness reduction. We see this work as a contribution to understanding the practical realizability of the Goldwasser white-box CLWE core, not as a new theoretical result. Comments: 13 pages, 4 figures, 4 tables Subjects: Cryptography and Security (cs.CR); Machine Learning (cs.LG) Cite as: arXiv:2609.16403 [cs.CR] (or arXiv:2609.16403v1 [cs.CR] for this version) https://doi.org/10.48550/arXiv.2609.16403 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-43] Bounded Adjustment with Reliability-Guided Embedding for Imbalanced Learning with Noisy Labels
链接: https://arxiv.org/abs/2609.16380
作者: Mushir Akhtar,Akarsh J.,M. Tanveer,Mohd. Arshad
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 20 pages, 2 figures, and 9 tables
Abstract:Class-balanced learning and label noise create a coupled failure mode: frequency correction prevents majority classes from dominating the decision rule, but can amplify incorrectly labeled minority examples. We introduce BARGE (Bounded Adjustment with Reliability-Guided Embeddings), a single-stage objective combining a bounded, prior-adjusted density-power score with reliability-guided angular geometry. Its classification score is strictly proper in the adjusted probability space and recovers balanced Bayes ordering under clean supervision and the true class prior. Under label contamination, its finite range bounds classification-risk perturbation at a fixed predictor, while its logit gradient redescends when the model confidently contradicts the supplied label. The adjusted target probability also weights class-equal feature compactness, and a one-sided separation term discourages aligned class directions. BARGE requires neither a noise rate nor a transition matrix, uses one network, and leaves inference unchanged. We evaluate it on CIFAR-10, CIFAR-100, and Tiny ImageNet under long-tail and step imbalance, clean labels, and 20% and 40% random incorrect-label replacement. Across 12 clean settings, BARGE ranks second overall and attains the lowest error in four. Under corruption, it achieves the lowest mean balanced error in all six dataset-corruption settings, reducing the six-setting average from 72.32% for the strongest competitor to 70.00%. It also obtains the highest macro-F1 and macro-AUPRC in every corrupted-label setting. Ablations show that class-equal angular compactness improves on the bounded score alone. These results support bounded predictive influence and reliability-guided geometry as complementary mechanisms for imbalanced learning with uncertain labels.
[LG-44] Certified Uncertainty Propagation in One-Shot Federated Bayesian Models via Posterior Event Transport
链接: https://arxiv.org/abs/2609.16373
作者: Mahyar Mohammadi,Mohammad Hossein Badiei,Abolfazl Yaghmaei,Hamed Kebriaei
类目: Machine Learning (cs.LG)
*备注:
Abstract:Probabilistic certification of Bayesian neural networks lower-bounds the posterior probability that a model satisfies a verifier-defined safety property. In one-shot federated Bayesian learning, however, the deployed model is obtained by aggregating parameters drawn from client-specific posterior distributions, so local certificates do not directly guarantee safety of the aggregated model. This paper develops a deployment-consistent certification framework by propagating local posterior events through the deployment aggregation rule, with an exact geometric characterization for Federated Averaging (FedAvg). Each client constructs disjoint hyper-rectangular regions in parameter space and computes their probability masses. The server forms Cartesian products of these regions, maps them through the deployment rule, and retains a product event only when its aggregation image is verified to satisfy the safety property. Under independent client posteriors, each product-event probability factorizes into local masses, and summing verified disjoint events yields a lower bound on safety probability of the deployed model. For FedAvg with nonnegative aggregation coefficients, the image of a Cartesian product of axis-aligned hyper-rectangles is exactly a weighted hyper-rectangle, introducing no set over-approximation. We distinguish the proposed transported-event certificate from direct certification under posterior distributions induced by FedAvg and Product-of-Gaussians aggregation. Experiments on MNIST and Fashion-MNIST under label-Dirichlet heterogeneity show that the transported FedAvg certificate ranges from 22.51% to 46.89%, while direct global certificates range from 72.05% to 91.39%. Results show that predictive accuracy and certifiable safety do not necessarily follow the same trend, and that global posterior constructions can exhibit distinct certification behavior across architectures.
[LG-45] Fast-Convergent Meta-RL via Gradient-Clustered BS Sampling for Edge Caching
链接: https://arxiv.org/abs/2609.16370
作者: Farnaz Niknia,Ping Wang
类目: Networking and Internet Architecture (cs.NI); Machine Learning (cs.LG); Systems and Control (eess.SY)
*备注:
Abstract:Wireless edge caching networks typically consist of many independent Base Stations (BSs), each facing its own request rate and content popularity profile. Training a Reinforcement Learning (RL) caching agent from scratch at every BS forces each agent to relearn, through slow trial and error, a decision problem that is structurally identical across the network. Meta-reinforcement learning removes this redundancy by learning a shared initialization that adapts to any BS in a few local updates; however, meta-training itself becomes the bottleneck at scale: the meta-gradient must be estimated from a small subset of BSs at each meta-iteration, and sampling this subset uniformly at random yields a high-variance estimate, an issue existing meta-RL caching frameworks leave unaddressed. This paper proposes a meta-reinforcement learning framework for caching across independent, non-overlapping BSs that directly targets this bottleneck. Each BS runs a local Proximal Policy Optimization (PPO) agent, formulated as a Semi-Markov Decision Process (SMDP) over content popularity, size, lifetime, and importance, while a shared meta-policy is learned via a Model-Agnostic Meta-Learning (MAML)-style loop. To scale meta-training and accelerate convergence, we introduce gradient-based clustering, which groups BSs by local gradient similarity and draws from every cluster, in proportion to its size, at each meta-iteration. We prove, via an Analysis of Variance (ANOVA)-style decomposition of gradient variance, that this strategy yields a strictly lower-variance meta-gradient estimator than uniform random sampling under BS heterogeneity.
[LG-46] Autonomous Droplet Navigation via Model-Based Reinforcement Learning
链接: https://arxiv.org/abs/2609.16369
作者: Rajneesh Anand,Mayuresh V. Kothare
类目: Machine Learning (cs.LG); Robotics (cs.RO); Systems and Control (eess.SY); Fluid Dynamics (physics.flu-dyn)
*备注: 43 pages, 15 figures, 3 tables including supplementary material. The source code is available via GitHub at this https URL . An archived version of all supplementary movies has also been uploaded to Google Drive: this https URL
Abstract:Precise manipulation of liquid droplets underpins lab-on-a-chip platforms for diagnostics, chemical synthesis, and biological assays. Yet autonomous droplet transport through confined geometries of varying complexity remains an open challenge. Droplets exhibit contact-angle hysteresis, deformability, and capillary pinning, which make their response to actuation nonlinear and history dependent, that classical controllers and pre-programmed trajectories cannot cope in multi-turn environments. Here we demonstrate autonomous navigation of a liquid droplet through geometries of increasing complexity on a gravity driven (Labyrinth) platform using model-based reinforcement learning. A thin silicone oil film reduces contact-line pinning while two-axis tilt supplies the gravitational driving force, and an overhead camera tracks the droplet in real time. An offline-trained policy discovers effective tilt strategies from limited physical interaction data, without simulation or analytical droplet models. The system operates under partial observability, as oil-film thickness, instantaneous contact angle, and droplet deformation state remain hidden from the controller. Despite these challenges, the learned policy achieves reliable navigation across straight, right-angle, and curved-arc paths, including outside-corner geometries. We further demonstrate that a policy trained on a simpler geometry transfers to complex ones, succeeding zero-shot on right-angle and staircase paths and reaching full success on a curved arc with a fifth of the training data. The findings suggest promising avenues for enabling droplet based microfluidic systems to serve as intelligent chemical laboratories.
[LG-47] EBL: Efficient Broad Learning for Distributed Adaptive Harmonic Analysis CEC
链接: https://arxiv.org/abs/2609.16358
作者: Changhong Li,Georgios Floros,Biswajit Basu,Shreejith Shanker
类目: Hardware Architecture (cs.AR); Machine Learning (cs.LG)
*备注: Accepeted by ICECS’26
Abstract:Renewable energy systems and electrified transport have found widespread adoption in recent years. The integration of these non-linear loads, dominated by electric vehicle (EV) charging, however, has introduced severe harmonic distortion into the power grid, impacting the efficiency and lifetime of substation equipment and switchgear in the distribution network. Rapid and high-precision harmonic analysis has hence become a prerequisite for effective harmonic control at the source of injection. This paper proposes an Efficient Broad Learning (EBL) framework for distributed adaptive harmonic estimation. As a quantised FPGA acceleration framework for BLS-style harmonic estimation, it offers high-accuracy estimation with half-cycle input, reconfigurable flexibility enabled by the FPGA implementation, and ultra-low latency, achieving 17.4 \times faster predictions than the nearest reported FPGA method. For harmonic prediction across multi-scenario charging and discharging nodes, the online transfer learning based on a closed-form solution rather than backpropagation in EBL demonstrates rapid adaptability. By exploiting bespoke quantisation and sparsity, the approach consumes 5.9% of the LUTs on the Zynq Ultrascale+ ZU7EV FPGA, using \approx 82% of the LUTs required by the state-of-the-art FPGA-accelerated estimator.
[LG-48] Federated stochastic bilevel optimization with fully first-order gradients IJCAI2025 IJCAI
链接: https://arxiv.org/abs/2609.16350
作者: Yihan Zhang,Rohit Dhaipule,Chiu C Tan,Haibin Ling,Hongchang Gao
类目: Machine Learning (cs.LG)
*备注: Accepted for publication in the Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence (IJCAI 2025). The official version is available at this https URL
Abstract:Federated stochastic bilevel optimization has been actively studied in recent years due to its widespread applications in machine learning. However, most existing federated stochastic bilevel optimization algorithms require the computation of second-order Hessian and Jacobian matrices, which leads to longer running times in practice. To address these challenges, we propose a novel federated stochastic variance-reduced bilevel gradient descent algorithm that relies solely on first-order oracles. Specifically, our approach does not require the computation of second-order Hessian and Jacobian matrices, significantly reducing running time. Furthermore, we introduce a novel learning rate mechanism, i.e., a constant single-timescale learning rate, to coordinate the update of different variables. We also present a new strategy to establish the convergence rate of our algorithm. Finally, the extensive experimental results confirm the efficacy of our proposed algorithm.
[LG-49] Multi-Label Proportion Learning for Sea-Ice Type Prediction
链接: https://arxiv.org/abs/2609.16347
作者: Samira Alkaee Taleghan,Younghyun Koo,Andrew P. Barrett,Farnoush Banaei-Kashani
类目: Machine Learning (cs.LG)
*备注:
Abstract:Sea-ice type prediction is important for climate monitoring, maritime navigation, and decision-making in polar regions. The main source of label data for this task is the ice chart, produced manually by ice analysts who interpret satellite imagery to delineate ice zones into polygons. Although ice charts are valuable, their production is labor-intensive and expensive, motivating recent efforts to automate the process using deep learning. However, deep learning models require patch-level (or pixel-level) label data for training, while ice charts provide only polygon-level annotations. As a workaround, supervised approaches often create approximate patch-level labels from polygon-level ice chart labels by assigning each sample the dominant ice type of its parent polygon. This approach enables supervised training but creates an ill-posed learning problem with intrinsically approximate solution. In this paper, we redefine sea-ice type prediction as a weakly supervised multi-label proportion learning problem to be able to directly use the polygon-level ice chart labels and avoid unnecessary label approximation for improved prediction accuracy. To address this problem, we propose a two-module framework where first Multiple Instance Learning (MIL) is used for water–ice classification, and then a multi-label proportion learning (MLPL) is introduced for ice-type composition prediction. We further extend this framework with a multimodal model that integrates SAR imagery with AMSR2 brightness temperatures and ERA5 reanalysis data through modality-guided auxiliary regularization. Evaluated on the AI4Arctic dataset, the SAR-only model reduces MAE by 14.5% and more than doubles mean ice-class F1 over the best supervised baseline. The multimodal model further reduces MAE by 21.5% and raises mean F1 by 41.2% over the SAR-only model, and by 52.7% over the supervised multimodal baseline.
[LG-50] Channel-Informed Neural Network for Physical Layer Key Generation
链接: https://arxiv.org/abs/2609.16341
作者: Jose Angel Sanchez Viloria,George Sklivanitis,Dimitris Pados,Elizabeth Serena Bentley
类目: Machine Learning (cs.LG); Signal Processing (eess.SP)
*备注:
Abstract:Physical-layer key generation (PKG) enables wireless devices to establish shared keys from reciprocal channel observations without directly exchanging the key. This capability is attractive for edge networks, where distributed and resource-constrained devices may require lightweight key establishment with limited access to centralized infrastructure. We introduce a channel-informed neural network for PKG that derives binary key features directly from received IQ measurements while explicitly grounding the learned representation in the underlying multipath channel. The proposed multi-task recurrent neural network jointly learns reciprocity-preserving binary features and an auxiliary channel estimate using a training objective that combines deep metric learning with channel-informed supervision. Structured channel sounding enables channel estimation from over-the-air measurements, while Sionna-RT ray tracing is used to augment training with additional propagation conditions. We evaluate the framework using indoor and outdoor software-defined-radio measurements collected on the POWDER radio testbed. Across all evaluated scenarios, the proposed model produces lower bit disagreement for reciprocal Alice-Bob observations than for Eve-related observations. Ray-traced data augmentation substantially improves key diversity, increasing the unique-key rate to 0.94, 0.99, and 0.99 across the indoor and two outdoor scenarios, respectively. Successfully reconciled channel-informed keys pass the selected NIST randomness tests prior to SHA-3 privacy amplification. The results demonstrate the potential of channel-informed representation learning for decentralized wireless key establishment while highlighting an important tradeoff between key diversity and reconciliation reliability.
[LG-51] Generative models for simulation based filtering: Formulations and Empirical Comparisons
链接: https://arxiv.org/abs/2609.16317
作者: Mohammad Al-Jarrah,Wei Deng,Bamdad Hosseini,Amirhossein Taghvaei
类目: Machine Learning (cs.LG)
*备注: 6 pages, 2 figures, 1 table
Abstract:This letter presents a unified formulation and a controlled numerical comparison of generative-model approaches to the nonlinear filtering problem. Under this formulation the analysis step is realized by a transport of the forecast distribution to the posterior, the approaches differing only in how that transport is selected and learned. We derive three new filters, based on stochastic interpolants, their deterministic flow-matching limit, and Schrödinger bridges realized through forward–backward SDEs. We develop a two-stage tuning procedure that separates the training of the generative model from its online refinement. The resulting methods are compared against the optimal transport filter (OTF), the Knothe–Rosenblatt filter (KRF), the sequential importance resampling (SIR) particle filter and the ensemble Kalman filter (EnKF), in terms of accuracy, computational time, and sensitivity to ensemble size and state dimension. The results indicate that every generative filter resolves multimodal posteriors that the EnKF and SIR do not, that no single generative framework dominates, the preferred method being set by the available online budget and ensemble size, and that the filters differ in the regularity of the particle trajectories they produce.
[LG-52] Robust Fault Detection in Mechanical Multimodal Time Series via Self-Supervised Cross-Modal Reconstruction
链接: https://arxiv.org/abs/2609.16314
作者: Magnus Munk Jensen,Dorte Hammershøi,Rafał Wiśniewski,Olga Fink
类目: Machine Learning (cs.LG); Applications (stat.AP)
*备注: Submitted to Reliability Engineering System Safety
Abstract:Fault detection is essential in industrial systems, enabling early identification of abnormal behaviour and improving safety, reliability, and operational efficiency. Modern systems increasingly rely on heterogeneous sensing modalities that capture complementary aspects of the underlying physical process. However, existing data-driven anomaly detection methods often process each modality independently or use simple feature-level fusion, limiting their ability to exploit cross-modal relationships that characterize normal system behaviour. Their performance also commonly assumes similar training and deployment distributions, whereas real-world operation is affected by changing operating conditions, environmental influences, and system degradation that induce distribution shifts and reduce detection performance, especially in unseen regimes. In this work, we propose a multimodal anomaly detection framework based on cross-modal reconstruction of heterogeneous time-series sensor data. Rather than modeling each modality independently, the framework learns system dynamics by reconstructing each modality from the others, thereby exploiting complementary information across modalities. This integrates information across sensing channels without requiring explicit temporal alignment or identical sampling rates, while improving robustness to sensor noise, missing measurements, and modality-specific disturbances. To address distribution shifts during real-world deployment, anomalies are identified using cross-modal reconstruction error and an adaptive test-time thresholding mechanism that adjusts to changing operating conditions. Experiments on three industrial case studies show strong fault detection performance and substantially improved robustness under out-of-distribution conditions, with the largest gains observed in the most challenging operating regimes. Comments: Submitted to Reliability Engineering System Safety Subjects: Machine Learning (cs.LG); Applications (stat.AP) Cite as: arXiv:2609.16314 [cs.LG] (or arXiv:2609.16314v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.16314 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-53] Agent ic Search Spaces for Tabular Machine Learning
链接: https://arxiv.org/abs/2609.16309
作者: Renat Sergazinov,Artem Chistyakov,Sergey Pankevich,Artem Babenko
类目: Machine Learning (cs.LG)
*备注:
Abstract:Despite the rapid progress of LLM-based agents for planning, code generation, and debugging, their practical value for tabular machine learning remains underexplored. In this paper, we investigate a concrete use case: whether state-of-the-art agentic AI systems can design extended HPO search spaces for established tabular models that outperform the standard search spaces provided by the model authors. Specifically, we represent each tabular model as a modular pipeline covering preprocessing, embeddings, architecture, training, and inference. We then task the agent to propose candidate code implementations for each module and use a classical HPO algorithm to jointly optimize over these candidates and the model’s default hyperparameters. Compared with the base HPO spaces, the expanded search spaces improve the performance of nearly every model family across a suite of 45 datasets, with average relative gains of 0.6%, rising to 2.0% on small-to-medium regression datasets. Notably, these gains come at no extra tuning cost: the enlarged spaces outperform the base under the same tuning and ensembling budgets. The gains transfer to the recent TabArena benchmark, where the agentic spaces improve the official Elo scores of four of the five model families and the two strongest agentic ensembles surpass the best AutoGluon ensemble of conventional models. Overall, our study suggests that LLM agents can provide practical value for tabular ML by expanding the design space.
[LG-54] Drift Field Net: Learning Ocean Lagrangian advection fields from in-situ and satellite observations
链接: https://arxiv.org/abs/2609.16288
作者: Théo Archambault,Pierre Garcia,Mattia Romero,Anastase Charantonis,Dominique Béréziat
类目: Machine Learning (cs.LG)
*备注: Submitted to Artificial Intelligence for the Earth Systems
Abstract:The North Pacific Subtropical Gyre (NPSG) is a major accumulation zone for floating plastic debris, resulting from basin-scale convergent ocean circulation. Effective cleanup strategies in this region rely on accurate forecasts of Lagrangian particle drift. Here, we introduce Drift Field Net (DFN), a deep neural network that predicts ocean surface flow fields from operational satellite observations. DFN is trained using a novel two-stage strategy that combines pretraining on simulated data with Lagrangian fine-tuning based on an advection-consistent loss function. This physics-informed optimization directly improves the accuracy of particle trajectory predictions. We evaluate DFN against an operational physics-based forecasting system and demonstrate the potential of deep learning for ocean surface flow prediction. On in situ drifter trajectories, DFN reduces the mean positioning error by 20 km after a 7-day forecast compared with the operational model. Furthermore, Lagrangian fine-tuning with the proposed advection loss further reduces the positioning error by 10 km, highlighting the benefits of incorporating Lagrangian constraints into the training process.
[LG-55] Differentially Private Semantic Plans for Aggregate Insight Generation
链接: https://arxiv.org/abs/2609.16283
作者: Behrooz Razeghi
类目: Machine Learning (cs.LG)
*备注:
Abstract:\textttURANIA provides end-to-end differential privacy (DP) for summaries of data-dependent clusters. However, its cluster–keyword release does not directly provide collection-wide aggregates for semantic concepts defined independently of the protected corpus. Records may express several concepts, records expressing the same concept may be assigned to different clusters, and cluster identities need not correspond across analyses. Consequently, cluster-level statistics do not directly provide comparable measurements of predefined concepts across collections or repeated analyses. We introduce \textttDP-SPIN, a trusted-curator framework for aggregate measurement and summarization over semantic concepts fixed independently of the protected target records. Each record is mapped to a bounded sparse nonnegative vector over these concepts, whose sum forms a semantic sketch. A differentially private mechanism releases a semantic plan containing admitted concepts and noisy masses; normalized semantic-support values and support bins are obtained by post-processing. For user-level privacy, each user’s aggregate contribution is clipped to a fixed bound. The language model receives only the plan and fixed decoding instructions, while a public verifier checks concept mentions, reported values, comparisons, and rank claims against the released plan. The final summary is differentially private by post-processing. We establish record- and user-level DP guarantees under add/drop and replacement adjacency. We evaluate \textttDP-SPIN under record-level privacy on CFPB complaint narratives, Amazon All Beauty reviews, and Yelp restaurant reviews, and under user-level privacy on Amazon and Yelp. We compare \textttDP-SPIN with non-private plan and summary references, DP keyword and category histogram baselines, and a \textttURANIA-style baseline with a fixed public keyword vocabulary.
[LG-56] Scaling Laws for Physics-Aware ACOPF Surrogate Learning
链接: https://arxiv.org/abs/2609.16282
作者: Yijiang Li,Emon Dey,Stefano Fenu,Massimiliano Lupo Pasini,Teja Kuruganti,Kibaek Kim
类目: Machine Learning (cs.LG)
*备注:
Abstract:Learning-based surrogates for AC optimal power flow (ACOPF) promise large speedups over classical solvers, but their operational value depends on physical feasibility as much as predictive accuracy. Physics-aware objectives such as the augmented Lagrangian (AL) improve constraint satisfaction at additional per-step cost, yet how this trade-off behaves with scale is uncharacterized. We sweep model and dataset sizes under both MSE and AL training, and characterize how constraint violation changes with network size across grids. Both objectives improve as power laws, but at different rates: MSE is governed primarily by model capacity, while AL is balanced across both. Violation grows roughly twice as fast with network size under MSE as under AL. On matched hardware, AL reduces violation by nearly 30\times for an order of magnitude more training time, with negligible added memory. The training objective determines not only where a surrogate lands but how its quality evolves with scale.
[LG-57] How I learned to stop worrying and love StopGrads: Stationarity Convergence and a case study on Flow Map Learning
链接: https://arxiv.org/abs/2609.16222
作者: Max W. Shen,Mark Goldstein,Zichu Wang,Aahlad Puli,Rajesh Ranganath
类目: Machine Learning (cs.LG)
*备注:
Abstract:Stopgrads are widely used in training machine learning models, but stopgrads can alter the gradient, stationary points and convergence guarantees of the original objective, which can make stopgrad training theoretically ungrounded. We introduce a stopgrad regression principle, which identifies a general template for stopgrad objectives with a closed-form characterization of stationary points and their uniqueness, unifying stopgrad objectives for flow maps, reinforcement learning, and diffusion samplers. We provide theoretical grounding for optimizing stopgrad flow map objectives by showing their unique stationary point is the true flow map, and showing positive convergence results for Eulerian and Lagrangian objectives, including MeanFlow and improved MeanFlow. Remarkably, we show that under functional semi-gradient flow, the learned flow map has a closed-form expression composing the initial flow map and the true flow map. We additionally use our stopgrad regression principle to propose modified stopgrad placements for flow map objectives which reduce training memory by 2x.
[LG-58] Skeletal Prototypes on Iterative Nerve Expansions
链接: https://arxiv.org/abs/2609.16170
作者: Jordan Eckert,Henry Schenck
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:Prototype reduction replaces a training set with a smaller representation, and the established methods return a finite set of points. We propose Skeletal Prototypes on Iterative Nerve Expansions (SPINE). The model for each class is an embedded 1-complex rather than a point set. Its initial edge set is a class-conditional Mapper graph, so the data decide which localized clusters are joined. Later phases fit the vertices under a classification objective, and an observation is assigned to the class whose complex is nearest. The segments therefore enter the decision rule and not only the fitting. We evaluate SPINE on seventeen benchmark datasets under stratified 10-fold cross validation, against seven other prototype reduction methods at a matched budget. SPINE attains the highest mean accuracy and the best average rank. It is significantly better than five of the seven competitors under Wilcoxon signed-rank tests with Holm correction. A budget sweep shows that the decision rule using the entire graph segments contribute most when prototypes are scarce, while the method as a whole competes best at moderate budgets. Construction cost places SPINE with the discriminative methods, and it is faster than generalized learning vector quantization on fourteen of the seventeen datasets.
[LG-59] LLM Inference in a Flash!
链接: https://arxiv.org/abs/2609.16161
作者: Sebastian Zhao,Minseo Kim,Coleman Hooper,Luca Manolache,Michael W. Mahoney,Yakun Sophia Shao,Kurt Keutzer,Amir Gholami
类目: Machine Learning (cs.LG)
*备注:
Abstract:Large Language Models (LLMs) have shown impressive capabilities across a range of natural language processing tasks, and LLM inference has emerged as a critical workload for enabling downstream applications. The demands of serving LLM inference are becoming increasingly challenging as requests shift toward longer sequences and heavier inference, driven by retrieval-augmented generation, inference-time compute scaling, and long-context applications. Additionally, these challenges are compounded by hardware trends, as memory capacity and communication bandwidth are not scaling as fast as increases in workload complexity. Compute-in-Flash is a promising solution to address memory bandwidth limitations by moving computation close to memory, and to exploit the large capacity of SSD technologies. However, it is challenging to deploy LLMs on these systems as they lack support for high-precision floating point operations and have limited write endurance. In our work, we aim to address these challenges by designing inference algorithms to enable LLM inference on Flash compute-in-memory devices. We present an end-to-end integer-only quantization approach to eliminate expensive floating-point computations. To address the limited write endurance, we design a dictionary-based KV cache compression strategy based on sparse dictionary coding that represents each KV vector as a linear combination of static dictionary vectors. These algorithmic improvements enable us to exploit the benefits of Compute-in-Flash for both model weights and KV cache, and to minimize expensive data transfer operations. Across Llama-3.1-8B and Qwen-2.5-7B, our combined method exhibits limited accuracy degradation while reducing dynamic KV cache traffic by 15 \times .
[LG-60] SWB-DM: A Calibrated Sliced-Wasserstein-Barycenter Aggregator with Delayed-Momentum Caching for Byzantine-Robust Federated Learning under Partial Participation
链接: https://arxiv.org/abs/2609.16099
作者: Saranraj S,Saranya M S,Alex David S,Ajay Kumar A
类目: Machine Learning (cs.LG); Cryptography and Security (cs.CR); Distributed, Parallel, and Cluster Computing (cs.DC)
*备注: 7 pages, 2 figures, 5 tables
Abstract:Robust aggregation methods for federated learning quietly rest on a fragile assumption: that whoever shows up in a given round is a fair sample of the full population. In practice, they rarely are. When only a handful of clients participate per round, even a modest fraction of adversaries can dominate that sample and silently invalidate the finite-sample guarantees that coordinate-wise median, Krum, Bulyan, and trimmed mean all depend on. We introduce SWB-DM to address this directly. SWB treats each slice of a client update as a one-dimensional distribution, computes a trimmed Wasserstein barycenter across clients, and recovers coordinate identity via a medoid-based gauge-fixing step – a heuristic we developed and do not claim it belongs to standard optimal-transport theory. DeMoA-style delayed momentum then caches updates across the full client population each round, decoupling robustness from whoever happened to be sampled. Trim ratio calibration is not cosmetic: under-trimming causes collapse at corruption levels a properly calibrated model survives. Across 448 CIFAR-10 configurations, plus CIFAR-100, FEMNIST, and a 500-client scalability run, we find several mechanistically distinct failure modes. Even-sample coordinate-wise median degrades to a deterministic wrong answer. Krum silently violates its own n greater than 2f+2 precondition and diverges without warning. Bulyan’s n greater than or equal to 4f+3 threshold produces a sharp pass/fail boundary. On attacks, IPM defeats order-statistic defenses – including SWB – more reliably than ALIE, confirmed through delta-space measurements against a convergence bound. SWB-DM’s cache carries a real warm-up cost, but extending all baselines to the same round budget shows its CIFAR-10 gains are disproportionately large. On CIFAR-100, FLTrust benefits more – for reasons entirely unrelated to caching. Comments: 7 pages, 2 figures, 5 tables Subjects: Machine Learning (cs.LG); Cryptography and Security (cs.CR); Distributed, Parallel, and Cluster Computing (cs.DC) Cite as: arXiv:2609.16099 [cs.LG] (or arXiv:2609.16099v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.16099 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Saranraj Sankar [view email] [v1] Mon, 14 Sep 2026 16:30:18 UTC (214 KB)
[LG-61] Evaluating Open-Weight E-Commerce Agents with Environment-Grounded Verification
链接: https://arxiv.org/abs/2609.16093
作者: Nimit Shah,Haitz Sáez de Ocáriz Borde
类目: Machine Learning (cs.LG)
*备注:
Abstract:A shopping conversation has many routes to the same cart, and a task-success rate reduces all of them to one score. We build a deterministic and reproducible e-commerce environment that precommits each trial’s customer and trajectory parameters, including the persona, difficulty, target cart, and an item reveal schedule. A simulated consumer attempts to buy a target cart from the environment with assistance from the evaluated model. The environment guides the simulator’s actions and records every assistant action alongside the environment state at that point. After the trial, these records allow the evaluator to assess individual parts of the conversation against the retained evidence. For example, the evaluator penalizes a search for failing to surface a target product only when the customer has already mentioned that product. We further use this evidence to apply different penalties to tool calls depending on how the assistant’s actions compare with an expected tool-call set. Our environment also interacts with the simulator bidirectionally, reading its output to stop the trial when the simulator determines that the customer has become too frustrated and injecting directives in real time that specify when to explore, defer buying an item, or recall a previous exchange. This interaction creates an open-ended and verifiable simulation. Across eight open-weight agents from 20B to 35B parameters, with 160 trials per agent and 44 metrics, the resulting capability profiles distinguish under-action, over-purchase, unsupported product attributes, and poor search, all of which terminal success obscures.
[LG-62] Distilling Foundation Models for Agent ic What-If Reasoning :Cost Latency and Governance in a Hybrid LLM SLM Architecture
链接: https://arxiv.org/abs/2609.16091
作者: Sourish Dey,Aditya Kumar
类目: Machine Learning (cs.LG)
*备注: 13 pages, 1 figure, 7 tables
Abstract:Tabular foundation models deliver strong zero-training predictive performance via in-context learning, but their high inference latency makes them impractical as hot-path decision backends in interactive agentic loops. We distill a TabPFN teacher into a compact feed-forward student across a business-decision simulation on UCI Adult and five OpenML benchmarks: the classification head compresses 53.2M parameters to 8,546 (6,220x); the deployed two-head loan pipeline compresses 111.4M parameters to 17,059 (6,532x). The student retains 95.4-100.5% accuracy and 96.8-100.0% AUC, with the lowest accuracy retention on credit-g at 95.4%; an alpha = 0 hard-label control shows that the teacher’s soft targets provide a 2.1-7.0 AUC point gain.
[LG-63] Is INT8 Portable? A Cross-Platform Measurement Study of Quantized Inference on Embedded and Automotive Accelerators
链接: https://arxiv.org/abs/2609.16085
作者: Yuyeong Shin
类目: Hardware Architecture (cs.AR); Machine Learning (cs.LG); Performance (cs.PF)
*备注: 22 pages, 3 figures, 8 tables. Artifact: this https URL
Abstract:Eight-bit integer (INT8) post-training quantization is the default recipe for edge deployment, under a widely held assumption: INT8 makes inference faster at a small, predictable accuracy cost, and a model quantized once can be carried to any target. We test that assumption with a controlled measurement study across seven hardware classes – ARM and x86 CPUs, a discrete GPU, an NVIDIA Jetson AGX Orin iGPU and its NVDLA cores, and two vendor NPUs (Qualcomm Hexagon HTP, DEEPX DX-M1) – holding the ONNX artifact and the quantization scales fixed so the integer kernel or ISA is the only free variable. Portability fails on three axes. (1) The sign of the INT8 speedup is set by the CPU’s dot-product ISA (ARM dotprod/SDOT, x86 VNNI): cores that have it speed up by up to 2.1x, cores that lack it slow down by 1.7x, for the identical model and runtime. (2) INT8 outputs are not portable, and the rule is an invariance rather than a gradient: FP32 predictions are bit-identical for every pair (1000/1000), while INT8 predictions agree 1000/1000 exactly when two targets share an integer kernel and 958-965/1000 whenever they do not – independent of whether the boundary is CPU-CPU or CPU-accelerator, and invisible to top-1 accuracy, which is preserved. (3) Vendor NPUs own quantization: a bring-your-own QDQ graph fails silently on one NPU (external scales ignored, accuracy 0.75 - 0.005 while it compiles, profiles and runs without error) and loudly on the other (the compiler refuses the graph), so only the vendor’s native path yields a correct engine. We further show that edge-NPU latency regimes are set by output/device-to-host transfer size rather than compute, and locate the transition with a fixed-compute sweep. We release the scripts and 32 reports. “Quantize once, deploy anywhere” is unsafe for embedded and automotive deployment, where per-input determinism and redundancy matter.
[LG-64] Predicting Social Media Engagement using Machine Learning
链接: https://arxiv.org/abs/2609.16082
作者: Ritwik Singh,Mayukh Majumdar,Subodha Kumar
类目: ocial and Information Networks (cs.SI); Machine Learning (cs.LG)
*备注: 10 pages, 2 tables, 1 figure
Abstract:Social media platforms are popular channels for disseminating information, owing to their large user bases and ease of access. Companies also use social media as an important aspect of the advertising process. By creating high-quality posts, companies can strengthen their engagement metrics and increase their follower count. While a growing body of research has examined social media engagement, fewer studies have jointly examined the visual, textual, and temporal features of image posts, even though these features collectively determine the performance of content on social media. To understand the important drivers of social media engagement, we collect image posts of furniture firms on Facebook and extract visual, temporal, and textual features from them using text and image analytics methods. We evaluate several machine learning models - including Random Forest, Light Gradient Boosting Machine (LightGBM), and eXtreme Gradient Boosting (XGBoost) - to assess the drivers and the prediction power of social media engagement using the features from our data. Our research quantifies the extent to which these features are associated with interactions and provides recommendations that organizations may consider.
[LG-65] A Dynamic Aggregation Strategy Enhanced Efficient Global Optimization Algorithm for Solving High-Dimensional Turbomachinery Design Problems DATE
链接: https://arxiv.org/abs/2609.16067
作者: Qineng Wang,Zhendong Guo,Yun Chen,Guangjian Ma,Liming Song,Jun Li
类目: Machine Learning (cs.LG); Computational Engineering, Finance, and Science (cs.CE)
*备注: Author manuscript updated to align core methods and results with the published article; 37 pages, 14 figures, 11 tables
Abstract:In order to solve the high-dimensional ( d \geq 30 ) expensive black-box problems within budget, an efficient global optimization (EGO) algorithm with a dynamic aggregation strategy is proposed, labeled as DA-EGO. Specifically, the DA-EGO decomposes the original high-dimensional design space into a set of low-dimensional subspaces for efficient surrogate-based optimization search, and the optimal solutions of subspaces are combined as an elite point for the global search. Most importantly, the subspaces are not fixed. Instead, the subspace variables are updated in each iteration, according to the variable interaction analyses in the sub- and full-spaces. The perturbation method and the analysis of variance are used to detect variable interactions. To further accelerate the optimization progress, the searching ranges of subspaces are also adaptively adjusted according to the analyses of subspace optimization results of the previous iteration. Tests on 21 benchmark instances, comprising seven functions at 30, 60, and 90 dimensions, show that DA-EGO is effective on separable and partially separable problems under a budget of 1500 function evaluations. Its advantage is case-dependent: on the non-separable shifted Rosenbrock function, GSGA performs better at 60 and 90 dimensions, while the 30-dimensional results are statistically comparable to IKAEA and GSGA. Moreover, the advantage of DA-EGO is also seen in the aerodynamic optimization of a transonic rotor blade with 28 variables as well as the compressor stage optimization with 60 variables. With the above, the effectiveness of the proposed DA-EGO has been well demonstrated.
[LG-66] A panoramic aerodynamic performance prediction method for turbomachinery cascades using transformer-enhanced neural operator DATE
链接: https://arxiv.org/abs/2609.16066
作者: Qineng Wang,Zhendong Guo,Liming Song,Tianyuan Liu
类目: Machine Learning (cs.LG); Computational Physics (physics.comp-ph); Fluid Dynamics (physics.flu-dyn)
*备注: Author manuscript updated to align core methods and results with the published article; 41 pages, 20 figures, 14 tables
Abstract:To enable flexible and rapid aerodynamic performance evaluation in turbomachinery design, this paper proposes a panoramic performance prediction framework. Unlike most previous prediction models that directly predict the objective functions of interest, our approach first predicts the basic parameters of the Navier-Stokes equations, such as temperature, pressure, and density. Utilizing these basic physical quantities, it subsequently predicts key performance parameters of the turbine stage meridian plane. By adopting this methodology, our proposed panoramic performance prediction framework functions similarly to a CFD simulator, capable of predicting various objective of interest to the designers. To enhance prediction accuracy, a transformer-enhanced neural operator (TNO) is introduced within this framework. Using the Rotor 37 blades as a reference, the proposed TNO is trained to predict the performance of a transonic compressor blade in the meridian plane. The TNO can accurately predict total quantities such as isentropic efficiency, mass flow, and distributions of total pressure ratio. Remarkably, the prediction error of TNO is observed to be smaller than that of state-of-the-art deep learning operators such as the FNO and DeepONet. Furthermore, the TNO is applied to downstream tasks, including sensitivity analysis and optimization of various objective functions. The results confirm that the TNO can operate almost like a CFD simulator, while reducing the computational cost of downstream tasks by four orders of magnitude. The effectiveness and reliability of the proposed TNO for solving different kinds of downstream tasks have been well demonstrated.
[LG-67] Signed p-adic Residual Encodings of Finite-Domain All-Different Systems with a Sudoku Case Study
链接: https://arxiv.org/abs/2609.16063
作者: Greg Baker
类目: Machine Learning (cs.LG)
*备注: 31 pages, 7 figures. Accepted for publication in p-Adic Numbers, Ultrametric Analysis and Applications
Abstract:We study signed, weighted affine p -adic residual objectives as native encodings of finite-domain constraints. For primes that separate the finite alphabet, sufficiently weighted positive unary rows pin each coefficient to its allowed set, while negative rows reward unequal endpoints or clause satisfaction. A coordinatewise domination theorem places every global minimiser in the finite domain; there the loss is, up to an additive constant, the all-different conflict count or the negative number of satisfied CNF clauses. Standard Sudoku provides an 81 -coefficient case study without a one-hot lift. A client-side implementation exposes the generated dataframes, arithmetic, diagnostics, and searches.
[LG-68] Digital Persuasion: Understanding the Impact of Online Influencers on Public Opinion
链接: https://arxiv.org/abs/2609.16062
作者: Omran Berjawi,Rida Khatoun,Giuseppe Fenza
类目: ocial and Information Networks (cs.SI); Machine Learning (cs.LG)
*备注:
Abstract:The studying of opinion dynamics and its propagation within social networks is crucial for addressing a wide range of challenges, including political polarization, public health, and marketing strategies. In this work, we study the problem of opinion dynamics by proposing a framework based on Friedkin-Johnsen (FJ) to identifies influential users and study their impact on dynamics opinions of community. The FJ model assume each individual have two opinions: initial and expressed. Through a series of initial opinion manipulation experiments, the proposed framework assesses the impact of influential versus random users on the overall community opinion. The proposed framework is validated using a tweet dataset representing the U.S. presidential election. The results shows that influencers with highest influencing score, significantly shift the overall community opinion. Moreover, the results shows that the impact of influencers not limited to direct neighbors , but beyond it, to their neighbors of neighbors . This study demonstrates how digital influencers on social media can shape public opinion regarding a subject or cause.
[LG-69] Measuring AI harms with multidimensional Lorenz Zonoids
链接: https://arxiv.org/abs/2609.16004
作者: Paolo Giudici,Jose’ Maria Sarabia,Sofia Vei
类目: Computers and Society (cs.CY); Machine Learning (cs.LG)
*备注: 26 pages, 6 figures
Abstract:While AI systems increasingly shape high-stakes societal domains, their governance is limited by the lack of risk management methods that operate on real harms, taking their severity, and not only their likelihood, into account. As a consequence, AI risk management models remain compliance-driven and provider-centric, offering limited insight into how harms are dangerous, and on what should be the priority of intervention. The problem is amplified by the nature of harm data which are typically ordinal and multidimensional. To solve the problem, and offer an effective risk assessment methodology, in this paper we propose to model harm data by means of Lorenz Zonoids and Gini indices. To this aim we propose to extend them in a multidimensional setting, and show how to practically calculate them for a real AI incident data repository, provided by the Massachusetts Institute of Technology. The empirical findings indicate that environmental, infrastructure, property, physical, and democracy-related harms attain the highest values under the two multidimensional Gini indices and therefore exhibit the strongest concentration in their joint direct, indirect, and inferred severity-frequency distributions. These concentration patterns may help identify categories that warrant closer examination when mitigation priorities are determined.
[LG-70] Bridging the Gap Between Homogeneous and Heterogeneous Asynchronous Optimization Is Surprisingly Difficult
链接: https://arxiv.org/abs/2609.17483
作者: Alexander Tyurin
类目: Optimization and Control (math.OC); Machine Learning (cs.LG)
*备注:
Abstract:Modern large-scale machine learning tasks often require multiple workers, devices, CPUs, or GPUs to compute stochastic gradients in parallel and asynchronously to train model weights. Theoretical results typically distinguish between two settings: (i) the homogeneous setting, where all workers have access to the same data distribution, and (ii) the heterogeneous setting, where each worker operates on different data distributions. Known optimal time complexities in these settings reveal a significant gap, with far more pessimistic guarantees in the heterogeneous case. In this work, we investigate whether these pessimistic optimal time complexities can be overcome under different assumptions. Surprisingly, we show that improvement is provably impossible under widely used first- and second-order similarity assumptions for any randomized algorithm. We then turn to the interpolation regime and demonstrate that the weak interpolation assumption alone is also insufficient. Finally, we introduce a minimal combination of irreducible assumptions, strong interpolation and the local Polyak-Lojasiewicz condition, to derive a new time complexity bound that matches the dependence on worker computation times in the best-known result in the homogeneous setting, without requiring identical data distributions.
[LG-71] Bias-Induced Crossover in Absolute Capacity of Dense Associative Memory
链接: https://arxiv.org/abs/2609.17477
作者: Yuto Sakurai,Takeaki Shimokawa,Kazunori Iwata,Kazushi Mimura
类目: Disordered Systems and Neural Networks (cond-mat.dis-nn); Machine Learning (cs.LG)
*备注: 17 pages, 5 figures
Abstract:The absolute capacity of dense associative memory has mainly been analyzed for unbiased patterns. Here we examine the effect of bias in centered binary patterns under the Krotov-Hopfield single-site criterion P_\mathrmerror=1/N , where P_\mathrmerror is the probability that a single-site flip lowers the energy of a stored pattern and N is the number of neurons. Each pattern component takes 1-q with probability q and -q otherwise, where 0q\le1/2 . For polynomial interactions of order n , a signal-to-noise analysis gives an absolute capacity of order N^n-1/\ln N at q=1/2 . For fixed q1/2 , however, the capacity is O(N^n/2) for even n\ge4 and O(N^(n+1)/2) for odd n\ge5 . For n=3 , both the unbiased and fixed-bias capacities remain O(N^2/\ln N) . For n\ge4 , these different asymptotic forms imply a nonuniform large- N limit near q=1/2 . Asymptotic matching predicts a bias-induced crossover in the region 1-2q=O(\ln N/N^\lfloor n/2\rfloor-1) . The crossover originates from a bias-dependent crosstalk mean that reduces the stability of sites carrying the more frequent value -q . Computer simulations are compared with the finite-size conditioned-Gaussian predictions. An activity-dependent control potential that cancels the conditional crosstalk mean restores the N^n-1/\ln N capacity for fixed 0q1/2 within the conditioned-Gaussian approximation.
[LG-72] Goal-oriented probabilistic forecasting for dynamic PRB allocation in 5G networks
链接: https://arxiv.org/abs/2609.17297
作者: Oier Larumbe-Lizarraga,Roberto Pereira,Cristian J. Vaca-Rubio
类目: ignal Processing (eess.SP); Machine Learning (cs.LG)
*备注:
Abstract:Efficient physical resource block (PRB) allocation in 5G networks requires accurate demand forecasting. Conventional methods minimize symmetric error metrics (MAE, RMSE), ignoring the operational cost asymmetry where under-provisioning (service degradation) is far costlier than over-provisioning (wasted capacity). We propose a goal-oriented probabilistic forecasting framework that aligns model training with the operator’s decision-making objectives. Specifically, we train DeepAR and Temporal Fusion Transformer (TFT) models using the Pinball Loss function and derive the optimal allocation quantile from the operator’s cost matrix. Evaluation on a real beam-level 5G traffic dataset shows that the proposed approach reduces operational cost compared to MSE-trained baselines while maintaining calibrated uncertainty estimates. The framework enables dynamic PRB allocation that explicitly balances service reliability against resource efficiency.
[LG-73] Conformal Policy Learning with Distribution-Free Safety Guarantees
链接: https://arxiv.org/abs/2609.17296
作者: Ying Jin,Naoki Egami
类目: Methodology (stat.ME); Machine Learning (cs.LG); Econometrics (econ.EM); Statistics Theory (math.ST); Machine Learning (stat.ML)
*备注:
Abstract:Policy learning aims to determine who should be treated based on individual characteristics. In high-stakes settings such as medicine and public policy where safety is a central concern, improving the average outcomes alone may not be sufficient: decision makers may also seek to protect individuals from harm, in line with the Hippocratic principle of ``do no harm.‘’ In this paper, we propose \textitconformal policy learning (CPL), a policy learning procedure with a new distribution-free safety guarantee that controls the probability of assigning treatment to an individual who would be harmed relative to control. CPL views each treatment decision as testing a hypothesis of counterfactual harm and assigns treatment by thresholding conformal p-values. These p-values use observable proxies and selective calibration to address the challenge that the potential outcomes under comparison are never simultaneously observed. For randomized experiments, under standard exchangeability conditions, CPL provides finite-sample safety guarantee at a user-specified level, without imposing any outcome modeling assumptions. Moreover, when the outcome model is consistently estimated, CPL achieves asymptotically optimal welfare subject to the safety constraint. In observational studies, CPL with learn-then-balance weights achieves doubly robust safety guarantees. We evaluate CPL through extensive simulations and apply it to an empirical study of AI-powered interventions designed to reduce conspiracy beliefs.
[LG-74] Optimization over covariance matrices with a parameterized metric
链接: https://arxiv.org/abs/2609.17089
作者: Yibang Li,Bamdev Mishra,Pratik Jawanpuria,Cyrus Mostajeran
类目: Optimization and Control (math.OC); Machine Learning (cs.LG)
*备注:
Abstract:The choice of Riemannian metric can strongly influence the convergence of gradient-based optimization over covariance matrices. Euclidean, Bures-Wasserstein and affine-invariant metrics are common choices, but their relative effectiveness depends on the objective. We introduce a two-parameter family defined by X^pLX^q+X^qLX^p=U , solved for L at each tangent vector U , that contains all three as exact members, at (0,0) , (1,0) and (1,1) , and extends past them. We treat the choice of member as a particular way of preconditioning for a given problem. To this end, we analyze the conditioning of the Riemannian Hessian at the solution. We show that it obeys a lower bound that depends on (p,q) only through the exponent r=p+q . When the Euclidean Hessian is a pure power that mixes no eigendirections, the member p=q=r/2 attains that bound, and a closed-form criterion identifies the other members that do. We discuss ways to tune r for a given problem. Experiments on real covariance data confirm the predicted conditioning and the benefit of tuning r . A task covariance example shows a further gain from tuning the shape.
[LG-75] Splitting the Difference: Interpretable Causal Forests for Treatment Effect Heterogeneity and Bias
链接: https://arxiv.org/abs/2609.16971
作者: Nicolas Alexander Ihlo,Merle Behr
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:In various fields, such as medicine and marketing, accurately predicting individual treatment effects holds significant promise. However, achieving reliable predictions alone is often insufficient for making informed decisions; it is equally important to understand why the treatment effect is higher for some individuals than for others. To address this two-fold challenge of prediction and interpretation, we introduce an algorithm based on decision trees and random forests for estimating individual treatment effects. Our algorithm is simple: it operates exactly like a standard random forest, but with a different splitting criterion, and requires no additional workarounds such as double machine learning or orthogonalization as used in Generalized random forests. It handles observational studies with varying treatment propensities without requiring separate estimation of the full propensity function. This is achieved by combining two splitting criteria—one targeting heterogeneity in the treatment effect, the other targeting bias correction for the average treatment effect—which together improve split point selection and automatically distinguish confounders from features responsible for heterogeneity. As a result, interpretation follows directly from the fitted tree structure itself, that is, from which features the trees split on and with which split statistics, without requiring separate post-hoc analysis. For the theoretical analysis of this algorithm, we consider a change point model with step functions for potential outcomes and treatment propensity and provide insights into the theoretical underpinnings of our approach. Simulation studies show that our simple algorithm achieves comparable, and often better, prediction accuracy than existing methods, while substantially improving interpretability.
[LG-76] On the disintegration of the stochastic majority vote: From PAC-Bayesian bounds to a self-bounding algorithm
链接: https://arxiv.org/abs/2609.16803
作者: Julien Bastian(LabHC),Benjamin Leblanc,Pascal Germain,Amaury Habrard(LabHC, UJM, MALICE),Guillaume Metzler(ERIC),Emilie Morvant(LabHC),Paul Viallard(MALT)
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:Weighted majority votes are central to many successful ensemble methods. PAC-Bayesian theory provides tight generalization guarantees for such models by analyzing the expected risk of stochastic classifiers, while analyzing the risk of deterministic majority votes relies on surrogate bounds. To avoid these surrogates, Zantedeschi et al. ( 2021) introduced guarantees for stochastic majority votes, but the resulting models remain randomized. In this paper, we propose a derandomization framework for stochastic majority votes. To do so, we apply recent advances in disintegrated PAC-Bayesian theory directly to the space of majority vote weight vectors, transforming stochastic guarantees into certificates for a single deterministic majority vote. We derive two families of high-probability generalization bounds, covering both data-independent and data-dependent constructions of the ensemble, which naturally lead to a self-bounding learning algorithm optimizing deterministic majority vote guarantees.
[LG-77] me-warping estimation via stationarity-based learning of the de-warped signal
链接: https://arxiv.org/abs/2609.16796
作者: Corentin Presvôts(Phys-ENS),Adrien Meynard(Phys-ENS)
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:Time-warping estimation is a fundamental problem in signal processing with applications in bioacoustics, radar, and biomedical analysis. This paper introduces a Time-Warping Estimation Trainable (TWET) model for estimating timewarping functions from a single observation. The proposed approach formulates time-warping estimation as a stationarization problem in the wavelet domain and leverages a hierarchical dilated convolutional architecture to estimate the time-warping functions. A differentiable stationarity criterion is introduced for end-to-end optimization. TWET is compared with existing approaches. Experimental results show improved deformation reconstruction accuracy together with significantly reduced computation time, making the framework compatible with low-latency applications.
[LG-78] Certified Inference and Training for Deep Equilibrium Networks: A Continuation Framework with Polynomial Complexity Guarantees
链接: https://arxiv.org/abs/2609.16485
作者: Alex Borisevich
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注: Python scripts and Lean formalization are included as ancillary files
Abstract:We develop a certified continuation framework for equilibrium computation and for training deep equilibrium networks (DEQs), with training formulated as interpolation to accuracy 2^-b . For inference, compact input homotopy selects a unique branch from a supplied start root, and a rounded Newton tracker follows it under certified boundary, conditioning, derivative, and tube-radius bounds. For training, we augment local-plus-low-rank recurrence with programmable dormant bilinear rank-one channels. Loaded Tikhonov solves diagnose a failed interpolation pass without spectral decomposition; an output-preserving repair aligned with the pass residual supplies the required direction. Training requires certified gate realization and column stability on each pass region, well-posed inference, and finite-update error budgets. With polynomial geometric, encoding, precision, and complete backend budgets, both certified inference and training have bit cost O(\operatornamepoly(L+b)) , where L is the encoded instance length. The trainer uses O(b+\ell) passes and reserve channels from an initial residual bounded by 2^\ell . These guarantees concern a certified promise class. Lean 4 verifies the quantitative core and concrete inference backend; numerical comparisons illustrate the loaded mechanism.
[LG-79] Learned Look-Ahead Splitting Rule for CART
链接: https://arxiv.org/abs/2609.16440
作者: Andrew Gao,Tianlin Liu,Ruichen Han,Lu Tian
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:Classification and regression trees are typically constructed using a greedy splitting rule that maximizes the immediate reduction in prediction error at each node. Although this strategy is computationally efficient, it can miss splits that yield small short-term gains but create substantial downstream improvements after further partitioning. We propose a look-ahead tree-building method that evaluates each candidate split by the prediction error reduction achieved after growing a conventional CART subtree below that split. Because the full look-ahead procedure can be computationally expensive, we also describe a smart look-ahead algorithm that learns downstream split values using node-level features. The proposed framework preserves the interpretability of recursive partitioning while improving split selection in hierarchical or interaction-driven settings. We conduct a simulation study comparing conventional, full look-ahead, and smart look-ahead methods under several settings and apply the proposed methods to analyze two real data examples demonstrating the merit of the new methods.
[LG-80] Nationally Consistent Locally Incomplete: A Bayesian Remote-Sensing Audit of Rooftop Photovoltaic Registries
链接: https://arxiv.org/abs/2609.16294
作者: Gabriel Kasmi,Yves-Marie Saint-Drenan,Laurent Dubus,Philippe Blanc
类目: Applications (stat.AP); Machine Learning (cs.LG)
*备注: 47 pages, 5 tables, 15 figures
Abstract:Tracking the energy transition requires reliable statistics on renewable deployment. Rooftop photovoltaics (PV) are especially hard to track, owing to their decentralised nature, and the resulting inaccuracies in official statistics are known but not quantified. Remote sensing offers an independent way to identify rooftop PV systems. We introduce a Bayesian framework to estimate the ground-truth rooftop PV capacity from remote sensing detections, turning an imperfect detector into an uncertainty-aware measurement instrument. Applied to France, the corrected detections estimate a capacity of 4.03 GWp [3.96–4.11] (99% credible interval) of rooftop PV below 36 kWp, matching the transmission system operator’s connection data within 3.3% nationally, while identifying local under-reports of up to 61% of local capacity. We also document and quantify a significant truncation bias in French rooftop PV open data. Beyond France, the approach paves the way for more reliable estimates of rooftop PV capacity worldwide.
[LG-81] Semantic-Aware Neural Video Codec for Error-Resilient Low-Latency Transmission
链接: https://arxiv.org/abs/2609.16279
作者: Matin Mortaheb,Homa Esfahanizadeh,Jinfeng Du,Harish Viswanathan
类目: Image and Video Processing (eess.IV); Information Theory (cs.IT); Machine Learning (cs.LG); Multimedia (cs.MM)
*备注:
Abstract:Emerging physical AI systems require low-latency, task-oriented video communication over unreliable channels. We propose a semantic-aware multi-level neural video coding method for robust low-latency video transmission over unreliable channels that are abstracted as multi-level packet erasure channels. Built upon the real-time DCVC-RT neural video codec, the proposed framework introduces a semantic- and feature-aware coding strategy that partitions encoded representations into packets carrying different levels of semantic and latent-feature importance and assigns these packets to different streams, each associated with a priority level when transmitted over unreliable communication channels. We also developed an error-resilient entropy model that removes inter-packet dependencies, allowing each packet to be decoded independently under packet losses. The complete system is trained end-to-end over the abstracted multi-level packet erasure channels, enabling learning of channel-aware representations together with importance-aware packet assignment while facilitating the network for differentiated packet prioritization. Experiments show that the proposed framework significantly improves robustness over baseline DCVC-RT under packet erasures, achieving graceful degradation in less important regions while better preserving task-relevant visual content.
[LG-82] owards Surrogate Based Dequantization of Quantum Reinforcement Learning
链接: https://arxiv.org/abs/2609.16266
作者: Pablo Rodriguez-Grasa,Sofiene Jerbi,Mikel Sanz,Ryan Sweke
类目: Quantum Physics (quant-ph); Machine Learning (cs.LG)
*备注:
Abstract:In recent years, the utility of parameterized quantum circuits as function approximators has been widely studied. In the context of reinforcement learning, this approach has led to variational quantum algorithms such as quantum Q-learning. While these methods show promising empirical results, and can provide provable advantages for artificial problems, it remains unclear whether they can provide a provable quantum advantage over classical approaches for problems of practical relevance. A natural way to investigate this question is through the lens of dequantization: The construction of efficient classical algorithms capable of matching the performance of quantum variational methods. Building on recent kernel-based dequantization results for supervised learning, we take steps towards extending this surrogate-based dequantization program to reinforcement learning. Specifically, we study the simplified setting of reinforcement learning with a uniform generative model in which uniformly random state-action samples are available, which models the regime of sampling from a large experience replay buffer after sufficient exploration. Within this setting, we provide finite sample guarantees for classical kernelized Fitted Q-Iteration, with classical kernels designed to match the inductive bias of particular parameterized quantum circuits. Using these results, we then provide a set of sufficient conditions, on the data-encoding strategy of a parameterized quantum circuit, the corresponding classical kernel, and the problem structure, under which kernelized Fitted Q-Iteration provides a meaningful dequantization of quantum Q-learning, in this simplified setting. Apart from providing rigorous dequantization guarantees when these conditions are met, these results also motivate the use of kernelized fitted Q-iteration as a dequantization heuristic when these sufficient conditions cannot be verified.
[LG-83] Compute-Optimal Pretrain–Fine-tune in Ridge Gradient Descent
链接: https://arxiv.org/abs/2609.16262
作者: Alex Buna,Fanghui Liu,Patrick Rebeschini
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:Pretraining followed by fine-tuning introduces a compute-allocation problem: under a fixed training budget, compute spent improving the upstream objective reduces the compute available for downstream adaptation. Despite its practical importance, this trade-off is not yet well understood theoretically, even in simple models. In this paper, we cast this allocation as a compute-split problem under a two-stage pretrain–fine-tune procedure with fixed total optimisation budget, using regularised least squares trained by gradient descent as a tractable setting. We characterise the optimal split under data-dependent evaluation geometries induced by the fine-tuning problem. Our results show that the allocation depends on how pretraining directions affect fine-tuning predictions and how fine-tuning shifts are seen through downstream data geometry. In particular, the relevant quantities are determined by prediction-relevant spectral components of the pretraining and fine-tuning empirical covariances. Technically, the analysis relies on a basis-invariant, eigenspace-level spectral decomposition, together with perturbative control of the non-commuting pretraining and fine-tuning dynamics.
[LG-84] Copula Adapted Directed Acyclic Graph for Cluster Representation of Biomedical Data
链接: https://arxiv.org/abs/2609.16240
作者: Heranga K. Rathnasekara,Norou Diawara,Manar D. Samad
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Computation (stat.CO)
*备注:
Abstract:Diagnostic errors and mislabeling are common in biomedicine, which compromise the reliability of predictive models and data-driven outcomes. Stratifying unlabeled biomedical data based on complex relationships between features eliminates the need for data labels and overcomes the limitations of supervised learning. Traditional clustering methods assume restrictive data distributions, making them suboptimal for capturing complex dependencies in high-dimensional biomedical data. This paper introduces a novel cluster-friendly data presentation framework that integrates the non-Gaussian and non-linear feature dependence of copula models with an ensemble of causal structure discovery (CSD) methods based on Directed Acyclic Graphs (DAGs). While copulas model flexible multivariate distributions by relaxing assumptions related to multivariate normality, linear dependence, and symmetric relationships, an ensemble of DAG-based CSD methods identifies stable causal relationships between features. When clustered using K-means, the new data representation obtained by the proposed copula-adapted DAG (CopDAG) ranks first among the 12 methods in normalized clustering accuracy and adjusted Rand index across 16 biomedical datasets. Our CopDAG method predicts ground-truth class labels directly from feature relationships without data annotations and supervised learning, while also providing cluster visualizations and explainable causal structures of the biomedical data features.
[LG-85] Improving Reduced-Order Rotating Detonation Engine Models with Data Assimilation and Machine Learning
链接: https://arxiv.org/abs/2609.16237
作者: Ashwin Suriyanarayanan,Romit Maulik
类目: Fluid Dynamics (physics.flu-dyn); Machine Learning (cs.LG); Dynamical Systems (math.DS); Computational Physics (physics.comp-ph)
*备注: 20 pages, 12 figures
Abstract:Rotating detonation engines (RDEs) exhibit strongly nonlinear, multiscale wave dynamics that set the observed thermal field. High-fidelity simulations (DNS/LES) resolve these structures but remain computationally prohibitive, while low-order models such as the one-dimensional Koch-Kutz model capture circumferential wave motion yet lack the expressivity for high-frequency content. We use continuous data assimilation (nudging) to synchronize the Koch-Kutz solver with processed high-fidelity temperature data, introducing the prediction-observation mismatch as a relaxation source in the conserved energy equation; where observations are temporally sparse, interpolation supplies a target at every source update. As the nudging strength increases, the reduced model is progressively drawn onto the high-fidelity trajectory, and the forcing recorded along it provides an explicit, state-dependent estimate of the correction the model requires. We then train a Jacobian-regularized closure a priori on this recorded source. With the observation term removed, the corrected model advances autonomously, remains bounded, and recovers the temperature spectrum and the marginal statistics of the conserved variables relative to the baseline.
[LG-86] Computer-assisted global regularity across nonlinear families of three-dimensional periodic Navier-Stokes flows
链接: https://arxiv.org/abs/2609.16157
作者: Jose Luis Lima de Jesus Silva
类目: Fluid Dynamics (physics.flu-dyn); Machine Learning (cs.LG)
*备注: 75 pages, 14 figures
Abstract:Numerical simulations reveal how vortices stretch and transfer energy, but establishing smooth evolution requires bounds that remain valid beyond the simulated resolution. Here I develop a computer-assisted framework that establishes global regularity for continuous families of three-dimensional periodic Navier-Stokes flows. Its central construction combines finite reference trajectories with a common error bound that covers an interval of centre fields and infinitely many smooth perturbation modes. The method retains the complete nonlinear residual before spectral truncation and controls the evolution until viscous decay guarantees regularity for all subsequent times. Applications to cyclic-shear, Arnold-Beltrami-Childress and three-component Taylor-Green fields yield explicit perturbation radii and include initial conditions outside the direct Fourier-Wiener smallness criterion. A parameter-uniform extension covers a connected family of non-Beltrami Taylor-Green centres without repeating the proof for individual parameter values. An ensemble of 4,096 configurations, supplemented by 1,600 refinement trajectories and public turbulence data, connects the mathematical observables to spectral transfer and vortex geometry. Matched neural-operator experiments show that physics-informed training improves physical prediction, while also revealing that these gains do not necessarily improve the discovery of proof-limiting initial conditions. Together, these results provide a reusable method for establishing regularity across prescribed flow families and a quantitative setting for evaluating how learned predictions can assist rigorous computation.
[LG-87] Molecular representation shapes the balance between target fidelity and exploration in flow based polymer generation
链接: https://arxiv.org/abs/2609.16028
作者: Tianren Zhang
类目: Chemical Physics (physics.chem-ph); Materials Science (cond-mat.mtrl-sci); Machine Learning (cs.LG)
*备注:
Abstract:Designing polymers with targeted properties requires navigating vast chemical spaces from limited labeled data. Here we introduce PolyLatentFlow, a framework based on continuous-time flow matching in latent space for unconditional and conditional polymer generation, together with LlamaUni, a multimodal representation combining polymer sequence and 3D structural information. In unconditional generation, PolyLatentFlow with LlamaUni produced the largest yield of valid candidates novel relative to PolyInfo among the evaluated unconditional generators while maintaining high diversity. For T_g conditioning, generated property distributions shifted systematically across a 200 °C target range. In multi-property tasks, molecular representations showed similar surrogate target fidelity but differed markedly in validity, training-set replay, and structural proximity to labeled polymers. PolyLatentFlow with LlamaUni consistently combined high validity with low replay and achieved the largest per-attempt yield of nonreplayed target hits for CO _2 /N _2 conditioning. These results demonstrate latent space flow matching for polymer inverse design and identify molecular representation as a key determinant of target control and exploration beyond labeled chemistry.
[LG-88] Nonsmooth Optimization via Orthogonalized Momentum
链接: https://arxiv.org/abs/2609.13677
作者: Lexiao Lai,Tianyi Lin,Jiayu Zhang
类目: Optimization and Control (math.OC); Machine Learning (cs.LG)
*备注: 32 pages, 3 figures
Abstract:Modern real application problems involve matrix-valued parameters, yet conventional optimizers treat them as vectors, thereby motivating matrix-aware methods that exploit input-output geometry, such as Muon which orthogonalizes the momentum matrices before parameter updates. Its empirical success raises a conceptual question: can orthogonalized momentum remain effective beyond smooth optimization? This paper studies this question for locally Lipschitz functions using a generalized derivative framework compatible with backpropagation. Our first contribution is to identify a key limitation: for every fixed momentum factor \beta\in[0,1) , Muon can fail to approach the global optimal solution of a convex Lipschitz objective from almost every initialization, when step sizes adapt to the full gradient history. The failure can occur even along bounded iterates. Our example is inspired by the one of Parshakova et al. which only covers \beta\in[0,\frac12) . Then, we show that the obstruction lies in fixed momentum rather than orthogonalization. Indeed, when the momentum factor is adaptive and approaches 1 together with a vanishing step size, Muon recovers asymptotic convergence for nonconvex nonsmooth optimization under the boundedness and regularity conditions. Moreover, we propose MAGD, which combines orthogonalized momentum with gradient, weighted based on their relative progress. MAGD retains asymptotic convergence in nonconvex settings and achieves an O(\min\m,n\epsilon^-2) rate in convex settings. A lower bound shows the optimal dimension dependence. Experiments on synthetic problems, image classification, and LLM pretraining show MAGD is a simple and practical alternative to Muon. Together, our results characterize when orthogonalized momentum fails without smoothness and how it can be made reliable and we hope that the analysis may be useful more broadly.
附件下载
