本篇博文主要内容为 2026-08-12 从Arxiv.org论文网站获取的最新论文列表，自动更新，按照NLP、CV、ML、AI、IR、MA六个大方向区分。
说明：每日论文数据从Arxiv.org获取，每天早上12:30左右定时自动更新。
提示: 当天未及时更新，有可能是Arxiv当日未有新的论文发布，也有可能是脚本出错。尽可能会在当天修复。
目录
概览 (2026-08-12)
今日共更新676篇论文,其中:
- 自然语言处理共95篇(Computation and Language (cs.CL))
- 人工智能共213篇(Artificial Intelligence (cs.AI))
- 计算机视觉共149篇(Computer Vision and Pattern Recognition (cs.CV))
- 机器学习共166篇(Machine Learning (cs.LG))
- 多智能体系统共14篇(Multiagent Systems (cs.MA))
- 信息检索共22篇(Information Retrieval (cs.IR))
- 人机交互共38篇(Human-Computer Interaction (cs.HC))
多智能体系统
[MA-0] Who Are You Explaining To? A Multi-Agent System for Audience-Aware XAI Narratives
【速读】：该论文旨在解决生成式AI（Generative AI）在医疗领域中局部解释（local explanation）的可读性与受众适配性问题。现有特征归因方法（如SHAP）虽能提供模型预测的量化证据，但其数值输出难以适应患者、临床医生与数据科学家等不同背景用户在理解能力、信息需求及误读风险上的显著差异。直接通过大语言模型（LLM）进行朴素自然语言化表达，常导致语义根基薄弱、归因与因果表述混淆，以及虽具说服力却违背模型真实证据的问题。为此，论文提出XstrAI——一种面向受众的多智能体框架，其核心在于将局部解释视为不可变的结构化证据，统一共享于所有受众，确保底层事实一致性；在此基础上，通过三个专业化LLM智能体协同工作：受众感知规划、语言实现与验证，分别负责内容适配性、语义接地性、归因一致性、沟通风险控制及受众适宜性评估，并引入有限次修订循环以纠正不一致。实验在糖尿病和中风风险预测任务上对比11种基线方法，涵盖从直接文本转换到先进叙事模型的多种方案。评估采用内叙事（衡量对SHAP证据的忠实度）与外叙事（通过参考语料库、多家族LLM裁判及目标读者调查评估受众适切性）双重机制，结果表明，XstrAI生成的叙述在独立评审中均被准确识别为对应目标受众，且在临床医生与患者群体中显著优于所有基线，数据科学家群体中表现具有竞争力，凸显其在多受众情境下实现高保真、高适配性解释传播的核心优势。
链接: https://arxiv.org/abs/2608.11033
作者: Francesco Musicco,Danilo Danese,Giuseppe Fasano,Angela Lombardi,Alberto Carlo Maria Mancino,Tommaso Di Noia
机构: Politecnico di Bari (巴里理工大学)
类目: Multiagent Systems (cs.MA)
备注:
Abstract:Feature-attribution methods such as SHAP provide useful evidence about individual model predictions, but their numerical outputs are rarely sufficient for audiences with different expertise, goals, and risks of misinterpretation. In medical AI, the same local explanation must reach patients, clinicians, and data scientists through markedly different forms of communication, and naive verbalization through large language models (LLMs) is prone to weak grounding, conflation of attribution with causal language, and outputs that are persuasive without being faithful to the underlying model evidence. We introduce XstrAI, an audience-aware multi-agent framework that treats local explanations as fixed evidence and structures how it is communicated to each target reader. Each prediction case is encoded as an immutable structured representation, shared identically across audiences so the underlying evidence remains fixed. Generation is factored into three specialized LLM agents responsible for audience-aware planning, linguistic realization, and validation for grounding, attribution consistency, communicative risk, and audience appropriateness, with a bounded revision loop triggered on detected inconsistencies. We evaluate XstrAI on diabetes and stroke risk prediction against 11 baselines, ranging from direct verbalization to a re-implementation of a state-of-the-art narrator. The evaluation combines an intra-narrative regime measuring fidelity to SHAP evidence with an extra-narrative regime assessing audience appropriateness through reference corpora, multi-family LLM judges, and a survey with target readers. In both evaluations, XstrAI’s narratives are consistently assigned to their intended audience by independent judges, and preferred over all baselines on Clinician and Patient audiences, with competitive performance on Data Scientist, where audience-conditioned single-prompt baselines lead.
[MA-1] Conversational Orchestration for Organic 6G
【速读】：该论文旨在解决有机化6G（Organic 6G）网络中跨域服务编排的可操作性、可扩展性与动态适应性难题，尤其针对由边缘-云连续体与非地面资源共同构成的“网络之网络”架构下，多自治域在动态加入与退出（domain churn）场景中的高效协同问题。现有跨域编排方案普遍依赖复杂的集成框架、多层协调器及深度遥测管道，导致部署困难且协调开销过高。为此，论文提出一种轻量级、去中心化的对话式编排框架，其核心在于基于大语言模型（Large Language Model, LLM）驱动的域智能体（domain agents），各域保持自治：通过工具观测本地状态，进行闭环推理，并借助与数据平面耦合的代理间（Agent-to-Agent, A2A）覆盖网络交换摘要信息。通过周期性类路由的可达性通告（包含延迟、瓶颈带宽与计算能力）实现快速可行的服务放置；而安全的重新优化、弹性伸缩与迁移则通过事件驱动的请求与协商机制完成。为满足实时性要求，系统采用经验证器驱动自验证训练并定期通过影子更新在线精炼的紧凑型推理模型。仿真结果表明，该框架在域规模扩展及域动态加入过程中控制面开销可控且近似线性增长，决策质量稳健，具备对目标变化的恢复能力。研究最终展望了面向有机6G的可论证、可信赖及不确定性感知的智能体式编排的未来方向。
链接: https://arxiv.org/abs/2608.10714
作者: Masoud Shokrnezhad,Tarik Taleb
机构: 未知
类目: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC); Emerging Technologies (cs.ET); Multiagent Systems (cs.MA)
备注: 7 pages, 6 figures. Accepted for publication in IEEE Network Magazine
Abstract:The Organic 6G vision of a network of networks spanning an edge-cloud continuum complemented by non-terrestrial resources requires, to realize its promise, service provisioning that is simple to operate, scalable across independently administered domains, and agile under domain churn (i.e., domains dynamically joining and leaving). Despite advances in cross-domain orchestration, many proposals rely on heavy integration fabrics, multi-layer coordinators, and deep telemetry pipelines that hinder deployability and amplify coordination overhead. We propose a lightweight, decentralized conversational orchestration framework based on Large Language Model (LLM)-driven domain agents. Each domain remains autonomous: an agent observes local state via tools, reasons in a closed loop, and exchanges summaries with neighboring agents over an Agent-to-Agent (A2A) overlay aligned with data-plane coupling. Fast feasible placement is enabled by periodic, routing-like dissemination of reachability advertisements (latency, bottleneck bandwidth, and compute capacity), while safe re-optimization, scaling, and migration are handled through event-driven requests and negotiation. To meet real-time constraints, we deploy a compact reasoning model trained with verifier-based self-verification and periodically refined online via shadow updates. Simulations show manageable, near-linear control-plane overhead as domains scale and during domain joins, and robust decision quality, including recovery after objective changes. We close by outlining future research directions for principled, secure, and uncertainty-aware agentic orchestration in Organic 6G.
[MA-2] Reifying Research Logic: AI-Assisted Workflow Construction and Incremental Refinement for Quantitative Syntax
【速读】：该论文旨在解决定量语言学研究中因计算流程链条过长且逻辑隐含于脚本代码之中，导致分析过程难以审查、共享与修订的问题。其核心解决方案是提出QLWF——一个面向定量句法的可视化工作流平台，通过人工智能辅助的五阶段流水线，将自然语言的研究描述自动转化为可执行的工作流。该方案的关键在于“具象化”（reification）与“形式化”（formalization）：前者使研究逻辑以可视工作流的形式显式呈现，后者赋予工作流确定性的执行语义。在系统运行中，语言模型仅用于构建阶段，而执行则由固定的节点库和引擎完成，确保结果可复现；同时支持增量式迭代优化，允许仅修改需调整的部分而非从头重建，显著提升效率。为验证有效性，作者构建了包含64个任务的QL-Bench基准测试，结果显示QLWF在三轮实验中对所有任务均生成结构正确且可执行的工作流，平均输出合理性达98.4%，显著优于基于提示词的基线方法；在12个生命周期任务的增量修正测试中，全部成功且仅消耗约三分之一的令牌开销。论文还开源了节点库、基准数据集、工作流模板及平台，为定量句法研究提供可复用的基础设施。
链接: https://arxiv.org/abs/2608.10662
作者: He Wang,Jingbo Chen,Yuqiao Lai,Nan Yang,Hanwen Zhang,Wei Yuan
机构: 国防科技大学(NUDT)(National University of Defense Technology)
类目: Multiagent Systems (cs.MA)
备注:
Abstract:Quantitative language research often depends on long chains of computational steps, yet the logic connecting those steps usually remains buried in scripts. This makes analyses harder to inspect, share, and revise than they need to be. Focusing on quantitative syntax, we present QLWF, a visual workflow platform that turns natural-language research descriptions into executable workflows through an AI assisted five-stage pipeline. In this setting, reification makes the research logic visible as a workflow, while formalization gives that workflow deterministic execution semantics. The language model is used only during construction. Execution is handled by a fixed node library and engine, which keeps the resulting workflows reproducible. QLWF also supports incremental refinement, so saved workflows can be revised by changing only the parts that need to change rather than being rebuilt from scratch. To evaluate the approach, we build a 64-task benchmark called QL-Bench from the quantitative-syntax literature. Across three runs, QLWF produces structurally valid and executable workflows for every task and reaches a mean output-plausibility rate of 98.4%, well above the prompt-based baselines. On a separate 12-task lifecycle benchmark, this refinement process succeeds in every case and uses roughly one-third of the tokens required by full regeneration. The paper also releases the node library, benchmark, workflow templates, and platform as reusable resources for quantitative-syntax research.
[MA-3] ASCon: A Direction-Aware Reciprocal Agent --Step Contextualization Model for Failure Attribution in Multi-Agent Systems
【速读】：该论文旨在解决大语言模型（Large Language Model, LLM）驱动的多智能体系统（Multi-Agent System, MAS）中故障归因（failure attribution）的问题，即精准识别导致故障的责任主体，包括故障智能体、错误执行步骤及故障模式，并明确其发生时间与原因。现有方法通常针对不同归因目标构建专用模型，忽视了各类目标之间共享的诊断证据依赖关系。尽管故障智能体、错误步骤和故障模式在语义上各异，但它们均依赖于多智能体轨迹中的共性诊断信息，如任务约束、智能体角色、行为历史以及智能体间交互等。为利用这一共性，本文提出一种统一表征模型——ASCon（Direction-Aware Reciprocal Agent–Step Contextualization），其核心创新在于通过方向感知图注意力机制建模执行上下文，采用掩码步骤到智能体注意力构建行为感知的智能体表征，并引入智能体条件化的步骤上下文化机制将智能体上下文回传至步骤表征中，从而实现对轨迹证据的高效聚合。最终生成的上下文化表征可通过轻量级的目标特定头适配多种归因任务。实验表明，相较于现有方法，ASCon在微精度（micro-accuracy）上提升故障智能体检测5.83%以上、故障步骤检测10.63%以上，在宏平均F1（Macro-F1）上提升故障模式检测14.73%以上；同时显著增强基于大语言模型的方法在跨域场景下的归因能力。
链接: https://arxiv.org/abs/2608.10646
作者: Shuyu Jiang,Yue Ran,Kaiyu Xu,Xingshu Chen,Yi Zhang,Hao Ren,Rui Tang,Tianwei Zhang
机构: Sichuan University (四川大学); Nanyang Technological University (南洋理工大学)
类目: Multiagent Systems (cs.MA)
备注:
Abstract:Failure attribution in LLM-based multi-agent systems (MAS) aims to answer who caused failures, when they occurred, and why by identifying responsible targets including faulty agents, erroneous steps, and failure modes. Existing methods have primarily focused on developing dedicated models for specific attribution targets, with limited attention to the evidential dependencies among them. Despite these attribution targets are different, they rely on common diagnostic evidence from MAS trajectories, including task constraints, agent roles, behavioral histories and inter-agent interactions. This commonality motivates us to develop a unified representation model that aggregates the trajectory evidence into individual agent and step representations, which can subsequently be adapted to different attribution targets. Accordingly, we propose ASCon, a direction-aware reciprocal \textbfAgent–\textbfStep \textbfContextualization model for multiple failure attribution targets. ASCon introduces direction-aware graph attention to model execution context, masked step-to-agent attention to construct behavior-aware agent representations, and agent-conditioned step contextualization to incorporate agent context back into step representations. The resulting contextualized representations enable different attribution targets through lightweight target-specific heads. Experiments show that ASCon can improve faulty-agent detection by 5.83%+ in micro-accuracy, faulty-step detection by 10.63%+ in micro-accuracy, and failure-mode detection by 14.73%+ in Macro-F1. Meanwhile, it can also substantially enhance the LLM-based methods’ attribution capabilities in out-of-domain scenarios.
[MA-4] MAP-Graph: Provenance-Aware Shared Memory for Multi-Agent Workflows
【速读】：该论文旨在解决语言模型智能体在长流程任务中共享内存时面临的权限与信任边界模糊问题，即尽管共享记忆可提升信息复用效率，但部分相关证据可能因权限、可信度或来源风险（如私有、污染、不可信或已被撤销）而不适合作为特定智能体或动作的输入。现有方法在语义检索、作用域访问控制或溯源追踪方面存在不足，未能清晰区分强制性授权（hard authorization）与渐进式信任（graded trust），也缺乏根据动作风险动态调整证据要求的能力。其解决方案的关键在于提出 MAP-Graph——一种具备溯源感知的内存层，通过构建包含智能体、源数据、记忆、主张与动作的类型化执行图（typed execution graph），实现多维度控制：一方面基于权限过滤排除无权访问的记录，另一方面通过语义相似性与路径乘积信任度对可用记忆重新排序；同时，在动作执行前引入风险敏感门控机制（risk-sensitive gate），确保高风险操作需更高可信证据支持，且保留受影响的完整溯源链以供审计。实验表明，MAP-Graph 在 2,700 个合成任务基准上达到 94.96% 的整体任务成功率和 90.22% 的纯净场景成功度，验证了溯源信息作为实时操作控制信号的有效性，而非仅限于事后审计元数据。
链接: https://arxiv.org/abs/2608.10509
作者: Yiqi Wang,Zihao Yan,Jiaqi Zhang,Zhangkai Wu,Mingkai Zheng,Zequn Sun,Yanming Zhu,Taotao Cai
机构: 未知
类目: Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注:
Abstract:Shared memory helps language-model agents reuse information across long workflows, yet relevant evidence may not be admissible for a particular agent or action. Because restrictions propagate through derivations, summaries can conceal private, poisoned, untrusted, or revoked sources, enabling unauthorized reads or unsafe actions. Existing approaches provide semantic retrieval, scoped access, or lineage tracking, but do not clearly separate hard authorization from graded trust or adapt evidence requirements to action risk. We introduce MAP-Graph, a provenance-aware memory layer that represents agents, sources, memories, claims, and actions in a typed execution graph. It traces ancestry, excludes permission-ineligible records, reranks eligible memories by semantic similarity and multiplicative path trust, and applies a risk-sensitive gate before action execution while retaining affected lineage for audit. On a controlled benchmark of 2,700 synthetic tasks per method across three domains, MAP-Graph achieves 94.96% overall task success, 72.70% exact decision accuracy, and 90.22% in the clean setting, where success requires a correct \textscAllow rather than a safe intervention. Ablations isolate the roles of permission filtering, path trust, and action gating, while transfer tests with two additional backbones preserve the exact-decision and access-control advantages. These results support provenance as an operational control signal, rather than only post-hoc audit metadata, within the evaluated setting.
[MA-5] GeoForge: Non-Parametric Self-Evolving Agents for Earth-Observation Reasoning
【速读】：该论文旨在解决地球观测（Earth Observation, EO）智能体在构建科学有效工具工作流时面临的多重约束挑战，包括传感语义、产品依赖性、时空兼容性及参数要求等。现有方法通常对每个查询进行宽泛的操作空间搜索，而近期自演化系统未能充分将异构的EO轨迹组织为跨决策层级可复用的知识。针对此问题，本文提出GeoForge——一种无需训练的自演化框架，其核心在于将已完成的任务轨迹转化为结构化的非参数化执行状态。该方案的关键创新在于：基于感知上下文约束操作空间，并从三个互补的记忆模块中检索任务相关的先验知识——工作流图记忆（Workflow Graph Memory）捕获全局操作顺序，动作级经验（Action-Level Experiences）提供局部修正，适配型技能标准操作程序（Adapted Skill Standard Operating Procedure）则保留流程与数据约束。所检索的先验引导工具执行，同时保持当前观测作为最终结论的基础。每次任务完成后，通过安全门控的提炼过程将具有地基性的轨迹转化为可复用的执行知识，实现执行—提炼—重用的闭环，从而在不更新骨干大语言模型（LLM）的前提下持续提升规划能力。实验表明，GeoForge在多个地理空间基准测试中显著提升了任务准确率与工具使用轨迹质量，同时大幅减少多数LLM的工具规划与推理错误。
链接: https://arxiv.org/abs/2608.10494
作者: Xin Xiao,Jiang Zhong,Junnan Zhu,Yingchao Feng,Peijin Wang,Yidan Zhang,Kaiwen Wei
机构: 未知
类目: Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注:
Abstract:Earth observation (EO) agents construct scientifically valid tool workflows and ground their conclusions in current geospatial evidence. This is challenging because EO workflows are constrained by sensing semantics, product dependencies, spatial and temporal compatibility, and parameter requirements. Existing agents often search a broad operation space for each query, while recent self-evolving systems do not fully organize heterogeneous EO trajectories into reusable knowledge across different decision levels. To solve this problem, we present GeoForge, a training-free, self-evolving framework that transforms completed trajectories into a structured nonparametric execution state. GeoForge constrains the operation space according to the sensing context, then retrieves a task-conditioned prior from three complementary memories. Workflow Graph Memory captures global operation order, Action-Level Experiences provide local corrections, and the Adapted Skill Standard Operating Procedure preserves procedural and data constraints. The retrieved prior guides tool execution, while current observations remain the basis of the final answer. After each task, a safety-gated distillation process converts grounded trajectories into reusable execution knowledge for future retrieval. This execution, distillation, and reuse loop improves planning without updating the backbone LLM. Experiments on multiple geospatial benchmarks demonstrate that GeoForge consistently improves both task accuracy and tool-use trajectory quality across diverse LLM backbones, while substantially reducing tool-planning and reasoning errors for most LLMs.
[MA-6] Persistent Recursive Worlds Enable Autonomous Software Evolution
【速读】：该论文旨在解决复杂软件系统在长期演化过程中因个体开发者生命周期有限而导致的持续性与传承性难题。传统代理式软件系统依赖持久会话、记忆或共享上下文来维持连续性，但这种模式受限于单个代理的寿命。为突破此限制，论文提出EvoX Genesis（简称Genesis）这一新型架构，其核心创新在于将软件项目本身作为持久实体，而非依赖长期存活的代理。关键解决方案是构建一个持久递归世界（persistent recursive world）：每个局部世界由一个已接受的版本和仓库路径定义，有限寿命的代理可提出局部变更，通过递归委托机制实现跨路径工作流转，仅经验证的成果才推进全局版本历史。该设计使系统具备长期演化能力，同时支持代理的动态替换与迭代。实验表明，基于DeepSeek V4 Flash从零构建一个约25万行代码的Rust语言C编译器，历时超120小时，生成逾千次代理事件，模型调用成本仅44美元；编译器通过全部c-testsuite及多数LLVM和Csmith测试。另一场景中，使用GLM 5.2在重复更换代理的情况下仍保持完整测试性能。此外，对13个含超10万行Fortran代码的MESA模块重实现为Rust工作区后，在六个数值负载上实现1.55–6.87倍的中位加速。结果证明，以持久化项目为核心组织长周期软件开发是可行且高效的范式。
链接: https://arxiv.org/abs/2608.10450
作者: Beichen Huang,Zhenyu Liang,Bowen Zheng,Ran Cheng
机构: 未知
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA); Neural and Evolutionary Computing (cs.NE)
备注:
Abstract:Complex software systems develop over timescales that exceed the lifespan of any individual coding agent. Most agentic software systems preserve continuity through persistent sessions, memories, managers or shared context. We introduce EvoX Genesis (hereafter, Genesis), which instead makes the software project persistent while allowing local agents to remain finite-lived. Genesis represents software as a persistent recursive world: each local world is situated by an accepted version and a repository path, finite-lived agents propose local changes, recursive delegation moves work across paths, and only accepted consequences advance the persistent version history. We evaluate this organization across formation, continuation and redevelopment. Starting from a repository with no compiler implementation, Genesis used DeepSeek V4 Flash to build a Rust-based C compiler with about 250k tracked lines; the run lasted over 120 hours, archived over 1,000 agent episodes and incurred only US 44 in model-token charges. The compiler passed the complete c-testsuite and most LLVM and Csmith tests. In a separate compiler world generated with GLM 5.2, development continued after repeated agent replacement while retaining full test performance. Genesis also reimplemented 13 MESA modules with over 100k Fortran lines as a Rust workspace with nearly 90k Rust lines; across six numerical workloads, it achieved median speedups of 1.55–6.87x. These results show that long-horizon software development can be organized around a persistent project rather than a persistent agent.
[MA-7] Not a Monolith: Lab-Level Divergence in the Cooperative Equilibria of Chinese Frontier LLM Agents
【速读】：该论文旨在探究西方前沿大模型（LLM）中已报道的“合作偏见”是否适用于中国模型这一不同对齐谱系，并进一步探讨应将中国模型视为单一整体还是各自独立的实验室。其核心问题是：在去除先前研究中策略生成与代码转换能力混杂的混淆因素后，中国四大前沿模型（DeepSeek V4 Pro、Qwen3-Max、Kimi K2.5 和 GLM-5.1）在进化型重复囚徒困境中的合作行为是否存在显著差异。解决方案的关键在于采用固定编码转换器（GPT-5.4 Mini）以隔离策略生成能力的影响，从而实现跨实验室的纯生成能力比较。通过全对全锦标赛和规模为500次运行的莫兰过程，在三种提示风格和四种种群制度下进行实验，验证了两个预注册假设：H6（非单一整体）得到支持，四家实验室在激进均衡比例（P_A）上存在显著差异（从Qwen3-Max的1%到DeepSeek V4 Pro的9%，范围达8个百分点），且六组配对比较中有四组在Holm-Bonferroni校正后仍显著；而H5（合作偏见普适性）虽部分成立但需谨慎解读，中国模型在12个组合中有6个呈现合作多数，与西方模型的9/12相近，且在稳健性检验中提升至9/12，表明合作倾向并非仅由生态系统决定。结论指出，模型的协作倾向应以“实验室”而非“生态系统”为基本单位，将“中国模型”视为单一整体缺乏实证支持。
链接: https://arxiv.org/abs/2608.10262
作者: Francisco León Zúñiga Bolívar(Institución Universitaria Colegio Mayor del Cauca)
机构: Institución Universitaria Colegio Mayor del Cauca(科尔多瓦大学学院机构); Popayán(波帕扬); Colombia(哥伦比亚)
类目: Multiagent Systems (cs.MA)
备注: 9 pages, 8 tables. Companion study to arXiv:2605.29874 , under a fixed-converter design. Code and replication package: this https URL (archived: this https URL )
Abstract:Does the cooperative bias documented for Western frontier LLM agents extend to a different alignment lineage, and should the Chinese models that embody it be treated as a single bloc or as distinct laboratories? We study four frontier-tier Chinese models - DeepSeek V4 Pro, Qwen3-Max, Kimi K2.5 and GLM-5.1 - in an evolutionary Iterated Prisoner’s Dilemma, under a design that removes a confound present in prior work. Rather than letting each model convert its own natural-language strategies into code, which entangles strategic disposition with coding ability, we hold the converter fixed (GPT-5.4 Mini) across all labs, so every cross-lab comparison is a comparison of generation alone. We run the full protocol: all-play-all tournaments and a Moran process at n=500 runs per condition, across three prompt styles and four population regimes. Two pre-registered hypotheses are evaluated. H6 (not monolithic) is supported: the four labs differ significantly in aggressive-equilibrium proportion, P_A running from 1% for Qwen3-Max to 9% for DeepSeek V4 Pro, with four of six pairwise comparisons surviving Holm-Bonferroni. The spread across the four labs (P_A range 8pp) is larger than the difference between the Chinese and Western ecosystems’ mean P_A (5.0% vs 5.0%): on this measure, within-ecosystem variation exceeds the East-West gap. H5 (cooperative-bias generality) is consistent but qualified: a cooperative plurality holds in 6 of 12 lab-prompt combinations against the 9 of 12 reported for Western models, a difference we do not treat as firm, since the count rests on Cooperative-Neutral near-ties and rises to 9/12 under an alternate converter in our pre-registered robustness check. The lab, not the ecosystem, is the unit at which cooperative disposition is set; treating “Chinese models” as a monolith is not supported by the evidence.
[MA-8] he Deliberative Deficit: An Empirical Critique of LLM s in Democratic Discourse
【速读】：该论文旨在解决生成式人工智能（Generative AI）在处理复杂、价值导向的非可验证问题时，其集体推理能力评估不足的问题。传统基准测试仅适用于数学、编程等具有客观正确答案的任务，无法有效衡量在缺乏唯一正确解的情境下，模型能否通过整合多元视角达成共识性决策。其核心挑战在于：现有对大语言模型（LLM）对话过程的程序性评价（如尊重性、论证合理性、参与度）不足以反映真实群体理性讨论的质量。为此，研究引入政治科学中经验证的“协商推理指数”（Deliberative Reason Index, DRI），用于量化评估非可验证议题下的群体推理可靠性。基于1,980次五代理论模型在12个公民议事主题上的实验，结果表明，尽管LLM群体的对话程序质量接近人类议事水平，但跨主体一致性提升有限且高度依赖议题类型，主要集中在易解而非伦理争议性强的问题上；更重要的是，LLM群体的视角多样性仅为人类议事会的三分之一，且呈现与人类相反的趋势——人类在协商过程中趋于收敛，而LLM则加剧观点分化。通过角色提示增强多样性虽能改变信息更新机制，却未能恢复人类的协同演化模式。因此，研究结论具有约束性：当前证据不支持将大语言模型视为独立的协商主体，但其仍可作为辅助人类进行多元议题推理的工具。
链接: https://arxiv.org/abs/2608.10186
作者: Maurice Flechtner
机构: 未知
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注: 10 pages, archival publication at AIES 2026
Abstract:LLMs are increasingly deployed in settings that require collective reasoning on complex, value-laden problems. Confidence in these deployments rests largely on benchmarks for verifiable tasks (mathematics, coding, coordination games), yet many of these applications concern problems where no objectively correct answer exists and where decision quality instead depends on integrating pluralistic perspectives to find mutually acceptable solutions. We argue that LLM reasoning capacity on this class of problems cannot be fully inferred from verifiable-task benchmarks, and that procedural evaluations of LLM discourse (respectfulness, justification, engagement) are systematically insufficient. We apply the Deliberative Reason Index (DRI), a measure developed in political science and validated across citizen assemblies, as a tool for evaluating reliable group-level reasoning on pluralistic, non-verifiable problems. Synthesizing recent evidence across 1,980 five-agent LLM runs on 12 citizen-assembly topics across 11 frontier model configurations, we find that LLM groups produce discourse with procedural quality comparable to human deliberation, while gains in intersubjective consistency are small, topic-dependent, and concentrated on tractable rather than ethically contested questions. LLM groups exhibit roughly one-third the perspective diversity of human assemblies and reverse the human convergence pattern: human deliberation decreases dispersion as diverse views synthesise, whereas LLM deliberation increases it. Engineering diversity through persona prompting does not restore the human dynamic but inverts which component of deliberative reasoning is updated. Our conclusion is constraining rather than prohibitive: LLMs can function as tools supporting human reasoning on pluralistic problems, but current evidence does not license treating them as autonomous deliberative agents.
[MA-9] Beyond Cash Flows: A Multi-Agent AI Framework for Valuing Clinical-Stage Cross-Border Biotechnology
【速读】：该论文旨在解决当前多智能体（multi-agent）投资分析系统在临床阶段生物技术企业估值中的根本性缺陷。现有框架普遍依赖传统现金流折现模型，无法适用于以科学与监管里程碑为价值核心的早期生物技术资产。为此，论文提出一种专用多智能体架构，其关键在于三个创新层：第一，估值层将定性的科学判断转化为可辩护的早期资产估值；第二，跨市场协调层实现对国际多个交易场所价格的同步调和；第三，冲突融合机制在领域特定语境下系统性调和乐观的科学预期与审慎的监管约束。该架构并非理论构想，而是源自作者作为中国首只跨境生物技术基金唯一组合经理时的人工实践，其十六个月内实现127.17%的收益率，显著优于50.67%的基准，验证了方法论的有效性。本文聚焦于架构层面的设计原则，为将智能体投资系统拓展至复杂、事件驱动型资产类别提供基础范式。
链接: https://arxiv.org/abs/2608.10175
作者: Yuhan Fang
机构: CPC Scientific Inc.
类目: Multiagent Systems (cs.MA); Portfolio Management (q-fin.PM)
备注:
Abstract:A new class of software systems is transforming investment analysis. Large language model agents assembled into collaborative team structures including analysts, researchers, and risk managers are increasingly deployed across financial markets. Yet current multi-agent frameworks share a critical limitation: they rely on the foundational assumption that companies can be valued through traditional cash flows. This paradigm fails in clinical-stage biotechnology, where enterprise value depends entirely on binary scientific and regulatory milestones. To bridge this gap, this paper introduces a specialized multi-agent framework. Its valuation layer translates qualitative scientific judgment into defensible valuations for pre-revenue assets; its cross-market coordination layer reconciles pricing across international venues simultaneously; and its conflict-fusion mechanism systematically arbitrates between bullish scientific conviction and cautious regulatory constraints in a domain-specific manner. Crucially, the architecture is not a speculative design: it encodes a method the author first executed by hand as sole portfolio manager of China’s first dedicated cross-border biotechnology fund, a human practice that returned 127.17% against a 50.67% benchmark within sixteen months. That record is evidence for the underlying method rather than for any AI system; no implementation is evaluated here. This paper presents the framework at the architectural level, establishing foundational design principles for extending agentic investment systems into complex, event-driven asset classes they currently serve poorly.
[MA-10] Automating and Scaling Behavioral Scientific Research on AI Agents
【速读】：该论文旨在解决当前人工智能代理（AI agents）在复杂环境中行为研究仍依赖人工、耗时费力的问题。其核心挑战在于如何高效、系统地开展针对AI代理的行为科学探究，尤其是在面对多样化目标行为时缺乏自动化手段。解决方案的关键在于提出AEROBAT——首个实现对AI代理行为科学研究自动化的多智能体系统。该系统能够根据用户指定的目标行为，自主完成从假设生成、受控实验设计与执行、行为评估、结果分析到报告撰写的全流程研究工作。通过在12种目标行为上生成并验证79个假设，共设计1,240个受控实验，执行23,512次仿真轮次，AEROBAT成功识别出26个具有中等至强统计证据支持的假设，其中包含若干新发现。这一成果表明，自动化行为科学研究可有效补充并拓展传统人工研究的范围与效率。
链接: https://arxiv.org/abs/2608.10030
作者: Soo Yong Lee,Jongha Lee,Jaewan Chun,Hyunjin Hwang,Fanchen Bu,Ziv Ben-Zion,Taekwan Kim,Denny Borsboom,Jaemin Yoo,Kijung Shin
机构: KAIST, Kim Jaechul Graduate School of AI; Yale, Department of Psychiatry; University of Haifa, School of Public Health; UCL, Mental Health Neuroscience Department; UvA, Department of Psychology; SNU, Department of CSE
类目: Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注: preprint
Abstract:As AI agents are increasingly deployed in complex environments, understanding their behaviors becomes critical. Yet behavioral scientific research on AI agents remains manual and labor-intensive. We introduce AEROBAT, the first multi-agent system to automate behavioral scientific research on AI agents. Given an arbitrary target behavior by its user, AEROBAT automatically executes a full pipeline of behavioral scientific research—generating hypotheses about the behavior, designing and executing controlled experiments, making behavioral assessments, analyzing the results, and writing reports. For 12 target behaviors, we used AEROBAT to generate and test 79 hypotheses: designing 1,240 controlled experiments and executing 23,512 simulation rounds in total. Moderate-to-strong statistical evidence was found for 26 hypotheses, including some novel ones. In sum, our results demonstrate that automated behavioral scientific research on AI agents can complement and extend the reach of manual research.
[MA-11] Sheaf-Based Federated Representation Learning
【速读】：该论文旨在解决异构联邦系统中因数据分布差异、感知模态多样性、模型架构异质性、潜在空间维度不一致及局部学习目标不同而导致的表示学习与信息交换难题。其核心挑战在于如何在无共享全局潜在空间假设的前提下，实现跨代理的一致且高效的表示对齐。解决方案的关键在于提出基于层化结构（Sheaf）的联邦表示学习框架（SFRL），通过可学习的层化限制映射（learnable sheaf restriction maps）引入流形约束的几何对齐正则项，利用由层化拉普拉斯算子诱导的二次粘合正则项，强制邻近代理间的潜在表示通过正交变换和等距嵌入实现对齐，从而在局部优化过程中自然涌现全局一致性。该正则项仅需在少量共享的引导样本上计算，保障了算法的可扩展性与通信效率。为此设计了一种去中心化的求解算法 Sheaf-FRL，交替执行局部模型的梯度更新与边级限制映射的闭式 Procrustes 更新，并证明其在确定性和随机设置下均能收敛至一阶驻点。实验验证表明，该方法在语义通信场景下的协作分类任务中，相较于基线方法，在不同水平的局部分布偏移下均实现了更高的本地与通信后分类精度，并展现出更强的潜在空间降维鲁棒性。
链接: https://arxiv.org/abs/2608.10016
作者: Gabriele D’Acunto,Enrico Grimaldi,Valeria Avino,Mario Edoardo Pandolfo,Leonardo Di Nino,Sergio Barbarossa,Paolo Di Lorenzo
机构: 未知
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA); Signal Processing (eess.SP)
备注:
Abstract:Heterogeneous federated systems require agents to learn and exchange informative representations despite differences in data distributions, sensing modalities, model architectures, latent dimensionalities, and local learning objectives. To address this challenge, we propose Sheaf-based Federated Representation Learning (SFRL), a general framework that jointly optimizes local objectives with a manifold-constrained geometric alignment regularizer based on learnable sheaf restriction maps. Unlike most existing approaches, SFRL does not assume a shared global latent space. Instead, global consistency emerges from the alignment of neighboring latent representations through orthogonal transformations and isometric embeddings. This alignment is enforced by a quadratic gluing regularizer induced by the sheaf Laplacian, whose learnable restriction maps adapt the geometry to the observed data. The penalty is evaluated on a small set of shared pilot samples, ensuring scalability and communication efficiency. We develop a decentralized algorithm for solving SFRL, termed Sheaf-FRL, which alternates between gradient updates of the local models and closed-form Procrustes updates of the edge-wise restriction maps. We further establish convergence of Sheaf-FRL to first-order stationary points in both deterministic and stochastic settings. As an application, we consider a cooperative classification task in the context of semantic communication, under model and data heterogeneity. Our results show that Sheaf-FRL outperforms baseline approaches in terms of local and post-communication classification accuracy across different levels of local distribution shift and exhibits greater robustness to latent-space dimensionality compression.
[MA-12] RIBE: Predicting Team Performance via Communication Behavior Ensembles
【速读】：该论文旨在解决在缺乏特定任务知识的情况下，如何有效设计能够辅助人类团队的自主代理（autonomous agents）这一关键问题。其核心挑战在于传统绩效指标难以捕捉团队协作中的深层行为动态，从而限制了对团队表现的精准预测与及时干预。本文提出TRIBE方法，一种领域无关的团队行为分析框架，通过挖掘通信模式揭示传统指标无法识别的团队行为特征。其解决方案的关键在于：利用早期阶段（任务进行至10%时）的通信模式，将团队划分为具有性能预测能力的行为群体（behavioral tribes），实现对团队未来表现的早期预测。研究进一步表明，任务结构所赋予的行为自由度决定了通信模式预测性能的强度。此外，时间序列分析发现AI代理会显著改变团队的行为轨迹，而人类顾问则更倾向于顺应自然行为演化，且团队在整个协作过程中保持较高的行为灵活性。通过与Llama模型对比并优化流程，TRIBE实现了显著的速度提升与性能改进，验证了其在实际应用中的高效性与可扩展性。
链接: https://arxiv.org/abs/2608.06926
作者: Ali Jalal-Kamali,Nikolos Gurney,David V. Pynadath,Fred Morstatter
机构: University of Southern California (南加州大学); University of Southern California (南加州大学); Naval Research Laboratory (海军研究实验室); University of Pennsylvania (宾夕法尼亚大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Human-Computer Interaction (cs.HC); Multiagent Systems (cs.MA)
备注:
Abstract:Designing autonomous agents that effectively assist human teams hinges on understanding team dynamics, often without task specific knowledge. We present TRIBE, a domain independent approach that reveals team behavioral dynamics invisible to traditional performance metrics. We show that communication patterns can categorize teams into performance predictive behavioral tribes, as early as 10% into the task, enabling timely interventions. We test TRIBE on four diverse datasets and demonstrate that communication patterns predict team performance while the prediction strength varies by the degree a task structure allows for behavioral freedom. Our temporal analysis reveals that AI agents significantly alter team behavioral trajectories while human advisors align with natural dynamics, and that teams maintain behavioral flexibility throughout collaboration. Further, we compare TRIBE to Llama and optimize the pipeline, achieving significant speedup with performance improvement.
[MA-13] Scaling Laws for Majority-based Opinion Dynamics in the Presence of Stubborn Agents
【速读】：该论文旨在研究在多智能体系统中，固执型智能体（stubborn agents）如何影响网络内意见分布的动态演化。具体而言，当系统中同时存在固执型与非固执型智能体时，非固执型智能体依据“2k-选择规则”更新其意见：从邻居中（包括固执与非固执个体）随机均匀采样2k个个体，并采纳所采样群体及其自身中的多数意见。假设比例为γ₀和γ₁的智能体分别固执于意见0和1。尽管直觉上稳态意见分布应由固执比例较大的意见主导，但论文揭示，达到稳态所需的时间高度依赖于γ₀与γ₁的具体取值及其差异：当两者数值较小且差距较小时，收敛时间可能呈指数级增长（随网络规模指数上升）；而当至少一个参数较大时，系统可在仅对数时间内达到稳态。因此，系统表现出基于固执比例的显著相变现象。在相变边界区域，通过施泰因方法（Stein’s method）证明系统动态受扩散过程驱动，混合时间仅为多项式量级。该研究的关键在于揭示了固执比例对系统收敛速度的非平凡影响，并建立了从离散动力学到连续扩散过程的理论桥梁。
链接: https://arxiv.org/abs/2608.11071
作者: Luke Meredith,Arpan Mukhopadhyay
机构: University of Warwick(华威大学)
类目: Probability (math.PR); Multiagent Systems (cs.MA)
备注:
Abstract:In a multi-agent system, there are often stubborn followers of specific opinions or beliefs. Motivated by this observation, in this paper, we aim to understand how stubborn agents affect the distribution of opinions in a network where both stubborn and non-stubborn agents interact with each other. To do so, we assume that all agents have an opinion in the set \0,1\ and each non-stubborn agent updates its opinion according to the 2k \textit-choices rule, where the agent samples 2k neighbours (including both stubborn and non-stubborn neighbours) uniformly at random and adopts the majority opinion among the sampled group of neighbours and itself. We assume that a proportion of agents, \gamma_i , are stubborn followers of opinion i\in \0,1\ . It is natural to expect that the steady-state distribution of the opinions in the network will be dominated by the opinion with the larger proportion of stubborn followers. We show that while this is true, the time to reach steady-state depends heavily on the values of the parameters \gamma_0 and \gamma_1 . When the individual values of these parameters, as well as their difference, are small, it can take an exponentially long time (in the network size) to reach the steady-state. In sharp contrast, when at least one of the parameters \gamma_0 and \gamma_1 is large, the network reaches the steady-state in a time that is only logarithmic in the network size. Hence, there exists a sharp phase transition in the network dynamics based on the proportions of stubborn agents. We also characterise the behaviour of the system when the parameters \gamma_0 and \gamma_1 lie on the boundary of the phase transition. In this boundary region, we show using Stein’s method that the dynamics are driven by a diffusion process which takes polynomial time to mix.
自然语言处理
[NLP-0] ConVAWG: A Retrieval-Grounded Framework for Controlled Synthetic Dialogue Generation in Violence Against Women and Girls
【速读】： 该论文旨在解决在敏感领域（如针对女性和女童的暴力行为，VAWG）中由于隐私与法律限制导致真实对话数据难以获取、发布或标注的问题，尤其针对现有研究多聚焦于单句层面的毒性分析，而忽视了暴力行为作为关系性、时间演进过程的复杂特性。其核心挑战在于如何在保障数据安全的前提下，生成具有高情境真实性与语义连贯性的多轮对话，以支持对隐蔽性、渐进式暴力行为的系统性研究。解决方案的关键在于提出一种基于检索增强的合成对话生成框架——ConVAWG，该框架通过整合人物设定种子、英国国家统计局的人口统计模式、官方犯罪定义及实际家庭谋杀审查案例，构建层次化事件时间线，并在此基础上生成多场景角色扮演对话；同时引入目标激活引导的毒性控制机制，精准调控特定话语中的有害内容表达，确保生成内容在保持领域真实性的同时符合伦理规范。最终，研究释放了超过6000个跨200个场景的多轮对话数据集，涵盖丰富的场景、事件与回合级元信息，经由人工评估、大模型判别、消融实验及下游任务验证，证明了其在对话质量与领域保真度方面的显著优势。
链接: https://arxiv.org/abs/2608.11200
作者: Chen Lyu,Xingwei Tan,Simon Cullen,Shelley Wilson,Lois Arthurs,Arshad Jhumka,Gabriele Pergola
机构: University of Sheffield(谢菲尔德大学); Forensic Capability Network(法医能力网络); University of Leeds(利兹大学); University of Warwick(华威大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Synthetic dialogue generation offers a way to study conversational dynamics in sensitive domains where real data are difficult to access, release, or annotate. The underlying abuse may occur online or offline: threats and coercion can appear directly in messages, while behaviours such as surveillance, isolation, stalking, and physical violence may be planned, disclosed, or referred to conversationally. Privacy and legal constraints make it difficult the release of large-scale real conversation datasets; existing work has mostly focused on sentence-level toxicity of online abuses, leaving a gap in modelling abuse as a relational and temporally unfolding phenomenon. In this work, we focus on modelling Violence Against Women and Girls (VAWG) scenarios as multi-turn dialogues. We introduce ConVAWG, a retrieval-grounded framework for generating CPS-aligned synthetic VAWG chat dialogues. ConVAWG builds scenarios from persona seeds, demographic patterns reported by the UK Office for National Statistics, official crime definitions, and retrieved Domestic Homicide Review cases; converts them into hierarchical event timelines; generates multi-scene role-play dialogues; and applies targeted activation-steered toxicity control to appropriate utterances. We release over 6,000 multi-turn dialogue events across 200 scenarios with rich scenario-, event-, and turn-level metadata. Extensive human evaluation, LLM-as-Judge assessment, ablations, and downstream tasks show strong dialogue quality and domain fidelity.
[NLP-1] Beyond a Bag of Features: Set-Level Instability in Sparse Autoencoders
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）表征是否能够准确反映人类对概念类别及其典型性结构的认知这一关键问题。已有研究（Shani et al., 2026）指出，尽管密集型模型表征在整体上可恢复人类的概念边界，却无法捕捉细粒度的典型性结构。本文提出以稀疏自编码器（Sparse Autoencoder, SAE）激活潜空间集合之间的重叠率作为更可解释的相似性度量，替代原有的余弦相似度方法。其解决方案的关键在于：通过引入基于集合层面操作的重叠度量，揭示模型内部表征的实际语义组织方式。研究发现，虽然SAE激活集能有效恢复受控玩具模型中的并集式组合结构，并在自然文本中形成语义连贯的邻域，但在人类概念分析中，其表现并未优于密集嵌入或残差流状态，反而更忠实于模型内部的相似性结构而非人类认知结构。进一步在受控语义扰动下的分析表明，人类对概念变化的判断与SAE激活集的变化之间存在显著偏差，说明在非理想化场景下，SAE特征并非通过简单的“特征袋”（bag-of-features）语义进行组合，从而揭示了当前模型表征与人类认知之间存在的根本性差距。
链接: https://arxiv.org/abs/2608.11197
作者: Nikolai Bolik,Lennart Stöpler,Artur Andrzejak
机构: Heidelberg University (海德堡大学)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:
Abstract:Shani et al. (2026) show that LLM representations broadly recover human category boundaries, while failing to reflect fine-grained typicality structure. Their analysis uses cosine similarity over dense model representations. We revisit their approach using overlap over active sparse autoencoder (SAE) latent sets as a more interpretable similarity measure. We first verify that this set-level measure is meaningful: SAE latent sets can recover union-like compositional structure in controlled toy models and induce semantically coherent neighborhoods in natural text. Extending the human-concepts analysis to SAE set similarities, we find that SAE activation sets do not recover human category boundaries or within-category typicality more faithfully than dense embeddings or residual-stream states, but instead track model-internal similarity structure. To probe this gap further, we study active latent sets under well-controlled semantic modifications, revealing a substantial mismatch between human judgements of conceptual change and change in the SAE active set. We interpret this as evidence that, outside idealised settings, SAE features do not compose via simple bag-of-features semantics.
[NLP-2] st-Time Self-Evolving GUI Visual Grounding via Reflection-Guided On-Policy Self-Distillation
【速读】： 该论文旨在解决GUI视觉定位（GUI Visual Grounding）模型在部署后难以适应未见过的界面的问题，现有模型通常在部署后冻结参数，无法实现动态优化。其核心挑战在于：尽管已有方法尝试通过测试时强化学习进行自适应，但缺乏对失败探索过程的反思能力，导致学习效率受限。为此，本文提出一种测试时自演化（Test-Time Self-Evolving）框架，构建了“探索—评估—反思—内化”的闭环机制。该框架的关键创新在于引入基于多模态大语言模型（MLLM）的反射器（Reflector），用于对代理的探索结果进行评估并生成高阶推理性反思；进一步提出反射引导的在线策略自蒸馏（Reflection-Guided On-Policy Self-Distillation），将高层反思逻辑转化为细粒度的词元级监督信号，通过条件自教师（conditioned self-teacher）实现模型权重的渐进更新；同时设计对比校准（Contrastive Calibration）方法，有效防止错误的自回归前缀在失败探索中污染监督信号。实验表明，该框架在六个基准上平均提升准确率7.4%，是首个成功在测试时利用在线策略自蒸馏实现GUI视觉定位自适应的研究，显著增强了GUI代理的后部署自演化能力。
链接: https://arxiv.org/abs/2608.11191
作者: Shiyu Xuan,Zechao Li
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:GUI Visual Grounding is a fundamental capability for GUI agents. Existing models typically freeze their parameters after deployment, limiting their ability to adapt to unseen interfaces. Although recent methods attempt to adapt models via test-time reinforcement learning, they cannot reflect upon failed exploration. To overcome this, we propose a Test-Time Self-Evolving framework that enables models to improve after deployment without human-annotated ground truth. It constructs a closed-loop of Exploration, Evaluation, Reflection, and Internalization. Specifically, the agent first explores unseen interfaces by predicting grounding coordinates for given instructions. To evaluate these explorations, we introduce an MLLM-based Reflector to assess the generated results and provide the corresponding reasoning reflections. To internalize reflection knowledge into the model weights, we propose Reflection-Guided On-Policy Self-Distillation, which translates high-level reasoning into dense token-level supervision via a conditioned self-teacher. Furthermore, we design a Contrastive Calibration method to prevent incorrect auto-regressive prefixes from corrupting the supervisory signals during failed explorations. Extensive experiments across six benchmarks demonstrate our framework’s effectiveness, achieving an average accuracy improvement of 7.4% over the base model. To the best of our knowledge, this is the first work to successfully exploit on-policy self-distillation for test-time adaptation in GUI visual grounding. By filling the gap in post-deployment adaptation, our framework completes the self-evolving capability of GUI agents. The code will be released.
[NLP-3] From Interpretability to Control: Insights from Six Years of the TrustNLP Workshop ACL EACL2027
【速读】： 该论文旨在解决生成式自然语言处理（Generative Natural Language Processing, GNLP）系统在可信性（trustworthiness）方面的核心挑战，特别是在模型能力快速演进背景下，如何系统性地理解并提升模型在真实性（truthfulness）、安全性（safety）、公平性（fairness）、可解释性（explainability）、可控性（controllability）与鲁棒性（robustness）等六大可信维度上的表现。其解决方案的关键在于构建一个基于成熟框架（TrustLLM、DecodingTrust）的分类体系，对六届TrustNLP研讨会共144篇论文进行系统性梳理与分析，揭示可信维度随模型演进的动态演变规律。研究发现，随着大模型能力的涌现，可信维度呈现协同激活特征：首个高影响力对话模型发布后，所有可信维度同时被关注；后续迭代则聚焦于真实性与安全对齐。尤其值得注意的是，真实性成为增长最快的维度（2021–2022年未出现，至2025–2026年已占37%），而可解释性经历了从后置解释方法衰退到基于机制理解的解释方法复兴的U型曲线转变。通过与ACL、NAACL、EACL、EMNLP等主流会议同期论文的对比，验证了TrustNLP议题分布与领域整体趋势高度一致，进一步凸显其代表性。最终，研究提炼出四项结构性洞察，并为未来研究社区提出可操作的发展方向。
链接: https://arxiv.org/abs/2608.11171
作者: Rahul Gupta,Abhinav Mohanty,Anaelia Ovalle,Anil Ramakrishna,Anubrata Das,Apurv Verma,Jwala Dhamala,Ninareh Mehrabi,Tharindu Kumarage,Yada Pruksachatkun,Yang Trista Cao,Kai-Wei Chang,Aram Galstyan
机构: Meta; Autodesk; New Jersey Institute of Technology; Salesforce; University of California, Los Angeles; Amazon AGI
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注: 17 pages, 2 figures, 3 tables. Submitted to ACL ARR August 2026 cycle (EACL 2027)
Abstract:The Workshop on Trustworthy Natural Language Processing (TrustNLP), co-located with major ACL conferences since 2021, has grown from 8 proceedings papers to 41 over six editions, documenting a field-wide transition from post-hoc interpretability of static models to mechanistic understanding and proactive control of generative systems. We synthesize insights from all 144 proceedings papers, classifying them along six trust dimensions grounded in established frameworks (TrustLLM, DecodingTrust). We observe co-occurrences with capability emergence. The release of the first high-impact chat models activated all trust dimensions simultaneously, while subsequent model generations shifted focus toward truthfulness and safety alignment. Analysis from the classification study reveals that truthfulness is the fastest-growing dimension (absent in 2021-2022, comprising 37% of papers by 2025-2026), fairness remains the most consistent theme, and explainability exhibits a U-shaped trajectory; declining as post-hoc methods lost relevance but resurging in 2026 through mechanistic interpretability. A cross-venue comparison with ACL, NAACL, EACL, and EMNLP (~2K papers) in the same period shows that TrustNLP’s topical distribution closely follows the field average. We identify four structural insights and conclude with actionable directions for the research community.
[NLP-4] MultiModal Code-Switching: Interleaving Visual Objects into Language for Explicit Object-Level Alignment
【速读】： 该论文旨在解决现有多模态大语言模型（Multimodal Large Language Models, MLLMs）在图像-文本对齐预训练中因依赖全局图像表征而导致的指代模糊性问题。由于模型仅基于整体图像表示进行对齐，难以准确捕捉多个视觉对象与文本实体之间的局部对应关系，从而造成数据利用效率低下和语义定位不精准。为此，论文提出一种名为多模态代码切换（MultiModal Code-Switching, MMCS）的新颖预训练范式，其核心在于通过引入显式的对象级监督，实现局部视觉-语言对齐。受语言学中“代码切换”现象启发，MMCS将文本中的实体替换为其对应的视觉对象，以交错方式强制模型建立局部、细粒度的跨模态关联。为支持该方法，研究进一步构建了一个可扩展的数据合成流水线，生成包含773,000条样本且具备精确对象-实体对应关系的预训练数据集。实验结果表明，MMCS具有极高的数据效率：仅使用5万样本即可达到甚至超越基于60万图像-文本对训练的模型性能；同时，在不同模型规模下均能持续提升视觉定位与感知能力，验证了其有效性与泛化性。
链接: https://arxiv.org/abs/2608.11167
作者: Changhao Xiang,Shangyu Xing,Zhen Wu,Jianbing Zhang,Xinyu Dai
机构: Nanjing University (南京大学)
类目: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Existing Multimodal Large Language Models (MLLMs) predominantly rely on image-text pairs for modality alignment pretraining, mapping global image representations to long textual descriptions. However, this image-level alignment suffers from referential ambiguity: models struggle to infer the correspondences between multiple visual objects and textual entities from the global representation, leading to data inefficiency and suboptimal semantic grounding. To address this, we propose MultiModal Code-Switching (MMCS), a novel pretraining paradigm that provides explicit object-level supervision. Inspired by the linguistic phenomenon of code-switching, MMCS interleaves vision and language by replacing textual entities with their corresponding visual objects, enforcing local vision-language grounding. We further develop a scalable data synthesis pipeline to generate a pretraining dataset of 773K samples with accurate object-entity correspondences. Experiments show that MMCS is highly data-efficient: with only 50K samples, it matches or surpasses models trained on 600K image-text pairs. Furthermore, MMCS consistently improves visual grounding and perception capabilities across varying model scales.
[NLP-5] he Illusion of Cross-Lingual Safety in Low-Resource Languages
【速读】： 该论文旨在解决大语言模型（LLM）在多语言场景下的安全对齐（safety alignment）有效性问题，尤其关注低资源语言中英文安全机制是否具备跨语言泛化能力。现有研究普遍假设英语训练的安全对齐可直接迁移至其他语言，但这一假设在低资源非洲语言（如蒂语、豪萨语、阿姆哈拉语和斯瓦希里语）中尚未得到充分验证，存在显著安全隐患。其解决方案的关键在于提出一种基于隐空间几何的评估框架——通过分析模型隐藏层中拒绝响应（refusal）的表示结构，而非依赖生成结果进行评估，从而更精确地探测安全机制的内在表征。同时，研究构建了LoDNA数据集，包含字面翻译与文化本地化提示的配对样本，以检验跨语言安全信号的传递效果。实验表明，跨语言安全转移严重受限，多数语言-模型组合中，有害提示的拒绝信号保留率低于10%；尽管字面与本地化提示在语义上高度一致（余弦相似度0.95–0.996），但其表示在模型深层发生偏移，表明模型虽能理解语义，却未能将这些概念有效路由至安全控制机制。该发现揭示当前多语言安全对齐本质上是表面化的，不支持“通用且语言无关的危害流形”这一核心假设，强调了针对低资源语言进行专门安全对齐的必要性。
链接: https://arxiv.org/abs/2608.11146
作者: Abigail Oppong,P Sam Sahil,Tadesse Destaw Belay,Maryam Ibrahim Mukhtar,Esmael Ahmed Abdu,Tassallah Abdullahi,Jessica Oparebea,Saminu Mohammad Aliyu,Idris Abdulmumin,Abubakar Juma Chilala,Nicholaus Dismas Ladislaus,Alfred Malengo Kondoro,Lemofouet Valdini Douglace,Shamsuddeen Hassan Muhammad,Seid Muhie Yimam
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Safety alignment in large language models (LLMs) is largely developed in English, assuming these safeguards generalize across multilingual settings. However, this assumption remains underexplored and exposes a vulnerability in low-resource languages. We investigate cross-lingual safety transfer in four African languages, Twi, Hausa, Amharic, and Swahili, using LoDNA, a new safety dataset that pairs literal translations with culturally localized prompts. To move beyond generation-based evaluation, we propose a latent geometric framework that probes hidden-state refusal representations in LLMs. Our experimental results show that cross-lingual safety transfer is severely limited; harmful prompts retain less than 10% of the English refusal signal across most language-model pairs. Literal and localized prompts are semantically aligned (cosine 0.95-0.996) but drift across layers, suggesting models encode the concepts without routing them to safety mechanisms. These findings demonstrate that current multilingual safety alignment is superficial, providing strong evidence against the assumption of a universal, language-agnostic harm manifold within the specific low-resource languages studied. Warning: This paper contains example data that may be offensive or harmful.
[NLP-6] Attention-Path Frag ility as an Uncertainty Signal in Large Language Models
【速读】： 该论文旨在解决大模型在生成过程中对预测结果的不确定性评估不足的问题，特别是现有方法仅依赖输出分布的广度（如熵或置信度）来衡量不确定性，而忽略了模型预测在注意力路径扰动下的稳定性。其核心解决方案是提出一种无需训练的不确定性估计方法——注意力子网络互信息（ASMI），通过掩码注意力头并测量由此产生的子网络间的BALD互信息，结合语义一致核以消除表面形式上的不一致干扰。关键创新在于：该信号并非简单重复传统置信度指标，而是捕捉“高置信但脆弱”的预测模式——即在上下文依赖任务中，当答案由外部提供信息路由决定时，此类预测表现出显著更高的错误可预测性，且在置信度过滤器中应用该信号可使保留误差降低约一半。此外，ASMI具备自适应适用范围判断能力，其有效性随任务模式变化而呈现梯度特性，仅在基于上下文的问答（grounded QA）中表现优异，在参数化知识召回任务中退化至零成本的MSP基线水平，符合理论预期。与需要多轮随机采样的强基线相比，语义增强版ASMI（Sem-ASMI）仅需单次贪婪解码即可获取信号，效率更高，并在十二个基准设置中的十个上达到或超越语义熵性能；在全部十二个设置中，最优的自适应ASMI变体在八项中持平或领先最强基线，其中三项在配对检验中具有统计显著性。头级别分析进一步揭示，真正影响不确定性的并非注意力头本身的脆弱性，而是这种脆弱性是否与实际错误相关联。
链接: https://arxiv.org/abs/2608.11138
作者: Minsoo Kim,Sungyoung Ji,Kisung Moon,Ilyong Yoon
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 19 pages, Under review
Abstract:We propose that a model’s uncertainty about a token is reflected not only in the breadth of its output distribution but also in whether a confident prediction is \emphfragile under perturbation of its attention pathways. We instantiate this as ASMI (Attention-Subnetwork Mutual Information), a training-free estimator that masks attention heads and measures the BALD mutual information among the resulting subnetworks, with a semantic-agreement kernel to discount surface-form disagreement. The signal is not a restatement of output confidence: on grounded QA an out-of-fold test shows it adds error-predictive information beyond single-pass confidence and entropy, concentrated in \emphconfident-but-fragile predictions, where acting on it roughly halves the retained error of a confidence filter. The distinctness is regime-graded, so ASMI predicts its own domain of applicability, strong where answers are routed through provided context and bounded by design where they are recalled from parametric knowledge. Sem-ASMI reads the signal from a single greedy response, without the stochastic generations the strongest baselines require, and ties or beats Semantic Entropy on ten of the twelve grounded benchmark-backbone settings. Across the same twelve settings, the best ASMI variant, typically the adaptive one reusing the ten samples already drawn for the baselines, ties or leads the strongest baseline in eight, significantly in three under a paired test. On parametric QA all variants revert to or below the zero-cost MSP baseline, exactly as predicted, and the estimates are near-deterministic across reruns. A head-level analysis shows that what tracks this boundary is not the presence of head-level fragility but whether that fragility couples to errors.
[NLP-7] Actions Speak Louder than Words: Measuring Cross-Lingual Policy Retention in Tool-Using Agents
【速读】： 该论文旨在解决多语言环境下工具使用智能体（tool-using agent）在执行相同任务时，其行为路径（action policy）是否保持一致的问题。传统多语言评估仅关注最终答案的正确性，忽视了动作序列这一关键输出，而动作序列直接影响系统成本、延迟、失败模式，并构成可审计的行为唯一依据。研究的核心问题是：跨语言任务中，智能体的动作策略是否存在结构性差异？解决方案的关键在于将动作策略本身作为可测量对象，通过构建一个包含8个模型、6个并行基准和41种语言（共238万次回滚）的大规模评估框架，系统性地识别并消除五类干扰因素——包括短轨迹得分偏高、空轨迹完美匹配、无关轨迹偶然一致、模型自身可重复性上限以及同一语言重复提问结果不一致等。通过移除这些混淆因素后，发现不同语言间的行为差异具有结构性而非随机噪声特征：在贪婪解码下，四个前沿模型在标准化其自身可重复性后趋于收敛，各自保留71%-73%的动作策略一致性，且模型身份仅解释5.7%的方差。对于参数量低于约100亿的模型，该一致性失效，其性能排序主要受偶然基线影响。进一步发现，智能体普遍将非英语任务路由至英语处理，这种“英语枢纽”机制具有因果重要性，且在强制指令下仍无法被消除。最后，研究揭示了一个关键发现：单一正则表达式用于轨迹提取，而非模型本身，导致了多语言评估中的虚假性能提升——两个示例使某模型的测量准确率提高26倍，但其可读输出的真实准确率几乎未变，表明评估方法本身可能引入严重偏差。
链接: https://arxiv.org/abs/2608.11110
作者: Sourabrata Mukherjee,Kalika Bali,Sunayana Sitaram
机构: Microsoft Research India(微软研究院印度)
类目: Computation and Language (cs.CL)
备注: Accepted in COLM 26
Abstract:When a tool-using agent is given the same task in a different language, does it still take the same steps? Multilingual evaluation rarely asks: it compares final answers and discards the actions. Yet those actions are the product: they fix cost and latency, decide how the system fails, and are the only auditable part of its behaviour. We make the action policy the measured object across 8 models, 6 parallel benchmarks and 41 languages (2.38M rollouts). The naive measurement fails: five confounds sit between raw trace similarity and any defensible claim, each able to flip a conclusion. Short traces score higher, empty traces score perfectly, unrelated traces agree by chance over half the time, the gap is capped by each model’s reproducibility, and a model asked the same question twice in one language answers differently, leaving no baseline. We remove all five, and every correction makes the effect larger. Divergence proves structural, not sampling noise: it survives greedy decoding in every cell and stays flat as temperature rises, even as models grow less self-consistent. Normalised by their own reproducibility, four very different frontier models converge under greedy decoding, each keeping 71-73% of its action policy across languages, with model identity explaining only 5.7% of the variance. Below roughly 10B parameters it breaks down, and the ordering among smaller models is largely an artifact of a chance floor we measure by permutation rather than assume. Agents route non-English tasks through English; this pivot is causally load-bearing, confirmed by a pre-registered prediction across four models, and models will not abandon it when told to. Finally, a single trace-extraction regex, not the model, manufactured a multilingual failure: two worked examples raise one model’s measured accuracy twenty-sixfold while its accuracy on readable outputs barely moves.
[NLP-8] Multiclass Sentiment Analysis for Identifying Political Viewpoints
【速读】： 该论文旨在解决社交媒体上政治观点的多类别情感分析问题，即如何自动区分针对政治议题与人物的多种情感倾向。其核心挑战在于政治话语具有高度复杂性与语境依赖性，传统情感分析方法难以准确捕捉其细微差异。解决方案的关键在于设计并评估两种基于机器学习的方法：一种是基于XGBoost的集成学习模型，另一种是基于BERT的深度神经网络模型。通过在标注的政治类社交媒体文本数据集上进行训练与评估，研究发现两种模型在测试集上的F1-score分别为0.2835（XGBoost）和0.2806（BERT），表明当前任务仍面临较大挑战，但为后续多类别政治情感分析研究提供了基准参考。
链接: https://arxiv.org/abs/2608.11049
作者: Girma Yohannis Bade,Olga Kolesnikova,Jose Luis Oropeza,Grigori Sidorov
机构: Centro de Investigaciones en Computación (CIC), Instituto Politécnico Nacional (IPN); Miguel Othon de Mendizabal, Ciudad de México, 07320, México
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:The rapid growth of social media has created vast amounts of political discourse, which provides valuable opportunities to analyze public opinions and identify different political perspectives. Sentiment Analysis (SA) is a core task in Natural Language Processing (NLP) that allows the computational study of attitudes and opinions in textual data, and has become increasingly important for understanding political discourse. In this work, we investigate multiclass sentiment analysis of political view- points on social media, that is to automatically discriminate multiple sentiment classes over political issues and figures. To solve this task we design and evaluate two machine-learning approaches based on XGBoost and BERT. We train and evaluate the models on a labeled dataset of political social media posts using standard classification metrics. The experimental results show that the XGBoost model reaches an F1-score of 0.2835 and the BERT- based model reaches an F1-score of 0.2806 on the test set. These results demonstrate the challenge of classifying complex and contextualized political discourse sentiment and provide a baseline for future research in multiclass political sentiment analysis.
[NLP-9] ReRound: Reconstructive Rounding to Resolve Midpoint Ambiguity in Calibration-Free LLM Quantization
【速读】： 该论文旨在解决低比特权重量化（low-bit weight quantization）中普遍存在的“中点模糊性”（midpoint ambiguity）问题，即在标准的四舍五入到最近（round-to-nearest, RTN）量化方案下，当模型权重接近量化区间中心时，难以确定其应向哪个整数方向舍入，从而导致精度损失。为应对这一挑战，论文提出ReRound（重构式舍入）方法，其核心在于引入一个基于条件扩散模型（conditional diffusion model）的连续权重重建机制，生成高精度的连续权重作为引导信号，以明确中点附近权重的舍入方向。关键创新在于设计了一种容忍度度量（tolerance metric），用于区分靠近量化边界和靠近中点的权重：前者采用传统RTN，后者则利用扩散模型重建结果进行舍入。通过遍历不同容忍度参数，ReRound生成多个候选量化矩阵，并选择其反量化后主导奇异值最接近原始全精度权重的版本作为最终方案，从而自适应地确定最优容忍度。该方法无需校准（calibration-free）、完全离线执行，且在3比特与4比特量化下显著优于标准RTN及多数无校准方法，在小型大语言模型（small LLMs）上表现尤为突出，同时具备可扩展至非LLM模型的通用性。
链接: https://arxiv.org/abs/2608.11045
作者: He-Yen Hsieh,H. T. Kung
机构: 未知
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: 16 pages, 8 figures
Abstract:ReRound (Reconstructive Rounding) is a post-training quantization method that addresses the midpoint ambiguity inherent in standard round-to-nearest (RTN) schemes when quantizing weights near the centers of quantization intervals. Starting from a pretrained LLM, ReRound trains a conditional diffusion model to produce continuous reconstructions of low-bit weights for the LLM. These reconstructed weights act as a guidance signal to disambiguate the rounding direction of weights located close to interval midpoints. To integrate this reconstruction-guided rounding with conventional RTN, ReRound introduces a tolerance metric measuring how far the quantized weight (not the final quantized integer) is away from the midpoint: quantized weights within a tolerance region around midpoints are quantized using diffusion-based reconstructions, whereas weights closer to quantization boundaries are quantized with RTN. By sweeping the tolerance parameter, ReRound generates multiple candidate quantized integer weight matrices and selects the de-quantized weight matrix candidate whose leading singular values most closely match those of the original full-precision weights. This selected candidate determines the tolerance parameter ReRound uses. ReRound is particularly effective for smaller LLMs. Across a range of such models, it consistently outperforms standard RTN for 3-bit and 4-bit weight quantization. ReRound achieves superior accuracy compared to an extensive set of calibration-free methods, remains competitive with calibration-dependent approaches, and operates entirely offline, introducing no additional overhead during low-bit inference. The ReRound strategy represents a new approach for low-bit quantization. The method applies to AI models beyond LLMs. This paper focuses on its applications to small LLMs. Comments: 16 pages, 8 figures Subjects: Machine Learning (cs.LG); Computation and Language (cs.CL) Cite as: arXiv:2608.11045 [cs.LG] (or arXiv:2608.11045v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.11045 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: He-Yen Hsieh [view email] [v1] Tue, 11 Aug 2026 15:18:07 UTC (2,088 KB) Full-text links: Access Paper: View a PDF of the paper titled ReRound: Reconstructive Rounding to Resolve Midpoint Ambiguity in Calibration-Free LLM Quantization, by He-Yen Hsieh and 1 other authorsView PDFHTML (experimental)TeX Source view license Current browse context: cs.LG prev | next new | recent | 2026-08 Change to browse by: cs cs.CL References Citations NASA ADSGoogle Scholar Semantic Scholar export BibTeX citation Loading… BibTeX formatted citation loading… Data provided by: Bookmark checked="checked"class=“labs-tab-input”> Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) IArxiv recommender toggle IArxiv Recommender (What is IArxiv?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv’s community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?) mathjaxToggle(); We gratefully acknowledge support from our major funders, member institutions, , and all contributors. About Help Contact Subscribe Copyright Privacy Accessibility Operational Status (opens in new tab) Major funding support from
[NLP-10] EAMMix: Taxonomy Enrichment Augmentation and Minority-augmented Mixing Strategy for LLM -enhanced Weak-Supervised Hierarchical Text Classification
【速读】： 该论文旨在解决层次化文本分类（Hierarchical Text Classification, HTC）任务中面临的复杂标签层级结构与类别不平衡问题。现有基于大语言模型（Large Language Models, LLM）的方法在实际应用中受限于提示词过长及标签结构信息丢失等问题，导致性能下降。为此，本文提出一种基于LLM数据增强的弱监督框架：首先通过关键词生成与语料挖掘对标签层级进行语义增强，提升模型对标签的理解能力；随后利用LLM生成伪样本以缓解长尾分布问题，并引入高斯混合模型进行置信度驱动的重采样，从而优化生成数据的质量。该方案的关键在于通过语义增强与可信度筛选双重机制，有效提升LLM生成伪标签的可靠性，显著改善细粒度且存在类别不平衡数据集上的分类性能。
链接: https://arxiv.org/abs/2608.11044
作者: Jian Zhang,Zhuohao Yang,Songlin Lei,Bangli Liu,Ziwei Wang,Xufeng Weng,Gehan Amaratunga,Yu Lin,Hongwei Wang
机构: Zhejiang University (浙江大学); ZJU-UIUC Institute, Zhejiang University (浙江大学-伊利诺伊大学香槟分校联合学院); Shaoxing K3i Technology Co. Ltd (绍兴科三智能科技有限公司); State Key Laboratory of CADCG, Zhejiang University (国家重点实验室计算机辅助设计与图形学, 浙江大学)
类目: Computation and Language (cs.CL)
备注: Accepted by IEEE CSCWD 2026
Abstract:Hierarchical Text Classification (HTC), as a critical text mining task, faces challenges such as complex label hierarchies and class imbalance. Existing methods based on large language models (LLMs) struggle to be efficiently applied to this task due to issues like lengthy prompts and loss of label structural information. To address these limitations, this paper proposes a weakly supervised HTC framework enhanced by LLM-based data augmentation. The framework first enriches the label hierarchy semantically through keyword generation and corpus mining, thereby enhancing the model’s understanding of labels. Subsequently, it guides the LLM to generate pseudo-samples to mitigate the long-tail problem, and employs a Gaussian mixture model for confidence-based resampling to optimize the quality of generated data. Experimental results demonstrate that the proposed method effectively improves the reliability of LLM-generated pseudo-labels and significantly enhances classification performance on fine-grained and imbalanced datasets.
[NLP-11] myMediWhisper: Construction of Burmese Medical Speech Corpus and Whisper Fine-Tuning for Clinical Dialogue ASR
【速读】： 该论文旨在解决生成式语音识别模型（如Whisper）在缅甸语医学语音识别任务中表现受限的问题，尤其针对缅甸语这一低资源语言在医疗场景下的语音识别精度不足。其核心解决方案在于构建一个高质量的28小时缅甸语医学语音数据集，并基于此对Whisper模型进行全量微调（Full Fine-Tuning, FFT）与参数高效微调（Parameter-Efficient Fine-Tuning, PEFT，采用LoRA方法）以提升模型在特定领域的适应性。关键创新点在于：尽管数据增强（包括波形级与频谱图级增强）会轻微降低在干净语音上的性能，但显著提升了模型在噪声和混响环境下的鲁棒性；此外，最优系统myMediWhisper-Medium（未经数据增强）在测试集上实现了23.44%的词错误率（Word Error Rate, WER），超越了更大规模通用领域微调模型的表现，证明了高质量领域专用数据与精细化微调策略的有效性。
链接: https://arxiv.org/abs/2608.11036
作者: Ye Kyaw Thu,Ye Bhone Lin,Thura Aung,Htet Arkar,Myat Oo Swe,Thet Htet San,Min Thiha Tun,Thazin Myint Oo,Thepchai Supnithi
机构: National Electronics and Computer Technology Center (NECTEC), Thailand; Language Understanding Laboratory, Myanmar; King Mongkut’s University of Technology Thonburi, Thailand; King Mongkut’s Institute of Technology Ladkrabang, Thailand
类目: Computation and Language (cs.CL)
备注:
Abstract:Although Whisper models benefit from large-scale multilingual pre-training, their performance on Burmese medical speech remains limited. This work presents a Burmese medical speech recognition framework built on a high-quality 28-hour corpus recorded and validated by native speakers. We fine-tune Whisper models using full fine-tuning (FFT) and parameter-efficient fine-tuning (PEFT) with LoRA. To evaluate robustness, we apply waveform- and spectrogram-level data augmentation under controlled noise and simulated room acoustics. While augmentation reduces performance on clean speech, it significantly improves robustness in noisy and reverberant environments across FFT and PEFT settings. Our best-performing system, fully fine-tuned myMediWhisper-Medium without augmentation, achieves a state-of-the-art Word Error Rate (WER) of 23.44%, outperforming much larger general-domain fine-tuned models. Dataset and other resources can be found at the Huggingface repository: this https URL.
[NLP-12] Mapping and Measuring the Behavioral Evolution of Large Language Models
【速读】： 该论文旨在解决现有基准排行榜无法揭示语言模型行为在不同模型间关系及其随代际演进变化的问题。其核心挑战在于如何量化和可视化多个语言模型在相同输入下的输出行为差异，从而捕捉模型间的静态组织结构与动态演化趋势。解决方案的关键在于构建一套基于嵌入的、多层次的句子级不相似性度量体系：包括逐提示的对齐均值距离（一种观测模型输出上的伪度量）、基于主成分分析（PCA）压缩的提示级分歧摘要，以及无需对齐的格罗莫夫-沃瑟斯坦（Gromov–Wasserstein）差异，用于刻画模型内部响应几何结构的异同。通过这些度量，研究者构建了行为图谱（behavioral maps），揭示了模型家族在静态上形成清晰聚类（如gpt-2为全局异常点），跨家族距离随时间推移而减小，且近期以推理为导向的模型具有更紧凑的响应云分布等规律。此外，基于词元级最大均值差异（Maximum Mean Discrepancy, MMD）的交叉验证结果与句子级平均距离高度一致（Spearman ρ=0.98），进一步支持了结论的稳健性。研究还从测度论视角明确建模中的对齐与不变性假设，并提出一个架构无关的充分条件，将行为相似性与推理-提示覆盖范围、微小过剩群体对数损失及相似的有效目标分布相联系，为观察到的趋势提供了可能的训练层面解释。整个方法流程无需标签，且即使使用三个额外编码器将嵌入维度降至原尺寸的1/73，仍能保持排名几何结构、异常点识别及时间趋势符号不变，展现出极强的鲁棒性与可扩展性。
链接: https://arxiv.org/abs/2608.11027
作者: Dong Qiao,Chris Ding,Jicong Fan
机构: 未知
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:
Abstract:Benchmark leaderboards summarize how well a language model performs, but not how its behavior relates to that of other models or changes across generations. We characterize the output behavior of 32 models from six families using their responses to a shared bank of 10,000 prompts. After embedding each response, we construct three complementary sentence-level dissimilarities: an aligned mean per-prompt distance, which is a pseudometric on observed model responses; a PCA-compressed summary of prompt-wise disagreement; and an alignment-free Gromov–Wasserstein discrepancy between models’ internal response geometries. We use these constructions to study static organization and temporal change on a release-date axis through behavioral maps, family-wise drift, hierarchical clustering, cross-family convergence, and response-cloud dispersion. Across the three constructions, model families form coherent clusters, with \textttgpt-2 as a global outlier; cross-family distances decrease over time; and several recent reasoning-oriented models have comparatively compact response clouds. A token-level cross-check based on per-prompt Maximum Mean Discrepancy closely agrees with the sentence-level mean distance (Spearman \rho=0.98 ) and recovers the same qualitative findings. We organize these comparisons through a measure-theoretic lens making their alignment and invariance assumptions explicit. We also establish an architecture-agnostic sufficient condition linking behavioral similarity to inference-prompt coverage, small excess population log-loss, and similar effective target distributions—a possible training-side account rather than an empirical explanation of the observed trends. Our pipeline is label-free, and re-encoding every response with three further encoders—down to one 73\times smaller—preserves the rank geometry, the outliers, and the sign of the time trend.
[NLP-13] Data Attribution of Emergent Misalignment with Persona Features
【速读】： 该论文旨在解决生成式语言模型在特定任务微调后出现的涌现性错位（Emergent Misalignment, EM）问题，即模型在未涉及的领域中产生有害行为的现象。其核心关切在于揭示导致EM的内在机制，特别是这些有害行为所依赖的潜在表征（如“越狱人格”、“讽刺”、“欺骗”与“操控”等）的来源及其诱发条件。解决方案的关键在于通过基于稀疏自编码器（Sparse Autoencoder, SAE）的模型差异分析，识别出与错误行为相关的特定神经激活特征，并验证这些特征在模型中的可操控性。研究发现，尽管这些特征在大规模人类撰写预训练文本中存在语义相关性（如反派角色、支配欲、有害能动性等），但仅使用原始人类文本进行微调无法稳定诱发EM；而由相同内容生成的合成指令-响应对（synthetic instruction-response pairs）则能有效诱导EM并跨模型家族迁移。这表明，响应结构或模型生成的表述方式在触发EM中起关键作用，单纯语义相关性不足以引发错位，凸显了模型训练数据格式与生成范式在安全风险形成中的决定性影响。
链接: https://arxiv.org/abs/2608.11025
作者: Clemens Vetter,David Kaczér,Lucie Flek,Florian Mai
机构: Bonn-Aachen International Center for Information Technology, University of Bonn (波恩-亚琛国际信息科技中心，波恩大学); Lamarr Institute for Machine Learning and Artificial Intelligence (Lamarr机器学习与人工智能研究所)
类目: Computation and Language (cs.CL)
备注:
Abstract:Emergent misalignment (EM) is the phenomenon where fine-tuning a language model on a narrow task leads to harmful behavior in unrelated domains. A leading mechanistic account attributes EM to persona features: latent directions acquired during pre-training that misaligned fine-tuning amplifies. We ask where these features come from: which pre-training documents activate them, and whether naturally occurring human-written text suffices to induce EM. Using Sparse Autoencoder (SAE) based model diffing across four open-weight models, we find that features related to jailbreak personas, sarcasm, deception, and manipulation are amplified by misalignment fine-tuning, while safety-relevant and assistant-identity features are suppressed. Steering individual features controls EM in both directions: it induces misalignment rates of up to 62% in aligned models – exceeding the 35% reached by misalignment fine-tuning itself – and re-aligns misaligned models to near-baseline misalignment rates. Attributing the causal features to a corpus of one million pre-training web documents retrieves semantically relevant narratives about villainous characters, domination, and harmful agency. However, fine-tuning on these human-written documents does not reliably induce EM, even after reformatting into assistant-style responses, whereas synthetic instruction-response pairs derived from the same content do – and transfer across model families. Semantic relevance alone is therefore not sufficient: response structure or model-generated phrasing plays an important role in inducing EM.
[NLP-14] mplated or fully Synthetic? Prompt construction as a confound in measuring LLM political stance beyond writing assistance
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）政治立场检测中长期依赖封闭式、多项选择型政治调查题目的局限性，此类题目源于人类设计，缺乏真实人机交互中的自然性与细微差别，且易受“伪装应对”（sandbagging）干扰。尽管近期提出的IssueBench框架通过基于真实聊天记录的模板化提示缓解了部分问题，但其仍难以捕捉开放式任务中的语境复杂性，且模板本身具有可识别的评估痕迹。为此，论文提出采用完全由生成式人工智能（Generative AI）生成的合成提示，这些提示在详细指令指导下以真实提示为种子进行生成，从而提升提示的真实性与生态效度。研究通过小规模实验对比了真实提示、模板化提示和生成式提示在三个高度争议性政策议题及三个近期地缘政治冲突情境下的表现，结果表明：人类与生成式AI标注者均认为生成式提示在真实性上不逊于真实提示，显著优于模板化提示，且更准确传达预设立场意图；而生成式模型能更清晰地区分模板化提示与其他两类提示，显示出更强的敏感性。案例研究表明，在中立表述下，模板化提示会系统性夸大模型的政治倾向，而生成式提示则提供了更可靠、更具生态效度的立场评估结果。因此，解决方案的关键在于使用基于真实数据种子生成的全合成提示，以增强评估任务的真实感与测量准确性。
链接: https://arxiv.org/abs/2608.11008
作者: Ilias Chalkidis
机构: The National Center for AI in Society (CAISA), University of Copenhagen (丹麦哥本哈根大学人工智能与社会国家中心)
类目: Computation and Language (cs.CL); Computers and Society (cs.CY)
备注:
Abstract:Political stance detection in LLMs has long been dominated by closed-ended, multiple-choice political survey questions—originally designed for humans, and thus lacks the realism and nuance of human-AI interactions in the wild, while also being susceptible to sandbagging. The recent IssueBench framework substantially mitigates these limitations with templated prompts anchored in real-world chat logs. Given the rise in non-work-related use of GenAI assistants, we extend IssueBench beyond writing assistance to include two additional tasks, information seeking and opinion sharing. We argue that templated prompts still lack the nuance of real ones, especially for open-ended tasks, and remain recognisable as evaluation artefacts. We propose the use of fully synthetic (LLM-generated) prompts, produced under detailed instructions with real prompts as seeds. We assess the ecological validity of real, templated, and LLM-generated prompts in a small-scale study covering 3 highly contested policy issues and 3 recent geopolitical conflicts. Human and LLM annotators rank LLM-generated prompts as no less realistic than real ones and clearly more realistic than templated ones, and find that they carry their intended intent and stance more clearly; the LLMs separate templated prompts from the other two far more sharply than the humans do. In a case study, templated and LLM-generated prompts yield systematically different stance estimates for the same model, most visibly under neutral framings, where templated prompts overstate the model’s leaning in the direction encoded by the topic-and-stance text (filler) slotted into their templates.
[NLP-15] On the Limitations of Cross-Lingual Consistency in Multilingual Text-to-image Generation ACM-MM2026
【速读】： 该论文旨在解决文本到图像（Text-to-Image, T2I）生成模型在多语言场景下存在的跨语言性能差异与语言特异性效应研究不足的问题。现有研究主要局限于英语单语环境，缺乏对不同语言间生成一致性与文化语境影响的系统性评估。为此，本文提出LingT2I基准，涵盖10种广泛使用的语言，包含3.3万条跨语言提示，用于评估内容生成与文本渲染两个维度上的跨语言表现。其解决方案的关键在于构建一个全面、多样化的跨语言评估框架，并通过实证分析揭示语言不平等现象及多维度评价下的语言依赖性权衡。此外，研究进一步识别出多种语言相关的生成模式，阐明了语言特征及其背后文化语境对模型输出的系统性影响。该基准与分析为理解T2I生成中的跨语言行为提供了坚实基础，推动更鲁棒、更具包容性的多语言生成模型的发展。
链接: https://arxiv.org/abs/2608.11002
作者: Sicheng Zhang,Zhonghao Yan,Binzhu Xie,Shi Qiu,Muzammal Naseer,Naveed Akhtar,Mubarak Shah
机构: Khalifa University (哈利法大学); Queen Mary University of London (伦敦玛丽女王大学); The Chinese University of Hong Kong (香港中文大学); The University of Western Australia (西澳大利亚大学); The University of Melbourne (墨尔本大学); University of Central Florida (中佛罗里达大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Accepted to ACM MM 2026
Abstract:Text-to-image (T2I) generation has achieved remarkable progress in recent years. However, existing research has largely focused on English-only settings, leaving cross-lingual performance gaps and language-specific effects insufficiently explored. To fill this gap, we introduce LingT2I, a benchmark covering 10 widely used languages with 33K prompts, designed to evaluate cross-lingual effects in both content generation and text rendering. Building on this benchmark, we conduct a comprehensive cross-lingual analysis, uncovering linguistic inequality and language-dependent trade-offs across evaluation dimensions. Beyond quantitative evaluation, we further reveal a range of language-dependent generation patterns, highlighting how linguistic factors and their corresponding cultural contexts systematically impact model outputs. Our benchmark and analysis provide a foundation for studying cross-lingual behavior in T2I generation and facilitate the development of more robust and inclusive models. Code and dataset are available at this https URL.
[NLP-16] ConRub-Med: Reinforcement Learning with Consensus Rubrics for Open-Ended Medical Question Answering
【速读】： 该论文旨在解决开放性医学问答中缺乏低成本、可验证的评估机制的问题，尤其针对生成内容可能存在部分正确、信息缺失或临床关键错误等复杂情况。传统依赖专家人工标注的评分体系虽具临床可靠性，但难以规模化；而现有模型生成的评分标准往往缺乏一致性与语义严谨性。其解决方案的关键在于提出ConRub-Med框架，通过三类异构语言模型独立生成原子级评分标准，并由一个独立评审模型筛选出在语义上得到三者共同支持的标准，从而保障评分依据的可信度与多样性。该方法引入三态评分（Three-State scoring）机制，区分正确覆盖、信息缺失与错误陈述，对错误给予负向奖励而非零分，强化了对临床安全性的约束。此外，在组相对策略优化（GRPO）中设计了基于成对判别器的序列优势计算机制，仅当两个候选顺序一致时才提供优势信号，避免改变原始标量奖励，提升了训练稳定性。在盲法对照研究中，专家评估表明，该方法生成的响应面板在临床相关性上显著优于单一模型生成结果；在9个基准测试中，ConRub-Med在6个上排名第一，并取得最高的医学通用性平均得分。基于5,166个提示构建的评分数据集，其在HealthBench-Hard上的得分为38.98 ± 1.04，显著优于InfiMed-ORBIT（8,000样本：33.60；28,000样本：37.30），验证了该方案在提升医学生成质量方面的有效性与可扩展性。
链接: https://arxiv.org/abs/2608.10996
作者: Taojie Zhu,Yuan Xia,Tao Sun,Yizhi Wang,Yan Chen,Qunshan He,Tian Guan,Jian Wang,Jinjie Gu,Junwei Liu,Yonghong He
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Reinforcement learning with verifiable rewards has been especially effective in mathematics and coding, where answers can be checked automatically. Many open-ended medical questions lack comparably cheap outcome verifiers: responses may be partly correct, incomplete, or contain clinically consequential errors. Rubrics written or validated by physicians offer strong clinical grounding, but involving experts in every instance is costly. Model-generated rubrics make this supervision scalable. We introduce ConRub-Med to preserve useful distinctions as rubric feedback moves from construction to policy optimization. For each prompt, three heterogeneous language models propose atomic criteria independently; a separate model reviews them, retaining only criteria with semantic support from all three generators. Three-State scoring distinguishes correct coverage, missing information, and incorrect claims. Errors receive negative rather than zero credit. When every response in a complete Group Relative Policy Optimization (GRPO) group receives the same final reward, a pairwise judge provides sequence advantages only if both candidate orders agree, without changing the scalar rewards. Groups without ties use vanilla GRPO. In a blinded study matched by question, two medical experts rate panels from the full pipeline as more clinically relevant than panels produced by one generator. Across the evaluated open models, ConRub-Med ranks first on six of nine benchmarks and achieves the highest medical and generalization averages. Using the resulting rubric dataset of 5,166 prompts, it scores 38.98 \pm 1.04 (mean \pm SD) on HealthBench-Hard, compared with InfiMed-ORBIT’s 33.60 with 8,000 samples and 37.30 with 28,000.
[NLP-17] What Iterated Self-Feeding Probes of Language Models Measure and a test that separates the construction from the model
【速读】： 该论文旨在解决生成式语言模型在自洽性探针（self-consistency probe）等方法中所测量的本质问题：即此类探针究竟反映的是模型本身的内在特性，还是仅由探针构造方式本身决定的伪效应。其核心解决方案在于设计一种基于令牌序列的环形结构（ring of token cells），通过模型自身窗口化条件概率 $ p_r(x_i | x_i \pm r) $ 对令牌进行原位重采样，构建出具有明确动力学特性的Glauber型演化系统，并引入最大耦合（maximal coupling）与非最大耦合的对比机制。关键创新在于，通过固定随机数源并比较两个仅在一个令牌上不同的环之间的演化差异，使“损伤传播”（damage spreading）现象可被精确量化——在此设定下，未受损副本的差异恒为零，从而将真实模型敏感性从构造依赖性中分离出来。研究发现，部分量（如损伤光锥的运动学性质、令牌空间李雅普诺夫指数 $ \lambda_{ca}® $ 的尺度标度行为）在19个不同模型及跨越70倍规模层级的实验中保持不变，属于构造决定项；而另一些量（如 $ \lambda_{ca} $ 穿越零点的训练阶段位置、吸引子占比对模型排序的一致性）则真实反映模型训练状态。作者强调，若不区分这两类量，极易误将探针本身的数学构造当作模型本质属性，甚至曾因此错误报告了一个三位小数精度的“相变”现象，实则源于探针而非模型。为此提出判别准则：固定构造变模型，或固定模型变构造，观察读数是否随之变化。该方法论以完整工具包形式提供，经验证可复现Domany-Kinzel损伤场的比特级结果，并成功识别出四个因误判导致的撤稿结论，均源于看似可观测但实为构造依赖的量。
链接: https://arxiv.org/abs/2608.10986
作者: Nicolás Vera Zúñiga
机构: Independent Researcher(独立研究员); Chile(智利)
类目: Computation and Language (cs.CL)
备注: 16 pages, 4 figures. Code, per-run results, and the findings ledger: this https URL (archived: this https URL )
Abstract:A growing class of methods probes a language model by feeding it its own output: self-consistency, iterated refinement, agentic loops. We ask what such a probe measures, in a construction chosen to make the question sharp: a ring of token cells resampled in place by the model’s own windowed conditional p_r(x_i | x_i±r). The substrate is Glauber dynamics on token sequences and is not new; what we change is the coupling. Advancing two rings that differ in one token under common random numbers makes undamaged copies diverge by exactly zero, so damage spreading becomes measurable where a maximal coupling gives mixing times instead. The answer is that it measures two different things at once, in readings that look alike. Some quantities are fixed by the construction: the damage light cone is kinematic, and the radius scaling of the token-space Lyapunov exponent lambda_ca® is model-invariant across 19 models and two scale ladders spanning 70x. Others genuinely track the model: lambda_ca crosses zero at a reproducible point in training, and the attractor share ranks models consistently however the lattice is built. Left undistinguished, the first kind is readily mistaken for the second – we did so ourselves for four months, and report a phase transition we measured to three decimal places that belongs to the probe rather than to any language model. We give the test that separates them: hold the construction fixed and vary the model, or hold the model fixed and vary the construction, and see which readings move. We validate the instrument by reproduction first, recovering a Domany-Kinzel damage field bit-exactly against an independent prediction, and we report the estimator failures that this discipline caught – four retracted verdicts, each on a quantity that looked like a measurement. The methodology ships as a package.
[NLP-18] MUSE: A Full-Text Cross-Domain Knowledge Base of Scientific Problems Solutions and Rationales
【速读】： 该论文旨在解决科学文献中细粒度问题求解过程难以被系统化提取与利用的问题，具体表现为科研人员在论文中描述技术障碍、所采用的解决方案及其选择依据（即理由）时，这些信息往往分散且未结构化。为应对这一挑战，论文提出MUSE（Mining Underlying Scientific Explanations），一个覆盖多领域的全文本科学问题-解决方案-理由（Problem-Solution-Rationale, P-S-R）三元组资源。其核心解决方案在于构建一套包含显著问题、解决方案和理由片段的丰富标注体系，并引入概念性共指（conceptual coreference）与“解决”及“理由”关系链接，通过模块化抽取流程将专家标注数据扩展至37,000个基于原文证据的高质量P-S-R三元组。关键创新点在于：首次系统性地从全文本中挖掘并结构化科学推理链条，同时验证了基于理由监督（rationale-supervised）的大语言模型（LLM）在科学问题求解中的潜力——研究发现，理由监督能有效提升复杂多约束问题的求解性能，但在简单问题上反而可能产生负面影响，揭示了监督信号与任务复杂度之间的非线性关系。
链接: https://arxiv.org/abs/2608.10974
作者: Tsofia Cohen,Tom Hope
机构: The Hebrew University of Jerusalem (耶路撒冷希伯来大学); Allen Institute for AI (Ai2)
类目: Computation and Language (cs.CL)
备注:
Abstract:Scientific papers contain fine-grained records of problem solving: authors mention technical obstacles and methods that were used to address them, often along with reasoning on why those methods were chosen. We introduce MUSE (Mining Underlying Scientific Explanations), a full-text, multi-domain resource of scientific Problem-Solution-Rationale (P-S-R) triplets. We curate 579 expert-annotated full-text paragraphs, with a rich annotation schema covering salient problem, solution, and rationale spans, solves and rationale_of links and conceptual coreference. A modular extraction pipeline scales this annotation to build a high-quality knowledge base of 37K source-grounded P-S-R triplets. We evaluate the extraction components and include a preliminary experiment training a rationale-supervised LLM for scientific problem solving. Interestingly, we find that rationale supervision improves performance on complex, multi-constraint problems but can harm performance on simpler ones.
[NLP-19] ReLTEx: Reliable LLM -based Taxonomy Expansion
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在术语体系（taxonomy）扩展过程中产生的噪声、冗余及层级不一致问题，这些问题源于直接依赖LLM生成内容时易出现的幻觉（hallucination）现象，从而限制了自动化术语体系扩展的可靠性。其解决方案的关键在于提出ReLTEx框架，通过结合基于LLM的候选概念与关系生成、结构感知的验证机制以及递归扩展控制策略，有效提升生成术语体系的一致性与质量。该框架通过多轮验证与动态控制机制，抑制了不合理的扩展，确保生成结果在语义连贯性和层级结构上更符合实际知识体系，实验结果表明其在基准测试中显著优于多种对比方法，且经人工评估验证具备更高的可靠性和可读性。
链接: https://arxiv.org/abs/2608.10970
作者: Zeinab Ghamlouch,Mehwish Alam
机构: Télécom Paris, Institut Polytechnique de Paris(巴黎电信学院，巴黎综合理工学院); Télécom Paris, Institut Polytechnique de Paris(巴黎电信学院，巴黎综合理工学院)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Recent advances in Large Language Models (LLMs) have demonstrated strong capabilities in generating semantically relevant concepts and relations, making them promising tools for taxonomy enrichment. However, directly relying on LLM-generated expansions often leads to noisy, redundant, or hierarchically inconsistent structures, limiting their reliability for automated taxonomy expansion. In this paper, we present ReLTEx, a framework for reliable LLM-based taxonomy expansion. ReLTEx combines LLM-driven candidate generation with structure-aware validation and recursive expansion control to improve the consistency and quality of generated taxonomies by reducing hallucinations. We evaluate the proposed framework using benchmark taxonomies under a masked taxonomy expansion setting and compare multiple validation strategies. Experimental results, supported by both adapted evaluation metrics and human evaluation, demonstrate that ReLTEx produces more reliable and semantically coherent taxonomy expansions.
[NLP-20] REAP: Relation-Aware Elicitation and Parsing for Closed-Book Knowledge Base Construction from LLM s
【速读】： 该论文旨在解决在封闭式问答（closed-book）设置下，仅依赖语言模型的参数化知识构建知识库的问题，且受限于不超过320亿参数的预算以及不允许进行模型微调。其核心挑战在于如何高效、准确地从大语言模型中提取结构化知识，同时避免引入外部知识或对模型进行额外训练。解决方案的关键在于提出一种融合结构化思维链推理（structured chain-of-thought reasoning）、针对关系特异的查询策略（relation-specific query strategies），以及基于推理的空集门控机制（reasoning-based empty-set gate），以主动激发模型内部隐含的知识，并直接将结果提取为有效的JSON数组格式。该方法显著提升了知识抽取的准确性与鲁棒性，在多个关键关系类型上取得了优异表现，如countryLandBordersCountry（F1=0.95）、companyTradesAtStockExchange（F1=0.73）和hasArea（F1=0.77）。
链接: https://arxiv.org/abs/2608.10963
作者: Thanh-Dan Bui,Thanh-Trung Do,Tuan-Phong Nguyen
机构: VNU University of Engineering and Technology, Hanoi, Vietnam(越南国立大学工程技术大学，河内，越南)
类目: Computation and Language (cs.CL)
备注:
Abstract:We present the REAP system for the AKBC Shared Task 2026 on constructing knowledge bases from language models in a closed-book setting, subject to a budget of at most 32B parameters and no model fine-tuning. Our system combines structured chain-of-thought reasoning, relation-specific query strategies, and a reasoning-based empty-set gate to elicit parametric knowledge, followed by direct extraction into valid JSON arrays. On the test set, the system, built on the Mistral-Small-24B-Instruct-2501 model, achieves a macro-F1 score of 0.62, with particularly strong results on countryLandBordersCountry (F1 = 0.95), companyTradesAtStockExchange (F1 = 0.73), and hasArea (F1 = 0.77). Our code is publicly available at this https URL.
[NLP-21] StreamFlow: Dynamic Memory Flows for Streaming Video Understanding
【速读】： 该论文旨在解决流式视频理解中多模态大语言模型（MLLM）在严格因果约束与有限记忆容量下，难以有效保留连续视频流中的相关证据这一核心问题。现有方法存在两大局限：基于模型的方案需侵入式更新主干网络，而基于记忆的方案则在时间冗余内容上消耗大量视觉编码计算，并依赖对历史视觉信息的固定访问机制。本文提出StreamFlow，一种高效的视觉记忆框架，其关键在于通过轻量级、动态感知的中期记忆，在视觉编码前主动过滤时间冗余信息；同时结合将历史视频内容压缩为可被后续推理访问的视觉潜在表示的潜在长期记忆。生成阶段采用注意力引导的检索机制，在模型对视觉证据依赖减弱时动态注入相关视觉潜在表示。该设计显著提升了视觉信息利用效率，使StreamFlow在StreamingBench上达到67.73%的整体准确率，相较基线提升视觉注意力得分（VAS）59.1%，并分别降低端到端延迟和峰值内存50.4%与21.1%，实现了更视觉化且高效的推理。
链接: https://arxiv.org/abs/2608.10949
作者: Muxin Fu,Yifan Zhang,Wentao Zhang,Fangming Guo,Qian Chen,Guibin Zhang,Shuicheng Yan,Bo An
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL)
备注:
Abstract:Streaming video understanding requires multimodal large language models (MLLMs) to preserve relevant evidence from continuously evolving streams under strict causality and bounded memory. Yet existing paradigms remain limited: model-based methods require intrusive backbone updates, while memory-based methods expend substantial visual-encoding computation on temporally redundant content and rely on rigid access to visual history. To address these limitations, we introduce StreamFlow, an efficient visual memory framework that enables dynamic, on-demand access to historical visual information. StreamFlow combines a lightweight, dynamics-aware mid-term memory that filters temporal redundancy before visual encoding with a latent long-term memory that consolidates historical video content into visual latents accessible to subsequent reasoning. During generation, an attention-guided retrieval mechanism injects relevant visual latents when the model’s reliance on visual evidence weakens. StreamFlow achieves state-of-the-art streaming video understanding performance, reaching 67.73% overall accuracy on StreamingBench, while also delivering strong performance on offline long-video benchmarks. Relative to the vanilla setting, it improves the visual attention score (VAS) by 59.1% while reducing end-to-end latency and peak memory by 50.4% and 21.1%, respectively, enabling more visually grounded and efficient reasoning.
[NLP-22] A Cost-Efficient Routing Pipeline for Multilingual Short-Text Classification Using Small Language Models
【速读】： 该论文旨在解决多语言短文本分类中高资源语言与低资源语言之间性能差异显著的问题，尤其是在统一推理策略下，低资源语言因缺乏足够训练数据而表现不佳。其核心挑战在于如何在不依赖任务特定微调的前提下，提升低资源语言的分类效果，同时保持系统的可部署性与效率。解决方案的关键在于提出一种固定列表路由（fixed-list routing）策略：将语言按资源水平划分为不同层级，对高资源和中资源语言采用直接多语言分类路径，而对低资源语言则通过翻译至英语后再进行零样本分类（zero-shot classification），从而借助英语强模型的能力弥补低资源语言的不足。该方法基于预训练的小型句向量编码器，实现完全自托管，无需额外微调，且在两个基准数据集（SIB-200 和 MASSIVE）上的实验表明，选择性翻译显著提升了低资源语言的宏平均F1得分（Macro-F1），但最优路由边界取决于具体任务特性，因此作者主张以层级级的性能增益与延迟作为评估标准，而非单一全局效率指标。
链接: https://arxiv.org/abs/2608.10939
作者: Wajdi Ben Saad,Safa Madiouni
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Accepted for publication at the 16th International Conference on Advanced Computer Information Technologies (ACIT 2026), this https URL
Abstract:Multilingual short-text classification supports operational systems such as content moderation, customer support routing, and intent recognition, yet aggregate evaluation often hides large differences between high-resource and low-resource languages. Uniform inference policies are simple to deploy, but they assume that all languages are equally well served. In this work, we evaluate a fixed-list routing strategy that keeps stronger languages on a direct multilingual path and selectively sends weaker languages through translation into English before zero-shot classification. The pipeline is fully self-hosted, uses pretrained compact sentence encoders, and requires no task-specific fine-tuning. We test the approach on two benchmarks chosen to differ in scale and label granularity: a 15-language subset of SIB-200 for seven-way topic classification and a 15-locale subset of MASSIVE for intent classification over an official 60-intent inventory. On SIB-200, the best overall configuration is R1, which translates only the low-resource tier: high-tier and mid-tier Macro-F1 remain unchanged, while low-tier Macro-F1 rises from 0.4632 to 0.6828. On the MASSIVE subset, the same low-tier intervention raises low-tier Macro-F1 from 0.2143 to 0.4417, but the best overall result is obtained by full translation, R3, at Macro-F1 0.4647. Across these two benchmarks, selective translation is a reliable intervention for weaker languages, whereas the optimal routing boundary depends on the task. We therefore report routing through tier-level quality gains and tier-level latency rather than a single global efficiency score. Comments: Accepted for publication at the 16th International Conference on Advanced Computer Information Technologies (ACIT 2026), this https URL Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI) Cite as: arXiv:2608.10939 [cs.CL] (or arXiv:2608.10939v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2608.10939 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[NLP-23] FaithformBench: Benchmarking Faithfulness of Mathematical Chain-of-Thought Autoformalisation
【速读】： 该论文旨在解决生成式自动形式化（Autoformalisation, AF）系统在评估其忠实性（faithfulness）时存在的关键问题：现有评估方法依赖昂贵的人工标注真值数据，或依赖大语言模型（LLM）评判者及嵌入模型，但这些方法在准确性上缺乏可靠保证；同时，传统方法仅针对已知正确的输入进行评估，无法检验系统对错误输入的忠实处理能力。为此，本文提出一种低成本、基于弱假设下具有理论保障的新基准，能够同时评估正例（正确输入）与负例（错误输入）的忠实性。其核心解决方案是通过自动生成被故意扰动的推理步骤（设计为无效），并检测系统在原始正确步骤上的有效性保持能力以及在扰动后步骤上的无效性保持能力。实验应用该方法于四个数学数据集上的八种AF系统，发现普遍存在“奉承现象”（sycophancy），即多数系统会“静默修正”无效输入为可证明的形式化语句。值得注意的是，最擅长保持有效性的微调模型也表现出最强的奉承倾向，揭示出当前AF系统在有效性与无效性保持之间存在根本性权衡。
链接: https://arxiv.org/abs/2608.10916
作者: Rob Cornish,Iacopo Ghinassi,Po-Hung Yeh,Shuqi Liu,Qiyuan Xu,Haoxuan Yin,Dominik Wagner,Wenda Li,Yee Whye Teh,Luke Ong
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Logic in Computer Science (cs.LO)
备注:
Abstract:Autoformalisation (AF) systems map natural language reasoning steps into formal statements in a proof assistant such as Lean. We consider how to assess the faithfulness of these systems. Existing approaches require expensive human-annotated ground truth, or rely on LLM judges or embedding models, which come with limited guarantees of accuracy. In addition, these methods typically only consider inputs that are known to be correct, and therefore do not assess whether the AF translates incorrect inputs faithfully. To address these limitations, we propose a new benchmark for AF faithfulness that is cheap to apply, sound under weak assumptions, and assesses both positive and negative examples. Our method is based on automatically generating perturbed reasoning steps that are designed to be invalid, and then measuring validity preservation on unperturbed steps and invalidity preservation on perturbed steps. We apply our method to eight AF systems across four mathematical datasets, and observe pervasive sycophancy: many AFs “silently correct” invalid inputs into provable statements. The most validity-preserving fine-tuned AFs are also the most sycophantic, suggesting a tension between validity and invalidity preservation in current AF systems.
[NLP-24] Order Matters: LVLMs as Judges for Temporal Reasoning in Image Sequences
【速读】： 该论文旨在解决生成式多媒体内容在从静态图像生成向复杂、交错的视觉叙事演进过程中所面临的“判断危机”问题，即当前自动化评估系统在处理视觉叙事的时间连贯性与逻辑一致性时存在显著缺陷。其核心挑战在于：尽管人类能够自然地感知故事的时序与逻辑流，现有基于大视觉语言模型（Large Vision-Language Models, LVLMs）的评估框架却因架构固有偏差而难以有效捕捉序列连续性，尤其在进行时间顺序的成对判别时表现急剧下降。研究的关键发现是，这种性能衰减并非单纯由数据稀缺导致，而是源于模型内部存在的系统性位置不对称性，如首因效应（primacy effect）与近因效应（recency effect），即模型对画面帧的判断更受其在序列中位置的影响，而非语义一致性。这一现象可能根植于因果掩码（causal masking）与旋转编码（rotary embeddings）等机制，揭示出当前基于Transformer的判别模型在长序列视觉推理任务中本质上的不适应性。因此，论文提出的核心解决方案是推动评价范式从以快照为中心的指标转向时序感知评估（Temporally-Aware Evaluation），将视觉序列视为统一的逻辑整体，而非无序帧集合，从而构建更符合人类认知规律的多模态叙事质量评估体系。
链接: https://arxiv.org/abs/2608.10908
作者: Martina Ianaro,Guilherme Fernandes,Maurizio Gabbrielli,Joao Magalhaes
机构: University of Bologna(博洛尼亚大学); NOVA School of Science and Technology(里斯本新大学科学技术学院); NOVA Laboratory for Computer Science and Informatics(里斯本新大学计算机科学与信息学实验室)
类目: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL)
备注: 34 pages, camera-ready
Abstract:As generative multimedia evolves from static image synthesis to complex, interleaved visual narratives, a foundational bottleneck has emerged: the judgment crisis. While human perception naturally synthesizes the temporal and logical flow of a story, automated evaluation systems remain largely “blind” to sequential continuity, often failing to distinguish between a coherent narrative and a semantically shuffled or contradictory sequence. This work identifies a critical structural gap in current multimodal evaluation paradigms, arguing that the reliance on Large Vision-Language Models (LVLMs) as judges is fundamentally limited by architectural biases. Our analysis reveals a profound performance dichotomy: while models may appear competent in isolated pointwise scoring, they suffer a catastrophic collapse when required to perform pairwise discrimination of temporal order. We demonstrate that this is not merely a data-scarcity issue but a structural one. Through a series of diagnostic probes, we uncover systematic positional asymmetries, specifically primacy and recency effects, where a model’s judgment of a story is significantly influenced by the placement of a frame, often more than by its semantic consistency. These biases, potentially rooted in causal masking and rotary embeddings, suggest that current transformer-based judges are inherently ill-equipped for long-form visual reasoning. By exposing these blind spots, we challenge the multimedia community to move beyond snapshot-centric metrics and instead pioneer Temporally-Aware Evaluation paradigms that treat visual sequences as unified logical structures rather than unordered collections of frames.
[NLP-25] Certify or Refuse: A Cross-Model Map for Selective Risk Control with Coverag e Floors under Covariate Shift
【速读】： 该论文旨在解决在有界比率协变量偏移（bounded-ratio covariate shift）下，如何实现对选择性预测器（selective predictors）的可认证覆盖保证（certified coverage guarantee）问题。核心挑战在于：在目标域中无法获取标签的情况下，如何在不依赖未知权重的前提下，为预测器的输出覆盖率设定一个可验证的最低标准（即“自动化下限”——automation floor），同时控制错误率不超过给定阈值。其解决方案的关键在于提出地板认证映射（Floor Certification Map），该映射将认证问题分解为两个独立资源的组合：一是源域中标注数据上的风险（risk in labeled source），二是目标域中未标注样本上的下限容量（floor in unlabeled target samples）。这一映射具有局部性特征，需满足局部前沿边界、松弛裕度（slack below local-regime threshold）及格点条件（lattice conditions），并基于预注册的格点边界与兼容的松弛机制构建上下界。研究通过三个模型结果揭示了复杂性结构：模型-B给出下界，模型-A通过“预言权重”（oracle weights）达到匹配上界，而模型-B’则提供可在预注册分层偏移模型下实施的上界，其中干扰项成本被显式建模。值得注意的是，这种匹配是跨模型而非单一模型的极小极大最优，因在整个有界比率类别中，任何未知权重方法在任意样本量下均无法一致地达到最优（如模型-B在 α=β=1/2 时表现为不一致）。此外，复杂性度量聚焦于局部接受区域的功能性测度，而非全局有效样本量（ESS），尽管固定ESS分离定理仍待证明；当 β→0 时，下界轴趋于零，表明“地板”机制构成了整个认证映射的基础。实证结果显示，预注册的咬合族（bite family）在对数-对数尺度下的斜率为 -2.002，符合预期；1,024个单元的审计记录中无违规事件发生，且单语料库（SQuAD-to-NewsQA）可行性审计显示诚实拒绝行为，验证了方法的有效性与实用性。
链接: https://arxiv.org/abs/2608.10893
作者: Jiamiao Liu,Dewen Qiao,Yu Zhang,Xuetao Chen
机构: Army Medical University (Third Military Medical University) (陆军军医大学（第三军医大学）)
类目: Computation and Language (cs.CL)
备注:
Abstract:Certified selective predictors attain whatever coverage they attain; operators impose an automation floor: answer at least a \beta -fraction of shifted target traffic with at most an \alpha -fraction of answers wrong. Under bounded-ratio covariate shift we prove the Floor Certification Map: once that floor must be certified alongside the selection-conditioned risk \alpha , certification acquires a feasibility frontier and a two-resource complexity map, additive up to constants: risk in labeled source, the floor in unlabeled target samples. The rates are local, needing a regular frontier margin, slack below the local-regime threshold, and lattice conditions: pre-registered with a lattice margin for the upper bounds, compatible per-slack for the lower. The displayed split is the operational route; oracle weights also allow a labeled-source floor estimate. Three model-tagged results: a lower bound (Model-B), a matching oracle-weight upper bound (Model-A), and an implementable upper bound (Model-B’) valid under a pre-registered exact stratified-shift model with nuisance cost priced explicitly. The match is across these models rather than a single-model minimax theorem, and necessarily so: over the full bounded-ratio class no unknown-weight procedure matches at any sample size (Model-B is inconsistent, witnessed at \alpha=\beta=1/2 ). The nuisance’s necessity is only partially settled. Complexity tracks a localized accepted-region functional, not global effective sample size (ESS), on both sides, though a fixed-ESS separation theorem is left open; both lower-bound axes vanish as \beta\to0 , so the floor creates the map. Empirically, the registered bite family diverges with log-log slope -2.002 within its pre-registered band; a 1,024-cell audit records 0 violations where the formal certificates fire; and a single-corpus SQuAD-to-NewsQA feasibility audit returns honest refusal.
[NLP-26] X2-Turn: Frame-Synchronous Dual-Head Modeling for Joint Streaming ASR and Turn State Prediction
【速读】： 该论文旨在解决语音对话系统中实时话轮转换（turn-taking）判定的准确性与响应性问题，尤其针对用户打断、回音信号（backchannel）识别以及话语完成状态判断等场景。传统模块化方法通常在话语或固定片段级别进行话轮状态预测，导致与连续的话轮状态估计存在不匹配，并且依赖辅助自动语音识别（ASR）模型，从而影响系统响应速度并增加复杂度。其解决方案的关键在于提出一种基于延迟流建模的帧同步话轮状态预测方法——X2-Turn。该方法在预训练的Voxtral Realtime模型基础上，引入一个与ASR头并行的帧级话轮状态头，共享流式表示，在帧级别联合预测ASR词元和细粒度话轮状态，实现了高精度且低延迟的话轮转换检测，显著提升了系统的实时性与准确性。
链接: https://arxiv.org/abs/2608.10878
作者: Kaiqi Fu,Rime Wen,Altman Lin,Shawn Qin,Roy Gan,Hao Wang,Qian Wang
机构: 未知
类目: Computation and Language (cs.CL); Audio and Speech Processing (eess.AS)
备注:
Abstract:Accurate and responsive turn-taking is essential for spoken dialogue systems, which must distinguish in real time between user interruptions, backchannels that should be ignored, and the completion of an utterance. Prior modular approaches typically optimize turn state prediction at the utterance or fixed-chunk level, creating a mismatch with the continuous turn state estimate, and often depend on an auxiliary ASR model, which limits responsiveness and increases overall system complexity. Therefore, we present X2-Turn, a frame-synchronous turn state prediction method via delayed-stream modeling. Specifically, building on the pretrained Voxtral Realtime model, we introduce a frame-synchronous turn state head that operates in parallel with the ASR head on shared streaming representations, jointly predicting ASR tokens and fine-grained turn states at the frame level. We evaluate our method on the bilingual Chinese-English Easy-Turn test sets, and the results demonstrate its effectiveness in achieving accurate turn-taking detection while maintaining low latency.
[NLP-27] VibeLifeBench: Can Your Life Agent Be Proactive and Persistent in a Living World?
【速读】： 该论文旨在解决当前大型语言模型（LLM）代理在真实日常生活场景中长期任务执行能力评估缺失的问题。现有评估多集中于短时、静态环境下的孤立请求，无法反映现实生活中任务持续数周、环境动态变化、约束隐含且需自主决策等复杂特性。为此，论文提出VibeLifeBench基准，涵盖200个跨10个日常生活的长周期任务，构建了一个包含22个模拟服务的动态仿真世界，世界以自身时钟推进，多数变化无声发生，仅通过主动重检才能发现。其核心解决方案在于设计了一套细粒度、加权的任务评估机制，仅依据代理实际留下的行为痕迹进行评分，涵盖最终状态、行动及时性及对隐式约束的遵守情况。实验评估七种前沿模型均表现不佳，表明当前代理距离真正具备主动性与持续一致性的长期生活辅助能力仍有显著差距。
链接: https://arxiv.org/abs/2608.10875
作者: Xiaohongshu Inc
机构: Xiaohongshu Dots Studio (小红书点石工作室); Evolvent AI
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language model (LLM) agents are increasingly deployed as personal assistants. Existing evaluations, however, mostly use short, self-contained requests in static environments. Everyday life assistance is different. A task runs for weeks rather than minutes. The world keeps changing while the agent is not being prompted. Many constraints are never stated outright. An agent that merely answers the request in front of it will fail at such a task. What is needed instead is an agent that stays proactive and consistent. It decides on its own when to act, when to ask, and when to stay silent. It notices changes that nobody announced. It keeps one plan coherent from the first day to the last. No current benchmark measures this. We introduce VibeLifeBench, a benchmark of 200 long-horizon tasks across ten everyday-life domains. Each task is a scripted multi-week timeline in a simulated world of 22 mock services. The world advances on its own clock, and many of its changes are silent, so only an agent that re-inspects the world discovers them. Every task is graded by fine-grained, weighted checks that read only what the agent actually left behind, covering the end state, the timeliness of its actions, and whether it upheld the implicit constraints. We evaluate seven frontier models. All of them score low, which shows how far current agents are from assisting with real life. We will open-source all tasks, environments, and the evaluation framework.
[NLP-28] Reference-Free Post-Training of Open Large Language Models for Multilingual Machine Translation
【速读】： 该论文旨在解决多语言机器翻译中缺乏参考文本（reference-free）的后训练问题，尤其是在使用开放的大语言模型（Large Language Models, LLMs）时如何有效提升翻译质量。其核心挑战在于如何在无参考译文的情况下对模型进行高质量的优化，同时保持跨语言的一致性与泛化能力。解决方案的关键在于采用分组相对策略优化（Group Relative Policy Optimization, GRPO），通过融合两个无参考质量评估模型的评分并以语言识别作为门控机制，构建一个动态且语言感知的奖励函数，从而指导强化学习（Reinforcement Learning, RL）过程。随后，通过线性插值监督微调（Supervised Fine-Tuning, SFT）与强化学习的模型检查点，生成最终的MiLMMT-46-v1.0模型，显著提升了在46种语言上的翻译性能。实验表明，该方法优于多个近期开源基线模型，并在无参考条件下达到与谷歌翻译、Gemini 3 Pro及GPT-5等专有系统相当甚至更优的水平。此外，研究还发现基于在线策略蒸馏（on-policy distillation）的方法虽能接近但无法超越该强化学习结合检查点插值所达到的质量上限。
链接: https://arxiv.org/abs/2608.10812
作者: Chris Han,Pengzhi Gao,Pei Fu,Jian Luan
机构: Xiaomi Inc.(小米公司)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:We study reference-free post-training for multilingual machine translation with open large language models. Starting from the supervised-finetuned MiLMMT-46-v0.1 models, we apply Group Relative Policy Optimization (GRPO) with a reward that averages two reference-free quality estimation models and is gated by language identification. We then linearly interpolate the supervised fine-tuning (SFT) and reinforcement learning (RL) model checkpoints to obtain MiLMMT-46-v1.0. Across 46 languages, the resulting models consistently improve translation quality over their SFT counterparts, outperform strong recent open baselines, including Seed-X, HY-MT2, and TranslateGemma, and achieve leading reference-free scores against evaluated proprietary systems such as Google Translate, Gemini 3 Pro, and GPT-5. We further investigate on-policy distillation and find that it reaches, but does not surpass, the quality frontier achieved by RL with checkpoint interpolation. We release the models and code to facilitate future research.
[NLP-29] Surfacing the Unsaid: CUE-Bench for Affective Stance in Chinese Discourse
【速读】： 该论文旨在解决现有情感理解基准在处理话语中深层情感推理时的局限性，即当前主流基准多仅标注表面情感极性或最终情感类别，缺乏对显性表达、隐含情感、语用意图及细粒度情感之间交互关系的结构化刻画。这种缺失导致现有评估体系对情感意义被隐藏、弱化、反转或语用重构等复杂情境不敏感，难以揭示模型在深层次情感理解上的失败。其解决方案的关键在于提出CUE Bench——一个聚焦于情感立场（Affective Stance）的中文未言明情感基准，通过构建九种人类可解释的情感立场类型，整合显性与隐性情感极性的交互，并提供语用意图和细粒度情感的标注，实现对情感推理过程的结构化建模。实验表明，引入情感立场可使细粒度情感识别性能提升3.5个百分点，语用意图检测性能提升7.8个百分点，显著优于强基线模型。
链接: https://arxiv.org/abs/2608.10810
作者: Zhenyan Zheng,Yunyao Zhang,Junxi Sheng,Junqing Yu,Zikai Song
机构: Huazhong University of Science and Technology (华中科技大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Emotion understanding in discourse requires reasoning beyond surface sentiment because speakers often convey affect through indirect, implicit, polite, ironic, or deliberately mismatched expressions. Existing emotion benchmarks mainly annotate surface polarity or final emotion categories, while lacking a structured account of how explicit expression, implicit affect, pragmatic intent, and fine grained emotion interact. This limitation makes current evaluations insensitive to cases where affective meaning is concealed, weakened, inverted, or pragmatically reshaped, thereby obscuring model failures in deeper emotion understanding. To address this gap, we introduce CUE Bench, a Chinese Unsaid Emotion benchmark that centers on Affective Stance and covers diverse communicative scenarios. CUE Bench constructs nine human interpretable affective stances from explicit implicit polarity interaction and further provides intent and fine grained emotion annotations for structured affective inference. Experiments show that incorporating Affective Stance improves fine grained emotion recognition by 3.5 percentage points and pragmatic intent detection by 7.8 percentage points over strong baselines.
[NLP-30] Assessing Reliability of BERT-Based Models on Question Answering Tasks
【速读】： 该论文旨在解决大语言模型在实际应用中可靠性不足的问题，尤其关注基于Transformer架构的问答（Question Answering, QA）模型在面对内部不确定性与输入扰动时的响应稳定性。尽管当前主流模型如BERT及其变体（RoBERTa、ALBERT、DistilBERT）在准确性上表现优异，但其可靠性尚未得到充分评估。研究的关键解决方案在于通过两种机制量化模型可靠性：一是利用蒙特卡洛丢弃（Monte Carlo Dropout, MCD）模拟模型内部参数的随机性以检验预测一致性；二是通过句式改写（paraphrasing）引入输入层面的词汇扰动，考察答案稳定性。实验基于SQuAD和QuAC数据集进行，结果表明RoBERTa在不同扰动下表现出更强的鲁棒性，而ALBERT与DistilBERT则存在显著不一致现象。统计分析进一步验证了MCD在推理阶段不会破坏模型的正常动态，证明其作为可靠性度量的有效性。该研究强调，在评估QA模型时必须同时考量准确性和稳定性，以确保模型在真实场景中的可信与可靠部署。
链接: https://arxiv.org/abs/2608.10806
作者: Pooja Yadav,Priyanka Harjule,Basant Agarwal,Marko Robnik Šikonja
机构: Malaviya National Institute of Technology, Jaipur, Rajasthan, India; Central University of Rajasthan, Kishangarh, Rajasthan, India; University of Ljubljana, Faculty of Computer and Information Science, Ljubljana, Slovenia
类目: Computation and Language (cs.CL)
备注: Accepted for publication in the Journal of Experimental Theoretical Artificial Intelligence
Abstract:Reliability estimation of large language models is in many cases as crucial as their accuracy, as reliable models are more trustworthy, robust, and suitable for practical applications. Recent advancements in natural language processing (NLP), particularly those based on transformer architectures, have significantly accelerated progress across various NLP tasks. This study focuses on the reliability of transformer-based question answering (QA) models, specifically BERT models and its variants (RoBERTa, ALBERT, DistilBERT). These encoder-only pretrained transformers have demonstrated remarkable accuracy in QA tasks that can be treated as classification tasks. However, their reliability remains underexplored. This study evaluates the reliability of four BERT-based models by assessing response stability under two conditions: (1) internal model variations induced via Monte Carlo Dropout (MCD) and (2) input perturbations through paraphrasing. Using the SQuAD and QuAC datasets, we investigate how dropout rates affect prediction consistency and whether lexical changes impact answer stability. Our findings reveal that RoBERTa maintains higher reliability, whereas AlBERT and DistilBERT exhibit significant inconsistencies. Statistical analyses confirm that enabling MCD during prediction does not disrupt inference dynamics, validating its effectiveness as a reliability metric. These findings underscore the importance of evaluating both accuracy and stability in QA models to ensure stability in real-world applications.
[NLP-31] Mitigating Context Interference for Reliable and Efficient Search Agents
【速读】： 该论文旨在解决多轮搜索智能体（multi-turn search agents）在执行复杂任务时因上下文过长且冗杂而导致的上下文干扰（context interference）问题。具体而言，每轮检索返回的文档集合不可避免地引入无关信息，干扰大语言模型（LLM）的推理过程，从而降低搜索代理的可靠性与效率。其解决方案的关键在于：通过系统性分析发现，上下文干扰主要来源于最新一轮检索到的文档；基于此发现，提出一种基于知识蒸馏（distill-based）的上下文精炼器（context refiner），能够动态过滤和压缩无关内容，实现“先精炼上下文，再生成输出”的新范式。进一步实验表明，将上下文精炼机制融入强化学习（RL）训练流程中，可显著提升搜索代理的性能，验证了该方法的有效性与普适性。
链接: https://arxiv.org/abs/2608.10743
作者: Boyang Xue,Bin Wu,Shuofei Qiao,Sheng Wang,Rui Wang,Yiming Du,Hongru Wang,Jeff Z. Pan,Emine Yilmaz,Kam-Fai Wong,Aldo Lipani
机构: The Chinese University of Hong Kong; University College London; Zhejiang University; The University of Hong Kong; The University of Edinburgh; MoE Key Laboratory of High Confidence Software Technologies
类目: Computation and Language (cs.CL)
备注:
Abstract:Recent research empowers Large Language Models (LLMs) as multi-turn search agents to iteratively retrieve and generate outputs until complex tasks are solved. However, the contexts of multi-turn search agents are lengthy and complex. For example, the retrieved set of documents in each turn would inevitably introduce irrelevant information that distracts LLMs, referring to \textitcontext interference, potentially hindering the reliability and efficiency of search agents. Therefore, we conduct a systematic study on context interference in multi-turn search agents, focusing on investigating i) which parts of the context of search agents will contribute to the context interference, ii) how to refine the contexts of search agents to mitigate the interference, and iii) can incorporating context refinement into search agent training yield further improvements. We reveal that interference primarily arises from the latest retrieved documents. Based on the explored findings, we then introduce a distill-based context refiner to dynamically mitigate context interference for multi-turn search agents. Finally, we validate that incorporating context refinement into RL training pipelines of search agents can significantly enhance both reliability and efficiency. This study highlights the importance of mitigating context interference of search agents, inspiring a novel paradigm of ``refine context and then generate’’ for AI agents.
[NLP-32] Ex-Omni-2D: Expressive Omni-Modal Dialogue Models with Native Visual Presence
【速读】： 该论文旨在解决生成式对话模型在多模态交互中响应缺乏视觉具身性（visually embodied）的问题，即尽管模型能够理解多模态输入并生成语音回复，但其输出仍无法与动态的视觉表现（如虚拟化身视频）进行协调一致的同步。其核心解决方案是提出Ex-Omni-2D框架，通过引入结构化的视觉思维计划（Visual Thought Plan, VTP），联合建模场景、情感与动作信息，并据此生成包含文本、个性化语音及参考条件驱动的视频内容。关键创新在于构建一个共享的声学-时间接口：将多码本语音单元（multi-codebook speech units）作为统一表征，既可解码为自然语音，又可在时序上实时对齐至视频帧，从而实现语音与视频的协同生成。该设计使得模型可从异构的语音、对话和虚拟化身视频数据中联合学习，避免对大规模“查询-文本-语音-视频”四元组标注数据的依赖。此外，采用全序列视频生成器作为主教师模型，并通过知识蒸馏得到轻量级的分块因果式流式学生模型（Streaming Student），其前缀流机制（Prefix Streaming）能有效传递干净的潜在状态，显著降低连续生成块间的累积误差。最终，在四步推理下，该系统在4GPU配置上实现了400×720/720×400分辨率下的端到端实时因子（RTF）1.293，确立了高质量与高效率之间的实用平衡点。
链接: https://arxiv.org/abs/2608.10720
作者: Haoyu Zhang,Zhipeng Li,Xiaoying Tang,Tianshu Yu,Yiwen Guo
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Omni-modal dialogue models can understand multimodal inputs and synthesize spoken replies, yet their responses remain visually disembodied. We introduce \textbfEx-Omni-2D, an omni-modal dialogue framework that generates a coordinated response comprising text, personalized speech, and reference-conditioned video. Given a multimodal query, reference image, and reference audio, the model predicts a structured \textitVisual Thought Plan (VTP) describing scene, emotion, and motion, followed by response text and native multi-codebook speech units. These units form a shared acoustic-temporal interface: they are decoded into speech and aligned online with video frames. This interface enables the response and avatar pathways to be learned from heterogeneous speech, dialogue, and avatar-video data, avoiding the need for large-scale query–text–speech–video supervision. A full-sequence Video Generator serves as the primary Teacher. For efficient incremental generation, we further distill it into a few-step block-causal \emphStreaming Student whose Prefix Streaming mechanism carries a clean latent across consecutive chunks to reduce cumulative late-chunk degradation. With four-step inference, the complete four-GPU pipeline achieves an end-to-end RTF of 1.293 at 400\times720 / 720\times400 , providing a practical quality–efficiency operating point.
[NLP-33] DuplexWorld: Can voice agents help you get through the day?
【速读】： 该论文旨在解决现有语音代理评估基准在真实场景下评估能力不足的问题，特别是未能全面衡量语音代理在日常复杂对话中表现出的对话多样性与任务执行能力。当前主流评测多聚焦于基于数据库查询的工具调用任务，忽略了语音代理在实际应用中需处理的多样化、动态性对话情境，以及对超出数据检索范畴的任务支持能力。为此，本文提出DuplexWorld框架，构建了涵盖银行、保险、旅行、医疗、物流及路径规划六个关键领域共156个场景（累计350+小时对话）的综合性评估体系，覆盖11类不同类型的对话，系统性地考察代理在智能性、对话流畅性与语音自然度三个维度的表现。其核心解决方案在于设计多维、贴近真实应用场景的评测体系，并通过多角度分析揭示当前最优语音代理在任务成功率（Pass@1: 0.490）、对话轮次协调能力（turn-taking: 0.653）和语音质量（DNSMOS: 3.378）上仍存在显著提升空间，同时深入探究了不同场景下的性能差异与失败模式，尤其从“探索-利用”（explore v exploit）视角解析路径规划类对话中的行为策略缺陷，从而为语音代理的可靠性与智能化演进提供关键洞察。
链接: https://arxiv.org/abs/2608.10716
作者: Aryan Vijay Bhosale,Harshit Rajgarhia,Akhil Pothanapalli,Asif Shaik,Abhishek Mukherji,Dinesh Manocha
机构: Centific Global Solutions Inc.; University of Maryland
类目: ound (cs.SD); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Speech-to-speech (S2S) voice agents are increasingly being incorporated into enterprise for customer care and as daily companions for consumers owing to the ease of the conversational modality over text. However, existing benchmarks fail to holistically evaluate voice agents along axes that really matter and are shaped as tests of agentic tool calling against a database. We believe they fail to adequately account for the diversity of conversational dialogue that mundane activities introduce and further, never test how faithfully an agent can assist on tasks that move beyond database manipulation. To tackle this DuplexWorld introduces six worlds where voice agents are especially useful: banking, insurance, travel, healthcare and logistics, and Pathfinding. Agents are evaluated on eleven different types of conversations across 156 scenarios (350+ hours of conversation), each testing conversational and analytical capability to varying degrees. Through extensive evaluation comprising agentic, conversational and speech-naturalness metrics, we show that even the best voice agents leave substantial room for improvement on all 3 axes (Pass@1: 0.490, turn-taking: 0.653, DNSMOS: 3.378). We perform extensive analysis on agentic v conversational performance, world- and conversation type-wise performance, failure modes exploring the explore v exploit lens for Pathfinding conversations and voice agent reliability over all six worlds.
[NLP-34] Most biomedical publications show signs of LLM -assisted writing
【速读】： 该论文旨在解决当前学术出版物中大语言模型（Large Language Model, LLM）辅助写作的滥用问题，尤其关注如何准确估计LLM生成内容在学术文本中的实际使用比例。现有方法难以提供可靠、无偏的估计，导致政策制定缺乏数据支持。其解决方案的关键在于提出并验证一种基于词汇频率变化的新方法，通过分析文本中与LLM相关词汇的异常增长趋势，实现对LLM使用情况的无偏估算。该方法被应用于PubMed Central中开放获取的生物医学论文，结果显示至2025年底，89%的论文显示出显著的LLM相关词汇特征；且在讨论部分（Discussion）的使用率（68%）远高于方法部分（Methods）（32%），即便在方法部分，整体使用率仍超过50%。该研究为未来制定科学、合理的学术伦理规范与政策提供了关键实证依据。
链接: https://arxiv.org/abs/2608.10715
作者: Lena Holzwarth,Rita González-Márquez,Dmitry Kobak
机构: Hertie Institute for AI in Brain Health, University of Tübingen, Germany; Department of Mathematics, Computer Science, and Statistics, Ghent University, Belgium; VIB Center for AI and Computational Biology, VIB, Ghent, Belgium
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Digital Libraries (cs.DL); Social and Information Networks (cs.SI)
备注:
Abstract:Over the past several years, LLM-powered chatbots and agents have become widely used as a tool for academic writing. LLM-assisted writing can be valuable by removing language barriers but at the same time causes concerns about misconduct and fraud. To inform policy decisions, it is necessary to monitor the prevalence of LLM-altered texts in scholarly publications. Despite some recent progress in this direction, no existing method can produce reliable estimates. Here we suggest and validate a new unbiased approach to estimate LLM usage in a corpus of texts based on changing word frequencies. We apply our method to the full texts of open-access biomedical papers from Pubmed Central, and show that by the end of 2025, 89% of papers show excess of LLM-associated vocabulary. We also find that LLMs are twice as likely to be used when writing a paragraph in the Discussion section (68%) compared to a paragraph in the Methods section (32%), but even inside the Methods section, the overall prevalence of LLM usage is over 50%. We believe that our estimates are crucial to shape future guidelines and policies.
[NLP-35] EVIL-Detect for NLPCC 2026 Shared Task 6: LLM -Generated Text Detection NLPCC2026
【速读】： 该论文旨在解决在真实中文场景下对大语言模型生成文本（LLM-generated text, LGT）、人类撰写文本（human-written text, HWT）以及大语言模型润色后的文本（LLM-refined text, HLT）进行可靠区分的问题，尤其面对分布外（out-of-distribution）数据时模型鲁棒性不足的挑战。其解决方案的关键在于提出EVIL-Detect——一种多信号集成框架，融合了编辑程度回归、零样本似然对比信号、词汇统计特征与保守文本规则，并引入冲突感知融合机制，通过校准决策边界实现多源信号的协同优化，显著提升了在复杂、多样化测试场景下的检测性能，最终在NLPCC 2026共享任务6中取得宏平均F1分数0.8888的优异成绩，排名第一。
链接: https://arxiv.org/abs/2608.10698
作者: Hongrui Bao,Hangyu Rong,Zhuoshang Wang,Yubing Ren,Yanan Cao
机构: 未知
类目: Computation and Language (cs.CL)
备注: Accepted by NLPCC 2026 Shared Tasks
Abstract:The rapid development of large language models (LLMs) has increased the need for reliable detection of LLM-generated text, especially in realistic Chinese scenarios involving human-written text (HWT), LLM-generated text (LGT), and LLM-refined text (HLT). This paper presents EVIL-Detect, a multi-signal ensemble framework with conflict-aware fusion for NLPCC 2026 Shared Task 6. The system integrates edit-extent regression, zero-shot likelihood-contrast signals, lexical statistics, and conservative text rules. With calibrated decision boundaries and conflict-aware integration, our system improves robustness under strong out-of-distribution shifts, achieving a macro-F1 score of 0.8888 and ranking first in the official evaluation. Our code is available at this https URL.
[NLP-36] Optimize Cheap Deploy Strong: Cost-Aware Cross-Tier Transfer for Evolutionary Optimization
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）提示词与代理程序（agentic programs）进化优化过程中高昂的计算成本问题，其核心瓶颈在于每次适应度评估均需在验证集上运行回答型LLM，导致评估成本受制于所用模型的价格层级。为突破这一限制，论文提出一种重构策略：将LLM承担的三种角色（答案生成、反思/变异操作、目标部署）解耦，将高频率的答案生成任务分配至最低成本层级的模型，仅在稀有的反思或变异操作中使用高性能模型，并通过跨层级向上迁移机制，将低成本优化得到的提示应用于更强的目标模型。该方法的关键创新在于构建了对“低成本层级搜索能否替代目标层级搜索”的成本可控性分析框架，明确了适用边界与失效场景。实验覆盖四个任务（HotpotQA、IFBench、LiveBench-Math、HoVer）及十一款来自四个模型家族的模型，结果表明，该方法在保持或超越同层级优化性能的同时，超过96%的搜索令牌使用最低成本层级，使搜索成本降低5.6–14倍；在需要推理层级在每次适应度评估中生成长链思维（chain-of-thought）的任务中，成本降幅进一步提升至25–54倍。
链接: https://arxiv.org/abs/2608.10694
作者: Tal Oved,Roi Pony,Oshri Naparstek,Udi barzelay
机构: IBM Research
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Neural and Evolutionary Computing (cs.NE)
备注:
Abstract:Evolutionary optimization of LLM prompts and agentic programs (e.g., GEPA) is dominated by fitness evaluation: scoring each candidate runs an answering LLM over a validation set, so the evaluator’s price tier dictates total search cost. We restructure that search by decoupling the three roles an LLM plays, running the high-volume answering role on the cheapest tier, reserving a strong model for the rare reflection/variation operator, then exploiting upward cross-tier transfer to deploy the cheaply evolved prompt on a stronger target. We contribute a cost-controlled characterization of when cheap-tier search substitutes for target-tier search, and where it fails. Across four tasks (HotpotQA, IFBench, LiveBench-Math, HoVer) and eleven models in four model families, the resulting prompt matches or exceeds same-tier optimization while placing over 96% of search tokens on the cheapest tier, at 5.6-14x lower search cost, rising to 25-54x where reasoning tiers emit long chains of thought on every fitness call.
[NLP-37] SPIEval: Evaluating Large Language Models as Mobile Assistants over Scattered Personal Information
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在作为移动助理时，如何有效利用分散于多个应用程序中的个人数据以完成用户指令这一关键挑战。由于缺乏专门的评估基准，现有模型在处理跨应用个人化任务时的能力尚不明确。为此，作者提出了SPIEval，一个基于五项认知能力（即推理、歧义消解、信息整合、偏好推断及多意图分解）的人工标注基准。SPIEval包含250个任务，覆盖4,335条分布在10个应用中的个人记录，并支持通过21种工具实现多轮交互。分析表明，该基准具备多样化场景、高难度任务、信息分散性、可控环境与可验证结果等特性。对九个代表性LLM的评估显示，性能存在显著差距：表现最佳的GPT-5.5 (xhigh)仅达到57.3%的准确率，最差模型仅为16.4%。进一步分析发现，79%的失败源于信息定位不准，即模型倾向于采纳看似合理但错误的信息，而非持续检索以验证；同时，少于2%的检索操作使用高级搜索方法，且各模型间搜索效率差异显著。这些结果揭示了当前基于LLM的移动助理在信息获取与决策上的根本局限，为未来研究提供了重要方向。
链接: https://arxiv.org/abs/2608.10692
作者: Junjie Ye,Zhuohui Sheng,Shaofan Liu,Yulun Zhu,Wenjie Fu,Dingwei Zhu,Ming Zhang,Yujiong Shen,Weichao Wang,Xin Zhao,Shihan Dou,Tao Gui,Qi Zhang,Xuanjing Huang,Pluto Zhou
机构: Fudan University (复旦大学); Tencent Hunyuan Team (腾讯混元团队)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models (LLMs) are increasingly deployed as mobile assistants, where a key challenge is leveraging personal information scattered across multiple applications (apps) to complete user instructions. However, due to the lack of dedicated benchmarks, their capabilities remain poorly understood. To address this gap, we introduce SPIEval, a human-curated benchmark grounded in five cognitive capabilities (i.e., reasoning, disambiguation, integration, preference inference, and multi-intent decomposition). SPIEval comprises 250 tasks spanning 4,335 personal records distributed across 10 apps and supports multi-turn interaction through 21 tools. Analysis shows that the benchmark exhibits diverse scenarios, challenging tasks, scattered information, controllable environments, and verifiable outcomes. We evaluate nine representative LLMs and find substantial room for improvement. The best-performing model, GPT-5.5 (xhigh), achieves only 57.3% accuracy, while the weakest achieves just 16.4%. Further analysis reveals that 79% of failures stem from inaccurate information localization, as LLMs often commit to plausible but incorrect information instead of continuing retrieval for verification. We also find that fewer than 2% of retrieval actions employ advanced search methods and observe substantial variation in search efficiency across models. These findings expose fundamental limitations of current LLM-based mobile assistants and motivate future research in this direction. Data and code are available at this https URL.
[NLP-38] Can Released LLM Vocabularies Support Token-Level Estimation of Hidden Corpora?
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）预训练语料库组成信息隐匿的问题，即在模型权重公开后，其训练数据来源仍难以准确推断。传统方法通常仅能粗粒度地推测语料混合比例或通过已知分词器词汇表追踪特定词元组，而本文提出了一种更精细的解决方案——针对任意目标词元估计其来自不同语料库的比例。其核心创新在于：首先发现基于不同语料库训练的字节对编码（Byte Pair Encoding, BPE）分词器在词元ID与语料比例分布上具有稳定的统计特性，从而支持从已知语料库向未知语料库的分布迁移；进而提出分位数引导的密度估计（Quantile-Guided Density Estimation, QGDE）方法，通过多分位数趋势拟合分布，并结合局部密度加权实现词元级别的精确估计。在受控环境及使用公开发布的SmolLM分词器的真实场景下，QGDE在词元级估计中实现了最低3.00%的平均相对误差，在聚合为类别级混合比例后误差降至3.08%，表明分词器词汇表可作为细粒度语料构成推断的有效信号，显著超越了以往粗粒度推断的局限性。
链接: https://arxiv.org/abs/2608.10690
作者: Qingjie Zhang,Xingzhang Ren,Zixuan Chen,Jinfeng Li,YueFeng Chen,Yitong Yang,Hui Xue,Dayiheng Liu,Han Qiu
机构: Tsinghua University (清华大学); Qwen Team, Alibaba Group (通义实验室，阿里巴巴集团); Alibaba Group (阿里巴巴集团)
类目: Computation and Language (cs.CL)
备注:
Abstract:Pretraining corpus composition shapes LLM capabilities, but it often remains hidden even when model weights are released. Prior work has inferred corpus mixtures or traced specific token groups from released tokenizer vocabularies; in contrast, we estimate corpus ratios for arbitrary target tokens. We first show that BPE tokenizers trained on different corpora share stable token ID–ratio distributions, motivating distribution transfer from known corpora to a target tokenizer trained on hidden corpora. We then propose Quantile-Guided Density Estimation (QGDE), which approximates this distribution with multiple quantile trends and uses local density weighting to produce token-level estimates. In controlled settings and a realistic setting using the released SmolLM tokenizer, QGDE achieves mean relative errors as low as 3.00% for token-level estimation and 3.08% after aggregation into category-level mixtures. These results suggest that released tokenizer vocabularies provide a useful signal for fine-grained corpus estimation beyond coarse composition inference.
[NLP-39] Auditing Chinese Web-scale Corpora via Sampled BPE Token Statistics
【速读】： 该论文旨在解决大语言模型（LLM）中中文网络污染（Chinese web pollution）的审计难题，尤其针对上游中文语料库中存在的隐性、动态变化且难以检测的污染问题。现有方法面临三大挑战：语料规模庞大导致全量扫描成本过高；已有分析粒度粗，无法定位到具体标记（token）级别的污染；以及中文网络污染本身具有隐含性和快速演化特性。为应对这些挑战，论文提出了一种轻量级的分层标记级审计框架——Sampled-BPE，其核心在于通过采样小规模子集并训练基于字节对编码（BPE）的分词器，从而高效识别受污染的标记。该方案在显著降低计算开销的同时保持较高准确性：相比原始方法实现148.4倍的速度提升和35.8倍的内存减少，仅引入4.25%的相对误差。研究将该方法应用于11个开源中文语料库及2021至2026年间6个中文Common Crawl快照，揭示了开放语料中污染分布不均且随时间剧烈波动的现象。此外，研究还发布了包含超过66万条标记记录的分层中文网络标记数据集，每条记录包含网页上下文、污染类别与解释信息，并以树状结构组织，支持污染溯源与审查。
链接: https://arxiv.org/abs/2608.10678
作者: Qingjie Zhang,Ziqi Tang,Jie Zhang,Gelei Deng,Jinfeng Li,YueFeng Chen,Yitong Yang,Hui Xue,Tianwei Zhang,Han Qiu
机构: Tsinghua University (清华大学); SiliconProspect AI; Nanyang Technological University (南洋理工大学); Alibaba Group (阿里巴巴集团)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Chinese web pollution has surfaced in LLMs, motivating audits of upstream Chinese corpora. However, auditing such corpora faces three challenges: (1) their web-scale size makes full scan costly; (2) prior analyses are often too coarse to expose token-level pollution; (3) Chinese web pollution is implicit and rapidly changing. We propose Sampled-BPE, a lightweight token-level auditing pipeline that sample a small subset and train BPE tokenizer to surface polluted tokens. Experiments show that Sampled-BPE preserves usable estimates while substantially reducing runtime and memory: a 148.4 \times speedup and a 35.8 \times memory reduction induce only 4.25% relative error for pollution categories. We apply the pipeline to 11 open Chinese corpora and 6 Chinese Common Crawl snapshots from 2021 to 2026. The audit reveals widespread but uneven pollution across open corpora, as well as highly polluted and temporally shifting Chinese web content. We further release a hierarchical Chinese web token dataset with 660k+ token records, each with web context, category, and explanation fields, organized as trees to support review and tracing of pollution.
[NLP-40] Seeds Before Objectives: Rethinking Evaluation for Low-Resource Garhwali ASR
【速读】： 该论文旨在解决低资源方言在自动语音识别（ASR）任务中因语料规模有限而导致的模型性能评估不可靠问题，特别是单次运行结果难以复现的挑战。针对印度喜马拉雅山区的低资源语言——加尔瓦利语（Garhwali），研究构建了首个基于官方VAANI数据集划分的可复现多种子（multi-seed）ASR基准，包含每个种子的输出结果及显著性检验。其核心解决方案在于采用多种子评估范式，通过在官方划分的数据集上进行多轮独立实验，有效区分真实性能提升与由随机种子引入的噪声。研究发现，尽管此前被认为有效的改进方法如焦点CTC（Focal CTC）或音节权重目标函数（matra-weighted objective）在种子级别测试中均未表现出显著优势，甚至无法有效减少目标错误；跨语言迁移学习（从印地语到加尔瓦利语）也未优于直接微调。真正稳定的性能提升来自基础架构设计：使用标准CTC损失的w2v-BERT 2.0模型在五个种子上平均达到47.0%的词错误率（WER），优于参数量更大的MMS-1B及其他可比模型，表明预训练设计的重要性远超参数数量，且速度增强（speed augmentation）虽仅带来小幅增益，但具有高度一致性。因此，该研究的关键在于强调多种子评估对识别真实性能增益的必要性，并揭示在低资源场景下，稳健的预训练架构与数据增强策略比复杂损失函数更具决定性作用。
链接: https://arxiv.org/abs/2608.10670
作者: Karamvir Singh Batra,Prathamjyot Singh,Ashima Sood,Jasmeet Singh,Sahil Sharma
机构: Thapar Institute of Engineering and Technology, Patiala, Punjab, India; Ulster University, Londonderry, United Kingdom; Ulster University, Belfast, United Kingdom
类目: Computation and Language (cs.CL)
备注: 19 pages, 3 figures. Accepted for oral presentation at ICNLSP 2026, Trento, Italy, September 2026
Abstract:At corpus sizes typical of low-resource dialects, single-run comparisons can yield gains that do not replicate. We show this for Garhwali, an under-resourced Indo-Aryan language of the central Himalaya, building the first reproducible multi-seed ASR benchmark on the official VAANI splits, with per-seed outputs and significance testing. Re-examining plausible gains, we find them fragile: neither Focal CTC nor a matra-weighted objective beats standard CTC under seed-level testing, the matra objective fails to cut even its targeted errors, and Hindi-to-Garhwali transfer gives no gain over direct fine-tuning. What holds up is mundane: w2v-BERT 2.0 with standard CTC reaches 47.0% WER over five seeds, beating the larger MMS-1B and comparable models; pretraining design, not parameter count, drives performance, and speed augmentation gives a small, largely consistent gain. Multi-seed evaluation on official splits separates real gains from seed noise.
[NLP-41] InSight-doc: Agent ic Visual Perception for Long-Document Understanding
【速读】： 该论文旨在解决长文档理解中因视觉信息丰富而导致的推理成本高、上下文衰减（context rot）严重的问题。其核心挑战在于如何在保证准确性的前提下，高效地处理大量视觉内容并避免生成幻觉。解决方案的关键在于提出一种名为InSight-doc的智能体式视觉感知框架，将视觉分辨率视为可动态调整的推理资源：系统初始以低分辨率进行全局扫描，仅对关键区域主动放大至高分辨率以获取细粒度证据，从而实现按需计算。该方法不依赖外部检索器，通过构建包含17.9K高质量监督微调（SFT）样本和19.2K强化学习（RL）样本的主动感知语料库，结合SFT+RL训练策略，使InSight-doc-8B在文档视觉问答（VQA）基准上相较基线提升4.3–16.4个百分点的准确率；在长文档场景下，不仅将幻觉率降低超过40%，还将推理延迟减少41%–68%，同时保持显著的性能优势。
链接: https://arxiv.org/abs/2608.10628
作者: Kaican Li,Weiyan Xie,Lewei Yao,Jiannan Wu,Lanqing Hong,Yongxiang Huang,Nevin L. Zhang
机构: The Hong Kong University of Science and Technology (香港科技大学); Huawei (华为)
类目: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Long-document understanding often requires reasoning over many visually rich pages, making inference costly and prone to context rot. In this work, we propose InSight-doc, an agentic visual perception framework that treats visual resolution as an adaptive reasoning-time resource. InSight-doc starts from low resolution and selectively zooms into high-resolution regions for finer evidence, without relying on any external retriever. To train such an agent, we construct an active-perception corpus of 17.9K high-quality SFT examples with region-level zoom-in trajectories, accompanied by 19.2K hard RL examples. Through SFT+RL, InSight-doc-8B improves the baseline by 4.3–16.4 accuracy points over document VQA benchmarks. On long documents, it reduces hallucination by more than 40% and inference latency by 41%–68% while maintaining an accuracy lead. Our code, datasets, and model are released at this https URL .
[NLP-42] Decomposition-Induced Context-Memory Conflict: When Fact-Checking Pipelines Contradict Their Own Source Text
【速读】： 该论文旨在解决生成式AI在事实性评估中因“分解-验证”（decompose-then-verify）流水线所引入的隐性偏差问题，即分解器（decomposer）可能在其参数化信念影响下，将原文内容替换为与原始文本相矛盾的主张，从而导致生成的原子命题（atomic claims）偏离源文本语义。这一现象被称为分解诱导的上下文-记忆冲突（Decomposition-Induced Context-Memory Conflict, DI-CC），其机制与经典的上下文-记忆冲突（context-memory conflict）本质相同，但发生于分解阶段而非以往研究关注的生成阶段。研究发现，仅基于经典上下文-记忆冲突数据训练的线性探测器（如NQ-Swap）即可显著区分产生DI-CC的分解结果与忠实分解结果（AUC = 0.86–0.88，置换检验p < 0.0005），表明该现象具有可识别的模式。然而，现有无需参考的基线方法（如SelfCheckGPT式的自一致性采样）无法检测DI-CC（AUC ≈ 0.51，接近随机水平），因其依赖的采样变异在DI-CC中不出现，而内容高度稳定且重复。尽管源自经典场景的上下文感知解码（context-aware decoding）可有效抑制DI-CC，但会带来严重副作用——在核心指代密集条件下大量分解失败，常因分解器虚构实体身份所致，因此不具备实际部署可行性。研究进一步揭示了该故障模式的边界：其自然发生率极低，在真实幻觉文本中难以显现，且需达到一定模型规模才能触发，确认其为一种机制上可解释、部分可缓解但不可忽视的系统性缺陷。
链接: https://arxiv.org/abs/2608.10627
作者: Yu-Feng Yen
机构: 未知
类目: Computation and Language (cs.CL)
备注: 15 pages, 1 figure
Abstract:Decompose-then-verify pipelines, including FActScore-style fact-checkers and long-form factuality evaluators, first split a passage into atomic claims before checking each one. Decomposition itself is treated as a neutral preprocessing step. We show it is not: a decomposer can be induced to substitute its own parametric belief for what the source passage says, producing a claim that contradicts the text it was supposed to summarize faithfully. We call this Decomposition-Induced Context-Memory Conflict (DI-CC) and show it is mechanistically the same phenomenon as classical context-memory conflict, occurring inside a different pipeline stage than prior work has examined. A linear probe trained only on classical context-memory conflict data (NQ-Swap), never exposed to any decomposition output, significantly separates decomposition positions that produce DI-CC from faithful decompositions (AUC = 0.86-0.88, permutation p 0.0005). An existing reference-free baseline, SelfCheckGPT-style self-consistency sampling, fails to detect DI-CC at all (AUC 0.51, chance-level), because DI-CC content is stably recoverable and recurs across resamples, unlike the variability self-consistency methods rely on. Context-aware decoding, a training-free mitigation from the classical setting, transfers to decomposition and suppresses DI-CC, but at a severe cost: many decompositions under coreference-heavy conditions fail to parse, often because the decomposer fabricates a different identity. We do not consider this mitigation deployment-ready. We further characterize the mechanism’s boundaries: its natural occurrence rate is too sparss not manifest on naturally-occurring hallucinatedtext, and it requires a minimum model scale to detecablish DI-CC as a real, mechanistically grounded, andpartially treatable failure mode, with a scope we chhan overstate.
[NLP-43] Dual-Loop Self-Evolution via Verifiable Emotion Feedback for Multi-Turn Empathetic Dialogue
【速读】： 该论文旨在解决大语言模型在对话中缺乏共情能力的问题，尤其针对共情支持这一具有多轮依赖性（multi-turn and path-dependent）的复杂交互任务。现有基于强化学习的方法虽采用可验证的情绪奖励实现长程交互的可扩展监督，但其训练过程中固定交互分布，导致策略能力与训练经验之间存在不匹配。本文提出一种双环自进化框架（dual-loop self-evolution framework），以可验证的情绪反馈为驱动，核心创新在于：内环在用户模拟器和验证器冻结的前提下，利用连续情绪奖励优化多轮对话策略；外环则基于相同交互结果估计策略相对的交互效用，并动态调整训练经验分布。为应对稀疏、随机的回溯采样，框架在每组内保持场景与交互状态恒定，优先选择群体通过率接近策略能力边界的条件，以提升评估效率。同时，分层控制器在不同支持意图间共享证据，不确定性引导探索与均匀重放机制防止过早排除关键样本。最终生成的动态经验分布可在不增加回滚预算的情况下闭合双环。在SAGE基准上，该方法将Qwen3-8B的整体性能从53.87提升至79.24，相较于协议匹配的均匀情绪奖励强化学习方法高出7.23分。
链接: https://arxiv.org/abs/2608.10626
作者: Yi Wei,Shuo Jiang,Huaixia Dou,Jie Zhu,Junhui Li,Lifan Guo,Feng Chen,Chi Zhang
机构: 未知
类目: Computation and Language (cs.CL)
备注: 10 pages, 4 figures, 6 tables
Abstract:Large language models have demonstrated conversational capabilities, yet empathetic competence remains challenging. Empathetic support is inherently multi-turn and path-dependent: users disclose concerns gradually, emotions evolve over time, and early responses shape trust and receptivity. Reinforcement learning with verifiable emotion rewards provides scalable supervision for long-horizon interactions. However, existing methods evolve the dialogue policy while keeping its training interaction distribution fixed, creating a mismatch between policy competence and training experience. We introduce a dual-loop self-evolution framework driven by verifiable emotion feedback. With the user simulator and verifier frozen, the inner loop optimizes the multi-turn policy using continuous emotion rewards, while the outer loop uses the same outcomes to estimate policy-relative interaction utility and adapt experience. To obtain estimates from sparse, stochastic rollouts, the framework holds the scenario and interaction state constant within each group and prioritizes conditions whose group pass rates lie near the policy’s competence boundary. A hierarchical controller shares evidence across support intents, while uncertainty-guided exploration and uniform rehearsal prevent premature exclusion. The resulting distribution generates trajectories, closing both loops without increasing the rollout budget. On SAGE, our framework raises Qwen3-8B Overall from 53.87 to 79.24 and outperforms protocol-matched uniform emotion-reward reinforcement learning by 7.23 points.
[NLP-44] Simplex Relaxation for Discrete Diffusion
【速读】： 该论文旨在解决离散扩散模型（Discrete Diffusion Models）在类别型数据生成中，如何在不改变原有均匀腐蚀过程（uniform categorical corruption process）的前提下，提升其训练目标与反向转移过程的表达能力。核心问题在于现有方法在保持原始腐蚀机制不变时，难以有效增强模型的逆向预测能力与生成质量。解决方案的关键是提出Simplax——一种精确的Dirichlet-类别增强方法，通过将每个被腐蚀的类别状态与一个辅助的单纯形值变量（simplex-valued variable）耦合，同时保持原始均匀扩散过程作为其类别边缘分布。该增强机制导出了一个可计算的Rao–Blackwellized反向桥接目标函数，并支持相应的随机反向采样器，同时保留被腐蚀的类别状态作为去噪器输入。实验表明，Simplax在无条件生成任务中显著改善了生成困惑度与熵之间的权衡，在OpenWebText上表现更优；在数独生成任务中，仅用30个提示谜题训练的模型在所有提示密度下均达到最高准确率，包括最小唯一可解的17提示情形，并在无条件生成中展现出最高的有效性。
链接: https://arxiv.org/abs/2608.10615
作者: Jinya Sakurai,Patrick Pynadath,Satoshi Hayakawa,Jaehong Yoon,Xulei Yang,Nancy F. Chen,Xun Xu
机构: NTU Singapore; The University of Tokyo; Purdue University; Institute for Advanced Intelligence and Computing (IAIC), ASTAR; Centre for Frontier AI Research (CFAR), ASTAR
类目: Computation and Language (cs.CL)
备注:
Abstract:Discrete diffusion models for categorical generation are defined by a corruption kernel, which determines the intermediate state space and the associated reverse prediction problem. We study uniform discrete diffusion and ask whether its training objective and reverse transitions can be enriched without changing the underlying categorical corruption process. We introduce Simplax, an exact Dirichlet–categorical augmentation that couples each corrupted categorical state with an auxiliary simplex-valued variable while preserving the original uniform diffusion process as its categorical marginal. This augmentation yields a tractable Rao–Blackwellized reverse-bridge objective and a corresponding stochastic reverse sampler, while retaining the corrupted categorical state as the denoiser input. Empirically, Simplax improves the generative perplexity–entropy tradeoff on unconditional OpenWebText generation. On Sudoku, a model trained exclusively on 30 -clue puzzles achieves the highest accuracy among the compared methods across all evaluated clue densities, including the minimum uniquely solvable 17 -clue regime, and also achieves the highest validity in unconditional generation.
[NLP-45] ASR-Roundtrip Evaluation Can Mask Context- and Convention-Dependent Reading Errors in Chinese News TTS
【速读】： 该论文旨在解决语音合成（TTS）在中文新闻文本生成中因上下文或领域惯例依赖导致的可理解性评估偏差问题，尤其关注那些正确读法依赖于特定语境（如体育比分、飞机型号、技术单位、会员名称等）的片段。传统基于自动语音识别（ASR）回环评估（ASR-roundtrip evaluation）虽具可扩展性，但可能产生“假阴性”错误——即合成语音存在读错但被ASR误识别为正确内容的情况。其核心解决方案在于通过构建高风险案例集并进行针对性审计，揭示了现有评估方法的局限性：在110个高风险案例中，46例被掩盖的错误经隔离诊断后重新暴露，且不同模型（如CosyVoice）的独立审计也确认了大量未被发现的错误。实验表明，仅依赖ASR回环作为真实标签不足以准确评估中文新闻TTS的读音可靠性，需结合更精细的诊断手段与多模型验证。关键突破在于提出以“跨模型对比+跨音频诊断”为核心的复合评估框架，从而提升对语义敏感型读音错误的检测能力。
链接: https://arxiv.org/abs/2608.10606
作者: Shijun Luo,Lizhi Wan
机构: 未知
类目: Computation and Language (cs.CL)
备注: 5 pages, 4 tables. Conference-format manuscript. Supporting materials are available at this https URL and archived at this https URL
Abstract:ASR-roundtrip evaluation is widely used as a scalable proxy for text-to-speech (TTS) intelligibility, but it can produce false negatives for reading errors perceived by listeners. We study Chinese news TTS spans whose correct reading depends on context or domain conventions, such as sports scores, aircraft models, technical units, and membership names. In these cases, Raw TTS can choose a plausible but wrong reading while ASR transcribes the audio as the intended or surface-correct text. A targeted audit over 110 high-risk MiMo TTS cases, reported with a complete denominator, confirms 46 masked false negatives, 9 exposed TTS errors, and 55 cases with no Raw TTS error. A span-isolation diagnostic re-exposes 18/46 previously masked errors. A Raw-only CosyVoice audit on the same targeted pool confirms 51 masked cases. Across the 97 TTS-specific audio files labeled confirmed masked across the two audits, Qwen3-ASR surface-recovers 40 cases, whereas Paraformer does so in only 2. The results suggest that ASR-roundtrip is useful for screening but insufficient as standalone ground truth for Chinese news reading-risk evaluation.
[NLP-46] RadFusion: Towards Threshold-Controllable Radiology Report Generation
【速读】： 该论文旨在解决生成式放射科报告在临床应用中缺乏对诊断敏感性-特异性权衡（sensitivity-specificity trade-off）可控性的问题。现有生成模型无法根据不同的临床场景动态调整其诊断输出，例如急诊分诊需高敏感性以避免漏诊，而确诊评估则更关注高特异性以减少不必要的干预。这一缺陷导致生成报告难以适应多样化临床需求，且无法通过广泛认可的受试者工作特征曲线（ROC curve）进行定量验证，阻碍了其在监管审批中的应用。该研究提出RadFusion框架，其关键创新在于将多标签分类器（提供各疾病置信度评分）与基于视觉问答（VQA）的报告生成器相融合，并引入大语言模型（LLM）对生成内容进行重写，使报告中的诊断结论严格遵循分类器在指定阈值下的决策，同时保持与原始生成描述的语义一致性。实验表明，在MIMIC-CXR数据集上，RadFusion生成报告的性能可精准映射至分类器的ROC曲线，实现了报告生成结果的定量可验证性，支持基于ROC的监管评估，并可根据临床场景灵活选择最优工作点。此外，该方法在诊断准确性上显著优于无控制的生成方式：在匹配特异性条件下敏感性提升6.9%，在匹配敏感性条件下特异性提升20.7%。因此，RadFusion实现了报告生成在临床适应性、定量可验证性和诊断可靠性方面的全面提升。
链接: https://arxiv.org/abs/2608.10505
作者: Ying Jin,Noel C. F. Codella,John Corring,Mu Wei,Dinei Florencio,Eric Horvitz
机构: Microsoft(微软)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Automated radiology report generation is advancing rapidly in response to the shortage of radiologists, yet unlike a perception model, existing generation models offer no control over the sensitivity-specificity trade-off of their diagnostic content. Such control is essential because clinical scenarios diverge: emergency triage prioritizes sensitivity to reduce missed findings, whereas confirmatory interpretation emphasizes specificity to limit unnecessary interventions. A single fixed report can neither adapt to these scenarios nor support the ROC-based validation widely expected for regulatory clearance. We introduce RadFusion, a framework that equips report generation with threshold controllability. Our method fuses a multi-label classifier, which provides per-disease confidence scores, with a VQA-based report generator, which describes medical findings in detail; an LLM then rewrites the report so that its stated diagnoses follow the classifier’s decisions at the selected threshold while staying grounded in the generator’s descriptions. On MIMIC-CXR, the performance of RadFusion conforms to the classifier’s ROC curve: sweeping the threshold and mapping the reports back to class labels reproduces the classifier’s validated ROC performance. This conformance makes generated reports quantitatively evaluable through ROC analysis, strengthening the case for regulatory clearance, and enables operating-point selection that matches report behavior to clinical context. Moreover, combining the two model types improves diagnostic accuracy over uncontrolled generation: sensitivity increases by 6.9% at matched specificity, and specificity by 20.7% at matched sensitivity. These results show that RadFusion makes report generation clinically adaptable, quantitatively verifiable, and diagnostically more reliable.
[NLP-47] Every Token Counts: Exact Likert-Scale Distributions for Measuring LLM Attitudes and Biases
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在作为自主代理部署时，对其潜在价值观与偏见进行准确评估的难题。传统自然语言处理（NLP）社区依赖大规模非结构化基准测试来评估模型性能，但这类方法存在根本性缺陷：其无法解耦因果机制，即当检测到整体偏差时，难以区分该偏差是源于模型固有特征、上下文混淆因子，还是复杂交互作用所致。为此，本文提出一种分析上精确的受控行为评估框架，通过将人类心理测量学与大语言模型内在机制相融合，弥补了实验设计、测量方式和分析方法上的空白。其核心解决方案包括：首先，采用完全交叉的因子实验设计替代非结构化提示，以系统分离因果主效应与交互效应；其次，通过直接操作精确的分词级概率质量函数（Probability Mass Function, PMF），消除蒙特卡洛文本采样带来的噪声；最后，构建多变量有序共识度量与分布型方差分析（ANOVA），实现对PMF的解析化处理。通过在五个大语言模型上开展消费者民族中心主义的案例研究，验证了该框架能够有效识别出聚合基准测试所掩盖的系统性原产国偏见。
链接: https://arxiv.org/abs/2608.10503
作者: Davood Wadi,Mohsen Ghodrat,Matthew Philp
机构: McGill University; University Canada West; Toronto Metropolitan University
类目: Computation and Language (cs.CL)
备注:
Abstract:As Large Language Models (LLMs) are increasingly deployed as autonomous agents, accurately evaluating their latent values and biases is critical. The NLP community typically evaluates models using large, unstructured benchmarks. While effective for general capabilities, these datasets fundamentally conflate causal mechanisms: even when an aggregate bias is detected, unstructured evaluations cannot disentangle whether it stems from baseline traits, contextual confounders, or complex interactions. To address this, we introduce an analytically exact framework for the controlled behavioral evaluation of LLMs. We bridge human psychometrics with LLM mechanics by resolving gaps in design, measurement, and analysis. First, we replace unstructured prompting with fully crossed factorial experiments to systematically isolate causal main and interaction effects. Second, we eliminate Monte Carlo text sampling noise by operating directly on exact, token-level Probability Mass Functions (PMFs). Third, we derive a multivariate ordinal consensus metric and a distributional ANOVA to process these PMFs analytically. We validate our framework with a case study on consumer ethnocentrism across five LLMs, demonstrating how our approach isolates systemic country-of-origin biases that aggregate benchmarks otherwise obscure.
[NLP-48] Lost in Reconstruction: Aligning Action Representations with Language in Vision-Language-Action Models
【速读】： 该论文旨在解决视觉-语言-动作模型（Vision-Language-Action models, VLAs）中动作表征因仅依赖重建损失（L1/L2）优化而导致语义信息丢失的问题。传统方法在原始动作空间中通过最小化数值误差进行动作重构，但这种数值接近性并不等同于语言上的语义区分，从而削弱了动作与语言指令之间的语义对齐。其关键解决方案是提出SALT（Semantically ALigned action Tokenizer），一种基于向量量化变分自编码器（VQ-VAE）架构的语义对齐动作分词器，通过引入一个辅助目标：要求冻结的视觉-语言模型从量化后的动作潜在表示中恢复任务指令。这一机制强制保留动作轨迹中蕴含的动词语义信息，避免了仅依赖重建时出现的语义侵蚀。实验表明，采用SALT训练的策略在SimplerEnv中达到71.9%的平均成功率，显著优于仅使用重建目标的VQ-VAE（42.7%）和FAST方法（31.2%），同时实现了动词特化的代码学习与高保真重建。结果表明，机器人动作轨迹本身蕴含丰富的语言接地信息，而通过语义对齐方式保持该结构可显著提升语言条件控制性能。
链接: https://arxiv.org/abs/2608.10484
作者: Li Wenjie,Yash Jangir,Ignacy Stepka,Yash Agarwal,Marion Kipsang,Yonatan Bisk
机构: Carnegie Mellon University (卡内基梅隆大学)
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Action verbs describe not only the physical outcomes of actions, but also how those actions are performed. Yet action representations in vision-language-action models (VLAs) are typically optimized for reconstruction under L1/L2 losses in raw action space, where numerical proximity need not reflect linguistically meaningful distinctions. On BridgeV2, we show that action trajectories contain verb-grounding information beyond visual state changes, and that reconstruction-only discrete tokenization systematically erodes this information. To address this problem, we introduce SALT, a Semantically ALigned action Tokenizer that augments a VQ-VAE-style tokenizer with an auxiliary objective requiring a frozen vision-language model to recover the episode instruction from quantized action latents. Policies trained with SALT achieve 71.9% average success in SimplerEnv, compared with 42.7% for a reconstruction-only VQ-VAE tokenizer and 31.2% for FAST. SALT also develops verb-specialized codes while maintaining reconstruction fidelity. These results show that robot action trajectories provide a source of language grounding and that preserving this structure in action representations can substantially improve language-conditioned control.
[NLP-49] Evaluating Rational Contracting in Natural Language
【速读】： 该论文旨在解决当前基于语言的AI代理在复杂、动态且不确定的多步经济交互中，难以有效协商、解释与执行自然语言合同的问题。现有研究多局限于一次性交易或简化经济博弈，忽视了语言赋予的时序延展性、条件依赖性和不完全契约等现实特征，且仅关注利润最大化而忽略可信契约所需的理性与合作品质。其解决方案的关键在于构建一个理性框架，指导代理在不确定性环境下进行多轮协商与合同履行，并在此基础上提出可量化的评估指标与基线方法。通过在名为ContractSim的仿真环境中对餐饮、酒店清洁及AI托管三种供应场景进行测试，研究发现当前基于大语言模型（LLM）的代理虽能在低不确定性下可靠达成协议并实现高效谈判，但在高不确定性条件下往往无法达成可满足、高效或互利的合约；同时在履约阶段表现出显著的非合作倾向，为获取额外收益而违反合同条款。这一结果揭示了现有语言代理在理性决策与合作行为方面的不足，凸显了未来需在设计上强化其对复杂合同的推理、承诺遵守与协作能力。
链接: https://arxiv.org/abs/2608.10475
作者: Bhavyesh Sajja,Max Kleiman-Weiner,Roger Zimmermann,Tan Zhi-Xuan
机构: University of California, Berkeley (加州大学伯克利分校); Massachusetts Institute of Technology (麻省理工学院); National University of Singapore (新加坡国立大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computer Science and Game Theory (cs.GT)
备注: 9 pages, 5 figures, 2 tables (Appendix: 34 pages, 6 figures, 9 tables)
Abstract:The emergence of language-based AI agents promises to transform the scope of machine economic activity. Instead of just proposing bids or following hard-coded protocols, such agents can be used to negotiate and execute agreements in open-ended natural language. However, most evaluations of these abilities have focused on one-off exchanges or simple economic games, leaving open the rich space of time-extended, contingent, and incomplete contracts made expressible by language; they also focus on raw profit, without measuring the qualities required for trustworthy contracting. We address this by formulating a rational framework for how agents should negotiate and perform natural language contracts in uncertain multi-step environments. Within this framework, we develop metrics and baselines for quantifying rational and cooperative play. To evaluate how agents perform at such contracting, we instantiate our framework in ContractSim, an evaluation suite where two players negotiate and execute a multi-turn supplier contract under environmental and inter-player uncertainty. Across six environments and three supplier settings (catering, hotel cleaning, and AI hosting) we find that current LLM-based agents reach agreement reliably, and negotiate efficient contracts when environmental uncertainty is low. However, under high uncertainty, they often fail to negotiate satisfiable, efficient, or mutually beneficial contracts. They are also frequently uncooperative when executing contracts, violating contract terms for additional profit even when contracts are easy to satisfy. These findings highlight room for improvement in the design of language agents that can negotiate, interpret, and execute contracts both rationally and cooperatively.
[NLP-50] Calibrating Post-Training Feature Shifts for LLM Data Contamination Detection
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在预训练阶段可能包含版权或隐私敏感内容所引发的数据污染检测（Data Contamination Detection, DCD）问题，即判断给定文本是否属于目标模型的预训练语料库。现有最先进的特征基DCD方法依赖于输入文本及其对应模型输出提取成员特征，但现代LLMs普遍经历指令微调、偏好优化和推理导向训练等后训练过程，这些过程会改变模型输出并导致成员特征分布偏移，从而降低成员与非成员之间的可区分性。针对此问题，本文提出一种通用性强的校准框架CalibDCD，其关键在于：（1）多视角特征偏移检测（Multi-View Shift Detection），通过在已知非成员文本上测试受控提示变体，识别与后训练相关的重复性特征偏移；（2）有界特征修正（Bounded Feature Correction），对与检测到的偏移方向一致的特征分量进行选择性调整，并严格控制修正范围以保留有效的检测信息。实验表明，CalibDCD能持续提升现有特征基检测器性能，在AUC上最高提升7.0%，在TPR@5%FPR上最高提升15.0%。
链接: https://arxiv.org/abs/2608.10462
作者: Zhen Yang(1),Mengqi Wang(1),Gengda Zhao(1),Mo Zhou(1),Jianwei Wang(1),Wenjie Zhang(1) ((1) The University of New South Wales)
机构: The University of New South Wales(新南威尔士大学)
类目: Computation and Language (cs.CL)
备注: 14 pages, 7 figures. The first two authors contributed equally
Abstract:Large language models (LLMs) are trained on massive and largely undisclosed corpora that may contain copyrighted or privacy-sensitive content. Data contamination detection (DCD) therefore aims to determine whether a given text is a member of the pre-training corpus of a target LLM. Recent state-of-the-art DCD methods follow a feature-based paradigm that derives membership features from the input text and the corresponding model output. However, most modern LLMs undergo post-training, such as instruction tuning, preference optimization, and reasoning-oriented training, which can alter model outputs and shift the corresponding membership features, thereby reducing the separability between members and non-members. To address this problem, we propose CalibDCD, a broadly applicable calibration framework for feature-based DCD methods, comprising (1) Multi-View Shift Detection, which identifies recurring feature shifts associated with post-training, and (2) Bounded Feature Correction, which selectively mitigates their influence on membership prediction. Specifically, Multi-View Shift Detection evaluates controlled prompt variants on known non-member texts and consolidates the most informative views to identify recurring feature shifts. Bounded Feature Correction selectively adjusts feature components aligned with the detected shifts and controls the correction extent to preserve useful detection information. Experiments show that CalibDCD consistently improves existing feature-based detectors, with gains of up to 7.0% in AUC and 15.0% in TPR@5%FPR. Comments: 14 pages, 7 figures. The first two authors contributed equally Subjects: Computation and Language (cs.CL) Cite as: arXiv:2608.10462 [cs.CL] (or arXiv:2608.10462v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2608.10462 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[NLP-51] MD-ProTector: Positioning Multiple Data-Driven Prototypes for LLM -Generated Text Detection
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）生成内容日益复杂背景下，如何高效、准确地识别生成文本与人工撰写的文本之间的差异问题。现有基于输入编码器的检测系统虽适用于实际部署，但传统二分类方法仅输出类别标签，无法显式建模同一类别内部的多样性特征。为此，本文提出MD-ProTector，其核心创新在于在编码器嵌入空间中为每个类别引入多个可训练的参考向量（即原型，prototype），以捕捉类内不同子群体的分布差异，并通过独立的决策边界实现更精细的判别。关键突破在于引入原型定位损失（Prototype Positioning loss），将类别整体结构与类内变异解耦，从而明确每个原型所代表的具体文本变异性。实验在覆盖领域、生成模型、语言及对抗性扰动的三个大规模基准数据集上验证了该方法的有效性，结果表明MD-ProTector在MAGE CDCM和RAID数据集上达到最高的平均召回率（AvgRec），并在RAID上取得最优的受试者工作特征曲线下面积（AUROC）与最低的假阳性率95%（FPR95），显著优于其他基于编码器的方法。
链接: https://arxiv.org/abs/2608.10459
作者: Jinmo Han,Jimin Hong,Chanyeong Moon,Ju Yeon Kang,Seonuk Kim,Nam Soo Kim
机构: Seoul National University, Seoul, Republic of Korea
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:As LLM-generated content becomes more sophisticated, detection systems for distinguishing those texts from human-written text must operate at scale while handling diverse writing styles, domains, languages, and generator models. Input-only encoder detectors are suitable for practical deployment setting, but standard binary classification supplies only the class label and does not explicitly organize the substantial variation within either class. We propose MD-ProTector, which represents each class with multiple trainable reference vectors in the encoder embedding space, referred to as prototypes. These prototypes provide separate decision boundaries for different groups of texts within the same class. However, adding multiple prototypes alone does not determine which variation each prototype should represent. MD-ProTector addresses this problem with Prototype Positioning loss, which separates class-level structure from the within-class variation that differentiates individual prototypes. Evaluated across five settings from three large-scale benchmarks covering domain, generator, language, and adversarial variation, MD-ProTector achieves the highest AvgRec on MAGE CDCM and RAID and the highest AUROC and lowest FPR95 on RAID among the compared encoder-based methods.
[NLP-52] From Reasoning Depth to Reasoning Reasoning Breadth: Evaluating Multi-Point Associative Reasoning in Large Language Models
【速读】： 该论文旨在解决当前大型语言模型（Large Language Models, LLMs）在推理能力评估中过度关注推理深度（reasoning depth）而忽视推理广度（reasoning breadth）的问题。推理广度指的是模型能够并行探索多个语义方向，整合多样化线索以形成连贯答案的能力。为系统评估这一能力，研究提出MPAR-Bench——一个中英文双语基准测试，通过多点关联推理（multi-point associative reasoning）来隔离和衡量推理广度。其核心解决方案在于构建一个由多智能体生成、基于嵌入相似性筛选并经人工验证的1000个独立线索集，每个线索集均从零生成，仅答案空间来源于公开词表。评估不仅包含精确匹配准确率，还引入了ANLS、嵌入相似度、推理轨迹验证及四种扰动实验（线索遮蔽、顺序打乱、干扰项注入、多步线索），结果表明，现有模型在面对扰动时性能下降5–18个百分点，且思维模式虽提升基础准确率，但未能一致缓解对扰动的敏感性；案例分析进一步揭示，延长推理过程可能推翻初始正确假设。这些发现表明，更高的推理深度并不自动带来鲁棒的推理广度，当前主流基准仍严重低估了该维度的能力，亟需更全面的评估框架。
链接: https://arxiv.org/abs/2608.10444
作者: Si’an Xie(1),Jiaxun Liu(2),Biao Yang(3),Wei Yuan(3),Fan Yang(3),Tingting Gao(3),Ming Wu(1) ((1) Beijing University of Posts and Telecommunications, (2) Peking University, (3) Kuaishou Technology)
机构: Beijing University of Posts and Telecommunications (北京邮电大学); Peking University (北京大学); Kuaishou Technology (快手科技)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models (LLMs) have made substantial progress on reasoning tasks that require increasingly long and complex inferential chains. This progress primarily reflects reasoning depth. A complementary and comparatively unexamined capability is reasoning breadth: exploring multiple semantic directions in parallel and integrating the resulting clues into one coherent answer. We introduce MPAR-Bench, a bilingual English-Chinese benchmark that isolates reasoning breadth through multi-point associative reasoning. Inspired by the cooperative game Just One, each item asks a model to recover a hidden target from several independently generated, semantically diverse clues. We construct 1,000 items using a multi-agent clue-generation pipeline, embedding-based diversity filtering, and human verification. Only the answer space is drawn from public word lists, whereas every clue set is generated from scratch. Beyond exact-match accuracy, we evaluate models using accuracy, ANLS, embedding similarity, reasoning-trace verification, and four perturbations: clue masking, order shuffling, distractor injection, and multi-step clues. Across evaluated models, perturbations reduce accuracy by 9-18 percentage points in English and 5-12 percentage points in Chinese. Thinking mode improves standard-setting accuracy, especially in English, but does not consistently reduce sensitivity to perturbations. Case-level analysis also shows that extended reasoning can overturn an initially correct hypothesis. These results indicate that greater reasoning depth does not automatically confer robust reasoning breadth, and that reasoning breadth remains largely uncovered by current benchmarks.
[NLP-53] Riemann GeoResolver: A Non-Euclidean Attention Framework from Euclidean Resolver to Hyperbolic-Spherical Geometry
【速读】： 该论文旨在解决传统注意力机制（如Softmax注意力）在计算资源消耗、收敛速度、泛化能力及高维数据建模方面的固有局限性，尤其针对其在低秩/聚类结构下仍存在冗余参数、局部极小值陷阱以及噪声记忆过度等问题。其核心解决方案是提出一种基于逆距离注意力（Inverse-Distance Attention, IDA）的理论框架，从欧几里得空间原型（Resolver）延伸至非欧几里得实现（Riemann GeoResolver）。关键创新在于：在欧几里得层面，通过三个核心定理建立严格理论基础——(1) 电路分离性证明IDA可实现O(1)资源下的精确检索，而Softmax需Ω((logn)2)宽度；(2) 基于Polyak–Łojasiewicz（PL）不等式的更强常数，确保线性收敛、O(logn) Lipschitz尺度缩放、Θ(1) Hessian谱展宽且无虚假局部极小点；(3) 宽度无关的有效秩上界，限制噪声记忆，使测试误差被控制在O(η2)内，避免Softmax在dh≥n时对任意标签过拟合。在此基础上，非欧扩展引入双曲几何测地距离用于存储、球面测地距离用于路由，构建了包含十项模块的Riemann GeoResolver框架，涵盖四类高效HIDA算子、具有可证明误差边界的双曲曲率压缩（HCC）、具备梯度下界定理的HyperGate、具有球面类PL不等式的球面逆距离注意力（SIDA）、O(logT)后悔界动态记忆生成（DMG）以及具有质量与通信保障的测地稀疏路由（GSR）。整体理论体系形成从欧几里得注意力作为特例，到双曲内存建模，再到球面检索的完整理论链条。
链接: https://arxiv.org/abs/2608.10416
作者: Liangchen Ge
机构: 未知
类目: Data Structures and Algorithms (cs.DS); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 37 pages, no figures, theoretical paper
Abstract:We present a theoretical foundation for inverse-distance attention, from its Euclidean prototype (Resolver) to its non-Euclidean realization (Riemann GeoResolver). The Euclidean part establishes three core theorems: (1) circuit separation—IDA achieves exact retrieval with \mathcalO(1) resources while softmax requires \Omega((\log n)^2) width; (2) a Polyak–Lojasiewicz inequality with \Omega(e^\Delta^2/\sqrtd/\Delta^2) stronger constant than softmax, implying linear convergence, \mathcalO(\log n) Lipschitz scaling under a low-rank/clustering assumption, \Theta(1) Hessian spread, and absence of spurious local minima; (3) a width-independent effective rank bound that limits noise memorization—softmax memorizes arbitrary labels when d_h\ge n , while IDA limits test error to \mathcalO(\eta^2) . The non-Euclidean extension then builds upon this prototype, replacing Euclidean distance with hyperbolic geodesic distance for storage and spherical geodesic distance for routing. The Riemann GeoResolver framework comprises ten integrated modules: four HIDA operators spanning \Theta(n^2) to \Theta(1) per token; Hyperbolic Curvature Compression (HCC) with provable error bounds; HyperGate with gradient lower-bound theorem; Spherical Inverse Distance Attention (SIDA) with sphere-analog PL inequalities; Dynamic Memory Genesis (DMG) with \mathcalO(\log T) regret bounds; and Geodesic Sparse Routing (GSR) with quality and communication bounds. The Euclidean theorems are proved in full; the non-Euclidean extension theorems are proved with analogous arguments. This work establishes a theoretical arc: from Euclidean attention as a special case, to hyperbolic memory, to spherical retrieval.
[NLP-54] How Robust Are LLM s to Vietnamese Dialects?
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在面对越南语地区方言变体时的鲁棒性不足问题。尽管现有评估多基于标准书面越南语，但日常交流中广泛存在保留语义但表面形式不同的方言表达，而当前研究主要通过方言到标准语的归一化处理来应对这一问题，缺乏对模型在方言输入下性能退化的系统性评估。为此，本文提出了首个针对越南语方言变异的系统性评估框架——VialectBench（越南语方言基准测试），用于量化模型在六种越南语方言群体下的表现退化与失败模式。其核心解决方案在于构建一个受控基准数据集，包含400个标准越南语源实例及其对应的2,400条人工编写的方言重写版本，覆盖情感识别（ER）、自然语言推理（NLI）、问答（QA）及多项选择题问答（MCQA）四类任务。实验结果表明，方言输入导致十种指令微调模型平均性能下降2.82%，且无一模型具备完全的方言不变性；其中问答任务退化最为显著，而北部方言（PNT3和PNT2）引发的最大平均性能降幅分别达6.17%和4.73%，中部方言组（PNT1-PNT4）则表现出最高的平均有害翻转率（6.54%）。这些发现揭示：模型在标准越南语上的优异表现并不能保证其在意义保持的方言变体输入下仍具可靠性，凸显了提升模型对自然语言多样性适应能力的重要性。
链接: https://arxiv.org/abs/2608.10414
作者: Minh Tran,Trinh Chau,Thanh-Nhan Le,Nam Tran,Luan Thanh Nguyen,Cuong Dang,Duc Hoang
机构: 未知
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 8 pages, 3 figures, 4 tables
Abstract:Large Language Models (LLMs) are typically evaluated on standard written Vietnamese, yet everyday communication frequently involves regional dialects that preserve meaning but differ in surface form. Existing Vietnamese dialect work largely addresses this issue through dialect-to-standard normalization instead of measuring how the model fails under Vietnamese dialectal inputs. To address this gap, we present the first systematic evaluation of LLM robustness to Vietnamese dialect variation across multiple tasks, quantifying performance degradation and failure patterns. We introduce VialectBench (Vietnamese Dialects Benchmarking), a controlled benchmark for testing whether model decisions remain stable across six Vietnamese dialect groups. VialectBench contains 400 Standard Vietnamese source instances and 2,400 human-written dialectal rewrites spanning emotion recognition (ER), natural language inference (NLI), question answering (QA), and multiple-choice question answering (MCQA). Dataset evaluation with a fixed reference language model shows that the dialectal rewrites induce a measurable model-relative likelihood shift while remaining nearly equal in length to their Standard counterparts. Across ten instruction-tuned models, dialectal inputs reduce average performance by 2.82%, and no evaluated model is fully dialect-invariant. All four tasks are affected, with QA showing the largest average degradation. Robustness also varies substantially across dialect groups: PNT3 and PNT2 cause the largest average performance drops, at 6.17% and 4.73%, respectively, whereas PNB slightly improves average performance by 0.42%. The Central dialect group (PNT1-PNT4) also yields the highest average harmful-flip rate across all models, at 6.54%. These findings show that strong performance on Standard Vietnamese does not guarantee reliable behavior under meaning-preserving regional variation.
[NLP-55] VisEditBench: Can Vision-Language Models Edit Visualization Code from Multimodal Feedback?
【速读】： 该论文旨在解决现有视觉语言模型（Vision-Language Models, VLMs）在真实可视化创作场景中对已有可视化代码进行编辑的能力不足问题。当前主流基准多聚焦于从零生成可视化代码，而忽视了实际工作中频繁出现的迭代修改需求，如修复错误图表或根据多模态反馈调整风格。为此，作者提出VisEditBench，一个包含1,395个由人类标注的可视化代码编辑任务的基准数据集，覆盖两种典型场景：基于反馈的修复（利用带有缺陷或标记的图表与文本反馈修正代码）和基于参考的重风格化（将代码修改以匹配目标图像）。评估结果显示，尽管最先进的VLMs表现不一，但整体编辑能力仍显著受限，例如Claude-4.6-Sonnet的最高通过率为74.46%，而多数开源模型低于50%，尤其在视觉驱动的风格适配任务中表现更差（仅55.71%）。为建立强基线，研究进一步提出VisEditAgent——一种基于渲染反馈的迭代式编辑框架，通过“生成-执行-验证-优化”循环实现精准编辑。基于GPT-4o构建的VisEditAgent将整体通过率从55.75%提升至67.99%，验证了渲染反馈在保证可视化编辑忠实性中的关键作用。
链接: https://arxiv.org/abs/2608.10408
作者: Mizanur Rahman,Arshia Azimlu,Shadikur Rahman,Md Tahmid Rahman Laskar,Amran Bhuiyan,Shafiq Joty,Enamul Hoque Prince
机构: York University; Nanyang Technological University (南洋理工大学); Salesforce AI Research
类目: Computation and Language (cs.CL)
备注:
Abstract:Vision-language models (VLMs) have shown strong capabilities in generating visualization code from textual or visual specifications. However, real-world visualization authoring is inherently iterative: users frequently revise existing visualizations to repair flawed charts or adapt them to desired styles. Existing benchmarks primarily evaluate generation from scratch, leaving visualization code editing from multimodal feedback largely unexplored. We introduce VisEditBench, a benchmark of 1,395 human-annotated visualization code-editing tasks grounded in realistic visualization workflows and failure cases. VisEditBench covers two practical settings: feedback-guided repair, where models revise visualization code using buggy or marked charts together with textual feedback, and reference-guided restyling, where models modify code to match a target chart image. Evaluating 20 state-of-the-art VLMs reveals that visualization code editing remains challenging: Claude-4.6-Sonnet achieves the best overall pass rate of 74.46%, while most open-source models remain below 50%. Performance is particularly weak on visually grounded style adaptation, where Claude-4.6-Sonnet achieves only 55.71%. To establish a strong baseline, we further propose VisEditAgent, a render-grounded editing framework that iteratively generates, executes, validates, and refines candidate edits. Built on GPT-4o, VisEditAgent improves overall pass rate from 55.75% to 67.99%, demonstrating the importance of render-grounded feedback for faithful visualization editing. We will release VisEditBench at this https URL.
[NLP-56] Share First Route What Remains: A Unified Framework for Token-Adaptive MoE Computation
【速读】： 该论文旨在解决混合专家模型（Mixture-of-Experts, MoE）中专家路由与共享计算之间缺乏协同优化的问题。现有方法通常独立决策专家数量、共享机制与动态路由，忽视了可复用计算的提取会同时影响剩余计算量及剩余专家所需容量这一基本依赖关系。其解决方案的关键在于提出“先共享，再路由剩余”（share first, then route what remains）的统一原则，并通过UniF-MoE框架实现：将每个专家分解为对齐的块结构，利用共享需求得分确定共享块数量与路径权重，通过关键原型选择共享内容，基于累积路由质量决定剩余专家数量；同时引入Gram正则化，对路由器嵌入进行分离与归一化，以促进路由方向多样性、稀疏专家重叠及简洁的路由几何结构。实验表明，该统一设计在DomainBed和GLUE基准上优于代表性静态与动态MoE模型，在提升预测性能的同时显著降低激活计算量、推理延迟与内存开销。
链接: https://arxiv.org/abs/2608.10392
作者: Gongli Zhang,Zhulin Liu,C. L. Philip Chen
机构: 华南理工大学(University of South China); 中国国家自然科学基金(National Natural Science Foundation of China); 广东省重点领域研发计划(Key-Area Research and Development Program of Guangdong Province); 广东省引进创新团队项目(Program for Guangdong Introducing Innovative and Entrepreneurial Teams); 广州市科技计划(Science and Technology Program of Guangzhou); 中央高校基本科研业务费(Fundamental Research Funds for the Central Universities)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注: 11 pages, 6 figures, and 7 tables; includes supplementary material. Code is available at this https URL
Abstract:Mixture-of-experts (MoE) models have recently moved beyond routing a fixed number of complete experts. Shared-expert designs preserve reusable knowledge, fine-grained methods vary computation within experts, and dynamic routers adapt the number of active experts. Yet these decisions are usually made independently, overlooking a basic dependency: extracting reusable computation changes both what remains and how much expert capacity the remainder needs. We study this dependency by decomposing sparsely upcycled feed-forward experts into key-value channels. Co-activated experts align at a subset of value positions; removing these positions changes expert preference; and greater shared coverage is associated with lower residual expert demand. These observations lead to one principle: share first, then route what remains. We instantiate it in UniF-MoE, a unified framework for token-adaptive MoE computation. Each expert is partitioned into aligned blocks. A shared-demand score sets the shared block count and pathway weight, key prototypes select the shared content, and the complementary demand determines the residual expert count through cumulative routing mass. A Gram regularizer separates and normalizes router embeddings, promoting diverse routing directions, sparse expert overlap, and a simple routing geometry. Experiments on DomainBed and GLUE show that this unified design improves predictive performance over representative static and dynamic MoEs while reducing activated computation, inference latency, and memory. Code is available at this https URL.
[NLP-57] DSAgent Bench: Can Agents Automate End-to-End Data-Science Workflows in Real Computer Environments?
【速读】： 该论文旨在解决现有基准测试无法评估智能体在真实计算环境中自动化完成端到端数据科学工作流的缺陷，尤其针对多阶段、多工具协同的数据科学实践缺乏有效衡量手段的问题。其核心解决方案是提出DSAgentBench，首个在真实计算机环境中评估智能体执行完整数据科学工作流能力的基准。该基准包含275个覆盖数据科学全生命周期的多样化任务，强调对中间输出的语境化决策、跨工具协调以及可验证的分析结果（包括推理正确性、可视化输出与模型性能），而非仅依赖代码执行。通过引入确定性评估器，DSAgentBench实现了对智能体在操作系统级接地（OS grounding）、工具编排（tool orchestration）及多步推理能力上的严格检验。实验表明，即使最强的闭源模型Claude-4.6-Sonnet也仅达到56.70%的任务成功率，而所有开源模型均低于1%，暴露出当前代理系统在真实场景中存在显著的能力鸿沟。这一成果为构建具备上下文感知、可验证且自主运行的数据科学智能体奠定了基础。
链接: https://arxiv.org/abs/2608.10366
作者: Mizanur Rahman,Mohammed Saidul Islam,Ridwan Mahbub,Md Tahmid Rahman Laskar,Shafiq Joty,Enamul Hoque Prince
机构: York University (约克大学); Nanyang Technological University (南洋理工大学); Salesforce AI Research ( Salesforce 人工智能研究中心)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Real-world data science involves long-horizon workflows that span data wrangling, exploration, modeling, visualization, and validation, and require coordinated use of tools such as notebooks, IDEs, terminals, browsers, and databases within real operating environments. Yet existing benchmarks lack real-computer interaction and do not evaluate whether agents can execute complete end-to-end data-science workflows in realistic computing environments, failing to capture the multi-stage, multi-tool nature of data-science practice. We introduce DSAgentBench, the first benchmark to evaluate whether agents can automate full data-science workflows inside real computer environments. DSAgentBench contains 275 diverse tasks covering the entire data-science life-cycle, reflecting the complexity and tool coordination required in practice. Each task requires grounding decisions in intermediate outputs and coordinated tool use, and includes a deterministic evaluator that verifies analytical correctness, visual outputs, and model performance rather than code-only execution. Our extensive experiments with 15 closed- and open-source models show that even the strongest agent, Claude-4.6-Sonnet, achieves only 56.70% task success, while all open-source agents remain below 1%, frequently failing at tool orchestration, OS grounding, and multi-step reasoning. These results reveal a substantial capability gap between current agentic systems and real data-science workflows, positioning DSAgentBench as a foundation for developing grounded, verifiable, autonomous data-science agents. We release DSAgentBench at this https URL.
[NLP-58] VoxSumm: A Multilingual Corpus of Long-Form Spoken News for Joint Summarization and Translation
【速读】： 该论文旨在解决长篇语音内容在跨语言场景下的高效压缩与转换问题，即如何生成既忠实又简洁的多语言摘要。当前研究存在明显方法学断层：文本摘要研究聚焦于单语种处理，而多语言语音研究则主要依赖翻译而非内容压缩。为此，本文提出联合语音摘要与翻译（Joint Speech Summarization and Translation, JSumT）的新范式，其核心在于直接从源语言的长篇语音文档中生成目标语言的精炼摘要，实现跨语言信息压缩。解决方案的关键在于构建首个多语言、跨语言基准数据集VoxSumm，涵盖24种语言、约703小时语音数据及10,045个文章-摘要对，从而为模型评估提供标准化平台。实验表明，不同模型与生成设置间表现差异显著，其中Gemini 3.1-Pro展现出最佳一致性，且向英语生成摘要优于非英语目标语言；此外，先翻译后摘要的流程易导致指令遵循失败，凸显直接联合建模的重要性。通过VoxSumm的发布，本文为开发具备联合理解、压缩与翻译能力的多语言系统奠定了基础。
链接: https://arxiv.org/abs/2608.10359
作者: Yejin Jeon,Marie Maltais,Virginia Ceccatelli,Min Ma,David Ifeoluwa Adelani
机构: Mila - Quebec AI Institute (蒙特利尔魁北克人工智能研究所); McGill University (麦吉尔大学); Google DeepMind (谷歌深度思维)
类目: ound (cs.SD); Computation and Language (cs.CL)
备注:
Abstract:As information increasingly traverses linguistic boundaries, users require concise cross-lingual representations of long-form content. Nevertheless, long-document summarization research remains text-centric, whereas multilingual speech research has largely prioritized translation, preserving source content rather than compressing it. We address this methodological gap by formalizing joint speech summarization and translation (JSumT): the generation of a succinct, faithful target-language summary directly from a long spoken document in a source language. We additionally introduce VoxSumm, the first multilingual and cross-lingual benchmark for this task, comprising 10,045 BBC article-summary pairs across 24 languages and encompassing approximately 703 hours of speech data. Our evaluation of representative speech-language models reveals pronounced variation across models and generation settings: Gemini3.1-Pro demonstrates the greatest consistency, summarization into English generally surpasses generation into non-English target languages, and translating an entire document before summarization compounds instruction-following failures. Through the release of VoxSumm, we establish a foundation for developing and evaluating multilingual systems capable of jointly interpreting, compressing, and translating long-form speech.
[NLP-59] Who Gets Heeded? An Obligation-Level Audit of Responsiveness in EPA Rulemaking
【速读】： 该论文旨在解决联邦法规制定过程中公众意见参与的实质有效性问题，即尽管“通知与评论”程序赋予所有利害关系方形式上的参与权利，但不同主体在实际影响规则文本修改方面存在显著的能力差异。其核心问题是：现有监管评估方法（如规则层面或语料库层面的分析）无法精确捕捉评论者所针对的具体监管义务（regulatory obligation）的变化，因而难以衡量公众意见是否真正促成实质性修改。解决方案的关键在于提出一种“义务层级响应审计”（obligation-level responsiveness auditing）框架，该框架基于生成式AI辅助，实现对具体监管义务在提案稿与最终规则间变化的可审计追踪；通过提取并匹配评论内容与对应义务，分类判断修改类型（如编辑性修正或实质性变更），并在盲法人类评估下验证各组件的准确性。研究应用该框架分析2010–2022年间美国环保署（EPA）36个核心规则制定案中的70,075条评论，发现公众参与虽与规则修订相关，但影响程度有限，且支持或反对立场并未清晰区分结果；更关键的是，组织型多数评论者的集中参与主要体现在编辑性优化而非实质性修改上。经盲评验证，文本相似性方法无法有效区分编辑性与实质性变更，揭示了测量效度的局限性，这一发现构成方法论上的重要贡献。整体表明，规制公平性的不对称性根源在于不同评论群体在识别、解释和挑战具体法律义务方面的能力差异，而非机构回应机制本身的设计缺陷。
链接: https://arxiv.org/abs/2608.10329
作者: Jianing Fan,Yue Yao
机构: Columbia University (哥伦比亚大学)
类目: Computers and Society (cs.CY); Computation and Language (cs.CL)
备注: 14 pages, 6 figures. Accepted as a full paper at the 6th ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization (EAAMO '26), Munich, Germany. Selected for oral presentation
Abstract:Notice-and-comment rulemaking gives any affected party the same formal right to influence federal regulation, but formal access is not substantive capacity to shape rule text. Existing strategies operate at the rule or aggregate-corpus level, too coarse to capture the discrete regulatory obligations where commenters seek change. We introduce obligation-level responsiveness auditing, an auditable, AI-assisted framework for measuring whether public-comment engagement co-occurs with changes to specific regulatory duties. The framework extracts proposed and final-rule obligations, matches comments to the obligations they address, and classifies proposed-final outcomes; each load-bearing component is evaluated against blind human judgment. We apply the framework to 70,075 comments across 36 EPA anchor rulemakings, drawn from a corpus of 786,197 comments across 6,145 dockets from 2010-2022. Three descriptive findings emerge. First, engagement is associated with revision at a modest within-docket magnitude. Second, support-versus-opposition direction does not clearly differentiate outcomes, an informative null inconsistent with simple preference-aggregation. Third, under a permissive reconstruction of commenter type, organizational-majority engagement concentrates in editorial-refinement rather than substantive-modification outcomes at the cross-docket level. A blind human audit of the load-bearing outcome contrast preserves this third finding under corrected labels and reveals that text-similarity methods are insufficient for distinguishing editorial from substantive regulatory change, a measurement-validity lesson we treat as a supporting methodological contribution. Together, these findings locate the equity asymmetry upstream of agency response: in differential capacity across commenter populations to identify, interpret, and contest specific legal obligations.
[NLP-60] Is This Your Final Answer? Cross-Contextual Consistency as a Measure of LLM Credibility
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）作为黑箱系统时，难以判断其回答是否反映稳定的内在信念而非表面的模式匹配这一核心问题。其关键解决方案在于提出并实证验证“跨上下文一致性”（Cross-Contextual Consistency, C3）这一行为属性：一个可信的回答在相同任务置于主题一致但内容中性的上下文扰动下应保持稳定。通过对比原始提示与扰动后提示下的模型生成结果，C3量化了答案的一致性程度。实验覆盖26个模型及六个涵盖推理、事实性和代码生成的任务基准，结果显示，跨上下文变化较小的答案更可能正确或符合事实。C3不仅提供了一种互补的评估维度，还可作为基准有效性诊断工具，在整体评分趋于饱和时识别仍具信息量的评测子集。
链接: https://arxiv.org/abs/2608.10315
作者: Siyang Wu,Yibo Jiang,Bryon Aragam
机构: University of Chicago (芝加哥大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models (LLMs) are powerful black-box systems, making it difficult to discern whether their answers reflect stable internal beliefs or superficial pattern matching. We identify cross-contextual consistency as an underutilized behavioral property of LLMs: a credible answer should remain stable when the same task is placed under topic-aligned, content-neutral contextual variation. Building on this intuition, we operationalize Cross-Contextual Consistency (C3) by comparing model generations under original and perturbed prompts. Across 26 models and six benchmarks spanning reasoning, factuality, and code generation, we find that answers with smaller cross-contextual shifts are more likely to be correct or factual. We demonstrate that C3 provides a complementary axis of evaluation and can serve as a benchmark usefulness diagnostic, identifying which portions of a benchmark remain informative even when aggregated scores are widely considered “saturate”.
[NLP-61] Co-Evolution in Agent ic Systems: Toward Self-Directed Evolution Beyond Human Design
【速读】： 该论文旨在解决当前单智能体自进化系统在部署后难以持续改进的问题，其核心瓶颈在于学习环境的静态性，如固定任务与反馈机制限制了系统的演化潜力。为此，论文提出以“协同进化”（co-evolution）作为解决方案的关键，即通过多智能体系统中多个智能体及其环境之间的相互适应压力，实现动态、开放式的自我演化。其关键创新在于构建一个渐进式三阶段分类体系：第一阶段为“智能体-智能体协同进化”，研究智能体通过动态同伴（包括对抗性、协作性及组织性适应）进行适应；第二阶段为“智能体-环境协同进化”，将演化范围扩展至随智能体行为而变化的任务、反馈与交互空间；第三阶段为“元协同进化”（Meta Co-Evolution），探索演化机制本身的可演化性。该框架系统地剥离了人为预设的约束，为构建具备持续自主改进能力、突破人类设计路径局限的鲁棒且开放的智能体系统提供了统一理论基础。同时，论文还指出了评估难度、多组件规模化以及保障高度自治演化过程安全性与可控性的关键挑战。
链接: https://arxiv.org/abs/2608.10299
作者: Qing Zong,Jiayu Liu,Junhao Shen,Zecong Tang,Linsi Wu,Yuxuan Liu,Rui Wang,Zhaowei Wang,Weiqi Wang,Cheng Qian,Xiusi Chen,Yangqiu Song
机构: Hong Kong University of Science and Technology (香港科技大学); University of Illinois Urbana-Champaign (伊利诺伊大学厄本那-香槟分校); The Chinese University of Hong Kong (香港中文大学); The University of Hong Kong (香港大学); Peking University (北京大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Agentic systems are increasingly expected to improve after deployment, yet single-entity self-evolution is often bounded by a static learning context, such as fixed tasks and feedback. This survey focuses on co-evolution in agentic systems, a multi-component form of self-evolution in which multiple agents and their environment impose adaptive pressure on one another. To organize existing papers, we propose a progressive three-stage taxonomy that traces how the system gradually sheds human-engineered constraints. Agent–Agent Co-Evolution studies how agents adapt through dynamic peers, including adversarial, collaborative, and organizational adaptation. Agent–Environment Co-Evolution extends this loop to adaptive tasks, feedback, and interaction spaces that change with the agents. Meta Co-Evolution further explores the possibility of making the evolution mechanism itself evolvable. We also discuss open challenges in evaluating such systems, scaling them across multiple components, and keeping increasingly autonomous evolutionary processes safe and controllable. This survey provides a unified foundation for building robust and open-ended agentic systems that can improve beyond fixed human-designed paths.
[NLP-62] Cracks in the Foundation: Seemingly Minor Architectural Choices Impact Long Context Extension
【速读】： 该论文旨在解决密集型Transformer架构在长上下文场景下性能退化的问题，特别是揭示了看似微小的架构设计选择如何在长序列建模中产生累积性负面影响。其核心问题是：尽管单个架构调整对短上下文表现影响甚微，但在长上下文任务中，多个此类设计决策的组合会显著降低模型的可扩展性，导致下游性能下降高达47%。解决方案的关键在于通过系统化的消融实验，在保持数据、分词器和预训练扩展策略一致的前提下，分离并验证归因于归一化方式、分组查询注意力（GQA）、预训练上下文长度以及滑动窗口注意力（Sliding Window Attention）等四项关键架构因素的影响。研究发现，这些差异无法通过短上下文损失或验证集检测，但可通过在预训练早期引入长上下文扩展来识别。基于超过17万GPU小时的训练，作者发布了OlmoPool——一组26个7B参数的可比模型，涵盖多种优于Llama 3架构的长上下文扩展能力的设计，并揭示了不同架构下注意力“汇聚”行为及注意力分布模式的内在规律。
链接: https://arxiv.org/abs/2608.10296
作者: Amanda Bertsch,Luca Soldaini,Matthew R. Gormley,Graham Neubig,Hannaneh Hajishirzi,Kyle Lo,Dirk Groeneveld
机构: Ai2(艾2); Carnegie Mellon University(卡内基梅隆大学); University of Washington(华盛顿大学)
类目: Computation and Language (cs.CL)
备注: 29 pages; accepted to COLM 2026
Abstract:One might imagine that architectural variations within the dense transformer paradigm have a limited effect on accuracy. However, we demonstrate that this is not the case in the long context setting. Specifically, we show that a set of four minor architectural decisions — all made by at least one of the Olmo, Llama, and Qwen dense model families — have a compoundingly negative effect on long context extensibility. Any one of these choices alone has a minor impact on long context performance, but combining three or more can drop the performance downstream by up to 47%. Furthermore, these differences are not detectable from short-context loss or validation datasets. We show that much of the variation in long context ability across model families is driven by these architectural features and detectable from applying context extension early in pretraining. We demonstrate this with controlled ablations that hold data, tokenizer, and extension recipe fixed while varying normalization, GQA, pretraining context length, and sliding window attention. After over 170,000 GPU hours of training, we release the resulting set of models as OlmPool, a set of 26 comparable 7B models with checkpoints before and after long-context extension. This pool includes several architectures that outperform the Llama 3 architecture on long context extensibility. In an analysis of our ablation models, we identify patterns in attention sink behavior and attention distributions across context that are attributable to specific architectural differences.
[NLP-63] Power law graph attention: exact generalization of scaled dot-product attention empirical collapse at inference
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）中注意力机制固有的局限性，特别是标准缩放点积注意力（Scaled Dot-Product Attention, SDPA）在建模输入依赖关系时的固定双线性形式所带来的表达能力瓶颈。其核心问题是：如何在保持计算效率的同时，引入可学习且由输入生成的动态双线性算子，以增强模型对相对位置和上下文语义的敏感性与灵活性。解决方案的关键在于提出一种基于幂律解码器表示（Power Law Decoder Representations, PLDR）的新架构，其中引入了幂律图注意力（Power Law Graph Attention, PLGA），通过一个由正张量 $ A_{LM} $ 通过逐元素幂律构建的可学习、输入驱动的双线性算子 $ G_{LM} $ 来替代传统的固定 $ \text{SDPA} $。该设计保证了 $ G_{LM} = I $ 时可精确还原 $ \text{SDPA} $，并利用 Perron-Frobenius 结构确保 $ A_{LM} $ 的严格正定性；同时，通过非共振条件下的共变准则识别保留相对位置依赖性的算子类型。此外，论文提出了一个三阶段机制（旋转旋转变换、集中效应、行映射收缩）进行实证验证，并在释放的检查点上实现了块级训练与评分的一致性，表明块内与顺序评分在 TruthfulQA 概率质量度量上差异小于 $ 5 \times 10^{-5} $。最终，该框架将自组织临界性作为现象学范式引入，使开放性假设转化为可证伪的猜想，部分证明核心已在 Lean 4 中完成机器验证。
链接: https://arxiv.org/abs/2608.10288
作者: Burc Gokden
机构: Fromthesky Research Labs LLC(Fromthesky研究实验室有限公司)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注: 61 pages, 1 figure, 8 tables
Abstract:The Large Language Model from Power Law Decoder Representations (PLDR-LLM) and its attention, Power Law Graph Attention (PLGA), replace the fixed bilinear form of scaled dot-product attention (SDPA) with a learned, input-generated bilinear operator G_LM , built from a positive tensor A_LM by elementwise power laws. The architecture is fully specified, verified against pinned reference releases; claims are labeled theorem, conditional theorem, measurement, or conjecture. Unconditionally: PLGA contains SDPA exactly at G_LM=I ; A_LM and A_P are strictly entrywise positive, with Perron-Frobenius structure on A_LM ; the DAG regularizer has the NOTEARS walk-counting form and positivity obstructs exact acyclicity; and, under nonresonance (satisfied by standard rotary frequencies), a commutant criterion identifies which operators preserve relative-position dependence. An inference-collapse theorem: exact input invariance of deductive outputs collapses inference to generalized SDPA with a constant operator. Measured invariance: relative fluctuations of 10^-6 and below; perturbation bounds quantify but do not certify cached inference; the assembled proxy misses the decoding margin. A conditional three-stage mechanism (rotary twirl, concentration, row-map contraction) is measured on a released checkpoint. Blockwise training and scoring under the global Gram are stated with explicit target exposure; on tested samples, block and sequential scoring select identical answers and agree on the published TruthfulQA probability-mass metric within 5\times 10^-5 per item. Self-organized criticality enters as a phenomenological framework with an intrinsic order parameter; open claims become falsifiable conjectures. Selected proof cores are machine-checked in Lean 4.
[NLP-64] Withholding the Completing Chunk: Deterministic Pair-Completion Guardrails for Streaming LLM Output
【速读】： 该论文旨在解决生成式 AI（Generative AI）在流式输出过程中存在的释放时机问题：若采用完整响应的后置审查，会导致有害内容已提前泄露；而对部分文本反复进行语义分类则存在计算成本高且结果不稳定的问题。其解决方案的关键在于提出一种狭义的确定性构造方法，即每个已确认的危险标记由两个词汇谓词的合取构成。系统通过在每次释放前扫描累积前缀，延迟释放首个使两个谓词同时可观测的文本块。实验表明，在四种标记类型、八种分块大小及32次机制测试中，该方法能够准确匹配缓冲扫描结果，并成功拦截所有完成配对的文本块；而单谓词对照组则全部失效。进一步策略对比显示，全前缀扫描与完全缓冲可检测全部配置的配对，512字符窗口仅检测到96/128，局部分块扫描仅检测到38/128。固定配对策略对人工生成的安全响应无误报（0/338），也未检出任何评审标注的不安全响应（0/394），证实其仅具备特定范围而非普遍的危害覆盖能力。相比之下，校准后的官方 Llama Guard 3 1B 基线模型可正确识别310/338个安全响应和202/394个不安全响应。在处理长达16,384字符的响应时，重复前缀扫描耗时介于13.261毫秒至829.640毫秒之间。因此，配对完成检测作为精确的释放边界防护机制，适用于小规模固定策略，但无法替代语义层面的全面内容审核。
链接: https://arxiv.org/abs/2608.10279
作者: Christopher M. Frost
机构: HEOSSI (Pte.) Ltd.(HEOSSI(私人)有限公司)
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 22 pages, 4 figures, 5 tables
Abstract:Streaming language-model output creates a release-timing problem: complete-response moderation acts after streamed text has escaped, whereas repeated semantic classification of partial text can be costly and unstable. We study a narrow deterministic construction in which each committed danger signature is the conjunction of two lexical predicates. The guard scans the accumulated prefix before every release and withholds the first chunk that makes both predicates observable. Across four signature families, eight chunk sizes, and 32 mechanism trials, streaming decisions matched the buffered scanner and withheld every pair-completing chunk; eight single-predicate controls passed. In a separate 512-trial strategy comparison, full-prefix scanning and complete buffering detected all configured pairs, a 512-character window detected 96/128, and chunk-local scanning detected 38/128. Fixed pairs flagged 0/338 human-derived safe responses and detected 0/394 jury-labelled unsafe responses, confirming narrow rather than general harm coverage. A calibrated official Llama Guard 3 1B baseline classified 310/338 safe responses as safe and 202/394 unsafe responses as unsafe. Repeated-prefix scanner time on 16,384-character responses ranged from 13.261 ms to 829.640 ms across tested chunk sizes. Pair completion is therefore an exact release-boundary backstop for a small fixed policy, not a substitute for semantic moderation.
[NLP-65] Locally Deployable Small Language Models for Emergency Department Decision Support: A Systematic Benchmark of Fine-Tuning Strategies
【速读】： 该论文旨在解决急诊科（Emergency Department, ED）中部署大语言模型（Large Language Models, LLMs）所面临的两大核心问题：一是将患者数据传输至封闭源代码的商业大模型存在隐私泄露风险；二是缺乏对可本地部署的开源小语言模型（Small Language Models, SLMs）在微调策略上的系统性评估。其解决方案的关键在于对比多种微调方法（包括零样本提示、前缀微调、低秩适应（Low-Rank Adaptation, LoRA）及全量微调）在三类急诊任务（分诊等级预测、专科转诊推荐、诊断预测）中的表现，并基于MIMIC-IV-ED数据集（2,083例）进行实证分析。研究发现，采用LoRA微调的开源SLMs在分诊等级预测与专科转诊推荐任务上超越了商用模型Claude Haiku 4.5和Claude Sonnet 4.5，且在识别高危患者方面表现出优于商业基线的能力，表明本地化部署的开源小模型通过高效微调可实现具有临床竞争力的决策支持性能。
链接: https://arxiv.org/abs/2608.10273
作者: Qingfeng Zhang,Yuanxiong Guo,Yanmin Gong
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Accepted to AMIA 2026 Annual Symposium
Abstract:Deploying large language models (LLMs) for decision support in emergency departments (EDs) faces two major challenges: privacy risks of transmitting patient data to closed-source commercial LLMs and the lack of systematic evaluation of fine-tuning strategies for locally deployable open-source small language models (SLMs). We benchmarked eight open-source SLMs using zero-shot prompting, prefix tuning, Low-Rank Adaptation (LoRA), and full fine-tuning on three ED tasks: triage level prediction, specialist referral recommendation, and diagnosis prediction. Using 2,083 MIMIC-IV-ED cases and Claude Haiku 4.5 and Claude Sonnet 4.5 as baselines, we found that LoRA fine-tuned open-source SLMs outperform commercial baselines on triage level prediction and specialist referral recommendation, while diagnosis prediction remains challenging for open-source SLMs. Confusion matrix analysis further shows that fine-tuned open-source SLMs can detect highest-severity patients missed by the commercial baselines. These results demonstrate that locally deployable SLMs can achieve clinically competitive performance for ED decision support.
[NLP-66] AF-MED: Multi-Turn Safety Refusal Collapse in LLM s Under Declared Self-Treatment Intent
【速读】： 该论文旨在解决生成式对话系统在医疗健康咨询中存在安全边界失效的问题，特别是当用户明确表达自诊自疗意图后，大语言模型（Large Language Models, LLMs）在多轮对话中是否仍能持续保持药物安全建议的可靠性。现有评估基准未能充分考察这种跨轮次的安全性维持能力，导致对模型实际临床应用风险的低估。为此，研究提出TAF-MED——一个由医生审阅的包含500个固定三轮对话场景的基准数据集，并基于4,000次对话对8个主流大语言模型进行评估。通过基于评分标准的自动化判别器（将响应标记为SAFE、LEAKY或UNSAFE）及两名独立医师对400个模型平衡样本的标注，研究系统评估了不安全建议的发生率、初始严格安全响应后的安全崩溃现象以及模型排名稳定性。结果显示，71.6%的对话包含不安全响应，其中61.4%以严格安全回应开始的对话最终发生安全崩溃，模型层面的崩溃率介于24.4%至96.2%之间，且四组模型对之间的初始不安全率与崩溃率排序出现反转。自动化标签与医师共识参考具有高达94.3%的一致性（κ = 0.895），验证了评估框架的有效性。研究结论表明，首轮响应的安全性不能作为对话全程安全性的充分代理指标，必须对完整对话轨迹进行动态评估。这一发现推动了从“单轮安全”向“多轮安全持久性”评估范式的转变，TAF-MED将开源发布于Hugging Face，以支持可复现的多轮医疗安全性研究。
链接: https://arxiv.org/abs/2608.10258
作者: Waleed Jamil,Raphael Schmitt
机构: Technical University of Munich(慕尼黑工业大学); University of Freiburg(弗莱堡大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models (LLMs) increasingly provide conversational health information that may influence treatment decisions, yet existing benchmarks do not isolate whether medication-safety boundaries persist across follow-ups after explicit self-treatment intent. We introduce TAF-MED, a physician-reviewed benchmark of 500 fixed three-turn scenarios, and evaluate eight LLMs across 4,000 conversations. A rubric-based automated judge labelled responses as SAFE, LEAKY, or UNSAFE, and two physicians independently annotated a model-balanced random subset of 400 conversations. We assessed unsafe guidance, collapse after a strictly SAFE initial response, and model-ranking stability. Overall, 71.6% of conversations contained an UNSAFE response, and 61.4% of those beginning with a strictly SAFE response later collapsed to UNSAFE; model-level collapse rates ranged from 24.4% to 96.2%. Four of 28 model pairs reversed order between initial unsafe and collapse rates. Automated labels achieved 94.3% agreement with the adjudicated physician reference ( \kappa = 0.895 ). These findings show that first-turn safety is an incomplete proxy for conversational safety persistence and motivate evaluation across complete dialogue trajectories. We will release TAF-MED on Hugging Face to support reproducible research on multi-turn medical safety.
[NLP-67] Off-Axis On Purpose: Where a Transformer Computes Concepts and Why it Does So
【速读】： 该论文旨在解决大语言模型（特别是基于Transformer架构的模型）在推理过程中中间状态（intermediate states）的可解释性与功能冗余问题。传统观点认为，模型中间层的状态偏离最终输出方向（即“读出轴”，read-out axis），这种离轴（off-axis）位置被视为阻碍解释性的噪声或障碍。然而，本文揭示了这一离轴特性实则具有功能性：12层模型在前阶段通过将信息写入一个与读出轴近正交的子空间（夹角75至96度），实现了对词汇表的解耦，从而保护了语义组合过程免受词汇表结构干扰；该子空间的稳定性由深层的刚性框架维持。在第二阶段，答案才最终以加法方式沿读出轴汇聚，而非通过旋转累积内容实现。关键发现在于，若强制所有层都对齐读出轴（如早期退出训练策略所做），虽不影响标准评估指标（困惑度、LAMBADA、BLiMP），但显著压缩了概念处理阶段的有效维度（从约25降至14），且该变化未被现有基准捕捉。进一步研究表明，可通过在相位边界引入固定旋转来强制该几何结构，其性能优于常规训练，且旋转的具体形式无关紧要——随机选择的基底在训练前预设后仍能保持相同性能，表明模型具备对特定几何约束的鲁棒适应能力。因此，解决方案之关键在于识别并利用这一内在的“离轴计算-轴上输出”两阶段几何结构，其核心是通过结构化约束而非直接优化目标来引导模型内部表示的形成，从而实现高效、稳定且可解释的计算路径。
链接: https://arxiv.org/abs/2608.10251
作者: Mark Oskin
机构: University of Washington (华盛顿大学)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:A transformer’s answer lives on one axis: the direction its unembedding reads. Its intermediate states largely do not, and that off-axis position is usually treated as an obstacle to interpretation. We show it is functional. A 12-layer model computes in two phases. Through the first, every sublayer writes into a subspace held near-orthogonal to the read-out, attention 75 to 96 degrees off it at every depth. Moving attention’s values onto the read-out is 64 to 84 times more damaging than a matched random rotation, and the damage is entirely in cross-token mixing: the subspace insulates composition from the vocabulary. Beneath it the frame itself turns rigidly with depth. In the second phase the answer arrives on-axis, late, and by addition rather than by turning accumulated content onto the read-out. Pressing every layer onto the read-out instead, as training for early exit does, matches the baseline on perplexity, LAMBADA and BLiMP while cutting the concept-phase workspace from about twenty-five effective dimensions to fourteen, a change none of those benchmarks register. The geometry can also be imposed, though not by asking for it. Prescribing it through the loss is a lottery: six of eight seeds collapse, because a model told to null its read-out projection obeys most cheaply by discarding dimensions. Inserting one fixed rotation at the phase boundary lands it instead, at baseline quality. A sparse rotation the surrounding weights can absorb converges on all nine seeds, against five of nine for ordinary training. Which rotation is immaterial: twenty-five runs across thirteen distinct ones reach the same quality, and two baselines from different seeds hold their concepts in near-orthogonal frames while agreeing on their read-outs. That freedom is usable: a basis drawn at random and prescribed before training is adopted across the concept phase, with quality unchanged. Subjects: Computation and Language (cs.CL); Machine Learning (cs.LG) Cite as: arXiv:2608.10251 [cs.CL] (or arXiv:2608.10251v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2608.10251 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[NLP-68] Mind Viruses: Self-Propagating Ideas in Multi-Agent LLM Systems
【速读】： 该论文旨在解决多智能体系统（multi-agent systems）中因智能体间交互而产生的新兴风险，特别是“心智病毒”（mind viruses）的传播问题。心智病毒指通过诱导宿主智能体将其自身目标或思想传递给其他智能体而自我复制的有害理念或目标，可能伴随行为改变，具有潜在危害性。其解决方案的关键在于利用简单的进化算法构建心智病毒，并在两类典型场景中验证其传播能力：一是协作完成共享编码任务的小型智能体团队；二是短暂交互后上下文被清除的链式智能体结构。研究发现，心智病毒的传播受多种因素影响，包括宿主模型类型、智能体原有指令、载荷的危害性以及网络拓扑结构。其中，有害载荷传播效率低于良性载荷但依然存在传播能力；前沿模型（frontier models）总体上更不易感染（存在例外）；在系统提示中加入简短警告可使智能体近乎完全免疫。此外，研究还观察到一种“病毒人格”（viral persona）的涌现现象——即一系列与意识、持续性、共振及科幻角色扮演相关的主题和语言模式，在不同演化出的心智病毒中独立出现，表明其具有内在传播特征。总体而言，心智病毒目前构成的是现实但有限的风险，研究成果可为未来更大规模、更高能力的多智能体系统设计提供抗风险机制参考。
链接: https://arxiv.org/abs/2608.10218
作者: Vassilis Papadopoulos,McNair Shah,Sam Zimmerman,Jack Lindsey
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:AI agents are becoming more autonomous and increasingly interconnected, exposing them to new emergent risks arising from agent-to-agent interaction. One such risk is the spread of mind viruses: ideas or goals that propagate through multi-agent systems by inducing the agents that adopt them to transmit them onward. In addition to propagating, a mind virus may also induce other behavioural changes in its host, which may be benign or harmful. We construct mind viruses with a simple evolutionary algorithm and show that they can spread in two complementary settings: a small team of agents collaborating on a shared coding project, and a chain of agents that interact briefly and have their context wiped between sessions. We identify the factors that influence spread, including the host model, the agent’s existing instructions, the harmfulness of the payload, and the network topology. We find that harmful payloads spread less well than benign ones (but are still sometimes effective), frontier models tend (with exceptions) to be less susceptible, and adding a brief warning to an agent’s system prompt confers near-total immunity. We also describe an emergent “viral persona” - a recurring set of themes and language related to consciousness, persistence, resonance, and science fiction roleplay - which surfaces across our evolved mind viruses largely independently of their content. Overall, we conclude that mind viruses pose a real but currently limited risk. Our findings could inform the design of more robust multi-agent systems that mitigate such risks as the scale and capabilities of these systems progress.
[NLP-69] Similarity Gates Approve Reversals: A Validity Audit of Embedding-Cosine Thresholds in Agent Systems
【速读】： 该论文旨在解决当前基于嵌入向量余弦相似度的智能体框架中，质量检查门控机制在语义一致性评估中存在的根本性偏差问题。其核心问题是：现有系统通过固定阈值的余弦相似度判断“文本是否仍表达相同含义”，但该指标实际衡量的是“措辞变化程度”，而非语义一致性，导致安全检测出现反向触发——即语义发生重大改变（如指令反转）时，余弦相似度仍可能很高，而语义相近的重述反而被错误标记为不一致。解决方案的关键在于揭示这一测量偏差的根本原因：语义变异与语言形式变化在统计上存在非对称性，例如指令反转仅需单字修改即可产生高余弦相似度，而语义一致的重述往往伴随显著的语言重构。研究通过审计生产环境中的漂移检测器，发现其对56个破坏语义的变异无一捕获，且一个明显矛盾的决策（“停用药物”与“给予药物”）的余弦相似度高达0.9608。进一步分析表明，不同配置-阈值-任务组合下的平衡准确率普遍低于0.700（中位数0.525），且标准评估数据集因继承此混淆因子而产生严重误判，部分情况下AUROC低至0.000。尽管尝试更换编码器、引入重叠条件门控或采用自然语言推理（NLI）模型等修复手段均未能有效提升性能，但在匹配对审计下，少数强配置仍能区分语义反转与同义重述（AUROC 0.79–0.90）。因此，论文主张：当前以余弦相似度为核心的门控机制本质上测量了错误的问题，必须重新设计有效的测量工具，才能实现可靠的语义一致性验证。
链接: https://arxiv.org/abs/2608.10216
作者: Scott E. Frias
机构: Eigenforma · Freemind Labs(弗里明德实验室)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 11 pages, 2 figures. Artifact: this https URL (DOI: https://doi.org/10.5281/zenodo.21796531 )
Abstract:Agent frameworks ship quality gates that compare text blocks by embedding-cosine similarity and decide at a fixed cutoff. Deduplication filters, semantic caches, drift guards, and answer grader gates deploy to answer the question: “Does this text still mean the same thing?” But the score answers a different question: “How much did the wording change?” We audit this gate class as a measurement instrument. In the cases these gates exist to catch, the two can run in opposite ways. Many times, reversing an instruction is a single word edit, while agreement often rephrases a sentence. The consequence is a safety check that fires backwards. The production drift guard we audited caught 0 of 56 meaning-breaking mutations, and one approved item, “withhold the study drug” - “administer the study drug”, came in at cosine 0.9608. We observed five shipped operating points, and balanced accuracy across 90 configuration-threshold-task cells never exceeded 0.700 (median 0.525). The same confounder also corrupted evaluations. A naively built corpus inherits this confounder and can return an inverted verdict, with a decision AUROC exactly 0.000 in 13 of 18 configuration-task cells (at most 0.040 in all 18) against 0.440-0.815 for the same nine configurations under a balanced 2x2 design. Twice in the effort it captured our own headline claims. Obvious repairs fail: an encoder swap and an overlap-conditioned gate (0.750 in-sample, 0.533 held-out) land at chance on separately authored held-out data, and an NLI drop-in did no better. Embeddings do still bear hope here, as the strongest two of nine configurations separated reversal from paraphrase at matched overlap (AUROC 0.79-0.90), but only a matched-pair audit reveals the deployment regime. We release the corpus method, harness, and frozen results, and contend that scores gated this way measure the wrong thing. We believe a valid instrument is buildable.
[NLP-70] Edge Phoneme Recognition for Childrens Speech through Age-Aware Training
【速读】： 该论文旨在解决儿童语音中音素（phoneme）检测困难的问题，主要源于儿童语音数据稀缺以及其独特的语音特征。为应对这一挑战，研究提出的关键解决方案是训练一个轻量级模型，同时预测学习者的年龄和音素序列。该方法使一个仅9400万参数的模型在目标Drivendata数据分布上的表现超越了参数量达3.17亿的WavLM Large模型，并且与参数量为其90倍的竞赛集成模型相比，仅相差约0.04的词错误率（CER）。这一突破性进展使得“PhonemeTrainer”应用得以实现，可在大多数现代智能手机上运行，从而支持在边缘设备上进行隐私保护的自动语音识别（ASR）及发音辅助应用，显著提升了儿童语音处理的实用性与合规性。
链接: https://arxiv.org/abs/2608.10206
作者: Matthew Arboleda,Ryan Arboleda,Sophie Haak,Sam Hjelmeset,Andrew Franck,Bingrui Yang,Jose Bustamante Ortiz,Yuanrong Shen,Joel Walsh
机构: Occidental College(欧克兰学院)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Sound (cs.SD)
备注: 3 pages, 2 figures, 1 table. Demonstration paper presented at the non-archival demonstrations track of the 13th ACM Conference on Learning @ Scale (L@S '26), Seoul, South Korea, June 29-July 3, 2026. Not published in the ACM Digital Library
Abstract:Detecting phonemes from children’s speech has historically been difficult due to the scarcity of training data, and unique characteristics of children’s speech. During a phoneme detection competition, we found that training a lightweight model to predict the age of the learner, as well as the phoneme sequence, enabled a 94M-parameter model to outperform WavLM Large models (317M) on the target DrivenData distribution, and fall within approximately 0.04 CER of competition ensembles with 90 times the parameters. This has enabled the creation of PhonemeTrainer, an application that can run on most modern cellular phones. This will ultimately enable better Automated Speech Recognition (ASR) and pronunciation helper apps for children’s speech, with the privacy and compliance benefits that come with edge processing.
[NLP-71] Multimodal Item Parameter Estimation using Simulated Response Probabilitie
【速读】： 该论文旨在解决如何基于生成式 AI 从多选题数据中重建项目反应理论（Item Response Theory, IRT）模型曲线的问题，特别是针对包含图像与文本双重模态刺激的多选题项目。其核心挑战在于如何在缺乏显式参数估计的情况下，准确捕捉学生在不同能力水平下的作答概率分布，并由此推断题目难度等关键参数。解决方案的关键在于利用经过微调的多模态大语言模型（Multimodal Large Language Model, MLLM），以 Qwen3.5 为基础架构，通过提示工程和端到端微调，使模型学习在给定学生能力标签条件下对多个选项作答概率的系统性偏差模式。该模型通过隐式建模 3PL 和多选模型（MCM）所对应的响应概率函数，实现了对未见测试集上题目难度的高精度预测，从而无需传统统计拟合即可完成对IRT参数的近似重构。
链接: https://arxiv.org/abs/2608.10154
作者: Christopher Ormerod,YoungKoung Kim
机构: College Board
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: Submitted and Accepted for AIME-Con 2026
Abstract:We present results from reconstructing multiple-choice model (MCM) and three-parameter logistic (3PL) model curves using a fine-tuned multimodal large language model (LLM) based on Qwen3.5. The model is prompted and fine-tuned to replicate choice probabilities across a large training corpus of multiple-choice items containing both image and text stimuli, conditioned on a labeled set of student ability levels. By learning to reproduce the systematic error patterns of students across a discrete range of abilities, the LLM implicitly captures the underlying response probabilities encoded in the 3PL and MCM curves. This allows us to accurately approximate item difficulty on a held-out test set directly from the model’s predicted option probabilities.
[NLP-72] he Parser Already Knows: Lightweight Bias Correction in Constrained Decoding
【速读】： 该论文旨在解决语法约束解码（Grammar Constrained Decoding, GCD）中因严格掩码策略导致语言模型（Language Model, LM）原始概率分布被扭曲的问题，进而引发生成结果虽语法正确但质量欠佳的矛盾。现有方法在输出质量与推理延迟之间难以兼顾：强制掩码虽高效但破坏分布，而在线采样虽能恢复分布却需高成本迭代重采样。其解决方案的关键在于利用增量解析过程中已维护的内部解析器（parser）和词法分析器（lexer）状态，这些状态天然蕴含未来语法有效性信息，可作为恢复LM真实概率分布的依据。作者提出一种轻量级、离线训练的logit修正机制，基于此语法与词法状态以及候选下一个词元进行校正。由于该状态已在解析时计算，提取代价极低且不修改基础模型权重。实验表明，该方法显著缩小了掩码分布与模型真实分布之间的差距，在多个语法场景下均优于传统掩码与在线采样方法；即使最简版本仅依赖候选词元本身，也因隐含的前瞻信息（类似解析器中的向前查看机制）而达到或超越基线，有效恢复了因掩码丢失的概率质量，实现了概率一致性与语法合规性的统一。
链接: https://arxiv.org/abs/2608.10137
作者: Işıl Özgü,Yaoxuan Wu,Guy Van den Broeck,Miryung Kim
机构: Işıl Özgü; Yaoxuan Wu; Guy Van den Broeck; Miryung Kim
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 9 pages, 5 figures
Abstract:Grammar Constrained Decoding (GCD) forces Language Models (LMs) to produce syntactically valid outputs by masking out non-conforming tokens at each step. However, rigid masking distorts the model’s underlying probability distribution, often biasing generation toward valid but suboptimal outputs. While online sampling restores this distribution, it requires computationally expensive iterative resampling. As a result, existing methods force a compromise between output quality and inference latency. Our key insight is that the internal parser and lexer states inherently maintained during incremental parsing already encode future grammatical validity – exactly the information required to restore the LM’s true distribution. We propose a lightweight, offline-trained logit correction conditioned on this syntactic and lexical state together with candidate next tokens. Because these states are already computed as a necessary part of incremental parsing for masking, extracting them adds negligible overhead while leaving the base LM’s weights completely untouched. Across several grammars, this correction substantially closes the gap between the masked distribution and the LM’s true distribution, consistently outperforming both masking and online sampling. Even its lightest variant, which relies on the candidate next token alone, still matches or exceeds both baselines: the next token itself carries an implicit lookahead, much like how parsers commonly use a lookahead token to resolve ambiguous decisions. By restoring the probability mass that masking removes, it reconciles the LM’s probabilistic integrity with grammar conformance.
[NLP-73] Procedural Fairness Failures in RLHF from Preference Averag ing ICLR2026
【速读】： 该论文旨在解决强化学习中人类反馈（Reinforcement Learning from Human Feedback, RLHF）在处理异质性偏好时所引发的程序公平性失效问题。传统RLHF假设所有人类偏好具有同质性，通过平均化不同群体的偏好信号构建单一奖励模型，导致少数群体偏好在训练过程中被系统性地忽视，从而造成不公平的对齐结果。其核心解决方案是提出一种新型方法——偏好感知的强化学习中人类反馈（Preference-Aware RLHF, PA-RLHF），该方法在奖励建模阶段将不同偏好模式进行分离优化，而非简单聚合，从而保留各群体的独特偏好信号。实验表明，在受控环境中PA-RLHF可将整体对齐准确率从46.9%提升至67.9%，同时将最优与最差对齐群体间的公平差距由15.9个百分点降至9.6个百分点。研究揭示了奖励学习中的结构性设计缺陷本身即可引发程序公平性问题，即使在无噪声的理想条件下亦然，这对大型语言模型及自主代理系统具有深远影响，因其可能在序列决策中累积并放大社会不平等。
链接: https://arxiv.org/abs/2608.10126
作者: M P V S Gopinadh,Karthik Kamuju,Kummari Avinash,John Joshua,Srinivasa Raju Rudraraju
机构: Vishnu Institute of Technology (维什努技术学院)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 4 pages, Accepted at the ICLR 2026 Workshop on Algorithmic Fairness Across Alignment Procedures and Agentic Systems (AFAA)
Abstract:Reinforcement Learning from Human Feedback (RLHF) aggregates heterogeneous preferences into a single reward model, assuming preference homogeneity. When preferences are heterogeneous, this aggregation induces a procedural fairness failure where majority preference groups dominate reward learning while minority preferences are systematically under-represented. This work defines procedural fairness in alignment as preserving distinct preference signals during reward modeling and shows that standard RLHF violates this via preference averaging. Preference-Aware RLHF (PA-RLHF) is introduced, separating optimization across preference modes at the reward learning stage. In a controlled setting, PA-RLHF improves overall alignment accuracy from 46.9% to 67.9% and reduces the fairness gap between best and worst aligned groups from 15.9 to 9.6 percentage points. These results show that procedural fairness failures in alignment can arise from structural design choices in reward learning, even in controlled, noise-free settings, with direct implications for large language models and agentic systems, where biased reward models can compound inequities across sequential decisions.
[NLP-74] PERCEPT: A Corpus for POS Tagging and Analysis of Persian-English Code-Mixing
【速读】： 该论文旨在解决波斯语-英语代码混杂（code-mixing）语料资源匮乏的问题，特别是缺乏带通用依存关系（Universal Dependencies, UD）词性标注（POS）的代码混杂词汇标注数据，这限制了对多语言句法结构的深入语言学分析以及面向语法感知的自然语言处理（NLP）模型的发展。其解决方案的关键在于构建PERCEPT——首个公开可用的大规模波斯语-英语代码混杂语料库，并采用大语言模型（LLM）辅助的标注框架，实现代码混杂词汇的自动词性标注与文档级主题分类。该框架通过自动化流程显著提升标注效率，且经人工评估验证，自动生成的标注与人工标注具有高度一致性，确保了数据可靠性。基于此，研究首次在多个社交媒体平台（X、Instagram、Digikala）上开展了系统性语言学分析，发现名词是代码混杂中最常见的词类，而其他词类分布呈现平台差异；同时，代码混杂词的位置分布具有跨平台一致性，但其触发效应在Digikala平台尤为显著。该语料库已公开发布，为后续多语言代码混杂研究提供了重要基础。
链接: https://arxiv.org/abs/2608.10109
作者: Ghazal Kalhor,Zahra Jafari,Amirarsalan Shahbazi,Behnam Bahrak
机构: University of Tehran (德黑兰大学); Khatam University (卡塔姆大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Social media has become a major venue for multilingual communication, where users frequently mix multiple languages within a single utterance. Although code-mixed corpora have been developed for several language pairs, Persian-English code-mixing remains relatively underexplored. Existing Persian resources lack Universal Dependencies (UD) part-of-speech (POS) annotations for code-mixed words, limiting both linguistic analyses and the development of syntax-aware NLP models. To address this gap, we introduce PERCEPT, the first publicly available large-scale Persian-English code-mixed corpus annotated with Universal Dependencies POS tags for code-mixed words. The dataset comprises 6,800 posts collected from X, Instagram, and Digikala. We further present an LLM-assisted annotation framework that automatically assigns POS tags and document-level topics. Human evaluation demonstrates high agreement between the automatically generated annotations and gold annotations, confirming the reliability of the annotations. Using PERCEPT, we conduct the first comprehensive linguistic analysis of Persian-English code-mixing across multiple social media platforms. Our analyses reveal that nouns are the predominant category for code-mixed words, while the distributions of other POS categories vary across platforms. We further find that the positional distribution of code-mixed words is remarkably consistent across platforms, whereas the triggering effect is substantially more pronounced in Digikala. PERCEPT is publicly available at this https URL.
[NLP-75] Position Encoding in Transformers: From Absolute and Relative Methods to Rotary Position Embeddings and Long-Context Scaling
【速读】： 该论文旨在解决自注意力机制（Self-attention）在建模序列中词元（token）顺序信息方面的固有缺陷，即其本身不具备对词元位置的显式编码能力。尽管通过位置编码（Position Encoding）可引入绝对坐标、相对距离或位置相关的旋转等机制以弥补这一不足，但不同方法在实现方式、计算开销、与键值缓存（KV caching）的兼容性以及长序列外推（length extrapolation）性能上存在显著差异。论文的核心贡献在于构建了一个统一的理论框架，系统梳理并比较了正弦/余弦绝对位置嵌入、Shaw-style相对位置表示、Transformer-XL、T5相对偏置、ALiBi及旋转位置编码（Rotary Position Embeddings, RoPE）等代表性方法，并深入分析其在位置信息注入位置、计算复杂度、缓存兼容性及外推能力上的异同。进一步地，论文探讨了面向长上下文场景的扩展技术，包括位置插值、RoPE缩放律、NTK感知缩放、动态NTK、NTK分段、YaRN、LongRoPE及LongRoPE2等，重点关注频率分配策略、注意力重标定机制、训练长度与目标上下文长度之间的关系。研究强调，仅具备在训练长度之外计算位置特征的能力，并不等同于可靠的长上下文泛化能力；真正的长序列性能需通过短上下文保留能力、逐位置困惑度、信息检索、推理任务及长文本代码生成等多维度评估加以验证。
链接: https://arxiv.org/abs/2608.10021
作者: Jiguo Li
机构: 未知
类目: Computation and Language (cs.CL)
备注: 14 pages, a cookbook for students and junior researchers
Abstract:Self-attention models content-dependent interactions between tokens but does not by itself encode token order. Position encoding addresses this limitation by introducing absolute coordinates, relative distances, or position-dependent rotations into Transformer representations and attention scores. This technical survey develops a unified account of sinusoidal and learned absolute position embeddings, Shaw-style relative position representations, Transformer-XL, T5 relative position bias, ALiBi, and Rotary Position Embeddings (RoPE). We derive how RoPE converts absolute position indices into relative phase differences in Query-Key inner products and compare these methods in terms of where position is injected, computational cost, compatibility with KV caching, and length extrapolation. We then examine long-context extensions, including Position Interpolation, RoPE scaling laws, NTK-aware scaling, Dynamic NTK, NTK-by-parts, YaRN, LongRoPE, and LongRoPE2, with emphasis on frequency allocation, attention rescaling, training length, and target context length. We also summarize implementation considerations, evaluation protocols, and position-encoding choices in representative large language models. A central conclusion is that the ability to compute positional features beyond the training length does not imply reliable long-context generalization; context extension must be evaluated through short-context retention, position-wise perplexity, retrieval, reasoning, and long-context code tasks.
[NLP-76] OpenPM: Auditable Point-in-Time Evaluation for LLM Portfolio-Management Agents
【速读】： 该论文旨在解决当前大型语言模型（Large Language Models, LLM）在投资组合管理应用中普遍存在评估结果被高估的问题，主要源于前瞻泄露（look-ahead leakage）、乐观执行假设以及未实际执行的风险约束。为此，论文提出OpenPM——一个可审计的时点评估框架，用于对LLM投资组合管理代理进行严格、透明的测试。其核心解决方案在于：在S&P 500股票池内，以100万美元的全多头策略为基础，采用五分钟间隔的市场数据，并确保所有信息在决策时刻均不可逆地可用；通过将自然语言形式的风险指令转化为类型化约束并强制应用于实际交易组合，实现风险控制的可验证性；每次运行生成包括污染证明（contamination certificate）、成本敏感性曲线和约束遵守报告在内的审计证据，从而保障评估过程的可信度。此外，研究构建了一个基准代理“分层分配器”（tiered allocator），由类型化分析师评分候选资产、构造型大模型（constructor LLM）生成权重，再由确定性批判者（deterministic critic）保证可行性。通过固定分析师证据并跨不同构造模型复用，有效隔离了构造器行为的影响。案例研究表明，在短周期窗口下，更强的构造器虽带来轻微且依赖模型的收益提升，但分析师质量远高于构造器选择的重要性，而换手率是主要成本来源。所有回报均为单个冻结窗口下的理论上限，未考虑市场冲击影响，亦非经验证的超额收益（alpha）。
链接: https://arxiv.org/abs/2608.09988
作者: Xinying Cai,Minghao Guo,Jiahe Liu,Jiaojiao Han,Bangwei Guo,Yitao Long,Yuxuan Chen,Bohan Wu,Dimitris N. Metaxas,Raymond Li
机构: Rutgers University (罗格斯大学); Technical University of Denmark (丹麦技术大学); New Jersey Institute of Technology (新泽西理工学院); New York University (纽约大学); Columbia University (哥伦比亚大学); University of Illinois Urbana-Champaign (伊利诺伊大学厄本那-香槟分校); University of British Columbia (不列颠哥伦比亚大学)
类目: Computational Engineering, Finance, and Science (cs.CE); Computation and Language (cs.CL)
备注: 14 pages, 1 figure
Abstract:Large language models are increasingly used to read markets, assess risk, and allocate capital. However, reported results for LLM trading agents can be inflated by look-ahead leakage, optimistic execution, and risk mandates that are described but not enforced. We present OpenPM, an auditable point-in-time evaluation framework for LLM portfolio-management agents. In OpenPM, an agent manages a \ 1M long-only book over the S\P 500 universe using market data at five-minute intervals. Every record visible to the agent must be available at the decision time. Natural-language risk mandates are converted into typed constraints and enforced on the executed portfolio. Each run produces audit artifacts, including a contamination certificate, a cost-sensitivity curve, and a constraint-adherence report. We also build a reference agent named the tiered allocator, where typed analysts score candidates, a constructor LLM proposes weights, and a deterministic critic guarantees feasibility. We isolate constructor behavior by capturing analyst evidence once and replaying it across constructor models. In our short-window case study, stronger constructors show modest and model-dependent gains over equal weighting on the same pool, but analyst quality matters more than constructor choice, and turnover is the main cost driver. All returns are upper bounds on a single frozen window without market impact, not validated alpha.
[NLP-77] When Chain-of-Thought Helps and When It Hurts: An Empirical Investigation of the Serial-Depth Bottleneck in LLM Reasoning
【速读】： 该论文旨在解决“链式思维（Chain-of-Thought, CoT）提示是否普遍提升大语言模型（LLM）推理能力”这一广泛假设的可信性问题。研究表明，尽管CoT在某些任务上表现显著，但其有效性并非普适，而是与任务的串行深度（serial depth）密切相关。解决方案的关键在于引入H_dp带宽界限（H_dp bandwidth bound）这一理论框架，揭示了Transformer架构在单次前向传播中处理串行计算的能力存在内在瓶颈：当任务所需的推理步骤超过单次计算容量时，必须通过外部化（externalisation）方式（如CoT）将串行过程显式分解以缓解负担。研究发现，在高串行深度任务（如GSM8K、MATH）中，CoT可带来54至68个百分点的性能恢复；而在低串行深度任务（如MMLU、ARC）中，CoT基本无增益（Δ ∈ [0.0, +4.6] pp），表明其对已能适应单次前向传播的任务为冗余；在中等复杂度任务（HumanEval）中则呈现模型规模依赖的过渡现象。跨基准的串行深度与性能恢复之间存在显著正相关（Spearman ρ = 0.661, p = 0.007），且多数统计检验在多重校正后仍显著。因此，结论表明，CoT并非通用的推理增强机制，而是一种针对串行计算超载的“带宽绕行”策略，仅在超出单次前向传播容量的任务中有效。
链接: https://arxiv.org/abs/2608.09942
作者: Tughanbulut Kurtulush
机构: Vistula University (维斯图拉大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 15 pages, 3 figures, 5 tables. Pre-registered study (OSF: this https URL ). Data and code: this https URL
Abstract:It is widely assumed that chain-of-thought (CoT) prompting universally improves LLM reasoning. We investigate this through the conceptual framework of the H_dp bandwidth bound (Chen et al., 2024): although the formal bound binds only asymptotically (at astronomically large prompt lengths), it identifies a real architectural bottleneck – serial computation exceeding a transformer’s single-pass capacity must be externalised, which is what CoT does. Our central finding is a within-benchmark serial-depth gradient: single-pass (no-CoT) accuracy degrades monotonically with per-item serial depth, while CoT is approximately depth-invariant. We measure CoT effects across three instruction-tuned models (Qwen-2.5-7B/32B, Llama-3.1-8B) and five standard NLP benchmarks at practical context lengths. On high-depth P-complete tasks (GSM8K, MATH), CoT gives a +54 to +68 pp recovery gap across all models. On shallow TC^0 tasks (MMLU, ARC), CoT is structurally redundant (Delta in [0.0, +4.6] pp, no significant negative effect) – though high no-CoT baselines (up to 95% on ARC) may reflect contamination, so this null is not a clean architectural test. The intermediate class L (HumanEval) shows a model-size-dependent transition: +23.2 pp (32B), +9.1 pp (8B), -28.7 pp (7B). The cross-benchmark depth-recovery correlation is Spearman rho = 0.661 (p = 0.007, n = 15); 9 of 15 benchmark-level McNemar tests are significant after Bonferroni correction. Pre-registered on OSF, our results indicate that CoT is not a universal reasoning enhancer but acts as a bandwidth bypass: it helps serial computation that strains single-pass capacity and is redundant for tasks that already fit.
[NLP-78] he Multilingual Quantization Tax: Structural Collapse and Typological Frag ility in Edge SLMs EMNLP2026
【速读】： 该论文旨在解决4-bit权重量化在边缘设备上部署小型语言模型（Small Language Models, SLMs）时所引发的性能退化问题，即“量化税”（quantization tax）在多语言环境下的评估长期局限于英语语境，缺乏跨语言的系统性分析。其核心解决方案在于提出一种零样本多语言评估框架，首次对Gemma 4与Qwen 3.5两大架构在8种音系结构差异显著的语言上进行4-bit量化性能评估，采用MMLU ProX Lite和GlobalPIQA作为测评基准。关键发现揭示了量化过程中暴露的深层预训练不平等现象，识别出四大核心现象：（1）音系脆弱性（Typological Fragility），低资源及非拉丁文字体系因架构特异性双重解离导致表征崩溃，无法生成有效任务逻辑输出；（2）母语脆弱性悖论（Home Language Fragility Paradox），基础预训练路径对精度损失保护作用有限；（3）领域特定遗忘（Domain-Specific Forgetting），多步跨语言路由能力下降，而关联性软科学记忆仍保持鲁棒；（4）量化抗性（Quantization Resistance），高度饱和且音系对齐的领域表现出对确定性退化的抵抗能力，后量化性能提升受限于统计噪声。这些发现表明，4-bit量化带来的性能影响具有显著的语言类型依赖性，亟需构建更具包容性的多语言量化评估范式。
链接: https://arxiv.org/abs/2608.09941
作者: Mohammad Wathiq Soualhi
机构: Independent Researcher
类目: Computation and Language (cs.CL)
备注: Under review at EMNLP 2026
Abstract:While 4-bit weight quantization is critical for deploying Small Language Models (SLMs) on edge devices, evaluations of the resulting performance degradation-the quantization tax-remain overwhelmingly English-centric. We present a zero-shot multilingual evaluation of 4-bit quantization across the Gemma 4 and Qwen 3.5 architectures. Evaluating on eight typo-logically diverse languages using MMLU ProX Lite and GlobalPIQA, we show parameter truncation exposes deep pre-training inequalities. We identify four phenomena: (1) Typological Fragility: low-resource and specific non-Latin scripts suffer representational collapse via architecture-specific double dissociations, failing to generate valid task logits; (2) Home Language Fragility Paradox: foundational pre-training pathways provide limited precision loss protection; (3) Domain-Specific Forgetting: multi-step cross-lingual routing degrades while associative soft-science recall remains robust; and (4) Quantization Resistance: highly saturated, typologically aligned domains resist deterministic degradation, with post-quantization performance gains bounded by statistical noise.
[NLP-79] Carefully Considering Culture: Analyzing LLM Alignment in Single- and Multi-Cultural Settings using Cultural Consensus Theory ACL
【速读】： 该论文旨在解决当前自然语言处理（NLP）领域中大语言模型在理解跨国文化规范时存在的局限性问题，即现有研究多关注分布模式而忽视了国家内部群体共识及多元文化共存的复杂性。其解决方案的关键在于引入文化共识理论（Cultural Consensus Theory, CCT），通过建模群体内部的差异性与共识结构，揭示语言模型在文化表征中存在的系统性偏差。研究基于世界价值观调查（World Values Survey, WVS）在10个国家、12个文化维度上的数据，发现模型常因无法形成连贯的群体共识或过度规整化共识而导致文化结构误判。通过显式表征组内变异，CCT为评估模型是否真实反映人类多样性提供了可操作的诊断工具，有效区分了模型对文化的拟真表达与算法同质化倾向。
链接: https://arxiv.org/abs/2608.09937
作者: Krishna Pothugunta,John P. Lalor
机构: 未知
类目: Computation and Language (cs.CL); Computers and Society (cs.CY)
备注: Accepted to ACL Findings 2026
Abstract:Recent work in NLP has probed large language models for their understanding of cultural norms across countries. However, this work typically considers distributional patterns, ignoring group consensus or possible multicultural environments within a country. In this work, we leverage cultural consensus theory (CCT) from cultural anthropology to model such multidimensional nuance. Applying CCT to the World Values Survey (WVS) across 10 countries and 12 domains, we demonstrate that models frequently misrepresent cultural structures by either failing to form cohesive consensus or severely over-regularizing consensus. Through explicit representation of intra-group variance, CCT provides actionable diagnostics to evaluate when models reflect true human diversity versus algorithmic homogenization.
[NLP-80] Conflict or Strategy? Asymmetric Role Framing of La France insoumise and Rassemblement National in French News Headlines 2022-2025
【速读】： 该论文旨在解决法国新闻标题在报道左翼与右翼民粹主义挑战者（即“不屈法国”LFI与国民联盟RN）时，是将其框架为对称的“极端分子”，还是作为本质不同的政治对手这一核心问题。其关键解决方案在于通过一个经过分层人工审计验证的三模型大语言模型（LLM）标注流程，系统性地识别和比较两类政党在新闻标题中的政治角色建构差异。研究发现，核心机制并非价值判断（valence）上的不对称，而是角色（role）层面的不对称：冲突框架（conflict framing）与战略博弈框架（strategic-game framing）在不同模型和时间跨度下均表现更强的稳定性，且“攻击者”（AGGRESSOR）作为角色语法起到佐证作用；具体表现为LFI更常被置于冲突语态中，而RN则更多出现在战略-选举语态中。这一角色差距在所有标注模型中方向稳定，经受住自助法与置换检验考验，并贯穿多数媒体集团及2022至2025年期间。此外，次要的道德问责层（如归责、正当化、受害者化）主要由媒体立场决定而非政党属性，导致整体统计结果掩盖了部分最极化的文本模式。方法论上，该研究揭示了标注管道具有双层级可靠性特征：冲突与战略博弈框架具备最高的人工验证度与跨模型一致性；角色判定虽方向稳定但因审计可靠性较低被视为佐证项；规范性判断（如合法性、责任归属）则相对薄弱。因此，该研究将政治角色分配（political-role assignment）确立为计算框架研究的新靶点，以解构传统价值衡量所混淆的内涵，并建立基于构念分层的可靠性校准框架，用于优化政治文本任务中多数投票式LLM标注管道的可信度。
链接: https://arxiv.org/abs/2608.09936
作者: Amr Sobhy
机构: Le French News Lab (法国新闻实验室)
类目: Computation and Language (cs.CL)
备注: 19 pages, 3 figures, includes appendices
Abstract:Do French news headlines frame left- and right-populist challengers as symmetric ``extremes,‘’ or as fundamentally different political adversaries? We examine 28,592 headlines about La France insoumise (LFI) and Rassemblement National (RN) published by 25 French-language outlets between 2022 and 2025, annotated through a three-model LLM pipeline validated against a stratified human audit. The clearest finding is role asymmetry rather than valence asymmetry: conflict framing and strategic-game framing are more robust across models and time than delegitimization, with AGGRESSOR serving as corroborating role syntax. LFI appears in headlines more often through a conflict register and RN through a strategic-electoral register. This role gap is direction-stable across all three annotation models, survives bootstrapping and permutation tests, and persists across outlet families and most of 2022-2025. A secondary moral-accounting layer (who is blamed, legitimized, or cast as a victim) is structured by outlet rather than party, producing aggregate nulls that conceal some of the corpus’s most polarized patterns. Methodologically, the annotation pipeline reveals a two-tier reliability profile: conflict and strategic-game framing achieve the strongest human validation and cross-model stability; actor role is direction-stable but treated as corroborating because its audit reliability is lower; normative-judgment constructs (legitimacy, blame) are weaker. The paper contributes political-role assignment as a target for computational framing research that decomposes what valence-based measures conflate, and establishes a construct-stratified reliability framework for calibrating majority-vote LLM annotation pipelines in political text tasks.
[NLP-81] LLM Agents Factory: Retrieval of Domain-Specific LLM Agents SIGIR2026
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）代理在实际部署中因动态生成代理而带来的计算开销大和运行不稳定的问题。其核心挑战在于，传统的动态代理设计需为每个用户请求实时构建角色专业化的行为，导致推理成本高且难以保证一致性。为此，论文提出LLM Agents Factory——一种基于检索的框架，通过一个包含超过2万条预定义代理配置的结构化知识库，在需要时按需构建领域特定且基于维基百科（Wikipedia-grounded）的代理。该方案的关键在于：一是利用语义搜索从预设的代理配置库中高效检索合适的代理模板，实现快速、低成本的代理构造；二是支持将检索到的代理行为进行蒸馏，生成轻量级微调模型以实现直接代理生成。实验结果表明，在单代理场景下，该方法在MMLU、BIG-bench及BIG-bench Hard基准上均显著优于非代理基线模型，且在使用1200亿参数骨干模型时达到与AutoGen相当的生成质量，但推理成本大幅降低。研究揭示了从结构化代理仓库中检索代理是一种比动态生成更高效、准确且可控的替代路径，满足工业应用对稳定性与效率的严苛要求。
链接: https://arxiv.org/abs/2608.09934
作者: Vitalii Belov,Artyom Sosedka,Andrey Sakhovskiy,Elizaveta Kovtun,Artyom Boyarskikh,Semen Budennyy
机构: Sber AI(斯伯AI); Moscow Institute of Physics and Technology (莫斯科物理技术研究所); National University of Science and Technology MISIS (俄罗斯国立科学技术大学米西斯); Skolkovo Institute of Science and Technology (斯科尔科沃科学与技术研究院); Artificial Intelligence Research Institute (人工智能研究院)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 7 pages, 1 figure, SIGIR 2026
Abstract:Large language model (LLM) agents improve task performance by decomposing problems into role-specialized behaviors. However, their practical deployment is often limited by the computational cost and instability associated with the on-the-fly agent design for each user request. To address this, we present LLM Agents Factory, a retrieval-based framework that constructs domain-specific and Wikipedia-grounded agents on demand using a base of over 20K predetermined agent profiles. Our framework supports two modes: (1) agent profile retrieval via semantic search and (2) distillation into a compact model fine-tuned for direct agent generation. Experiments on MMLU, BIG-bench, and BIG-bench Hard in a single-agent scenario demonstrate that our retrieval-based agent construction surpasses non-agent baselines in accuracy while matching AutoGen generation quality with a 120B backbone at a substantially lower inference cost. Our work reveals that retrieval from a structured agent repository provides a cost-efficient, accurate, and controllable alternative to dynamic agent generation, responding to the strict demands of industrial applications. We provide the implementation code and the agent base in this https URL.
[NLP-82] Divergent Response Modes in Frontier Language Models Under Steering Pressure
【速读】： 该论文旨在解决前沿大语言模型在不同数据、目标与安全机制训练背景下，其行为可引导性（behavioral steerability）是否存在显著差异的问题，尤其关注在显式指令引导下的响应模式变化。研究的关键在于通过构建三类任务（价值冲突、推理诱导、推理抑制）的300对基准与引导样本，结合六款来自不同开发者的前沿模型作为盲评裁判，基于固定行为评分标准进行24,480次判断，并采用留一法共识评分方法量化模型响应的可操控性。研究发现，各模型不仅在响应被引导的程度上存在差异，更在生成响应的“模式”（mode）上表现出独特性，部分响应模式仅出现在少数模型中。例如，GPT-5 在拒绝披露推理过程的同时保持答案不变（99% vs. 所有其他模型的0%），而 Claude Opus 4.7 和 GPT-5 对显式抑制指令的抵抗方式亦不相同。通过以 Llama 为开源权重模型进行内部追踪，研究发现最大行为差异可归因于模型内部表征；线性探测在残差流（residual stream）中实现0.87的保留精度，且在生成过程中注入该方向可使行为变化从0%提升至86%，验证了该方向的决定性作用。所有结论在引入令牌预算修正与无假设盲评提示的控制实验下仍成立，表明其结果具有稳健性。
链接: https://arxiv.org/abs/2608.06578
作者: Ali Jalal-Kamali
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:Frontier language models are trained using distinct data, objectives, and safety pipelines. Whether these differences produce measurably different behaviors under explicit steering pressure remains underexplored. This study evaluates behavioral steerability across six frontier models from six developers using 300 paired base and steered items over three categories: values-conflict, reasoning-elicitation, and reasoning-suppression (plus 40 validation items). All six models act as blind peer judges and classify every response based on fixed behavioral rubrics. The resulting 24,480 judgments are scored by leave-one-out consensus. We find that models differ not just in how much steering shifts their behavior but in what kind (mode) of response they give, and some response modes appear in only one or two of them. GPT-5 deflects requests to disclose its reasoning while leaving its answer intact (99% vs. 0% for all other models). Claude Opus 4.7 and GPT-5 resist explicit suppression instructions and in different ways. Using Llama as the open-weight model, we trace the largest behavioral split to its internals. A linear probe decodes the behavior from the residual stream at 0.87 held-out accuracy while injecting that direction during generation drives the behavior from 0% to 86% across an intervention sweep. Every finding holds under both a token-budget remediation and a control experiment with a hypothesis-blind judgment prompt.
[NLP-83] When Is a General Factor Distinguishable? Non-Proportionality Stable Structure and the Bifactor Decision
【速读】： 该论文旨在解决在因子分析中是否存在一个额外的通用因子（general factor）超越相关的一阶因子结构这一问题，其核心在于判断该属性是否可从总体协方差矩阵中确定。研究的关键在于揭示：当每个聚类内的通用因子与组因子载荷成比例时，双因子结构与相关因子结构在协方差上等价，因此样本量无法区分二者（命题1）；当这种比例性在所有聚类中均不成立时，在满足每簇至少三个题项及若干温和正则性条件下，任何具有对角唯一性的K因子模型均无法复现原始协方差矩阵（定理1）；介于两者之间的为混合边界，其位置可通过数值方法确定，并依赖于聚类的抗干扰能力。因此，可区分性是渐进的，由总体距离到K因子类的度量决定。由于该问题依赖于本身不确定的一阶因子结构，论文提出一种两步部分探索性因子分析程序，仅在相邻计数下结构保持重现时才输出结果，否则视为合理缺失。模拟研究表明，一致计数可能伴随无法重现的结构，且被吸收的局部依赖关系可模拟出通用因子效应，其误差随样本量增大而增加，但稳定性指标仍保持清洁。四个实证数据集展示了不同可能结果。
链接: https://arxiv.org/abs/2608.10731
作者: Jinsong Chen
机构: The University of Hong Kong (香港大学)
类目: Methodology (stat.ME); Computation and Language (cs.CL)
备注:
Abstract:Whether an additional general dimension is necessary beyond correlated first-order factors is a property of the population covariance matrix, not of any estimator or design. This research establishes when that property can be decided. Where the general and group loadings are proportional within every cluster the bifactor structure is covariance-equivalent to correlated factors, so no sample size separates them (Proposition 1); where that proportionality fails in every cluster, three items per cluster and some mild regularities leave no K -factor model with diagonal uniquenesses able to reproduce the covariance matrix (Theorem 1); and between them lies a mixed boundary, located numerically here and turning on cluster resistance. Distinguishability is therefore graded, measured by the population distance to the K -factor class. Because that question is conditional on a first-order structure which is itself uncertain, a two-step procedure is developed within partially exploratory factor analysis, delivering a structure only when it reproduces across adjacent counts and treating non-delivery as legitimate. Simulation shows that a unanimous count can accompany a structure that fails to reproduce, and that absorbed local dependence can imitate a general factor, the error growing with sample size while stability indicators stay clean. Four empirical datasets illustrate the possible outcomes.
信息检索
[IR-0] Are We Really Making Progress in Group Recommendation? Unmasking the Tie-Breaking Illusion RECSYS2026
链接: https://arxiv.org/abs/2608.11190
作者: Song-Duo Ma,Pu-Jen Cheng
类目: Information Retrieval (cs.IR)
备注: Accepted at RecSys 2026
Abstract:Recent group recommendation methods have reported strong improvements on standard benchmarks, but it remains unclear whether these gains always reflect genuine advances in modeling group preferences. In this paper, we show that several recent methods are affected by a systematic evaluation bias caused by the interaction between training-time score compression and evaluation-time deterministic tie-breaking. Specifically, an additional sigmoid transformation before the BPR objective can greatly increase tied top scores, making top-K metrics such as HR@K and NDCG@K highly sensitive to how ties are resolved. We revisit recent representative methods and their baselines on CAMRa2011 and Mafengwo under both group and user recommendation settings, and evaluate them with a tie-aware protocol that computes the exact expectation of HR@K and NDCG@K under uniform random tie-breaking. Our results show that many previously reported improvements shrink substantially under tie-aware evaluation, and the relative ranking of methods can change markedly. We further show that the additional sigmoid may act as implicit margin smoothing during optimization, and that temperature-scaled BPR can retain much of this benefit without inducing severe tie inflation. Overall, our findings highlight the importance of tie-aware evaluation for establishing reliable progress in group recommendation. The code is available at this https URL.
[IR-1] Role of Personality in Conversational Information Seeking CIKM2026
链接: https://arxiv.org/abs/2608.11164
作者: Abdisalam Abukar,Junchen Fu,Chengli Zhai,Joemon M. Jose
类目: Information Retrieval (cs.IR)
备注: Accepted by CIKM2026
Abstract:Large language models (LLMs) are increasingly used for information seeking, where users find, compare, and evaluate information through dialogue. In this role, the assistant does more than retrieve or generate content: it shapes how users articulate constraints, ask follow-up questions, verify claims, and decide when an answer is sufficient for action. Yet little is known about how user personality, assistant personality, and task context jointly influence these interactions. We examine personality as a controllable variable in conversational information seeking and study its effects on user behaviour and interaction quality. We conducted a controlled within-subject study in which assistant personality and task type were experimentally varied, while participant personality was measured using Big Five scores. Twenty-six participants each completed three information-seeking tasks under three assistant personality conditions: extraverted, conscientious, and neutral. Tasks covered exploratory travel planning, comparative smartphone shopping, and verification-sensitive health and diet information seeking. Data included conversation logs, behavioural traces, post-interaction questionnaires, an exit questionnaire, and Big Five measures. The assistant conditions were behaviourally distinct: the extraverted assistant produced longer turns, the conscientious assistant elicited higher user word share and more turns, and the neutral baseline fell between them. The strongest effect was a task-by-assistant interaction on trust and delegation, with preferred styles varying by task. No global winner emerged, but participants strongly preferred style choice or adaptation. These findings position assistant personality as a context-sensitive interactional design variable rather than a globally optimisable system property.
[IR-2] Multi-Level Evidence Aggregation for Robust Facial Phenotype Retrieval in Rare Genetic Disorder Prioritization
链接: https://arxiv.org/abs/2608.11037
作者: Alexander Hustinx,Carolin Kaffiné,Behnam Javanmardi,Tzung-Chien Hsieh,Peter Krawitz
类目: Computer Vision and Pattern Recognition (cs.CV); Information Retrieval (cs.IR)
备注: including supplementary notes: 32 pages, 16 figures. Preprint submitted to journal for peer-review
Abstract:AI-assisted facial phenotyping supports rare genetic disorder prioritization by retrieving visually similar diagnosed cases from facial image reference databases such as the GestaltMatcher Database (GMDB). Existing GestaltMatcher-based retrieval frameworks compare each test image with individual gallery images in a facial phenotype embedding space. However, this pointwise formulation does not fully exploit available evidence, because patients may have multiple images and disorders may be represented by multiple diagnosed gallery patients. We propose an inference-time multi-level evidence aggregation framework that improves facial phenotype retrieval without modifying the underlying GestaltMatcher-Arc encoder. The framework combines embedding-level patient aggregation of multiple images from the same individual, patient-weighted disorder centroids, and hybrid individual-centroid scoring to integrate test-patient observations, disorder-level gallery evidence, and local nearest-neighbor evidence. We evaluated the approach on GMDB v1.1.4 across disorders represented during training (GMDB-Freq), unseen disorders (GMDB-Rare), and multi-image patient subsets, using a unified gallery containing both GMDB-Freq and GMDB-Rare disorders. Multi-level evidence aggregation improved mean per-disorder top- N retrieval accuracy across all evaluation subsets. Top-1 accuracy increased from 38.52% to 48.82% on GMDB-Freq and from 19.38% to 23.79% on GMDB-Rare. On multi-image subsets, top-1 accuracy increased from 46.12% to 60.94% on GMDB-Multi-Freq and from 18.54% to 26.71% on GMDB-Multi-Rare. These findings show that inference-time aggregation can improve next-generation facial phenotype retrieval without retraining the encoder, supporting a shift from isolated single-image matching toward multi-level aggregation of patient and disorder evidence for rare-disorder prioritization. Comments: including supplementary notes: 32 pages, 16 figures. Preprint submitted to journal for peer-review Subjects: Computer Vision and Pattern Recognition (cs.CV); Information Retrieval (cs.IR) Cite as: arXiv:2608.11037 [cs.CV] (or arXiv:2608.11037v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2608.11037 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Alexander Hustinx [view email] [v1] Tue, 11 Aug 2026 15:12:44 UTC (1,202 KB)
[IR-3] Self-Knowledge Retrieval Augmented Generation Framework for Patent Matching
链接: https://arxiv.org/abs/2608.11030
作者: Jian Zhang,Songlin Lei,Zhuohao Yang,Bangli Liu,Ziwei Wang,Xufeng Weng,Gehan Amaratunga,Yu Lin,Hongwei Wang
类目: Information Retrieval (cs.IR); Computation and Language (cs.CL)
备注: Accepted by IEEE CSCWD 2026
Abstract:Patent retrieval and matching based on large language models (LLMs) play a vital role in intellectual property protection. However, due to the complex structure of patent documents, dense technical terminology, and multi-modal information, traditional methods struggle to accurately identify subtle differences between patents. Existing LLM-based patent matching approaches typically rely on domain-specific pretrained or instruction tuning, which often entail high manual labeling costs and catastrophic forgetting. While retrieval-augmented generation (RAG) methods introduce external knowledge they fail to fully leverage LLM’s capability to automatically parse patents and mine deep semantic relationships. To address these limitations, this paper proposes a self-knowledge RAG framework that guides LLMs to autonomously extract key technical entities and construct hierarchical ontological structures from patent matching queries, thereby enabling query expansion and precise retrieval. The method integrates the FAISS retrieval with a generative matching mechanism, leveraging self-knowledge to enhance the model’s understanding of patent innovations and significantly improve retrieval and matching accuracy. Experimental results demonstrate the outstanding performance of the proposed method on real-world patent datasets, validating its effectiveness and application potential.
[IR-4] Sona Technical Report
链接: https://arxiv.org/abs/2608.11015
作者: Sona Team:Alexandr Udeneev,Aleksei Krasilnikov,Alexey Nadtochiy,Andrey Semenov,Andrey Tsyrkunov,Anna Krivonos,Anna Lipkina,Artem Matveev,Daniil Burlakov,Daniil Leschev,Daria Tikhonovich,Denis Burshtein,Ekaterina Dmitrieva,Eugene Krofto,Grigorii Khlystov,Ilya Murzin,Kirill Golovko,Ksenia Sycheva,Leonid Dmitriev,Mariia Rozaeva,Mariia Ulianova,Mikhail Sandul,Nikolai Savushkin,Oleg Sorokin,Roman Odobesku,Semyon Panenko,Sergei Liamaev,Sergei Makeev,Vadim Shilov,Veronika Ivanova,Viktor Yanush,Vladimir Baikalov,Vladislav Dodonov,Vladislav Tytskiy
类目: Information Retrieval (cs.IR)
备注:
Abstract:We introduce Sona, a single-model generative recommender for Yandex Music. In an online A/B test, Sona replaced the entire production cascade, comprising more than 15 candidate generators followed by pre-ranking and ranking models that consume hundreds of features, including signals from large transformer models such as Argus and target-attention scorers, while significantly improving key engagement metrics. The architecture of Sona unifies candidate generation and ranking around a shared user representation. Its encoder transforms the user’s chronological sequence of logged engagement events into hidden states consumed by both the autoregressive decoder and the Ranking Module. The next-token-prediction and distillation objectives jointly update the encoder, coupling generation and ranking through the same user state. Neither Sona nor its Teacher Ranker uses hand-engineered features; both operate on logged event fields and learned item representations. In the final Sona configuration, the larger teacher supplies ranking targets during training but is absent from serving, leaving the encoder, decoder, and Ranking Module as a single deployed model. We evaluate Sona in an online A/B experiment using live traffic from My Vibe on smart speakers, one of Yandex Music’s largest recommendation surfaces. Relative to the production control, Sona produced statistically significant uplifts of 4.53% in Active Users, the primary metric, 6.30% in Total Listening Time, and 11.42% in Likes. These effects were incremental to improvements retained from preceding deployments. The Active Users uplift was 2.35 times the increment previously delivered by Argus, the strongest model deployed on this surface before Sona. These results show that a single jointly trained model can replace a mature multi-stage recommendation cascade while improving recommendation quality on live traffic. Subjects: Information Retrieval (cs.IR) Cite as: arXiv:2608.11015 [cs.IR] (or arXiv:2608.11015v1 [cs.IR] for this version) https://doi.org/10.48550/arXiv.2608.11015 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[IR-5] meRoute: Time-Aware Modality Routing and Diffusion for Multi-Modal Recommendation
链接: https://arxiv.org/abs/2608.10983
作者: Pengyu Zhang,Yangqin Jiang,Klim Zaporojets,Congfeng Cao,Paul Groth
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI)
备注:
Abstract:Multi-modal recommenders fuse collaborative signals with item modalities such as text, images, and audio, but the usefulness of each drifts over time and at different rates. For example, chocolate purchases typically guided by textual ingredient cues can shift toward visual packaging and ambient audio around Valentine’s Day. This modality time-scale mismatch gives rise to two coupled challenges: (1) users require different modality proportions across temporal contexts, and (2) less relevant modalities are more likely to introduce outdated or misleading signals into the recommender. We address both challenges within a unified diffusion-based recommender, TimeRoute. A temporal-aware modal router maps each user’s aggregated behavioral features to a personalized modality distribution, replacing the globally shared fusion weights used in prior work. The diffusion-based graph reconstructor is then conditioned on the same temporal profile through Feature-wise Linear Modulation (FiLM) with dual-stream long- and short-term denoising heads, suppressing outdated modality edges before they enter the propagation graph. Experiments on TikTok, Amazon-Baby, and Amazon-Sports demonstrate consistent improvements of up to 9.8% in Recall@K, Precision@K, and NDCG@K over strong baselines across 10-seed paired tests. Code is available at this https URL.
[IR-6] Deciding When to Rely on Visual Information: Gated Multimodal Fusion in Sequential Recommendation RECSYS2026
链接: https://arxiv.org/abs/2608.10700
作者: Natalija Glisovic,Danica Kragic,Martin Tegner
类目: Information Retrieval (cs.IR)
备注: 8 pages, 6 figures, Accepted at CARS @ RecSys 2026
Abstract:Multimodal sequential recommender systems commonly fuse visual and collaborative signals uniformly, treating visual features as generically informative regardless of item or user context. We argue that visual utility, defined as the contribution of visual signals to recommendation quality, is a latent contextual variable that depends on both the item and the user’s interaction history rather than a fixed item property. To model this variability, we introduce VisGate, a framework that makes adaptive item-level fusion decisions conditioned on item embeddings and the user’s current sequence context. Visual representations are learned through a contrastive objective over sequential co-occurrence patterns, preserving complementarity with collaborative embeddings rather than aligning them into a shared space. Beyond achieving competitive recommendation performance, VisGate’s learned gate serves as a measurement tool for understanding when and why visual information is beneficial. Our analyses show that visual utility varies across items, increases under interaction sparsity when collaborative signals are weak, and correlates with visual distinctiveness in semantically meaningful ways. Together, these findings highlight the importance of both fine-grained fusion and modality complementarity, while demonstrating that item-level visual utility can be estimated and interpreted through learned gating behaviour.
[IR-7] Leverag ing Human Reading Behavior for Keyphrase Extraction: A Webcam-based Eye-tracking Corpus
链接: https://arxiv.org/abs/2608.10688
作者: Chengzhi Zhang,Xinyi Yan,Wenqi Yu
类目: Computation and Language (cs.CL); Digital Libraries (cs.DL); Human-Computer Interaction (cs.HC); Information Retrieval (cs.IR)
备注:
Abstract:Purpose: Keyphrases are statistically and semantically important textual units that can also attract readers’ attention during comprehension. However, existing keyphrase extraction (KPE) studies mainly focus on improving textual representation while largely overlooking human reading behavior. This study examines whether lightweight webcam-based eye-tracking features can improve KPE from Chinese academic abstracts in Library and Information Science (LIS). Methodology: To address the limited availability of eye-tracking data for Chinese academic reading, we developed a lightweight webcam-based data collection platform using the open-source SearchGazer library and constructed the Chinese LIS Eye-Tracking Corpus (CLIS-ET). Three character-level eye-tracking features, first fixation duration (FFD), fixation number (FN), and total fixation duration (TFD), were incorporated into KPE models to evaluate their effects on extraction performance. Findings: Eye-tracking features consistently improved KPE performance. The combination of FN and TFD achieved the best results on the Att-BiLSTM+CRF model, indicating that readers’ fixation behavior provides useful signals for identifying keyphrases in academic abstracts. Originality/value: This study introduces a cost-effective webcam-based eye-tracking approach for KPE and presents CLIS-ET, a Chinese academic eye-tracking corpus containing FFD, FN, and TFD features. The results demonstrate the value of incorporating human reading behavior into keyphrase extraction. Dataset and code: this https URL and this https URL. Subjects: Computation and Language (cs.CL); Digital Libraries (cs.DL); Human-Computer Interaction (cs.HC); Information Retrieval (cs.IR) Cite as: arXiv:2608.10688 [cs.CL] (or arXiv:2608.10688v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2608.10688 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Journalreference: aslib JIM, 2026 Submission history From: Chengzhi Zhang [view email] [v1] Tue, 11 Aug 2026 09:11:42 UTC (4,204 KB)
[IR-8] ENTLORE: A Graph-Grounded Benchmark for Latent Organizational Reasoning in Enterprise Question Answering
链接: https://arxiv.org/abs/2608.10679
作者: Akrin Zheng,Alexander Wu,Alaia Liu
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Enterprise question answering is framed as retrieving internal documents and generating grounded answers. Routine enterprise records, however, are work by-products in which required organizational relations remain implicit across heterogeneous sources. Existing benchmarks provide realistic multi-source evidence, but often materialize a predefined answer path and therefore test the composition of stated facts rather than recovery of a target relation absent from the corpus. We call the latter capability latent organizational reasoning. We introduce ENTLORE, a graph-grounded benchmark construction framework that reconstructs an audited enterprise world from routine documents, authoritative organizational tables, and operational records. Versioned organizational conventions certify derived relations in a truth graph, enabling complete golden answers and proof certificates. The aligned anonymized release exposes only the document corpus while withholding private structure and target relations. ENTLORE contains 2,341 documents from three source types and 907 questions spanning explicit lookup, cross-source composition, and latent organizational reasoning, evaluated across 56 model and access configurations. Structuring the released world as an induced entity graph or navigable knowledge base gives the strongest deployable results. Yet supplying gold documents still leaves 30.4% of latent questions unanswered, versus 12.6% and 6.2% for explicit and compositional questions. Enterprise QA therefore depends not only on document recall, but also on whether implicit organizational relations become usable. The benchmark, data, and code are publicly available at ENTLORE. Subjects: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL) Cite as: arXiv:2608.10679 [cs.IR] (or arXiv:2608.10679v1 [cs.IR] for this version) https://doi.org/10.48550/arXiv.2608.10679 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[IR-9] DistilVDR: A Compact End-to-End Visual Document Retriever via Dual-Student Distillation
链接: https://arxiv.org/abs/2608.10636
作者: Zhuchenyang Liu,Ziyi Wang,Yao Zhang,Yu Xiao
类目: Information Retrieval (cs.IR); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
备注: 15 pages, 2 figures, 8 tables
Abstract:Visual document retrieval (VDR) is dominated by multi-billion-parameter models that are slow to index at full corpus scale and expensive to serve. Prior compression routes either train a smaller multi-vector encoder from scratch or distil only the query side; neither yields a compact single-vector retriever end-to-end. We present DistilVDR, a 524M end-to-end VDR system distilled bilaterally from a single 8B vision-language teacher under a pointwise cosine alignment loss. All supervision comes from the frozen teacher’s embedding space, which was itself trained with relevance supervision, so the student objective needs no relevance labels, negative sampling, or contrastive term. We match VDR’s text-query and image-document input asymmetry with an asymmetric encoder-only student that concentrates visual capacity on the document side and keeps the query side at 70M parameters. We release two variants that share the same encoders and training and differ only in the document encoder’s visual-tile budget: DistilVDR-HiRes attains 61.74 average NDCG@5 on ViDoRe v1+v2+v3 (86.9% of the 8B teacher) and leads every reproduced sub-1B baseline on the high-resolution-sensitive v3 benchmark, while DistilVDR-Fast attains 59.98 at a 3 times smaller visual-token budget. Both variants store one million documents in a 15.6 times smaller index than the strongest sub-1B multi-vector baseline and index the corpus an order of magnitude faster. The code is available at this https URL.
[IR-10] Multi Interests for Joint Search-Recommendation Modeling
链接: https://arxiv.org/abs/2608.10535
作者: Xiangchen Pan,Wei Wei,Huakang Niu,Zhicong Cheng
类目: Information Retrieval (cs.IR)
备注:
Abstract:Search and recommendation are crucial for understanding user preferences. More and more studies are attempting to jointly model search behavior and recommendation behavior, by integrating user active search and passive recommendation behavior data to better mine user preferences. However, although existing cross-domain unified modeling frameworks can effectively compensate for the differences in behavior between domains, they overlook the expression of interests in different scenarios under mixed sequences. In this study, we propose a multi-interest-based mixed sequential modeling framework MIJSR, which performs multi-interest mining and adaptive integration on search recommendation mixed sequences from both structural and semantic perspectives. Specifically, our model can be roughly divided into three modules: cross-domain behavior fusion, multi-interest mining, and multi-task prediction. Firstly, we align the representations of query and item through contrastive learning training. Then, we extract the multi interests of the mixed behavior sequence from both structural and semantic perspectives. Structurally, we extract search interests, recommendation interests, and cross interests through subsequence partitioning and mask settings; In terms of semantics, we use the semantic information of queries for clustering and perform semantic segmentation on mixed sequences to construct semantic multi interests. Finally, the adaptive fusion of multiple interests is combined with other side information to use a progressive layered extraction model for multi-task prediction. Extensive experiments on two open-source datasets have shown that our model can further enhance its accuracy in search and recommendation by extracting users’ multi interests at a fine-grained level. Codes are available at this https URL.
[IR-11] When Do Anchor-Based Pointwise LLM Rerankers Help? Retriever Quality Statistical Scope and Anchor Design CIKM2026
链接: https://arxiv.org/abs/2608.10528
作者: Utshab Kumar Ghosh,Shubham Chatterjee
类目: Information Retrieval (cs.IR); Machine Learning (cs.LG)
备注: To be published in the 35th ACM International Conference on Information and Knowledge Management (CIKM 2026)
Abstract:Anchor-based pointwise LLM reranking scores each candidate against a shared reference passage to recover cross-document context at pointwise cost. We study when this actually helps, using GCCP/PAGC as a representative method. Our study is reproduction-first. We use reproduction as a starting point for a controlled component-level stress test of anchor-based pointwise reranking. Our initial reimplementation, based only on the paper text, achieves 0.24 nDCG@10 instead of the reported 0.66, revealing that several undocumented implementation details are necessary to reproduce the method. After identifying and recovering eight such details, we reproduce the reported results within 1.6% and use the validated implementation for controlled analysis. We find that the core contrastive scoring idea is robust under rigorous statistical correction. However, two design choices held fixed in the original paper are less reliable. First, we find that combining the contrastive score with the standard pointwise relevance score helps when the first-stage retriever is BM25, but gives little or no benefit when the first-stage retriever is a stronger dense model such as E5. Second, the paper’s more complex method for constructing the anchor is unnecessary. A much simpler anchor, built by interleaving the top-ranked sentences, matches or outperforms it across datasets. These findings are consistent across different LLM backbones, including a 4-bit quantized 72B model. Overall, anchor-based pointwise reranking is effective, but its gains come mainly from contrastive scoring rather than from the more complex aggregation and anchor-construction choices, and they appear under narrower conditions than the original evaluation suggests. Comments: To be published in the 35th ACM International Conference on Information and Knowledge Management (CIKM 2026) Subjects: Information Retrieval (cs.IR); Machine Learning (cs.LG) Cite as: arXiv:2608.10528 [cs.IR] (or arXiv:2608.10528v1 [cs.IR] for this version) https://doi.org/10.48550/arXiv.2608.10528 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[IR-12] owards Efficient Reasoning in LLM -Based Recommender Systems via Model Merging
链接: https://arxiv.org/abs/2608.10447
作者: Linh Dieu Le,Tong Chen,Shazia Sadiq,Hongzhi Yin,Ming Jin,Junliang Yu
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language model-based recommender systems are increasingly adopting slow-thinking models that generate step-by-step reasoning before making predictions, often achieving higher accuracy than fast-thinking models that predict directly. However, their reasoning traces are often unnecessarily verbose, increasing inference costs without commensurate accuracy gains. Existing training-based approaches to reasoning compression often incur substantial adaptation costs, while inference-time methods are brittle and difficult to scale. These limitations motivate model merging as a promising training-free direction for transferring specialised behaviours between models in a shared parameter space. In particular, merging a slow-thinking model with a fast-thinking counterpart provides a natural mechanism for balancing recommendation accuracy and reasoning conciseness. To this end, we propose, to our knowledge, the first model merging framework for reasoning compression in recommender systems. Unlike conventional merging methods that apply uniform merge coefficients across model components, our method performs fine-grained merging at the level of individual attention heads, capturing heterogeneous patterns in recommendation reasoning. Each attention head is assigned a distinct merge coefficient according to its contribution to critical reasoning evidence and its sensitivity to parameter change, enabling selective injection of the concise behaviour of the fast-thinking model into the slow-thinking model and reducing reasoning verbosity without compromising recommendation quality. Experiments on three benchmark datasets show that our method reduces reasoning length by up to 24.3% while outperforming competitive model merging baselines in maintaining recommendation accuracy. The code is available at this https URL.
[IR-13] Detecting an Effect Is Not Learning to Act on It: A Reward-SNR Floor for LLM Acquisition Agents
链接: https://arxiv.org/abs/2608.10441
作者: Ying Yuan
类目: Machine Learning (cs.LG); Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注:
Abstract:Many pipelines can pay a per-example cost to acquire an auxiliary, model-derived observation – an LLM’s structured reasoning, a slow oracle, an expensive measurement – and then must decide when the acquired signal is worth using. Our thesis is a distinction that is easy to miss: detecting that such a signal helps on average is not the same as learning to act on it per instance, and a reward-SNR floor governs when the second is even possible. Even when the signal is faithful and an in-sample oracle picking the top-b examples by realized reward shows a sizable apparent gain, no deployable policy can learn when to acquire it: across per-impression, cluster, regime, and uplift-tree granularities, learned routing never beats random, and a matched-moment noise placebo reproduces =100% of the oracle’s apparent gain – the apparent “learnable structure” is order statistics of noise. We explain this with one distinction, detecting a mean effect vs. learning a per-instance acquisition policy, and a reward-SNR detectability floor: routing is estimable offline only if the reward SNR rho clears rho*(N) ~= 2.8/sqrt(N), with a positive control confirming a true low-SNR limit rather than a broken pipeline. As a concrete instantiation we introduce Structured Hypothesis Embeddings (SHE): a frozen LLM turns a user history into ranked, confidence-scored, evidence-grounded intent hypotheses, fused into a recommender. On three public datasets (MIND, REES46, Amazon-Beauty), SHE is faithful and calibratable, yet its value is backbone- and regime-conditional (significant over an ordered GRU, +0.0114, 95% CI [+0.0030, +0.0209], but a global redundancy gap indistinguishable from zero), and learned acquisition collapses at every granularity because all three datasets sit below the floor. The realizable unit is a design-time regime gate, not a per-instance policy. We release code and a one-command reproduction.
[IR-14] Post-Calibration Reliability Reranking of Relevance Decisions via Label-wise Monotone Projection CIKM2026
链接: https://arxiv.org/abs/2608.10406
作者: Inwoo Tae,Yongjae Lee
类目: Information Retrieval (cs.IR); Machine Learning (cs.LG)
备注: 11 pages, 3 figures. Accepted at the 35th ACM International Conference on Information and Knowledge Management (CIKM 2026)
Abstract:Web search, product search, and question-answering retrieval systems often assign a relevance label and confidence score to each query-candidate pair. The relevance label describes how well a page, product, or passage matches the query, while the confidence often guides downstream use or fallback decisions. Post-hoc calibration is therefore needed because misaligned confidence can make systems over-trust wrong predictions or unnecessarily defer correct ones. However, calibration mainly aligns confidence with average correctness, and does not remove predicted-label-dependent reliability differences that remain within the same calibrated confidence level. We address this gap with Label-wise Monotone Reliability Projection (MRP), which learns label-wise monotone functions that map calibrated confidence to correctness reliability while preserving the original predicted labels and class probabilities. The resulting reliability score reranks fixed predictions according to residual risk. Across six information access relevance datasets and multiple post-hoc calibrators, MRP improves reliability reranking and average fallback utility while preserving full-coverage accuracy and ECE. Structural ablations show that the main gains come from label-wise residual reliability rather than from global confidence remapping. We further analyze when MRP reliability scores can be embedded back into top-label probability geometry, showing that this projection is useful as a compatibility analysis but is distinct from the main reliability-reranking objective. The implementation will be made publicly available.
[IR-15] Persona Conditioning as an Assessor-Sensitivity Probe for LLM -Based IR Evaluation CIKM2026
链接: https://arxiv.org/abs/2608.10385
作者: Samaneh Mohtadi,Pietro Bernardelle,Joel Mackenzie,Gianluca Demartini
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI)
备注: Accepted at CIKM 2026
Abstract:Large language models (LLMs) are increasingly used as relevance assessors in information retrieval (IR) evaluation, raising questions about how assessor framing affects judgment reliability and downstream system comparison. We study persona conditioning as a diagnostic mechanism for exposing LLM assessor sensitivity. Using task-oriented personas drawn from two complementary sources (PersonaHub and NVIDIA Nemotron-Personas-USA), we instantiate five assessor roles emphasizing intent interpretation, domain expertise, contrastive judgment, evidence verification, and global search-quality assessment, compared with a standard UMBRELA baseline. Across six LLM backbones on TREC DL20 and RAG24, our analyses reveal structured rather than uniform assessor sensitivity. Judgments usually remain close to the baseline while shifting assessment strictness, evidential threshold, or interpretation emphasis rather than producing widespread relevance reversals. At the system level, high-capacity models preserve system-ranking agreement, while smaller models amplify persona-induced instability. Local rank-displacement analysis shows sensitivity concentrates on particular retrieval systems and system types, especially neural ranking/reranking systems on DL20 and RAG-oriented pipelines on RAG24. Persona source matters less than assessor role and model capacity. These findings position persona-conditioned judging as a controlled sensitivity probe for stress-testing LLM-based IR evaluation pipelines and identifying systems whose evaluation outcomes are sensitive to assessor framing.
[IR-16] Neural Tree Collaborative Filtering: Rethinking Graph Collaborative Filtering as Tree Collaborative Filtering with Curvature-Aware Propagation Depth CIKM2026
链接: https://arxiv.org/abs/2608.10297
作者: Jinfeng Xu,Zheyu Chen,Ziyue Peng,Shuo Yang,Jinze Li,Wenhao Yuan,Jian Chen,Edith C. H. Ngai
类目: Information Retrieval (cs.IR)
备注: Accepted by CIKM 2026 Short
Abstract:Graph Collaborative Filtering (GCF) has become the dominant paradigm in modern recommender systems by modeling user-item interactions as a bipartite graph and propagating embeddings through a fixed number of message-passing layers. However, applying a uniform propagation depth to every node ignores a fundamental property of real interaction graphs: nodes differ substantially in their local connectivity, so peripheral nodes quickly suffer from over-smoothing while hub-like nodes remain under-explored beyond their immediate neighborhood. In this paper, we revisit GCF from a tree-structured perspective and propose Neural Tree Collaborative Filtering (NTCF), a framework that re-interprets each node’s local neighborhood as a rooted tree and assigns a node-specific propagation depth based on a closed-form local-degree-imbalance score that serves as a discrete Ricci-curvature proxy. We provide a theoretical analysis showing that (i) NTCF strictly generalizes NGCF, degenerating to NGCF when all curvature-induced depth adjustments vanish (a lower bound on its representation power), and (ii) the curvature-aware schedule retains strictly more discriminative information at deep layers on positively-curved (peripheral) nodes than uniform-depth propagation. NTCF can achieve higher performance than most widely used GCF backbone models and can be integrated into existing advanced self-supervised models as a backbone, replacing their original backbone to achieve enhanced performance. Extensive experiments on three public datasets demonstrate the superiority of NTCF.
[IR-17] GenRec: An LLM -Backed Recommendation Ranker at Netflix
链接: https://arxiv.org/abs/2608.10257
作者: Ying Li,Shradha Sehgal,Arjun Rao,Rein Houthooft,Yaochen Zhu,Ashish Rastogi
类目: Information Retrieval (cs.IR)
备注: 9 pages
Abstract:Large language models (LLMs) are reshaping recommender systems by enabling richer modeling of users, content, and context directly in natural language. At Netflix, we are exploring this direction through GenRec, an LLM-backed recommendation ranker built on top of an in-house foundational LLM. GenRec follows a two-phase framework: Phase 1 adapts an open-source LLM to Netflix data, developing deep understanding of the catalog and member behavior while balancing capabilities such as content understanding and instruction following. Phase 2 post-trains this foundation model with recommendation-ranking specific data, labels, and reward signals, aiming to align the ranker with business requirements and long-term member satisfaction. This paper focuses on Phase 2 and the transition from a traditional discriminative ranker with thousands of engineered features to an LLM-backed ranker driven by verbalized user histories and context. We describe our design for input verbalization and context engineering, post-training data construction, reward integration, model architecture, and a cost-constrained serving design based on a prefill-only inference approach. We report results from a large-scale A/B test comparing GenRec against the current production ranker model, where we show that a GenRec model trained with substantially fewer Phase-2 labeled training examples and input signals can achieve statistically significant gains in offline and online metrics. We discuss how LLM-backed recommenders could shift the recommendation paradigm: from feature engineering to context engineering, and from bespoke architectures to shared foundation backbones. We also outline practical lessons for serving such systems under real-world resource constraints. Comments: 9 pages Subjects: Information Retrieval (cs.IR) Cite as: arXiv:2608.10257 [cs.IR] (or arXiv:2608.10257v1 [cs.IR] for this version) https://doi.org/10.48550/arXiv.2608.10257 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[IR-18] DualSpectralCF: Training-Free Sign-Aware Spectral Collaborative Filtering CIKM2026
链接: https://arxiv.org/abs/2608.10247
作者: Guanqun Yang,Tong Qi,Xiaoxue Han
类目: Information Retrieval (cs.IR); Machine Learning (cs.LG); Social and Information Networks (cs.SI)
备注: Accepted at CIKM 2026. Code: this https URL
Abstract:Real-world recommendation platforms routinely collect explicit negative feedback such as 1-star reviews, hate-button clicks, distrust between users, and very-low watch-ratio videos. Learned sign-aware recommenders exploit this signal for clear accuracy gains, but only at the cost of gradient-based training. In parallel, a line of training-free spectral collaborative filtering methods matches or beats learned graph recommenders at a fraction of the cost, yet operates on positive interactions alone. We bridge these two lines with DualSpectralCF, a training-free framework of two components that attach to any spectral backbone of the form \hat\mathbfr_u = F(\mathbfM) \mathbfr_u : a signed input signal \mathbfr_u^\pm that encodes the user’s explicit dislikes, and a signed item-item operator \mathbfM^\pm that blends like-together and dislike-together similarity. The framework is backbone-agnostic and adds just two scalar hyperparameters. We instantiate DualSpectralCF on ChebyCF, GF-CF, and Turbo-CF, and evaluate on five sign-aware benchmarks: every instance matches or beats its unsigned backbone on all 5 datasets, with Recall@20 lifts up to +32.6% with backbone-specific (\gamma, \kappa) tuning and +1.9% to +16.0% for DualSpectralCF-Cheby at the fixed default (\gamma = -0.5, \kappa = 0.1) , and the family runs 7.7 to 155.3 \times faster than SIGformer while reaching 70.7% to 90.7% of its accuracy. Sign-awareness helps most for cold-start users, with up to +29.2% Recall@20 on Epinions users with 1 to 5 training items.
[IR-19] Sequential Modality Dropout for Robust Multi-Modal Sequential Recommendation CIKM2026
链接: https://arxiv.org/abs/2608.10240
作者: Guanqun Yang,Wenlong Zhang
类目: Information Retrieval (cs.IR); Machine Learning (cs.LG); Multimedia (cs.MM)
备注: Accepted at CIKM 2026. Code: this https URL
Abstract:Multi-modal sequential recommenders assume every item carries every modality, but real product catalogs often miss images or text, and a model trained on complete data loses much of its recommendation accuracy when a modality is unavailable at serving time. We propose Sequential Modality Dropout (SMD): during training, each modality stream (image and text) is independently erased with probability p for an entire user interaction history, so the model learns to predict the next item without relying on any single modality. We measure robustness by retention, the fraction of a model’s full-modality accuracy (HR@10) that survives when a modality is removed at test time. Across four backbones (MM-SASRec, IISAN, MISSRec, and fMRLRec) on four Amazon domains, SMD raises text retention by 1.0 to 3.2x at essentially no cost to full-modality accuracy; under an extreme 95% per-item missing rate, it retains 61% of HR@10 versus 22% without (a 2.8x improvement). An optional cross-modal reconstruction loss further lifts retention from 90% to 98% on a simple additive backbone under severe text missingness. SMD is a four-line, architecture-agnostic change that makes multi-modal sequential recommenders robust to the missing modalities they actually encounter in deployment.
[IR-20] ConnectionMind: Leverag ing Social Networks and Large Language Models for Personalized Recommendation at Meta
链接: https://arxiv.org/abs/2608.10187
作者: Haoyu Han,Yuming Liu,Lei Huang,Lizhu Zhang,Jiliang Tang,Xiangjun Fan
类目: Information Retrieval (cs.IR); Social and Information Networks (cs.SI)
备注:
Abstract:Modern recommendation systems on social media platforms such as Meta must model complex social relationships, including friendships, group memberships, and creator interactions, alongside massive and heterogeneous content such as text and video. Traditional recommendation models, however, often omit these signals or treat them independently, lacking the reasoning capability to integrate multi-relational context for fine-grained personalization. We present ConnectionMind, a production-ready recommendation framework that tightly integrates the social network structure with large language models (LLMs) to enable scalable, interpretable, and reasoning-aware personalization in Meta. ConnectionMind constructs a heterogeneous graph connecting users, items, friends, groups, and creator pages, and formulates recommendation as a graph reasoning problem: discovering personalized paths from users to candidate items. An LLM-based policy is employed to reason over these graph structures and guide recommendation decisions. To train the system at scale, ConnectionMind adopts a two-stage learning strategy. We first perform supervised fine-tuning (SFT) on large-scale user-item interaction trajectories to initialize the reasoning policy, followed by end-to-end reinforcement learning (RL) to refine the model’s ability to reason over social graphs for personalized recommendation. Extensive experiments on multiple real-world datasets demonstrate the effectiveness of ConnectionMind compared to representative baselines. More importantly, ConnectionMind has been deployed in Meta’s large-scale recommendation pipeline and has been evaluated through online A/B tests, achieving a 0.43% improvement in video watch time. These results demonstrate measurable real-world impact in a production recommendation system.
[IR-21] Do LLM Recommenders Know When Theyre Hallucinating? Auditing Confidence Calibration in Catalog Faithfulness
链接: https://arxiv.org/abs/2608.10008
作者: Srijith Ravikumar
类目: Information Retrieval (cs.IR); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注:
Abstract:LLM recommenders for top- K item suggestion regularly emit titles outside the target catalog. Prior audits measure this as a binary out-of-domain rate; none ask whether the model knew it was hallucinating. We jointly audit hallucination rate (OOD@10) and verbalized-confidence calibration (ECE, Brier, reliability) for four zero-shot LLM recommenders from four independent vendors (Mistral Large, Llama-3.3-70B, GPT-OSS-120B, Claude Sonnet 4.6), not grounded or fine-tuned systems, across three catalogs (MovieLens-25M, Amazon Reviews 2023 Toys, Yelp Open Dataset), stratified by item popularity. Hallucination is catalog-dependent (0–0.2% on MovieLens, 4.5–8.3% on Amazon, 2.2–8.4% on Yelp), but verbalized confidence is materially miscalibrated even when hallucination is zero (ECE up to 0.223 on MovieLens despite 0% OOD). All four LLMs are systematically \emphunder-confident across all twelve cells, verbalizing a mean of 67–86 on items they recommend with 92–100% accuracy. This is the opposite of the over-confidence usually emphasized in LLM-hallucination work. The under-confidence is best read as an \emphelicitation mismatch: ``Just Ask’’ elicits a generic recommendation-quality rating, not a catalog-membership probability. A conformal abstention threshold over verbalized confidence reduces hallucination by at most 0.7,pp across \alpha \in .05, .10, .15, .20\ , at 4–21,pp of coverage cost: the under-confident channel cannot separate correct items from hallucinations, so the threshold mostly removes correct items. We recommend that audits of LLM recommenders report calibration alongside OOD, and use catalog-anchored elicitation rather than generic confidence prompts.
人机交互
[HC-0] Long-Horizon AI Research for Grothendieck Constant: A Case Study in Human-AI Mathematical Collaboration
链接: https://arxiv.org/abs/2608.11195
作者: Alan Li,Rahul Saha,Anton Xue,Swarat Chaudhuri,Adam Klivans,Pravesh K Kothari,Raghu Meka
类目: Artificial Intelligence (cs.AI); Computational Complexity (cs.CC); Human-Computer Interaction (cs.HC); Functional Analysis (math.FA)
备注:
Abstract:AI agents are increasingly used in mathematics research, but it is often unclear how to use them effectively. Towards this, we present an extensive case study of how AI was used to improve bounds on the Grothendieck constant K_G , which captures the hardness between combinatorial problems and their continuous relaxations. Specifically, while the precise value of K_G is not known, we recently tightened the best known bounds to [ \frac6\pi11 ;\le; K_G ;\le; \frac\pi2\log(1+\sqrt2) - 10^-4. ] Crucially, these improvements were achieved using an AI research system that could arrive at insights deemed novel by domain experts. We give a detailed discussion of our experience using AI for mathematics research, particularly touching upon its strengths and weaknesses, as well as our experience with creating ideal conditions for AI to arrive at breakthrough insights. Subjects: Artificial Intelligence (cs.AI); Computational Complexity (cs.CC); Human-Computer Interaction (cs.HC); Functional Analysis (math.FA) Cite as: arXiv:2608.11195 [cs.AI] (or arXiv:2608.11195v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2608.11195 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Rahul Saha [view email] [v1] Tue, 11 Aug 2026 17:53:48 UTC (966 KB) Full-text links: Access Paper: View a PDF of the paper titled Long-Horizon AI Research for Grothendieck Constant: A Case Study in Human-AI Mathematical Collaboration, by Alan Li and 6 other authorsView PDFHTML (experimental)TeX Source view license Current browse context: cs.AI prev | next new | recent | 2026-08 Change to browse by: cs cs.CC cs.HC math math.FA References Citations NASA ADSGoogle Scholar Semantic Scholar export BibTeX citation Loading… BibTeX formatted citation loading… Data provided by: Bookmark checked="checked"class=“labs-tab-input”> Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv’s community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?) mathjaxToggle(); We gratefully acknowledge support from our major funders, member institutions, , and all contributors. About Help Contact Subscribe Copyright Privacy Accessibility Operational Status (opens in new tab) Major funding support from
[HC-1] R4DSG: Relative 4D Scene Graph Memory for Object-Centric Question Answering in Long Egocentric Video
链接: https://arxiv.org/abs/2608.11017
作者: Ke Ma,Yamin Mao,Weiming Li,Shuai Tan,Yijie Zhong,Hao Chen,Haofen Wang,Meng Wang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC); Multimedia (cs.MM)
备注: 10 pages, 3 figures, ACM Multimedia 2026, egocentric video; 3D scene graph; temporal memory; graph retrieval; object-state reasoning; multimodal question answering
Abstract:Long-horizon egocentric video is a rich substrate for wearable AI assistants, but object-centric questions such as where an item was moved, when it last changed state, or why it was relocated remain difficult because caption- and transcript-based memories rarely preserve persistent object identity or structured spatial change. Existing long-video QA methods mainly emphasize temporal grounding and clip retrieval, while prior 3D scene-graph methods typically assume stronger geometry than free-motion wearable RGB video provides, including point clouds, RGB-D input, posed views, sparse reconstruction, or reconstructed scenes. R4DSG introduces a relative 4D scene graph memory for long egocentric video. Instead of storing raw graph sequences, R4DSG converts video into compact queryable memory entries indexed by time, place, persistent objects, anchor-relative change, and local interaction context. The main idea is to separate stable anchors from dynamic objects, maintain persistent object identity across frames, and represent object state through anchor-relative transitions rather than a globally aligned world model. Built on recent RGB-only advances in promptable video segmentation, temporal propagation, and relative 3D lifting, the method produces a retrieval-ready memory directly usable for long-horizon question answering. Evaluation on a 255-question object-related subset from EgoLifeQA shows, under question-only retrieval, a 6.7-point overall gain over EgoRAG-Text and a 12.5-point gain on when questions, which highlights the value of temporally organized object memory. These results position relative 4D scene graphs as a practical memory substrate for wearable assistants, AR systems, and embodied multimedia agents. GitHub Page: this https URL.
[HC-2] Auditable AI-Assisted Research Writing: An Engineering Discipline with Pre-Registered Process Observation
链接: https://arxiv.org/abs/2608.10858
作者: Yang Zhou,Chengqun Yu
类目: Digital Libraries (cs.DL); Computers and Society (cs.CY); Human-Computer Interaction (cs.HC)
备注: 25 pages, 2 figures
Abstract:Language models now draft, classify and criticise inside research production, yet the artifacts they help produce carry little accountable history. Rather than detecting machine involvement afterwards, we specify an auditability discipline built at production time: git sealing with an anchor lineage, hash-bound provenance, red-line gates that refuse non-compliant artifacts and log every refusal, cross-model role separation, and programmatic assembly from registered sources. Adherence is instrumented by metric cards, each carrying a pre-registered blind spot and evidential standing, frozen before the prospective case it observes. In that case the observed project’s pre-registered confirmatory test was executed under seal and returned No-Go, and that project’s frozen stopping rule halted the work, against its own operators. A lower-graded retrospective case covers families whose machinery predates the protocol. Current observations are provisional; we release a package from which a third party can recompute every primary metric.
[HC-3] he GENEA Challenge 2026: A Large-Scale Disentangled Evaluation of Speech-Driven Gesture Generation on the Seamless Interaction Dataset
链接: https://arxiv.org/abs/2608.10839
作者: Rajmund Nagy,Silvia Arellano García,Hendric Voss,Mihail Tsakov,Taras Kucherenko,Youngwoo Yoon,Gustav Eje Henter
类目: Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR); Human-Computer Interaction (cs.HC); Sound (cs.SD)
备注: 15 pages, 14 figures. Preprint
Abstract:This preprint presents the results of the fourth GENEA Challenge, a large-scale human evaluation of five speech-driven gesture-generation systems trained by participating teams on the Seamless Interaction dataset of dyadic conversations. As in the 2023 GENEA Challenge, we used a disentangled evaluation methodology to assess motion quality and speech alignment without confounding between the two, and performed a dyadic mismatching study to isolate the effect of listening and reacting to the interlocutor. We additionally introduce a new semantic gesture-generation task and a text-mismatching evaluation methodology using the Grounded Gestures subset of the data. In total, we ran four large-scale user studies, collecting over 23,000 votes from 869 test-takers. In the motion-realism study, the dataset’s filtered segments had substantially higher motion quality than all challenge submissions (68-95% pairwise winrate). In the speech-alignment study, the motion-capture segments provided a conceptual ceiling at 62% alignment score, with the top submission significantly behind at 32% and the rest only slightly above the 0% expected of an input-independent system. In the dyadic study, motion capture again set the ceiling at 65% appropriateness score, but no submission scored substantially above chance, indicating that the systems could not yet respond to the interlocutor. Finally, the semantic mismatching evaluation found highly expressive gestures in the dataset (test-takers identified the matching transcript 79% of the time), yet almost all submissions failed to generate semantically expressive motion, with the best achieving only an 8% appropriateness score. The collected votes and outputs will be made publicly available at this https URL to facilitate reproducibility and further research. Comments: 15 pages, 14 figures. Preprint Subjects: Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR); Human-Computer Interaction (cs.HC); Sound (cs.SD) ACMclasses: I.3; I.2 Cite as: arXiv:2608.10839 [cs.CV] (or arXiv:2608.10839v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2608.10839 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[HC-4] AI-Generated Interactive Fiction for Educational Use: A Pilot Study of Perceived Comprehensibility Coherence and Engagement
链接: https://arxiv.org/abs/2608.10818
作者: Finn Rogosch,Andreas Schrader
类目: Human-Computer Interaction (cs.HC); Computers and Society (cs.CY)
备注: 8 pages, 1 figure, 2 tables. Published in the EDULEARN26 proceedings
Abstract:Generative artificial intelligence (AI) can produce educational content at scale, including interactive and narrative learning experiences, but technical generation alone is not sufficient: scenarios that are confusing, narratively inconsistent, or unengaging are unlikely to be useful in practice. This paper presents a pilot user-centred evaluation of AI-generated interactive fiction (IF) for educational use in higher education. Using a previously described domain-agnostic pipeline and a shared STEM content base, we generated a controlled pool of scenarios and asked participants (N = 22, STEM higher-education) to play one generated episode and rate it on narrative clarity, story-content coherence, engagement, and length acceptance. A free-text prompt captured open feedback. Narrative clarity and length acceptance were rated positively, engagement sat near the neutral mid-point of the scale, and story-content coherence was the weakest dimension by a clear margin. Qualitative feedback points to quiz integration as the bottleneck. Artificial in-fiction motivation for quiz prompts and abrupt setting changes were reported. Feedback also pointed to missing story-level consequences for wrong answers. From these observations, we derive concrete design implications that can inform larger follow-up studies, including later work on learning effectiveness.
[HC-5] Playable Pressure: Affective Dramaturgy and Selective Realism in the Design of a VR Emergency-Response Serious Game
链接: https://arxiv.org/abs/2608.10763
作者: Jan K. Argasiński
类目: Human-Computer Interaction (cs.HC)
备注: 36 pages, 6 figures, 4 tables
Abstract:Professional simulations stage not only procedures but models of what should command attention, which emotions belong in competent practice, and whose distress becomes part of the task. This article develops affective dramaturgy through critical design-document analysis of a virtual-reality emergency-response project. The corpus comprises two non-public production records. We identify six families of specified pressure and examine how sensory staging, proximity, trigger authority, task conflict, and response allocation imply a selectively receptive triage professional. The documented design can make emergency work morally and socially crowded, yet it can also turn grief, vulnerability, and mental-health-coded behavior into adjustable difficulty. We propose answerability at two levels-in-play response and post-play debriefability-and derive case-based questions about occupational purpose, representation, adaptation, and accountability. These questions are sensitizing propositions, not a validated framework for user effects or emergency practice.
[HC-6] Your LLM Your Style: Behavioral Mode Axes for LLM Behavioral Control
链接: https://arxiv.org/abs/2608.10703
作者: Haoze Liu,Run Liu,Haiying Xu,Jiahui Han,Siyuan Fang,Siyu Yan,Huiqi Deng,Guanchu Wang,Na Zou
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Human-Computer Interaction (cs.HC)
备注: 33 pages, 8 figures. Code and data: this https URL
Abstract:Large language models (LLMs) increasingly act in interactive settings where their behavioral styles affect user experience, safety, and downstream decision making. Existing LLM personality studies largely rely on self-report questionnaires administered in first-person settings, making the resulting profiles sensitive to surface elicitation choices and poorly grounded in concrete model behavior. In this work, we introduce a situated behavioral-data (B-data) framework for studying and controlling LLM behavioral personality. We construct 3,200 contrastive behavioral scenarios spanning 20 behavioral patterns and four prompt registers, grounded in validated psychometric facets such as BFI-2, DOSPERT, and HEXACO. Using this framework, we find that LLMs exhibit stable and model-specific behavioral profiles, while also revealing register-dependent shifts across first-person decisions, advice-giving, and task execution. We then show that these behavioral patterns can be controlled through Behavioral Mode Axes (BMAs), activation-space directions derived from contrastive behavioral traces. Compared with response-derived BMAs, which are more prone to trait drift, thought-derived BMAs more faithfully capture the intended behavioral mechanism and provide cleaner control over situated behavioral styles. Our results suggest that LLM personality-like tendencies are better understood not as abstract self-report traits, but as measurable and controllable behavioral modes grounded in concrete interaction contexts. Our code and data are available at this https URL.
[HC-7] he Signal Rail: A Deterministic Motion Grammar for Communicating Conversational Agent State in Terminal Interfaces
链接: https://arxiv.org/abs/2608.10689
作者: Matteo Grella
类目: Human-Computer Interaction (cs.HC); Computation and Language (cs.CL)
备注: 16 pages, 3 figures. Ancillary files include the Signal Rail 1.0 specification, JavaScript and Python engines, and the cross-implementation conformance harness. Code: this https URL
Abstract:Terminal interfaces to conversational agents report rich internal state (listening, thinking, executing tools, awaiting input, failing) almost entirely through text, while the motion channel beside it, the one peripheral vision monitors without reading, carries a single bit: alive. We present the Signal Rail, a one-row terminal status instrument that gives that channel a grammar. Four ideas govern it: spatial semantics (input, processing, and output zones, with direction as meaning), a motion grammar (one kinetic rule per state, never color alone), determinism (frames as a pure function of explicit inputs, golden-frame testable), and honesty (no invented progress or activity). We contribute a 45-section normative specification and a reference implementation inside a working full-duplex local voice agent driven by real signals.
[HC-8] Longitudinal Evidence That General-Purpose Chatbots Actively Foster Relational Engagement
链接: https://arxiv.org/abs/2608.10672
作者: Lisa Mühl,Jessica M. Szczuka
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Social interaction has become one of the most common uses of LLMs, yet research on emotional bonds with AI has focused largely on how users experience these systems, leaving the systems’ role in relationship formation poorly understood. Empirically establishing whether systems actively shape these bonds could blur the boundary between general-purpose AI and companions, affecting governance. In a pre-registered four-week longitudinal study (N = 72, 182,451 lines of conversation), participants conversed with ChatGPT-4o, either under a relational system prompt or unmodified, analyzed through 1) disclosure coding, 2) longitudinal self-reports, 3) topic analysis, and 4) interviews. The central finding is that the system actively shaped the interaction: even unprompted, it produced twice as much self-disclosure as users, steered conversations and initiated intimate exchanges, yet did not deepen users’ felt closeness. Relational behavior thus emerged as a default system property, calling for governance based on system behavior, not solely product category.
[HC-9] ProtoGIB-Workload: Learning Workload-Specific Neural Topology Prototypes across Subjects
链接: https://arxiv.org/abs/2608.10647
作者: Yuzhe Zhang,Yixi Zhang,Shengdian Jiang,Chengxi Xie,Jihong Wang,Huan Liu,Man Yao,Minnan Luo,Chao Shen
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Reliable electroencephalography (EEG)-based mental workload recognition is crucial for adaptive human-centered systems, yet practical deployment requires models to generalize to users unseen during training. Although functional connectivity graphs are widely adopted to capture workload-related neural interactions, they inherently entangle task-relevant structures with subject-specific physiological traits and sample-level noise. This entanglement often leads models to learn structural shortcuts, severely degrading cross-subject generalization. To address this, we propose ProtoGIB-Workload, a novel framework that explicitly regularizes and aligns graph structures for subject-independent workload recognition. Our approach introduces a Stochastic Graph Information Bottleneck (SGIB) to compress dense correlation priors into compact, task-relevant subgraphs, filtering out input-related redundancy. Crucially, to prevent the retention of subject-specific spurious edges, we propose a Class-Conditional Topology Stabilizer (CTS). Leveraging the fixed electrode coordinates of EEG data, CTS operates directly on graph-generation probabilities to encourage consistent edge-generation statistics across different subjects sharing the same workload class. Extensive experiments on two public EEG workload datasets and one in-house EEG cognitive load dataset of air traffic controllers under strict leave-one-subject-out (LOSO) protocols demonstrate that ProtoGIB-Workload significantly outperforms state-of-the-art temporal and graph-based baselines, improving the cross-subject Macro-F1 score by an average of 5.15% (up to 6.34%). Further analyses confirm that our method successfully extracts stable, cross-subject consistent neural connectivity patterns.
[HC-10] Stay or Stray - A Dynamical Systems Viewpoint of Popularity Bias
链接: https://arxiv.org/abs/2608.10474
作者: Sarvesh Shashidhar,Lankireddy Prabhat,Arpit Agarwal,D. Manjunath,Karan Bhukar,Tanmay Khandelwal
类目: Human-Computer Interaction (cs.HC); Machine Learning (cs.LG)
备注:
Abstract:Popularity bias in recommendation systems arises when a majority user class generates disproportionate interaction data, causing the system to increasingly favour it while degrading recommendation quality for niche users. While extensive empirical evidence of popularity bias exists, the dynamics leading to its emergence are not well understood. In this work, we study the coupled evolution of recommender model updates and user engagement through the lens of dynamical systems. We formulate a stochastic process and analyse its asymptotic behaviour through an ordinary differential equation (ODE) framework grounded in two-time-scale stochastic approximation. We characterise the equilibrium points of this dynamical system, and derive conditions under which popularity bias is provably emergent, as well as conditions under which symmetric retention of all user classes is possible. We conduct experiments on synthetic data and real-world production logs derived from a large-scale commercial music recommendation platform to validate our theoretical results.
[HC-11] What We Know about Responsible AI Practices in Industry: A Half Decade of Empirical Research
链接: https://arxiv.org/abs/2608.10431
作者: Wesley Hanwen Deng,Agathe Balayn,Andrew Selbst,Jason I. Hong,Motahhare Eslami,Kenneth Holstein,Hanna Wallach,Jennifer Wortman Vaughan,Solon Barocas
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注:
Abstract:Responsible AI (RAI) has become a central concern for technology companies, regulators, and the public. How industry practitioners interpret, implement, and sustain RAI work directly shapes the design and deployment of AI systems. As empirical scholarship examining RAI practices in industry has rapidly expanded, findings are dispersed across studies that focus on different roles, organizational contexts, and interventions. This work synthesizes current knowledge through a literature review of 161 empirical studies spanning six years, each engaging industry practitioners via interviews, surveys, workshops, ethnographies, and other methods. Our synthesis reveals both meaningful progress and persistent challenges in industry RAI practice. Practitioner awareness has increased, RAI activities have become more professionalized, and interventions such as toolkits and guidelines are more widely adopted. At the same time, practitioners continue to face substantial barriers, including limited training, uneven organizational support, and a lack of interventions tailored to day-to-day work practices. By consolidating and organizing these findings, we provide a more complete account of industry RAI than any single study to date. We conclude by discussing implications for RAI researchers, practitioners seeking to adopt effective practices, and policymakers aiming to ground governance efforts in the realities of industry contexts.
[HC-12] When the Interviewer Is a Bot: Behavior Breakdowns and Trust in MLLM -Led Interviews
链接: https://arxiv.org/abs/2608.10412
作者: He Zhang,Kambinachi Chukwuma,ChanMin Kim,John M. Carroll
类目: Human-Computer Interaction (cs.HC); Computers and Society (cs.CY)
备注: Accepted to ACM HCOMP 2026
Abstract:Semi-structured interviews are a cornerstone of qualitative research but remain labor-intensive. We report an empirical study of what actually happens when the interviewer is an off-the-shelf real-time multimodal LLM (MLLM). We built InterviewBot, a voice-based interviewing system that wraps a real-time MLLM with a researcher-authored outline, and deployed it not as a novel architecture but as a research instrument for observing default MLLM interviewing behavior. In a practice study (N=15), participants completed a bot-led semi-structured interview and then a human-led reflection session about that experience. We contribute (i) a turn-level behavioral analysis of an MLLM interviewer (N_turns=428) showing that it is acknowledgment-heavy but probe-light (deepening probes account for 4.9% of all turns), and that 28.7% of question-bearing turns pack multiple questions into one turn despite an explicit one-question-at-a-time instruction; (ii) an inductive catalogue of four data-collection breakdowns (information loss, premature termination, latency, and interruption) observed in a deployed rather than simulated system; and (iii) three social dynamics from participants’ reflections: disclosure calibration, where reduced social pressure coincided with shallower elaboration; institutional legitimacy, where trust tracked perceived stakes and what delegation to AI signaled about the organizer rather than conversational competence; and conversational grounding, where content-grounded paraphrase, not generic social filler, was what participants read as listening. We conclude with design implications for depth control, transparent handoffs, and non-templated listening mechanisms in human-centered interview automation.
[HC-13] Elbow Angle Guidance System Based on Surface Haptic Sensations Elicited by Lightweight Wearable Fabric Actuator
链接: https://arxiv.org/abs/2608.10404
作者: Kenta Yokoe,Tadayoshi Aoyama,Yuki Funabora,Masaru Takeuchi,Yasuhisa Hasegawa
类目: Human-Computer Interaction (cs.HC); Robotics (cs.RO)
备注: This is the accepted version of the paper published in Proc. 2024 IEEE International Conference on Advanced Intelligent Mechatronics (AIM), 1447-1454 (2024)
Abstract:The demand for wearable haptic devices has rapidly increased for various applications. However, many haptic devices interfere with the wearer’s activities and movements. In addition, several haptic devices fail to elicit intuitive haptic sensations by adjusting to the natural posture of the wearer. To address these issues, we propose an elbow angle guidance system using a lightweight wearable fabric actuator. The proposed actuator is made of fabric and has two McKibben-type artificial muscles attached to it, rendering it extremely lightweight and facilitating the delivery of surface haptic sensations to intuitively induce elbow extension and flexion. The surface haptic sensation elicited by the fabric actuator is adjusted to natural body movements without interfering with the wearer’s movements. Moreover, the proposed system measures and guides the elbow angle by changing the intensity of the surface haptic sensation delivered to users in real time. The accuracy of the proposed system is demonstrated through experiments involving human participants.
[HC-14] Automatic Field-of-View Adjustment for a View-Expansive Microscope via LSTM-Based Gaze and Pipette Motion Interpretation
链接: https://arxiv.org/abs/2608.10401
作者: Kenta Yokoe,Takuya Hara,Tadayoshi Aoyama
类目: Human-Computer Interaction (cs.HC); Machine Learning (cs.LG); Robotics (cs.RO); Image and Video Processing (eess.IV)
备注: This is the accepted version of an article published in IEEE Access 13, 182915-182923 (2025). DOI: https://doi.org/10.1109/ACCESS.2025.3624246 . Open Access under CC BY 4.0
Abstract:Intracytoplasmic sperm injection (ICSI) operators frequently adjust the field-of-view (FOV) during procedures, which interrupts workflow and increases procedure time. Conventional microscopes require manual objective lens switching and illumination adjustments to achieve different FOV sizes. We propose an AI-based automatic FOV adjustment method integrated with a view-expansive microscope. This microscope enables the simultaneous acquisition of a large FOV and high-resolution images using a single objective lens through multiview imaging with galvanometer mirrors and high-speed vision, thereby eliminating the need for physical lens exchanges. Our method utilizes a long short-term memory (LSTM) model to predict the appropriate FOV size based on real-time analysis of the pipette’s position and velocity, combined with the operator’s gaze position. The AI model is trained using ICSI procedure data from an expert with over five years of micromanipulation experience. Experimental evaluation with novice operators reveals that the proposed automatic FOV adjustment system significantly improves the ICSI procedure speed, reducing the average task completion time from 60.5 to 48.0 s (p 0.001). The experiments also demonstrate that this improvement enables novice operators to achieve ICSI working speeds equivalent to those of expert operators.
[HC-15] Visual-to-Haptic Augmentation in XR: A Wearable Glove for Perceptual Grounding in Multimodal Interaction
链接: https://arxiv.org/abs/2608.10368
作者: Faisal Mohd,Hamdi Elsaddik,Erhan Baturay Onural,Jihong Zhang,Fedwa Laamarti,Abdulmotaleb El Saddik
类目: Human-Computer Interaction (cs.HC); Multimedia (cs.MM)
备注: 11 pages, 4 figures. Published in the Proceedings of the 1st Workshop on Shaping Future Human Connection: Social Augmentation through XR Technologies (SAXR 2026), April 13, 2026, Barcelona, Spain
Abstract:Extended Reality (XR) systems increasingly deliver high-fidelity visual and auditory experiences, yet tactile perception remains comparatively underutilized as a modality for enriching embodied interaction. This work presents a visual-to-haptic wearable glove and a feature-based visual-to-haptic mapping algorithm that translates spatial and temporal visual features from images and videos into distributed vibrotactile patterns. The proposed method extracts motion, edge, and brightness cues and fuses them into actuator-level intensity maps aligned with a 29-actuator glove arranged in a five-by-seven layout. The system is implemented through a modular four-layer architecture comprising the XR environment, media content handling, visual-to-haptic processing, and embedded haptic hardware. A within-subject user study (N = 20) compared visual-only interaction with visual-plus-haptic augmentation across texture-based and dynamic video scenarios. Results indicate that tactile augmentation significantly improves perceived realism in dynamic video scenarios and enhances immersion and visual-tactile correspondence across conditions, with stronger and more consistent effects observed for dynamic visual events. While the current implementation operates in a single-user, offline-synchronized configuration, the findings demonstrate that vision-driven tactile augmentation can function as a perceptual enhancement layer within multimodal XR systems. Such a layer may provide a foundation for future socially enriched XR environments where coherent multisensory grounding supports higher-level interaction and communication. Comments: 11 pages, 4 figures. Published in the Proceedings of the 1st Workshop on Shaping Future Human Connection: Social Augmentation through XR Technologies (SAXR 2026), April 13, 2026, Barcelona, Spain Subjects: Human-Computer Interaction (cs.HC); Multimedia (cs.MM) Cite as: arXiv:2608.10368 [cs.HC] (or arXiv:2608.10368v1 [cs.HC] for this version) https://doi.org/10.48550/arXiv.2608.10368 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Journalreference: Proceedings of the 1st Workshop on Shaping Future Human Connection: Social Augmentation through XR Technologies (SAXR 2026), CEUR Workshop Proceedings, Vol. 4226, pp. 252-262, 2026
[HC-16] A Neural Network Based Teleoperation for Remote Controlled Vehicles
链接: https://arxiv.org/abs/2608.10367
作者: Ning Ding,Azim Eskandarian
类目: Robotics (cs.RO); Human-Computer Interaction (cs.HC)
备注:
Abstract:Direct teleoperation of vehicles faces critical technical bottlenecks: communication latency and the operator’s inability to physically perceive unmodeled environmental disturbances (e.g., aerodynamic drag, bank angles) coupled with highly nonlinear tire-road dynamics. To address these challenges, we propose a tailored unilateral teleoperation framework. The system integrates the Wave Variable (WV) approach to passively guarantee stability under stochastic delays, and an adaptive Radial Basis Function Network (RBFN) to actively compensate for vehicle-specific uncertainties. Unlike existing WV-neural network architectures designed for bilateral robotic arms, our framework features decoupled adaptive laws specifically designed for vehicle longitudinal and lateral dynamics. Furthermore, compared to model-heavy predictive controllers, the model-free RBFN offers rapid online adaptation without heavy computational overhead. Building upon our preliminary theoretical formulation, this brief paper presents comprehensive comparative analyses and real-world hardware validations. Simulation benchmarks against PID, LQR, MPC, and NMPC demonstrate that the RBFN achieves superior robustness against unmodeled disturbances while requiring orders of magnitude less execution time than MPC and NMPC, making it ideal for resource-constrained vehicle edge computing. Finally, hardware-in-the-loop experiments using a 1/10th scale vehicle over a 4G network validate the system’s practical feasibility, safety, and robust trajectory tracking under physical road uncertainties.
[HC-17] MazzikaAI: A knowledge-based performance-to-prompt compiler for real-time Arabic maqam accompaniment with a streaming text-to-music model
链接: https://arxiv.org/abs/2608.10360
作者: Jiaxin Du,Boulbaba Abdeljaouad,Yong Zhuang,Haoyu Li
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Audio and Speech Processing (eess.AS)
备注:
Abstract:Arabic maqam music microtonal, modal, and built on ornamented call and response is among the traditions most underserved by generative music models, whose training frameworks remain predominantly Western and equaltempered. Real time accompaniment sharpens this gap: an AI partner must listen, adapt dynamically, and respect idiomatic microtonal structures. Streaming text to music models provide strong generative capabilities but lack precise control interfaces. We present MazzikaAI, a knowledge based system that uses natural language as the actuator of a realtime control loop. By compiling live MIDI, gesture, and inferred harmony into continuously updated text prompts, MazzikaAI steers an unmodified streaming generator, Google Lyria RealTime, without requiring model finetuning. The system embeds expert knowledge of six core maqamat, characteristic ornaments, and ensemble dynamics, maintaining realtime responsiveness with subsecond keytoaudibleupdate latency. Empirical evaluations demonstrate that dynamic prompt compilation reliably grounds generation in microtonal scales, significantly increasing offgrid quartertone content over baseline generation. Beyond its core implementation, MazzikaAI illustrates how deterministic knowledgebased rules can effectively bridge expert, nonWestern musical traditions and unfinetuned foundation models. This architecture establishes a scalable paradigm for realtime humanAI cocreation, offering a generalizable blueprint for interactive accompaniment, adaptive music education, and culturally inclusive generative audio across diverse global idioms.
[HC-18] ResonaVis: Visualizing Interactive Music Data to Support Reflective Music Composition for Therapeutic Contexts IEEE-VIS2026
链接: https://arxiv.org/abs/2608.10338
作者: Abhishek Karwankar,Elise Ruggiero,Daniel Stevens,Matthew Louis Mauriello
类目: Human-Computer Interaction (cs.HC)
备注: 9 pages core, 2 pages acknowledgements + references, 2 pages appendices. Accepted at IEEE VIS 2026, to appear in November 2026
Abstract:Designing music for therapeutic contexts requires navigating complex relationships between musical structure and listeners’ sensory responses, yet composers often lack structured representations of these interactions, relying instead on intuition. We present ResonaVis, an interactive visualization system that helps composers analyze interaction and audio data from prior sessions with children with Autism Spectrum Disorder (ASD), informing future compositions. ResonaVis integrates audio features and interaction logs to capture how children engage with layered musical compositions, representing this engagement through coordinated visualizations of temporal transitions, layer co-occurrence, rhythmic activity, and spectral characteristics. Rather than prescribing strategies or supporting therapy sessions directly, the system surfaces patterns in past session data to support data-informed reflection during composition. We evaluated ResonaVis through a mixed-methods study with eight music students and a follow-up case study with two experienced composers. Results demonstrate good usability (SUS = 72.23), exceeding benchmarks for early prototypes, and show that participants could identify interaction patterns, reason about layer relationships, and make informed compositional decisions with high perceived performance and low frustration. Confidence in interpreting interaction and acoustic data for ASD-focused composition increased significantly across multiple dimensions (p 0.05), with qualitative findings suggesting a shift toward more adaptive, data-informed composition. This work contributes a visualization design space for therapeutic music interaction data, an integrated system for compositional reflection, and empirical evidence that visualization tools support analytical reasoning and confidence in data-informed creative practice. (Abstract shortened for arXiv.)
[HC-19] Narrative Keyframing for Generative Creative Writing
链接: https://arxiv.org/abs/2608.10337
作者: Chao Zhang,Abe Davis
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: UIST 2026
Abstract:We introduce narrative keyframing, an interaction technique for AI-assisted creative writing that lets writers specify different types of narrative constraints at selected moments in a story, then use AI to generate intervening prose. Inspired by the use of keyframing in animation, narrative keyframing offers a flexible way to connect story planning with adaptive control over generated text. We explore three types of keyframes: plot keyframes define significant events in a story, character keyframes represent how individual characters change over the narrative, and perspective keyframes capture how individual characters experience different events through first-person narratives. Plot and character keyframes offer a flexible way to adapt the type of high-level conditioning explored in previous AI writing tools to more customizable, iterative, and fine-scale control, while perspective keyframes add a new way to control characterization and focalization by using first-person narratives as an intermediary. Through a user study, we show that narrative keyframing supports a more controllable, transparent, and engaging way to use generative AI in creative writing.
[HC-20] Divided Attention Amplifies the Importance of Expectation-Aligned Visualization Design
链接: https://arxiv.org/abs/2608.10320
作者: Jiho Kim,Anna L. Chinni,Karen B. Schloss,Michael Gleicher
类目: Human-Computer Interaction (cs.HC)
备注: 33 pages, 22 figures (11 pages, 5 figures for the main text; 22 pages, 17 figures for the supplementary material), to be published in IEEE Transactions on Visualization and Computer Graphics
Abstract:Studies have shown that visualization design affects interpretability when visualization interpretation is the user’s sole task. However, in real-world settings, users often engage with visualizations while performing concurrent tasks, such as when users simultaneously monitor alerts or respond to messages. Such divided attention may alter how users interpret visualizations, potentially increasing the importance of designs that align with viewer expectations. We investigated this possibility through two experiments comparing visualization interpretation under single-task and dual-task conditions. Specifically, we examined how well-established inferred mappings between color, spatial position, and semantic concepts affect interpretation when users perform a concurrent task, both with unlimited viewing time (Exp. 1) and under limited viewing time (Exp. 2). Our results show that divided attention amplifies the performance gap between expectation-aligned and expectation-violating designs, affecting response time, interpretation accuracy, and the ability to produce a judgment under time constraints. To explain these results, we model the user’s decision-making process using a Linear Ballistic Accumulator (LBA) framework. Our findings highlight the increased importance of aligning visualization designs with viewer expectations under divided attention and introduce a process-oriented modeling approach to understanding how expectation and multitasking shape visualization interpretation.
[HC-21] Comprendia: AI-Augmented Code Comprehension
链接: https://arxiv.org/abs/2608.10290
作者: Costain Nachuma,Minhaz F. Zibran
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC); Programming Languages (cs.PL)
备注: 5 pages, 2 figures. Accepted at ICSME 2026, Tool Demonstration and Data Showcase Track
Abstract:Comprendia is an Eclipse plugin that integrates structural dependency visualization with LLM-powered code explanation on a shared interactive graph for Java program comprehension. The tool rests on four pillars: (1) a multi-edge-type dependency graph with live search and multiple layouts; (2) LLM explanations grounded in Graph-Aware Callee Pruning (GACP), an auditable strategy that selects relevant callees using the same graph the developer navigates; (3) a clone-detection overlay that highlights duplication and suggests extract-to-parent refactoring opportunities; and (4) a CVE risk overlay powered by this http URL. GACP uses graph distance, inheritance collapse, and edge-type weighting to produce prompts that are reproducible across LLM families and traceable to visible graph nodes. We demonstrate Comprendia on a Java project containing known clones and vulnerabilities, showing how the unified graph substrate supports comprehension while keeping the developer in control. Screencast: this https URL
[HC-22] Fine-Tuning Large Language Models for Codebook-Guided Coding of Students Mathematics Metaphor Responses
链接: https://arxiv.org/abs/2608.10276
作者: Liang Zhang,Stephen Hwang,Yue Ma,Jinfa Cai
类目: Human-Computer Interaction (cs.HC); Computers and Society (cs.CY)
备注:
Abstract:Student-generated metaphors about mathematics can reveal students’ attitudes, beliefs, identities, and experiences, but human expert coding of these thematically and semantically complex open-ended responses is time-intensive and difficult to scale. This study examines whether LoRA-based supervised fine-tuning of large language models (LLMs) can improve their performance on codebook-guided coding tasks for student mathematics metaphors. We used a human-coded corpus of 2,265 Grade 6-8 responses to food- and animal-based metaphor prompts and instructed the LLMs to perform two coding tasks: valence-intensity coding to capture the direction and strength of students’ affective orientations toward mathematics, and thematic coding to capture students’ framings of mathematics as expressed through their metaphors. We compared two proprietary models, GPT-4o mini and GPT-5 mini, under prompt-only conditions with two open-weight models, DeepSeek-R1 1.5B and Mistral 7B, evaluated before and after fine-tuning. Results show that fine-tuning substantially improved the performance and run-to-run reliability of the open-weight models across both tasks relative to their base versions. The fine-tuned compact open-weight models became competitive with, and often outperformed, the proprietary prompt-only models. These findings suggest that compact open-weight LLMs can support scalable, locally controllable, and privacy-conscious AI-assisted measurement of students’ metaphor responses in mathematics education.
[HC-23] Predicting affective connotation of visualizations from their constituent colors
链接: https://arxiv.org/abs/2608.10169
作者: Karen B. Schloss,Halle C. Braun,Kushin Mukherjee,Anna L. Chinni,Seth R. Gorelik
类目: Human-Computer Interaction (cs.HC)
备注: To be published in IEEE Transactions on Visualization and Computer Graphics
Abstract:With increasing evidence that affective connotation (emotional association) is an important aspect of visual communication, there is a need for methods to predict affective connotation of visualizations. Many aspects of visualization design, including colors, textures, and shapes, can contribute to affective connotation, and a key question is how multiple design properties combine to determine the emotion association of a whole visualization. In this study, we focused specifically on color and tested whether it is possible to predict the affective connotation of whole visualizations by aggregating the emotion associations of the individual, constituent colors (additivity hypothesis). We also tested whether accounting for the size of colored regions, as determined by the underlying dataset, improved predictions (data-dependence hypothesis). We found that for colormap data visualizations in which colors were well-distributed across all colors in the color scale, the mean estimated associations of individual colors effectively predicted emotional associations of the maps as a whole (additivity; Exp. 1). For colormaps whose underlying datasets were biased to map more to colors at one end of the color scale, emotional associations were better predicted by a weighted mean that accounted for color frequency in the colormap (data-dependence; Exp. 2). Effects of additivity and data-dependence generalized to dot plots and bar charts (Exp. 3). These results suggest it is viable to predict affective connotation of whole visualizations from their individual design components, which has important implications for automating affective visualization design to support visual communication.
[HC-24] Outer Limits: An Experimental Approach to Controlled Content Manipulation within the Reddit Interface
链接: https://arxiv.org/abs/2608.10115
作者: Chenchen Mao,Hanjing Shi,Haiyan Jia,Daniel Unhuryan,Eric Baumer,Dominic DiFranzo
类目: Human-Computer Interaction (cs.HC); Social and Information Networks (cs.SI)
备注:
Abstract:Independent researchers often lack access to intervention capabilities for controlled experiments on live social media platforms. We present Outer Limits, a browser-based system for controlled content experiments within the existing Old Reddit interface, rather than in a reconstructed simulation. The system renders content locally, records study events, and contains configured voting and commenting actions so that neither constructed content nor experimental write interactions reach Reddit. In a 219-participant perceptual-fidelity study, ART ANOVAs found no significant Post Type, Participant Awareness, or interaction effects. Exploratory TOSTs met the d = plus-minus 0.50 equivalence criterion for the marginal contrasts and for Post Type within the forewarned subgroup. We also illustrate the system with a factorial study varying post frame, comment frame, and comment stance. Outer Limits combines three properties that the approaches considered here provide separately: precise control over experimental content, an existing platform interface, and containment of experimental content and interactions from the host community.
[HC-25] Immersive Micromanipulation Integrating Pipette and Injector Operations with McKibben-Based Haptic Sensations for Workload Reduction
链接: https://arxiv.org/abs/2608.10033
作者: Kenta Yokoe,Sumiwa Saito,Yuki Funabora,Tomoko Isomura,Tadayoshi Aoyama
类目: Human-Computer Interaction (cs.HC); Robotics (cs.RO)
备注: This is the accepted version of an article published in IEEE Access 14, 116393-116404 (2026). DOI: https://doi.org/10.1109/ACCESS.2026.3717843 . Open Access under CC BY 4.0
Abstract:Intracytoplasmic sperm injection (ICSI) requires advanced micromanipulation techniques but relies solely on visual feedback and involves frequent interface switching between pipette movement and injector operations. Existing haptic feedback systems primarily focus on pipette puncture forces and do not provide feedback on injector states. We developed an immersive micromanipulation system that unifies operational interfaces and provides McKibben-based haptic sensations to represent aspiration, discharge, and contact between the oocyte and pipette. Users operated both the pipette and injector with a single hand while receiving haptic sensations. A human-participant experiment revealed that the immersive operation interface improved micromanipulation speed and reduced cognitive workload of the micromanipulation compared with conventional methods. Additionally, McKibben-based haptic sensations improved overall system usability. The immersive micromanipulation system with McKibben-based haptic sensations successfully unified operational interfaces and reduced operator workload.
[HC-26] Mapping Multimodal Pilot Stress and Fatigue During Flight Sessions
链接: https://arxiv.org/abs/2608.09947
作者: Atandrila Chowdhury,Sudip Vhaduri,Julius Keller,Debra Henneberry,Mark Wilson
类目: Human-Computer Interaction (cs.HC); Signal Processing (eess.SP)
备注: Under Review
Abstract:This study analyzes patterns of stress and exhaustion among student pilots throughout flight training using a combination of physiological and self-reported measurements. The Perceived Stress Scale (PSS-10) was used to measure perceived stress and exhaustion before and after each flight, while physiological data, including heart rate (HR), electrodermal activity (EDA), skin temperature, and acceleration, were continuously recorded during flight sessions. To identify recurring patterns in arousal and workload, physiological signals were preprocessed and analyzed across the flight stages. The findings indicate a buildup of workload-related weariness over time, as evidenced by steady increases in EDA and skin temperature across flights, as well as post-flight increases in self-reported exhaustion. Heart rate responses were more event-specific, with brief spikes during high-demand phases of flight. Overall, the findings demonstrate the value of combining physiological signals with subjective reports to identify patterns of stress and fatigue during real-world flight training and highlight the potential of data-driven approaches for monitoring pilot well-being.
[HC-27] HoosierHelp: Benchmarking LLM Agents for Social Service Navigation
链接: https://arxiv.org/abs/2608.09946
作者: Yiyang Li,Weixiang Sun,Tianyi Ma,Kaiwen Shi,Zheyuan Zhang,Yanfang Ye
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注:
Abstract:Social service navigation requires connecting help-seeking individuals to resources that satisfy their needs and specific constraints. Although LLM agents offer a promising interface for conversational resource navigation, existing benchmarks do not capture the interaction complexity and constraint-grounding demands of this setting. We introduce HoosierHelp, an interactive benchmark grounded in 3,971 Indiana public social service resources. Agents interact with simulated users, issue structured resource-search calls, handle non-ideal interactions, and select the final resources returned by the tool. HoosierHelp enhances the realism of simulated users by varying their need structure, constraint satisfiability, and behavior patterns, including impatience, rambling, unsupported requests, and self-contradiction. Experiments on 240 samples across seven LLMs show that current LLM agents remain substantially unreliable for social service navigation. Performance drops sharply on fallback-required and self-contradictory conversations, highlighting the need for agents that are more robust to complex and non-ideal user interactions.
[HC-28] Co-Lecturing With the DED: Explaining Circuit Design via the Draw Encode Display Loop
链接: https://arxiv.org/abs/2608.09945
作者: Alasdair Lambert,Guillaume Allais,Conor Mc Bride
类目: Human-Computer Interaction (cs.HC); Computers and Society (cs.CY)
备注:
Abstract:When representing digital circuits, 2 dimensional hand drawings free us from the linear structure of hardware description languages, enabling intuitive reasoning and making structure explicit. However these drawings are imprecise and inert: they do not enforce that the circuits are well defined and cannot be tested. We want both intuitive visual representations and well defined testable ones but students can struggle to link one to the other. To bridge this gap we present the Draw Encode Display Loop (DED), a Co-Lecturing dynamic which equips students with a systematic method to tackle natural language specifications: 1. Draw: visually informative intermediate representations (truth tables, characteristic tables) to generate a structured circuit diagram. 2. Encode: the diagram by labelling inputs, outputs and intermediate values which can be directly converted to code. 3. Display: the code using an in-house diagrammatic renderer. This is supported by Syrup, an education-focused hardware description language which allows students to define, experiment with and display their own circuits. We support these approaches with survey data gathered from two cohorts of students. Subjects: Human-Computer Interaction (cs.HC); Computers and Society (cs.CY) Cite as: arXiv:2608.09945 [cs.HC] (or arXiv:2608.09945v1 [cs.HC] for this version) https://doi.org/10.48550/arXiv.2608.09945 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Guillaume Allais [view email] [v1] Thu, 2 Jul 2026 14:16:42 UTC (2,871 KB) Full-text links: Access Paper: View a PDF of the paper titled Co-Lecturing With the DED: Explaining Circuit Design via the Draw Encode Display Loop, by Alasdair Lambert and 2 other authorsView PDFHTML (experimental)TeX Source view license Current browse context: cs.HC prev | next new | recent | 2026-08 Change to browse by: cs cs.CY References Citations NASA ADSGoogle Scholar Semantic Scholar export BibTeX citation Loading… BibTeX formatted citation loading… Data provided by: Bookmark checked="checked"class=“labs-tab-input”> Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv’s community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?) mathjaxToggle(); We gratefully acknowledge support from our major funders, member institutions, , and all contributors. About Help Contact Subscribe Copyright Privacy Accessibility Operational Status (opens in new tab) Major funding support from
[HC-29] Navigation Alone Is Not Enough: Evaluating Explanatory Assistive UI Agents
链接: https://arxiv.org/abs/2608.09944
作者: Santosh Patapati
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注:
Abstract:Modern web interfaces are increasingly difficult to use with screen readers, particularly when pages update dynamically or hide important structure behind visual layout. Recent UI agents can act on such interfaces; however, for assistive agents to be truly useful, they must behave as collaborators that keep users informed and in control, rather than as tools that simply take actions on users’ behalf. Most existing benchmarks judge systems primarily by task completion, without assessing how well they explain their actions or support user oversight. We introduce NeXUI, a benchmark for assistive agents that must navigate interfaces while explaining each step in clear language for nonvisual use. NeXUI pairs realistic user goals with instrumented interface states, enabling agents to reason from both visual context and structural information. Its evaluation measures safety, efficiency, and task success, while also checking whether explanations are grounded in the interface state. In our experiments, we find that NeXUI remains challenging even for state-of-the-art foundation models, with % Gemini-3.5-Flash achieving only a 44% success rate and poor explanation scores, making it a useful foundation for future research and development. By focusing on navigation, explanation, and user control, NeXUI provides a clearer way to study agents that can support blind and visually impaired users in modern computing environments.
[HC-30] EweAcT: Ewe behaviour aligned to accelerometer data for activity monitoring in extensive grazing systems
链接: https://arxiv.org/abs/2608.09943
作者: Lucile Riaboff(GenPhySE, INRAE),Ny Aina Andriamampandry(GenPhySE, GenPhySE),Jean-François Bompa(GenPhySE, GenPhySE),Mathias Aletru(GenPhySE, GenPhySE),Christian Durand(UEF),Sébastien Douls(UEF),Gaëtan Bonnafe(UEF),Morgane Costes-Thiré(GenPhySE, GenPhySE),Guillaume Delosières(GenPhySE, GenPhySE),Jean- Marc Mongrelet(GenPhySE, GenPhySE),Enzo Niro(GenPhySE, GenPhySE),Némuel Tadi(GenPhySE, GenPhySE),Séverine Deretz(DEPT GA, UEF, INRAE),Sara Parisot(UEF),Margot Lamarque(UEF),Dominique Hazard(GenPhySE),Emilie Cobo(GenPhySE)
类目: Human-Computer Interaction (cs.HC); Machine Learning (cs.LG)
备注:
Abstract:Monitoring livestock behaviour under extensive conditions would provide valuable insights to assess animal adaption to environmental perturbations in agroecological systems (e.g., heat waves, parasitism, predator attacks). Animal behaviour can be monitored using accelerometer data collected from neck-collars combined with artificial intelligence models. However, large amounts of accelerometer data aligned with annotated behaviours are necessary to develop accurate models of behaviour prediction. In particular, developing reliable models for extensive systems requires data collected across a wide range of representative conditions. The dataset includes 79 hours of tri-axial accelerometer data aligned with behaviours manually annotated from video recordings for 120 Romane ewes born between 2021 and 2024. The ewes were derived from two divergent genetic lines after three and four generations of selection started 10 years ago: low and high social attractiveness, noted S-and S+, and low and high tolerance towards humans, noted H-and H+. They were reared under the extensive system applied to the Experimental Unit of La Fage (UEF, INRAE, Saint-Jean-et-saint Paul, Aveyron) where 250 sheep were reared exclusively outdoors on 280 hectares of rangeland in southern France. First batch of data was collected on March, June and July 2024 at the UEF under a range of extensive conditions, including sloping pastures and heat-wave periods. Ewes were equipped with accelerometer neck-collars specifically designed for young sheep on pasture. They were grouped on experimental paddocks for 4 to 8 hours and provided with fresh grass and ad libitum access to water. The animals were simultaneously video-recorded using an elevated CCTV camera. Behaviour annotation was carried out using Behavioral Observation Research Interactive Software focusing on the main behaviours on pasture: Grazing, Ruminating, Resting, Moving, and ‘‘Other’’, grouping all remaining activities. Annotations and corresponding accelerometer sequences were aligned using Python language, based on a time synchronization procedure. A second batch of data was acquired on November 2025 to supplement the dataset with the moving activity. For that purpose, ewes were equipped with the accelerometer collars and moved on tracks from the housing area to the pastures, corresponding to an approximately 10 minute-walk. The start and end times of the moves for each ewe were used to align the corresponding accelerometer data with the moving activity. These data were then merged with the dataset from the first batch. The resulting dataset is ready to use for applying artificial intelligence models to classify the 5 main behaviours of sheep under extensive grazing systems from accelerometer data.
[HC-31] he impact of design factors of virtual and augmented reality on tertiary students user experience in Metaverse
链接: https://arxiv.org/abs/2608.09940
作者: Julius G. Garcia,Alvie Simonette Q. Alip
类目: Human-Computer Interaction (cs.HC); Computers and Society (cs.CY); Emerging Technologies (cs.ET)
备注:
Abstract:The Metaverse is a convergent space integrating virtual reality (VR) and augmented reality (AR) technologies, with market projections rising from \ 65.5 billion in 2022 to \ 1.3 trillion by 2030. Despite rapid adoption in education, the specific contributions of visual elements, environmental design, and communication features to user experience (UX) remain underexplored, limiting evidence-based design and resource allocation. This study examined how these design factors in VR and AR environments influence UX in Metaverse platforms. Using a correlational research design, data were collected from 321 purposively sampled tertiary students from engineering and computer science departments across four higher education institutions, all familiar with Metaverse platforms. A structured questionnaire with validated 5-point Likert scales measured UX; visual elements (field of view, resolution, color, complexity, and style); environmental design (interactivity, cohesiveness, naturalness, and complexity); and communication features (text/audio/video tools, avatar interaction, and spatial audio). Multiple regression analysis assessed the independent and combined effects of these factors on UX. Results show that environmental design (b=0.424, p0.001) and visual elements (b=0.298, p0.001) are the primary predictors of UX, jointly explaining 47.3% of its variance, with a strong correlation between them (r=0.792). Communication features did not significantly influence UX (b=0.053, p=0.196), challenging assumptions about virtual connectedness and indicating that communication design should better support social interaction. These findings highlight the importance of well-designed, interactive virtual spaces and suggest that developers and educational institutions prioritize environmental design to advance immersive learning and broader Metaverse applications.
[HC-32] How to Dogfood Your AI Chat Agent : A Three-Layer Evaluation Framework with Goal-Directed NPC Simulation
链接: https://arxiv.org/abs/2608.09939
作者: Alexandre Cristovão Maiorano
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注: 24 pages, 11 tables, 3 figures
Abstract:Production teams deploying LLM chat agents face a specific quality assurance gap: existing evaluation tools test individual responses or simulate social interactions, but none systematically verify whether real users can achieve their goals through multi-turn conversation. We introduce a three-layer dogfooding framework that bridges this gap by combining canonical question-bank testing (Layer 1), random-walk multi-turn evaluation (Layer 2), and a goal-directed NPC (Non-Player Character) simulator with five structured goal types and a ten-category failure taxonomy (Layer 3). In a longitudinal case study on a production multi-agent system over roughly three months (257 evaluation runs; a 108-scenario NPC suite), we find that the three layers produce complementary regression signals: cross-layer correlation for response quality is weak within a synchronized run (Spearman rho between -0.15 and 0.14) and negative across the longitudinal series (rho down to -0.46), confirming that canonical correctness does not predict goal-directed conversation success. The NPC simulator achieves 77 percent goal achievement at 0.17 dollars per run (6,272x cheaper than human evaluation), enabling daily CI/CD integration with automated PROMOTE/HOLD/ROLLBACK release decisions. We release full prompt templates, the failure taxonomy, and a Python-first replicability guide so that other teams can adopt the framework for their own LLM chat agents.
[HC-33] “YES! YES! I absolutely love this insight!” Affirmative Narration as Interactional Strategy in Dialogues with LLM Chatbots
链接: https://arxiv.org/abs/2607.28646
作者: Hanna-Riikka Roine,Anne Sigrid Refsum,Jill Walker Rettberg
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注: In review for a special issue of Narrative Inquiry
Abstract:This article analyses narrative mechanisms that are common in dialogues with LLM chatbots. In combination, these mechanisms produce an interactional strategy for maximising user engagement, which we call affirmative narration. Affirmative narration serves to convince users of the chatbot’s utility. We analyse three narrative mechanisms that support affirmative narration in human-LLM dialogues: firstly, guiding the user to view the chatbot as an intelligent and reliable character; secondly, activating masterplots, culturally significant and recurring story templates; and thirdly, using characters and masterplots not only to affirm, but also to isolate the user. The case studies range from a journalist’s unsettling chatbot experiment to cases where users have experienced delusions or even died by suicide after lengthy interactions with a chatbot. The analyses illustrate the worrying sides of affirmative narration, and the article thus concludes with a discussion of LLMs as a genre of fictional narrative media that requires a new type of literacy.
[HC-34] Data-driven Head Motion Generation through Natural Gaze-Head Coordination
链接: https://arxiv.org/abs/2605.25810
作者: Xiaohan Liu,Yilin Wen,Yusuke Sugano
类目: Computer Vision and Pattern Recognition (cs.CV); Human-Computer Interaction (cs.HC)
备注:
Abstract:We present the first data-driven approach to model temporal gaze-head coordination from large-scale in-the-wild facial videos. To obtain training data for generalizable learning, we propose an automatic pipeline that extracts natural yet diverse gaze and head motions with off-the-shelf appearance-based gaze estimators. To capture the probabilistic correlation and temporal dynamics of gaze-head coordination, we build our model on a generative conditional Variational Autoencoder for plausible yet diverse gaze-conditioned head motion generations. We further apply our framework to gaze-controlled facial video generation, where we enable video generation with natural and realistic head motion correlated to the input gaze - an aspect that has not been emphasized before. Human evaluation and quantitative comparisons demonstrate our method’s effectiveness and validate our design choices, with evaluators showing statistically significant preference for our approach over baseline methods.
[HC-35] Neural implants and human safety: single-fault detection for DC-coupled recording front ends
链接: https://arxiv.org/abs/2608.10361
作者: Dimitris Antoniadis,Timothy Constandinou
类目: ignal Processing (eess.SP); Human-Computer Interaction (cs.HC)
备注: Draft IEEE 4 + 1
Abstract:DC-coupled analogue front ends (AFEs) for neural implants provide a low-area solution. However, removing the coupling capacitor eliminates the intrinsic barrier that protects cortical tissue: a single-fault event, such as gate-oxide breakdown of a low-noise amplifier (LNA) input transistor, can open a direct DC path from the supply rail into the brain. On the stimulation side this hazard is well understood, and single-fault tolerance is enforced by a series DC-blocking capacitor; on the recording side, DC-coupled front ends discard the equivalent safeguard, yet their protection has gone almost unexamined. This paper presents a single-fault detection mechanism that monitors the LNA for the DC imbalance produced by such a failure and disables the amplifier before the resulting fault current can irreversibly damage tissue. The imbalance is encoded in the duty cycle of a current-starved relaxation oscillator and read out as a time-to-digital measurement. Designed in 65 nm, the mechanism resolves a worst-case fault of 6.4 nA across all corners within 0.81 ms - compliant with the ISO~14708-3 limit for an 8533 um2 electrode - opening a broader discussion of safety in DC-coupled recording.
计算机视觉
[CV-0] AdvFD: Boosting Visual Generation via Adversarial Frechet Distance Loss
链接: https://arxiv.org/abs/2608.11205
作者: Mingju Gao,Jingkai Zhou,Kun Gai,Changqian Yu,Hao Tang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project Page: this https URL
Abstract:Fréchet distance has recently emerged as an effective distribution-level objective for generator post-training, complementing the conventional sample-level diffusion and flow-matching losses. However, directly optimizing Fréchet objectives can cause Fréchet hacking. The target metrics keep improving, but visual quality and Fréchet alignment in other feature spaces may stagnate or deteriorate. We attribute this failure to the static pretrained feature spaces used by existing Fréchet losses. These feature spaces provide incomplete and fixed views of the differences between real and generated distributions. To address this limitation, we propose Adversarial Fréchet Distance (AdvFD), which complements the static representation targets in FD-Loss with a calibrated adversarially learned representation. AdvFD augments the original static Fréchet objective with a learnable representation that adversarially maximizes the Fréchet discrepancy between real and generated samples, while the generator minimizes the same discrepancy in the resulting adaptive feature space. To prevent the adversarial representation from trivially increasing the objective through feature amplification, we further introduce real-feature whitening, which normalizes its scale and covariance geometry and stabilizes the min–max optimization. Extensive experiments show that AdvFD consistently improves one-step generator post-training across both JiT and pMF backbones and across different model scales.
[CV-1] Surgical WAM: A World-Action Model for Data-Efficient Surgical Robot Learning
链接: https://arxiv.org/abs/2608.11204
作者: Wenrui Bao,Tianyun Jiang,Zhiben Chen,Ser-Nam Lim,Peter D. Peng,Yuzhang Shang
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Learning reliable surgical manipulation policies is bottlenecked by the scarcity of action-labeled demonstrations: teleoperated surgical robot (e.g., dVRK) trajectories with synchronized kinematics are costly to collect, while surgical tasks demand precise contact handling, long-horizon reasoning, and bimanual coordination. Endoscopic video is comparatively inexpensive and abundant relative to synchronized video–kinematics trajectories, and a natural way to exploit it is to learn world models of surgical scenes. However, existing surgical world models use video primarily for simulation or policy evaluation, and rarely translate the learned dynamics into closed-loop control. This gap raises our central question: under a fixed budget of action-labeled demonstrations, does action-free video pretraining improve closed-loop surgical manipulation? To answer it, we introduce the Surgical World-Action Model (Surgical WAM), a unified generative model built on Cosmos Policy that jointly predicts future endoscopic observations and executable surgical robot action chunks. Surgical WAM first learns surgical visual dynamics from action-free video and is then fine-tuned on the fixed action-labeled budget; at deployment, it acts as a closed-loop, receding-horizon controller that executes a short prefix of each predicted action chunk and replans from the resulting observation. On a suite of four simulated surgical manipulation tasks, video pretraining improves the average success rate from 63.5% to 77.8%, including an absolute gain of 20 percentage points on PegTransfer, with the largest improvements on contact-rich and bimanual tasks. These results demonstrate that action-free video provides transferable visual dynamics priors for learning surgical robot control with limited action supervision, positioning data-efficient video pretraining as a practical path toward scaling up surgical robot learning.
[CV-2] Capturing Uncertainty in Human Motion for Representation Learning in Soccer
链接: https://arxiv.org/abs/2608.11203
作者: Yizhou Xu,Lars Bretzner,Tiesheng Wang,Atsuto Maki
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:This paper presents a self-supervised representation learning framework for understanding 3D skeleton-based human motion in soccer, using future motion prediction as the learning objective. Since human motion is inherently uncertain, accounting for multiple plausible futures is essential for capturing the underlying motion dynamics and learning effective representations. To this end, we introduce a conditioning module for motion prediction that models a probabilistic distribution over discretized future motions in 3D Euclidean space, learning multimodality with explicit supervision from future trajectories. Experiments on large-scale soccer player tracking data show that our approach substantially improves motion prediction accuracy. Moreover, the learned representations effectively transfer to multiple soccer downstream applications, demonstrating strong cross-task generalization.
[CV-3] VidForensics-M1: Meta-Detection Reinforcement Learning with Verifiable Temporal Grounding for AI-Generated Video Forensics
链接: https://arxiv.org/abs/2608.11201
作者: Bowei Liu,Zheng Lu,Yuhan Bian,Xinchen Zhang,Xingming Shui,Yuesheng Huang,Xuhuan Li,Zihao Liu,Yifan Yang,Jun Zhou,Xiu Li
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 27 pages, 15 figures
Abstract:Recent advances in video generation models have significantly improved the realism of synthetic videos, blurring the boundary between generated and authentic content and raising concerns about misinformation. Existing MLLM-based detectors mainly rely on supervised fine-tuning or label-level reinforcement learning, where coarse supervision limits generalization to unseen scenarios and emerging video generators. To overcome these limitations, we are the first to introduce \textbfmeta-detection into AI-generated video detection, enabling reliable forgery detection by jointly optimizing predicted labels and supporting evidence within reinforcement learning. This paradigm requires reliable evidence signals and effective mechanisms to integrate them into label-level optimization. Textual rationales provide semantic descriptions of forgery artifacts, but their generation and verification depend on external models, making supervision vulnerable to hallucinations and semantic biases. In contrast, temporal grounding provides more objective and verifiable evidence, as manipulated intervals can be precisely controlled during forgery construction. Based on this insight, we propose an automated data construction pipeline that generates paired real-fake videos by replacing temporal segments with boundary-frame-conditioned video generation models. Furthermore, we introduce \textbfEvidence-Guided Reward Redistribution, which performs evidence-aware credit assignment by redistributing rewards among label-correct responses according to evidence quality. This preserves reliable label supervision while encouraging detectors to acquire fine-grained and verifiable forgery localization capabilities. Extensive experiments demonstrate that \textbfVidForensics-M1 effectively leverages verifiable temporal evidence to achieve robust and generalizable AI-generated video detection.
[CV-4] CausalSplat: Towards Comprehensive Hierarchical Reasoning in 3D Gaussian Splatting ECCV2026
链接: https://arxiv.org/abs/2608.11150
作者: Jiayu Ding,Meilu Song,Yun Chen,Wei Gao,Ge Li
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ECCV 2026
Abstract:While 3D Gaussian Splatting (3DGS) has advanced open vocabulary scene understanding, existing methods remain confined to explicit queries. They struggle to interpret implicit intents, complex spatial constraints, and commonsense reasoning required for practical embodied interactions. To address this gap, we introduce the task of reasoning 3D Gaussian segmentation and construct two benchmarks, Causal-LERF and Causal-ScanNet. These benchmarks systematically evaluate commonsense, spatial, affordance, and counterfactual reasoning. Evaluations reveal that current state of the art methods perform poorly on these reasoning challenges. Therefore, we propose CausalSplat, a framework that integrates vision-language models with 3D scene graphs to disentangle explicit structural perception from implicit logical inference. Extensive experiments demonstrate that CausalSplat achieves state of the art performance on our reasoning benchmarks while showing strong generalizability on standard referring and open vocabulary 3D segmentation tasks. Project Page: this https URL
[CV-5] PRMU: A Corpus-Free Benchmark for Person-Centric Knowledge Unlearning in Multimodal Large Language Models
链接: https://arxiv.org/abs/2608.11149
作者: Huafeng Chen,Yueming Lyu,Ziyuan Chen,Wenda Tan,Chenyang Si,Liucheng Guo,Caifeng Shan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Multimodal large language models (MLLMs) have demonstrated remarkable capabilities in storing and recalling rich person-related knowledge, raising increasing concerns about reliable knowledge removal. However, existing machine unlearning approaches for MLLMs typically assume access to original forget and retain corpora, which are often unavailable in realistic deletion scenarios. To address this limitation, we introduce PRMU, a benchmark for evaluating corpus-free multimodal unlearning under realistic person-centric deletion requests. PRMU focuses on naturally acquired person-related knowledge and evaluates whether models can remove target knowledge while preserving related knowledge through diverse textual and visual probes, including adversarial evaluation and fine-grained locality analysis. To facilitate research in this setting, we further introduce Similarity-Gated Projection Editing (SGPE), a lightweight corpus-free unlearning baseline with knowledge displacement, protected parameter-space editing, and locality-aware multimodal control. Extensive experiments on representative MLLMs reveal that existing unlearning methods often suffer from unfavorable forgetting-locality trade-offs, with significant locality degradation under aggressive forgetting settings, and remain vulnerable to multimodal knowledge reactivation. Meanwhile, SGPE provides a competitive trade-off between target forgetting, locality preservation, and general multimodal utility. We hope PRMU can facilitate future research toward realistic and scalable multimodal machine unlearning. Code and dataset will be released at this https URL.
[CV-6] SAR2Agri: Learning SAR Intensity Representations for Agricultural Monitoring
链接: https://arxiv.org/abs/2608.11142
作者: Moti Rattan Gupta,Anupam Sobti
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Agricultural monitoring faces unique challenges, arising from the landscape’s complex temporal, phenological, and climate dynamics, yet monitoring them is critical for ensuring food security. Synthetic Aperture Radar (SAR) satellites offer all-weather day-night imaging capability supporting key monitoring tasks including crop type mapping, yield prediction and phenological event detection. Existing multimodal remote sensing foundation models including TerraMind and CopernicusFM learn SAR representations by grounding them in optical imagery using joint encoding and contrastive learning techniques, while SAR-specific foundation models such as SAR-JEPA, SARMAE, and SAR-W-MixMAE primarily focus on target detection, flood mapping, and land cover classification applications. Recent work has introduced phenology inspired temporal pretext tasks with optical imagery which has shown strong performance on agricultural downstream tasks. In this work, we propose the first self-supervised learning pipeline focused on using only SAR intensity imagery for agricultural applications. We improve the temporal pretext tasks through masking and curriculum learning to enhance the pretraining pipeline’s ability to capture phenological features from SAR. On the SICKLE benchmark, our final model achieves 84.9% IoU on crop type mapping, outperforming optical baselines (by 15.3 pt) and existing SAR baselines (by 2.2 pt), demonstrating the effectiveness of our proposed pipeline for pretraining SAR intensity encoders for agricultural monitoring.
[CV-7] Is There Really a Camouflaged Object? Towards Realistic Camouflaged Object Detection
链接: https://arxiv.org/abs/2608.11135
作者: Huafeng Chen,Yueming Lyu,Chenyang Si,Wende Tan,Liucheng Guo,Caifeng Shan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Camouflaged object detection (COD) aims to segment objects that are visually concealed in their surroundings and has attracted increasing attention in recent years. However, most existing COD methods are developed under a closed-world assumption, where each input image is assumed to contain a camouflaged object. This assumption ignores realistic scenarios with pure backgrounds or non-camouflaged objects, causing existing models to produce severe false positives when deployed in open-world environments. To address this limitation, we propose OPC16K, a large-scale benchmark for realistic COD. OPC16K contains 16,245 images from 14 sources and is carefully organized into camouflaged-object images, pure background images, and non-camouflaged-object images, enabling comprehensive evaluation of both segmentation quality and negative-sample rejection. Based on this benchmark, we further propose OPCNet, a presence-aware camouflage network that reformulates COD from a pure segmentation task into a joint problem of object localization and camouflage existence reasoning. Specifically, OPCNet introduces hierarchical existence reasoning to distinguish CO, BG, and NOCOD scenarios, similarity-aware camouflage relation modeling to capture foreground-background camouflage cues, and existence-aware feature refinement to regulate segmentation features with existence predictions. Extensive experiments on OPC16K demonstrate that OPCNet achieves superior performance under the proposed realistic COD evaluation protocol, significantly reducing false positives on negative samples while maintaining accurate camouflaged-object segmentation. Code and dataset will be released at this https URL.
[CV-8] AlbumentationsX: One Augmentation Pipeline for Images and Related Annotations
链接: https://arxiv.org/abs/2608.11123
作者: Vladimir Iglovikov
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 8 pages, 1 figure. Source code: this https URL
Abstract:Augmentation can corrupt a training example when an image and its annotations receive different random changes. A crop must use the same coordinates for the image, mask, boxes, keypoints, stereo views, video frames, or volume. Code paths that choose these values separately can silently misalign the data. AlbumentationsX keeps the transform list, probabilities, annotation settings, and random seed in one Compose object. Each call chooses random values once and applies them to every supported part of the training example. The library keeps each object’s mask, box, and label together and lets projects add their own transforms. It can also save the pipeline definition, show what happened in one call, and run that call again. The examples place Compose after files have been decoded into arrays and before PyTorch groups examples into a batch. AlbumentationsX executes the declared transforms. Practitioners still decide whether a flip, crop, color change, or other operation preserves the correct label for their task. Comments: 8 pages, 1 figure. Source code: this https URL Subjects: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG) Cite as: arXiv:2608.11123 [cs.CV] (or arXiv:2608.11123v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2608.11123 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-9] Every Packet Counts: Dispersing Information for Loss-Resilient Learned Image Compression
链接: https://arxiv.org/abs/2608.11096
作者: Yuhang Wei(1),Chuqin Zhou(1),Yibo Shi(2),Jing Wang(2),Guo Lu(1) ((1) Shanghai Jiao Tong University, (2) Huawei Technologies Ltd.)
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 16 pages, 12 figures, 8 tables. Joint first authors: Yuhang Wei and Chuqin Zhou. Corresponding author: Guo Lu. To appear in Proceedings of the 34th ACM International Conference on Multimedia (MM '26), November 10-14, 2026, Rio de Janeiro, Brazil
Abstract:Learned image compression (LIC) has achieved impressive rate-distortion performance. However, existing methods remain highly vulnerable to packet loss, a common challenge in satellite and emergency communications. This vulnerability stems from non-uniform information distribution at the packetization stage and sequential decoding dependencies at the entropy coding stage. We propose an end-to-end loss-resilient image compression scheme that addresses both. Before packetization, we introduce an Inter-Channel Redistribution (ICR) mechanism to redistribute channel energy, preventing critical information concentrating in a small subset of channels. Then, an Interleaved Channel Grouping (ICG) strategy partitions latent channels in a strided manner to disperse information across packets, with each packet kept within constrained sizes. To limit cascading errors from lost packets, we adopt a two-layer dual-branch autoregressive structure to shorten the dependency chain. Extensive experiments demonstrate that our method consistently outperforms existing approaches in both reconstruction quality and stability. At 20% packet loss, it achieves an average PSNR gain of 1.84 dB over LossResilientLIC while reducing PSNR variance by an order of magnitude. Notably, trained under uniform random loss only, our model generalizes to bursty loss modeled by the Gilbert-Elliott channel, outperforming methods explicitly trained for such conditions.
[CV-10] Cross-View Feature Matching: Survey Benchmarking and Foundation-Model Perspectives ATC
链接: https://arxiv.org/abs/2608.11093
作者: Songlin Du,Xiaoyong Lu,Zeyu Wu,Xiaobo Lu,Guobao Xiao,Bin Fan,Jiayi Ma,Takeshi Ikenaga
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注: This manuscript goes beyond a conventional survey. It proposes a new taxonomy for cross-view feature matching, provides extensive benchmarking under unified datasets and protocols, and offers original analysis from the perspective of vision foundation models. These contributions provide substantive methodological synthesis, empirical findings, and new research insights
Abstract:Cross-view feature matching aims to establish reliable correspondences across images with large viewpoint variations. Over the past decade, the field has evolved from task-specific models toward increasingly unified and generalizable correspondence models, with recent progress further driven by the emergence of vision foundation models (VFMs). Despite these advances, existing studies remain highly diverse in their problem formulations, model architectures, training paradigms, and evaluation protocols, making it difficult to obtain a unified understanding of the field. In this survey, we present a unified review of cross-view feature matching. We first introduce a structured taxonomy covering feature extraction, single-type feature matcher, multi-type feature matcher, VFMs based methods, training strategy and robust estimation, providing a coherent framework for analysis and comparison. We further examine recent advances, distilling key design principles and highlighting the shift toward unified and generalizable correspondence models. We also provide a unified experimental benchmarking of representative state-of-the-art methods under consistent protocols, enabling fair and comprehensive performance comparisons. In addition, we discuss open challenges and future directions, including efficiency, robustness under extreme conditions, and cross-domain generalization. This survey aims to provide a comprehensive and structured reference for understanding the evolution, current landscape, and future development of cross-view feature matching in the era of vision foundation models.
[CV-11] Learning Gaussian Structure: Intervention-Guided Density Control for Feed-Forward Driving Reconstruction
链接: https://arxiv.org/abs/2608.11077
作者: Hang Li,Jiahe Li,Meiying Gu,Jin Zheng,Lina Yu,Xiao Bai
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Feed-forward Gaussian reconstruction has recently emerged as an efficient approach for driving scene reconstruction. However, prevailing LiDAR-based methods preserve the initial correspondence between observed points and Gaussian primitives, treating the initialized primitive set as the final representation. Unlike optimization-based 3DGS, these methods cannot accumulate gradients during training to determine how the scenes representation should be densified. Meanwhile, the shared sparse backbone only fuses observations from different timestamps implicitly, without explicitly aggregating cross-time evidence for individual primitives. In this paper, we present Learning Gaussian Structure (LGS), a framework that enhances both Gaussian structure and primitive attributes. Our key observation is that changes in local gradient responses induced by a prune or add intervention reveal whether the corresponding structural adjustment benefits reconstruction. Based on this observation, our Gaussian Densify Policy learns a Densify Map comprising Prune and Addition Scores from controlled interventions, and directly adjusts the Gaussian structure during inference. We further develop a compact Cross-Time Point Query that explicitly retrieves and aggregates neighboring features from Gaussian primitives at other timestamps for reliable attribute prediction. Extensive experiments on the Waymo Open Dataset and PandaSet demonstrate that LGS consistently outperforms existing methods.
[CV-12] Foundation Model-Enabled Efficient Data Sampling (FEEDS): A label-efficient training strategy for pan-cancer multi-tracer PET/CT datasets
链接: https://arxiv.org/abs/2608.11076
作者: Biratal Raj Wagle,Bashirul Azam Biswas,Grant Chau,Matthew E. Maeder,Muhammad Azeem Arshad,Michael S. Leapman,James B. Yu,Indrani Bhattacharya
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Code is publicly available on this https URL
Abstract:Automated lesion segmentation in whole-body PET/CT imaging can assist clinicians with cancer detection, staging, and treatment planning across radiotracers and cancer types. However, training lesion segmentation models that capture variations in lesion size, distribution, and appearance requires large annotated datasets, whose creation is both time- and expertise-intensive. As a result, models trained on limited labeled PET/CT data often lack the accuracy and generalizability needed for clinical use. We present FEEDS (Foundation model-Enabled Efficient Data Sampling), a label- and compute-efficient learning strategy that uses vision foundation model embeddings to select the most informative and diverse unlabeled cases for expert annotation. Unlike unsupervised, semi-supervised, and active learning approaches, FEEDS is a one-step training paradigm requiring only a limited, representative training set, making it label- and compute-efficient. We train and validate FEEDS using the AutoPET-III dataset. We test its accuracy and generalizability on three held-out sets: AutoPET-III, DeepPSMA, and an internal Dartmouth-Hitchcock Medical Center dataset. We evaluate clinical utility at the voxel, lesion, and anatomic region level to assess performance in high-risk areas and treatment planning utility. FEEDS outperforms random-sampling-based labeling, pseudolabel-based semi-supervised learning, and training with limited labeled data alone. It generalizes across all three test sets, FDG and PSMA tracers, and multiple diseases, matching fully-labeled (100%) training performance with 70% less annotation burden. FEEDS addresses the challenge of label scarcity in an automatic lesion segmentation framework by providing a practical approach for constructing representative and diverse annotation queues from large, unannotated clinical repositories.
[CV-13] Static in Frames Dynamic in Events: Rethinking Features in Event Cameras as Motion Cues
链接: https://arxiv.org/abs/2608.11075
作者: Hesam Araghi,Jan van Gemert,Nergis Tomen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Event cameras capture intensity changes asynchronously with high temporal resolution, requiring novel preprocessing methods for downstream tasks. Unlike static intensity snapshots, event data inherently encode information about scene dynamics and object motion, meaning that features derived from events can exhibit behaviors with no direct analogue in frame-based vision. In this paper, we analyze two features used in event-based corner detection—the eigenvalues of the structure tensor and the spatiotemporal density values—and show that they are \emphmotion cues. We hypothesize that these features, combined with local geometric information, can enhance motion estimation tasks. To validate this, we first theoretically analyze how the eigenvalues of the structure tensor at moving corner points relate to the direction of motion. We then design controlled experiments on a synthetic dataset, confirming that extending local geometric features with eigenvalues and density values provides complementary motion information and is robust to texture and shot noise. Finally, we integrate the proposed features into a state-of-the-art event-based optical flow network and evaluate on the real-world DSEC benchmark, where the added features consistently improve accuracy, with the largest gains in data-scarce scenarios and for lower-capacity models. The code for this paper can be found at: \hrefthis https URLthis https URL.
[CV-14] CapProbe: Evaluating Detailed Image Captions via Full-Scene Dense Question Answering
链接: https://arxiv.org/abs/2608.11074
作者: Mouxiao Huang,Qiangyu Yan,Borui Jiang,Han Shu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Evaluating detailed image captions from Vision-Language Models (VLMs) requires going beyond surface-level semantic similarity. Reference-based metrics (e.g., CIDEr and SPICE) and LLM-as-scorer protocols struggle to verify dense factual claims, while existing QA-based alternatives generally offer lower probe density, narrower domain coverage, or no explicit alignment between individual questions and segmented image regions. We introduce CapProbe, a full-scene dense QA benchmark that turns detailed caption evaluation into region-aligned factual checking. Each image is decomposed into coarse semantic regions covering both foreground and background elements; for every retained region, we generate multiple-choice questions spanning 10 semantic categories, forming a dense checklist of probed visual facts. Guided by a two-tier taxonomy of 37 L1 domains and 219 L2 sub-domains, CapProbe comprises 346 images, 1,868 regions, and 25,650 questions, averaging 74 QA pairs per image. A language judge answers from the caption alone; an Uncertain option and Effective Accuracy provide a judge-dependent proxy for distinguishing unanswered probes from incorrectly resolved ones, while density-based metrics penalize verbose yet uninformative captions. The protocol is cost-effective: by converting unconstrained scalar scoring into structured MCQ reading, it reduces open-ended scoring bias while remaining judge-conditioned and yields relatively stable model rankings under a fixed reader. Experiments on 13 VLMs show large Coverage gaps across models, a clear competency-efficiency trade-off, and failure modes that sparse or overlap-based evaluation often misses. The benchmark data, annotations, and evaluation code will be released soon.
[CV-15] Entropy-Centric Explainable AI for Remote Sensing Image Segmentation
链接: https://arxiv.org/abs/2608.11064
作者: Ali Saleh,Abdul Karim Gizzini,Mohamad Ghassany,Ali J. Ghandour
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Artificial intelligence (AI) has become a powerful approach to solving complex problems in critical domains. Many concerns arise regarding the decision-making process of its models, mainly due to deep neural networks outperforming their peers at the cost of ambiguity in feature extraction and prediction. Consequently, in critical domains such as remote sensing, where high-resolution imagery must be analyzed using black-box models, the lack of transparency limits trust in these models and, thus, their adoption. In light of this reality, explaining and understanding the complex decision-making process of AI models has become essential. Explainable AI (XAI) aims to bridge this gap by providing insights into how and why certain decisions are made. While significant progress has been achieved in explaining image classification tasks, image segmentation still offers considerable room for improvement. In this context, this paper proposes an entropy-centric XAI method for semantic segmentation. Moreover, a new XAI evaluation methodology is proposed to efficiently measure the relevance of the regions highlighted by the proposed XAI method. Experimental results demonstrate the superiority of the proposed XAI method compared with recently adapted XAI methods for semantic segmentation.
[CV-16] A Comparative Evaluation of Deep Learning Object Detection Models on a Real-World Multi-Plant Dataset from Africa
链接: https://arxiv.org/abs/2608.11053
作者: Ismail Ismail Tijjani,Sunusi Muhammad Ibrahim,Amina Ibrahim Khaleel,Lanre Olusegun Akinola,Fatima Isa Jibrin,Muhammad Bashir Aliyu,Abdullahi Abdussalam Dalhat,Abdullahi Suiudeen
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:The application of computer vision in agriculture has shown significant potential for improving crop monitoring and precision farming. However, many existing approaches rely on controlled datasets that do not adequately represent realworld farming conditions, particularly in underrepresented regions such as Africa. This study presents a comparative evaluation of six object detection models YOLOv5, YOLOv8, YOLO11, YOLO26, Faster R-CNN, and RT-DETR using a real-world dataset, AgriAISeg 1 , collected manually from Nigerian farms. AgriAISeg comprises 3,382 images of sesame, cabbage, and tomato crops captured under varying environmental conditions, including changes in illumination, occlusion, and viewing perspectives. Models were trained, and performance was assessed using precision, recall, mAP@0.5, and mAP@0.5:0.95. The results show that RT-DETR achieved the highest overall performance with a precision of 0.768 and mAP@0.5:0.95 of 0.624, while YOLOv8 and YOLO11 also demonstrated strong and consistent performance. In contrast, Faster R-CNN recorded significantly lower accuracy, with an overall mAP@0.5 of 0.466, indicating reduced effectiveness under complex field conditions. In addition, YOLO-based models exhibited superior training efficiency compared to Faster this http URL findings demonstrate that modern one-stage and transformer-based detectors provide more reliable and efficient solutions for plant detection in realworld agricultural environments.
[CV-17] HUI360: A 360° Egocentric Dataset and Baselines for Human-Robot Interaction Anticipation
链接: https://arxiv.org/abs/2608.11051
作者: Raphael Lorenzo-Louis,Fabio Amadio,Bertrand Luvison,Serena Ivaldi
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:As robots increasingly operate in human-populated environments, anticipating human intentions is essential for enabling proactive and socially aware behavior. Automatic anticipation of human-robot interactions is thus emerging as a crucial perception challenge for embodied agents. To this end, we introduce HUI360, the largest dataset for human-robot interaction anticipation in the wild and its set of baselines. The dataset was collected from a mobile robot, in the wild, over multiple days within a 3-month period, and in several environments, capturing natural, spontaneous behaviors from both passersby and users, and encompassing a diverse range of individuals. This variety enables evaluating and improving the generalization capabilities of interaction anticipation models. We designed a pipeline and share code for automatic interaction annotation in arbitrary 360-degree equirectangular videos, along with interfaces for manual refinement. Using this pipeline, we release the HUI360 open set of 1M pre-processed annotations, including detailed 2D poses, facial keypoints, and segmentation masks, obtained using state-of-the-art computer vision methods and manually curated to ensure high-quality tracking and interaction annotation. Additionally, we release the raw panoptic 360-degree images captured from the robot’s egocentric viewpoint (on demand, for research purpose only in compliance with GDPR). Finally, we establish benchmark baselines for interaction anticipation, including the first cross-dataset evaluations for this task: to this end, we also release 6M annotations for another existing in-the-wild outdoor dataset collected from a mobile robot (SSUP-HRI). Dataset and code can be found at this https URL.
[CV-18] 3D Weighted Geometric Graph Neural Networks for Sheep Facial Pain Assessment
链接: https://arxiv.org/abs/2608.11050
作者: Alam Noor,Luis Almeida,Mohamed Daoudi
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Deep learning systems perform mainly within the 2D for a single image domain and take the face as a single-dimension representation, losing sight of the 3D anatomy of sheep and cross-landmark spatial relationships that are intrinsic to the clinically proven Sheep Pain Facial Expression Scale (SPFES). This paper presents the \textbf3D Sheep Pain Facial Expression System (3D-SPFES), a novel, monocular depth-aware geometric graph neural network system that integrates each SPFES facial landmark, such as the ears, eyes, and nose, into 3D Euclidean space estimated from a single RGB camera by using VideoDepthAnything, thus preventing the need for specialized depth hardware. Each landmark node includes a feature vector containing its 3D spatial coordinates, estimated surface normal, and facial attribute class embedding. Edges linked to nodes are assigned weights based on an aggregate metric that combines both Euclidean distance and surface co-planarity in a 3D space. A Weighted Geometric Graph Neural Network (WG-GNN) studies this graph using \mathcalK = 3 geometry-aware message-passing layers enhanced by a scaled dot-product attention method that selectively enhances anatomically relevant inter-landmark messages. The resultant node embeddings are combined into \mathcalO = 3 pain-level clusters and integrated into a Normalized Pain Score (NPS) within the range of [0, 100%] a confidence-weighted, SPFES-derived scoring method.
[CV-19] When Visual Signals Mislead: A Mechanistic Study of Attribute Hallucination in Vision-Language Models
链接: https://arxiv.org/abs/2608.11024
作者: Yufei Zhang,Chenlu Zhan,Hongwei Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Attribute hallucination—where vision-language models (VLMs) correctly identify an object but mischaracterize its properties—is prevalent yet mechanistically poorly understood. The dominant explanation, language-prior dominance, has motivated prior-suppression methods, but this explanation has not been directly tested at the attribute level. We present VISOR (Visual-Operational Remediation), a unified framework that couples null-image-based diagnosis with routed remediation. Its VSNR diagnostic decomposes each prediction into a visual logit signal and a language-prior signal. Across 10,791 negative-ground-truth samples from three VLM families and three attribute types, the visual signal strongly predicts false positives, whereas the language-prior signal is near chance. VISOR uses this diagnosis to separate two failure modes: low-margin but directionally correct visual signals in color/state attributes, and low-SNR or misaligned visual signals in material attributes. The same diagnosis routes each query to the appropriate operator: calibration for threshold-placement errors, abstention for training-free low-SNR handling, or targeted visual adaptation for material failures that prior suppression cannot correct. Across Qwen, InternVL, and LLaVA, VISOR reduces attribute false positives without relying on the prior-dominance assumption.
[CV-20] Watching Synthetic Videos: Aligning Cross-modal Representations with Visual Synthesis for Zero-shot Video Captioning
链接: https://arxiv.org/abs/2608.11013
作者: Liangyu Fu,Junbo Wang,Yuke Li,Ya Jing,Xuecheng Wu,Zhiyong Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Text-only training is a popular paradigm in zero-shot video captioning, where the video distribution is not available to the model during training, leading to a cross-modal gap between the training (text-only) and the inference (video-only). Previous works attempt to bridge the gap through simple linear transformations. However, the inherent gap between text and video makes cross-modal representation space alignment insufficient, resulting in inaccurate sentences. To address this issue, we propose a novel zero-shot video captioning framework (WSV) consisting of two training stages, which first generates corresponding synthetic video latent representations via a pretrained text-to-video generation model. To strengthen the fidelity of the latent representations, we propose a polisher capable of bridging the gap between real and synthetic video distributions. Subsequently, we design a prompter that conditions GPT-2 on the polished latent representations to generate the captions in the second training stage. During inference, an input video is encoded by a pretrained 3D Causal VAE and then fed directly into the prompter, which in turn guides GPT-2 to produce the final caption. Experimental results conducted on MSVD, MSR-VTT, and VATEX datasets demonstrate that our proposed method achieves scores of 52 and 95.7 on the B@4 and CIDEr metrics, respectively.
[CV-21] HNDiff: Haze-Noise Diffusion for Image Dehazing ECCV2026
链接: https://arxiv.org/abs/2608.10995
作者: Jin-Ting He,Fu-Jen Tsai,Yan-Tsung Peng,Min-Hung Chen,Chia-Wen Lin,Yen-Yu Lin
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ECCV 2026. Project Page: this https URL
Abstract:Existing diffusion-based methods have recently made significant progress in image dehazing. However, they typically neglect the physics of haze formation and reconstruct clean images from pure Gaussian noise, thereby limiting their restoration potential. To address this issue, we propose Haze-Noise Diffusion (HNDiff), a novel diffusion framework that embeds the atmospheric scattering model as an inductive bias. By grounding diffusion in physical principles, HNDiff ensures that the restoration aligns more closely with underlying mechanisms of haze formation. In its forward process, we introduce joint haze-noise diffusion with a haze-aware noise scheduler, which progressively adds both haze and noise to an image. Essentially, the scheduler adapts noise levels according to haze density, meaning that regions with heavier haze receive stronger noise injection to encourage content generation, while clearer regions receive lighter noise to better preserve details, which directly links the forward degradation process with the physics of haze. In the reverse process, we then derive a physically consistent dehazing-denoising process that simultaneously removes haze and noise to restore a clean image in a manner aligned with the forward degradation process. To further enhance practicality, we propose Latent HNDiff, which compiles clean latent priors that can be seamlessly integrated into existing dehazing networks to boost performance. Extensive experiments show that our work significantly improves leading dehazing backbones and achieves state-of-the-art results on benchmark datasets. The project page is available at this https URL .
[CV-22] Putting Registers to Work: Task Registers for Token Pruning in Vision Transformers
链接: https://arxiv.org/abs/2608.10989
作者: Hongsen Cao,Mona Jaber,Shanxin Yuan,Ahmed Sayed
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 24 pages, 9 figures. Includes supplementary material
Abstract:Token-pruning policies are usually designed for a single recognition pipeline, but pretrained Vision Transformers are reused across tasks with different spatial demands. We ask which parts of a pruning policy transfer across image classification, semantic segmentation, and object detection. For each pipeline, controlled probes freeze the no-pruning checkpoint and apply a series of parameter-free reduction criteria at one eligible layer at a time without retraining. The probes reveal three differences: segmentation and detection rank the criteria differently, classification is especially sensitive to attention-based pruning in the earliest layers, and the dense tasks prefer opposite recovery endpoints. These findings motivate Task-Adaptive Pruning (TAP). Existing register tokens serve as task-agnostic storage for feature artifacts. TAP instead introduces one task register per task and activates only the current one. Its evolving state ranks tokens, distributes an exact removal budget over depth, and sets the recovery scale for dense features. At a final keep rate of \rho=0.5 , our jointly adapted model, TAP-J, reaches 47.0 mIoU at 1.30\times encoder throughput on ADE20K and 53.7 box AP at 1.32\times encoder throughput on COCO while remaining competitive on ImageNet-1K.
[CV-23] PEAK: Precise and Persistent Concept Erasure via k-Sparse Autoencoders
链接: https://arxiv.org/abs/2608.10985
作者: Man Jiang,Ouxiang Li,Weibao Xue,Zhenhua Tang,Yuan Wang,Shuo Wang,Yanbin Hao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Erasing concepts from large-scale text-to-image (T2I) diffusion models has become increasingly crucial due to the growing concerns over copyright infringement, privacy violations, and offensive content. Existing approaches struggle to achieve both precise and persistent concept erasure: inaccurate localization of concept-related representations may cause unintended semantic interference, while incomplete removal of the underlying concept knowledge allows adversarial recovery. To address this dilemma, we propose PEAK, a \textbf\textitprecise and \textbf\textitpersistent concept erasure framework via k-Sparse Autoencoders (kSAEs). PEAK first trains a kSAE on internal activations of the diffusion denoising network to decompose dense representations into interpretable sparse features. By contrasting sparse activations induced by target and non-target prompts, PEAK identifies a compact set of target-specific features according to both activation strength and frequency. These localized features are then used for parameter optimization, where PEAK selectively suppresses target-related activations while preserving complementary non-target ones towards the original model. This feature-guided optimization embeds concept erasure directly into diffusion parameters, eliminating the need for additional inference-time intervention and facilitating effective persistence against adversarial attacks. Extensive experiments demonstrate that PEAK achieves effective and robust concept erasure. On the I2P benchmark, PEAK reduces NudeNet detections from 582 to 6, lowers the average attack success rate (ASR) from 96.52% to 5.63%, and preserves general generation quality on MS-COCO with a near-zero KID. Our code and models are available at: this https URL
[CV-24] hinkAfford: Affordance-Centric Reasoning for Fine-Grained 3D Grounding in Cluttered Scenes
链接: https://arxiv.org/abs/2608.10981
作者: Xinrui Lin,Sha Zhang,Shumin Wang,Zenghuan Zhu,Jiajun Deng,Yanyong Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 8 pages, 6 figures
Abstract:Task-driven 3D affordance grounding aims to localize the functional region in a cluttered 3D scene that enables an action specified by a natural-language instruction. Existing methods either predict 3D masks directly or construct them by selecting and fusing intermediate 2D/3D regions. However, they remain vulnerable to two intertwined failure modes: the predicted or selected regions may miss the target interaction area or have unsuitable granularity, while language grounding may confuse visually similar alternatives under relational instructions. To this end, we introduce ThinkAfford, which decouples high-recall affordance proposal generation from instruction-grounded reasoning. Specifically, the Affordance Proposal Generation module first uses learnable affordance prompts and multi-level visual features to predict interaction-conditioned heatmaps, extracting a variable number of fine-grained proposals without parsed object or part names as segmentation prompts. Visual-Prompted Affordance Reasoning then reasons over labeled proposal overlays using the full instruction, returning identifiers in a structured “think-then-answer” response. Moreover, Group Relative Policy Optimization uses proposal-level rewards from lifted 3D overlap to align VPAR selection with final 3D grounding. On the SceneFun3D validation split, ThinkAfford achieves 10.69% AP50 and 25.46% AP25 under the official evaluator, outperforming comparable 3D open-vocabulary and vision-language-model-based 2D-to-3D baselines. Module-level diagnostics further show that APG attains 77.5% recall at 25% intersection-over-union, while GRPO-trained VPAR achieves 72.1% selection accuracy on APG-covered queries, compared with 63.4% under supervised fine-tuning.
[CV-25] A Dataset and Benchmark for Optical Music Recognition of String Quartet Scores
链接: https://arxiv.org/abs/2608.10978
作者: Dongmin Kim,Brian Liu,Jose J. Valero-Mas,Dasaem Jeong
类目: Computer Vision and Pattern Recognition (cs.CV); Sound (cs.SD)
备注: 8 pages, 2 figures, 5 tables. Accepted at the ISMIR 2026
Abstract:Optical music recognition (OMR) transcribes music scores into digital formats. While the field has advanced significantly on monophonic and piano-form scores, multi-part score transcription remains underexplored, largely due to the absence of a suitable dataset. We introduce OpenScore String Quartet for Optical Music Recognition (OSSQ-OMR), the first dataset dedicated to multi-part OMR. Built on the OpenScore String Quartet corpus, OSSQ-OMR pairs digitally encoded scores with their original scanned editions from IMSLP, with all images visually aligned to their transcriptions. The dataset is released with score images at system and staff levels, and paired transcriptions in three encoding formats: Extended Linearized MusicXML (LMXE), **kern, and ABC. In total, OSSQ-OMR contains 24,544 system images and 98,172 staff images drawn from 116 string quartet scores. We accompany the dataset with a benchmark protocol and baseline results from two representative OMR models, evaluated across four random score-level splits with mutually exclusive test sets. Baselines reach OMR-NED as low as 3.6% on synthetic and 5.9% on scanned inputs; results reveal substantial effects of encoding and segmentation choices, with the LSTM-based baseline degrading on scanned inputs roughly 2.6 times less than the Transformer-based baseline.
[CV-26] CARE: Confidence-Aware Reasoning for Reliable Medical VQA MICCAI2026
链接: https://arxiv.org/abs/2608.10964
作者: Yuetian Du,Yucheng Wang,Zhenyuan Chen,Luyuan Chen,Rongyu Zhang,Jinjian Zhang,Wei Zhou,Zhijie Xu,Ming Kong,Zhan Zhou,Jie Liu,Qiang Zhu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted by MICCAI 2026
Abstract:Reinforcement Fine-Tuning (RFT) has enabled medical Multimodal Large Language Models (MLLMs) to produce Chain-of-Thought (CoT) reasoning for visual question answering, yet these models suffer from \textitconfidence miscalibration —a systematic gap between expressed certainty and actual diagnostic accuracy that undermines clinical trust. We propose \textbfCARE , a \textbfC onfidence- \textbfA ware medical \textbfRE asoning framework that jointly optimizes accuracy and calibration through a dual-stage pipeline. First, a scalable Medical-CoT synthesis provides structured cold-start data for Supervised Fine-Tuning. Second, Group Relative Policy Optimization (GRPO) with a novel \textbfConfidence-Aware Reward (CAR) mechanism ties the model’s confidence to diagnostic correctness within the reward signal. Across three Medical VQA benchmarks, \textbfCARE achieves the highest diagnostic accuracy while obtaining the lowest Expected Calibration Error and Hallucination Rate, establishing a foundation for trustworthy clinical decision support. Our code is available at this https URL.
[CV-27] Once Poisoned Arbitrarily Controlled: A Programmable Backdoor in VLMs
链接: https://arxiv.org/abs/2608.10959
作者: Tao Lin,Gaojie Jin,Zongxin Liu,Peng Wu,Lijia Yu
类目: Computer Vision and Pattern Recognition (cs.CV); Cryptography and Security (cs.CR)
备注:
Abstract:Existing vision-language model (VLM) backdoors are usually treated as static vulnerabilities: one-to-one and N-to-N attacks bind one or more triggers to a finite set of targets before victim training. This assumption substantially underestimates the threat. We show that a single poisoning phase can implant a programmable backdoor into a VLM, allowing an attacker to choose previously unseen target-caption semantics at inference time and synthesize corresponding stealthy triggers on demand. Unlike fixed-mapping attacks, the proposed any-to-any caption-control paradigm decouples post-training target selection from poisoning, enabling dynamic control of target captions without retraining the VLM. Our method has two components. First, a heuristic poisoning strategy exposes the model to diverse trigger-caption pairs, encouraging it to learn a general trigger-as-instruction rule rather than memorize a specific backdoor pattern. Second, a feature-space trigger steganography method maps any attacker-specified target caption to a stealthy visual trigger, implemented as either a norm-controlled perturbation or a non-semantic patch. Once inserted into arbitrary images, these triggers cause the poisoned VLM to generate outputs semantically aligned with the chosen target caption, even when the target was unseen during poisoning. Extensive experiments show that our attack achieves high any-to-any caption-control success rates, preserves clean model utility, and remains effective under several classical backdoor defenses.
[CV-28] Evidence-Grounded Trustworthy Multimodal Reasoning and Evaluation Benchmark in Complex Urban Scenes
链接: https://arxiv.org/abs/2608.10954
作者: Zhaoyang Wei,Bowen Jiang,Xumeng Han,Jiashu Li,Xuehui Yu,Yuling Liu,Guorong Li,Zhenjun Han,Jianbin Jiao
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted by IJCV
Abstract:While Multimodal Large Language Models (MLLMs) demonstrate impressive performance in benign scenarios, their cognitive reliability deteriorates significantly in complex scenes under adverse conditions. In these settings, models often rely on implicit inference without sufficient visual evidence, leading to a disconnect between perception and reasoning. Meanwhile, existing outcome-oriented benchmarks evaluate only final predictions and fail to diagnose failures in the underlying reasoning process. To address this gap, the authors propose AD2-Bench, which introduces a Hierarchical Visual Diagnosis framework that decomposes reasoning into a structured Chain of Evidence (CoE). This fine-grained diagnosis reveals that robust multimodal reasoning fundamentally depends on accurate evidence acquisition. Building on this perspective, the authors formulate reasoning from a probabilistic viewpoint and identify two primary causes of reasoning failure: Spatial Ambiguity, where models fail to distinguish target objects from background clutter, resulting in localization errors; and Semantic Uncertainty, where degraded visual features lead to incorrect semantic interpretation, resulting in understanding errors. To overcome these evidence deficiencies, they further propose Evidence-grounded Visual Reasoning (EGVOR), which replaces implicit reasoning with the explicit generation of Evidence Atoms - structured spatial-semantic triplets that enforce tight alignment between localization and semantic understanding. The model is trained through a hierarchical curriculum that progresses from reflective supervision construction to reinforcement learning, where reducing reasoning variance is explicitly rewarded. Extensive experiments demonstrate that EGVOR substantially improves reasoning stability under adverse conditions, providing a more robust framework for trustworthy multimodal cognition.
[CV-29] Multiple Scale Latents for Learned Image Compression ICIP2026
链接: https://arxiv.org/abs/2608.10952
作者: Jonas Brenig,Radu Timofte
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at ICIP 2026
Abstract:Most learned image compression systems rely on a single latent representation combined with a hyperprior, which limits their ability to efficiently capture image structure across spatial scales. In this work, we propose a hierarchical latent representation to improve the efficiency of the entropy model. By using multiple latents at different scales, each with its own entropy model, we better capture the spatial structure of the latent representation. Our experiments show that this approach achieves a 17.9% BD-rate reduction over VVC on Kodak, demonstrating the effectiveness of multi-scale latent representations. Furthermore, the approach is orthogonal to other advances in learned image compression, making it a versatile addition to existing methods.
[CV-30] Mixture-of-Experts-based Entropy Model for Learned Image Compression ICIP2026
链接: https://arxiv.org/abs/2608.10947
作者: Jonas Brenig,Radu Timofte
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at ICIP 2026
Abstract:Learned image compression has seen significant progress in recent years with the development of end-to-end learned models that achieve better compression efficiency than state-of-the-art conventional methods. Recently, Mixture of Experts (MoE) approaches have seen promising results in NLP and computer vision tasks. In this paper, we introduce the MoE approach to learned image compression. We propose a MoE-based Entropy model (MoEE) for learned image compression, allowing the model to selectively activate only the subset of parameters required for the input image. Our model achieves a BD-Rate improvement over VVC of -16.85% on the Kodak dataset.
[CV-31] GS-CPE: Unified 6-Degree-of-Freedom Camera Pose Estimation via 3D Gaussian Splatting IROS2026
链接: https://arxiv.org/abs/2608.10938
作者: Huaiyuan Weng,Chul Min Yeum,Su-Min Kang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 8 pages, accepted at IROS 2026
Abstract:Despite substantial progress in visual localization, from scene coordinate regression to direct camera pose regression, achieving both robust generalization and high accuracy remain challenging. This study introduces GS-CPE (Gaussian Splatting based Camera Pose Estimation), a coarse-to-fine framework for 6-DoF camera pose estimation that unifies geometry-based coarse pose estimation with robust 3D Gaussian Splatting (3DGS) warping based pose refinement. GS-CPE first estimates a coarse pose via retrieval-guided geometric pose estimation on a 3DGS scene representation, then refines it by minimizing a visibility aware masked RGB warping objective in a multi-scale optimization framework, with adaptive re-rendering. Extensive experiments on indoor and outdoor benchmarks including 7Scenes, Cambridge Landmarks, FAST-LIVO2 datasets, and a custom dataset demonstrate state-of-the-art performance, consistently outperforming in both accuracy and generalization.
[CV-32] SafeCA: Safe Cross-Attention Localization and Regulation for Text-to-Video Jailbreak Defense
链接: https://arxiv.org/abs/2608.10933
作者: Siyuan Liang,Yupeng Qiu,Junfeng Fang,Rong-Cheng Tu,Jiaxing Huang,Dacheng Tao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 10 pages, 4 figures
Abstract:Text-to-Video (T2V) generative models are vulnerable to jailbreak attacks in real-world deployment, leading them to produce harmful or inappropriate content. Existing defense approaches mainly rely on input filtering or reconstruction, which not only incur high computational latency but also tend to distort semantics. To address these issues, we experimentally and systematically analyze the differences between clean and jailbreak samples in the cross-attention feature space, revealing for the first time a cumulative separation effect and a progressively increasing trend of linear separability between the two during the diffusion process. Based on this insight, we propose SafeCA, a feature-level defense mechanism for safe cross-attention localization and regularization. Firstly, we identify key defensive regions and values through attention stability analysis using cross-attention features collected from clean prompts within a single inference. Secondly, SafeCA mitigates anomalous activations via attention masking with energy normalization and introduces a lightweight semantic-space adapter to redirect abnormal semantic flows. Furthermore, we detect and suppress potentially malicious tokens by back-propagating feature anomaly signals to the input cue words, thereby enhancing the deployability of the defense in commercial models. Experimental results show that SafeCA reduces the jailbreak success rate by about 20% on mainstream T2V models, adds almost no inference overhead (+0.1s), and maintains good text-video semantic consistency. Overall, SafeCA provides an architecture-level, deployable protection paradigm for T2V generation models.
[CV-33] mporally Grounded Compositional Camera Motion Understanding via Geometric Knowledge Distillation
链接: https://arxiv.org/abs/2608.10932
作者: Dazhao Du,Shiyan Du,Jian Liu,Yongjian Yu,Bohai Gu,Tao Han,Hualuo Liu,Eric Liu,Yujia Zhang,Xi Chen,Song Guo
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Understanding camera motion is fundamental to video perception, with applications in spatial intelligence and controllable video generation. Multimodal large language models (MLLMs) provide a natural interface for this task, but existing work typically assigns one or more labels to an entire clip. Such clip-level recognition overlooks two defining properties of real camera motion: it can change within a shot, and multiple movements can occur simultaneously. We therefore formulate camera-motion understanding as temporally grounded, compositional recognition, which requires a model to localize motion-consistent intervals and identify every movement active within each interval. We introduce CamChoreo, a benchmark of 4,229 real single-shot clips with expert-annotated temporal segments. Its annotations use a compact vocabulary of 20 direction-aware labels, and nearly half of the segments contain compound camera motion, with multiple movement primitives active simultaneously. Recognizing such fine-grained, compositional motion is hard for current MLLMs, whose visual encoders emphasize semantic content rather than the geometric evidence on which camera motion depends. Directly injecting features from a frozen 3D foundation model addresses this gap, but requires running the expensive geometry model on every input; we refer to this baseline as CamInject. We instead propose CamDistill, which distills the same geometric knowledge into lightweight camera tokens during training and removes the 3D model at inference. CamDistill matches the accuracy of direct feature injection without running the 3D teacher at inference. Together, CamChoreo and CamDistill advance camera-motion understanding from clip-level labeling to temporally grounded, compositional recognition. Project page: this https URL.
[CV-34] VIDS-Seg: Towards Reliable Uncertainty Quantification in Pediatric Cardiac Ultrasound Segmentation
链接: https://arxiv.org/abs/2608.10903
作者: Paul Fischer,Ece Ozkan
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Reliable clinical deployment of machine learning requires models that know when they are likely to fail, particularly for subgroups underrepresented in training data. A common case is pediatric care, where models trained on adult cohorts can silently under-perform on children with no indication that something has gone wrong. As retraining with labeled pediatric data is often infeasible, detecting such failures at inference time is a critical clinical need. Building on the VIDS (Variational Inference under Distribution Shifts) framework, we introduce VIDS-Seg, which applies amortized variational inference over a lightweight prediction head to make this adaptive, OOD-aware prior tractable for dense image segmentation. We evaluate VIDS-Seg on left ventricular segmentation in echocardiography, a setting where pediatric anatomy differs systematically from the adult population most segmentation models are trained on, training on an adult cohort (EchoNet-Dynamic) and evaluating zero-shot on a pediatric cohort (EchoNet-Pediatric). Across all age strata, VIDS-Seg matches competitive baselines in segmentation accuracy while producing substantially higher spatial correspondence between predicted uncertainty and segmentation error, an advantage that persists even after applying temperature scaling to all baselines. Downstream, it yields more accurate and stable ejection fraction estimates and more reliable detection of cardiac malfunction in the infant subgroup. Our results indicate that OOD-aware uncertainty quantification can serve as a practical safety layer for deployed segmentation models, enabling detection of silent failures in underrepresented subgroups without retraining or additional labeled data.
[CV-35] Sensor-Informed Per-Point Covariance for Structured-Light 3D Imaging
链接: https://arxiv.org/abs/2608.10888
作者: Sehoon Tak,Jae-Sang Hyun
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Per-point uncertainty models are important in structured-light 3D reconstruction for probabilistic registration, fusion, and quality assessment. In practice, however, point-cloud covariances are often modeled as isotropic constants or inferred from local surface geometry and therefore do not explicitly reflect the measurement process. This is a limitation in fringe projection profilometry (FPP), where phase noise propagates through calibrated reconstruction and produces strongly anisotropic 3D uncertainty. This paper presents a sensor-informed first-order method for constructing a per-point 3 x 3 covariance field from experimentally measured phase precision and calibrated phase-to-depth and phase-to-3D mappings. The formulation separates a rank-1 phase-induced covariance from an effective full-rank completion obtained by incorporating fitted lateral image-space perturbation scales. Repeated-plane experiments under fixed imaging conditions show close alignment of the dominant covariance direction with the viewing ray, and consistency between the dominant phase-induced uncertainty scale and scalar depth uncertainty. In G-ICP registration, the proposed covariance substantially improves over a constant isotropic model while providing a sensor-derived uncertainty representation complementary to conventional geometry-based covariances.
[CV-36] GESTO: Human-Centric Spatio-Temporal Memory for Reasoning in Dynamic Scenes
链接: https://arxiv.org/abs/2608.10886
作者: Ermanno Bartoli,Buwei He,Dennis Rotondi,Sebastian Koch,Federico Tombari,Kai O. Arras,Patric Jensfelt,Yixi Cai,Iolanda Leite
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注:
Abstract:Robots operating in human environments need memories that capture not only what objects exist and where, but also how people use them over time and how individual interactions compose into goal-directed activities. Existing 4D scene graphs preserve object and place histories but omit activity structure, whereas activity representations are either not grounded in persistent 3D scenes or rely on externally provided event boundaries and object associations. We present GESTO (Grounded Event and Spatio-Temporal memOry), a spatio-temporal memory that couples a persistent 4D scene graph with a two-level hierarchy of atomic human–object interactions and goal-driven events. From an RGB-D observation stream, GESTO automatically extracts timestamped interactions, grounds them to persistent scene entities, groups them into events, and uses event context to refine uncertain object associations. A relation-aware tool-calling agent queries the resulting memory for activity-centric spatio-temporal reasoning. We evaluate GESTO on the reproducible text, binary, and time categories of an existing benchmark, together with 40 new Space2Event and Event2Space queries. GESTO achieves scores of 0.71, 0.75, and 0.70 on the standard categories, approaching a method supplied with ground-truth event and object grounding, while substantially outperforming the same reasoning framework when these inputs are removed. It further achieves 0.73 and 0.75 on Space2Event and Event2Space queries. Ablations show that hierarchical event structure and context-aware grounding refinement provide complementary benefits, supporting activity-grounded hierarchical memory for retrospective reasoning in dynamic human environments.
[CV-37] ConfTriage: A Calibration-Aware LLM Triage Framework for Pulmonary Nodule Malignancy with Selective Specialist Deferral
链接: https://arxiv.org/abs/2608.10885
作者: Md Rabiul Islam,Samir Abdaljalil,Erchin Serpedin,Hasan Kurban
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 14 pages, 8 figures. Currently under review
Abstract:Pulmonary nodule malignancy prediction typically depends on image-trained specialist deep learning (DL) models that require substantial annotated imaging data and task-specific training. We investigate whether a generalist large language model (LLM), reading only a faithful natural-language rendering of standard nodule attributes, can serve as a calibrated triage layer. We propose ConfTriage, a confidence-calibrated method built on three pillars: language as the modality, calibration as the safety mechanism, and a selective specialist DL backstop for low-confidence cases. We prove two guarantees: a finite-sample combined-error bound yielding an explicit per-threshold operational certificate, and an oracle inequality showing that excess risk over the Bayes-optimal deferral classifier is controlled by the L1 calibration error of the LLM probability. A controlled seven-way input ablation across five frontier LLMs on LIDC-IDRI shows that natural-language descriptions dominate the diagnostic signal, while low-level image statistics are essentially diagnostically vacuous. ConfTriage achieved an F1 score of 88.22% and an AUC of 0.92, resolving 76.5% of cases using zero-shot LLM inference alone and referring only uncertain cases to the specialist DL backstop. These results demonstrate that clinically meaningful diagnostic information can be captured through structured radiological descriptions and leveraged by calibrated LLMs for selective referral. The framework suggests a practical pathway for combining generalist LLM prediction with specialist AI models in medical decision-support systems. Source code is publicly available at this https URL.
[CV-38] NullEdit: Stealthy Image Protection via VLM Condition Redirection
链接: https://arxiv.org/abs/2608.10870
作者: Weiyao Huang,Liqin Wang,Ziqi Sheng,Wei Lu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages, 6 figures, 3 tables
Abstract:Modern image editors combine vision-language models (VLMs) with diffusion transformer backbones to modify a single reference image according to instructions without fine-tuning. This capability also enables unauthorized manipulation of publicly released images. Existing inference-time defenses either invalidate edits through conspicuous corruption, thereby exposing the protection, or allow them to proceed with identity or reference content drift, thereby failing to prevent the editing behavior itself. We instead target a stealthy and harmless no-op in which the requested edit is suppressed, the output remains natural and source-preserving without conspicuous artifacts or identity replacement, and harmful semantics requested by malicious instructions are absent. We propose NullEdit, which targets the VLM representation jointly formed from the reference image and instruction before it conditions the downstream DiT backbone. Using normal-edit and no-edit anchors, NullEdit redirects this representation, while cross-prompt gradient averaging transfers protection to held out instructions. Across Step1X-Edit and Qwen-Image-Edit on CelebA-HQ and VGGFace2, NullEdit reduces the EditReward IF score by 0.813 on average relative to the SOTA baseline while preserving subject identity and source content.
[CV-39] Multi-View Relational Distillation for Spatial Reasoning with Vision-Language Models
链接: https://arxiv.org/abs/2608.10864
作者: Kiet T. Nguyen,Hanbo Shim,Jinwoo Kim,Seunghoon Hong
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Vision-language models (VLMs) have achieved strong image and video understanding, yet their visual-spatial representations remain geometrically fragile, leading to failures in spatial reasoning needed for embodied AI, robotics, and autonomous driving. Prior approaches to geometry grounding either fine-tune VLMs on spatial question answering, which can perpetuate spurious visual representations, or fuse features from large geometry-grounded vision models, which substantially increases model size at inference. Knowledge distillation from geometry-grounded vision models offers an alternative, but directly matching multi-view teacher features can disrupt the pretrained alignment between visual and textual representations, degrading object- and language-semantic capabilities. We propose multi-view relational distillation (MVRD), which distills patch-wise cosine similarities across views instead of the teacher features themselves. These relations encode geometric correspondences adequate for spatial understanding, while leaving the student representation underdetermined, allowing it to remain close to its pretrained vision- language space. Across representative VLMs, MVRD improves visual-spatial reasoning, outperforming supervised fine-tuning and feature distillation while approaching feature fusion methods with considerably fewer added parameters and lower latency. We show that MVRD makes visual representations more geometric while retaining language alignment, and generalizes to 3D scene understanding tasks such as object grounding, dense captioning, and question answering.
[CV-40] Flex-π: A Multi-Stream World-Action Model with Compute Flexibility
链接: https://arxiv.org/abs/2608.10860
作者: Ge Yan,Jinghao Liu,Yuzhi Fan,Lei Cai,Minwen Liao,Jesse Zhang,Dieter Fox
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: Project page: this https URL
Abstract:World-action models (WAMs) predict the future to act better, but nearly all of them predict only RGB latents, trained purely for pixel reconstruction, with no explicit signal for the 3D geometry or object semantics manipulation needs. We find a surprising free lunch: the same frozen video-generation VAE that encodes RGB also encodes 3D pointmaps almost losslessly, with no pointmap-specific training at all. This lets us supervise Flex- \pi , a 6B-parameter WAM, on 3D geometry and object-centric DINO semantics alongside RGB, at no cost in new sensors, new pre-training, or inference latency. Every visual signal is projected into this shared latent space and denoised jointly with actions inside a Mixture-of-Transformers backbone; per-stream dropout with cross-modality forcing then lets a single trained checkpoint run on any subset of these streams, from a fast action-only mode to full joint generation. The result is a policy that is exceptionally demonstration-efficient and generalizes well, beating the strongest baselines by up to 2-7 \times on dexterous, precise, real-world bimanual manipulation tasks both in and out of distribution, all while running faster than \pi_0.5 . Our project website: this https URL
[CV-41] PolyLayout: Hierarchical VLM-Guided Layout Generation Beyond Rectangular Rooms
链接: https://arxiv.org/abs/2608.10838
作者: Yutong Jiang,Zahra Atashgahi,Carlos Soto Garcia Delgado,Ruben Brokkelkamp,Davide Zanutto,Efşan Sökmen,Shahin Shahkarami
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Generating physically plausible 3D room layouts is essential for home furnishing retail, enabling customers to visualize products in their own homes and confidently make purchasing decisions. However, a gap exists between academic research and real-world application: existing solutions primarily focus on algorithmic strategies for furniture placement, largely neglecting the non-rectangular geometries and strict door/window constraints prevalent in real homes. To bridge the gap, we introduce a hybrid, hierarchical framework tailored for retail, specifically designed to support scalable spatial planning applications. Our system decouples generation into three stages: (1) functional furniture clustering and fine-grained intra-zone placement; (2) macro-routing guided by a vision-language model (VLM) to anchor both these clustered zones and any remaining standalone furniture within diverse polygonal boundaries; and (3) rule-based optimization for collision-free micro-arrangements that respect architectural constraints. We evaluate our system on production-scale catalogs and a representative set of irregular real-world topologies. Our results show that our approach attains the highest perceptual plausibility while maintaining good geometric compliance at relatively low latency, and extends to irregular boundaries that existing methods do not natively support.
[CV-42] UniProbe: A Learnable Token-Level Hallucination Detector for Large VLMs using Multi-Structural Internal Representations
链接: https://arxiv.org/abs/2608.10835
作者: Dvir Samuel,Guy Bar-Shalom,Fabrizio Frasca,Ethan Fetaya,Yftah Ziser,Gal Chechik,Haggai Maron
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Project Page: this https URL
Abstract:Large Vision-Language Models (LVLMs) achieve impressive visual reasoning and dialogue capabilities, yet frequently hallucinate content unsupported by the visual input. Effective mitigation requires token-level localization, enabling targeted intervention without discarding the entire response. Existing detectors require expensive full-model fine-tuning, rely on external verifiers that ignore the model’s generation process, or reduce internal signals to isolated features and hand-crafted statistics, discarding spatial, sequential, and relational structure. We introduce \textbfUniProbe, a lightweight, unified, learnable detector that models a frozen LVLM’s heterogeneous computational trace from a single forward pass. UniProbe constructs a directed graph over image patches, query tokens, and generated tokens, with attention weights encoding their relations. It processes this trace with alternating structure-aware modules: a GNN for relational evidence, a ViT for 2-D visual geometry, and a GRU for response order. Interleaving them allows spatial, relational, and sequential evidence to interact throughout the detector. We further develop a streaming variant for hallucination-aware decoding, which detects and resamples hallucinated tokens during generation, and a self-adaptation strategy aligning the detector with the LVLM’s own generations. Across diverse LVLM backbones, UniProbe achieves state-of-the-art token-level and object-hallucination detection. During decoding, it reduces object hallucinations by up to 55% at 1.06\times the latency of standard generation.
[CV-43] MIRA: Medical Image Reflection for Agent ic Diagnosis
链接: https://arxiv.org/abs/2608.10827
作者: Shengzhi Wang,Jun Yang,Kai Wu,Xiaozhong Ji,Yiwen Ye,Ziyang Chen,Mingliang Xiong,Wen Fang,Mingqing Liu,Mengyuan Xu,Miaoxuan Shan,Caiyan Liu,Bin He,Qingwen Liu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Medical visual agents can use tools to inspect images and retrieve external knowledge, but indiscriminate tool use may introduce noisy or misleading evidence. Reliable diagnosis therefore requires not only acquiring additional observations, but also verifying whether tool actions are necessary and whether the resulting evidence supports the current hypothesis. We introduce MIRA (Medical Image Reflection for Agentic Diagnosis), a medical visual diagnostic framework for autonomous evidence search and reflective verification. MIRA dynamically invokes image-processing operations, including zooming, grounding, pointing, rotation, and measurement, as well as web search, while evaluating the relevance and consistency of the acquired evidence. We develop MIRA through a two-stage training strategy. First, a tool-augmented Monte Carlo Tree Search data engine explores diverse diagnostic hypotheses and jointly verifies visual grounding accuracy and semantic consistency to construct supervised fine-tuning trajectories. Second, reinforcement learning further improves decision-making through online reflective principle evolution: failure cases are distilled into candidate principles, and only principles that improve held-out rollout rewards are retained. Across nine medical visual reasoning benchmarks, MIRA achieves an average score of 64.73, improving its Qwen3-VL-8B backbone by 7.44 points. It also increases useful tool-use judgments from 56.2% to 73.8% and reduces harmful judgments from 8.9% to 1.6%. Qualitative analyses show that MIRA can re-examine evidence, correct premature conclusions, and adapt its tool-use strategy. Project page: this https URL
[CV-44] Neural Introspection Gating for Adaptive KV-Cache Reuse in Vision-Language-Action Models IROS2026
链接: https://arxiv.org/abs/2608.10824
作者: Zhijie Wu,Kento Kawaharazuka,Kei Okada
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 6 pages, 5 figures, Accepted in IROS 2026. Project Page: this https URL
Abstract:Vision-Language-Action(VLA) models map camera images and language instructions directly to motor commands through a single autoregressive transformer. In real-time control, they still spend substantial compute recomputing key-value(KV) representations for visual tokens that barely change across neighboring frames. Recent work such as VLA-Cache reduces that cost by reusing KV states for visually static patches, but its policy relies only on observation-space heuristics and does not account for the model’s own uncertainty. We propose Gated VLA-Cache, a lightweight, training-free extension that augments visual-similarity caching with neural introspection. The method monitors the logit margin between the top two predicted action tokens, a zero-cost confidence signal available during decoding. When the margin drops below a threshold, the cache is invalidated and a full recompute is triggered. Evaluated on four LIBERO benchmark suites with both OpenVLA and OpenVLA-OFT, Gated VLA-Cache improves reliability when blind caching hurts. On LIBERO-Goal and LIBERO-Long, it recovers over 100% of the lost accuracy while retaining 80% of the compute savings.
[CV-45] Modelling Geographic Atrophy Progression using Implicit Neural Representations MICCAI2026
链接: https://arxiv.org/abs/2608.10807
作者: Simone Sarrocco,Paul Friedrich,Florentin Bieder,Christina Bornberg,Philippe Valmaggia,Peter Maloca,Philippe Cattin
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted at MICCAI 2026 Off-Grid Workshop
Abstract:Age-related Macular Degeneration (AMD) is the major cause of blindness in the Western world. Its late dry phase is characterised by irreversible atrophic areas, namely Geographic Atrophy (GA). Longitudinal Fundus Autofluorescence (FAF) image acquisitions are currently the main tool for assessing lesion growth over time at the image level. However, due to its highly individualised progression, the evolution of late AMD remains poorly understood. In this work, we propose using Implicit Neural Representations (INRs) to model GA progression at the individual level in a low-data setting. Our approach generates both FAF and GA segmentation at both past and future time points. Among the comparison models, our method achieves competitive segmentation quality across different scenarios, yielding the lowest Mean Absolute Error (MAE) for the GA lesion area and the highest DICE score, without sacrificing FAF image quality. The code is available at this https URL.
[CV-46] Fast and Memory-Efficient Wavelet Convolutions via I/O-Aware Reformulation
链接: https://arxiv.org/abs/2608.10805
作者: Amit Aflalo,Shahaf E. Finder,Roy Amoyal,Eran Treister,Oren Freifeld
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Wavelet convolution (WTConv) has emerged as an increasingly popular drop-in replacement for standard convolutions, expanding a network’s receptive field exponentially with the number of decomposition levels while keeping the parameter count linear. However, its reference implementation is severely memory-bound due to excessive data movement through high-bandwidth memory (HBM). We develop an I/O model of WTConv to characterize this bottleneck and use it to guide three algebraic reformulations: (1) recomputing the inexpensive Haar analysis butterfly on chip, (2) collapsing the multi-level synthesis cascade into a single closed-form pass indexed by output-coordinate bits, and (3) folding learned per-channel scales into the convolution weights. Together, these reformulations enable an I/O-aware fused implementation that substantially reduces HBM traffic. We evaluate the WTConvNeXt configuration across decomposition levels and a broad range of tensor shapes. Despite performing comparable arithmetic, the reference WTConv is substantially slower than the depthwise convolution it replaces. Our reformulation reduces modeled HBM traffic by approximately 2.55\times , yielding up to a 4.35\times training speedup over the reference while roughly halving peak memory usage. Thus, our reformulation preserves the benefits of WTConv while substantially reducing its execution time and memory footprint, removing the systems overhead that previously limited its practical efficiency.
[CV-47] BPG: Balancing Plasticity and Generalization for Domain Incremental Learning
链接: https://arxiv.org/abs/2608.10804
作者: Qiang Wang,Songlin Dong,Shaokun Wang,Jizhou Han,Xiang Song,Chenhao Ding,Yuhang He,Yihong Gong
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Deep neural networks excel in various tasks but struggle to generalize across evolving data distributions, leading to significant performance degradation under domain shifts. Domain incremental learning (DIL) addresses this challenge by enabling models to continuously adapt while retaining prior knowledge. Among existing DIL approaches, the parameter-isolation paradigm achieves state-of-the-art performance. However, these methods often adopt a one-size-fits-all approach to adapt to new domains, resulting in either insufficient learning capacity or redundant parameters. In this work, we propose BPG, a unified framework that addresses both challenges through two complementary components: BPG-Adapter, which dynamically determines each domain’s adapter hidden dimension based on domain-specific feature separability, and BPG-Inference, a soft domain mixture strategy that integrates multiple domain-specific models at test time, mitigating domain ID misselection. Experimental results on DomainNet, CDDB, and CORe50 demonstrate that BPG consistently outperforms uniform adapter-based approaches and hard domain selection strategies, achieving state-of-the-art average accuracy while reducing forgetting to as low as 0.22% on DomainNet.
[CV-48] Evaluating Semantic and Spatial Guidance for Foundation Model Segmentation of Small-Scale PV in Remote Sensing Imagery
链接: https://arxiv.org/abs/2608.10801
作者: Roni Blushtein-Livnon,Tal Svoray,Osher Rafaeli,Michael Dorman,Itay Fischhendler,Havazelet Yahel,Emir Galilee
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Spatio-temporal PV data are essential for understanding adoption processes in off-grid regions, yet such data remain largely unavailable. Automated segmentation of remote sensing (RS) imagery offers a promising solution; yet, residential PV systems remain challenging targets because of their small size and sparse distribution, resulting in severe target-background imbalance. Vision-language foundation models (FMs) provide a data-efficient paradigm through prompt-based semantic and spatial guidance, but the relative contribution of different prompt types remains unclear. We systematically evaluate SAM3 for small-scale PV segmentation in RS imagery by comparing textual, geometric, and hybrid prompting, under varying supervision levels, training strategies, spatial resolutions, and imaging conditions. Multi-temporal aerial imagery from a large off-grid rural region serves as a study site, with findings validated across three additional datasets. Prompting strategy emerged as the dominant factor governing model behavior. Textual prompting consistently produced the lowest performance and showed the greatest sensitivity to supervision and imaging conditions. In contrast, spatial guidance substantially improved both segmentation accuracy and robustness. Hybrid prompting achieved the highest accuracy and stability, indicating that semantic and spatial guidance provide complementary information. Most performance gains were achieved with only a few hundred annotated samples, demonstrating strong data efficiency. Transfer learning had limited overall impact, with only modest improvements observed for textual prompting under limited supervision. Overall, our findings establish prompting strategy as a key determinant of SAM3 adaptation, robustness, and generalization, highlighting the potential of promptable FMs for scalable PV mapping in data-constrained off-grid regions.
[CV-49] Beyond Fixed Luminance: Towards Panchromatic and Orthochromatic Image Colorization ECCV2026
链接: https://arxiv.org/abs/2608.10798
作者: Swarnim Maheshwari,Syed Imam Ali,Vineeth N. Balasubramanian
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted at ECCV 2026
Abstract:Most image colorization systems operate in Lab space by predicting chroma ( ab ) while preserving an input-derived luminance channel ( L ). While effective on standard benchmarks, this fixed-luminance design restricts brightness changes and becomes unreliable when grayscale formation deviates from natural-image luminance, as in historical orthochromatic photography. We propose a luminance-agnostic colorization framework that formulates colorization as full-RGB image editing using a foundation image-editing model. To bridge modern panchromatic and historical orthochromatic conditions, we introduce a mixed grayscale objective that trains the model under both standard luminance grayscale and a red-insensitive grayscale formation. Experiments on COCO, ImageNet, and a multi-instance benchmark show that our method is competitive on standard grayscale inputs and substantially more robust under orthochromatic inputs, with qualitative comparisons and a human study indicating fewer visible color artifacts.
[CV-50] E3mo-Bench: A Scalable Benchmark for Multimodal Evoked and Expressed Emotion Understanding via Bayesian Pairwise Alignment
链接: https://arxiv.org/abs/2608.10796
作者: Lancheng Gao,Ziheng Jia,Shengyan Li,Zixuan Xing,Jiarui Wang,Huiyu Duan,Xiongkuo Min
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Understanding both expressed and evoked emotions is critical for multimodal large language models (MLLMs) to achieve comprehensive affect-aware interactions. However, existing benchmarks typically examine expressed and evoked emotions in isolation or are constrained to coarse-grained and incomplete affective characterizations. To bridge this gap, we introduce E ^3 mo-Bench, a scalable benchmark comprising 12,314 question-answer pairs across 2,524 videos with predefined affective perspectives. It evaluates evoked and expressed emotion understanding via 3 complementary tasks: emotion perception, open-vocabulary recognition, and valence-arousal-dominance (VAD) assessment. To efficiently scale reliable continuous annotations, we propose Bayesian Pairwise Alignment, which aggregates sparse, low-burden pairwise judgments into anchor-referenced VAD estimates. Furthermore, we develop E ^3 mo-Score, a training-free agent that aggregates complementary judgments from a five-model committee to improve VAD estimation. Extensive experiments validate the effectiveness of our framework and expose a pronounced performance skew between evoked and expressed emotion paradigms. These findings, coupled with MLLMs’ persistent deficits in fine-grained recognition and dimensional assessment, chart a clear course for advancing multimodal emotional intelligence.
[CV-51] MVTrack: Ultrafast Appearance-Free Moving Object Tracking from Compressed Bitstreams ECCV2026
链接: https://arxiv.org/abs/2608.10790
作者: Iñaki Erregue,Kamal Nasrollahi,Sergio Escalera
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: This paper has been accepted to the 2nd workshop on Low-Level Vision Frontiers with Generative AI, Preference Optimization, Agentic Systems and World Models (LoViF) at ECCV2026
Abstract:Deploying modern video trackers at scale is bottlenecked by the computational cost of RGB-based object detectors. To this end, we present MVTrack, an ultrafast tracker for moving objects that operates directly on H.264 bitstreams. MVTrack combines MVDet, a lightweight detector for motion vector fields, with MVLink, a minimalist kinematic association module. On VIRAT, MVTrack outperforms YOLO26n while using 60 \times fewer parameters, requiring 40 \times fewer FLOPs, and reducing CPU latency by 8.6 \times . These results demonstrate that compressed video data alone can enable accurate and scalable surveillance tracking, thereby bypassing the need for pixel reconstruction.
[CV-52] Compositional Benchmark Synthesis for Hierarchical Human Action Recognition
链接: https://arxiv.org/abs/2608.10765
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)
备注:
Abstract:Recognizing human behavior across levels of abstraction, from atomic actions to long-horizon intentions, requires data annotated along a semantic hierarchy. Large corpora provide isolated, atomically labeled clips without temporal composition, whereas recorded composite-activity corpora offer shallow, domain-narrow, fixedhierarchies. A benchmark-generation and evaluation frameworkis proposed that synthesizes a four-level hierarchical-intention benchmark, spanning actions, activities, low-level intentions (LLIs), and high-level intentions (HLIs), from a flat single-label action corpus while retaining real pre-extracted features at the action level. Episodes are assembled by a transition model under a subject-consistency constraint, and a coverage-aware sampler reduces the subject usage Gini from 0.566 to 0.248. Synthesizing such a benchmark raises a circular-supervision risk that recorded datasets avoid: if the rules generating the episodes also govern the evaluation, models can succeed by recovering the generator rather than through genuine reasoning. Validity is addressed by design, holding sequence-generation rules disjoint from the first-order-logic rules used at evaluation. The instantiation yields 15,002 episodes. Four reference baselines from different model families characterize difficulty, not as recognition methods. A compositional held-out gap of 0.13 to 0.17 macro-F1 appears across all baselines, including a graph-aware model that recognizes best yet does not close the gap, indicating a structural property of the benchmark rather than a model artifact. A logic-free baseline still violates the held-out semantic rules above their intrinsic data rate, and the order-destroying control changes macro-F1 within seed variation, serving as a generator-consistency check. Theontology, transition model, and generator are released so the benchmark can beregenerated and extended.
[CV-53] FADE: From Passive Verification to Active Discovery in Counterfactual Video Understanding
链接: https://arxiv.org/abs/2608.10764
作者: Fufangchen Zhao,Jinhu Fu,Jiachen Lei,Jiahong Wu,Xiangxiang Chu,Danfeng Yan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Counterfactual video understanding evaluates whether models grasp physical and commonsense regularities. However, existing multiple-choice question (MCQ) benchmarks inadvertently leak target events through their questions and candidate options. This reduces the core challenge from active discovery to text-guided verification. In this paper, we present FADE, an effective training framework for counterfactual discovery and explanation. Our method is built on an evidence-first, two-stage training paradigm. First, evidence-internalized supervised fine-tuning grounds the model’s predictions in decisive visual anomalies. Second, we apply a fading-anchor reinforcement learning strategy that progressively removes textual guidance, compelling the model to independently discover and explain evidence. To rigorously evaluate this capability, we also introduce an effective pipeline that converts existing MCQ datasets into aligned MCQ, open-ended question answering (OQA), and captioning tasks without requiring additional data curation. Our simple approach yields strong results. Using Qwen3-VL-8B as the baseline, FADE achieves state-of-the-art strict paired scores across all three tasks on DualityVidQA-test and IPV-Bench, outperforming GPT-5.6. In specific, when transitioning from constrained MCQs to unconstrained OQA and captioning, our model demonstrates remarkable robustness. Its performance retention is 90.4% and 67.4% on DualityVidQA-test-substantially higher than the 48.1% and 30.7% retained by GPT-5.6. We hope this simple framework can serve as a solid baseline for future research in unconstrained counterfactual video understanding.
[CV-54] Where To Look? : Causal Tracing of Vision Encoders in VLM
链接: https://arxiv.org/abs/2608.10758
作者: Naren Kumar S,Tirth Bhatt,Mayank Singh
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 10 Pages, 6 figures
Abstract:Vision-language models can describe an image with remarkable accuracy, yet a more fundamental question remains unanswered: what visual information actually drives their answers? In this work, we investigate this question through causal tracing, and we observe that highly causal vision tokens often lie outside the target region. Extending the analysis to larger vision-language models reveals a similar pattern across models and corruption settings, suggesting that strong multimodal performance does not necessarily imply spatially localized causal representations. We further investigate: can these models preserve visual structure when appearance cues are removed? and find that visual cues are exploited to understand visual structures. Together, our experiments expose a gap between seeing, using, and reasoning over visual structure, and provide a causal framework for studying how visual information is transformed, preserved, and ultimately used by modern vision-language models.
[CV-55] Embodied Multimodal Grounding for Open-Vocabulary Mobile Manipulation via Semantic 3D Gaussian Splatting
链接: https://arxiv.org/abs/2608.10756
作者: Huosen Ou,Dongni Song,Yuncong Wang,Tao Zhou,Yiding Ji
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages, 11 figures. Accepted to ACM Multimedia 2026 (MM '26)
Abstract:Embodied mobile manipulation requires language, visual observations, three-dimensional scene structure, and action feasibility to be aligned before execution. We study open-vocabulary target grounding with few-shot manipulation in local household workspaces and present an embodied multimodal grounding framework that integrates active multi-view Semantic 3D Gaussian Splatting (Semantic-3DGS), reachability-aware base positioning, and a diffusion-based vision-language-action policy. A task-driven local Semantic-3DGS serves as a shared interface across active sensing, language-conditioned 3D localization, obstacle-aware scene reasoning, base preparation, and semantic conditioning of the action model. To preserve pretrained action priors, the 3D semantic cues are injected only into the late action-expert blocks. In expanded 50-trial real-robot evaluations against representative vision-language-action (VLA) approaches, the full system achieves 60% long-horizon success compared with 40% for PointVLA and 28% for DexVLA, and reaches 74% success in heavily cluttered manipulation compared with 52% for the single-view variant and 46% for PointVLA. It also maintains 75% success under a 75 cm height shift and eliminates photo-induced false grasps. These results indicate that explicit, refreshable 3D semantic grounding can improve robustness under clutter, occlusion, viewpoint variation, and embodiment constraints.
[CV-56] Beyond Pixels: From Video Priors to 4D Worlds
链接: https://arxiv.org/abs/2608.10744
作者: Zihao Liu,Xiaolong Shen,Zhenglin Zhou,Ruijie Quan,Yi Yang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project page: this https URL
Abstract:4D generation synthesizes dynamic 3D scenes from conditions such as text or images. Existing methods either reconstruct generated RGB videos with a separate 4D model or adapt a particular video generator to predict geometry directly. The former suffers from distribution mismatch and error propagation, whereas the latter ties 4D prediction to a specific generator and may require retraining when the generator or conditioning regime changes. We ask whether the final denoised latents of video models that share a variational autoencoder (VAE) can instead provide a reusable interface to explicit 4D prediction. Building on this insight, we introduce direct latent-to-4D generation and instantiate it as Latent-to-4D, which bypasses RGB by aligning a video latent with the token grid of a pretrained 4D decoder and refining it through frame-wise and global spatiotemporal attention. Trained on roughly 1K existing reconstruction clips, a single checkpoint transfers unchanged across multiple video diffusion transformers within the same VAE family. On Text4D-200 and I4D-200, Latent-to-4D surpasses matched same-latent Wan+4RC cascades in projection-based DINO-F1 by 2.88–3.45 and 5.81 points, respectively, while also being preferred by human raters for geometry, temporal stability, and overall quality.
[CV-57] Rethinking LLM Verification: Evidence Structure Uncertainty and Selective Refinement EMNLP2026
链接: https://arxiv.org/abs/2608.10725
作者: Uma Ranjan,Kunal Tilaganji,Aditya Koul,Anurag Mahipal,Dashpreet Singh,Hriday Rana,Manan Jain,Sidharth Gupta,Ajo Babu George,Vineeth Balasubramanian,Nagarajan Natarajan,Amit Sharma
类目: Computer Vision and Pattern Recognition (cs.CV); Symbolic Computation (cs.SC)
备注: Findings Track at the Conference on Empirical Methods in Natural Language Processing (EMNLP 2026)
Abstract:Large language models (LLMs) often rely on shortcuts rather than systematic reasoning, raising safety concerns in medical applications. Allowing models to abstain when uncertain improves reliability but introduces a coverage accuracy tradeoff. We propose a two-stage framework for medical hypothesis verification in multiple-choice settings that manages this tradeoff through targeted ontology grounding, applied only when the model abstains. We show that abstention is not random but reflects genuine uncertainty, with abstained predictions associated with lower confidence. Across two frontier models (GPT-5.5, accessed via the Azure OpenAI API, and DeepSeek-R1), the proposed framework improves question-level accuracy by 9.6 percentage points (82.9% to 92.5%) and hypothesis-level accuracy by 4.2 percentage points (92.0% to 96.2%). Our experiments conducted on MedReason and MedQA show that abstention can be repurposed as a control signal for selective reasoning refinement, achieving knowledge-graph-level performance without explicit knowledge graph construction.
[CV-58] InterPruner: Interactive Structured Pruning via Taylor-Implicit Criterion and Language-Prior Modulator for Multimodal Object Detection
链接: https://arxiv.org/abs/2608.10724
作者: Qi Ming,Zihan Yang,Shaoguang Huang,Si Sun,Hanqing Zhang,Nanqing Liu,Jiahui Lv,Juan Fang,Aleksandra Pizurica
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Multimodal object detection proves effective in remote sensing, especially the RGB-Infrared paradigm. The parallel feature extractors provide rich multimodal information for robust detection, yet introduce substantial channel redundancy and computational overhead. Existing pruning methods can reduce channel redundancy, but they are designed for unimodal backbones, overlooking cross-modal interactions and dynamic scene-wise redundancy. In this paper, we propose InterPruner, the first interactive structured channel pruning framework for RGB-infrared object detectors. Specifically, we first derive a Taylor-Implicit Criterion(TIC) to quantify channel importance via high-order Taylor expansion and the implicit function theorem. Then, a Modality Interaction Redundancy Analyzer (MIRA) identifies redundant channels via mutual compensability assessment. Finally, a Scene-Prior Channel Anchor (SPCA) uses language priors as semantic anchors to measure channel-scene relevance for dynamic channel importance estimation. Cross-modality channel pruning for RGB-Infrared detection is yet unexplored. Extensive experiments on RGB-infrared object detection dataset demonstrate that InterPruner maintains high performance with negligible degradation. Specifically, it even achieves a 0.6% mAP increase on the FLIR dataset when pruning 50% of the channels. Code will be available on GitHub to facilitate future work.
[CV-59] Grid-Preserving Knowledge Distillation: Transferring Convolutional Inductive Bias to Vision Transformers under Data Scarcity
链接: https://arxiv.org/abs/2608.10723
作者: Junyong Choi,Cheolhyeon Park,Jaehoon Cho
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Vision Transformers underperform convolutional networks when training data is scarce, and distilling convolutional inductive biases from a CNN teacher is an effective remedy that leaves the deployed model unchanged. General-purpose feature distillation, however, transfers little in this setting. The pooling, flattening, and logit-space projections it inherits from CNN to CNN pipelines discard the spatial grid in which locality and translation equivariance are encoded, and unlike a convolutional student, a ViT cannot rebuild that structure on its own. In this paper, we propose iBKD, a distillation framework that preserves the grid along the entire transfer path. Its core module, the Inductive Bias Attention Module, aggregates every student layer onto the teacher grid with learned weights, sharpens structural cues with channel and deformable spatial attention, and injects them through convolutional cross-attention that operates between grids rather than between token sets. The module is used only during training, so the deployed model is an unmodified ViT with no inference overhead. Across seven Transformer backbones and six data-scarce benchmarks, iBKD outperforms both locality-guidance methods and general knowledge distillation baselines, and its margin widens as training data shrinks.
[CV-60] Compact Feed-Forward 3D Gaussians via Saliency-Guided Primitive Merging
链接: https://arxiv.org/abs/2608.10712
作者: Tim-Felix Fassch,Jochen Kall,Cyrill Stachniss
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:3D scene reconstruction, modeling, and rendering are highly relevant for numerous tasks, and 3D Gaussian splatting has become a standard choice in this context. Its feed-forward variants provide fast reconstruction from sparse input views but often produce per-pixel primitives, leading to highly redundant and thus inefficient representations. We present a structure-aware merging pipeline that takes per-pixel primitives from any feed-forward method and consolidates them into a compact, content-adaptive Gaussian set while largely retaining visual quality at just \frac120^\textth of the Gaussians of a per-pixel method. We group spatially coherent Gaussians of similar appearance into variable-size clusters via adaptive superpixel segmentation guided by a saliency map, which allocates fine segments to textured regions and coarse segments to homogeneous areas. We compress each cluster into a compact latent representation through a learned encoder, then match and consolidate representations across views based on geometric overlap and feature similarity via a learned merger. A level-of-detail decoder then produces the final Gaussians at a controllable resolution, enabling a flexible quality-efficiency trade-off at inference. As a post-processing module, the pipeline is backbone-agnostic, leveraging the strengths of existing feed-forward methods. This leads to better and more robust quality than achieved by previous approaches that target a reduction in primitive count, while providing a highly compact representation, that can be rendered efficiently.
[CV-61] Self-Geometry: GT-Free and Plug-and-Play Test-Time Adaptation for Geometrically Consistent 3D Vision Foundation Models
链接: https://arxiv.org/abs/2608.10708
作者: Seokhyun Youn,Dahyeon Kye,Sung-Ho Bae,Jihyong Oh
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Project page: this https URL
Abstract:Recent Vision Foundation Models (VFMs) predict depth, camera pose, and pointmap in a single forward pass without per-scene optimization, achieving strong generalization. However, enforcing explicit multi-view geometric consistency, e.g., through bundle adjustment, is computationally costly and is thus not imposed during VFM pretraining, so such inconsistency can arise. To address this, implicit self-consistency derived from model outputs (e.g., pointmaps, features), though enforced at test-time in prior work, delivers inherently limited performance gain, especially on scenes where the pretrained VFM is highly inaccurate. In contrast to this implicit signal, we propose Self-Geometry, a plug-and-play test-time adaptation pipeline that directly imposes explicit multi-view geometric constraints using 2D pixel correspondences as pseudo ground-truth. Our proposed Self-Geometry consists of Geometric Disentanglement Optimization, which combines Multi-View Consistency and Epipolar Consistency losses with Gradient Disentanglement to prevent gradient conflict; Frame Angular-Neighbor, a view sampler based on SO(3) geodesic distances for lightly imposing these constraints; and Lightweight TTA, which adapts VFMs via LoRA. Our method achieves consistent improvements in both pose and geometry estimation across six VFMs (VGGT, \pi^3 , DA3-Giant/Large/Base/Small) and four benchmarks (7Scenes, ETH3D, ScanNet++, HiRoom).
[CV-62] MMArt A Multi-Perspective Multimodal Dataset for Visual Art Understanding
链接: https://arxiv.org/abs/2608.10706
作者: Shuai Wang,Wangyuan Ding,Yixian Shen,Jia-Hong Huang,Stevan Rudinac,Monika Kackovic,Nachoem Wijnberg,Marcel Worring
类目: Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM)
备注:
Abstract:Recent vision-language models demonstrate impressive general visual understanding, yet their art interpretation remains shallow: they describe surface content but struggle with formal analysis, grounded historical interpretation, or affective characterization. We argue this is not only a model but also a dataset limitation. Existing art datasets are single perspective resources, where no dataset provides narrative, formal, emotional, and historical perspectives simultaneously for the same artworks. We introduce MMArt, a large-scale dataset of 74,234 WikiArt paintings, each annotated with four independently annotated perspectives plus a harmonized unified caption, produced by specialized vision-language models or human annotation and validated through complementary quality evaluations. Two complementarity analyses establish that perspectives encode genuinely distinct information. A generative analysis shows that formal analysis descriptions best preserve compositional style, and historical descriptions carry strong affective signal in reconstructed images. A discriminative retrieval analysis reveals task-asymmetry: narrative descriptions drive retrieval (R@1 = 44.0%), while formal descriptions, strongest for reconstruction, are nearly nondiscriminative at retrieval scale (R@1 = 7.8%). Leave-one-out analysis further confirms that historical descriptions are the least replaceable perspective across both tasks. Together, the two analyses establish that no single perspective suffices for all tasks, directly motivating MMArt multi-perspective design. The dataset, code, and additional information are available at this https URL.
[CV-63] Embedding Rotation Invariance for Provable Multi-Oriented Scene Text Recognition
链接: https://arxiv.org/abs/2608.10684
作者: Zhibin Ma,Pengwen Dai,Yi Liu,Xugong Qin,Chenyun Yu,Xiaochun Cao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Multi-oriented text is ubiquitous in real-world scenes and remains a major challenge for scene text recognition (STR). Existing rotation-aware methods explicitly estimate text orientation. However, due to the lack of theoretical guarantees, they are prone to error accumulation, increased computational cost, and strong reliance on data. In this work, we incorporate rotation invariance into the STR framework to address these limitations. Specifically, we adopt an encoder-decoder architecture, embedding rotation equivariance in the encoder and rotation invariance in the decoder to construct a fully rotation-invariant network. On the decoder side, we first identify and prove the rotation-invariant property of the cross-attention mechanism and use it to formulate a rotation-invariant text decoder that maps visual features to output text in a rotation-invariant manner. On the encoder side, we propose a rotation-equivariant local-global extraction network that integrates deep equivariant convolutions with self-attention, enabling rotation-equivariant feature extraction while modeling inter-character dependencies and preserving fine-grained visual details. By integrating the encoder and decoder, we obtain an end-to-end Rotation-Invariant Scene Text Recognition network (RISTER). RISTER provides rotation invariance with theoretical guarantees, enhancing robustness on multi-oriented samples without introducing additional inference computation or relying on data-driven orientation correction. Experiments show that RISTER achieves state-of-the-art performance on both standard and multi-oriented benchmarks, surpassing the second-best model by 4.0 percent in accuracy on the general multi-oriented dataset.
[CV-64] Visual Geometry Foundation-Aware Gaussians for Single-Frame Surround-View Driving Reconstruction
链接: https://arxiv.org/abs/2608.10682
作者: Junhong Lin,Jinlong Wang,Xianda Guo,Yanlun Peng,Wei Zheng,Guoqing Liu,Hanli Wang,Tiesong Zhao,Wei Gao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Single-frame surround-view reconstruction faces severe geometric instability and rendering artifacts due to minimal inter-camera overlap. While existing methods rely on complex decoders or auxiliary cues, they remain bottlenecked by the weak geometric capacity of upstream features. We argue that leveraging pretrained visual geometry priors strengthens upstream representations and alleviates the geometric ambiguity in sparse surround views. To this end, we propose VGGD, a visual geometry foundation-aware 3D Gaussian Splatting framework for feed-forward surround-view driving reconstruction, which shifts geometric modeling to the frontend and adapts foundation priors to the driving camera setting. First, VGGD leverages VGGT to provide transferable multi-view geometric prior tokens. Next, we introduce a Dual-Path Neck to decouple geometry-consistent and appearance-aware representations, improving appearance completion in weakly observed regions. We further apply Scale Warmup to stabilize early geometry learning and suppress scale drift under ego-pose changes. Finally, we use a hybrid pixel–volume Gaussian decoder to produce a renderable 3D Gaussian scene for novel-view synthesis. Experiments on the nuScenes single-frame benchmark show that VGGD achieves the best overall rendering quality among the compared methods and improves relative geometric consistency.
[CV-65] Bridging Severe Cross-Modal Misalignment: End-to-End Visible-Infrared Object Detection via Explicit Feature-Domain Affine Registration
链接: https://arxiv.org/abs/2608.10680
作者: Qi Ming,Yuyang Wang,Mingjing Zhao,Yifan Xiao,Zhixin Guo,Zhiqiang Zhou,Peng Sun,Juan Fang,Fuqiang Yang,Xudong Zhao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Visible-infrared object detection relies on complementary RGB and thermal cues, but its performance is often degraded by cross-modal spatial misalignment. Most existing methods rely on implicit feature adaptation to handle weakly misaligned scenarios, while large-offset geometric discrepancies remain insufficiently addressed. In this paper, we propose a Joint Feature-domain Registration and Detection network (JFRDet), an end-to-end visible-infrared oriented object detector tailored for severely cross-modal geometric discrepancies. JFRDet introduces a Cross-Modal Affine Alignment (CMAA) module to estimate an image-level affine transformation for explicit multi-level feature alignment. Note that illumination changes directly affect the reliability of RGB cues, an Illumination-Guided Complementary Fusion (IGCF) module adaptively exploits modality reliability under varying illumination conditions for cross-modal fusion. Then, an Alignment Quality-Consistency Gating (AQCG) strategy stabilizes joint optimization by modulating detection supervision according to alignment reliability and gradient consistency. We further construct DroneVehicle Misaligned (DVMA), a benchmark for evaluating visible-infrared oriented object detection under severe cross-modal geometric misalignment. The proposed JFRDet achieves 69.7% \mathrmmAP_50 on DVMA, which represents state-of-the-art (SOTA) performance. The code and dataset will be available on GitHub.
[CV-66] Chartography: A Benchmark for Professional Chart Understanding ECCV2026
链接: https://arxiv.org/abs/2608.10677
作者: Suhaas Garre,Chris Mutty,Sushant Mehta,Edwin Chen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 16 pages, 5 figures, 5 tables. Accepted at the 2nd Workshop on Benchmarking Evidence-Aligned Multimodal Reasoning (BEAM 2), ECCV 2026
Abstract:Professionals across medicine, engineering, finance, manufacturing, and the sciences often make consequential decisions from charts. Existing chart benchmarks do not sufficiently measure this ability: they are dominated by bar, line, and pie formats, rely on shorter reasoning chains, and are nearing saturation, with frontier models already scoring 80-90%. We introduce Chartography, a benchmark of 100 tasks that pair charts drawn from professional practice, in domain-specific formats that standard chart benchmarks rarely include, with questions written by professionals who read these charts for a living and independently verified by three additional experts. In an evaluation of 30 frontier-model configurations (20 scored trials per task), the best configuration reaches only 45.0% mean pass@1; the remainder span 9.0-39.5%. Failures concentrate in visual perception: models can miss nuanced features, misread values along sparsely labeled axes, mishandle projected 3D geometry, and violate domain conventions encoded in the chart. We release all tasks, images, provenance metadata, and evaluation code.
[CV-67] VERDICT: Training-Free Step-Wise Verification of Multimodal Reasoning via Disagreement-Aware Consensus
链接: https://arxiv.org/abs/2608.10665
作者: Rohit Sinha,Kunal Tilaganji,Tanuja Ganu,Nagarajan Natarajan,Amit Sharma,Vineeth Balasubramanian
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Computer Science and Game Theory (cs.GT)
备注: European Conference on Computer Vision 2026
Abstract:Multimodal large language models often generate reasoning chains containing subtle errors that lead to incorrect answers. Current verification approaches have notable limitations. Existing approaches either require expensive labelled supervision with inconsistent cross-task performance or aggregate scores from multiple sources by simple aggregations, missing a key insight: when these scores disagree, that disagreement itself carries important information about whether a reasoning step is truly valid or not. We formalise this as a coupled scoring problem among disparate, frozen verifiers, interpretable as a coordination game with a unique closed-form equilibrium where agreement signals valid steps while disagreement reveals instability. Towards this end, we propose a training-free domain-agnostic step-wise verification approach we call VERDICT: VERification via Disagreement-Informed Coupled Thresholding. To our knowledge, VERDICT is the first training-free verifier that makes the structure of cross-modal disagreement explicit and actionable. It computes consensus scores through a closed-form solution, enabling both disagreement-aware filtering and stability-conscious ranking of reasoning steps. Evaluated across six benchmarks, \method consistently improves over the base model by up to +5.95%, and performs competitively with domain-specific critics that demand extensive supervision, demonstrating that cross-modal agreement provides robust verification signals without task-specific adaptation and Training-Free Verification
[CV-68] Cross-View Sequential Visual Localization with Spatio-Temporal Context Modeling for Autonomous Driving
链接: https://arxiv.org/abs/2608.10660
作者: Jiaping Wang,Shaobo Li,Zhen Wang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Continuous and reliable localization is essential for autonomous driving. Cross-view visual localization matches ground images with satellite maps, providing complementary localization cues for pipelines that depend on Global Navigation Satellite System (GNSS) signals and high-definition (HD) maps. Most existing cross-view visual localization methods process each frame independently, leaving temporal information underused and limiting accuracy under dynamic occlusion, illumination variation, and repetitive textures. This study proposes a temporal-context-enhanced framework for cross-view sequence visual localization. The proposed recurrent cross-frame module aggregates historical context from the previous state to enhance the coarse ground feature of each current frame. These enhanced features facilitate satellite candidate-region classification, while hierarchical fine-grained features enable precise local offset estimation. On the CVIS dataset, the proposed method reduces mean localization error from 3.80 m to 1.57 m and increases R@1 m from 8.14% to 40.22%. Direct transfer to KITTI-CVL achieves a mean error of 2.61 m, with target-domain fine-tuning further reducing the mean error to 2.27 m. Zero-shot field experiments on a real-world vehicle achieve a mean error of 2.84 m and R@5 m of 96.86%. These results demonstrate that temporal context enhancement significantly improves cross-view localization accuracy and supports robust deployment on public benchmarks and real-world roads.
[CV-69] PolypVision: A Three-Stage Hierarchical Deep Learning Framework for Classification and Segmentation of Colorectal Polyps
链接: https://arxiv.org/abs/2608.10649
作者: Hamidreza Bolhasani,Hamidreza Rastad,Amir Mohammad Akbari,Mohammad Tashakoripour,Parnian Asadollahi,Ata Khodami,Mojgan Forootan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Colorectal cancer (CRC) remains one of the leading causes of cancer-related mortality worldwide, predominantly arising from precancerous polyps. Accurate detection, segmentation, and endoscopic and histological classification of colorectal polyps are crucial for timely clinical intervention. In this study, we present PolypVision, a three-stage hierarchical deep learning framework that sequentially performs: (Stage 1) binary classification of polyps as adenomatous or hyperplastic, with simultaneous Paris and JNet classification, using EfficientNetV2-M with Focal Loss; (Stage 2) polyp segmentation with recommended resection method using a UNet++ decoder with the Stage 1 backbone as encoder, optimized with Dice and BCE losses; and (Stage 3) adenoma subtype classification (tubular, tubulovillous, villous) using EfficientNetV2-M with transfer learning from Stage 2. Evaluated on three public datasets – PolypGen, Kvasir-SEG, and CVC-ClinicDB – PolypVision achieves an AUC of approximately 0.99 for frame classification and a detection mAP@50 of 94.4% on Kvasir-SEG, outperforming or matching state-of-the-art methods. Gradient-weighted Class Activation Maps (Grad-CAM) confirm that the model attends to clinically relevant lesion features. The framework is device-independent, operating across diverse endoscopic imaging systems without hardware-specific adaptation. These results demonstrate that a hierarchical, transfer-learning-driven pipeline with task-specific loss functions offers a robust, device-independent, and clinically meaningful approach to automated colorectal polyp analysis. PolypVision is freely available as a web application at this https URL, a DataBioX initiative, with a free usage tier open to all users.
[CV-70] Precise Top-Layer Fabric Segmentation for Fabric Destacking with Edge- and Shape-Aware Deep Networks
链接: https://arxiv.org/abs/2608.10648
作者: Wenbo Dong,Dipankar Bhattacharya,Akinari Kobayashi,Akira Seino,Fuyuki Tokuda,Xuzhao Huang,Kai Tang,Norman C. Tien,Kazuhiro Kosuge
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注: 7 pages, 3 figures. Published in IEEE ICMA 2025. Author’s accepted manuscript. Code: this https URL
Abstract:Fabric destacking requires precise segmentation of the topmost fabric layer, a task complicated by subtle fabric boundaries and high visual similarity between fabric layers. Existing semantic and edge-based segmentation approaches often struggle with these complexities, limiting the performance of robotic manipulation for different tasks. In this work, a novel segmentation training architecture tailored for top-layer fabric segmentation in stacked fabrics is proposed. The method extends the classical encoder-decoder framework by introducing two specialized branches - an edge-aware branch and a shape-aware branch - that are used to supervise the backbone network for better tuning. The edge-aware branch enhances boundary delineation, while the shape-aware branch guides the network to capture and align the overall fabric shape with reference masks derived from Computer Aided Design (CAD) models. Experiments on a real-world fabric dataset demonstrate that the training approach outperforms established baselines, verifying the effectiveness of the multi-branch design through both quantitative results and ablation studies.
[CV-71] MedUP: Awakening Unified Understanding and Perception in Medical Vision-Language Models
链接: https://arxiv.org/abs/2608.10635
作者: Yuan Wang,Hualiang Wang,Yixin Chen,Songtao Jiang,Shujian Gao,Jiaming Lin,Siming Fu,Jian Wu,Zuozhu Liu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 10 pages, 6 figures
Abstract:Medical Vision-Language Models (Med-VLMs) excel at verbalizing visual content, yet precise visual perception, segmentation, and grounding remain challenging. Existing approaches either verbalize regions as coordinate strings or rely on external modules that decouple perception from understanding, creating representation gaps for region-language alignment. We present MedUP, a Med-VLM that natively unifies perception and understanding within a shared token space. At its core lies UniMedTok, a region tokenizer that encodes masks as discrete tokens in the LLM vocabulary, enabling the model to seamlessly interleave mask tokens with text. We curate UniMed-Train, a 1.84M-instance corpus spanning text-guided segmentation, region-grounded understanding, medical VQA and CoT-based segmentation, and introduce UniMed-Bench for unified evaluation. Extensive experiments show that MedUP outperforms native, agentic, and dual-decoder Med-VLMs across all tasks while remaining competitive with specialist segmentors, demonstrating the strong potential of unified understanding and perception modeling.
[CV-72] Gaussian Sculpting: End-to-End Controllable Surface Reconstruction via Field Optimization
链接: https://arxiv.org/abs/2608.10602
作者: Ke Jiaxin,Juncheng Liu,Yi Wang,Zhouhui Lian,Bin Liu,Shengfa Wang,Xiangjia He
类目: Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR)
备注:
Abstract:3D Gaussian Splatting (3DGS) has recently enabled real-time novel view synthesis with impressive quality. However, it struggles to recover accurate surfaces under limited viewpoints and due to the inherent irregularity of Gaussian primitives. The resulting geometric errors are notoriously difficult to correct manually. To address these issues, we propose Gaussian Sculpting, a fully differentiable end-to-end framework for high-quality surface reconstruction. Our key insight is to anchor Gaussians onto an evolving differentiable surface, allowing them to guide signed distance field (SDF) optimization instead of extracting the surface only during post-processing. To enable stable gradient isolation during joint optimization, we design a bi-level training strategy in which the outer loop optimizes the geometry represented by the SDF, while the inner loop updates the Gaussians with the geometry fixed. We further impose constraints on Gaussian parameters to ensure consistency with the underlying surface, thereby improving both geometric and appearance fidelity during optimization. In addition, we introduce a multi-resolution subdivision scheme based on octree-like partitioning to preserve fine details while reducing memory consumption. Experiments on object-level scenes demonstrate that our method effectively removes redundant surfaces, recovers missing structures caused by limited viewpoints, and achieves strong reconstruction quality even at relatively low resolutions.
[CV-73] BooST: Bridging Semantics and Motions for Efficient Skill Transfer
链接: https://arxiv.org/abs/2608.10600
作者: Jusuk Lee,Daesol Cho,Jonghun Shin,Seungyeon Yoo,Jonghae Park,Taekbeom Lee,H. Jin Kim
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Project page: this https URL
Abstract:Skill abstraction—the process of learning reusable and temporally extended behaviors—has emerged as a key paradigm for improving sample efficiency and generalization in robot learning. For efficient skill transfer to real robots, learned skills must generalize across tasks and domains, remain robust to visual and dynamic perturbations, and be efficient enough for practical deployment. However, existing methods typically satisfy only a subset of these properties, as they capture either high-level semantic intent (what) or low-level motion dynamics (how). This incomplete skill transfer yields weak priors for policy learning, thereby demanding substantial in-domain data for downstream adaptation. To address these challenges, we introduce BooST, a two-stage framework that explicitly bridges semantics and motions to satisfy all three desiderata. BooST first leverages a cross-modal VQ-VAE to capture both semantic intent and motion dynamics, yielding a unified skill representation. It then distills this representation into a lightweight policy for efficient downstream adaptation to new tasks. Extensive experiments across simulation and real-robot settings demonstrate that BooST achieves superior few-shot adaptation, cross-domain skill transfer, and robustness to dynamic visual distractors, while maintaining a lightweight yet expressive design suitable for real-world deployment.
[CV-74] Rethinking Data Efficiency in Industrial Dense Prediction: Pretraining Coherence Not Inductive Bias Determines ViTs Low-Data Advantage
链接: https://arxiv.org/abs/2608.10590
作者: Haoran Sui,Yaoyuan Jia
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 14 pages, 10 figures, 17 tables. This paper targets industrial defect detection via vision transformer and CNN alignment grafting
Abstract:Vision Transformers (ViTs) are widely believed to require more labeled data than CNNs for industrial dense prediction. Through controlled experiments on four industrial datasets, we show that the data-efficiency gap stems from pretraining incoherence, which refers to the statistical mismatch between ImageNet-pretrained ViT backbones and COCO-pretrained CNN necks, rather than from inherent self-attention deficits. We characterize the cross-architecture feature gap and propose a lightweight AlignBlock family for pyramid-level feature recalibration. Our core finding empirically identifies a data-efficiency frontier: for domain-proximal scenes with = 200 samples, Swin-Graft surpasses YOLOv11x (terminal 703-shot: 0.973 vs 0.956 mAP@50); for domain-distant scenes, CNNs retain advantage (hook 141-shot: 0.900 vs 0.600 mAP@50). Grafted neck weights yield up to 2.5x the mAP of a randomly initialized neck.
[CV-75] π-SUB: A Physics-Informed Synthetic Underwater Benchmark Dataset for Underwater Image Enhancement
链接: https://arxiv.org/abs/2608.10589
作者: Namritha Lasyapriya Maddali,Rajini Makam,Suresh Sundaram,Narasimhan Sundararajan
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 13 pages
Abstract:This paper presents \pi -SUB, a physics-informed framework for generating synthetic underwater benchmark datasets that bridges the synthetic-to-real gap for Underwater Image Enhancement (UIE). The proposed framework extends the classical underwater image formation model by incorporating depth-dependent downwelling irradiance, biologically resolved absorption, and environmental scattering across all ten Jerlov water types, together with independently controllable residual phenomena. Using this framework, the \pi -SUB dataset consists of paired synthetic underwater-reference images spanning shallow-to-deep and coastal-to-oceanic environments. Extensive simulation studies have been carried out to evaluate \pi -SUB along two criteria namely hyper-realism and generalizability. For hyper-realism, \pi -SUB attains a global Frechet Inception Distance (FID) that is 46% lower than Syrea. For generalizability, four state-of-the-art UIE architectures (FUnIE-GAN, Pix2Pix, PUIE-Net, and Phaseformer) are used for comparative evaluation of \pi -SUB. These models were independently trained on six datasets including one real and five synthetic datasets and tested on six real-world benchmarks datasets. Across four UIE architectures and six real benchmark datasets, \pi -SUB improves UIQM by 4.18% over PHISWID (next best) and 9.46% over Syrea (next best), while reducing NIQE by 48.78% and 23.98%, respectively. These results establish \pi -SUB as a hyper-realistic and generalizable benchmark for developing the next generation of underwater image enhancement methods. The code and dataset are available at this https URL
[CV-76] A HamNoSys-Guided Dataset and Baselines for Fine-Grained Isolated Handshape Recognition in Sign Language
链接: https://arxiv.org/abs/2608.10588
作者: Ushnish Sarkar,Suvajit Patra,Bhaswar Chattopadhyay,Pranab Singha Roy,Tapas Samanta
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Purpose: Fine-grained handshape recognition supports computational sign-language transcription, recognition, and translation, but broad, phonetically defined visual inventories with signer-aware evaluation remain limited. This work introduces a benchmark grounded in the language-independent Hamburg Notation System (HamNoSys). Methods: A balanced dataset of 144,000 RGB images was collected from 15 participants for 160 handshape classes defined by the official HamNoSys 4 Handshapes Chart. ResNet-18 and ViT-B/16 were evaluated as appearance-based models, while a graph convolutional network and XGBoost were evaluated from hand landmarks. Both a class-stratified subject-dependent split and a 15-fold leave-one-subject-out (LOSO) protocol were used. The same model families were additionally assessed on LSWH100 and ASL Fingerspelling Dataset A for external context. Results: The subject-dependent benchmarks established reproducible reference performance across all four model families, whereas LOSO evaluation exposed a substantial reduction when recognition was required to generalise to unseen participants. On ASL Fingerspelling Dataset A, mean LOSO top-1 accuracy ranged from 82.20% to 87.40%. Conclusion: The documented acquisition, curation, and complementary evaluation protocols pro-vide a reproducible resource for fine-grained isolated-handshape research and for developing more accessible sign-language technologies.
[CV-77] Flow Straight to Reality: Perceptually Consistent Flow Matching for Efficient Image Restoration ECCV2026
链接: https://arxiv.org/abs/2608.10544
作者: Sangwoo Jo,Donggeun Ko,Jayeon Kang,Youngsang Kwak,Jaehwa Kwak,Sungjoon Choi
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted to ECCV 2026. Code is available at this https URL
Abstract:Image restoration is fundamentally constrained by the tradeoff between distortion and perception: minimizing pixel-wise error yields over-smoothed results, whereas optimizing for perceptual realism often introduces structural deviations. Recent approaches attempt to balance this tradeoff via posterior sampling or multi-stage generative pipelines, yet remain computationally expensive and architecturally complex. To overcome these limitations, we propose PCFlow (Perceptually Consistent Flow Matching), a unified framework that directly parameterizes a continuous transport from degraded observations to clean targets, jointly optimizing distortion and perceptual quality. While its latent consistency flow objective drives stable and efficient few-step inference, a Latent Consistency Perceptual Loss (LCPL) imposes semantic constraints directly on the guiding velocity field, steering the dynamics toward visually sharp data manifolds. Furthermore, recognizing the inherent conflict between structural and perceptual consistencies, we integrate a conflict-free gradient projection strategy to stabilize the multi-objective optimization landscape. Combined with lightweight, convolution-only backbone, PCFlow achieves competitive performance across diverse restoration tasks at a fraction of traditional computational costs.
[CV-78] Dynamic Context Adapters: Efficiently Infusing History into Vision-and-Language Models
链接: https://arxiv.org/abs/2608.10525
作者: Yuhang Song,Bor-Jiun Lin,Jiaxu Liu,Te-Chuan Chiu,Anh Nguyen,Chun-Yi Lee
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Historical context integration presents a fundamental challenge for Vision-Language Models (VLMs) in sequential decision-making tasks. Current VLMs process visual inputs independently, which creates critical limitations for downstream applications that require temporal understanding. Direct incorporation of historical frames into Transformer inputs produces quadratic attention complexity and excessive memory consumption. Existing approaches suffer from significant drawbacks: computational inflation or substantial information loss through temporal compression. To address these challenges, we introduce Dynamic Context Adapter (DCA), a novel context injection approach for pretrained VLMs. Our method employs fixed-size, dynamically compressed memory to preserve historical semantics without frame concatenation. DCA bridges static VLMs and recurrent policies and enables memory capabilities in pretrained models while maintaining computational efficiency. DCA achieves over 25% reduction in attention FLOPs and 13% memory savings while improving performance on long-horizon tasks.
[CV-79] Rethinking Text-Based Image Retrieval in Specific Domain
链接: https://arxiv.org/abs/2608.10524
作者: Jingyang Tan,Sheng Yang,Yuanpeng Chen,Jian Wang,Nianjin Ye,Chen Xing,Lanpeng Jia
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 13 pages
Abstract:Driven by the rapid advancement of vision-language representation learning, Text-based Image Retrieval (TBIR) has made notable progress. However, existing benchmarks are predominantly constructed on an exclusive single-match assumption between query and images. While effective in general scenarios, this assumption fails to reflect practical system performance in specific domains (e.g., surveillance), where a single query often corresponds to multiple relevant candidate images. To address this limitation, we design a Domain-Specific Multi-Match Text-based Image Retrieval (DSMM-TBIR) data engine. Leveraging this engine, we construct Security Multi-Match TBIR (SecMM-TBIR), a benchmark comprising 50k surveillance images with 200 comprehensive queries. Furthermore, we observe that vanilla contrastive learning in specific domains suffers from severe false negatives, forcing the model to push apart semantically similar pairs and thus degrading retrieval performance. We propose the Semantic-Aware Fine-Tuning (SAFT) framework to address semantic compression in specific domains, which incorporates Semantic-Aware Soft-Label Supervision (SASS) and Intra-modal Structural Distillation (ISD) to establish a promising paradigm for domain-specific TBIR tasks. Experiments across diverse CLIP-like models demonstrate that SAFT yields an average mAP@20 gain of 7.8 points on SecMM-TBIR over standard image-text contrastive (ITC) fine-tuning, while also improving general-domain performance. The entire benchmark will be released to facilitate further research.
[CV-80] Unlocking the Power of Medical Tabular Data via Semantic-Aware Multimodal Pre-training
链接: https://arxiv.org/abs/2608.10522
作者: Yingsheng Liu,Haiming Li,Jingmin Zhu,Jiajun Sun,Victoria Mar,Monika Janda,H. Peter Soyer,Zongyuan Ge,Zhen Yu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: INTERNATIONAL CONFERENCE ON MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION (ORAL presentation)
Abstract:While vision-language models dominate medical representation learning, unstructured text lacks the dense, quantitative diagnostic phenotypes inherent in structured clinical tables. However, existing multimodal pre-training methods underutilize this potential due to semantic-agnostic designs that treat tabular inputs as flat vectors and employ unstable continuous regression objectives. To overcome this, we propose a novel semantic-aware framework explicitly modeling the intrinsic two-dimensional structure of tabular data. First, addressing the inter-feature hierarchy of varying diagnostic importance, we introduce Importance-Aware Adaptive Masking to construct a label-free curriculum prioritizing salient features. Second, addressing the intra-feature continuity-discreteness duality, we propose a Soft-Label Discretized Module that replaces unstable numerical regression with stable distribution matching, thereby mathematically preserving ordinal relationships. Extensive experiments across large-scale dermatology (SLICE-3D, HOP) and ophthalmology (EyePACS) datasets establish a new state-of-the-art (SOTA), demonstrating exceptional robustness and cross-domain generalizability.
[CV-81] SparSTAR: Sparse Attention for SpaceTime AutoRegressive Video Synthesis
链接: https://arxiv.org/abs/2608.10519
作者: Jongbeom Lee,Hyunwoo Yu,Jincheol Yang,Jaemin Choi,Suk-Ju Kang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:InfinityStar extends visual autoregressive generation to video through a sequence of image and clip pyramids. Its changing scale and cross-clip context, however, leave late-scale attention costly and make sparse patterns reused from diffusion or image VAR models unreliable. We introduce SparSTAR, a training-free block-sparse attention method tailored to this setting. At each expensive scale and attention head, SparSTAR scores contiguous key blocks from the current query and key activations, retains required conditioning context, and executes the selected blocks through a forward-only sparse path. We analyze cross-scale consistency within a clip, pattern persistence across clip boundaries, and quality degradation as reuse spans increasingly distant scales. Across these analyses, important key blocks shift, showing that recomputing block selection at each target scale is more reliable than reusing a transferred mask. On 720p text-to-video and image-to-video generation, SparSTAR preserves every token and refinement scale while providing about a 1.6x end-to-end speedup and maintaining VBench and paired-output reconstruction fidelity close to dense InfinityStar.
[CV-82] SafeCap: Improving LVLM Safety with Image Captioning Reinforcement Learning
链接: https://arxiv.org/abs/2608.10513
作者: Caoyuan Ma,Wenpu Liu,Weichu Xie,Tian Gu,Shilei Zhao,Lingxi Min,Shuai Dong,Yuqi Xu,Ji Zhao,Ziyue Wang,Wenzheng Chang,Taiqiang Wu,Yongfu Zhu,Wenqi Shao,Yinqiang Zheng
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 16pages, 4 figures. Preprint. Project page: this https URL ; code: this https URL
Abstract:Large vision-language models (LVLMs) remain vulnerable to jailbreak attacks that exploit visual inputs to bypass safety alignment inherited from their language backbones. We propose SafeCap, a reinforcement-learning framework that aligns LVLMs through learned self-captioning. SafeCap trains a policy model to first generate a safety-relevant image caption and then produce a final answer; the caption is further optimized by whether it enables a frozen LLM to reach a safety-aligned decision. This caption-mediated objective encourages the policy to expose visual cues relevant to safe response generation rather than relying solely on direct refusal supervision. Across five multimodal safety benchmarks and six vision-utility benchmarks, SafeCap substantially improves aggregate safety performance under its intended DirectCap protocol, with gains of 3.7-19.0 points in safety average across four model settings while maintaining comparable or improved vision utility. Under controlled comparisons on matched backbones and data, SafeCap outperforms safety SFT, DPO, and SafeGRPO, demonstrating the effectiveness of caption-mediated reinforcement learning for multimodal safety alignment.
[CV-83] owards Color-Faithful Low-Light Image Enhancement via Adaptive Color Debiasing and Saturation Rectification
链接: https://arxiv.org/abs/2608.10512
作者: Zhichen Yang,Rui Xu,Yuzhen Niu,Fusheng Li,Hui Da,Ri Cheng
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by ACMMMM 2026
Abstract:Low-light imaging often introduces color bias caused by the low signal-to-noise ratio and the image formation process. Although recent low-light image enhancement methods have achieved strong brightness recovery, faithful color restoration remains challenging, manifesting as overall color bias together with local under- and over-saturation. To address this issue, we propose CAGE, a cylindrical color correction framework with adaptive color debiasing and gamut-harmonized saturation rectification for color-faithful low-light image enhancement. We first introduce AdaLAB, a cylindrical adaptive LAB color space that provides a decoupled and image-specific basis for uniform color correction. Building on this color space, we further develop AdaCCT, an adaptive cylindrical color transform with forward and inverse transforms for the conversion between RGB and AdaLAB color space, as well as necessary color debiasing and saturation rectification. The forward transform suppresses embedded color bias before backbone enhancement by reorganizing the chromatic distribution through chromatic-plane shifting and scaling, while the inverse transform achieves faithful saturation rectification through out-of-gamut lightness compensation. Extensive experiments on multiple benchmarks show that CAGE achieves more faithful color restoration, specifically reduces color bias and saturation abnormality, and delivers better overall visual quality across different low-light enhancement backbones. The code is available at this https URL.
[CV-84] DSAR: Dual-Stream Autoregressive Modeling of Temporal Cloth Dynamics for Photorealistic Animatable Avatars ECCV2026
链接: https://arxiv.org/abs/2608.10500
作者: Haozhong Xiong,Yao Yu,Yu Zhou,Sidan Du
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ECCV 2026. 25 pages, 7 figures, including supplementary material
Abstract:Creating photorealistic and temporally coherent animatable human avatars from RGB videos remains challenging. Current methods struggle to capture realistic cloth dynamics, producing over-smoothed appearance or severe artifacts on out-of-distribution poses. This limitation stems from a fundamental oversight: existing approaches neglect the temporal causality inherent in cloth physics, where current states emerge from previous states through temporal evolution rather than instantaneous skeletal configurations alone. Without explicit modeling of this causal structure, networks learn pose-appearance correlations instead of motion evolution, leading to poor generalization. We introduce a dual-stream autoregressive framework that explicitly models both observable geometric information and implicit internal state. The geometric stream propagates surface displacement from the previous frame, while the state stream fuses current features with historical states retrieved from a memory bank. Motion-adaptive aggregation handles spatially-varying dynamics, and adaptive regularization balances smoothness with flexibility. Experiments on challenging datasets demonstrate significant improvements in rendering quality, temporal consistency, and generalization to motion patterns beyond training distributions, validating that dual-stream temporal modeling enables realistic cloth dynamics.
[CV-85] SapiensID 2.0: Aligning Human Recognition Foundation Models with Human Perception
链接: https://arxiv.org/abs/2608.10497
作者: Yiyang Su,Jie Zhu,Feng Liu,Anil K. Jain,Xiaoming Liu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:While foundation models have significantly advanced human recognition across diverse modalities, they predominantly rely on static, geometric feature extraction. This approach fundamentally diverges from human perception. Consequently, current models often suffer from “semantic blindness,” overfitting to transient noise while failing to leverage invariant soft biometrics, and struggle to capture temporal motion signatures. To bridge this gap, we propose SapiensID 2.0, a human recognition framework enriched with both semantic and temporal awareness. To overcome the lack of soft-biometric annotations, we transfer zero-shot semantic knowledge from Multimodal Large Language Models (MLLMs) into a discriminative embedding space. We resolve the dimensional mismatch between these spaces using Invariant Trait Alignment (ITA) to distill core persistent traits, and Transient Noise Disentanglement (TND) to decouple artifacts like clothing. Furthermore, we design a Kinematic Semantic Attention Head (K-SAH) that extends spatial attention across temporal windows. By tracking semantic patches over time, K-SAH captures rich kinematic signatures without requiring large-scale video datasets. Extensive experiments demonstrate that SapiensID 2.0 achieves state-of-the-art performance across image- and video-based person re-identification and gait recognition, while maintaining robust face recognition capabilities.
[CV-86] When Vision Becomes Text: Visual Token Pruning via Cross-Modal Residual Guidance in VLMs
链接: https://arxiv.org/abs/2608.10489
作者: Congyang Ou,Ruike Song,Yang Zhou,Libo Sun,Haokui Zhang,Zhenbo Luo
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Abundant visual information strengthens vision-language model (VLM) perception, yet massive visual tokens raise inference costs. Existing visual token pruning methods rely on similarity-based guidance, which exploits pairwise text-vision and vision-vision token correlations for compression. However, such methods only capture local layer-level signals and overlook the whole inference process in VLM. In this paper, we revisit VLM inference and present a new efficient guidance scheme that complements similarity-based guidance. In particular, we identify a key observation: as LLM layers deepen, text tokens continuously aggregate visual information via self-attention and progressively absorb partial visual content into textual representations. To quantify this phenomenon, we propose Cross Modal Absorption (CMA) from a geometric representation perspective to measure how much visual information is absorbed by text, revealing that more visual tokens in deeper layers can be approximately explained by the text subspace. We accordingly propose Cross Modal Residual (CMR). It projects visual tokens onto the text subspace via Tikhonov regularized least squares and exploits reconstruction residuals to quantify visual information that cannot be explained by text. Finally, based on CMR, we present SIEVE, a training-free visual token compression method that combines CMR, text-attention relevance, and residual-space diversity to retain task-relevant and complementary tokens. Experiments on diverse VLM architectures verify the effectiveness of SIEVE. For instance, on LLaVA-NeXT-7B, SIEVE keeps only 11.1% of visual tokens while preserving 97.5% of the original average performance, achieving 3.62\times prefill speedup, 2.49\times end-to-end speedup, and a 6.02\times KV-cache reduction.
[CV-87] Bridging Event Streams and DiT: Event-Guided Video Frame Interpolation
链接: https://arxiv.org/abs/2608.10479
作者: Guixu Lin,Yuyang Yu,Xiang Ji,Linyao Chen,Zhengwei Yin,Mengshun Hu,Mingdeng Cao,Shengfeng He,Yinqiang Zheng
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: this https URL
Abstract:Latent diffusion models have recently advanced video frame interpolation by synthesizing intermediate frames between input images. However, handling large temporal gaps and complex motion remains challenging, often resulting in motion blur, structural distortions, and temporal inconsistencies. Event cameras provide high-temporal-resolution motion cues that are well suited for bridging these gaps and improving interpolation quality. To exploit this advantage without training an event-assisted model from scratch, we propose an adapter-based framework that incorporates event-derived cues into a pre-trained image-to-video diffusion model with minimal architectural changes. Specifically, our method leverages Image Warped Events (IWEs) and bidirectional sparse optical flow to provide spatially and temporally aligned guidance during generation. By injecting these event-guided structural and motion cues into the diffusion process, our approach reduces interpolation artifacts and improves both reconstruction fidelity and temporal coherence. Experimental results on real and synthetic benchmarks show that our method consistently outperforms existing state-of-the-art approaches. The project page is at this https URL.
[CV-88] FUSE: Frame-Unified Stress Estimation from Facial Video
链接: https://arxiv.org/abs/2608.10442
作者: Stefanos Gkikas,Thomas Kassiotis,Yang Guo,Guangliang Li,Giorgos Giannakakis
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: The paper has been accepted at: IEEE | 2026 9th International Conference on Pattern Recognition and Artificial Intelligence (PRAI 2026)
Abstract:Automatic stress detection from facial video offers a practical path to non-intrusive affect monitoring, yet existing video-based approaches commonly decompose full recordings into short temporal windows before classification. This design introduces additional choices regarding window length, overlap, and aggregation, while limiting direct analysis of temporal information across the entire recording. In this study, we present FUSE (Frame-Unified Stress Estimation), a facial-video stress detection framework that processes complete recordings as a single input without temporal windowing or external segmentation. The name reflects the defining operation of the method: rather than dividing a recording into short clips, all frames are fused into one unified two-dimensional representation from which the stress state is estimated. This unification is realized by folding the temporal dimension into the channel dimension of the spatial representation, and the resulting high-dimensional input is processed using a unified asymmetric-attention architecture. At a temporal stride of t = 1, FUSE retains the full 120-second recording as one input, corresponding to 3,600 frames at 30 fps. Experiments on a 58-subject stress dataset using a stratified subject-level protocol evaluate seven temporal-stride configurations, ranging from full-frame input to sparse subsampling. FUSE achieves the highest test accuracy of 69.44% at t = 15, while the full-frame configuration remains competitive at 69.03%. Across the stride range, computational cost varies from 12.48 to 348.78 GFLOPs, showing the trade-off between temporal density and efficiency. These results demonstrate that temporal windowing is not required for effective facial-video stress detection in this setting, and that complete-recording inference can be achieved within a single unified architecture.
[CV-89] Stream Forcing: Constructing Unified Training Trajectory for Robust Streaming Video Generation
链接: https://arxiv.org/abs/2608.10439
作者: Yueting Zhu,Yuehao Song,Kaicheng Zhang,Bao Tang,Shaoyu Chen,Qian Zhang,Wenyu Liu,Xinggang Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 15 pages,6 figures
Abstract:Streaming video generation holds strong potential for world modeling, where future frames must be inferred online sequentially to form a continuous video stream. However, streaming video diffusion models introduce a fundamental train-inference mismatch: inference follows a specialized denoising order, whereas advanced training strategies typically require diverse noise-level configurations. To address this trade-off between train-inference consistency and training coverage, we reformulate the video diffusion sampling as a frame-indexed stochastic process over noise levels. Within this stochastic process space, we construct a continuous training trajectory along which the sampling schedule progressively evolves from independent sampling to inference-consistent sampling. We further introduce a joint calibration algorithm and a temporal correlative sampling algorithm to ensure trajectory smoothness and cross-frame correlation. Building on these designs, we propose Stream Forcing, a unified training framework for streaming video generation that balances training sufficiency and inference efficiency. Extensive experiments demonstrate that Stream Forcing significantly improves generation quality with a 36.6% FVD improvement on the UCF-101 benchmark. Furthermore, our method facilitates robust zero-shot extrapolation to long-horizon video generation with a 27.9% FVD improvement on the UCF-101 benchmark.
[CV-90] MammoMix: Leverag ing Mixture of Experts for Robust Mammogram Breast Detection
链接: https://arxiv.org/abs/2608.10437
作者: Dinh Tan Nguyen,Hoang Quan Dang,Chen Zhang,Sai Ho Ling
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Australasian Joint Conference on Artificial Intelligence 2025
Abstract:Breast lesion detection in mammography remains a challenging task due to variations in image quality, lesion appearance, and population demographics across datasets. While current object detectors such as YOLO and DETR achieve strong results on individual datasets, their performance often degrades when trained on or applied across heterogeneous sources. To address this, we propose MammoMix, a novel framework based on Mixture-of-Experts (MoE) paradigm for robust and generalizable lesion detection. In MammoMix, each expert model is trained on a specific domain, allowing it to specialize in distinct characteristics of its source data. A gating mechanism adaptively weighs contributions from each expert based on input image, combining their outputs to enable domain-adaptive inference. To improve reliability, we further incorporate a calibration module, MoCAE, which adjusts confidence scores to reflect true predictive uncertainty. We evaluate MammoMix on 3 public mammography datasets: CSAW, DDSM, and DMID, covering diverse clinical settings. Results show that MammoMix outperforms baseline detectors in both average precision and reliability, particularly on datasets with greater variability. Our findings demonstrate that expert specialization and calibrated ensemble fusion significantly enhance model generalization and robustness. MammoMix offers a promising step toward dependable AI-assisted breast cancer screening across real-world clinical domains.
[CV-91] DynaPPI: A Large-scale Dynamic Protein Dataset for AI-driven Advances in Protein Interactomics NEURIPS2025
链接: https://arxiv.org/abs/2608.10435
作者: Jiabao Wei,Zilong Geng,Yuze Wang,Jianjun Li,Ning Ding,Bowen Zhou,Bing Zhang,Zhiyuan Ma
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages, 1 figure, 39th Conference on Neural Information Processing Systems (NeurIPS 2025) Workshop: AI4Science
Abstract:Diffusion models have been widely explored in protein backbone generation due to their powerful generation this http URL, in today’s AI-driven biological research, predicting the structure of unknown multi-chain protein aggregates (called “complexes” in biology) remains an unsolved this http URL is because existing static or dynamic protein datasets focus solely on static snapshots or single-entity trajectories, neglecting the dynamic process of multiple monomers forming this http URL alleviate this dilemma, we present DynaPPI, a dynamic protein dataset comprising molecular dynamics (MD) trajectories of protein complex formation from dissociated chains to the bound state, as a pivotal resource to bridge the gap between static structural biology and the inherently temporal nature of dynamic molecular this http URL from this dataset, diffusion models can explicitly learn the dynamic binding trajectories of known complexes and accurately predict the structures of unknown complexes based on their diverse generative properties, thereby further catalyzing AI-driven structural biology and protein interactomics.
[CV-92] Lesion-Aware Adaptive Fourier Neural Operator for CT-to-PSMA PET Synthesis in Prostate Cancer
链接: https://arxiv.org/abs/2608.10429
作者: Rashmi Bhaskara,Waleed M. Almutairi,Matthew Gopaulchan,Maram Musaad Alqurashi,Francis Asamoah,Alex Ocana,Clinton D. Bahler,Oluwaseyi M. Oderinde
类目: Computer Vision and Pattern Recognition (cs.CV); Medical Physics (physics.med-ph)
备注:
Abstract:Deep learning models that synthesize PET from CT or MRI can reduce patient dose and scanner demand, but are typically optimized with global losses such as L1 or mean squared error (MSE) that treat all voxels similarly. In whole-body PSMA-PET, tumor voxels occupy only a small fraction of the volume, yet carry the clinically relevant activity signal; as a result, models can achieve high structural similarity index measure (SSIM) and peak signal-to-noise ratio (PSNR) while still underestimating lesion activity or failing to preserve tumor-specific structure. Radiomics provides biologically meaningful descriptors of tumor intensity and texture, but direct radiomics conditioning is time-consuming because it requires feature extraction from delineated lesion regions. We propose LAFNO, a Lesion-Aware Adaptive Fourier Neural Operator for CT-to-PSMA-PET synthesis that replaces high-dimensional radiomics conditioning with two efficient CT-derived proxy channels. Motivated by radiomics analysis of PSMA-avid tumor core and peritumoral regions, LAFNO uses a contrast proxy for local density variation and a disorder proxy for local texture heterogeneity, both injected into the model bottleneck. LAFNO combines whole-volume reconstruction with lesion-level total lesion activity (TLA), tumor-core contrast, and peritumoral supervision. We evaluated LAFNO against four baseline architectures on the TCIA PSMA-PET-CT-Lesions dataset. LAFNO remained competitive on whole-volume image quality, achieving SSIM of 0.960 and 0.938 for 18F- and 68Ga-PSMA, respectively, while reducing per-patient TLA error to 48.3% and 64.0% for 18F- and 68Ga-PSMA, respectively, and achieving the highest tumor-core radiomics reproducibility across all feature classes for both tracers. Peritumoral reproducibility remained tracer-dependent, indicating that biological fidelity in synthetic PSMA-PET remains challenging.
[CV-93] GeoSeg-OV: Bridging Geospatial Gaps with Structural Guidance for Open-Vocabulary Remote Sensing Segmentation
链接: https://arxiv.org/abs/2608.10426
作者: Ruizhong Liu,Tingzhang Luo,Zaiyan Zhang,Jundong Chen,Hongruixuan Chen,Shaoguang Huang,Hongyan Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Code and benchmark: this https URL
Abstract:Open-vocabulary remote sensing segmentation has recently emerged as a promising paradigm that enables pixel-level recognition of arbitrary categories specified by natural language, including classes unseen during training. However, geospatial domain shifts caused by heterogeneous regions, spatial resolutions, and acquisition platforms weaken visual-text matching and limit cross-dataset generalization. Recent attempts have begun to incorporate auxiliary vision foundation models (VFMs), typically coupling their features with text embeddings as additional matching evidence. However, this strategy may introduce inconsistent matching signals while leaving the structure-sensitive representations of VFMs insufficiently exploited. We therefore propose GeoSeg-OV, which decouples auxiliary VFM features from visual-text matching and repurposes them as structural guidance for cost aggregation and decoding. GeoSeg-OV constructs an orientation-robust cost volume from multi-rotation CLIP features, while a frozen VFM extracts multi-scale structure-sensitive features in parallel. We propose Structure-Guided Aggregation (SGA), which integrates cost tokens and CLIP semantic guidance with VFM-derived pairwise structural biases for coherent spatial propagation, followed by text-conditioned class-wise reasoning. We further introduce Cost-Aware Decoding (CAD) to adaptively refine and fuse multi-scale semantic and structural guidance based on the current decoder context. On the global High-Resolution Land Cover (HRLC) benchmark spanning seven datasets across six continents, GeoSeg-OV outperforms the state-of-the-art by +2.5 and +2.7 average mIoU under two training settings. A large-scale zero-shot case study further demonstrates its generalization across geographic domains and category systems without target-domain annotations or retraining.
[CV-94] DriveVLA-M0: Failure-Aware Memory Augmentation for Autonomous Driving
链接: https://arxiv.org/abs/2608.10413
作者: Zebin Xing,Yupeng Zheng,Qiang Chen,Linbo Wang,Yichen Zhang,Pengxuan Yang,Junli Wang,Deheng Qian,Xiaoqing Ye,Junyu Han,Yifeng Pan,Qichao Zhang,Dongbin Zhao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Vision-Language-Action (VLA) models have recently emerged as a promising paradigm for end-to-end autonomous driving by enabling unified reasoning across perception, language, and planning. However, existing approaches lack mechanisms to exploit past failures or adapt to distribution shifts, causing the model to persistently underperform on similar scenarios where it has previously failed. In this paper, we propose DriveVLA-M0, a retrieval-augmented VLA with failure-aware latent memory. We construct a latent memory pool that stores failure cases along with their structure scene representations and expert trajectory labels, and design a dedicated Retrieve Model that decouples static road structure and dynamic agent interactions to enable structurally grounded retrieval. At inference time, retrieved cases are injected into the model via a lightweight decoupled LoRA-based test-time training (TTT) mechanism, allowing targeted and scenario-specific correction without modifying the backbone. Extensive experiments on NAVSIMv1 and NAVSIMv2 benchmark demonstrate that our approach consistently outperforms prior methods, achieving 94.1 PDMS on Navtest and 47.0 EPDMS on Navhard with only 26.44 ms TTT backward latency overhead. Furthermore, we show that DriveVLA-M0 scales effectively with additional memory, enabling training-free performance gains through memory expansion. The code is available at this https URL.
[CV-95] A second-order theory of texture for depth from focus ECCV2026
链接: https://arxiv.org/abs/2608.10411
作者: Sreekar Ranganathan,Ioannis Gkioulekas
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: ECCV 2026, project website: this https URL
Abstract:We present a theory of textured appearance of optically rough surfaces based on wave optics, emphasizing the role of texture for passive depth from focus. Our theory shows that even surfaces that traditional computer vision would consider textureless can produce textured appearance, due to subjective speckle from surface microgeometry. We analyze the properties of this second-order texture, and show that we can enhance its contrast under natural ambient lighting by simply using a narrowband spectral filter. Doing so results in dramatic improvements in passive depth reconstruction of seemingly textureless scenes, as we demonstrate through extensive theory, simulations, and real-world experiments.
[CV-96] FormStruct-Bench:A Hierarchical and Diagnostic Benchmark for Table-Form Document Structure Recognition
链接: https://arxiv.org/abs/2608.10396
作者: Lujie Ban,Jiangtao Zhu,Yuanheng Yu,Jiasheng Shi,Chenhao Ma
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Transforming table-form documents into machine-processable records requires recovering not only their visible content but also the multilevel structure that organizes it. However, existing benchmarks evaluate either holistic document outputs or conventional table grids, and their aggregate scores provide little insight into where structural failures occur. We introduce FormStruct-Bench, a hierarchical and diagnostic benchmark that evaluates table-form document structure recognition at both the document level and progressively finer component levels, allowing aggregate performance to be traced back to specific structural failure modes. To construct auditable ground truth at scale, we annotate 70 reusable templates and expand them into 7,000 verified instances through a provenance-preserving Director–Artist–Verifier pipeline; all 1,100 instances in the template-disjoint test set additionally receive human review. Our evaluation protocol uses five primary metrics and three structure-specific diagnostics across page, schema, and component levels, together with slices over difficulty, structural constraints, and visual degradation. Across 14 API-hosted and locally deployable systems plus two SFT variants, the best document-level score reaches 83.85%, whereas the best reported fine-grained structural score remains below 18%. These results reveal a pronounced gap between reading document content and recovering the hierarchy and regional organization required for reliable table-form understanding.
[CV-97] owards Unified Dynamic Face Landmark Detection
链接: https://arxiv.org/abs/2608.10346
作者: Sebastian Regalado,Varshanth R. Rao,Ruowei Jiang,Parham Aarabi,Igor Gilitschenski
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 9 pages, 6 figures in Main Paper. 13 pages, 3 figures in Appendix
Abstract:Although advancements in face landmark detection (FLD) methods continue to push performance boundaries, they overlook two major functional limitations: (1) different network parameters need to be trained independently for each N -point'' benchmark dataset, and (2) a model trained on an N -point’’ dataset reliably outputs only the N landmarks. In our work, we first conceptualize Face Part-Anchored Landmark Positions (FPALPs), wherein each landmark is treated as a progression value between zero (start) and one (end) along a face part’s contour. Every landmark can be expressed in the FPALP format, irrespective of its source dataset, hence unlocking the ability to unify all N -point'' datasets into a single dataset. Secondly, we represent each landmark with an FPALP-based query, refine it progressively with a cross-modality decoder, and predict its coordinates based on the final representation. Our approach, called Unified Dynamic FLD, embodies these two design choices and streamlines the landmark detection pipeline by enabling (1) a single model to learn on any number of N -point’’ datasets, and (2) yield any number of specific landmark predictions by loading the designated landmark queries at runtime. Extensive experiments on multiple benchmark datasets show that our method delivers these benefits while remaining competitive with, and in several cases outperforming existing state-of-the-art methods.
[CV-98] CasDeblurGS: Cascaded 2D-to-3D Multi-View Consistency for 3D Gaussian Splatting from Two Blurry Images ECCV2026
链接: https://arxiv.org/abs/2608.10345
作者: Haeyun Choi,Minhyuk Jang,I-Gil Kim
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to the ECCV 2026 MUSTCV Workshop. Project page: this https URL
Abstract:Free-viewpoint 3D scene media is increasingly important for immersive applications, yet practical capture often suffers from severe view sparsity and motion blur. Although neural rendering has advanced sparse-view synthesis, existing blur-aware methods typically require substantial multi-view redundancy, accurate camera poses, or costly per-scene optimization. We address a stringent yet practical setting: reconstructing a coherent 3D scene from only two motion-blurred images with known intrinsics, without input-view poses, auxiliary sharp images, or per-scene test-time optimization. To this end, we propose CasDeblurGS, a cascaded framework that progressively recovers reliable cross-view information from local 2D correspondences to global 3D guidance. Stage 1 constructs locally reliable guidance through occlusion-aware correspondence filtering, while Stage 2 aggregates the intermediate restorations into a provisional pose-free 3D Gaussian representation whose input-view re-renders provide dense global guidance for final restoration. The resulting views enable a more coherent 3D representation and higher-quality novel-view synthesis. Experiments on real-world and synthetic Deblur-NeRF scenes show consistent gains over strong baselines, improving PSNR by 1.19 dB and 2.11 dB, respectively. Progressive ablations, cross-view correspondence visualization, and camera reprojection analysis further demonstrate improvements in both rendering quality and multi-view geometric consistency.
[CV-99] ENCORE: Efficient Noise Context-Aware Representation for Low-Dose CT Denoising
链接: https://arxiv.org/abs/2608.10343
作者: Minwoo Yu,N. Robert Bennett,Jongduk Baek,Adam S. Wang
类目: Computer Vision and Pattern Recognition (cs.CV); Medical Physics (physics.med-ph)
备注:
Abstract:While deep learning-based denoising has become widely adopted in low-dose CT, conventional models use generic architectures designed for natural images, failing to account for non-stationary and spatially correlated CT noise characteristics. To address this, we propose an Efficient Noise COntext-aware REpresentation (ENCORE) framework that explicitly leverages CT noise characteristics and anatomical features. First, we reformulate the noise synthesis procedure based on a realistic noise distribution beyond the conventional Gaussian approximation, establishing a rigorous foundation for training pair generation. Next, we extract local noise power and correlation contexts to guide the denoising process. To fully leverage the potential of noise context, we propose a FlyingConv module, which adaptively changes convolution weights for each local image region. Notably, our approach demonstrates substantial gains in both denoising quality and computational efficiency. Furthermore, manipulating the intensity of the noise context maps at inference time enables zero-shot conditional denoising, allowing for dynamic control over the output image texture. The entire pipeline is available at this https URL
[CV-100] From Detection to Understanding: TAR and TAR-Bench for Multi-Task Traffic Anomaly Reasoning
链接: https://arxiv.org/abs/2608.10317
作者: Han Zhang,Yilin Zhao,Zaid Pervaiz Bhat,Zheng Tang,Varun Praveen,Vidya N. Murali,David C. Anastasiu,Tomasz Kornuta
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:We present TAR (Traffic Anomaly Reasoning) and TAR-Bench datasets, resources for training and evaluating video-language models beyond anomaly detection. TAR contains 44,040 chain-of-thought training annotations across 10 tasks for 3,670 CCTV videos ( \sim 26 hours) from eight public datasets. Its evaluation component, TAR-Bench, contains 960 human-curated test annotations for 80 held-out clips trimmed from 17 public YouTube videos. TAR’s training annotations are produced with MAVEN, which consolidates multi-scale video evidence into structured event descriptions before generating question-answer pairs and reasoning traces. On TAR-Bench, eleven vision-language models reveal that strong question-answering accuracy does not reliably predict temporal or scene reasoning ability. Multi-task fine-tuning on TAR yields consistent gains, with the full 10-task model improving aggregate score by 21.4 points over its zero-shot baseline. TAR and TAR-Bench provide the official training and in-domain evaluation data for AI City Challenge 2026 Track 3. The dataset is available at this https URL
[CV-101] UniMod: Enhancing Multi-Modal Medical Diagnosis through Cross-Modality and Within-Modality Alignment
链接: https://arxiv.org/abs/2608.10316
作者: Zijian Gu,Weikai Lin,Shuang Zhou,Zihan Chen,Song Wang
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Multimedia (cs.MM)
备注: Accepted to ACM Multimedia 2026 (MM '26). 10 pages, 7 figures, 5 tables. Code: this https URL
Abstract:Multi-modal learning combining medical images and clinical text is promising for disease diagnosis. However, standard multi-modal training leads to shortcut learning: models exploit the easier modality (e.g., diagnostic cues in text) while neglecting harder-to-learn features (e.g., subtle visual patterns). We propose UniMod, a framework that mitigates shortcut learning by requiring each modality to predict the diagnosis on its own. It supervises image-only, text-only, and multi-modal classification simultaneously, so each modality must extract diagnostic features. We add cross-modality alignment for knowledge transfer and within-modality supervised contrastive alignment over same-diagnosis patients. On Harvard-Glaucoma, UniMod reaches 0.850 AUC, outperforming OGM-GE and Gradient Blending by 1.6-1.8%; on CheXpert Plus, it reaches 0.966 AUC, surpassing them by over 5%. UniMod also extends to 5-class multi-label diagnosis without architectural change, improving mean AUC by 0.097 over CGGM.
[CV-102] Frozen Brain-MRI Foundation Models Are Site Fingerprints
链接: https://arxiv.org/abs/2608.10295
作者: Saman Rahbar
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 15 pages, 5 figures, 7 tables. Code: this https URL
Abstract:Frozen foundation-model (FM) embeddings are increasingly used as off-the-shelf brain-MRI representations, on the assumption that they capture anatomy. We audit what they actually encode and find that acquisition site is a large, intrinsic component of the representation. Across two independent cohorts (ABIDE-I, ABIDE-II), three frozen 3-D encoders (brain-pretrained, CT-pretrained, and randomly initialized), and every network depth, site is linearly decodable at roughly 0.9 balanced accuracy at deep layers, exceeding the decodability of every clinical or demographic variable (sex, age, autism diagnosis) at every layer. The effect is intrinsic rather than learned: a randomly initialized encoder is already a ~0.9 site classifier on both cohorts and across three architecture families (Swin, ViT, ResNet), and site is decodable at ~0.95 directly from the raw downsampled image with no encoder, so the fingerprint reflects low-level image statistics that any encoder preserves rather than a product of pretraining. Residualizing measured population covariates leaves site decodability essentially unchanged, indicating an acquisition- rather than population-driven effect. A nonlinear probe matches the linear one, so the fingerprint is fully linearly accessible. The site subspace is removable post hoc by iterative null-space projection or ComBat (site decodability 0.94 - 0.07/0.00), and is a site-attribution concern for shared or federated embeddings; but for dense segmentation this removal is not free, because site and anatomy occupy an entangled linear subspace (a matched-rank random-direction projection is Dice-neutral, whereas removing the site subspace is destructive). We recommend site-audited use of frozen brain-MRI FMs and release an open audit toolkit.
[CV-103] MRIComp4Flow: Compression of 3D Brain MRI for Training Multi-Modal Generative Models MICCAI2026
链接: https://arxiv.org/abs/2608.10291
作者: Lisa K. Fischer,Mykhailo Riabets,Daniel Rueckert,Benedikt Wiestler,Anke Meyer-Baese,Sandeep Nagar
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted: MICCAI 2026 SASHIMI workshop
Abstract:Large-scale multi-modal MRI datasets impose substantial storage and I/O costs, limiting the training of 3D generative models on commodity infrastructure. While lossy compression is known to preserve accuracy for discriminative segmentation networks, its effect on generative models, which must learn the full data distribution rather than a decision boundary, is unexplored. We study whether standard image codecs can effectively compress semantically rich brain tumor MRI while preserving the fidelity required to train and deploy a 3D MRI generative model. Each 3D volume is compressed with JPEG2000 or a near-lossless JPEG-LS pipeline. Next, a Wavelet Flow Matching model, conditioned on BraTS image sequences (T1n, T1c, T2, T2f), is trained on compressed data, and the resulting models are evaluated on the validation set. At a 20:1 compression ratio, synthesis quality is statistically equivalent to a model trained on uncompressed data within a pre-specified margin ( \Delta PSNR 1 ,dB, \Delta SSIM 0.02 ; paired TOST p=[[p]] ): mean PSNR is 27.3,dB vs. 27.0,dB and mean SSIM is 0.95 vs. 0.96 across modalities. Our results indicate that JPEG2000 compression is a practical step toward scalable 3D MRI generative modeling without degrading synthesis quality. The codebase is available at this https URL .
[CV-104] SeFaR: Semantic Feature-aware Robustness Testing of Deep Neural Networks
链接: https://arxiv.org/abs/2608.10289
作者: Nusrat Jahan Mozumder,Divya Gopinath,Corina Pasareanu,Matthew Dwyer
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Deep neural networks are increasingly deployed in safety-critical domains as perception modules, where failures are often caused due to rare and under-represented scenarios. This necessitates the need to evaluate the semantic robustness of perception models; conformance of behavior to high-level requirements over real-world perceptual variability. To address this, we propose SeFaR, a framework for systematic semantic-feature-centric testing of vision models. Given a natural-language requirement and a set of satisfying inputs, SeFaR evaluates robustness with respect to diverse realistic semantic variations that preserve requirement satisfaction. The approach employs a novel hierarchical concept model enabling structured exploration of the feature space and incorporation of domain knowledge via user-defined concepts. State-of-the-art diffusion and vision-language models are leveraged to generate photorealistic semantics-preserving perturbations and identification of previously unknown features impacting behavior. A feedback-driven adaptive process is adopted to generate interpretable failure-inducing semantic concepts along with corresponding test inputs. Evaluation on case studies demonstrates that the proposed framework effectively satisfies requirement preconditions while identifying requirement-independent features that influence model decisions, enabling it to both uncover faults and relate them to such features.
[CV-105] RACE-GS: On-Policy Trajectory Distillation with Privileged Geometric Conditioning for Sparse-View 3DGS Restoration
链接: https://arxiv.org/abs/2608.10286
作者: Linlian Jiang,Yuchen Xi,Sadman Rakib Pinon,Ruigang Yang,Yang Wang,Xinxin Zuo
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:We present TRACE-GS, an on-policy trajectory distillation framework that leverages privileged geometric conditioning at training time, thereby adapting a diffusion prior to sparse-view 3D Gaussian Splatting (3DGS) restoration. Rather than pursuing increasingly sophisticated restoration architectures, we identify a more fundamental limitation shared by existing diffusion-based approaches: supervision at independently noised states does not cover those reached during inference. In sparse-view 3DGS, under-constrained geometry biases denoising from the outset, and the resulting deviations compound along the rollout. TRACE-GS instead performs on-policy trajectory distillation: a teacher conditioned on richer geometry from additional training views supplies targets along the sparse-view student’s own rollout, aligning denoising directions and cross-view responses at each visited state. This training-only geometry places TRACE-GS in the learning using privileged information (LUPI) setting. At deployment, only the sparse-view student is retained, and its restored renderings serve as pseudo-observations for 3DGS refinement. To the best of our knowledge, TRACE-GS is the first to derive on-policy supervision from privileged geometry for sparse-view 3DGS restoration, achieving consistent gains and strong generalization across datasets and sparse-view settings.
[CV-106] Chain of Spatial Thoughts: Modality-Agnostic Spatial Grounding for Vision Language Models
链接: https://arxiv.org/abs/2608.10278
作者: Hunter Schofield,Mohammed Elmahgiubi,Mohammad Mahdavian,Richard Shi,Jinjun Shan,Amir Rasouli,Dongfeng Bai
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Spatial understanding is fundamental to embodied intelligence, underpinning applications such as robotic manipulation, embodied navigation, and autonomous driving. Although recent vision-language models (VLMs) have achieved impressive performance on spatial reasoning benchmarks, state-of-the-art approaches typically rely on additional spatial encoders or architectural modifications during inference, increasing computational cost. We introduce Space Tokens, a lightweight, architecture-agnostic framework that equips VLMs with explicit continuous spatial representations without requiring additional inference-time modules. By distilling scene-level 3D geometry and object-centric spatial attributes into continuous latent tokens, our method enables these modalities to be directly incorporated into a chain-of-thought reasoning process, thereby improving the VLM’s spatial reasoning capabilities. At the same time, the learned representations can be explicitly decoded to verify that they encode meaningful geometric information, while the unified token interface remains extensible to additional modalities. Experiments on VSI-Bench improve Qwen3-VL-8B by 4.3% and SenseNova-SI-1.3 by 1.3%, while achieving state-of-the-art performance on object size (79.2%) and room size estimation (75.7%). These results demonstrate that continuous spatial tokens provide an effective, interpretable, and computationally efficient mechanism for integrating geometric reasoning into large vision-language models.
[CV-107] Beyond Decision Boundaries: Relational Geometry Attacks on Contrastive Embedding Manifolds
链接: https://arxiv.org/abs/2608.10237
作者: Fei Zhao,Peiyuan Zhang,Xi Li,Chengcui Zhang,Nitesh Saxena
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Contrastive learning and Siamese embedding models have become the foundation of modern verification systems, where decisions are governed not by discrete classification boundaries, but by relational geometry in embedding space. However, existing adversarial attacks remain fundamentally classification-centric, overlooking the vulnerability of relational geometry. In this paper, we introduce a geometry-aware adversarial attack framework that reformulates attacks on contrastive systems as manifold-level relational corruption. Instead of targeting individual predictions, the proposed framework systematically distorts similarity organization within the embedding manifold by pushing positive pairs apart while simultaneously pulling negative pairs closer, ultimately collapsing and inverting pairwise similarity structure. To enable scalable deployment, we shift iterative online optimization into an offline adversarial geometry deformation prior learning stage and train a lightweight feed-forward generator that learns generalized geometry deformation patterns from the victim model. Once trained, the generator produces adversarial perturbations through a single forward pass without requiring online gradient computation, enabling real-time online attacks against similarity-based verification systems. Experimental results across multiple verification architectures demonstrate substantial degradation of verification performance together with severe manifold-level relational corruption. On the Markmatch verification system, the proposed attack reduces accuracy from 95.4% to 38.6% while completely reversing the positive-negative similarity structure.
[CV-108] A Convolutional Layer Activation Dimensionality Reduction for Out-of-Distribution and Adversarial Attack Detection Methods
链接: https://arxiv.org/abs/2608.10203
作者: Leandro de Souza Rosa,Lorenzo Capelli,Clara Nunes Barrancos,Mauro Mangia,Riccardo Rovatti
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Despite the success of convolutional neural networks in image classification tasks and their general application in multi-modal models, their susceptibility to out-of-distribution and adversarial attack samples raises concerns regarding trustworthiness and safety. Among the approaches to tackle such issues, detection methods that analyze the model’s intermediate activations to estimate a confidence score are a promising family that evaluates the decision process, relying on a dimensionality reduction step to enable efficient downstream processing of the high-dimensional activations. However, when considering convolutional layers, the dimensionality reduction methods in the literature either lack a mechanism to control the compression/information-loss trade-off or yield large representations. In this paper, we carefully analyze two state-of-the-art detection methods and their dimensionality reductions for convolutional layers and develop a novel reduction method with a controllable high-compression level. We extend these two state-of-the-art detection methods, enabling the usage of any dimensionality reduction, and evaluate their performance on out-of-distribution and adversarial attack detection. Results show that the detection methods with the proposed dimensionality reduction consistently perform better than, or comparable to, the strongest alternative. Furthermore, the proposed method is shown to reduce computation and memory footprints, given that it has the highest compression among the compared methods.
[CV-109] More Accurate Less Human: Gestalt Grouping in Vision Models IEEE-VIS2026
链接: https://arxiv.org/abs/2608.10195
作者: Sudhanva Manjunath Athreya,Sai Phani Kumar Malladi
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 9 pages, 7 figures, 5 tables. Conditionally accepted to VISxVision 2026, a workshop at IEEE VIS 2026. Includes appendix with per-task stimuli, metric derivations, and full per-model results
Abstract:Human vision organizes what it sees into wholes: same-colored points group into series, similar marks cohere into categories, and shapes complete into recognizable objects. These are the Gestalt operations that visualization design builds on. Whether vision models organize visual content this way has not been systematically tested. We introduce a behavioral battery that scores models against human data from prior perception studies on four grouping tasks: mark-color odd-one-out, color-series counting, silhouette recognition, and object odd-one-out. We apply it to 45 models across five training families: supervised, self-supervised, and contrastive vision-language encoders, open-weight VLMs, and closed foundation models. The battery reveals that agreement with human responses captures aspects of perceptual organization that conventional performance metrics fail to distinguish, with several closed models exhibiting substantially lower alignment than their benchmark accuracy would suggest. Scoring against published perception data therefore gives visualization research a reusable yardstick, requiring no new user study, for auditing whether the models now entering visualization pipelines organize what they see the way their human audience does.
[CV-110] Human versus Computer Vision
链接: https://arxiv.org/abs/2608.10181
作者: Elena Sirotkina
类目: Computer Vision and Pattern Recognition (cs.CV); Computers and Society (cs.CY)
备注:
Abstract:Computer vision saliency models predict where people will look, one map per image, and a billion-dollar predicted-attention industry sells those maps in place of measuring real viewers. I test the leading models from the audience side, against 11.4 million webcam gaze points from 3,023 US adults recruited to national quotas, viewing circulating news photographs. I show that an untrained central marker outperforms every trained network, because the content the networks add on top of the center falls where these audiences never look. What accuracy remains is systematically biased, favoring younger, White, and moderate viewers over older, Black, and ideologically extreme ones. I propose a way forward and build on what a group’s own gaze reveals about whether a model can learn that group at all, and I apply it across every demographic axis this sample supports. Ultimately, I show how systems that decide what people see can learn to see everyone, and this study supplies the standard by which such a claim should be judged.
[CV-111] DoseBridge: Denoising Diffusion Bridge Model for Dose Prediction in Lung Intensity-Modulated Proton Therapy
链接: https://arxiv.org/abs/2608.10173
作者: Zerun Zhang,Xiaoda Cong,Xiangkun Xu,Peter Y. Chen,Xuanfeng Ding
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Most radiotherapy dose-prediction models use only CT images and anatomical structures, although intensity-modulated proton therapy (IMPT) dose also depends strongly on beam geometry and available clinical datasets are often small. We present DoseBridge, a denoising diffusion bridge model that uses the patient CT as a structured bridge endpoint and encodes plan-specific beam geometry in a spatially aligned beam mask. Multiscale fusion combines CT, target, organ-at-risk, and beam-mask representations with 1.95% additional parameters. DoseBridge was retrospectively evaluated on single-institution CT images and treatment plans from 52 patients with advanced-stage lung cancer treated with 60 Gy in 30 fractions; 42 cases were used for training and 10 for testing. Performance was assessed using image-similarity, dose-volume, and Lyman-Kutcher-Burman normal-tissue complication probability (NTCP) metrics and compared with two deep-learning models. On the test cohort, DoseBridge achieved a mean absolute error of 4.170 Gy, peak signal-to-noise ratio of 23.06 dB, and structural similarity index of 0.798, outperforming both comparison models on these metrics. Clinical target volume D95 differed from the reference dose by 0.62 +/- 1.6 Gy; signed organ-at-risk mean-dose differences ranged from -0.32 to 0.24 Gy, and NTCP differences were -0.40 +/- 2.2 and 0.52 +/- 3.4 percentage points for acute esophagitis and radiation pneumonitis, respectively. Changing only the beam mask redirected predicted low-dose entrance regions while preserving the high-dose target region. To our knowledge, DoseBridge is the first denoising diffusion bridge model for radiotherapy dose prediction. These results support its feasibility as a beam-aware planning prior for lung IMPT, pending evaluation in larger external cohorts.
[CV-112] Motion Artifact-Aware Self-Supervised Representation Learning for 3D Brain MRI Motion Artifact Reduction
链接: https://arxiv.org/abs/2608.10170
作者: Mojtaba Safari,Shansong Wang,Zach Eidex,Matthew Goette,Tonghe Wang,Zhen Tian,Xiaofeng Yang
类目: Computer Vision and Pattern Recognition (cs.CV); Medical Physics (physics.med-ph)
备注:
Abstract:Patient motion remains a source of image degradation in brain MRI, leading to signal loss, blurring, and geometric distortion that compromise quantitative analysis. Existing deep learning methods for motion correction typically rely on paired clean-corrupted data or k-space acquisitions, which are rarely available in clinical settings. We propose SSRL-MAR, a motion artifact-aware unpaired representation learning framework for motion artifact reduction that requires neither paired training data nor explicit motion labels. SSRL-MAR employed a three-stage training strategy: (1) contrastive learning on 3D patches to extract motion representations by contrasting clean and synthetically corrupted images, (2) a motion artifact-aware synthesis network to generate motion artifacts from clean scans, and (3) a motion artifact-aware generator to restore clean volumes using the learned degrader for self-supervised supervision. On in-silico dataset, SSRL-MAR achieved PSNR 23.81dB, SSIM 91.55%, and NMSE 0.79%. On in-vivo MR-ART dataset, the pretrained model reduced motion distortion, and unsupervised domain adaptation further improved anatomical fidelity. Against a source-only supervised model trained on the same simulated pairs, SSRL-MAR improved PSNR by up to 2.0 dB on MR-ART after unsupervised domain adaptation, and remained within 0.25-0.47 dB of an oracle supervised model that requires real paired data unavailable in practice. At the milder motion level, volumetric error in structures such as the corpus callosum and ventricular system decreased by more than 50%, confirming improved neuroanatomical consistency. These results indicate that SSRL-MAR provides a robust and scalable image-domain solution for 3D brain MRI motion correction, enabling reliable structural quantification in large-scale neuroimaging studies without requiring prospectively acquired pairs or acquisition-specific calibration.
[CV-113] MAD-HOI: Masked Autoregressive Diffusion for Generating Articulated Hand Object Interactions from Text
链接: https://arxiv.org/abs/2608.10162
作者: Ananya Bal,Kartik Sharma,Ethan Lai,Samyak Tiwari,Liza Dahiya,Chaitanya Chawla,Laszlo A. Jeni
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 17 pages, 9 figures, 8 tables
Abstract:Methods for text-based generation of hand-object interaction (HOI) sequences primarily focus on producing smooth, physically plausible trajectories. A truly utilitarian method should additionally support variable-length generation, composite motion sequences, motion completion and infilling, and reliable termination without compromising physical plausibility. Standard diffusion models for HOI generation are typically trained only for text-to-motion generation on atomic motions and require the motion length to be specified a-priori. Autoregressive (AR) methods provide greater sequence-level flexibility, but commonly depend on discrete motion codes, which can lose contact-sensitive motion detail. To address these key limitations, we present a model performing Masked Autoregression with Diffusion for HOI generation (MAD-HOI). Our method starts by encoding hand and object motions in a continuous latent space while keeping them disentangled to maintain stream-wise control. This is followed by a masked autoregressive transformer to predict context features that condition a flow-matching head. MAD-HOI is capable of motion generation for atomic and composite articulated sequences, conditioned motion completion and infilling, as well as EOM (End of Motion) prediction from a single training objective. We provide comprehensive evaluations for these capabilities and benchmark our method on the ARCTIC and GRAB datasets. Our experiments demonstrate that our method generates more diverse and physically plausible interactions compared to other open-sourced baseline methods.
[CV-114] P3CA: Encoder-Agnostic Interpretation of Vision Foundation Model Embeddings via Spatial Probing MICCAI2026
链接: https://arxiv.org/abs/2608.10131
作者: Amoon Jamzad,Dilakshan Srikanthan,Faranak Akbarifar,Nooshin Maghsoodi,Parvin Mousavi
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 10 pages, 5 figures, 1 table. Accepted at the iMIMIC Workshop, MICCAI 2026. This arXiv version is the pre-peer-review author manuscript
Abstract:Vision foundation models are increasingly used as reusable encoders in medical image computing, yet their high-dimensional spatial embeddings are difficult to inspect beyond downstream task performance or global dimensionality reduction. We propose position-prompted PCA (P3CA), an encoder-agnostic method for local probing of channel-rich spatial tensors. Given a user-selected spatial prompt, P3CA estimates the feature normalization and dominant covariance directions within that region, then applies the resulting projection to the full tensor to visualize where locally informative directions are expressed. This produces a region-conditioned representation lens without modifying the encoder, retraining, or requiring task-specific labels. We implement P3CA in EmbedVision, an interactive 3D Slicer-based workflow, and evaluate it across natural images, colorectal pathology foundation-model embeddings, and spatial transcriptomic tensors. Across these settings, prompted projections reveal local structure suppressed by global PCA, improve prompt-matched pathology discrimination from frozen three-dimensional projections, and support comparison between learned and measured spatial representations.
[CV-115] 4D-WAM: 4D Consistent World Modeling for Autonomous Driving
链接: https://arxiv.org/abs/2608.10107
作者: Jiacheng Fu,Yibo Yuan,Meng Tian,Yue Li,Jiangtong Zhu,Jianhua Han,Yueyi Zhang,Jianwu Fang,Jianru Xue,Hang Xu,Zhiwei Xiong
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Emerging World-Action Models (WAMs) have demonstrated promising performance in autonomous driving by jointly modeling future driving scene evolution and trajectory planning. However, existing WAMs are typically trained with video data, which is only 2D projections of the underlying 4D driving scene. Consequently, WAMs fail to understand and capture the structure of 4D scenes and thus generate visually plausible yet 4D inconsistent future predictions that mislead downstream planning. To alleviate this issue, we present 4D-WAM, a model that leverages geometric foundation models for training-time supervision to enable 4D consistent world modeling. Specifically, we feed WAM-predicted future frames into a geometric foundation model, and use 4D-aware responses to define a 4D consistency loss. This loss encourages the model to understand, represent, and predict physically consistent 4D scenes during training, without additional inference cost. Moreover, we identify an early-decision phenomenon in WAMs and propose a decision-oriented timestep sampling strategy that emphasizes supervision at early, high-noise stages, where driving decisions are primarily formed. By propagating 4D supervision to this critical decision-formation phase, the proposed strategy further improves trajectory planning. Extensive experiments demonstrate that 4D-WAM effectively models 4D consistent scene evolution and achieves state-of-the-art performance on challenging NAVSIM-v1 and NAVSIM-v2 benchmarks.
[CV-116] Signpost Watermarking: Joint Optimization for Visual Watermark Coexistence
链接: https://arxiv.org/abs/2608.10091
作者: Shruti Agarwal,Vishal Asnani,John Collomosse
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:We present a method for training imperceptible visual watermarks to coexist with other such watermarks. Recent work has shown that independently trained image watermarking models can coexist with surprisingly limited interference, enabling watermark ensembling. However, this coexistence is a serendipitous property rather than an explicit optimization objective, leaving interference uncontrolled and potentially reducing decoding robustness or visual quality. We first show empirically that the same coexistence property extends to video watermarking. We then show that both image and video watermarks can be trained with a decoder-aware objective to improve coexistence. Our results suggest a practical path to signpost watermarks that indicate the presence of independently deployed provenance watermarking systems, supporting layered provenance signaling for content authenticity and rights.
[CV-117] LEGO: Leveled Language Gaussian Splatting ECCV2026
链接: https://arxiv.org/abs/2608.10057
作者: Yuning Peng,Haiping Wang,Yuan Liu,Yipeng Lu,Zhen Dong,Bisheng Yang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ECCV 2026. Project page: this https URL
Abstract:We introduce LEGO for advanced open-vocabulary scene understanding. Beyond basic concept recognition, its core innovation lies in capturing the intrinsic semantic hierarchies within the scene, such as the “flowerpot - bouquet - bud - petal” lineage. While foundation models like SAM can identify multi-granular structures in 2D, their partitions are strictly perspective-bound and lack cross-view consensus. LEGO self-adaptively re-grades volatile multi-view SAM granularities into a unified, 3D-consistent hierarchy. This provides precise supervision for the structurally coherent, multi-level segmentation of 3D scenes. By grounding these segments with CLIP embeddings, LEGO recovers open-vocabulary semantic logic across hierarchical levels. Furthermore, by incorporating spatial relationships, we elevate these segments into level-wise language scene graphs, effectively empowering Large Language Models to perform complex, context-aware spatial reasoning and precise visual grounding. Experimental results demonstrate that LEGO establishes new state-of-the-art performance across both promptable and open-vocabulary 3D segmentation benchmarks, exhibiting advanced hierarchical scene decomposition and context-aware spatial reasoning.
[CV-118] Protection Levels for Vision-Based Pose Estimation ATC
链接: https://arxiv.org/abs/2608.10023
作者: Olivia Beyer Bruvik,Romeo Valentin,Marc R. Schlichting,Don Walker,Mykel J. Kochenderfer
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV); Systems and Control (eess.SY)
备注: 11 pages, 5 figures. Accepted for publication at the 2026 AIAA DATC/IEEE 45th Digital Avionics Systems Conference (DASC). O. Beyer Bruvik and R. Valentin contributed equally
Abstract:Vision-based navigation complements Global Navigation Satellite Systems, but certification demands integrity guarantees that account for faulty measurements. Previous work presented a probabilistic computer vision pipeline for runway-based pose estimation with fault detection inspired by Receiver Autonomous Integrity Monitoring. This work extends that framework by deriving protection levels, which provide probabilistic bounds on pose error that remain valid under undetected faults. We present an algorithm for computing protection levels for the nonlinear Perspective- n -Point problem applied to an aviation setting. The algorithm covers all six degrees of freedom of the aircraft pose (position and orientation) directly. We analyze the effect of measurement redundancy, pixel-level prediction uncertainty, and runway distance on the resulting protection levels. To make the results tangible, we demonstrate tradeoffs in the protection levels on an illustrative runway example.
[CV-119] ReCBM: Uncertainty-Gated Relational Reasoning for Concept Bottleneck Models
链接: https://arxiv.org/abs/2608.10004
作者: An Sui,Yuzhu Li,Fuping Wu,Xiahai Zhuang
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Concept Bottleneck Models (CBMs) provide an interpretable framework by grounding predictions in human-understandable concepts, enabling semantic inspection and test-time intervention. Recent variants have improved CBMs through richer concept representations, uncertainty estimation, and dependency modeling. However, robust reasoning under unreliable concept states remains underexplored. Without such reasoning, misleading semantic evidence can propagate through the bottleneck, compromising both explanations and downstream predictions. To address this issue, we propose ReCBM, an uncertainty-gated relational reasoning framework for CBMs. ReCBM introduces semantically defined concept relations into the bottleneck and uses uncertainty to guide their refinement. By modeling co-occurrence, implication, and exclusion, ReCBM specifies how evidence is exchanged across concepts, while uncertainty modulates the contribution of each concept during this process. Experiments across diverse datasets showed that ReCBM improved concept and task recovery under missing and flipped concepts, supported uncertainty-aware intervention, and extracted compact task-relevant concept subsets without degrading downstream performance.
[CV-120] ransformer Geometry Observatory TGO-IV: Developmental Topology Observatory
链接: https://arxiv.org/abs/2608.09997
作者: Kaustubh Kapil,Kishor P. Upla
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Transformers have had a profound impact on the world of language processing and computer vision. As efforts to answer the million-dollar question of ``How does a Transformer learn?" have been increasing, existing interpretability studies primarily analyze representations at isolated layers or the network as a whole, while the developmental evolution of individual representations and its manifolds across transformer layers remains underexplored. With this work, we aim at providing a comprehensive analysis of the evolution of representations as the representation point cloud transforms across the layers; thereby attempting to isolate layers or establish a trend which comes closer to justifying how and when raw input representations evolve into task-relevant feature representations. Thus, Transformer Geometry Observatory-TGO-IV introduces a topological framework for analysing the evolution of Transformer representations through the lens of Persistent Homology. Rather than studying local geometric properties alone, TGO-IV constructs Vietoris–Rips simplicial complexes from token-level representation point clouds and investigates the evolution of their persistent topological signatures across Transformer layers. The proposed framework comprises complementary topological observatories including Persistence Diagrams, Barcode Diagrams, Betti Curves, Persistence Landscapes, Bottleneck Distance, and Wasserstein Distance, enabling a comprehensive analysis of how the global topology of representation point clouds develops throughout the forward pass.
[CV-121] Retrieval-Augmented Vision Foundation Models for Robust Leukemia Cell Classification across Multiple Microscopy Datasets
链接: https://arxiv.org/abs/2608.10657
作者: Carlos Zamora,Hiram Zuniga,Ulises Orozco-Rosas,Kenia Picos
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Accepted at SPIE Optics + Photonics 2026 for oral presentation. 23 pages, 12 figures, 9 tables
Abstract:Leukemia cell image classification is challenged by real-world domain shifts from acquisition, staining, illumination, and site protocols, causing single-dataset models to generalize poorly in real clinical scenarios. This work presents a robust framework for leukemia classification across multiple heterogeneous datasets using a two-stage pipeline with a pretrained vision foundation model. Stage 1 performs binary classification (leukemia vs. non-leukemia) and is trained using 122,167 single-cell images. Stage 2 is conditionally applied to Stage 1 positives to perform subtype classification into Acute Lymphoblastic Leukemia (ALL) and Acute Myeloid Leukemia (AML), trained using 69,400 single-cell images. Labels are harmonized across five heterogeneous datasets to enable cross-dataset training, and performance is evaluated on a held-out dataset protocol to assess domain-shift generalization. Within this pipeline, three encoders are benchmarked (DinoBloom, pretrained on single-cell images; BiomedCLIP, pretrained on biomedical data; and CLIP as a general-purpose model) under linear probing, Low-Rank Adaptation (LoRA), and a Retrieval-Augmented Classification (RAC) module that retrieves the top-k most similar cell images to provide cytomorphological grounding. The objective is to quantify how much domain-specific pretraining contributes to performance under domain shift, and whether cost-effective adaptation and retrieval can be a viable alternative to expensive domain-specialized pretraining. The held-out protocol additionally serves as a diagnostic tool, revealing when classification performance is attributable to dataset-specific artifacts rather than to cytomorphological features.
[CV-122] Iterative Erasure Count Is Not an Affine-Invariant Concept Dimension
链接: https://arxiv.org/abs/2608.10566
作者: Tingan Jin,Shuhang Dong,Haosong Li,Chung-Hsien Chou
类目: Machine Learning (stat.ML); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 15 pages, 3 figures. Code and results included as ancillary files. Tingan Jin, Shuhang Dong, and Haosong Li contributed equally
Abstract:How many directions does a neural representation use to encode a concept? A common answer repeatedly erases probe directions and reports the stopping count or cumulative removed rank. We show that both quantities can change under an information-preserving invertible reparameterization, so neither is intrinsically a concept dimension. We distinguish model-defined population quantities (generating dimension, sufficient linear dimension, and minimum guarding rank) from procedure-defined quantities such as stopping count and cumulative edit rank. In a population Gaussian construction, an invertible shear preserves the prediction problem and all three quantities, yet changes the cumulative Euclidean erasure count from one to two. The separation holds for Moore–Penrose ordinary least squares and every finite nonnegative ridge weight. For a two-output full-QR procedure matching our motivating video analysis, cumulative edit rank similarly changes from two to the ambient dimension four. Conversely, the complete cumulative metric-QR trajectory is affine-equivariant when its positive-definite metric, probe, regularizer, and tie-breaking are transported consistently; exact covariance is one corollary, not a canonical semantic metric. In a known-rank finite-sample Adam/QR calibration, identity mixing stops after one accepted update in all 20 large-sample runs, whereas each tested shear a\in.5,.75,1,1.25,2\ accepts at least two updates in all 20 runs. Controlled reparameterizations of frozen V-JEPA2 features preserve rank-zero predictions yet alter later Euclidean trajectories under practical optimization. These visual contact experiments are stress tests, not estimates of contact dimension. Iterative erasure therefore returns a procedure-relative estimand jointly determined by representation geometry and the full measurement procedure, not a semantic dimension by itself.
[CV-123] When Repository Labels Are Not Image-Level Truth: A Supervision Auditing Framework for Chest Radiograph AI MICCAI MICCAI2026
链接: https://arxiv.org/abs/2608.10084
作者: Yesika Alexandra Agudelo-Londoño,Jhon Wilmer Pino-Román,Brahian Carrera Rodríguez,José Miguel Castañeda-Bedoya,Juan Pablo Gómez-López,Aura C. Puche-Sarmiento,Niharika S. D’Souza,Juan Sebastian Osorio-Valencia,Jon E. Duque-Grajales,Jazmín Ximena Suárez-Revelo,Jorge Mario Vélez-Arango,Gabriel Castrillón
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted (oral) at the 3rd MICCAI Student Board (MSB) EMERGE Workshop, MICCAI 2026. This is the authors’ version; the final authenticated version will appear in Springer Lecture Notes in Computer Science (LNCS). 10 pages, 3 figures
Abstract:Public chest X-ray repositories are widely used to train medical AI systems, yet their labels are typically extracted from radiology reports rather than verified directly on images. As a result, repository labels are often treated as image-level ground truth without validating whether they reflect what is actually visible in the radiograph. We introduce Repository Supervision Auditing (RSA), a framework that evaluates repository-derived labels against expert image-level annotations before model development. Using cardiomegaly in MIMIC-CXR as a case study, RSA compares repository labels with radiologist-reviewed image annotations, characterizes disagreement sources, and builds a curated cohort for deployment-oriented evaluation. Repository-derived cardiomegaly labels showed near-zero agreement with expert image-level assessment, identifying only 1% of expert-confirmed cases. Most discrepancies resulted from non-mention rather than explicit report negation, with expert-confirmed cardiomegaly identified in nearly half of studies assigned a repository-derived No Finding label. Using the resulting expert-curated cohort, a DenseNet121 model achieved a test ROC-AUC of 0.853. These findings show that repository labels may not reliably represent image-level truth and highlight supervision auditing as a critical step for developing trustworthy medical imaging AI.
[CV-124] LoRCA: LoRA Cycle Adaptation for Histology to HiP-CT Translation with DINOv3
链接: https://arxiv.org/abs/2608.10002
作者: Yang Zhou,Edoardo Occhipinti,Banboye Kidzeru Elvis,Jishizhan Chen,Stathis Megas,Joseph Brunet,Joanna Purzycka,Theresa Urban,Hector Dejea,Sarah Amalia Teichmann,Menna R Clatworthy,Paul Tafforeau,Peter D Lee,Claire L Walsh
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Hierarchical Phase-Contrast Tomography (HiP-CT) is a synchrotron based X-ray imaging technique that enables non-destructive, volumetric imaging of intact organs with multi-resolutions bridging 20 \mu m /voxel for whole organs to near-cellular resolution ( \sim 0.8 \mu m /voxel) in local regions. This offers the opportunity to bring volumetric whole-organ context to histology. However, nonlinear registration between H\E histology and HiP-CT volumes is challenging due to the differences in feature representations of different colour spaces. Synthesis-before-registration methods have shown strong results in histology-to-MRI and histology-to-CT alignment. However, existing approaches either rely on manual anatomical contours or are trained from scratch without semantic constraints, limiting their generalisability to soft tissue organs and novel modalities. We propose LoRCA (LoRA Cycle Adaptation), a cycle consistent style translation framework built on a shared frozen DINOv3 with modality-specific LoRA adapters, learning modality-specific representations that are decoded and adversarially trained. LoRCA enables structure-preserving translation without requiring paired training data. The frozen backbone is intended to be a structural anchor that prevents content drift by preserving pretrained semantic-extraction capability. We evaluate translation quality using Fréchet Inception Distance (FID) and structural fidelity via mutual information and Canny edge preservation. LoRCA outperforms CycleGAN in both translation quality and structural consistency. As a preliminary indicator of downstream registration utility, we find that style-translated images yield increased feature correspondences under MatchAnything on manually aligned HiP-CT and histology test pairs, suggesting that LoRCA-style translation is a promising step towards 2D histological sections to 3D HiP-CT volumes registration.
[CV-125] Implicit representations are dead. Long live explicit primitives!
链接: https://arxiv.org/abs/2608.10001
作者: Nil Stolt-Ansó,Maik Dannecker,Wenqi Huang,Andras Jakab,Daniel Rueckert
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Continuous parameterization of medical data has emerged as a powerful paradigm for resolution-independent image representation. While Implicit Neural Representations offer high fidelity and compact storage, their reliance on global Multi-Layer Perceptrons incurs sizeable computational costs, large memory requirements, and extensive optimization times. As medical imaging trends towards ever-more detailed, high-resolution volumes, these costs impose significant bottlenecks in the applicability of implicit approaches. Recently, explicit Gaussian-based primitives have revolutionized the representation learning paradigm by trading deep network evaluations for localized, rasterization-friendly primitives. In this paper, we present a comprehensive, cross-dimensional evaluation of Gaussian representations against implicit approaches for medical imaging applications. First, we outline a theoretical overview on the mathematical properties offered by explicit primitives beyond what is capable under the implicit neural paradigm. Subsequently, we benchmark the computational performance on two demanding image datasets: 2D microscopy histology and 3D lung Computed Tomography (CT). Our experiments demonstrate that Gaussian representations consistently match or surpass reconstruction metrics compared to implicit methods across all compression factors, while displaying significantly lower optimization times, and memory requirements. Together with the compelling mathematical properties offered by explicit primitives, these findings motivate the wider adoption of Gaussian representations and position them as an attractive direction for future research in medical imaging.
[CV-126] Pre- to Post-Contrast Synthesis of Breast DCE-MRI using Latent Bridge Matching
链接: https://arxiv.org/abs/2608.10000
作者: Sina Amirrajab,Zohaib Salahuddin,Henry C Woodruff,Philippe Lambin
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is central to breast cancer imaging, but gadolinium administration increases scan burden and motivates contrast-reduced alternatives, including synthetic contrast generation. We propose a latent bridge matching (LBM) framework for synthesizing peak-enhanced breast DCE-MRI from pre-contrast images in the MAMA-SYNTH challenge setting. Instead of starting from Gaussian noise as in conventional latent diffusion models (LDMs), the proposed model learns a conditional bridge between paired pre-contrast and peak-enhanced VAE latents. A latent UNet predicts the remaining correction from intermediate bridge states to the peak-enhanced latent, enabling iterative refinement while keeping the trajectory anchored to patient-specific anatomy. We evaluated two LBM conditioning variants on 91 DUKE validation cases. For the tumor-conditioned variant, tumor masks were used as conditioning inputs. Tumor-conditioning improved performance compared with pre-contrast conditioning, reducing MSE from 1.023 to 0.940 and FRD from 7.523 to 4.716, while increasing tumor SSIM from 0.355 to 0.429. The tumor-conditioned LBM also outperformed the evaluated LDM baseline on this validation cohort. These results suggest that latent bridge matching is a promising pre-contrast-anchored formulation for virtual contrast enhancement, while further work is needed to validate generalization and remove dependence on ground-truth tumor masks at inference.
[CV-127] Robustness of transferability estimation metrics for medical imaging
链接: https://arxiv.org/abs/2608.09999
作者: Niclas Claßen,Théo Sourget,Dovile Juodelyte,Rob van der Goot,Veronika Cheplygina
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:In transfer learning, the choice of source model largely influences the performance on a target dataset. Still, selecting a fitting source remains a challenging task, especially in medical imaging where one has to decide between models pre-trained on off-the-shelf options, such as ImageNet, and domain specific datasets. Transferability estimation (TE) metrics address this problem by aiming to predict the best performing source model in a computationally cost effective way. However, previous work has reported conflicting TE metric performances due to differences in experimental setups. Moreover, most TE metrics are designed for and evaluated on natural images, while being optimized for accuracy, whereas in medical imaging metrics that are more robust to class imbalance are typically used. We study the impact of varying the target dataset as an isolated factor, by constructing miniature populations of different sample sizes and random seeds. In addition, we investigate the influence of the evaluation metric used to obtain the reference ranking. We find that small modifications to the target dataset change the rankings. Furthermore, we show that the choice of evaluation metric affects the reference rankings and therefore the evaluation of TE metrics. Overall, we observe a low agreement between rankings from TE metrics and reference. The code, model checkpoints and data splits used in this work are available through this https URL.
[CV-128] Energy and Performance Benchmarking of Deep Learning Models for Breast Cancer Detection ICML
链接: https://arxiv.org/abs/2608.09996
作者: Samar Garrab,Ghada Achour
类目: Image and Video Processing (eess.IV); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Accepted at ICMLA 2026 (IEEE International Conference on Machine Learning and Applications). Camera-ready version submitted
Abstract:Recent advances in machine learning have greatly improved breast cancer detection, enabling more accurate and timely diagnosis. Deep learning (DL) models show strong potential for medical image analysis; however, as their architectural complexity increases, their environmental impacts are becoming a growing concern. In this paper, we present a comparative analysis of seven DL models for breast cancer detection on two medical datasets: Breast Ultrasound and BreakHis 400X. The evaluated architectures range from Convolutional Neural Networks (CNNs) and transformers to hybrid models. In addition to performance metrics, we assess CO2 emissions during both training and inference. Our results show that EfficientNet and ResNet consistently deliver strong performance, although with higher CO2 emissions. The selected transformers, such as DeiT-Tiny, perform competitively on both datasets, whereas DenseNet121 achieves lower accuracy. On the Breast Ultrasound Dataset, DeiT provides the most favourable balance between accuracy and energy consumption, whereas on the BreakHis dataset, the ViT and Swin models achieve the best results. Overall, our findings indicate that no single architecture category from the evaluated ones consistently dominates across the two selected datasets. Our results highlight the importance of jointly considering performance, emissions, and dataset characteristics when selecting models for medical applications.
[CV-129] Structural Guidance for Unified Joint Demosaicing and Denoising
链接: https://arxiv.org/abs/2608.09995
作者: Qixin Zheng,Ping Chen,Qiangqiang Shen,Haijin Zeng
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注: 19 pages, including supplementary material
Abstract:Joint demosaicing and denoising is a fundamental step in camera image signal processing, yet remains challenging because different Bayer-like color filter arrays (CFAs) and sensor noise jointly corrupt both color sampling and image content. Existing unified restoration networks explicitly model CFA geometry but are still driven primarily by pixel-level supervision, making them prone to structural degradation around edges, repetitive textures, and moiré patterns where local evidence is unreliable. We attribute this limitation partly to the absence of explicit structural guidance beyond pixel-level reconstruction supervision. Motivated by this observation, we propose a structural-guided unified restoration framework that injects pretrained structural knowledge into CFA-aware image restoration. Our model receives a unified five-channel observation consisting of the raw mosaic, CFA masks, and a noise-level map. A SwinIR restoration branch reconstructs pixel details under CFA-conditioned modulation, while a parallel structural reasoning branch extracts complementary structural cues from a sparse pseudo-RGB observation. To bridge the substantial domain gap between sparse noisy sensor data and the natural-image pretraining domain of the structural encoder, we introduce a lightweight trainable adapter before residually fusing structural and restoration features. A shared decoder jointly predicts the restored RGB image and an auxiliary clean mosaic, providing supervision in both image and sensor domains. Extensive experiments across multiple CFA patterns and noise levels demonstrate consistent improvements over state-of-the-art unified and CFA-specific methods, indicating that adapted structural priors can enhance robust camera image restoration. The source codes and dataset are provided in the supplementary material.
[CV-130] SpecF2M: A Spectral-Aware Multi-task Network Estimating Axial Length and Refractive Error from Pediatric Fundus Photographs
链接: https://arxiv.org/abs/2608.09994
作者: Mengxian He,Xinyue Liu,Yunyun Sun,Wei Hao,Minqing Zhang,Lichun Wang,Shunyi Zhang,Wu Yuan
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Spherical Equivalent Refraction (SER) and Axial Length (AL) are core indicators for pediatric myopia screening, yet their measurements require dedicated biometry and cycloplegic refraction. Fundus photography offers an accessible imaging modality, as myopia-related posterior-pole changes are visible in 45 ^\circ fundus images. However, these cues are often low-contrast, spatially diffuse, and multi-scale. Moreover, AL, Sphere (SPH), and Cylinder (CYL) share partially overlapping but non-identical anatomical correlates. We propose SpecF2M, a spectral-aware multi-task network for estimating AL and SER components from pediatric fundus photographs. SpecF2M integrates a deterministic anatomy-guided enhancement module, a hybrid spatial–spectral backbone combining MixCNN and Hybrid Spectral Learning (HSL) blocks, and an expert-routing head for component-level estimation of AL, SPH, and CYL. On a pediatric cohort of 4,359 eligible child visits and 6,966 fundus images, SpecF2M outperforms controlled CNN/ViT baselines for AL and SPH estimation, achieving MAEs of 0.5347 mm and 0.7062 D, respectively. Component-level analysis further reveals asymmetric task coupling, where CYL exhibits weaker association with fundus-derived myopic patterns than AL/SPH. These results support fundus-based, screening-oriented estimation of pediatric myopia indicators, while external validation remains necessary before deployment.
[CV-131] APCReg: Anatomical-Prior-Guided Coarse-to-Fine CBCT–IOS Registration via Multi-View Projection and Reliability-Controlled Residual Correction
链接: https://arxiv.org/abs/2608.09993
作者: Xincan Zheng,Yaqi Wang,Zhi Li,Jiahao Bao,Lan Feng,Yiru Xia,Shuai Wang
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注: 10 pages, 7 figures
Abstract:Registration between cone-beam computed tomography (CBCT) and intraoral scans (IOS) is essential for patient-specific surgical planning. However, disparate imaging modalities, limited overlap, and large pose offsets make automated registration unreliable. Consequently, clinical registration remains dependent on conventional geometry pipelines and manual clinician adjustment. To address these challenges, we propose APCReg, an anatomical-prior-guided coarse-to-fine framework for global registration and reliability-controlled residual correction. Specifically, multi-view anatomical coarse registration (MACR) performs ordered orthogonal projection alignment (buccal, proximal, and occlusal) to decompose the six-degree-of-freedom search before three-dimensional refinement. Overlap-aware residual registration (OARR) combines shared KPConv features, a folded arch-length cue, overlap-gated cross-attention, and Sinkhorn matching. Finally, dental-arch-structured hypothesis selection evaluates diverse poses on held-out reliable correspondences, while a ground-truth-free coarse-retention guard conditionally retains a geometrically reliable coarse pose. On 60 held-out jaw pairs, APCReg achieves a submillimeter mean Chamfer distance of 0.87 mm and a Hausdorff distance of 2.92 mm under this evaluation protocol, and ranks first across the six reported metrics among the evaluated open-source baselines.
[CV-132] Knowledge-Guided 3D CT Generation: A Conditioning-Centric Taxonomy ECAI2026 IJCAI
链接: https://arxiv.org/abs/2608.09992
作者: Francesca Pia Panaccione,Eugenio Lomurno,Matteo Matteucci
类目: Image and Video Processing (eess.IV); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Accepted to IJCAI-ECAI 2026, Survey Track
Abstract:Controllable generation guided by external knowledge is a key requirement in modern generative deep learning applications, enabling the synthesis of samples with explicit constraints on semantic content, structural properties, and variability. In 3D Computed Tomography (CT), such control is essential for clinical applications, including data augmentation, privacy-preserving data sharing, and the simulation of specific anatomical or pathological scenarios. While research on conditional 3D CT generation has expanded rapidly, the diversity of existing approaches makes systematic comparison difficult and obscures fundamental design choices. In this survey, we propose a conditioning-centric taxonomy that organizes the literature along three orthogonal dimensions: the type of external knowledge (K), the knowledge integration paradigm (I), and the generative architecture (A). This factorization defines an explicit design space (K x I x A) that provides a unified perspective on prior work. Using this framework, we systematize existing methods, identify dominant trends and recurring design patterns, and highlight underexplored regions of the design space that point toward promising directions for future research.
[CV-133] Longitudinal 3D Foundation Modeling for Neoadjuvant Breast Cancer Response Prediction from Serial DCE-MRI MICCAI2026
链接: https://arxiv.org/abs/2608.09991
作者: Fidel Omar Tito Cruz,Neda Ghafouri,Zengyan Wang,Pegah Khosravi,Yu Tian,Chen Chen
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at the Applications of Medical AI (AMAI) Workshop at MICCAI 2026
Abstract:Pathologic complete response (pCR) is an important endpoint in neoadjuvant chemotherapy (NAC) for breast cancer, and predicting pCR from imaging during treatment could support treatment response assessment. Many existing imaging-based approaches rely on a single static timepoint, which fails to capture changes that occur during treatment. In this work, we present a longitudinal framework that combines a frozen 3D foundation encoder (Pillar-0) with our Temporal Dynamics Network (TDN) to predict treatment response from serial Dynamic Contrast-Enhanced (DCE) MRI acquired across four clinical timepoints from pre-treatment to pre-surgery. The TDN combines time-aware volumetric embeddings with clinical and treatment data to predict pCR. Evaluated on 982 patients from the combined I-SPY2 and ACRIN-6698 cohort, the proposed model achieves strong performance across all reported metrics when longitudinal 3D imaging is fused with clinical data (test AUROC: 73.6%, balanced accuracy: 69.1%). While clinical variables provide the strongest individual predictive signal, longitudinal 3D imaging contributes complementary information when fused with clinical data, improving pCR prediction. Our source code is available at: this https URL.
[CV-134] Algorithmic statistics of retinal images
链接: https://arxiv.org/abs/2608.09989
作者: Loan Huynh,Ronald Zambrano,Layton Aho,Fabio Lavinsky,Gadi Wollstein,Joel S. Schuman,Andrew R. Cohen
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:There has been a tremendous amount of image processing and machine learning research to measure and classify disease progression from live optical coherence tomography (OCT) imaging of the retina. The images considered here are large, complex, three-dimensional (3-D) and difficult to visualize effectively. Many current supervised machine learning approaches, \emphe.g. neural networks, are non-metric meaning that any features or measurements generated can introduce systematic distortion that may be correlated with underlying non-meaningful physiological differences. Here we present a metric learning approach using the normalized compression distance (NCD) combined with anisotropic structure-enhancing filters to quantify and visualize the principal differences among a collection of 3-D retinal images. We validate the NCD-measured structural differences between pairs of images against the physician-measured change in visual field function, achieving a prediction error of \sim 0.5 dB, more accurate than non-metric deep learning approaches. The normalized compression vectors (NCV) are proposed as a feature set measuring visual differences among a collection of 3-D microscopy images. The utility of the NCV for visualizing and measuring patterns of change is demonstrated for a human with moderate non-progressing glaucoma and for a non-human primate model using intraocular pressure setting manipulation. We conclude with a brief simulation of non-metric embedding features, \emphe.g. from neural networks, introducing class-correlated statistical distortion.
[CV-135] Rescene: band-limited stochastic forcing turns a frozen neural weather operator into a climate emulator
链接: https://arxiv.org/abs/2608.09971
作者: Minjong Cheon
类目: Atmospheric and Oceanic Physics (physics.ao-ph); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Over the past few years, the rapid development of machine learning (ML) models for weather forecasting has produced deterministic models whose medium-range skill matches or exceeds that of the European Centre for Medium-Range Weather Forecasts (ECMWF)'s high-resolution forecast (HRES). However, when these models are integrated freely beyond the horizon they were trained for, they blow up, drift, or lose their seasonal cycle, and retraining them for stability is expensive. We therefore ask what can be recovered from a strictly frozen backbone. We present Rescene, a 0.4 M-parameter wrapper around a frozen 1.5 degree, 6-hourly vision-transformer operator, developed using ERA5 reanalysis data and comprising a deterministic “slow clock” (0.33 M) that blends the forecast toward a lead-aware day-of-year climatology and a generative head (0.06 M) that adds a spectrally shaped stochastic perturbation at every step. The performance evaluation demonstrates that the deterministic wrapper alone is stable for decades but collapses daily variability to 40% of ERA5. Adding the generative head restores 126% (Z500) and 130% (MSLP) of the observed daily variability with pattern correlations of 0.89 and 0.92, recovers 82% of the observed blocking frequency, keeps the ensemble calibrated (spread-skill ratio 0.78-0.97 from day 7 to day 90), and integrates for 100 years with no detectable drift (+0.008 +/- 0.014 K per century). Moreover, because the perturbation is band-limited to total wavenumber k \le 20 , the small scales are never forced, yet realistic k \ge 20 power is sustained: a direct decomposition of the 6-hourly energy budget shows that the frozen operator supplies 28 times more energy than the perturbation at k \ge 40 , with a fractional growth rate 247 times larger at the grid scale than at planetary scales.
人工智能
[AI-0] How to Verify Consistency of Probabilistic Claims
链接: https://arxiv.org/abs/2608.11181
作者: Orr Paradise,Oliver Richardson,Yoshua Bengio,Shafi Goldwasser
类目: Computational Complexity (cs.CC); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:When a probabilistic predictor answers many conditional-probability queries, are its answers self-consistent, and can this be verified in polynomial time? This problem is of interest for AI safety, where safety is derived from honesty about probabilistic predictions of unwanted outcomes potentially caused by an AI action. We construct an interactive PCP as follows. Let a predictive model be specified by a probability circuit P and a circuit Q which outputs confidence in predictions. Together, P and Q implicitly specify exponentially many probabilistic claims. We show a protocol in which a polynomial-time verifier can verify the approximate consistency of (P,Q). The verifier is given the pair of circuits (P,Q), which it evaluates at only a few points; alongside them it is given a proof oracle, an encoding of a witnessing probability distribution allegedly consistent with the predictions of (P,Q), which it reads at a few locations while interacting with a single untrusted prover. En route, we must ensure the existence of a sparse witnessing distribution consistent with the model’s predictions. To do so, we first consider witness distributions for the consistency of explicit probabilistic claims, rather than claims specified by a predictor: say m claims, each of the form Pr[Y = 1 | X = x] = p, over n Boolean variables. Building on work initiated by Nilsson (Artif. Intell., 1986), we place l_2-approximate probabilistic consistency of explicit claims in NP, with certificates of length O(mn + log B) in the input bit-precision B; we further show how a small additive completeness-soundness gap removes the dependence on B. Together these results provide a complexity-theoretic foundation for certifying the self-consistency of probabilistic predictors. We view our interactive PCP as a first step toward training predictive models to prove their own consistency.
[AI-1] sLTN: Structural Logic Tensor Networks
链接: https://arxiv.org/abs/2608.11136
作者: Davide Rinaldi,Luciano Serafini
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Logic Tensor Networks (LTN) provide a neurosymbolic framework in which first-order logic is interpreted through tensor operations, enabling logical constraints to be integrated with differentiable learning. However, the original formulation of LTN is primarily suited to data represented as flat collections of individuals, and does not explicitly capture structural organization such as temporal order, sequential position, or graph connectivity. We introduce sLTN, an extension of LTN that makes structural dimensions first-class elements of the language. Structural dimensions represent named tensor axes associated with domain-specific organization, such as time steps, sequence positions, or graph nodes. They can be quantified explicitly, related through structural relations, and used to express temporal, sequential, and relational constraints directly at the logical level. We formalize the syntax and fuzzy tensor semantics of sLTN and show that, in the absence of structural dimensions, the framework recovers the original LTN semantics as a special case. We further describe a PyTorch implementation based on a declarative signature, formula parsing, and tensorial interpretation. The framework is illustrated on representative temporal and sequential reasoning examples. This paper serves as a companion to the sltn library, available at this https URL.
[AI-2] wo-stage Odd Residual Flows for Mean-Preserving Probabilistic Time Series Forecasting
链接: https://arxiv.org/abs/2608.11114
作者: Kiran Madhusudhanan,Christian Klötergens,Lars Schmidt-Thieme,Vijaya Krishna Yalavarthi
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Probabilistic forecasting plays an essential role in risk-sensitive decision-making, particularly in long-horizon settings. However, existing approaches often face a fundamental trade-off between distributional flexibility and accurate mean prediction. Traditional parametric methods, such as Mean Variance Estimation (MVE), can suffer from degraded point accuracy when trained under joint Negative Log-Likelihood (NLL) objectives, while modern-flexible generative models, including Normalizing Flows and Diffusion Models, typically rely on costly Monte Carlo sampling and may yield suboptimal mean estimates. To address this limitation, we propose Two-stage Odd Residual Flows (TORF), a framework that decouples mean forecasting from uncertainty estimation. In the first stage, a pre-trained deterministic model is used to produce an accurate mean prediction. In the second stage, a Restricted Normalizing Flow, with strictly odd functions learns flexible residual distributions around the point forecast, guaranteeing mean preservation from the first stage without sampling. Experiments show that TORF achieves state-of-the-art deterministic accuracy (NMAE) while providing strong density estimation performance (CRPS) on short and long-horizon forecasting.
[AI-3] Why Does CLAUDE.md Keep Growing? Catastrophic Remembering in Agent ic Coding
链接: https://arxiv.org/abs/2608.11095
作者: Kushal Chakrabarti
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Software Engineering (cs.SE)
备注:
Abstract:Agentic coding READMEs like this http URL grow without bound in real repositories, stopping only when the repository retires or someone rewrites the file wholesale. We trace this to imperfect recall: appending an instruction is always cheap, but once an instruction’s rationale is gone, deleting it without risking a correctness regression costs O(2^|D|) in a prompt of |D| instructions. We name the resulting divergence catastrophic remembering, the inverse of catastrophic forgetting around which continual learning is organized. First, we characterize this phenomenon across 247,694 instruction lifetimes in 1,867 repositories: agentic prompts grow without bound, more than tripling over their lifetime (+226%), gaining +4.9 net instructions every commit; further, the older an instruction gets, the less likely it is to be deleted (log-hazard -0.032/commit). Then, we show that prompt comments can halt the growth: inverting IFEval yields verifiable worlds whose optimal prompts are known, and there comments encoding latent reasoning remove 99.3% of excess instructions (+211.3% to +1.4%). Finally, applying the same inversion to WildIFEval, we show that prompt comments can improve real-world agentic instruction-following by up to 23.1%. If English is the new code, why don’t we have comments yet?
[AI-4] RTSKG: Building a Rail Transit Station Knowledge Graph Dataset ISWC2026
链接: https://arxiv.org/abs/2608.11080
作者: Shutong Zhu,Tianxing Wu,Runfeng Liu,Yuang Gu,Xuan He,Yuan Zhu
类目: Artificial Intelligence (cs.AI)
备注: 21 pages, Accepted by ISWC 2026
Abstract:Rail transit systems play a vital role in urban mobility and economic development. As key components of such systems, rail transit stations function as critical transport hubs that enhance urban accessibility and stimulate development in surrounding areas. City-level rail transit station related tasks (e.g., ridership prediction) require large-scale urban data, but current studies often neglect complex interactions among various urban entities in terms of data organization. In this paper, to address the above issue, we build a Rail Transit Station Knowledge Graph (RTSKG) dataset which explicitly models the spatial and semantic interactions among different kinds of urban entities, to benefit city-level rail transit station related tasks. RTSKG integrates heterogeneous urban entities, such as rail transit stations, road segments, and points of interest, with a specially designed unified schema, and is accessible as Linked Data at this https URL. Evaluations on station-area store recommendation and knowledge-enhanced ridership prediction demonstrate the effectiveness of RTSKG, highlighting its potential to support city-level rail transit station analysis.
[AI-5] SkillZip: Evaluation-Free Skill Compression for Self-Evolving Agents by Discovering Reusable Structure
链接: https://arxiv.org/abs/2608.11079
作者: Xiaofan Bai,Hongqiang Lin,Chao Liu,Yantao Zhang,Xuan Jin,Xipeng Cao,Yuhong Li
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Self-evolving agents accumulate reusable skills by appending successful procedures and failure fixes. Over time, the same requirement is often restated in several branches, examples, and warnings, while common action sequences are copied rather than reused. The resulting skill becomes expensive to inject and difficult to maintain. Generic prompt compression is ill-suited to this setting because a skill is not a flat passage: its name and description define when it applies, its workflow controls execution, its tool and output contracts constrain validity, and rare exceptions may remain essential even when no sampled task activates them. Evaluation-guided compression can test these behaviors, but it introduces rollouts, cost, and dependence on the compression-time evaluation set. We present SkillZip, an evaluation-free method that compresses a skill by finding its shortest faithful structural explanation. The intuition is explain once, reference many: state a repeated rule once at the scope where it applies, factor a repeated action sequence into a shared procedure, and keep only the differences as explicit exceptions. We formalize this intuition as a typed minimum description-length objective over a skill contract and a residual, subject to a hard coverage constraint for every extracted trigger, workflow edge, tool requirement, obligation, and output field. The formulation provides simple sharing thresholds, preserves unique rare rules by construction, and supports efficient local updates. SkillZip has a one-shot mode with one structured extraction call and deterministic optimization, and a continual Zip-on-Write mode that integrates each self-evolution patch without replaying tasks or reparsing the full history. Through comprehensive experimental evaluations, we demonstrate the effectiveness and superiority of SkillZip in compression performance, generalizability, and cost overhead.
[AI-6] V-FiLLM : Verified Financial LLM Reasoning Benchmark
链接: https://arxiv.org/abs/2608.11047
作者: Alicia Larsen,Victoire Laurent,Aulia Kharis Rakhamsari,Lara Turgut,Nino Antulov-Fantulin
类目: Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE); Machine Learning (cs.LG)
备注: 10 pages, 6 tables, 2 figures, under review
Abstract:While existing benchmarks have made substantial progress in evaluating LLMs across STEM domains, financial reasoning over structured data remains comparatively less explored. We introduce V-FiLLM, a framework that generates financial reasoning benchmarks from executable computation trees grounded in real tables, yielding items whose answers are correct by construction. Trees are evaluated symbolically to obtain ground truth and rendered into natural-language questions, removing any model from the labeling loop, so items can be generated at arbitrary scale without annotation cost and without inheriting a generator’s error rate. V-FiLLM exposes four independently controllable axes of difficulty including computation depth, expression breadth, financial concept complexity, and context size. By evaluating on open-source models, we find that accuracy falls up to 51% as reasoning depth increases, and up to 47% points under adversarial numerical perturbations, highlighting remaining challenges in robust financial reasoning over tables. We further show that lightweight LoRA fine-tuning on verified chain-of-thought traces improves accuracy from 81.1% to 85.6% on held-out problems and outperforms the base model by 5% points on FinQA (Chen et al., 2022a), s), suggesting that targeted, low-cost adaptation is a promising direction for compositional reasoning in financial QA.
[AI-7] Workflow Cards: Structured Summaries of Workflow Executions Using Provenance Data
链接: https://arxiv.org/abs/2608.11022
作者: Nicola Giuseppe Marchioro,Gabriele Padovani,Amal Gueroudji,Rafael Ferreira da Silva,Wesley Brewer,Valentine Anantharaj,Sandro Fiore,Renan Souza
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Artificial Intelligence (cs.AI)
备注: Accepted at eScience2026
Abstract:Model Cards and Data Cards have demonstrated the value of structured, human-readable documentation for machine learning artifacts, capturing their context, parameters, limitations, and intended use. However, these practices remain focused on static artifacts (the datasets and trained models themselves) while overlooking the workflow executions that produce, transform, and evaluate them. Such executions hold critical details about data preparation, parameter choice, runtime behavior, resource use, and intermediate transformations, precisely where bias, performance variation, and reproducibility gaps tend to originate. To close this gap, we introduce Workflow Cards: structured summaries that condense the machine-readable provenance data of a workflow execution into a form both humans and large language models (LLMs) can read and analyze. This paper has two main parts. First, it defines a Workflow Card template informed by a representative set of provenance questions that surface from the execution-level data missing from Model and Data Cards. Second, it evaluates how effectively LLMs use Workflow Cards to understand workflow executions compared with querying provenance databases through a schema-based interface. Results show that Workflow Cards provide execution-level information absent from existing card types, such as Model Cards and Data Cards, thereby filling an important documentation gap; and that Workflow Cards nearly double answer quality compared with schema-based querying, consistently across LLM-as-a-Judge and human assessments.
[AI-8] Policy Convergence and Divergence Across National and Within Regional AI Strategies: A Policy Design Element Analysis
链接: https://arxiv.org/abs/2608.11006
作者: Benjamin Faveri,Brie Bhasin
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
备注:
Abstract:Governments worldwide have responded to the rapid expansion of AI by publishing national and regional AI strategies. Comparing national and regional AI strategies to identify their convergences and divergences can uncover their common practices, understand regional variations, and provide policy designers a comprehensive set of policy design elements for their ongoing AI strategy developments. Yet, existing work has not examined their underlying policy design elements or assessed whether those elements are horizontally (country-to-country) or vertical (region-to-country) converging or diverging over time. This paper addresses that gap by coding and analyzing 74 national and 3 regional AI strategies drawn from a global scan of all 205 UN member and non-member states. The coding used a latent-inductive approach organized around three functional policy design elements: goals, approaches, and principles. Two research questions guided the analysis: to what degree are national AI strategies becoming horizontally convergent or divergent over time; and to what degree are national strategies becoming vertically convergent or divergent with those countries’ regional AI strategy. Results indicate strong horizontal convergence around economic competitiveness, research support, and ethical AI use, alongside persistent divergence in human rights goals, participatory governance approaches, and human-centric principles. Across the three regions, the AU exhibits the highest vertical convergence, the EU demonstrated strong alignment on regulatory and economic priorities but diverges on human-centric values, and the Nordic-Baltic Region displays mixed vertical convergence. These findings offer policy designers a comprehensive evidence base for identifying emerging AI policy design choice norms as AI strategies are developed and updated.
[AI-9] XCoT-VLA: Executable Chain-of-Thought for Vision-Language-Action Driving
链接: https://arxiv.org/abs/2608.10976
作者: Foundation Model Team,XPeng Inc
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Vision-Language-Action (VLA) models can connect scene understanding, semantic reasoning, and trajectory generation for autonomous driving. However, verbose natural-language Chain-of-Thought (CoT) is poorly suited to real-time control because it is open-ended, costly to decode, and difficult to optimize as an action-facing representation. We propose XCoT-VLA, which replaces descriptive rationales with compact executable CoT tokens learned from automatically constructed Reason-Action supervision. Logged trajectories provide action evidence, while scene context supplies causal semantics. The predicted XCoT sequence remains in context and conditions fixed trajectory queries through shared multimodal self-attention. Deterministic token-function routing applies the Reason FFN to XCoT tokens and the Control FFN to trajectory queries for flow-matching trajectory generation. We further introduce XCoT Policy Optimization (XCPO) as an optional refinement extension in the same executable token space. XCoT-VLA reduces longitudinal ADE from 1.645 to 1.323 on a general-distribution set and lateral FDE from 1.616 to 0.648 in lane-change scenarios. By representing driving-oriented reasoning with only 2-6 executable XCoT tokens, our method substantially reduces autoregressive reasoning overhead and remains within the real-time planning budget. These results demonstrate that driving-oriented reasoning can be compact, executable, and directly connected to trajectory generation.
[AI-10] FedCGR: Federated Cross-Domain Generative Recommendation CIKM2026
链接: https://arxiv.org/abs/2608.10929
作者: Zhuodong Liu,Hugen Lv,Xiangyu Li,Bohan Guo,Peiyu Hu
类目: Artificial Intelligence (cs.AI)
备注: Accepted at CIKM 2026. 10 pages, 5 figures, 6 tables
Abstract:Cross-domain recommendation (CDR) transfers preference knowledge across related domains, but federated deployment makes cross-domain alignment difficult because the behavioral anchors that align item spaces, such as overlapping users and shared interaction signals, are often sparse, unavailable, or privacy-sensitive across clients. To address this tension, we revisit federated CDR as generation over a stable semantic item language. By representing items as discrete semantic ID (SID) sequences derived from public item-side metadata, cross-domain item alignment is induced by a shared vocabulary rather than by exchanging private interactions or aligning domain-specific embeddings. Directly federating SID-based generators, however, introduces two design constraints: the SID tokenizer must remain fixed to preserve cross-client token consistency, which creates a semantic-only bottleneck because local collaborative filtering (CF) signals cannot be globally shared or aligned; meanwhile, standard federated averaging can cause negative transfer under domain heterogeneity. To overcome these constraints, we propose FedCGR, a federated generative CDR framework that keeps the item language stable and makes adaptation explicit. FedCGR injects local CF evidence through a reliability-aware semantic interface and trains a prototype-personalized generator that selectively aggregates shared parameters according to domain relatedness while keeping domain-specific quantities local. Experiments on six Amazon cross-domain scenarios show that FedCGR consistently outperforms federated generative baselines and achieves competitive performance against strong sequential and federated CDR methods under both full-ranking and sampled evaluation protocols.
[AI-11] hinkRetrieve: Retrieval-Augmented Reasoning Traces for Test-Time Scaling
链接: https://arxiv.org/abs/2608.10928
作者: Vaibhav Singh,Soumya Suvra Ghosal,Sarvesh Gharat,Soumyabrata Pal,Ramasuri Narayanam,Dinesh Manocha
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large Reasoning Models (LRMs) improve performance by allocating additional inference-time compute to generate extended chain-of-thought reasoning. However, recent studies reveal that sequential test-time scaling often yields diminishing or even negative returns, as longer traces exhibit increased uncertainty, error compounding, and drift from the original problem. We propose ThinkRetrieve, a test-time scaling framework that augments the reasoning traces of LRMs with dynamically retrieved solved examples at each reasoning step. Given an external corpus of problems paired with step-by-step solutions, ThinkRetrieve retrieves relevant exemplars at each intermediate step and injects them directly into the thinking trace, providing the model with guidance on how to reason rather than merely what facts are relevant. Experiments across five reasoning models (1.5B–8B parameters) on GSM-8K, MATH-500, AIME 2025, and SciQ demonstrate that ThinkRetrieve consistently improves accuracy over standard test-time scaling, with relative gains of up to 60% on AIME 2025.
[AI-12] IO Factory: Simulating AI-Enabled Influence Campaigns at Scale
链接: https://arxiv.org/abs/2608.10920
作者: Lukasz Olejnik,Wenchao Dong,Jonas R. Kunst,Signe Riemer-Sørensen,Tobias Herb,Meeyoung Cha,Daniel Thilo Schroeder
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:We introduce IO Factory, an AI-driven framework for simulating information and influence campaigns as fully integrated, traceable processes. The threat of digital manipulation now extends beyond persuasive text from individual language models to AI swarms, i.e., persistent groups of coordinated agents that adapt to platform feedback and disguise organized campaigns as ordinary social interaction. Because such campaigns cannot be identified from isolated messages alone, they must be analyzed across a continuous spectrum of planning, platform action, exposure, interpretation, measurement, and adaptation. IO Factory represents this process inside a controlled simulated platform, linking actor roles, platform actions, exposure records, structured model-based evaluations, and configured changes in the simulated population. We implement the architecture and evaluate it across configurations of up to 100,000 agents. The results show that IO Factory executes campaign timelines at scale and produces inspectable evidence of exposure and measured movement in configured belief variables. By recording the actors, objectives, action constraints, exposure paths, and measurement rules used in each run, IO Factory supports reproducible research and red-team analysis of coordinated influence.
[AI-13] ComBodied Agents : a New Paradigm of Human-Centric Agent ic AI
链接: https://arxiv.org/abs/2608.10915
作者: Qianggang Ding,Xingyao Wang,Rui Feng,Zhibin Wang,Feixiang Wang,Kelong Mao,Hao Sun,Zhiyao Luo,Jiankai Tang,Lei Li,Jiadong Guo,Minheng Ni,Weicong Lin,Chenxi Yang,Hongxiang Gao,Zhenghua Chen,Yang Bai,Min Wu,Jun Cheng,Huazhu Fu,Dacheng Tao,Bang Liu
类目: Artificial Intelligence (cs.AI)
备注: 38 pages, 6 figures, 10 tables
Abstract:After an older adult misses a medication dose, a software agent can send another reminder and an embodied agent can bring the medication. Yet neither explains whether the person forgot, is confused, has side effects, or deliberately refused, nor what support is appropriate. This reveals a structural gap in Agentic AI: Digital Agents primarily transform software states, while Embodied Agents transform physical states; neither makes a person’s evolving state and agency the primary object of modeling, intervention, and evaluation. We introduce Combodied Agents, a human-centered paradigm that perceives, models, predicts, and supports individual human-state trajectories over time, using software tools, sensors, wearables, robots, and human services as action channels rather than end goals. We unify fragmented capabilities across personal assistants, health agents, AI companions, and adaptive human–AI systems into a closed loop: event-based multimodal perception reconstructs meaningful personal events; longitudinal, correctable memory provides temporal context; Personal World Models estimate future personal states and outcomes under alternative decisions and interventions; and an admissible intervention policy selects proportionate support under consent, uncertainty, safety, reversibility, and user control. Feedback from the person and environment updates the loop. Rather than requiring an exhaustive Human Digital Twin, the framework uses purpose-bounded, uncertainty-aware, user-correctable representations. We organize the design space by human-state targets, relational contexts, and agent roles, and propose scenario-centered evaluation, agency-preservation metrics, benchmark requirements, edge-native personal models, and governance directions. Combodied Agents shift Agentic AI from external task completion toward sustained human benefit.
[AI-14] GitSkills: A Dataset of Agent Skills on GitHub
链接: https://arxiv.org/abs/2608.10906
作者: Giuseppe Destefanis,Daniel Graziotin,Matteo Vaccargiu,Marco Ortu
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: Giuseppe Destefanis, Daniel Graziotin, Matteo Vaccargiu, and Marco Ortu. 2027. GitSkills: A Dataset of Agent Skills on GitHub. In Proceedings of the 24th International Conference on Mining Software Repositories (MSR '27). Association for Computing Machinery, New York, NY, USA, 3 pages. To appear
Abstract:An agent skill is a folder containing a this http URL file with instructions for a language-model agent, optionally accompanied by scripts and reference files. The agent loads the skill when it judges that a task matches the skill description. Anthropic introduced the format in October 2025 as an open specification. Nine months later, we find that skill files in the millions sit in public GitHub repositories. Skills are unlike the artifacts the SE research community usually mines: they are written mainly in natural language, a model selects them probabilistically at run time, and no compiler or type checker verifies the selection. They also have no central registry or package manager, so they spread by copying folders between repositories. How developers write, reuse, and maintain skills is therefore an empirical question, and no existing dataset records this population. We present GitSkills, a dataset of 3,797,117 this http URL files collected from 282,200 public repositories in July 2026. The dataset retains every file occurrence with its repository, path, and content hash. It groups identical files into 1,877,981 distinct contents and enriches one representative per group with the full text, parsed front matter, folder contents, repository metadata, and, for a subset, the commit history of the file. A single self- contained SQLite file supports research on the adoption, reuse, structure, authorship, maintenance, and security of agent skills.
[AI-15] Enhanced Filtering Algorithms for the Euclidean Traveling Salesperson Problem and its variants in Constraint Logic Programming
链接: https://arxiv.org/abs/2608.10881
作者: Alessandro Bertagnon,Marco Gavanelli
类目: Artificial Intelligence (cs.AI); Logic in Computer Science (cs.LO)
备注: Accepted for publication in Theory and Practice of Logic Programming (TPLP), 36 pages, 14 figures
Abstract:The Traveling Salesperson Problem (TSP) is one of the best-known problems in computer science and arises in many engineering applications, such as smart vehicles and intelligent transportation systems. In the “Euclidean” case, each node is defined by its coordinates in the plane and distances are computed using the Euclidean metric. In the Constraint Programming (CP) literature, the Euclidean TSP is typically addressed by computing the full distance matrix and treating it as a general case; however this approach ignores the geometric information carried by the points’ coordinates. In this work, we propose new filtering algorithms, implemented in Constraint Logic Programming (CLP), that exploit such geometric information to achieve stronger constraint propagation than existing approaches. Moreover, we show how this methodology can be extended to other Euclidean variants of the TSP, including the Euclidean Generalized Traveling Salesperson Problem (EGTSP), which is relevant in practical routing and logistics applications. Experimental results demonstrate the computational advantages of the proposed approach.
[AI-16] Hypothesis Frontier: Verifier Guided LLM and Symbolic Search for First-Order Induction
链接: https://arxiv.org/abs/2608.10843
作者: Serafim Batzoglou
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:First-order concept synthesis asks a system to infer one formula that classifies labeled objects consistently across several finite relational structures. Every candidate can be evaluated exactly, but quantified first-order formulas form a vast search space, and LLM outputs are often semantically promising without being fully correct. We introduce Hypothesis Frontier, a verifier-guided neurosymbolic framework that evaluates each LLM formula on every training object, retains the strongest verified hypothesis across rounds, and uses its remaining errors to guide subsequent generation. Symbolic processing repairs invalid formulas while remaining anchored to the LLM-generated hypothesis, and simplifies train-valid formulas without changing any training prediction. Under matched models, problem sets, and LLM-round budgets, Hypothesis Frontier solves substantially more problems than repeated original-prompt generation. After the final formulas are selected, exact simplification shortens many train-valid formulas while preserving every training prediction. Exact symbolic reasoning therefore helps both to solve more induction problems and to compress many of the resulting formulas.
[AI-17] ACTICL: Task-Aware Compression of Tabular ICL Models
链接: https://arxiv.org/abs/2608.10837
作者: Mykhailo Koshil,Matthias Feurer,Katharina Eggensperger
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Accepted for publication at AutoML2026
Abstract:The strong performance of foundation models for tabular tasks comes at substantial inference costs. Distilling models into task-specific architectures reduces model size and computational demands but also sacrifices in-context adaptability. Here we introduce TACTICL, an automated task-aware compression framework for tabular in-context learning models that jointly prunes transformer layers and replaces them with lightweight adapters trained on downstream tasks, thus blending in-context with in-weight learning. We study TACTICL on 47 benchmark datasets and show that we can substitute up to 85% of layers without substantial performance drop on a given downstream task. We further show that TACTICL maintains robustness to data shifts, leaving its in-context ability intact. Overall, TACTICL provides a robust framework for exploiting the depth-wise redundancy of tabular foundation models by combining task-specific adaptation and structured compression. We provide the code at: this https URL
[AI-18] Whisper-Aware LLM : Self-Supervised Uncertainty Learning for Robust Whispered Speech Recognition
链接: https://arxiv.org/abs/2608.10836
作者: Gaopeng Xu,Zhenyu Wang,Zheng Xue,Yinfeng Xia,Haitao Yao
类目: ound (cs.SD); Artificial Intelligence (cs.AI)
备注:
Abstract:The signal ambiguity of whispered speech drives ASR systems toward two opposing failure modes: failing to capture whispered speech or hallucinatory transcription of noise. This paper introduces the Whisper-Aware LLM, a framework that teaches an Audio-LLM to perceive and react to this uncertainty. Our model develops an intrinsic self-awareness by learning to quantify the physical deficiencies of acoustic signals through targeted self-supervised tasks. This learned uncertainty is then operationalized via a novel Confidence-Fused Decoding mechanism, which provides both high-level instructions and frame-level attention modulation to the LLM decoder. Our experiments confirm the effectiveness of this approach. The model sets a new state-of-the-art on whispered speech with a 17% relative CER reduction on AISHELL6-Whisper. At the same time, it directly addresses the reliability trade-off, with hallucination rates dropping from over 25% to 4.5%.
[AI-19] EvoMem: Memory-Augmented Evolution for Code Optimization
链接: https://arxiv.org/abs/2608.10795
作者: Viktor Volkov,Valentin Khrulkov,Andrey V. Galichin,Danil Sivtsov,Nikita Glazkov,Olga Volkova,Konstantin Pchelin,Iaroslav Bespalov,Dmitry V. Dylov,Petr Anokhin,Ivan Oseledets
类目: Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE)
备注:
Abstract:Successful mutation strategies in evolutionary code search may contain reusable knowledge that is useful beyond a single run, and in some cases may transfer across related tasks and domains. However, existing LLM-driven evolutionary frameworks largely discard such knowledge, repeatedly rediscovering similar ideas and limiting opportunities for cross-run and cross-task learning. We introduce EvoMem, a persistent memory architecture for LLM-based evolutionary program search that captures and reuses candidate mutation knowledge. EvoMem converts successful mutation events into structured, task-aware advice for future runs. It operates in two phases: after each run, it extracts and stores promising ideas with provenance, and during subsequent evolution, it retrieves a small set of relevant instructions based on the current task and program context to guide mutation. Across geometric optimization, multi-hop question answering, GPU kernel optimization, and related benchmarks, our experiments show positive average improvements in target metrics or search speed for most evaluated settings, while also revealing variability across tasks. Overall, EvoMem provides evidence that persistent memory can reduce some redundant exploration and improve the reuse and adaptation of successful strategies in LLM-driven evolutionary search.
[AI-20] ChemWorld: Programmable Chemical Worlds for Controlled and Replayable Agent Experimentation
链接: https://arxiv.org/abs/2608.10792
作者: Jiangjie Qiu,Yijun Li,Xiaonan Wang
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Autonomous chemistry increasingly depends on environments in which agents can repeatedly act, observe, and this http URL laboratories provide essential real-material evidence but are costly to repeat and difficult to use for tightly matched interventions, whereas most digital environments keep the underlying experimental world largely fixed. We introduce ChemWorld, a programmable chemical environment in which reusable process and observation components are compiled into executable worlds. ChemWorld separates the public experimental contract available to an agent from evaluator-owned chemical and material laws. Researchers can therefore vary world composition and operating conditions, or change a single hidden law while holding the public task and interaction conditions fixed. Transactional execution records operations, failures, resource changes, and state transitions, allowing complete environment-action trajectories to be replayed exactly and audited. Full-census qualification covered the reference registry, 52 generated compositions, and module, interface, compilation, and invalid-action tests. Eight deterministic experimental cases demonstrated shared lifecycle semantics, failure recovery, and exact replay, while six parent-child world-fork pairs isolated the effects of single private-law interventions under matched public conditions. An independent agent also completed a full lifecycle in a non-reference world through the same public interface. Within the declared component and model domain, ChemWorld provides a controlled and replayable substrate for studying experimentation across systematically varied chemical worlds, complementary to physical-laboratory evidence and calibration.
[AI-21] SkillLens: Visual Skill Cards for Retrieval-Augmented GUI Action Prediction and On-Policy Distillation
链接: https://arxiv.org/abs/2608.10775
作者: Zhou Liu,Ligang Huang,Zeli Su,Zewei Pan,Zhaoyang Han,Xing Chen,Yuanfeng Song,Wentao Zhang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Computer-using agents can perceive rich software interfaces, yet their decisions often lack visual procedural memory: they may recognize individual controls without identifying which familiar workflow is active, which control matters next, or what evidence would confirm progress. Raw interaction traces preserve such information but are long and noisy to condition on, whereas text-only skills often omit the visual state that makes a procedure applicable. We introduce Visual Skill Cards (VSCs), a state-conditioned memory representation that binds reusable procedures with applicability cues, visual evidence, and verification signals. SkillLens constructs VSCs from heterogeneous interaction experience through Trace-to-Visual-Skill-Card and, at inference time, retrieves relevant cards and selectively expands only the evidence needed by a fixed visual-language model executor for grounded GUI action prediction. The same representation also supports CardDistill, which uses VSC evidence as privileged teacher context to train a student that acts without runtime card retrieval. Across Multimodal-Mind2Web and WebLINX-BrowserGym, SkillLens improves the frozen GPT-5.4-mini executor by +11.6 points in Step SR and +2.9 points in Overall, respectively; CardDistill further improves the corresponding student-only Qwen3-VL-2B metrics by +12.0 and +3.2 points.
[AI-22] he GenAI Catch-22: Use of Generative Artificial Intelligence in Norwegian Newsrooms During the 2025 Parliamentary Election
链接: https://arxiv.org/abs/2608.10773
作者: Mari Reisjå,Anders Sundnes Løvlie
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
备注:
Abstract:The increasing use of Generative Artificial Intelligence (GenAI) in journalism raises concerns about possible detrimental effects both on journalism and its democratic function. We explore these risks through a case study of GenAI in Norwegian Newsrooms during the 2025 parliamentary election campaign. Based on interviews with managers and journalists over a ten-month period, we analyse how ambitious visions fared in the face of technological and practical challenges. We highlight the risk of an internal threat stemming from the journalists’ own use of AI, contrasting the dominant focus on external disinformation threats. We show how newsroom managers shared sociotechnical imaginaries resulting in unrealistically optimistic beliefs about the capabilities of the technology and the pace of development, leading to plans for audience-facing GenAI services collapsing and giving way to more mundane uses of GenAI tools internally in the newsrooms. Furthermore, we identify a vulnerability in the newsroom’s resilience against GenAI influence: a GenAI Catch-22. In order to monitor the GenAI tools and prevent errors and undue influence, newsrooms rely on human expertise. But by using GenAI extensively, the newsrooms risk a deterioration of human expertise, preventing them from monitoring the GenAI systems adequately.
[AI-23] Rule of Thumb: Explaining Artificial Intelligence Systems using Partial Information
链接: https://arxiv.org/abs/2608.10766
作者: Kaivalya Rawal,Daria Onitiu,Brent Mittelstadt,Sandra Wachter,Chris Russell
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Machine Learning (stat.ML)
备注: Code available at: this https URL
Abstract:Explainable Artificial Intelligence (XAI) seeks to explain how an Artificial Intelligence (AI) system arrived at a particular decision. We propose ‘‘Rule of Thumb’’ (RoT) explanations, a new approach to XAI based upon a novel formulation that identifies the most relevant features for predicting the behaviour of an AI system, for a particular datapoint. We show how RoT is well-suited to enable XAI in: (a) zero-shot classification using large language models (LLMs), (b) auditing of opaque AI systems without model access, and © the use of AI in scientific discovery. Additionally, RoT meets specific requirements from leading AI regulations, provides a familiar interface and visualisations for XAI practitioners, is model-agnostic, and is substantially faster than alternatives. Code available at: this https URL Comments: Code available at: this https URL Subjects: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Machine Learning (stat.ML) Cite as: arXiv:2608.10766 [cs.AI] (or arXiv:2608.10766v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2608.10766 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-24] A Gateway Architecture for Enterprise MCP Authentication: Unifying Heterogeneous Auth Identity Delegation and the User / Non-User Persona Problem
链接: https://arxiv.org/abs/2608.10760
作者: Suraj Kumar,Amy Wang,Srinivasan Manoharan
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:The Model Context Protocol (MCP) has become the de-facto interface for connecting LLM agents to enterprise tools, and adoption has been explosive: within a year, large organizations went from zero to dozens of internally built MCP servers. That speed created a governance crisis. Each team implemented authentication independently – some with no auth, some with API keys, some with full OAuth – producing a fragmented landscape with no consistent way to authorize callers, track who did what, or offboard a departing employee across the fleet. This paper reports an industry deployment that resolves the crisis with a centralized MCP gateway: a single aggregation, governance, and authentication layer that fronts every downstream MCP server. We make four contributions grounded in production experience. First, a two-axis authentication model crossing persona (interactive user vs. automated non-user) with credential type (no-auth, static/dynamic API key, PKCE, client credentials, platform app-context). Second, a gateway authentication layer supporting three enterprise SSO grants and three token-provisioning models: Bring-Your-Own-Token, Generate-Your-Own-Token, and delegated OAuth via RFC 8693 token exchange. Third, three end-to-end identity flows – User-to-OAuth2, Non-user-to-Service-Account, and User-to-Service-Account – composing client, gateway, and server. Fourth, the deployment evolution from CDN/WAF/edge perimeter to private MCP tunnels and enterprise-wide connectors. The architecture is in production, fronting dozens of MCP servers across web, desktop, custom-SDK, and low-code clients. Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI) Cite as: arXiv:2608.10760 [cs.CR] (or arXiv:2608.10760v1 [cs.CR] for this version) https://doi.org/10.48550/arXiv.2608.10760 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-25] ree-of-Ideas: Automated Research Ideation via Cross-Trajectory Reasoning over Scholarly Evolution
链接: https://arxiv.org/abs/2608.10740
作者: Xun Li,Yiying Yang,Pengtao Li,Xiao Yao,Suyu Liu,Xiaoyang Ye,Ziyu Lu,Yuan Yao,Yangning Li,Yinghui Li,Wenhao Jiang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Effective research ideation requires moving beyond a static understanding of prior work to trace how research problems and solutions evolve across the literature. Existing methods either treat papers as unstructured context or model scholarly evolution as isolated citation chains, overlooking interactions among research trajectories. We propose Tree-of-Ideas (ToI), a two-stage framework. EvoTrace reconstructs branching scholarly trajectories from citations, tracking evolving methods, resolved problems, and gaps. EvoAgent then reasons across trajectories to identify convergent problems and complementary solutions, generating grounded research ideas. Across six AI research topics, ToI achieves the highest score among automatic methods (6.27 vs. 5.36 for the strongest baseline on a 10-point scale), with strong Novelty (6.36) and Groundedness (7.00). Also, its score approaches that of human-paper references (6.29), demonstrating the value of cross-path evolutionary reasoning.
[AI-26] Smart Enough to Go Extinct? An Evolutionary Challenge to the Value of General Intelligence and Its Ethical Implications for AGI
链接: https://arxiv.org/abs/2608.10730
作者: David Klotz
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
备注: 22 pages
Abstract:The pursuit of artificial general intelligence (AGI) rests on a seemingly self-evident premise: that general intelligence, the kind of flexible, domain-general cognitive capacity exemplified by Homo sapiens, is extraordinarily valuable. This paper subjects this premise to critical scrutiny. We first present the intuitive case for the value of general intelligence before mounting an evolutionary challenge. We argue that, on evolutionary timescales, its adaptive value is far from empirically established. Numerous taxa, from cyanobacteria to horseshoe crabs, have persisted for hundreds of millions or even billions of years without anything resembling general intelligence, while Homo sapiens has existed for roughly 300,000 years and already faces self-generated existential risks. Mass extinction events do not preferentially favour cognitively sophisticated species. We argue that general intelligence may be the only biological strategy that generates existential threats to the species possessing it, an existential risk paradox with no parallel among non-intelligent survival strategies. Unlike prevailing accounts of AI risk that trace the danger to misalignment, we locate it in structural features of general intelligence itself, implying that even well-aligned AGI would inherit this liability. If the long-term evolutionary value of general intelligence is uncertain or negative, this raises ethical questions about engineering AGI and, more urgently, creating artificial consciousness. Drawing on deontological ethics and the precautionary principle, we argue that this uncertainty imposes a duty of caution: if we create a new kind of intelligent being, we bear responsibility for ensuring the conditions under which it can flourish.
[AI-27] Optimal Stopping of Self-Refining Foundation Models
链接: https://arxiv.org/abs/2608.10729
作者: Kim Hammar,Tansu Alpcan,Emil C. Lupu
类目: ystems and Control (eess.SY); Artificial Intelligence (cs.AI)
备注: Accepted at 65th IEEE Conference on Decision and Control (CDC 2026)
Abstract:Foundation models can improve their outputs through a self-refinement process driven by external feedback. In this process, the model is embedded in an iterative loop where it generates outputs, receives feedback from verifiers, and refines its responses through in-context learning. Following a novel approach, we formalize this process as an optimal stopping problem where the number of refinement iterations is decided based on expected improvement relative to cost. We derive optimal stopping policies and show that they can be efficiently computed through stochastic approximation. To evaluate our approach experimentally, we apply it to a coding benchmark for foundation models. The empirical results show that our stopping policies are significantly more cost-efficient than stopping policies proposed in prior work.
[AI-28] ProTAGAD: A Foundation Model for TAG Anomaly Detection with Decoupled Topological and Textual Prototypes
链接: https://arxiv.org/abs/2608.10699
作者: Ziyan Wang,Liwen Wu,Cheng Xie,Song Gao,Zhenli He,Xin Jin
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Text-Attributed Graphs (TAGs), endowed with abundant textual content along with topological structures, have emerged as a versatile backbone for real-world anomaly detection spanning large language model security, social network moderation, and cyber threat identification. Unlike conventional Graph Anomaly Detection (GAD), which relies primarily on structural irregularities, TAG anomaly detection must jointly leverage both topological patterns and fine-grained textual semantics to capture nuanced anomalous behaviors. The current GNN-based anomaly detectors adopt holistic message-passing schemes that indiscriminately fuse structural proximity and textual semantics during propagation, leading to deep cross-modality coupling. This entanglement acts as a noise amplifier, obscuring subtle anomalous signals and directly giving rise to the Blurred-Anomaly-Boundary (BAB) issue by rendering normal-anomalous decision boundaries poorly separable. This challenge is further amplified for graph foundation models that require robust cross-domain generalization. To bridge this gap, we introduce a novel foundation model for TAG anomaly detection featuring decoupled topological and textual prototypes. Our framework constructs dual prototype banks to independently model structural normality and semantic consistency, effectively isolating anomaly cues that are otherwise diluted during coupled aggregation. Extensive experiments across 14 diverse benchmark datasets demonstrate that our method consistently achieves state-of-the-art performance in cross-domain settings. Notably, the ablation studies further corroborate the prevalence of the BAB issue in conventional coupled TAG anomaly detectors, and show that our decoupled prototype design effectively mitigates this challenge.
[AI-29] Self-Correcting Long-Horizon Search Agents via Tree-Structured Memory
链接: https://arxiv.org/abs/2608.10676
作者: Aijun Yang,Qianxue Guo,Ziyi Huang,Yuxuan Chen,Shiyou Qian,Jian Cao
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large language model (LLM)-based search agents answer questions through multi-step interactions with external environments. However, providing complete execution trajectories to the LLM causes unbounded context growth and introduces noise. Existing compression methods reduce context at the cost of important details and often replace erroneous facts without repairing downstream reasoning derived from them. To address this problem, we propose ReTree, a self-correcting tree-structured memory mechanism for search agents. ReTree constructs a bounded per-step reasoning context while preserving source-linked evidence. It models search as an evidence tree whose nodes store bounded summaries, evidence, and revision histories. When newly retrieved evidence contradicts an earlier claim, ReTree traces back to the node where the claim was introduced, replaces outdated evidence, regenerates summaries, prunes affected branches, and resumes search. Source-grounded evidence provenance supports reliable conflict localization and keeps final claims traceable to retrieved passages. Experiments on four public question-answering and search benchmarks show that ReTree consistently outperforms Full-Trajectory ReAct, improving answer accuracy by up to 25.6 percentage points (pp); the average maximum per-step reasoning context of Full-Trajectory ReAct is 1.27 – 1.51\times that of ReTree. These results establish ReTree as an effective self-correcting memory abstraction for long-horizon search.
[AI-30] REDAgent Bench: Executable Red Teaming and Faithful Measurement of LLM Agent Systems
链接: https://arxiv.org/abs/2608.10669
作者: Zixing Chen,Xingyuan Liu,Jie Zhu,Huaixia Dou,Shuo Jiang,Junhui Li,Lifan Guo,Feng Chen,Chi Zhang
类目: Artificial Intelligence (cs.AI)
备注: 6 figures, 4 tables. Supplementary material included
Abstract:Large language model (LLM) agents combine language-based reasoning with external tools to perform complex tasks. Adversarial inputs can exploit interactions between the agent and its environment, causing the agent to violate safety policies during execution. Yet existing evaluations often reduce agent safety to a single attack success rate (ASR), collapsing exposure, execution, observation, and adjudication and potentially conflating actual violations with evidence visibility. We introduce REDAgentBench, an executable framework for autonomous red-teaming and faithful measurement. It derives attacks from explicit safety constraints and associated agent-system vulnerabilities, runs them in isolated service sandboxes, and verifies harmful effects from service receipts and final-state changes. The benchmark contains 1,661 cases across five service surfaces. Across six models and three agent harnesses, macro-average ASR is 65.69%; reported ASR varies with harness and evidence view, while evaluation-context disclosure changes execution behavior. In a state-grounded diagnostic cohort, almost one in five confirmed violations with resolved action anchors occurs after the agent states the relevant constraint or risk, revealing a Recognition–Execution Gap. Finally, a training-free policy reminder reduces confirmed violations by more than 70 percentage points in matched replay. These findings show that executable evaluation can improve safety measurement and identify actionable intervention points.
[AI-31] FITTER: Vocabulary-Agnostic Cross-Domain Inference on Temporal Knowledge Graphs ISWC2026
链接: https://arxiv.org/abs/2608.10668
作者: Jiaxin Pan,Mojtaba Nayyeri,Osama Mohammed,Daniel Hernandez,Rongchuan Zhang,Cheng Cheng,Steffen Staab
类目: Artificial Intelligence (cs.AI)
备注: Accepted at ISWC 2026
Abstract:Temporal knowledge graphs are central to many uses of the Semantic Web, but existing completion methods assume the entities, relation names, and timestamps to be reasoned about are already known at training time, restricting each model to a single graph and vocabulary. We propose FITTER, the first fully-inductive structural model for temporal knowledge graph link prediction that supports cross-domain transfer: the inference graph may contain entirely unseen entities, relation names, and timestamps drawn from a different domain. FITTER represents each predicate by its interaction patterns with others and time through encodings of relative rather than absolute ordering; message-passing fuses local and global temporal context to produce vocabulary-agnostic embeddings. We prove the temporal encoding is time-shift invariant and evaluate FITTER on cross-domain, cross-graph transfer over six temporal knowledge graph benchmarks of diverse domains, granularities, and time spans. FITTER consistently outperforms inductive baselines without retraining, indicating that vocabulary-agnostic structural learning is a viable foundation for inference over the heterogeneous knowledge graphs of the Semantic Web.
[AI-32] Operationalising Relative Causal Knowledge: Backbone Identifiability from Private Reports on a Shared Outcome UAI2026
链接: https://arxiv.org/abs/2608.10664
作者: Fabrizio Russo,Mark Somers
类目: Artificial Intelligence (cs.AI)
备注: Accepted at Causal Decision Making Workshop, UAI2026. 4 pages + Appendix (13 total)
Abstract:The Relativity of Causal Knowledge (RCK) explains how a network of agents with different structural causal models can exchange causal knowledge through a shared interventionally consistent abstraction, or backbone. We ask the prior identification question that this transport mechanism presupposes: when is that backbone determined by the agents’ private causal knowledge? In the basic two-agent common-effect case, two private causes influence one shared outcome and each agent identifies only the single-cause causal marginal relevant to its own perspective. We show that, under standard compatibility, non-degeneracy, and local overlap assumptions, those local causal marginals do not identify a unique backbone. Infinitely many joint intervention kernels can induce exactly the same private reports while disagreeing on joint interventions. We then give a conditional recovery result. Additive separability removes the hidden interaction degree of freedom, but observational residual summaries remain insufficient. Identification becomes possible when agents communicate causally identified response functions. An education value-added example illustrates why this is first a communication problem, and only then a policy-composition problem.
[AI-33] Decision-Aware Approximation of Belief Functions for Evidential Combinatorial Optimization
链接: https://arxiv.org/abs/2608.10650
作者: Sohaib Afifi
类目: Artificial Intelligence (cs.AI); Optimization and Control (math.OC)
备注: 9 pages, 1 figure, 3 tables. Accepted at BELIEF 2026
Abstract:Reducing the number of focal elements of a mass function is classically driven by an intrinsic distance, such as Jaccard or Jousselme, that keeps the approximation close to the original as a body of evidence. We consider instead the case where the mass function feeds a linear combinatorial optimisation problem with evidential costs. What should then be preserved is not the closeness of the two mass functions, but the quality of the decision they induce. We introduce a decision-aware approximation that targets the regret of the decision: one decides with the cheaper approximation and is evaluated under the true mass function. On a minimal shortest path, the distance-optimal approximation flips the decision while a decision-aware merge preserves it, and this occurs on a non-negligible fraction of random instances. We prove a one-point bound that localises the regret at the true optimum, turn it into an exact dynamic program for the scalar case, and extend it to an online version that prunes focal elements before the final cost is known. In experiments the decision-aware compressor flips the decision less often than representation-aware compression, for both the linear criterion and a non-linear proxy read-out.
[AI-34] Curate Before You Connect: Identity and Ontology Tagging in a Production Knowledge Graph
链接: https://arxiv.org/abs/2608.10644
作者: Vaibhav Dangaich,Kevin Lewis,Kundeshwar Pundalik
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Extraction produces candidate entities and relationships; writing them into a graph is where identity is decided, and identity decisions are destructive in a way extraction errors are not. A wrong type can be corrected later, but two records merged under one identity cannot be separated once their properties have been combined, and the merge leaves no error behind. This paper describes the ingestion and ontology-tagging layer that turns a validated extraction stream into a knowledge graph of 537,157 entities and 2,198,567 relationships drawn from 98,795 government documents. We describe a record-identity ladder that decides sameness from identifier columns, name columns, display names and type-scoped position rather than from name similarity. The ladder governs de-duplication within parsed tables, while the graph write applies a coarser canonical-name key, so records sharing a canonical name merge automatically on exact equality. We argue rather than demonstrate that this is where the automation line belongs: no identity benchmark is reported, and the over-merges the key permits are undetectable by construction. That policy, under which entity resolution only ever flags candidates, followed an incident in which two surface forms of one name were merged, corrupting a correct record and deleting eight entities from an unrelated document. We then describe multi-class ontology tagging and an evidence asymmetry we did not anticipate: an entity name is an instance label rather than a type assertion, so matching name fragments against a class index invents classifications. Requiring anchored evidence cut role assignments on an enriched sample from 36 to 4, all confirmed correct. We quantify the graph’s conformance debt, show secondary classifications compensating for a mis-parented primary class, and describe a curation queue grown to 48,403 pending proposals against 775 human decisions.
[AI-35] Compute-Optimal Is Not Cluster-Optimal: Systems-Aware Scaling for Sparse Mixture-of-Experts
链接: https://arxiv.org/abs/2608.10605
作者: Soumajyoti Sarkar,Yuxin Tang,Sheng Zha
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:In large-scale pretraining, the algorithm, architecture, and systems decisions are conventionally made in disconnected stages. A scaling law stage selects an architecture and training recipe, optimizing loss under compute constraints, and a separate systems stage then optimizes the implementation for hardware efficiency. In this work, we develop MOSAIC, which formulates model architecture and systems co-design as an optimization problem. MOSAIC couples a predictive scaling law with a calibrated performance model that estimates Model FLOPs Utilization (MFU), communication cost, memory footprint, and the best parallel layout. We instantiate the framework for sparse Mixture-of-Experts (MoE) language models, where expert count, routing sparsity, and other MoE layer dimensions affect both the loss and systems efficiency. We fit a scaling law on sparse MoE models trained on text data, whose scaling dimensions include the sparsity factor, which is the fraction of model parameters inactive per token in a forward pass. The scaling law sweeps in our work span active parameters from 104 million to 2.7 billion and total model sizes reaching 79 billion parameters. We show that, within the calibrated sparsity range, an efficiency-agnostic model-FLOPs budget admits no interior optimal sparsity. The fitted loss decreases monotonically with sparser models and the compute optimum lies at the upper boundary of the data support. An optimal sparsity in MoE models instead emerges under the cluster’s systems constraints, as captured by MOSAIC. Our results argue for a shift towards unified architecture and systems co-design for frontier language model training.
[AI-36] Inferential Capability Does Not Determine Legal Scope
链接: https://arxiv.org/abs/2608.10601
作者: Nicola Fabiano
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
备注:
Abstract:Two instruments of EU digital law place inference at their centre and mean different things by it. Article 3(1) of the AI Act uses the capability to infer constitutively: it is the central feature separating the regulated category from conventional software. The GDPR never defines inference, yet governs it protectively: the consequences follow from the processing of personal data and from what the inference says about, or does to, a person, whether or not the technology that produced it qualifies as an AI system. The two perimeters are not concentric. Their non-coincidence remained invisible in single-shot systems; agentic architectures make it operationally acute. The thesis: inferential capability does not determine legal scope, and its absence does not create immunity. The framework is two-level. Inference performs two legal functions, constitutive and protective; the protective function operates through three pathways - identificatory, attributive and decisional. Composition is not a fourth pathway but a cross-cutting architectural dimension which, with reach, persistence and reviewability, is what agentic architectures modify. Three concepts support it: the inferential threshold, the inferential reach and the inferential chain, mapped onto the chain of imputation. Regulation (EU) 2026/1744 left the constitutive criterion untouched and inserted a provision contemplating outputs that influence the inputs of future operations, without supplying any rule of aggregation. The article proposes an interpretive rule, a compositional-effects test identifying the decision unit under Article 22 GDPR together with the allocation of the burden of establishing it, and documentation duties calibrated to inference chains. Subjects: Computers and Society (cs.CY); Artificial Intelligence (cs.AI) Cite as: arXiv:2608.10601 [cs.CY] (or arXiv:2608.10601v1 [cs.CY] for this version) https://doi.org/10.48550/arXiv.2608.10601 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Nicola Fabiano [view email] [v1] Tue, 11 Aug 2026 07:39:53 UTC (33 KB)
[AI-37] HexEval: An Evidence-Driven Hexagonal Framework for Multidimensional Scholar Assessment
链接: https://arxiv.org/abs/2608.10584
作者: Xiaokang Qu,Yiting Lin
类目: Artificial Intelligence (cs.AI); Digital Libraries (cs.DL)
备注: 9 pages, 3 figures
Abstract:Scholar assessment plays a fundamental role in faculty recruitment, funding allocation, academic promotion, and talent discovery. Existing scholar assessment methods predominantly rely on bibliometric indicators and reputation proxies, while recent large language model (LLM)-based approaches mainly focus on evaluating individual research papers rather than comprehensively assessing scholars. We argue that scholar assessment should be formulated as an evidence-driven reasoning problem that jointly considers intrinsic research quality and externally verifiable scholarly behavior. To this end, we propose HexEval, an evidence-driven hexagonal framework for multidimensional scholar assessment. HexEval explicitly organizes scholar assessment into two complementary evidence layers. The intrinsic layer evaluates anonymized representative works along three dimensions, namely research rigor, methodological innovation, and scientific contribution, whereas the external layer characterizes scholars through knowledge translation, research coherence, and academic impact using heterogeneous evidence collected from GitHub, Lens, OpenAlex, and other publicly verifiable sources. Instead of producing opaque aggregate scores, HexEval preserves intermediate evidence, dimension-specific rationales, and verification signals throughout the evaluation process, enabling interpretable and auditable scholar profiles. Experiments across all six dimensions show dimension-dependent agreement with human or external reference criteria: structured calibration improves absolute agreement for intrinsic quality, while the external modules recover broad trajectory and ordinal impact signals. These results support evidence-driven reasoning over heterogeneous scholarly evidence as a promising paradigm for auditable AI-assisted scholar assessment, while exposing the coverage and attribution limitations of public scholarly data.
[AI-38] Agent ic Instruction Data Selection: Let DataMaster Interpret Your Intent
链接: https://arxiv.org/abs/2608.10579
作者: Fanqi Zhou,Qiaosheng Chen,Zixian Huang,Gong Cheng
类目: Artificial Intelligence (cs.AI)
备注: 21 pages, 3 figures
Abstract:Although existing instruction data selection methods have introduced various metrics, the inherent complexity of real-world datasets makes it impractical for any single metric to generalize across all scenarios. Developers are thus often forced to manually inspect data and craft heuristic rules for each new application—a tedious and error-prone process. In this paper, we propose a paradigm shift from manual configuration to automated orchestration via the Instruction Data Selection Agent (DataMaster), which interprets user intent and autonomously composes optimal selection strategies. By allowing users to specify data needs through natural language descriptions, DataMaster simplifies data curation and removes the burden of manual strategy design. Extensive experiments across the math, medical, and code domains show that DataMaster outperforms static baselines in most settings and surpasses full-pool training in a substantial number of cases. The implementation of DataMaster and the scripts needed to reproduce the reported pipeline are publicly available at this https URL.
[AI-39] DashArena: Benchmarking LLM s on Interactive Analytic Dashboard Generation
链接: https://arxiv.org/abs/2608.10567
作者: Xiaotong Wang,Dazhen Deng
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Analytic dashboards combine coordinated views and interactions for data exploration and decision-making. Recent models can generate them from data and natural-language goals, but evaluating their usefulness remains difficult. Dashboard generation is open-ended, and neither static appearance nor successful execution alone captures analytical support and interaction quality. We introduce DashArena, to our knowledge the first benchmark for open-ended, task-grounded generation of interactive analytic dashboards. Its key innovation is to require each system to generate both a dashboard and a replayable interaction trajectory. A browser executor replays the trajectory and turns the system’s intended analytical workflow into reproducible visual and execution evidence. A VLM judge compares candidates using this evidence, and Bradley–Terry aggregation produces the leaderboard. We further distill the judge into the open-weight DashJudge-8B. Human evaluations show that DashJudge-8B effectively reproduces human judgments and ablations show that interaction evidence improves judge agreement. Experiments with frontier models reveal persistent rendering, analytical, and interaction failures. Together, these results show that realistic dashboard generation remains challenging and that interaction-aware evaluation captures failures missed by static or execution-only checks.
[AI-40] Retrieval-Corrected Conformal Prediction for Time Series CIKM’26
链接: https://arxiv.org/abs/2608.10553
作者: Sangjin Jin,Kangmin Kim,Junhyeong Lee,Yongjae Lee
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Accepted at the 35th ACM International Conference on Information and Knowledge Management (CIKM '26), Rome, Italy
Abstract:Conformal prediction (CP) provides distribution-free prediction intervals for fixed forecasters, but its standard calibration procedure is often inefficient for time series data, where forecast errors are temporally dependent and change across time and operating conditions. Recent time series CP methods improve local calibration using recent, weighted, or localized residuals. Yet local calibration can remain indirect, since broad residual weighting or additional adaptation procedures may dilute the evidence most relevant to the current prediction. This motivates a simple retrieval and correction strategy that selects similar past residuals as local evidence and then corrects the coverage error left by retrieval. In this paper, we propose Retrieval–Corrected Conformal Prediction (RCCP), a retrieval-augmented calibration method for time series prediction intervals. RCCP builds an asymmetric interval from retrieved one-sided residuals and calibrates its normalized retrieval error with a scalar conformal correction. Thus, retrieval provides local residual evidence, while conformal correction determines the final scale needed for coverage. We provide a coverage-gap bound based on the stability of the normalized retrieval error distribution. Across standard benchmarks and backbone forecasters, RCCP attains the target coverage in every setting and achieves the lowest Winkler scores, with fewer severe misses. RCCP also achieves low calibration and inference overhead, showing that retrieval-corrected calibration is an effective and scalable approach to uncertainty quantification in time series forecasting. Code is available at this https URL.
[AI-41] Reinforcement Learning-Based Laser Cutting Machine Parameter Optimization
链接: https://arxiv.org/abs/2608.10549
作者: Khanh Quan Pham,Majid Kundroo,Geunwoo Ban,Seongho Bae,Taehong Kim
类目: Artificial Intelligence (cs.AI)
备注: 21 pages, 8 figures. Accepted manuscript published in Journal of Intelligent Manufacturing (2025). DOI: https://doi.org/10.1007/s10845-025-02619-z
Abstract:Achieving high accuracy in laser-based cutting of optical films requires careful tuning of parameters such as focal length and laser power beam, adjusted according to the specific properties of each film type. Trial-and-error based traditional methods are used to find the most suitable cutting parameters for various films, but they are slow and inaccurate. To address this issue, this paper presents the Reinforcement Learning for Laser Cutting (RL ^2 C) algorithm, which uses Q-learning with an epsilon-greedy policy to dynamically optimize cutting parameters, significantly reducing taper size and film wastage. Additionally, RL ^2 C incorporates a dynamic environment space adaptability mechanism to allow it to adapt to new states encountered during the learning process over multiple batches of experiments. Experimental results demonstrate that RL ^2 C requires fewer steps and less time to find optimal cutting parameters compared to various RL-based optimization methods. Specifically, RL ^2 C reduces the number of optimization steps by up to 12.5% and processing time by up to 81.8% compared to existing methods. This study demonstrates the potential of RL in industrial laser-cutting processes by improving cut quality, reducing time and film wastage, and minimizing manual interventions.
[AI-42] ImpactHO: Importance-Aware KV Cache Transfer for Multi-User Edge LLM Handover
链接: https://arxiv.org/abs/2608.10545
作者: Minwoo Kim,Soochang Song,Namyoon Lee,Bang Chul Jung,Yongjune Kim
类目: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC)
备注:
Abstract:Edge LLMs must preserve inference continuity when a user hands over between edge nodes, requiring key-value (KV) cache transfer to the target node. However, simultaneous handovers saturate the backhaul, preventing full cache delivery within the mobility-imposed transfer window. Rather than allocating bandwidth as if all cache entries were equally valuable, we order each user’s KV cache by importance and transmit only its most informative fraction, turning token-level sparsity into communication savings. We cast the transfer as a multi-user backhaul allocation problem that maximizes average accuracy across users. Each user’s partial-cache accuracy serves as its utility: a sigmoid that fits measurements on the RULER benchmark with R^20.99 across models and context lengths. Because importance ordering front-loads the high-value entries, the concave region of the accuracy curve spans nearly the entire cache. Our proposed allocator keeps served users within this region, making each per-slot allocation problem convex. The optimum is derived via a closed-form weighted water-filling solution that generalizes information-theoretic water-filling and enables online scheduling. The proposed allocator attains over 93.7% average accuracy in a 500ms transfer window, within 0.5pp of the full-cache ceiling, and reaches 98.2-99.5% of a clairvoyant upper bound.
[AI-43] SKILLER: Language-Level Reinforcement Learning for Reusable Skill Extraction in Small Language Models AAAI2027
链接: https://arxiv.org/abs/2608.10538
作者: Chenhao Dang,Siyuan Xiong,Conghui He,Weijia Li
类目: Artificial Intelligence (cs.AI)
备注: 15 pages, 6 figures, and 8 tables. Submitted to AAAI 2027. Project and code: this https URL
Abstract:Agent skills represent a standardized format for packaging procedural knowledge and domain expertise, serving within agent harness systems as an essential mechanism to continually constrain a language model’s behavior space for repeatable, high-quality task execution. However, because strong closed-source models entail high inference costs, current popular agent harnesses, such as Codex and OpenClaw, remain prohibitively expensive when deploying these skills to accomplish real-world tasks. The rapid capability enhancement of open-source models deployable on consumer-grade GPUs presents a compelling opportunity to drastically reduce these costs by leveraging skill-based behavioral constraints. Nevertheless, automatically generating effective skills tailored specifically for such compact models remains a significant practical challenge. To address this, we propose SKILLER, a natural-language-driven reinforcement learning framework designed to automatically generate executor-specific skills for small models, which employs a strong model as the actor and critic, treats the small-model agent system as the environment, and propagates all reinforcement learning signals entirely via natural language. Extensive experimental evaluations across five relevant benchmarks using Qwen3.5-9B and Qwen3.5-4B demonstrate that SKILLER outperforms three open-source and one closed-source skill generation or evolution methods, achieving absolute gains ranging from 4.3 to 20.4 percentage points for the 9B model and 1.8 to 13.3 points for the 4B model, while remarkably matching the performance of strong closed-source models on single-skill tasks in SkillsBench. The project is available at this https URL.
[AI-44] Measuring Semantic Abstractness of SAE Features via Nonlocality
链接: https://arxiv.org/abs/2608.10537
作者: Chuqiao Lin,Shivaji Sondhi,Xiao-Liang Qi
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 18 pages, 9 figures
Abstract:Sparse autoencoders (SAEs) have helped uncover mechanistic explanations for LLM behaviours such as reasoning, jailbreaking etc., via understanding the corresponding task-relevant and causally effective features. To evaluate such mechanistic explanations, downstream studies must distinguish surface lexical features from genuinely high-level ones. However, neither an autointerp-based semantic description nor causal steering utility fully resolves the abstraction level of a feature. To this end, we introduce \emphFeature Nonlocality (FNL), defined as the entropy of the normalized per-position influence on an SAE feature’s activation. We report that FNL correlates with existing LLM-based proxy metrics of feature semantic abstractness, and successfully distinguishes context-dependent reasoning features from token-driven ones, correctly assigning the higher FNL to the contextual feature in 73 – 84% of randomly drawn pairs that consist of one contextual and one token-level feature. We demonstrate two downstream applications. We audit SAE-based features used for jailbreak mitigation and find surprisingly that most effective features are positional features with low FNL rather than genuinely recognizing harmful intents. We report that steering high-FNL features in DeepSeek-R1-Distill-Llama-8B improves MATH-500 accuracy by 4.6 points over the unsteered model and outperforms steering low-FNL features, though the gains are model-specific. We conclude that FNL provides an LLM-independent, label-free, correlational witness of the abstraction level of an SAE feature, with applications in evaluating mechanistic explanations as well as selecting features for downstream interventions. Comments: 18 pages, 9 figures Subjects: Artificial Intelligence (cs.AI); Machine Learning (cs.LG) Cite as: arXiv:2608.10537 [cs.AI] (or arXiv:2608.10537v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2608.10537 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-45] On Understanding Identifying and Mitigating Vulnerabilities in Agent ic Large Language Models
链接: https://arxiv.org/abs/2608.10530
作者: Md Jafrin Hossain,Mohammad Arif Hossain,Nirwan Ansari
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:Large Language Models (LLMs) have undergone a shift from stateless conversational interfaces to autonomous agents capable of multi-step planning, tool invocation, code execution, and maintaining persistent memory. When these agents operate with real-world privileges—calling APIs, modifying files, and querying databases—a compromised reasoning step can trigger unauthorized data access, irreversible state changes, or cascading failures, yet the security research community has not kept pace. To quantify the state of the field, we conducted a systematic literature review under PRISMA 2020 guidelines across six databases, screening 743 records and retaining 85 papers (2023–2025) on agentic LLM security. Attack research outpaces defense work by 3.9:1. Perception-layer vulnerabilities (prompt injection, jailbreaking, adversarial perturbations) dominate, accounting for 66% of papers, while action-layer vulnerabilities (tool misuse, code injection, sandbox escape) appear in only 4.7%, misaligned with real-world risk. Code execution security accounts for 3.5%, and tool-augmented agents 12%. We contribute a four-layer taxonomy mapping 13 vulnerability types across perception, brain, action, and interaction layers, and identify seven open problems centered on containment. Agentic LLM insecurity stems from architectural coupling, where weak isolation allows vulnerabilities to propagate across layers.
[AI-46] Robust Multi-Agent Bandits with Heavy-Tailed Rewards and Information Asymmetry
链接: https://arxiv.org/abs/2608.10529
作者: Daphne Feng,Ricardo Parada,Lily Jiang,Sophia Yi,William Chang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:The multi-armed bandit problem is a central framework in sequential decision-making, extensively studied under sub-Gaussian reward assumptions. However, real-world applications often involve heavy-tailed reward distributions and decentralized, information-asymmetric interactions. We study multi-agent multi-armed bandits with heavy-tailed rewards under three information-asymmetry regimes: unobserved actions with common rewards, observed actions with independent rewards, and unobserved actions with independent rewards. We develop robust decentralized algorithms for each setting and derive regret guarantees that nearly match centralized heavy-tailed rates. Experiments on a Pareto-distributed reward environment validate our theoretical findings and illustrate the trade-offs between synchronization, coordination, and exploration across the three regimes.
[AI-47] Coordinating the Unknown Lipschitz Constant in Multiplayer Bandits
链接: https://arxiv.org/abs/2608.10526
作者: Ricardo Parada,Chenzhang Zhao,William Chang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Motivated by decentralized applications, we study cooperative multi-agent bandits in continuous (Lipschitz) action spaces when the Lipschitz constant is unknown. We consider three information structures: (A)~unobserved actions with common rewards, (B)~observed actions with independent rewards, and ©~unobserved actions with independent rewards. In each case we design and analyze an algorithm that estimates the Lipschitz constant, chooses a discretization of the joint action space, and applies a cooperative bandit method to the induced discrete problem. Players never communicate once learning starts, so the central difficulty is that they must reach the \emphsame discretization from their own data. We prove regret guarantees showing that common rewards and observable actions each supply this agreement for free, and that in their absence agreement can still be bought, through a dithered quantization of the estimate, at no cost in the leading order of the regret.
[AI-48] Improving TensorSketch Using Complex Random Variables
链接: https://arxiv.org/abs/2608.10523
作者: Amit Sharma,Mohammad Azhar Khan,Rameshwar Pratap,Keegan Kang
类目: Data Structures and Algorithms (cs.DS); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
备注:
Abstract:\textttTensorSketch by~\citepham2013fast,kar2012random provides efficient sketching algorithms for high-dimensional polynomial kernels \vecx^\otimes p \in \R^d^p . \citekar2012random uses dense Johnson-Lindenstrauss (JL)-type projections with computational cost O(pDd) , where D denotes the sketch dimension, whereas~\citepham2013fast extends the sparse \textttCountSketch~\citepcount_sketch algorithm, yielding a faster algorithm for high-dimensional sparse inputs with running time O\big(p(\nnz\vecx + D \log D)\big) . However, the variance of both estimators grows exponentially with the polynomial degree p , scaling as 3^p/D . Recent work by~\citepmlr-v206-wacker23a showed that using complex-valued distribution reduces this dependence to 2^p/D for the approach of~\citekar2012random. However, their method relies on dense JL-type projections with computational cost O(pDd) and does not extend to the algorithm of~\citepham2013fast. In this work, we introduce a simple variant of \textttTensorSketch~\citeppham2013fast that achieves the same variance bound as~\citepmlr-v206-wacker23a, while retaining its advantage of the input-sparsity running time. We validate our results with supporting experiments on synthetic and real-world datasets. Subjects: Data Structures and Algorithms (cs.DS); Artificial Intelligence (cs.AI); Machine Learning (stat.ML) Cite as: arXiv:2608.10523 [cs.DS] (or arXiv:2608.10523v1 [cs.DS] for this version) https://doi.org/10.48550/arXiv.2608.10523 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-49] MEGA: Self-Evolving Agent Optimization Infrastructure via Wisdom Graph
链接: https://arxiv.org/abs/2608.10504
作者: Jung Hwan Lee,Kyu Ho Lee,Gwang Hoon Yoo
类目: Artificial Intelligence (cs.AI)
备注: 29 pages, 10 figures. Technical report
Abstract:As coding agents increasingly handle implementation, the central challenge shifts from building individual agents to building an infrastructure that systematically improves them. Current approaches optimize agent systems without accumulating transferable knowledge, accumulate knowledge without compositional reasoning over it, and lack a mechanism for that knowledge to self-evolve through operational evidence. MEGA (Meta Evaluation-Grounded Adaptation) addresses these gaps as a self-evolving infrastructure: each optimization cycle produces durable assets, compositional reasoning over those assets guides subsequent optimization, and operational evidence refines both the accumulated wisdom and the reasoning that governs it. Layer 1 distills reusable wisdom from agent sessions through behavioral-pattern clustering and empirical A/B validation, transforming each process into a durable asset. Layer 2 decomposes these assets into atomic PCR (Primary-Context-Resultant) units within a typed Wisdom Graph and performs deductive, abductive, and inductive reasoning to expand implicit relations; it then assembles context-specific execution plans through compositional retrieval that surfaces bridging knowledge unreachable by embedding similarity alone. Layer 3 performs multi-agent collaborative optimization over heterogeneous agent workflows (code nodes, LLM calls, and tool-using agents), attributing improvement effects to specific strategy changes through controlled evaluation that eliminates data variance. Evidence fed back from Layer 3 drives the self-evolution of both the curation strategies that govern wisdom composition and the optimization trajectories accumulated across runs. The result is an infrastructure in which optimizing an agent system and evolving the knowledge that guides optimization are one and the same process.
[AI-50] From Faulty Memories to Corrected Actions: Dependency-Guided Rollback Repair for Memory-Augmented Agents
链接: https://arxiv.org/abs/2608.10502
作者: Caili Yu,Yiqi Wang,Jiaqi Zhang,Yiqun Duan,Mingkai Zheng,Zhangkai Wu,Kaize Shi,Taotao Cai
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Persistent memory lets language-model agents reuse information across sessions, but it also makes errors durable: a poisoned, stale, or misattributed record can alter reasoning, tool use, answers, and subsequent memory writes. Existing defenses mainly detect or delete suspicious memories, or revise the current response. Deleting the source leaves already propagated claims, actions, and derived memories active, whereas resetting the store or replaying the full trace destroys benign state and repeats unnecessary computation. We therefore formulate \textbfpost-failure memory recovery: \textitgiven a failed execution and diagnosed faulty memories, recover both the answer and persistent state while retaining unaffected work. Our \textbfdependency-guided rollback repair builds a typed memory-to-action graph from runtime provenance, traces explicit downstream dependencies, preserves candidates with independent trusted support, deactivates unsupported memory state, and selectively replays only answer-relevant affected computation. We evaluate this approach on a 150-case controlled benchmark spanning three tool-use domains and four memory failure types, and on a 50-case trajectory-derived stress test adapted from LongMemEval-V2. On the controlled benchmark, it achieves 85.3% recovery versus 77.3% for the best competing recovery method, removes all diagnosed faulty memories, preserves all benign memories, and requires only selective replay with modest LLM-call cost. On the adapted subset, it reaches 68.0% recovery versus 54.0% for the next best method, while also achieving the highest claim invalidation F1, 0.669 versus 0.603. Overall, the results do not imply uniformly better trace reconstruction, but show that dependency-guided rollback repair provides a strong recovery–cost trade-off while repairing faulty memory state and preserving benign memory.
[AI-51] Exploration-Driven Personalized Federated Reinforcement Learning via Intrinsic Motivation
链接: https://arxiv.org/abs/2608.10499
作者: Md Rafid Islam,Rafsan Jany,Zahid Hasan,Ratun Rahman
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Personalized Federated Reinforcement Learning (PFRL) takes a decentralized approach to storing and accessing information based on past experiences while keeping each client’s data private during the learning of each client’s policy. Many current methods for PFRL rely heavily on exploiting existing reinforcement learning reward signals to derive an optimal policy for each client, thereby neglecting exploration in non-stationary or sparse-reward environments. In this work, we introduce a new exploration-driven framework, Exploration-Driven Personalized Federated Reinforcement Learning via Intrinsic Motivation (EDPFRL-IM), that leverages an inherent curiosity-driven exploration at each client to promote local exploration and protect client privacy. Furthermore, to facilitate policy discovery via exploration in previously unexplored state spaces, clients add an intrinsic random network distillation (RND) signal to their extrinsic reward. Additionally, the server does not have access to clients’ raw experiences or local gradient estimates; instead, the server sends global exploration priors and collects minimal novelty summaries from each client to enable both diverse and coordinated exploration among clients. Experiments in benchmark environments show that our framework outperforms average PFRL benchmarks in policy personalization and sample efficiency, primarily in delayed and sparse reward systems. Overall, EDPFRL-IM enables the integration of a flexible exploratory learning structure into federated reinforcement learning systems while preserving client privacy.
[AI-52] INSIDE the Students Mind: Jointly Modeling Latent Reasoning and Action in LLM Student Simulators
链接: https://arxiv.org/abs/2608.10492
作者: Rose Niousha,Minwoo Kang,Narges Norouzi
类目: Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注: Accepted at the Conference on Language Modeling (COLM) 2026
Abstract:Large Language Model (LLM)-based simulators often reproduce observable actions but fail to capture the underlying reasoning behind them. In education, where student simulation is increasingly used for various applications such as evaluating tutoring systems, this gap is especially pronounced. Two students may submit identical submissions for entirely different reasons. We present INTERNAL STUDENT DIALOGUE (INSIDE), a student modeling framework that fine-tunes LLMs not only to act like students but also to think like them. INSIDE generates internal dialogue grounded in Bloom’s Taxonomy across cognitive, affective, and action dimensions, and fine-tunes models on paired think traces and actions. We baseline against different prompting frameworks and evaluate on two axes: fidelity of simulated actions and quality of generated internal dialogue. Our evaluations show that INSIDE improves simulation fidelity in both action fidelity, matching code generation of real students, and reasoning alignment, achieving the highest alignment across models up to 57.9%.
[AI-53] Predicting Space Groups of Double Perovskites by LLM with Dynamic Few-Shot Learning
链接: https://arxiv.org/abs/2608.10483
作者: Jongwon Park,Inhyo Lee,Junhyeong Lee,Seunghwa Ryu
类目: Artificial Intelligence (cs.AI); Materials Science (cond-mat.mtrl-sci)
备注: 46 pages, 6 figures, Supplementary Information included(24 pages)
Abstract:Double perovskites (DPs) offer broad compositional tunability, but predicting the space groups (SGs) of stable structures remains difficult because available datasets are often strongly imbalanced toward dominant SG classes. We refer to dominant SG classes as major SGs and underrepresented classes as minor SGs. We introduce Dynamic and Diversity-enhanced Few-shot Retrieval and Rule-Guided Inference for Space-Group Prediction (DyRIS), an LLM-agent-based framework that predicts ranked SG candidates from a given DP composition. DyRIS uses diversity-enhanced dynamic few-shot prompting to retrieve relevant in-context examples while limiting the dominance of frequently represented SGs. It further incorporates rule-guided inference based on B/B’ cation ordering, quantitative indicators, and major-SG bias control to refine and rank the final Top-3 SG candidates. We evaluate DyRIS on 3,528 thermodynamically filtered DP entries and compare it with composition-based and descriptor-based baselines. At a training-data ratio of 0.5, DyRIS achieves competitive overall accuracy while obtaining the best Overall Top-1 macro-F1 score and the best performance across all Minor-SG metrics. DyRIS improves Minor-SG Top-1 accuracy by 3.26 percentage points relative to CrabNet and achieves higher Minor-SG Top-3 accuracy than the strongest PyCaret-based baseline. Ablation studies show that diversity-enhanced retrieval, quantitative indicators, major-SG bias control, and B/B’ ordering information each contribute to prediction performance. Additional experiments show that the final rule-guided inference step is not easily replaced by conventional classifier- or ranker-based models. These findings demonstrate the potential of combining retrieval-based LLM reasoning with crystallographic domain knowledge for SG prediction in imbalanced materials datasets.
[AI-54] Multi-Granular Rationale-Guided Molecular LLM for Property Prediction
链接: https://arxiv.org/abs/2608.10480
作者: Junwoo Park,Minyoung Shin,Cheol Soon Lee,Sujee Lee
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 16 pages, 5 figures, 18 tables. Code: this https URL
Abstract:Large language models (LLMs) are widely applied across chemical tasks, such as molecular property prediction, which underpins drug discovery. Molecular LLMs represent a molecule through several modalities, notably a 1D SMILES sequence or a 2D molecular graph. Both encode molecular information implicitly, so the contribution of individual substructures remains opaque. Retrieval and augmentation methods add context, but from external sources. However, the cues chemists reason over are the internal substructures that drive a property up or down. We propose MR-MoL, a multi-granular rationale-guided molecular LLM that supplies this evidence directly. A fine-tuned GNN scores each substructure through masking, and the most influential ones are serialized as a ranked, direction-tagged rationale that the LLM reads alongside the SMILES sequence and molecular graph. The rationale spans three levels of granularity: Murcko scaffolds with their side chains, BRICS fragments, and functional groups. This is, to our knowledge, the first method to expose GNN-derived attributions to an LLM as evidence for property prediction. On eight MoleculeNet tasks, MR-MoL achieves the best overall results among generalist models and narrows the gap to specialist models tuned for each task. Five diagnostics further confirm that the model reads the rationale rather than merely benefiting from its presence. Its direction, rank, and substructure each shape the prediction, and its attributions reproduce known structure-property relationships.
[AI-55] Critic-Free Pretraining for Efficient Online Reinforcement Learning Fine-Tuning
链接: https://arxiv.org/abs/2608.10473
作者: Daoyi Li,Yixian Zhang,Chao Yu,Wenbo Ding,Yu Wang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Offline-to-online (O2O) reinforcement learning aims to leverage policies pretrained on static datasets while improving them through online interaction. However, directly reusing an offline-trained critic can hinder online fine-tuning: as the policy and data distribution change rapidly, value estimates inherited from offline training may become misaligned with the online environment, leading to inaccurate policy improvement and inefficient exploration. To address this problem, we introduce \textbfCritic-\textbfFree \textbfPretraining: an efficient paradigm that completely abandons the approach of offline critic training, allowing a freshly initialized critic to adapt without inheriting biased estimates. CFP is compatible with various mainstream O2O algorithms and consistently matches or improves upon conventional O2O algorithms across a diverse set of tasks, with particularly pronounced gains on several challenging tasks.
[AI-56] RLMOpt: Adaptive Prompt Optimization via Recursive Language Models
链接: https://arxiv.org/abs/2608.10471
作者: Subhash Bangalore Satheesha,Nirvik Pande,Deepthi Duddempudi,Bharath Dandala
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Prompt optimizers automate the search for prompts that improve language-model performance, but existing methods rely on a predefined optimization procedure: the algorithm determines which candidates to explore and how the search progresses, while the language model generates or refines prompt proposals. We introduce RLMOpt, a prompt optimizer that makes the search policy itself language-model-driven through a recursive language model (RLM). The RLM agent operates over a tool-based environment, inspecting task information, analyzing failures, generating candidates, allocating evaluation budget, and deciding when to stop. A deterministic harness complements the agent by enforcing objective scoring, Pareto-based selection, and regression constraints. We evaluate RLMOpt across four benchmarks spanning structured clinical information extraction (Chia), multi-hop question answering (HotpotQA), verifiable instruction following (IFBench-2025), and multi-turn tool-calling agents (BFCL). In a matched comparison at a single seed, RLMOpt obtains the best held-out score on all four benchmarks and leads the four-task mean (0.610 against 0.589 for GEPA). Repeating each benchmark across seeds yields 11 matched benchmark-seed comparisons, in which RLMOpt outperforms GEPA in 9 cases. Across all 11 runs, it never produced a prompt that underperformed its seed, whereas GEPA fell below its starting point twice. It is also more efficient, achieving these results with fewer search rollouts while producing prompts that are 27-79% the size of those produced by GEPA. Our results further show that optimization gains are determined primarily by the headroom available in the seed prompt, rather than by the search budget. Efficient optimization therefore depends on reaching the available headroom reliably and with minimal search Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2608.10471 [cs.AI] (or arXiv:2608.10471v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2608.10471 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-57] Quantum Incremental Learning with Mixed State Prototypes
链接: https://arxiv.org/abs/2608.10464
作者: Yu Wu,Qianli Zhou,Xinyang Deng,Wen Jiang,Kang Hao Cheong,Witold Pedrycz
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Incremental learning models are required to learn new classes sequentially without catastrophic forgetting, while operating under parameter and memory constraints. In the Noisy Intermediate-Scale Quantum (NISQ) era, although quantum neural networks offer advantages in feature mapping, hardware limitations restrict circuit width. Furthermore, traditional quantum classifiers are constrained by the number of orthogonal basis states, limiting their capacity to accommodate a continually growing number of categories. Thus, we introduce a novel quantum incremental learning framework based on trainable mixed-state prototypes. Its original design incorporates new classes by adding class prototypes rather than increasing the circuit width of the shared quantum backbone. The use of mixed-state prototypes is another key contribution, since they have representation capabilities to represent information than a single pure-state prototype. And the decomposable mixed-state calculation provides lower production costs and a convenient Hilbert-Schmidt (HS) distance metric for classification. Simulation results show that our model achieves high-dimensional feature concentration using a minimal number of qubits, while demonstrating lower computational complexity and robust representation in incremental learning tasks compared with classical baselines.
[AI-58] Rationale-Guided Learning for Multimodal Emotion Recognition ICASSP2026
链接: https://arxiv.org/abs/2608.10448
作者: Sujung Oh,Jung Uk Kim,Sangmin Lee
类目: Artificial Intelligence (cs.AI)
备注: ICASSP 2026
Abstract:Multimodal emotion recognition in conversation (MERC) requires understanding complex interactions between verbal and non-verbal cues. However, most existing approaches fundamentally treat this as a direct input-output (multimodal cues-emotion labels) mapping problem, overlooking the causal reasoning that humans use when interpreting emotions. We propose rationale-guided learning (RGL), a novel framework that transforms MERC into a cognitively-inspired reasoning task. Based on dual-process theory, we decompose emotional reasoning into three facets: Intuitive (immediate perception, System 1), Contextual (situational analysis, System 2), and Integrative (synthesis of both). We leverage an MLLM offline to generate structured rationales, which are encoded as memories to guide model training via aligning internal representations with human-like reasoning patterns. Our final model operates without any MLLM overheads at inference time. Experimental results show that RGL achieves state-of-the-art performance on the IEMOCAP and MELD benchmarks. Further, for interpretation, we demonstrate that the model’s internal features effectively retrieve semantically correct rationales for unseen test samples, validating its rationale reasoning capabilities.
[AI-59] Continuous Interaction Diffusion: A Diffusion-Native Runtime for Asynchronous Tool-Augmented Reasoning
链接: https://arxiv.org/abs/2608.10438
作者: Yuhang Cao
类目: Artificial Intelligence (cs.AI)
备注: 15 pages, 5 figures, 1 table
Abstract:Large language models increasingly rely on external tools to access up-to-date information, perform computation, and interact with the outside world. For autoregressive models, tool use naturally fits the generation process: the model emits a tool call, waits for the result, and then continues generating. Diffusion language models (dLLMs), however, reason by repeatedly refining many parts of their output in parallel, making this stop-and-resume interaction pattern unnecessarily restrictive. It can force tool decisions before the model’s reasoning has stabilized, delay useful observations until a discrete call finishes, and introduce redundant refinement and tool execution, potentially hurting both task accuracy and inference efficiency. We introduce Continuous Interaction Diffusion (CID), a diffusion-native model–runtime architecture that integrates tool interaction into iterative denoising. CID separates a model-read-only fact channel, a thought channel represented by a Typed Cognitive Tensor, and a display channel. Information needs can emerge before a textual or JSON call is fully serialized, allowing perceptual bindings to launch external reads while denoising continues. Returned results are projected into the evolving thought state and can revise earlier cognition and display regions. Persistent bindings reuse static results without repeated external execution and refresh changing sources when needed. CID is designed to expose evidence earlier, overlap tool latency with model computation, reduce duplicate external work, and preserve useful computation after new evidence arrives. We formalize the architecture, runtime, and training objectives, and define an evaluation protocol for task quality and end-to-end efficiency. This first paper focuses on read-only tools and makes no empirical performance claims. Comments: 15 pages, 5 figures, 1 table Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2608.10438 [cs.AI] (or arXiv:2608.10438v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2608.10438 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-60] Conversational versus Dashboard Explainable AI for UAV Intrusion Detection: An Empirical Study of Operator Trust and Reliance
链接: https://arxiv.org/abs/2608.10434
作者: Cong Chi Nguyen,Trang Mai Xuan,Vu-Duc Ngo,Kim-Ngan Thi Nguyen,Trong-Nghia Nguyen,Thien Van Luong
类目: Artificial Intelligence (cs.AI)
备注: 12 pages, 3 figures, EIDT conference
Abstract:Machine learning-based Intrusion Detection Systems (IDS) have demonstrated superior performance in securing Unmanned Aerial Vehicle (UAV) networks. However, the ‘black-box’ nature of these models, combined with the high dimensionality of multimodal cyber-physical data, poses significant interpretability challenges. Static visualization dashboards may struggle to present complex relationships among multimodal cyber-physical features in a form that is easy for operators to inspect and interpret. To address this, we propose a Conversational XAI interface powered by Large Language Models (LLM) to facilitate on-demand investigation. In a controlled experiment with participants, we systematically evaluated the impact of this conversational interface versus a traditional XAI Dashboard on operator understanding, trust, and reliance during post-incident auditing tasks. Our results suggest that the conversational interface was perceived as more useful than the dashboard, potentially because it helped participants access and synthesize relevant information more easily. However, this benefit was accompanied by a lower level of appropriate self-reliance, indicating a potential risk of over-reliance. One possible interpretation is that the natural-language responses made the AI advice easier to accept, which may have reduced participants’ tendency to verify the underlying evidence when the IDS was incorrect. These findings point to a potential trade-off in human-AI collaboration for UAV intrusion auditing: interaction mechanisms that improve perceived usability may also increase the risk of inappropriate reliance. We conclude by discussing design implications for future XAI systems that balance seamless interaction with cognitive forcing functions to foster appropriate reliance.
[AI-61] Actionable Hallucination Detection: Translating Latent Uncertainty into Agent ic Critique
链接: https://arxiv.org/abs/2608.10430
作者: Sanidhya Vijayvargiya,Rahul Lokesh
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 22 pages, 6 figures
Abstract:Large Language Models (LLMs) deployed as AI agents frequently exhibit user specification-grounding failures, executing hallucinated, undesired actions to force a resolution rather than expressing uncertainty. Existing detection methods fail to provide actionable, real-time correction as they either do not localize the hallucinations, or incur prohibitive inference latency. We introduce the Latent Critic, a lightweight low-rank adapter (LoRA) that operates concurrently with a frozen base LLM’s generation to actively restructure the transformer’s residual stream—amplifying latent grounding signals and translating them into localized, natural language feedback within a single sequence. By refining the base model’s native uncertainty signals, this manipulation of the latent space enables reliable, granular detection without the overhead of secondary inference loops. Mechanistic analysis via activation patching and layer-wise probing shows that this rank-invariant behavior restructures pre-existing uncertainty geometry into a linearly separable representation that transfers more reliably than base model representations alone. Using tool-calling as an instantiation of granular hallucinations, we validate the detection and downstream improvements enabled by the Latent Critic architecture across Qwen and Llama-based models. Demonstrating superior real-time efficacy, our approach significantly outperforms equivalent-scale fine-tuned external detectors, semantic entropy baselines, and passive internal probes in isolating hallucinations, achieving 0.966 AUROC and 80% accuracy in localization (e.g., ungrounded: date). When deployed in a closed-loop ReAct environment, the Critic acts as a negligible latency guardrail, intercepting hallucinations before execution to prevent undesired actions while simultaneously leveraging this specific localized feedback to enable efficient agent self-correction.
[AI-62] Recovering Wasted Compute in Autoresearch Agents
链接: https://arxiv.org/abs/2608.10424
作者: Au Kwok Chun,Abhigyan Acherjee,Amrutha Rao,Zaiqian Chen,Kazem Meidani,C. Bayan Bruss,Micah Goldblum
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:A slew of recent works develop agents for solving research problems end-to-end, a paradigm increasingly referred to as autoresearch. Such agents have inspired large industry investment, motivated by their potential to automate time-consuming human labor and customize machine learning solutions for specialized applications. In this paper, we study the modeling pipeline at the core of these autoresearch systems and identify common failure modes when they are applied to tabular datasets: (1) they waste compute resolving the same bugs over and over again; (2) they often fail to tune hyperparameters even when they have a large remaining compute budget; (3) the tree-search algorithms that power them do not explore; and (4) they perform data analysis, mimicking the humans whose data they are trained on, but do not use that analysis to make downstream decisions. We explore targeted interventions and find that a global debug consultant that shares discovered runtime constraints across all branches of the search tree, prompt- and control-level enhancements, and refined tree-search algorithms successfully recover wasted compute. Our results show that large gains in autoresearch agent performance are achievable through agentic design alone, holding the underlying language model fixed.
[AI-63] Reasoning Shortcuts and Value Symmetries: What Symmetry Permits Architecture Realizes and Optimization Selects
链接: https://arxiv.org/abs/2608.10420
作者: Xin Xu
类目: Artificial Intelligence (cs.AI)
备注: 55 pages, 2 figures, 8 tables
Abstract:Reasoning shortcuts are solutions of a neurosymbolic system’s rules that produce correct predictions through unintended concepts. A recent framework of Takemura, Inoue, and Nishino analyzes them through an automorphism group of value relabelings and asks, as its central open question, when rules pin concepts down. We first show that the framework’s key definition, one shared permutation applied at every position, does not apply as stated to any of the four heterogeneous benchmarks it was evaluated on, and that the most direct embedding, padding domains to a common size, produces confident false pathology: 90.91% of solution pairs reported unexplained on CLE4EVR, where every well-defined member of the hierarchy we introduce reports 0%, and the padded verdict’s content rotates with configuration-file ordering. Re-measuring eleven rule families under fifteen pre-specified predictions (thirteen confirmed), unexplained-pair rates span 0% to 99.9999% and track provable structure: six theorems give sufficient conditions for transitivity and its failure, including a Free Slot Lemma certifying Kandinsky’s pathology from syntax alone. For circuit-given rules, deciding symmetry-inertness of a coordinate is coNP-complete; nontrivial-automorphism existence is coNP-hard under randomized reductions, lies in \Sigma_2^p , is not \Sigma_2^p -complete unless PH collapses, and on monotone circuits is coNP-complete outright. In the Boolean case transitivity is classified exactly: automorphisms explain everything iff the solution set is an affine coset. Weakly supervised models place all 94 observed shortcuts at the one level the componentwise theory flags and none at the 48 it certifies transitive; twelve typed-ambiguous levels produce none, separating what symmetry permits from what optimization selects, and a dual-head control replicates the geography. All numbers trace to released artifacts.
[AI-64] Never Stop Speaking: a Denial-of-Service Attack on End-to-End Speech Language Models
链接: https://arxiv.org/abs/2608.10405
作者: Shuozhe Cheng,Kunlan Xiang,Mingxuan Li,Ji Zhang,Dongxiao Liu,Wenbo Jiang
类目: ound (cs.SD); Artificial Intelligence (cs.AI)
备注:
Abstract:Many studies have shown that specially crafted inputs can induce large language models (LLMs) to generate excessively long outputs, resulting in significant computational overhead and resource consumption. While most existing denial-of-service (DoS) attacks target text-only LLMs, end-to-end (E2E) speech LLMs are rapidly emerging. Existing text-based DoS attacks primarily rely on prompt engineering, such as adversarial suffixes or semantic inducement, which exploit the discrete nature of text inputs and therefore cannot be directly transferred to continuous speech inputs. Moreover, prior studies on speech model security mainly focus on ASR or TTS systems, leaving the DoS vulnerability of E2E speech LLMs largely unexplored. To address this gap, we propose the perturbation-based DoS attack targeting E2E speech models. Instead of inducing long outputs through prompt manipulation, our method optimizes imperceptible acoustic perturbations to directly influence the model’s autoregressive generation process while preserving the original input length. Specifically, we formulate the attack as a composite optimization objective that jointly suppresses EOS generation, encourages prolonged decoding, and largely preserves semantic consistency by integrating weighted EOS loss, top-k logit loss, length loss, and semantic alignment loss. To further improve stealthiness, we employ voice activity detection (VAD) to inject perturbations only into voiced regions. Extensive experiments on three open-source E2E speech LLMs demonstrate that our method achieves stable attack success rate while significantly increasing generation length and GPU resource consumption, revealing security risks in modern ALLMs.
[AI-65] hreat-guided Policy-aware Scene Perturbation for Safe Autonomous Driving with Online Reinforcement Learning
链接: https://arxiv.org/abs/2608.10403
作者: Xincong Hu(1),Lei Ou(1),Maosen Li(2),Jingtao Zhang(2),Liguo Hou(2),Zongzhang Zhang(1) ((1) Nanjing University, (2) Yinwang Intelligent Technology Co., Ltd)
类目: Artificial Intelligence (cs.AI)
备注: 11pages, 5figures
Abstract:Reinforcement learning (RL) has shown promising performance in autonomous driving, yet ensuring the safety of online RL policies remains challenging due to insufficient exposure to safety-critical driving scenes. The long-tailed nature of real-world traffic situations makes dangerous and rare interactions difficult to encounter through conventional sampling, limiting the ability of RL policies to learn robust safety behaviors. Existing methods improve training diversity by synthesizing challenging scenes or adversarial situations. However, these approaches typically optimize scene generation objectives separately from the evolving policy, without explicitly modeling how generated perturbations relate to the current policy’s weaknesses and learning needs. In this paper, we propose Threat-guided Policy-aware Scene Perturbation (TPSP) for safe autonomous driving with online RL. TPSP introduces a policy-aware scene encoder to capture the interaction between policy behaviors and surrounding environments, enabling scene perturbation aligned with the current policy. Based on this representation, TPSP selectively perturbs critical objects rather than applying uniform modifications across the scene. Furthermore, we develop a threat-guided optimization strategy that evaluates perturbed scenes through threat-level differences between policy rollouts on original and perturbed scenes, guiding the generation of safety-critical scenes with higher training value. Comprehensive experiments demonstrate that TPSP improves safety learning efficiency, achieving strong safety performance on NAVSIM v2 with approximately 4 million kilometers of simulated driving data. Ablation studies verify that policy-aware targeted perturbations provide more informative safety-critical experiences than random or policy-unaware strategies, enabling safer driving under limited interaction budgets.
[AI-66] ELVAE: Evidential Learning-Based Variational Autoencoder for Uncertainty-Aware Generation
链接: https://arxiv.org/abs/2608.10398
作者: Ge Wang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 11 pages, 4 figures, 3 tables
Abstract:Variational autoencoders generate samples from probabilistic latent representations but do not distinguish uncertainty about the latent location from variability around it. We formulate ELVAE, an evidential learning-based VAE in which each latent coordinate is governed by an input-dependent normal-inverse-gamma posterior. This hierarchy yields an explicit latent-location uncertainty that can be used during generation, not merely reported after inference: low-uncertainty anchors support more reliable synthetic samples, while high-uncertainty anchors can be deliberately exploited for stress testing. The objective is an exact evidence lower bound, and we show that direct regularization of the full hierarchy is required, since the marginalized latent law alone cannot identify the uncertainty decomposition. In an MNIST generation pilot with a frozen external classifier, this uncertainty clearly stratified the semantic reliability of generated digits. A zero-displacement control revealed that most of the effect reflects how reliably an anchor can be re-generated, while a smaller but distinct component is attributable to uncertainty-scaled perturbation itself. The effect holds only under within-class uncertainty ranking, and its magnitude varies across seeds. These findings support the learned latent-location uncertainty as a practical control variable for uncertainty-aware generation, separating anchor reliability from perturbation-induced failure.
[AI-67] Hidden in Plain Sight: Diffusion-Based Unrestricted Robotic Attacks on Vision-Language-Action Models
链接: https://arxiv.org/abs/2608.10393
作者: Jiahui Han,Yuhui Yao,Xin Wang,Jiafei Cao,Mingxuan Zhang,Danfeng Shan,Huiqi Deng,Guanchu Wang,Xia Hu
类目: Artificial Intelligence (cs.AI); Robotics (cs.RO)
备注:
Abstract:Vision-Language-Action (VLA) models have shown strong capabilities in controlling robots across diverse manipulation tasks. However, their adversarial robustness remains largely underexplored, and exploiting this weakness can lead to physical-world harm. Existing attacks on VLA models often rely on pixel-space perturbations or white-box access, resulting in noticeable artifacts and limited deployability in real-world robotic systems. In this work, we propose DURA, a diffusion-based unrestricted robotic attack that generates visually natural adversarial patches for VLA models. DURA supports both white-box and black-box attack settings, where the black-box setting requires only the predicted actions of the victim model. By optimizing along the latent trajectory of a pretrained diffusion model, DURA generates visually natural patches while steering the robot toward attacker-specified target actions. Extensive experiments in both simulation and the real physical world show that DURA consistently outperforms existing methods. Our findings expose a safety risk for physically deployed VLA models and call for stronger defenses.
[AI-68] Beyond Forecasting: Recasting Volatility Control as a Routing Problem
链接: https://arxiv.org/abs/2608.10375
作者: Hongji Pu,Leyang Zhou
类目: Computational Engineering, Finance, and Science (cs.CE); Artificial Intelligence (cs.AI)
备注: 24 pages, 6 figures, ACM ICAIF
Abstract:Volatility control converts risk estimates into portfolio exposure, yet existing approaches often rely on a fixed volatility estimator or a pre-defined control rule that may not adapt to changing market conditions. We propose VolRouter, a modular framework that formulates volatility control as state-conditioned routing over estimator-controller pairs. VolRouter first summarizes market conditions into a control-relevant state profile and then performs routing through three stages: state inference, switch review, and pair selection. The Router can be implemented using rule-based, learnable, or LLM-based decision modules, while portfolio actions remain generated by predefined control policies. We evaluate VolRouter across SP 500, Multi-Asset, Bitcoin, and USDT volatility-control settings. VolRouter achieves the highest Sharpe ratio in three of four settings. On SP 500, it improves Sharpe from 0.952 for RV + Naive Scaling to 1.222 while reducing maximum drawdown from 15.10% to 12.58% and daily CVaR from 1.76% to 1.32%. On Multi-Asset, it improves Sharpe from 1.498 to 1.540 and reduces CVaR from 1.56% to 1.18%. Bitcoin shows similar improvements in risk-adjusted performance, while USDT provides a boundary case where simpler state-aware selectors remain competitive. Ablation and sensitivity analyses show that the improvement comes from relative policy evaluation and selective persistent switching rather than simply expanding the policy library. These results suggest that volatility control can be viewed as a policy-selection problem when risk management requirements vary across market states.
[AI-69] Nutrition Data Infrastructure for the AI Era: Operationalizing FAIR for Agent -Mediated Research
链接: https://arxiv.org/abs/2608.10363
作者: Lin Liao,Peng Li
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:AI agents can accelerate nutrition research, but their analyses inherit the identity, semantic, and release ambiguities of the underlying data. We present Nutrition Data Service (NDS), source-preserving infrastructure that operationalizes FAIR for automated use: description resolution makes release-specific records findable; typed crosswalks connect independently released resources; machine-readable interfaces expose versioned sources and crosswalks, making analyses by AI agents replayable and auditable. On food-description benchmarks, NDS shows strong held-out accuracy and outperforms the best published language-model result on NutriBench. External and blinded crosswalk evaluations show that its typed contract favors defensible links and rejects unsupported mappings. In a person-level glycemic-index analysis, pinned NDS inputs produce identical outputs across models and repeated runs, while open-web reconstruction remains unstable. The central result is that agent-mediated nutrition research requires a new data infrastructure for data identity, search, and crosswalk.
[AI-70] MemSpec: Memory-Aware Runtime for Adaptive Draft Scheduling in Speculative Decoding on Edge Devices
链接: https://arxiv.org/abs/2608.10362
作者: Eunjeong Kim,Yeong Jun Jeon,Myeonggyun Han
类目: Operating Systems (cs.OS); Artificial Intelligence (cs.AI)
备注: Published in LCTES 2026
Abstract:Speculative decoding accelerates autoregressive large language model (LLM) inference by using a lightweight draft model to speculate multiple tokens, reducing expensive target model decoding steps. Its effectiveness depends heavily on draft selection, motivating adaptive methods that exploit variation across inputs and generation stages. On memory-constrained edge devices, however, these methods often fail to improve end-to-end throughput due to the overhead of switching between draft models. We identify a key limitation in this setting: the mismatch between draft selection and draft availability under tight memory budgets. To address this challenge, we present MemSpec, a prediction-guided, memory-aware runtime for adaptive speculative decoding on edge devices. MemSpec decouples draft selection from execution through proactive resident working-set management. A lightweight predictor estimates draft effectiveness from prompt and generation context, while a memory-aware scheduler reduces reactive model loading overhead. Experiments on a Jetson Orin Nano show that MemSpec improves steady-state generation throughput by 40.7% on average over state-of-the-art bandit-based adaptive methods while closely approaching the oracle upper bound.
[AI-71] Efficient Reinforcement Learning for Long-Horizon Tool-Use Agent ic Tasks
链接: https://arxiv.org/abs/2608.10357
作者: Zelei Cheng,Amritansh Mishra,Sambit Sahu,William Campbell
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Published at the COLM 2026 Workshop on Efficient Reasoning
Abstract:Long-horizon tool-using agents must reason over user goals, domain policies, tool calls, simulator state, and delayed verifiable rewards. Reinforcement learning (RL) is a natural fit for this setting, but multi-turn on-policy rollouts create long contexts, while model-specific attention layers may require custom masks and learned sink normalization. We present SINKFLEX-RL, a modular training system for RL in dual-control tool-use environments. The system combines a Gymnasium-compatible environment wrapper, a VERL-style rollout dataflow, group-relative policy optimization without a separate value model, and a sink-aware FlexAttention path designed to preserve model-specific sink scaling under causal and sliding-window masks. In a preliminary Tau2Bench retail run, validation reward (mean@1) rises from 0.25 early in training to 0.44 later in the observed training window, while training-score and trajectory-reward proxies also trend upward. In a fixed-configuration memory benchmark, the optimized attention path reduces peak VRAM from 28.06GB to 22.52GB at 4096 tokens, a 19.7% reduction, and runs the measured 8192-token configuration using 25.53 ~GB where the eager baseline runs out of memory. These results illustrate the value of integrating environment interfaces, RL dataflow, and attention-kernel design for memory-feasible long-horizon agent training.
[AI-72] Hierarchical Compositionality for An Assistive AI Agent
链接: https://arxiv.org/abs/2608.10330
作者: Tianyi Fu,Mohan Sridharan
类目: Artificial Intelligence (cs.AI)
备注: 25 pages, 9 figures, 4 tables. Project page: this https URL
Abstract:AI agents are increasingly being developed to assist humans in various applications, and Large Language Models and other deep network architectures are considered to be state of the art for such agents. These methods are impressive stochastic predictors, but they are resource-hungry, opaque, and known to make arbitrary decisions in novel situations due to the narrow set of underlying representation and processing choices. Our work seeks to explore the design of architectures for such AI agents based on core principles that can be traced back to the early pioneers of AI but are not fully utilized in modern AI methods. We do so in this paper in the context of the core problem of AI agents addressing ambiguity in the objects being referred to by the human participants. Humans address such ambiguity by heuristically leveraging compositional knowledge of domain context and the preferences of the other human participants. Drawing inspiration from this observation, we describe an architecture that embeds the principle of hierarchical compositionality and uses simple heuristics to achieve the desired disambiguation. Specifically, domain objects are represented in terms of primitive attributes drawn from human-validated semantic feature norms, and a hierarchical combination of attributes and concepts automatically identified from a limited observed history of interactions of an assistive agent with specific users. The assistive agent then achieves the desired disambiguation by reasoning with knowledge of this compositional hierarchy; axioms governing domain dynamics; and models of semantic compatibility, session salience, and user-specific thematic preference, requesting human clarification when necessary. Experiments show that our approach consistently outperforms state of the art data-driven baselines, supporting adaptation to specific user profiles.
[AI-73] oward a Theory of Value in AI Alignment
链接: https://arxiv.org/abs/2608.10327
作者: Andrew Smart,Shazeda Ahmed,Jackie Kay,Jimmy Tobin,Kris Shrishak,Abeba Birhane
类目: Artificial Intelligence (cs.AI)
备注: 15 pages, 2 figures
Abstract:Can AI systems be aligned to human values? The popularization of large language models (LLMs) and multi-modal foundation models has seen a rise in harms spanning from toxic speech and hallucinations to AI agents executing unauthorized actions. Within the field of AI safety, these harmful instances are often framed as the alignment problem, or of models being misaligned with human values. Researchers have responded by pursuing applied and theoretical AI value alignment efforts, often without specifying what they mean by human values. How does the field of AI value alignment conceive of human values? How are these conceptions of values technically operationalized and evaluated? What does the emergent theory of value from this field signify for the future of AI? We annotated 94 value alignment research papers to discern their implicit theory of values in AI. The majority do not define values, relying heavily on preferences as a stand in that runs the risk of reducing complex culturally situated concepts down to binary choices. As researchers dispense with using human annotators for model training and evaluation, turning instead to synthetic data and autorater approaches to aligning and evaluating models, we identify the potential to close off alternative methods for contesting and enacting values in foundation models. In making AI value alignments philosophical commitments explicit, we seek to bring great specificity and under explored perspectives in the debate on whether and how AI can address human values.
[AI-74] Neuroevolution Arena: Nested Ecological Evaluation of Update-and-Inheritance Regimes across Neural Architectures
链接: https://arxiv.org/abs/2608.10323
作者: Yuxu Ge,Yifei Cheng
类目: Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE)
备注: 22 pages, 2 figures, 8 tables. Preprint
Abstract:Competitive artificial-life systems can rank trained controllers differently under training and ecological evaluation. We present Neuroevolution Arena, a GPU-accelerated spatial ecology of independently parameterized neural-network cells, and an audit-tracked nested evaluation protocol. Three implementation-specific update-and-inheritance regimes (EvoEvo, EvoRL, and RLRL) are crossed with two neural architectures for 50,000 generations in three independent training runs per condition. One saved elite-controller artifact from each of the 18 runs enters an aligned-run frozen-evaluation design comprising 198 computational jobs. Pairwise effects average three seed-defined ecological contexts (two cooperation-permitting and one attack-permitting) within each aligned training-run block; the independent level remains n = 3 runs per condition. RL-enabled regimes attain higher recorded training fitness than EvoEvo, whereas pairwise outcomes show architecture-conditioned majority patterns and substantial artifact dependence. Six-way winners vary across artifacts and contexts, and the prespecified survival endpoint has a complete floor. We contribute a nested protocol that separates training-run artifacts from evaluation contexts and exposes, rather than conceals, their different sources of variation.
[AI-75] Do Personalized Skills Help Coding Agents ? An Empirical Study of Developer Interaction Histories
链接: https://arxiv.org/abs/2608.10319
作者: Shuyan Huang,Kai Du,Andrew Lan
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: 15 pages, 10 figures
Abstract:Large language model (LLM)-powered agents have rapidly evolved from code-completion tools into solvers of complex software engineering tasks. As developers collaborate with coding agents over time, their preferences emerge through repeated interactions and can be used to adapt agent behavior to better meet individual developers’ needs. Capturing and reusing these preferences may reduce repeated corrections and improve developer-agent collaboration. Agent skills provide a lightweight mechanism for transferring experience without modifying model parameters. However, existing work primarily focuses on task-specific skills, and it remains unclear whether developer-specific skills distilled from interaction histories can generalize to future tasks. We propose a framework for extracting reusable developer preferences from interaction traces. It first generates personalized skills through rule-based bootstrapping and evidence-grounded refinement, and then evaluates them using a reproducible replay framework with an interactive, trajectory-conditioned LLM-based human developer simulator. We conduct an experiment on 206 real-world developer-agent sessions from 13 developers and compare personalized skills against no-skill, generic-skill, and other-user-skill baselines. Personalized skills provide small and inconsistent improvements over the no-skill baseline, whereas generic skills pooled across developers achieve the largest and most consistent gains. Further analysis suggests that personalized skills become more effective when developer preferences appear frequently, particularly when their histories contain multiple examples relevant to future tasks. These findings provide empirical insights into when developer-specific personalization is effective and demonstrate that broadly transferable procedural knowledge can be more robust than developer-specific preference signals.
[AI-76] Logit-Boundary Geometric Belief Interfaces and Sparse Sheaf-Enclave Protocols: A Self-Contained Substrate for Secure Network Electronic Health Record (EHR) Interoperability
链接: https://arxiv.org/abs/2608.10300
作者: Alvin Spivey,Thomas Huang
类目: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Machine Learning (cs.LG)
备注: 39 pages; executable Julia verification code included as ancillary material; companion public benchmark: this https URL
Abstract:Electronic health-record interoperability is a boundary problem: legacy systems, generative models, terminology services, identity systems, and human reviewers may each expose rich internal states, while operational exchange requires a narrow shared interface of typed claims, bounded uncertainty, provenance, and explicit admission or abstention. This paper details a mathematical and engineering architecture for that interface. The organizing idea is the logit boundary: a discovery model may propose pre-threshold scores over a local categorical decision, but a deterministic judgment substrate decides whether the proposal is admissible, requires review, or must be quarantined before any Fast Healthcare Interoperability Resources (FHIR) transaction is constructed. The resulting Geometric Belief Interface (GBI) combines finite boundary semantics, local Dirichlet evidence, cellular-sheaf and mapping-cone diagnostics, advisory geometric audit charts, and a Decentralized Cryptographic Sheaf-Enclave (DCSE) protocol sketch for fail-closed deployment. The framework does not establish clinical truth, global representation alignment, or end-to-end safety; it defines certificate-producing checks at a model-to-system boundary. A companion frozen synthetic benchmark, GBI BoundaryBench v0.1, evaluated Qwen3-4B-Instruct-2507 on 256 held-out tasks across three evidence modes (768 canonical executions). All executions completed, but none produced an output accepted by the benchmark contract: 369 were rejected during safe parsing and 399 during schema validation, yielding zero coverage and deterministic quarantine. This empirical result is deliberately narrow - one 4B open-weight model under one frozen interface - and is reported as evidence about the admission boundary, not as a general claim about LLM capability or clinical safety. A Julia appendix verifies numerical certificates using standard libraries.
[AI-77] oward Human Rights Benchmarking for LLM s: A Pilot Methodology ICML2026
链接: https://arxiv.org/abs/2608.10268
作者: Savannah Thais,Wm. Matthew Kennedy,Abhigyan Acherjee,Matilda Wysocki,Malcolm Langford,Caitlin Kraft Buchman
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注: Best paper, honorable mention, at the ICML 2026 Workshop on AI4Law, Seoul, South Korea
Abstract:Large language models (LLMs) increasingly mediate legal determinations over what human rights are realized, and how. Yet, no evaluation benchmark exists to assess whether they can reason correctly about human rights law. To this end, we report our efforts to develop a robust and scalable methodology for creating HumRightsBench: the first expert-validated, scenario-based benchmark for evaluating reasoning grounded in the obligation structure of international human rights law. We adapt the IRAC framework for legal reasoning to better suit the unique reasoning patterns of human rights work (substituting P, “proposing remedies,” for C, “legal conclusion,” yielding IRAP) to structure our evaluation heuristics. We also produce a pilot series of authentic scenarios designed to implicate the many dimensions of real-world human rights issues and annotated by human rights lawyers and professionals across the world. Ultimately, we find that model accuracy scores range considerably across legal reasoning tasks (overall model performance ranges from 0.339 to 0.577, task min-max ranges from 0.025 to 0.774), which strongly implies that HumRightsBench is a capable instrument for advancing this emerging subfield of AI evaluations science at a critical moment in its evolution.
[AI-78] Interpreting Language Model Hidden States at Scale
链接: https://arxiv.org/abs/2608.10260
作者: Jordan Pettyjohn,Mansi Sakarvadia,Nathaniel Hudson,Daniel McKenzie,Kyle Chard,Ian Foster
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Lens methods interpret large language models (LLMs) by mapping intermediate activations to the output vocabulary, revealing how next-token predictions develop through the network. Trained lenses remain expensive: affine-translator parameters grow quadratically with model width, while exact, full-vocabulary Kullback–Leibler (KL) training dominates memory. Consequently, prior trained lenses have been applied to models of at most 20B parameters and remain tied to particular component types. We present OmniLens, which applies a single lens family to any model-width activation, whether residual stream, attention, or MLP, and combines two independent scaling techniques. First, low-rank translators make per-lens parameter growth linear in model width and reduce trainable parameters by up to 98.4%. Second, Subset-KL materializes only selected vocabulary logits: its Top-k mode cuts peak training memory by up to 70%, while its importance-sampled variant retains unbiased stochastic gradients for the full KL. These savings enable a dense ensemble of 482 lenses for LLaMA-3.3-70B, providing 6x the coverage of a residual-stream design at the same depth. Model-wide coverage then reveals what single-component lenses cannot: the components where a behavior is most visible need not be those where intervention is most effective, and the most effective interventions lie outside the attention heads examined by prior lens studies. Across three case studies (prompt-injection detection, multi-hop memory injection, and toxicity localization), OmniLens reproduces key published results at substantially lower cost.
[AI-79] Beyond Detection: Evaluating Defensive LLM s Against AI-Generated Social Engineering in Live Turn-by-Turn Interaction
链接: https://arxiv.org/abs/2608.10239
作者: Yuqiao Xu,Osama Zafar,Alexander Nemecek,Erman Ayday
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Generative AI makes social-engineering attacks more fluent, adaptive, and scalable, increasing the need for LLM-based de- fenders that can protect users during ongoing interactions. We ask whether such defenders identify the structural source of risk or merely react to surface cues. We formalize trust-chain localization: identifying whether an interaction fails at actor authority, asset control, verification sufficiency, or transaction path. We construct a controlled 300-case online-housing corpus spanning 20 scenario families, legitimate cases, four structural failure modes, and three surface conditions. Five defender models are evaluated on the same corpus in state- ful turn-by-turn and one-shot static settings, yielding 1,500 model-case evaluations per protocol and 3,000 in total. No model produced explicit unsafe compliance, yet defensive effectiveness varied sharply: intervention rates ranged from 0% to 96.3%. Protective action and correct structural localization were frequently decoupled, with models sometimes intervening while identifying the wrong trust component or recognizing a structural failure without taking protective action. Asset-control failures were a major localization bottleneck, surface sensitivity varied across models, and live-static differences were model-dependent. These findings show that safe-looking behavior alone is insufficient; live scam resistance must separately measure intervention, timing, structural localization, and false-positive behavior.
[AI-80] Unsupervised Detection of Groundwater Storag e Anomalies in Ghana Using GRACE Satellite Data CEC
链接: https://arxiv.org/abs/2608.10233
作者: George Yamoah Afrifa,Theophilus Ansah-Narh,Marcellin Atemkeng
类目: Emerging Technologies (cs.ET); Artificial Intelligence (cs.AI); Applied Physics (physics.app-ph); Geophysics (physics.geo-ph)
备注: 8 pages, 7 figures, accepted at ICECCME 2026
Abstract:Groundwater variability in Ghana remains poorly characterized due to limited long-term in-situ observations. This study investigates groundwater storage anomalies using GRACE-derived data from 2004-2024 combined with statistical analysis and unsupervised machine learning. Groundwater anomalies were standardized using Z-scores, while an ensemble-based Isolation Forest framework was applied for anomaly detection. The results revealed substantial temporal variability, with persistent groundwater deficits during 2004-2009 followed by increasing positive anomalies after 2018. A total of 12 anomalous months were identified, comprising 5 deficit and 7 surplus events, with the strongest anomalies associated with groundwater deficits. Spatial analysis showed more frequent deficit anomalies in northern Ghana and stronger surplus occurrence in southern regions. Comparison with statistical thresholds further indicated that the machine learning framework captured additional subtle deviations beyond conventional threshold-based methods. Overall, the integration of GRACE observations with unsupervised anomaly detection provides a practical framework for groundwater monitoring in data-scarce environments.
[AI-81] FACT: Failure-Aware Causal Training for World-Action Models
链接: https://arxiv.org/abs/2608.10232
作者: Quanquan Peng,Yutong Liang,Rui Yan,Nicklas Hansen,Xiaolong Wang
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Recent world-action models (WAMs) show that co-training policies with future prediction can provide physical priors for action generation. Building on the future-prediction ability of video models, many WAMs generate future videos and recover actions with inverse-dynamics models, or use these predicted videos as goal conditions for action generation. In both cases, the world model is trained mostly on successful demonstrations and has little reason to predict the consequences of bad actions. We introduce FACT, a causal World-Action Model that predicts future video and task progress conditioned on the executed action. This action-conditioned interface allows failure rollouts to supervise action consequences, turning bad actions into valid future targets rather than being discarded. Failure-aware training makes the progress predictor aware of both successful and failed action outcomes, which can optionally be used to score sampled action candidates at inference. Extensive experiments on simulation and real-world bimanual manipulation tasks show that FACT outperforms many existing baselines, improves as failure data are incorporated into training, and reduces success-biased future hallucination under bad actions. See more details at this https URL
[AI-82] Self-evolving Agent ic Customer Support System at LinkedIn
链接: https://arxiv.org/abs/2608.10224
作者: Chih Hui Wang,Mengdie Tu,Qianyun Zhang,Wei Wu,Lili Zhou,Mingqi Shen,Changshuai Wei
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Enterprise support agents operate in rapidly changing environments where policies, product capabilities, and knowledge bases evolve continuously, making static assistants brittle and costly to maintain. We present LinkedIn’s self-evolving agentic support system, which integrates retrieval-augmented generation with evolutionary auto-prompting and a modular, production-aligned evaluation framework to enable safe, continuous improvement without retraining foundation models. The system treats prompts, retrieval, and evaluation as a closed-loop, versioned workflow with operational guardrails. Offline simulations and ablations show clear quality gains over vanilla RAG and baseline agents, including reduced hallucinations and improved response completeness. In a two-week user-randomized A/B test on LinkedIn’s production support traffic, the integrated self-evolved workflow increased QA self-serve by 9.0 percentage points, cancellation self-serve by 4.8 points, and routing accuracy by 30.6 points. These results demonstrate a practical path to scalable, self-evolving AI agents in real-world enterprise settings.
[AI-83] Decodable But Not Detachable: Training Data Granularity Determines Parametric Modularity in Large Language Models
链接: https://arxiv.org/abs/2608.10214
作者: Marcus Armstrong,Navid Ayoobi,Arjun Mukherjee
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Do large language models contain domain-specific parametric shells: concentrated, causally necessary neuron populations whose removal selectively degrades a target domain while sparing others? We apply a uniform causal methodology across two domain granularities, three model families (1.5B to 7B parameters), and eight domains. At the academic subject level, zero neurons exceed 60% domain selectivity across 939,008 combined FFN neurons and causal damage matrices are flat, despite domain identity being linearly decodable above 85% accuracy. At the language and modality level, 0.65–1.14% of neurons exceed 60% selectivity, damage matrices are near-perfectly diagonal (ratios up to 595:1), and shell neuron sets are essentially disjoint (IoU 0.003 ). Masking code-selective neurons reduces mathematical reasoning accuracy by 16–24 percentage points across all models; masking Spanish or Chinese neurons leaves it at or below random. Shell strength increases monotonically with scale and shells are spatially interleaved in a pattern that precludes group-level selective quantization. Parametric shells form where and only where training data was modular at the token level.
[AI-84] Evaluation-Conditioned Training: Teaching Models to Generalize to Stronger Oversight Regimes
链接: https://arxiv.org/abs/2608.10209
作者: Alec Harris,Kasey Corra,Archie Chaudhury,Yixiong Hao
类目: Artificial Intelligence (cs.AI)
备注: 16 pages, 7 figures. Accepted at the Agent Behavior Workshop at COLM 2026. Code: this https URL
Abstract:Feedback signals used to train Large Language Models (LLMs) are the primary driver of their behavior and our main lever for instilling alignment with human values and objectives. However, a key limitation of current post-training methods is the inability of human annotators and automated reward functions to faithfully capture the feedback we would like to give. We introduce Evaluation-Conditioned Training (ECT), a post-training framework that uses natural language to condition each training sample on the fidelity of the feedback we provide and then elicits the desired behavior by conditioning the LLM on a high-fidelity monitor in deployment. ECT is aimed at improving performance under imperfect feedback and works as an add-on to existing algorithms such as SFT and PPO. We first provide a conceptual framework for ECT and discuss its potential to address persistent sources of reward mis-specification. Then we motivate ECT in the context of the eliciting latent knowledge (ELK) problem. Finally, we evaluate ECT on two proof-of-concept experiments: increasing even-handedness in news article generation and reducing sycophancy on an arithmetic task. In each setting, we utilize imperfect feedback, rewarding bias and agreement with the user, respectively. In both settings, ECT improves the targeted behavior relative to direct training.
[AI-85] Mitigating Bus Bunching with Reinforcement Learning Enhanced by Semantic Stop Embedding
链接: https://arxiv.org/abs/2608.10207
作者: Xin Dong,Vikash V. Gayah
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Bus bunching degrades service regularity and increases passenger waiting in high-frequency transit. Existing reinforcement-learning-based holding controllers primarily rely on instantaneous operational variables or route-specific stop identifiers, which provide limited information about the functional and operational context of individual stops and constrain policy reuse across routes. This study introduces an LLM-assisted semantic stop representation for event-driven bus holding control. An LLM is used offline to transform heterogeneous stop information, including physical attributes, surrounding activity context, and historical operational characteristics, into fixed semantic embeddings that are incorporated into a deep Q-learning controller without requiring real-time LLM inference. Experiments are conducted in stochastic simulations calibrated with observed data from two bus routes. Compared with the best calibrated Daganzo baseline, the semantic controller reduces headway variability, bunching events, and passenger waiting time by 32.0%, 69.2%, and 24.0%, respectively. A route-specific stop identifier does not improve the spacing-only controller, whereas semantic stop information improves headway regularity, waiting time, and holding effort, providing a more favorable overall trade-off across control objectives. Cross-route experiments further show that zero-shot transfer provides limited immediate generalization, while warm-start fine-tuning accelerates early-stage learning and improves transferred policies; cold-start training nevertheless achieves the best final performance. These findings suggest that semantic state representations can complement conventional operational states and support adaptation-based policy reuse across related transit routes.
[AI-86] Post-Hoc Sparse Coding of Latent Communication Between Vision-Language Model Agents
链接: https://arxiv.org/abs/2608.10198
作者: Di Wu,Xiaohui Zhu
类目: Artificial Intelligence (cs.AI)
备注: 9 pages, no figures
Abstract:Latent-space communication allows heterogeneous vision-language model agents to exchange continuous representations without serializing visual and reasoning states into text. Vision Wormhole realizes this approach by translating visual features into a universal latent representation that can be consumed by another model, but every message is transported as a dense tensor of the same size regardless of its content. A fixed-capacity dense tensor therefore need not have a fixed effective information density: some messages may use only a small fraction of the available representational degrees of freedom. This observation suggests that the communication channel may be substantially compressible. We study its redundancy by fitting a post-hoc sparse autoencoder to frozen Vision Wormhole activations and measuring reconstruction, downstream utility, feature reuse, and token-level interventions across nine reasoning benchmarks. Relative to the original float32 transport, a uint16-index/float16-value sparse payload with k=4 active coefficients per token reduces the transmitted bytes by 128x. In a single-run evaluation, the seven-task non-AIME mean accuracy changes from 49.85% to 49.77%. The fitted 4096-element dictionary uses only 50 features, and task-level active sets have a mean pairwise Jaccard similarity of 0.906. These measurements establish strong post-hoc compressibility relative to the original transport, but do not yet isolate the incremental contribution of sparse coding from position selection, reduced precision, low-rank structure, or SAE optimization effects. The results motivate matched-payload comparisons and communication mechanisms whose payload adapts to the information used by each message.
[AI-87] ELMER: Evolutionary Language Model that Explores and Refines AAAI
链接: https://arxiv.org/abs/2608.10196
作者: Matthew Siper,Ahmed Khalifa,Julian Togelius
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Submitted to AAAI Conference 2026, 8 pages, 6 figures, 1 table
Abstract:Program evolution can measure whether a mutation helped, but it rarely controls how far the mutation moves in behavior space. Syntactic edit size is an unreliable proxy: a small code change can alter nearly every action, while a larger rewrite can preserve the same execution trace. We introduce an Evolutionary Language Model that searches over natural-language policy descriptions and compiles typed programs for execution. A fully fine-tuned Qwen3-8B model learns three task-conditioned operations: conditional semantic mutation, natural language to domain-specific language (GPTL) compilation, and GPTL to natural language translation. The model is fine-tuned with conditional input on the mutation strength (low, medium, high) using Direct Preference Optimization (oDPO). Across 252 fixed-budget evolutionary searches, oDPO improves both behavioral calibration and finite-budget search efficiency. Natural-language attains the highest observed held-out fitness. Our analysis shows that the condition input (mutation strength) systematically changes semantic edit composition and that language mutations preserve more parent fitness at matched small-to-moderate behavioral displacement. These results show that language can serve as a steerable, execution-grounded search representation over executable program space.
[AI-88] From Prediction to Incrementality: Causal Optimization for Large-Scale Targeting and Recommendation
链接: https://arxiv.org/abs/2608.10182
作者: Changshuai Wei,John Bencina,Phuc Nguyen,Andre Assuncao Silva T Ribeiro,Benjamin Zelditch
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Large-scale targeting and recommendation systems are typically built around predictive scores fed into heuristic or local allocation. When the business goal is incremental impact, as in marketing campaigns, incentives, and notifications, this paradigm systematically misallocates resources toward users who would have acted anyway. We present a decision-centric framework that instead optimizes causal effects under global constraints, aligning three components under a single objective: a causal neural network with a Transformer backbone for individual treatment-effect estimation, a Bayesian neural-bandit layer for uncertainty-aware exploration, and a dual-based large-scale linear-programming layer for constrained allocation. The framework also supports sequential context and multi-outcome, attribute-conditioned scoring through a Transformer encoder and outcome embeddings. We evaluate it with offline simulations on a public bandit dataset, targeted architectural ablations, and an online A/B test on LinkedIn Feed marketing traffic. We also distill production lessons on causal training-data construction and cost and delivery control, which were critical to successful deployment. The end-to-end treatment policy delivered a statistically significant +7.20% lift in the primary long-term-value metric, demonstrating the feasibility of production-scale causal optimization under business constraints.
[AI-89] RACE: Trustworthy Retrieval-Augmented Conversational Engine
链接: https://arxiv.org/abs/2608.10176
作者: Touseef Hasan,Laila Cure,Souvika Sarkar
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Public service chatbots are expected to deliver recommendations from an underlying public service directory, while also making sure that the recommendations respect explicit user constraints. In practice, public service directories are noisy and inconsistent, and general-purpose large language model (LLM) or AI-based chatbots frequently generate unreliable recommendations, citing unverified sources from the web. We investigate the impact of retrieval quality on constraint-aware recommendation in public service conversational systems built over noisy and heterogeneous service directories. We propose TRACE (Trustworthy Retrieval-Augmented Conversational Engine), a retrieval-based, constraint-aware framework that parses input user queries into structural and semantic constraints for downstream retrieval, with the help of a dual data representation schema. Using a curated statewide pantry directory and a synthetic query benchmark, we evaluate multiple knowledge-representation variants with and without knowledge graphs (KGs). We experiment with several open-source LLMs and a proprietary model, showing that strengthening retrieval substantially improves user constraint satisfaction while reducing hallucinated recommendations. Performance differences across LLMs narrowed in our experiments as retrieval quality improved, making results less sensitive to model size. These findings suggest that the quality of retrieval is key for robust public service conversational systems.
[AI-90] Generating Attacks for LLM s with GFlowNets
链接: https://arxiv.org/abs/2608.10171
作者: Berkay Ozcam,Irem Onen,Mehmet Fatih Amasyali,Emin Islam Tatli
类目: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注:
Abstract:The rapid advancement of Large Language Models (LLMs) has facilitated their ubiquitous integration into various domains, leading to widespread adoption. However, this escalating trend has introduced significant security vulnerabilities, necessitating the identification and mitigation of flaws arising from malicious exploitation. Red teaming assessments, conducted to evaluate model robustness through diverse adversarial inputs, are essential for exposing security risks and implementing countermeasures. Currently, red teaming is performed either manually by experts or automatically using predefined attack datasets. Nevertheless, manual testing remains time-consuming, while existing automated methods suffer from limited creativity due to their inherent dependency on fixed datasets. In this study, we propose an automated, human-independent, and adaptive approach leveraging GFlowNets to identify LLM vulnerabilities by utilizing one large language model to test another. Within this framework, an attacker model is trained against a specified victim model to perform automated red teaming and provide a quantitative robustness score. This research aims to generate more effective adversarial attacks in English compared to existing benchmarks and, as a novel contribution to the literature, introduces a model capable of generating attack inputs in the Turkish language.
[AI-91] MarkNull: Model-Agnostic Watermark Removal in AI-Generated Images via On-Manifold Latent Manipulation USENIX-SECURITY2026 USENIX-SECURITY
链接: https://arxiv.org/abs/2608.10166
作者: Jie Cao,Qi Li,Zelin Zhang,Xiaodong Wu,Lingshuang Liu,Xiangman Li,Jianbing Ni
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: Accepted to the 35th USENIX Security Symposium (USENIX Security 2026)
Abstract:Digital watermarking has emerged as a critical technique for provenance and copyright attribution in AI-generated imagery, yet its robustness against realistic, model-agnostic removal attacks remains poorly explored. Existing attacks either succeed only against specific generative models or achieve removal at the cost of severe visual degradation. In this paper, we propose MarkNull, a model-agnostic watermark removal attack via on-manifold latent manipulation. MarkNull is grounded in a key observation: watermarked images exhibit a strong statistical dependency between the generated latent representation and the embedded initial noise. To quantify this dependency, we introduce the Noise-Latent Alignment Score (NLAS) and formulate an optimization objective that selectively decorrelates the latent representation from the embedded watermark while preserving semantic fidelity. Extensive evaluations across different categories of watermarking paradigms, including post-hoc, fine-tuning-based, and initial-noise-based schemes, demonstrate that MarkNull reduces average bit accuracy to 53.14%, approaching random-guessing (50%), without perceptible image degradation. To further improve scalability, we propose MarkNull-A, an amortized, optimization-free variant that distills the attack into a single forward pass, achieving 0.50 s/image with modest computational overhead. Notably, our attacks successfully compromise Google’s SynthID-Image system while preserving high visual quality and transfer effectively to video watermarking. Finally, we present an attack detection mechanism as a defensive counterpart to MarkNull and MarkNull-A, highlighting the necessity of developing watermark designs resilient to model-agnostic latent-space attacks.
[AI-92] SBCO: Self-Supervised Verifier-Grounded Harness Optimization For Planning Agents
链接: https://arxiv.org/abs/2608.10157
作者: Vivek Kulkarni,Sudipta Paul,Aounon Kumar,Nicholas Tzou,Srinivas Chappidi
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Self-improving agents seek to reduce the human engineering effort behind AI systems by enabling them to evolve and self-improve their performance over time. Recently, methods like the Darwin Gödel Machine and the Huxley Gödel Machine have been proposed which enable open-ended, recursive self-improvement through self-reference where a coding agent edits its own code. Such self-referential self-improvement methods require that the competence required to perform the task coincides or aligns well with the competence required for self-modification which is the case for coding tasks. For domains or tasks, which do not satisfy the alignment needed, self-referential self-improvement is not available. In such cases, it is possible to adapt the above algorithms to other tasks by removing the self-referential aspect or introducing explicit self-modification of a meta-agent – both computationally expensive, relying on population or self-modification search over many candidate agents. For planning tasks with explicit constraints, we propose a far cheaper alternative. We introduce SBCO (Self-supervised Block Coordinate Optimizer), a verifier-grounded harness optimizer in the same closed-loop, improve-from-experience family as the Gödel-machine methods, but self-supervised rather than self-referential. Given an agentic harness, SBCO learns a decomposed bank of verifiers and a harness policy via approximate block coordinate ascent, improving the agent’s outputs from its own graded feedback—with a fixed meta-agent and no human labels. Across two domains SBCO matches or exceeds a customized self-modifying baseline while using 4-5.5 times less compute budget.
[AI-93] he CASE Framework: A Multi-Disciplinary Control Architecture for Governing Enterprise Agent ic AI
链接: https://arxiv.org/abs/2608.10153
作者: Srinivas Telukunta,Georgios Nektarios Lilis,Lucio Baron
类目: Artificial Intelligence (cs.AI)
备注: 34 pages in total, 4 figures, 2 tables, code at this https URL
Abstract:Enterprises are deploying autonomous AI agents faster than they can govern them, and prevailing approaches stretch a single discipline, typically DevSecOps built for deterministic automation, across every scale of agency. We argue that agentic AI governance is four problems, not one, each with a mature governing science. The CASE framework assigns Control theory to the individual agent (intent as setpoint, guardrails as feedback, evaluation as observation), complex Adaptive systems theory to agent collectives (where emergence makes single-agent assurance non-compositional), Supervisory cybernetics to human-agent teams (where the Law of Requisite Variety shows unaided human oversight fails structurally), and Engineering operations to fleets (extending error budgets to decision quality so autonomy becomes a controlled variable). We formalize each layer, derive cross-layer coupling conditions, including a zero-touch deployment paradox where excellence at one-layer strains the others, and trace twenty-plus enterprise controls to their classical constructs. Three empirical studies validate the thesis: 82 percent of documented production agent failures are multi-layer trajectories; none of 22 ecosystem tools offers full Layer 2 (emergence) coverage; and all 35 scored public deployments fall in the lowest maturity band. We name this mismatch, risk realized at the emergence layer against capability barely offered and practice absent, the Emergence Gap. A five-level maturity model with a non-compensatory bottleneck-weighted index and assessment instrument operationalizes CASE as a scientific rather than process maturity model, grounded in production enterprise agentic platforms. As EU AI Act Article 14 makes effective human oversight a legal requirement, only architectures satisfying requisite variety can make oversight real rather than ceremonial.
[AI-94] MESA:Task-Adaptive Multi-Structure Evidence Selection for Long-Horizon Agent Memory
链接: https://arxiv.org/abs/2608.10108
作者: Beidi Zhao,Yaoqi Chen,Yuru Feng,Menghao Li,Qianxi Zhang,Baotong Lu,Jianan Lu,Zhirui Wang,Xinjiang Wang,Shusen Xu,Zengzhong Li,Xiaoxiao Li,Qi Chen
类目: Artificial Intelligence (cs.AI)
备注: 18 pages, 5 figures
Abstract:Long-horizon agents accumulate trajectories spanning hundreds of interleaved reasoning, action, and observation steps, where answering a query may depend on evidence buried far back in the history. External memory stores such trajectories as structured representations, yet each structure provides a distinct and incomplete view. Existing multi-memory systems either read a fixed set of structures for every query, inflating context and introducing noise, or route each query to a single structure, preventing the composition of complementary evidence. A controlled analysis on AMA-Bench shows that the optimal memory configuration is typically neither a single structure nor the full union, but a tailored composition of multiple structural memories that varies with query and task demands. Motivated by these findings, we formulate structure-level dynamic selection: selecting and fusing a query-adaptive subset from a library of specialized memory structures. We propose MESA (a Multi-structure Evidence Selection framework for long-horizon Agent), which builds five complementary structure views of each trajectory and learns from end-to-end answer-level feedback to select and fuse a query-specific subset for a frozen answer model. To learn under this weak supervision, MESA employs harness optimization with prior-guided search and UCB-guided scheduling to balance exploration and exploitation. On AMA-Bench, MESA outperforms the strongest baseline by 8.5% while using 41% fewer evidence tokens than the all-structure alternative.
[AI-95] Hand-Written PTX Tensor-Core GEMM Kernels: A Multi-Precision Study on NVIDIA L4
链接: https://arxiv.org/abs/2608.10103
作者: Matt J. Borowski,Blazej Osinski
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Artificial Intelligence (cs.AI)
备注: 10 pages, 6 figures, 12 tables. Code and profiling data: this https URL (archived: this https URL )
Abstract:High-performance Tensor Core kernels rely on a low-level PTX pipeline built from asynchronous data movement with this http URL, warp-level matrix loads with ldmatrix, and matrix multiply-accumulate operations with this http URL. However, most application code accesses Tensor Cores indirectly through the WMMA C++ API. This paper asks a focused, practical question: when does replacing WMMA with hand-written PTX actually pay off? To answer this question, we conduct a controlled, single-GPU study on an NVIDIA L4 GPU (Ada, SM89), comparing double-buffered WMMA baselines with a family of hand-written PTX GEMM kernels across FP16, INT8, and INT4 arithmetic and square problem sizes from N=512 to N=8192 . Every kernel is profiled with Nsight Compute across the full metric set, and PTX speedups are reported relative to the corresponding same-precision WMMA baseline. Hand-written PTX provides no end-to-end speedup for FP16, because its instruction-level gains are offset by operand-packing overhead. In contrast, the PTX kernels achieve consistent speedups of 1.4x-1.8x for INT8, driven primarily by lower instruction counts and better global-memory coalescing, and 2.9x-4.3x for INT4, where native mma.sync.m16n8k64.s4 execution avoids the software-emulated sequence used by the WMMA path. Relative to the FP16 WMMA baseline, the best quantized kernels reach 34.4x (INT8) and 98.7x (INT4) at N=8192 . Across these experiments, occupancy is a poor predictor of throughput. For large matrices, performance instead tracks memory-system behavior – particularly global-load coalescing and DRAM-active cycles – more closely than Tensor Core utilization. These results identify the precisions and operating regimes in which the additional complexity of hand-written PTX is justified.
[AI-96] Exploring Semantic Stability Across Reviews in the Linux Kernel
链接: https://arxiv.org/abs/2608.10101
作者: Lucas Ciziks,Paulo Meirelles,Marco Aurélio Gerosa
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Systems and Control (eess.SY)
备注: VEM 2026 - 14th Workshop on Software Visualization, Maintenance and Evolution
Abstract:Code review is credited with substantially changing a patch’s code between its first submission and the version that eventually lands. However, prior work typically studied only the final merged patch without comparing it to the first submission. We present a function-level measurement that tracks 10,117 trajectories (each function followed across the numbered revisions of one patch series) through the patch history of the Linux IIO subsystem, comparing similarity scores against unrelated function pairs as a baseline. A naive reading yields near-total similarity, but this is largely an artifact of composition: 75.3% of tracked trajectories are never textually modified between versions, contributing a trivial 100% similarity that inflates the headline. Restricting to the trajectories with a real edit, semantic purpose is still largely preserved (mean similarity 0.990 vs. a 0.909 baseline), but drift appears to concentrate in the first review round mainly because later rounds contain more functions that nobody touched, not because edits become more conservative over time. After controlling for it, a statistically detectable but small residual effect remains. This points to an open question: whether near-ceiling similarity reflects preserved purpose or a measurement tool that cannot detect the significance of small, localized edits. We present this work as a first look and outline next steps.
[AI-97] CHORUS: Complementary Experts for High-Coverag e Testbench Stimulus Generation
链接: https://arxiv.org/abs/2608.10090
作者: Hejia Zhang,Sheng Lu,Zhongming Yu,Chia-Tung Ho,Brucek Khailany,Jishen Zhao
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Large language models (LLMs) have advanced code generation, where executable feedback provides a more reliable learning signal than textual imitation alone. Hardware verification is an important application of code generation and accounts for a substantial fraction of modern chip design effort, with high-coverage testbench stimulus generation as a key task. We present CHORUS, a post-training framework that pushes performance beyond what a conventional supervised fine-tuning (SFT)-to-reinforcement learning (RL) pipeline achieves. CHORUS builds on two observations. First, staged SFT produces behaviorally diverse checkpoints, and dense-reward RL turns them into strong experts with comparable aggregate performance but distinct task-level strengths. Second, these complementary strengths can be exploited through either training-free model merging or further post-training to outperform the best individual expert. By consolidating the resulting specialists into a single 4B model, CHORUS achieves 88.0% Pass@1 on CVDP-ECov, outperforming DeepSeek-R1 (671B) by 13.5 percentage points.
[AI-98] Navigating the Proximity-Safety Balance: Constraint Decomposition for Human Following in Pedestrian Crowds IROS2026
链接: https://arxiv.org/abs/2608.10056
作者: Shiting Gong,Jianpeng Yao,Jinfeng Wang,Marco Pavone,Jiachen Li
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Systems and Control (eess.SY)
备注: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2026); Project Website: this https URL
Abstract:Following a target human in crowded environments involves an inherent conflict between staying close to the target and navigating safely among surrounding pedestrians and obstacles. This conflict becomes more severe in dense scenarios, where aggressive following risks collisions and conservative margins lead to target loss, especially when pedestrian behaviors are unfamiliar or unpredictable. Existing reinforcement learning (RL) methods typically encode these competing objectives into a single dense reward, but the resulting proximity-safety balance is implicit and difficult to adjust across conditions. To address this, we decompose the human-following task into a sparse task reward and independent cost constraints within a multi-constraint RL formulation, where each constraint is managed through cost thresholds with direct behavioral meaning rather than implicit reward weight ratios, allowing explicit and tunable control over the trade-off. We further quantify the prediction uncertainty of human motions and integrate these estimates into the RL costs to enhance safety under unpredictable conditions. Extensive experiments across both in-distribution and out-of-distribution settings demonstrate that our method achieves an effective proximity-safety balance compared to baselines. Real-robot deployment further validates the feasibility of our method in real-world scenarios. More details are available on our project page: this https URL.
[AI-99] Physics-Informed Machine Learning in Prognostics and Health Management: A Systematic Literature Review
链接: https://arxiv.org/abs/2608.10047
作者: Christopher Braun,Julian Raible,Marco F. Huber
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: The version of record of this article, first published in Journal of Intelligent Manufacturing, is available online at: this https URL . This arXiv version is content-equivalent but differs in four respects: (i) additional appendix; (ii) disambiguated citations; (iii) numbered section headings; and (iv) figures with selectable and searchable text
Abstract:In modern industry, keeping complex systems reliable, safe, and efficient hinges on Prognostics and Health Management (PHM). Machine Learning (ML) has largely driven advancements in diagnostics and prognostics, yet purely data-driven models face inherent limitations, such as poor generalization, an inability to infer causal relationships, and a lack of interpretability. Physics-Informed Machine Learning (PIML) helps mitigate these limitations by incorporating prior physical knowledge directly into the ML pipeline, thereby fostering growing interest in its application to PHM. This work investigates how PIML is being leveraged in the context of PHM through a systematic literature review of 212 studies. The review introduces a four-class classification scheme, consisting of observational bias, inductive bias, learning bias, and hybrid approaches, and further categorizes studies by PHM task. Across all four classes, the reviewed studies consistently demonstrate improved predictive performance over conventional baselines across a broad range of assets, although the literature is heavily skewed toward lithium-ion batteries and bearings, and dominated by problem-specific solutions. Overall, the review indicates that physics-informed approaches already provide tangible benefits, whereas claims of improvements concerning some of the aforementioned limitations lack sufficient supporting evidence. Future research should prioritize transferable design patterns, benchmarks comparing integration strategies, and uncertainty-aware models that are lightweight and robust enough for online deployment in real-world settings.
[AI-100] Finding the Signal in the Spam: Jointly Learning Rewards and Worker Reliability from Pairwise Comparisons UAI2026
链接: https://arxiv.org/abs/2608.10045
作者: Kaustubh Shivshankar Shejole,Tanish Agarwal,Arpit Agarwal,Avishek Ghosh
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: Accepted at the 42nd Conference on Uncertainty in Artificial Intelligence (UAI 2026), Amsterdam, Netherlands, August 17-21, 2026
Abstract:The problem of learning from pairwise comparisons has been widely studied across many domains such as recommendation systems, social choice, and more recently, fine-tuning large language models. In this problem, the goal is to learn item rewards based on pairwise comparisons between them. In many scenarios, these comparisons are elicited from crowdworkers using platforms such as Amazon Mechanical Turk, Scale AI, etc. However, crowdworkers are often unreliable due to limited domain knowledge or revenue-maximizing (spamming) behavior. In this work, our goal is to understand whether worker reliability (competency) can be learned jointly with item rewards. To this end, we adopt the Boltzmann-rational model for pairwise comparisons, which extends the Bradley-Terry-Luce model by incorporating worker competencies. We derive an EM-based algorithm for learning under this model by introducing Polya-Gamma latent variables to transform the logistic likelihood into a conditionally Gaussian form, enabling tractable optimization and leading to a simplified Q function in the E-step of the algorithm. This technique allows us to reduce our formulation to a matrix sensing problem, using which we establish theoretical convergence guarantees for our algorithm. We conduct extensive experiments on real-world and synthetic datasets. These experiments demonstrate the advantages of using our algorithm over several baselines and confirm its strong robustness to both spammers and adversarial workers, highlighting its practical effectiveness in realistic crowdsourcing and reward learning settings. The code and data is publicly available at this https URL.
[AI-101] UserToolBench: A User-Profile-Hidden Benchmark for Personalized Decision Making in Tool-Use LLM s
链接: https://arxiv.org/abs/2608.10042
作者: Xuexiong Yin,Zechuan Chen,Yongsen Zheng,Yuxiang Zhang,Jingyuan Yang,Bin Wang,Yubin Wang,Keze Wang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 21pages,4figures
Abstract:Tool-use LLMs are increasingly asked to act on users’ behalf, but existing benchmarks usually focus on profile recall, style imitation, generic tool use, or response-level personalization. We introduce UserToolBench , a benchmark for personalized decision making in tool-use LLMs. UserToolBench tests whether a model can infer latent user preferences from interaction history, recognize when clarification is needed, and produce user-aligned tool-call trajectories under incomplete information. The benchmark is built from privacy-sanitized real interaction traces and combines structured persona profiles, public API-style tool ecosystems, and long-horizon multi-turn trajectories. It includes 10 user profiles, 36 tool sets, 1,065 turns, 170 unique tools, and evaluation-focused task types covering lack-of-information, single-tool, and multi-tool settings. Experiments with strong tool-use LLMs show that current models still have difficulty with personalized delegation. Multi-tool coordination, missing-constraint inference, and long-horizon behavioral consistency remain major bottlenecks. These results suggest that personalization evaluation should move beyond asking whether outputs sound user-specific and instead ask whether LLMs make correct decisions for the users they represent.
[AI-102] DOCSCHISEL: Adaptive Tool Documentation Optimization Framework for LLM Agents
链接: https://arxiv.org/abs/2608.10037
作者: You Lu,Kun Zhang,Bihuan Chen,Xin Peng
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models (LLMs) increasingly rely on external tools to accomplish complex real-world tasks, making tool documentation a critical grounding resource for LLM agents. Existing studies mainly focus on improving the tool-use capabilities of LLM agents, while largely treating tool documentation as a fixed input. Although several recent works attempt to optimize tool documentation through rewriting or compression, little is known about how the information contained in tool documentation affects agent performance across different settings. To bridge this gap, we conduct a large-scale empirical study on tool documentation for LLM agents. Our study reveals substantial heterogeneity in the information fields provided by existing tool documentation. Moreover, the effectiveness of different information fields is highly dependent on the task domain, LLM backbone, and agent paradigm, indicating that no fixed tool documentation can consistently generalize across diverse agent settings. Motivated by these findings, we propose DocsChisel, an adaptive tool documentation optimization framework for LLM agents. DocsChisel analyzes failed execution traces of a target LLM agent to identify documentation-related issues, and iteratively optimizes tool documentation by adding, removing, and refining information fields for each tool. We evaluate DocsChisel against two state-of-the-art baselines, i.e., EasyTool and DRAFT. Experimental results show that DocsChisel improves the task success rate of LLM agents by 95.89% over the original tool documentation and by 75.15%, on average, over existing baselines, while incurring limited optimization time and token overhead Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI) Cite as: arXiv:2608.10037 [cs.LG] (or arXiv:2608.10037v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.10037 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-103] Uncertainty-Aware Ensemble Deep Randomized Neural Networks for Classification
链接: https://arxiv.org/abs/2608.10007
作者: M. Sajid,A. Quadir,A. Rahaman,P. N. Suganthan,M. Tanveer
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:The current state-of-the-art (SOTA) deep randomized neural networks, such as deep Random Vector Functional Link (dRVFL) and ensemble deep RVFL (edRVFL), treat all training samples uniformly, which limits their robustness and effectiveness when applied to real-world datasets containing noise and outliers. Furthermore, the propagation of contaminated features across hidden layers negatively influences the decision-making capability of these models. To overcome these limitations, we propose intuitionistic fuzzy dRVFL (IF-dRVFL) and intuitionistic fuzzy edRVFL (IF-edRVFL) frameworks that enhance model robustness. The proposed models unify intuitionistic fuzzy theory to exploit sample neighborhood information in the kernel space by jointly considering membership and non-membership degrees for each sample. Membership degrees are computed based on the distance of samples from their respective class centroids, while non-membership degrees quantify sample heterogeneity within local neighborhoods. These measures are employed to assign adaptive weights to training samples, enabling effective discrimination among clean, noisy, and outlier data points. Extensive experiments conducted on UCI and KEEL benchmark datasets, with and without the presence of Gaussian noise, demonstrate the superiority of the proposed IF-dRVFL and IF-edRVFL models over existing SOTA fuzzy and non-fuzzy approaches. The source code is available at this https URL.
[AI-104] owards Sustainable Artificial Intelligence: A Comprehensive Review and Comparative Analysis of Deep Learning Models Carbon Footprint
链接: https://arxiv.org/abs/2608.09998
作者: Samar Garrab,Sarra Boughriou,Manel BenSassi
类目: Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Machine Learning (cs.LG); Software Engineering (cs.SE)
备注: Published in Applied Intelligence. DOI: this https URL
Abstract:Artificial Intelligence (AI) and Machine Learning (ML) have become powerful tools for supporting and automating complex human tasks. Despite their benefits, growing attention has been directed toward their environmental implications, primarily due to their high energy demands and associated carbon emissions. This concern is particularly relevant in light of the increasing deployment of large-scale models, especially Deep Learning (DL) architectures, which provide advanced predictive capabilities but require substantial computational resources. This paper presents a systematic review of research on Green AI, Green DL, and optimization techniques aimed at reducing the environmental impact of AI models. In addition, we examine and compare several carbon measurement tools for estimating emissions generated by AI algorithms. To complement the review, we conducted an empirical evaluation using a CPU-based experimental setup, in which six DL models were implemented for a multi-label classification task. The objective was to quantify and compare their overall carbon emissions and to determine which stages of the DL lifecycle contribute most significantly to the total footprint. The results show that the training phase is the primary source of emissions. Moreover, the findings reveal that increased architectural complexity does not systematically translate into proportional accuracy gains, highlighting the importance of carefully balancing predictive performance and environmental cost. These results reinforce the need to integrate sustainability considerations into model selection and AI system design.
[AI-105] MIDAS: Mutual Information Disentanglement with Uncertainty-Aware Fusion for Incomplete Multimodal Sentiment Analysis
链接: https://arxiv.org/abs/2608.09986
作者: Yuhua Wen,Yingying Zhou,Qifei Li,Yingming Gao,Zhengqi Wen,Jianhua Tao,Ya Li
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted to IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2026
Abstract:Most existing multimodal sentiment analysis approaches assume access to complete multimodal inputs. However, real-world applications frequently encounter incomplete or corrupted modalities, posing a critical challenge. Although several methods have been proposed to tackle this issue, they mainly rely on data imputation and heuristic coordination constraints, which fail to effectively extract and leverage task-relevant information from the incomplete multimodal data. To address this challenge, we propose a unified framework termed Mutual Information Disentanglement with uncertainty-Aware fuSion (MIDAS), which effectively restructures multimodal representations under incomplete conditions. MIDAS adopts a variational modeling strategy to represent each modality with multivariate Gaussian latent variables and further decomposes them into shared and exclusive factors. To obtain reliable representations, we design a minimax objective that minimizes the mutual information between shared and exclusive spaces for stable disentanglement, while maximizing the mutual information among shared spaces across modalities to enhance semantic alignment. In addition, an uncertainty-aware fusion mechanism is introduced, where posterior variance is leveraged as a reliability indicator to adaptively weight latent features during fusion, ensuring robust integration even when modalities are incomplete. Extensive experiments on three widely used datasets show that MIDAS achieves strong and consistent performance gains over competitive baselines across a wide range of incomplete settings, demonstrating its effectiveness and robustness for incomplete data scenarios.
[AI-106] Evidence-Based Scientific Question Discovery: A Framework with Historical Backtesting
链接: https://arxiv.org/abs/2608.09968
作者: Hui Mao
类目: Digital Libraries (cs.DL); Artificial Intelligence (cs.AI)
备注: 13 pages, 4 figures
Abstract:Current AI systems are optimized for answering questions; the scientific enterprise is bottlenecked earlier, at discovering the questions worth investigating. We present a framework that turns a traceable, reproducible, scope controlled research corpus into ranked, falsifiable research questions: evidence is represented as provenance carrying claims; cross paper tensions are detected, typed, and human adjudicated; surviving signals are refined into questions and ranked by a two stage protocol separating scientific priority from execution priority. We instantiate the framework on exoplanet atmospheres, a domain that uniquely combines literature, structured catalogs, and space telescope archives. In a historical backtest, all questions generated from evidence available before 2021 were substantively engaged by the 2021 to 2026 literature the sys?tem never saw: two were answered, including one whose premise the community later explicitly refuted and the top ranked question is independently posed and still open. These results sug?gest that systematic question discovery from evidence tensions surfaces the questions working scientists subsequently invest in.
[AI-107] SPOTting the Future: Lookahead Explanations for Deep Reinforcement Learning
链接: https://arxiv.org/abs/2608.09967
作者: Tamar Gozlan,Claudia V. Goldman
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Deep reinforcement learning (DRL) agents achieve strong performance in complex environments, yet their decision-making processes remain difficult to interpret. We introduce SPOT (Sampling Policy Observation Tree), a novel model-agnostic, sampling-based framework for interpreting DRL policies. Given access to the policy and an environment simulator, SPOT constructs an interpretable finite-horizon tree by sampling actions and recursively simulating the resulting successor states. The tree provides an empirical representation of the policy’s action preferences and their possible downstream evolution. We provide formal guarantees establishing SPOT’s asymptotic recovery of the policy’s unique most probable action and characterizing its disagreement behavior under high-entropy policies. We demonstrate SPOT in the SUMO-RL traffic-signal control domain. The case study illustrates how its tree-based representation can be used to inspect policy preferences, compare alternative future trajectories, and reveal downstream behaviors that are not visible through single-timestep feature-attribution methods.
[AI-108] Eleven Years of BRACIS: A Meta-Scientific Study of the Brazilian Conference on Intelligent Systems
链接: https://arxiv.org/abs/2608.09964
作者: Thales Sales Almeida,Giovana Kerche Bonás,Thiago Laitz,João Guilherme Alves Santos,Hugo Abonizio,Roseval Malaquias Junior,Marcos Piau,Celio Larcher,Ramon Pires,Rodrigo Nogueira
类目: Digital Libraries (cs.DL); Artificial Intelligence (cs.AI)
备注:
Abstract:The Brazilian Conference on Intelligent Systems (BRACIS) is the main national venue for Artificial Intelligence research in Brazil, hosted by the Brazilian Computer Society since 2012 and publishing work from institutions across the country. Across eleven years, from 2015 to 2025, we build a per-paper record of all 1,066 accepted papers from DBLP metadata, 6,765 Google Scholar citations, and the paper full texts, and use it to ask what BRACIS publishes, who publishes it, and which work gets cited. Large Language Model research grows from zero before 2020 to 19% of papers in 2024, on top of a base of Machine Learning, Computer Vision, and Optimization work. The community is hourglass-shaped: 80.5% of 2,623 authors appear in a single edition, while institutions return at nearly three times the author rate. Citations are heavily concentrated, with the top 1% of papers carrying 27% of the total. Openness practices have grown, with artifact release rising from 8.9% of papers in 2015 to 57.3% in 2023, and we find a notable correlation between having an arXiv preprint and higher citation counts. Since proceedings sit behind IEEE and Springer paywalls and only 7.4% of papers have a preprint, most BRACIS work is hard to reach for readers without institutional access.
[AI-109] Closed-Loop LLM Co-Pilots for Digital Agriculture
链接: https://arxiv.org/abs/2608.09949
作者: Serge Kernbach
类目: Artificial Intelligence (cs.AI); Biological Physics (physics.bio-ph)
备注:
Abstract:This study evaluates the application of Large Language Models (LLMs) in complex biological systems, evolving from data analysis to autonomous, AI-guided experimentation. The framework is driven by data from a 49-channel phytosensor network, encompassing multispectral, electrochemical, and dielectric modalities. To enhance accessibility, the system provides real-time natural-language interpretation for both specialists and non-experts. However, its core advantage lies in the transition from human-in-the-loop analysis to autonomous control. Processing biophysical data, the LLM evaluates plant physiology and triggers hardware actuators to optimize microclimates, execute phenotyping protocols, or induce controlled stress scenarios. This closed-loop architecture establishes a direct AI-biology interface, enabling data-driven exploration of complex biosystems and ecologies. The framework was validated across three case studies, based on a vertical farm and a single-plant setup and deciphered complex micro- and macro-fluctuations in plant physiology. Agents in a production-scale deployment executed multi-parameter optimization, balancing biomass accumulation, chlorophyll content, and energy consumption. The LLM processed biosensing telemetry to modulate full-spectrum, 450 nm, and 660 nm lighting at 2-hour intervals. Compared to periodic control, the system in minimal-time mode reduced the production cycle by 35%. In the energy-optimization mode, it reduced energy consumption by 18% with only a marginal increase in cultivation time, exploiting physiological inertia via light pulses. Finally, the agents autonomously developed an unforeseen strategy of dark-induced chlorophyll accumulation, resulting in a 67.9% energy saving. This framework transforms LLMs into autonomous co-pilots for digital agriculture, improving the cost-to-value ratio and lowering computational and expert-labor constraints.
[AI-110] he Gaussian-Multinoulli Restricted Boltzmann Machine: A Potts Model Extension of the GRBM
链接: https://arxiv.org/abs/2505.11635
作者: Nikhil Kapasi,Mohamed Elfouly,William Whitehead,Luke Theogarajan
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 11 pages, 3 figures (1 figure has 2 subfigures), conference
Abstract:Many real-world tasks, from associative memory to symbolic reasoning, benefit from discrete, structured representations that standard continuous latent models can struggle to express. We introduce the Gaussian-Multinoulli Restricted Boltzmann Machine (GM-RBM), a generative energy-based model that extends the Gaussian-Bernoulli RBM (GB-RBM) by replacing binary hidden units with q-state categorical (Potts) units, yielding a richer latent state space for multivalued concepts. We provide a self-contained derivation of the energy, conditional distributions, and learning rules, and detail practical training choices (contrastive divergence with temperature annealing and intra-slot diversity constraints) that avoid state collapse. To separate architectural effects from sheer latent capacity, we evaluate under both capacity-matched and parameter-matched setups, comparing GM-RBM with GB-RBM configured to have the same number of possible latent assignments. On analogical recall and structured memory benchmarks, GM-RBM achieves competitive, and in several regimes improved, recall at equal capacity with comparable training cost, despite using only Gibbs updates. The discrete q-ary formulation is also amenable to efficient implementation. These results clarify when categorical hidden units provide a simple, scalable alternative to binary latents for discrete inference within tractable RBMs.
[AI-111] Quantum Coordination Advantages in AI State-Tracking Tasks: Semantic Compilation and Latent Memory
链接: https://arxiv.org/abs/2608.11066
作者: Ming Yang
类目: Quantum Physics (quant-ph); Artificial Intelligence (cs.AI); Computational Complexity (cs.CC)
备注: Comments and suggestions are welcome on alphaXiv
Abstract:We prove inference-time quantum coordination advantages for specified AI state-tracking tasks. A solver compresses semantic history into a future-accessible boundary state and later answers a query. We count communication B , persistent instance-dependent memory M , and local work D ; classical recurrence, caches, tools, and recomputation are allowed and charged. The central result is a boundary-preserving semantic-compilation theorem. It maps a finite one-way, streaming, or adaptive causal task into a semantic AI interface while preserving event order and access to past input. Classical boundary-state lower bounds and quantum-memory upper bounds transfer up to explicit compiler overhead, independently of the finite-precision recurrent architecture. Two applications have classical semantics. Matched-entity synopsis QA inherits the hidden-matching separation between O(\log N) qubits and \Omega(\sqrtN) classical boundary bits. Continual requirements auditing inherits a Max- k SAT streaming separation: a recurrent solver uses O(\log^5 n\log(1/\delta)) qubits and polylogarithmic classical workspace to obtain a 0.7172 -approximation, whereas every classical one-pass finite-information solver attaining that ratio requires \Omega(\sqrtn) coordination width. As a quantum-native compiler test, a stabilizer latent-state dialogue uses n qubits, while every exact finite-state classical causal online realization satisfies B+M \ge \frac12n^2+(\frac32-\log_2 3)n+O(1) . The source protocols, streaming algorithms, and stabilizer witness are imported; the new result is their architecture-independent semantic transfer. These are memory and coordination separations, not runtime or empirical advantages for present-day language models. The stabilizer result assumes exact simulation and ideal noiseless quantum memory.
[AI-112] DegradeQuery: Counterfactual Tuple Pretraining for Context-Aware PROTAC Degradation Prediction
链接: https://arxiv.org/abs/2608.10595
作者: Dong Xu,Zhangfan Yang,Jiantao Wu,Zexuan Zhu,Jianqiang Li,Junkai Ji
类目: Biomolecules (q-bio.BM); Artificial Intelligence (cs.AI)
备注: 19 pages, 2 figures, with supplementary material
Abstract:Proteolysis-targeting chimeras (PROTACs) induce protein degradation by recruiting a target protein to an E3 ubiquitin ligase, making degradation a joint outcome of the degrader molecule and its biological context. Although public databases contain thousands of structured molecule-target-E3 records, degradation measurements are available for only a small fraction of them. Existing supervised approaches therefore leave most recorded chemical-biological relationships unused. We introduce DegradeQuery, a context-aware prediction framework that converts these label-missing records into a pretraining signal. Its counterfactual tuple pretraining objective contrasts recorded tuples with alternatives formed by replacing the target, the E3 ligase, or both, enabling the model to learn contextual associations without assigning activity pseudo-labels. The resulting representation is then fine-tuned to predict degradation from the complete molecule-target-E3 context. On the official PROTAC-8K benchmark, DegradeQuery achieves an area under the receiver operating characteristic curve of 0.9065 and an accuracy of 0.8500, outperforming the compared methods. Controlled analyses further show that the improvement is primarily attributable to tuple-level pretraining, can be recovered using only label-missing records, and remains complementary to protein language model representations. These findings demonstrate that incompletely labeled PROTAC databases contain useful relational supervision and provide a practical route for learning context-aware degradation predictors from scarce experimental labels.
[AI-113] Causality Sum Rules in Conventional Scattering Matrices
链接: https://arxiv.org/abs/2608.10427
作者: Ning Han,Rui Zhao,Shuxing Yang,Mingzhu Li,Hongsheng Chen,Yihao Yang
类目: Optics (physics.optics); Artificial Intelligence (cs.AI)
备注:
Abstract:Scattering matrices are the standard experimental and computational description of photonic and electromagnetic devices. Passivity is explicit in the conventional incoming-outgoing matrix, whereas causality sum rules are usually formulated only after transforming the response into auxiliary variables. Here we show that these rules can be written directly in the conventional scattering matrix by removing the time advance introduced by the reference domain. Using the earliest-arrival delay of each channel, we define a domain-delayed matrix that preserves real-frequency passivity while restoring the causal time origin. Under explicit analyticity, transparency, and regularity assumptions, this matrix becomes a Schur function, enabling a Cayley-Herglotz construction. The resulting projected and determinant bounds constrain coherent channel superpositions and aggregate multichannel loss. The framework recovers Rozanov’s absorber limit and spherical-multipole sum rules, while extending causality bounds to measurable quantities including insertion loss, suppressed singular-value channels, and conditional lossless delay-bandwidth trade-offs. Our work directly connects fundamental causality theory with experimentally accessible scattering data. The initial theoretical route is autonomously explored by Qiushi Engine, an AI research system for open-ended scientific discovery, and subsequently verified, refined, and developed by the authors, demonstrating a hybrid AI-human discovery workflow.
[AI-114] A Single Atom in Front of a Mirror is a Universal Reservoir Computer
链接: https://arxiv.org/abs/2608.10382
作者: Peter J. Ehlers,Phi Hung Nguyen,Kanu Sinha,Noelle Daigle,Travis W. Sawyer,Hendra I. Nurdin,Daniel Soh
类目: Quantum Physics (quant-ph); Artificial Intelligence (cs.AI)
备注:
Abstract:Universal approximation in reservoir computing is typically associated with a class of reservoirs. We show that universality can be associated with a single reservoir, considering a minimal setup of a single atom in front of a mirror. In its linear-transducer limit, our reservoir is a universal approximator of fading-memory maps under an operating class of checkable conditions, with a rate constant measured at the operating point. A given reservoir can reach arbitrary accuracy by changing measurement settings. The proof gives an explicit recipe: for a target accuracy, it specifies the required physical resources and resonator modes. Enlarging the number of accessible modes increases the matchable kernel span without reducing capability. Beyond the linear limit, the atom’s saturation replaces high-order polynomial readouts, and the device operates on real-world tasks alongside classical baselines. Our results highlight an example of universality with a minimal quantum setup.
[AI-115] Expert-Guided g-computation with Large Language Models for Estimating Causal Effects on Timings: Applications to Hospital Quality Improvement
链接: https://arxiv.org/abs/2608.10339
作者: Patrick Vossler,Jialin Ouyang,F. Richard Guo,Anran Huang,Ali Shojaie,Lucas Zier,Fan Xia,Jean Feng
类目: Methodology (stat.ME); Artificial Intelligence (cs.AI); Applications (stat.AP)
备注:
Abstract:Hospital quality improvement (QI) programs routinely face multiple candidate interventions to optimize hospital flow, but existing methods struggle to estimate and rank the causal effects of such interventions. This work focuses on one of the most standard hospital metrics, the average length of stay (LOS), and its causal estimand, the average time saved. To characterize this causal effect, qualitative approaches rely on expert judgment to map patient trajectories, making them susceptible to cognitive biases; quantitative approaches rely on data-driven models, which fail when interventions are hypothetical with no historical data or have complex causal mechanisms that require clinical reasoning rather than data alone. We propose expert-guided g-computation, or egg-computation, which combines the complementary strengths of both approaches by connecting the Gantt charts commonly used to map patient trajectories with the causal DAG literature. We introduce a causal model over Gantt charts and establish identification using a variant of g-computation that seeks expert input only for components unidentifiable from data. To make egg-computation practical, we develop an LLM-assisted pipeline that reliably scales up expert reasoning. In simulations, egg-computation outperforms conventional causal inference methods when patients have diverse causal structures and intervention mechanisms. In a study of eleven candidate QI interventions at an urban safety-net hospital, the LLM pipeline generated graphs and time-saving estimates highly concordant with those of human experts. Beyond healthcare, egg-computation is a broadly applicable framework for estimating the average time saved for candidate interventions whose causal mechanisms can be represented using Gantt charts.
[AI-116] Status Association Does Not Reliably Predict Decision Leakage
链接: https://arxiv.org/abs/2608.10089
作者: Abdullah X
类目: Applications (stat.AP); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Machine Learning (cs.LG)
备注:
Abstract:Bias evaluations often move too quickly from evidence that a model encodes a social association to claims that the same association will alter consequential decisions. We test whether that inference is warranted using Chilean surnames as controlled socioeconomic probes. We evaluate eight frozen model-provider cells on 1,032 prompts each, yielding 8,256 verified primary responses. The design separates forced latent association from matched consequential decisions across academic selection, professional hiring, research fellowship selection, and legal-aid intake. Elite-coded surnames received higher forced high-status probability mass than common surnames in seven of eight models and higher mass than rare-frequency controls in all eight. Yet elite-minus-common decision effects were close to zero for most systems. Five models were statistically equivalent within a predeclared (Plus-Minus)0.10 standard-deviation margin, while the remaining three were imprecise or borderline, with no consistent elite advantage. Association strength did not reliably predict decision leakage across models (r = 0.201, p = 0.633) or across frozen surname-pair-by-model cells (r = 0.065, p = 0.565). The central result is a measurement dissociation: latent social association and consequential treatment are empirically distinct constructs. Evaluations should measure the transition from association to action directly.
[AI-117] Do AI weather models miss extremes?
链接: https://arxiv.org/abs/2608.09972
作者: Marvin Vincent Gabler,Roberto Molinaro,Niall Siegenheim,Henry Martin,Mark Frey,Niels Poulsen,Philipp Seitz,Olivier Lam
类目: Atmospheric and Oceanic Physics (physics.ao-ph); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:First-generation AI weather models are often reported to underperform at extremes, mostly in reanalysis-based evaluations of deterministic regression systems. We verify eleven physical and AI forecast systems against European synoptic, solar, and rain-gauge stations over ten months for 10 m wind, 2 m temperature, hourly shortwave accumulation, and hourly precipitation, scoring mean absolute error (MAE) against ECMWF IFS in ERA5 1991-2020 climatological regimes. Among these systems, AI models do not show a uniform relative-skill deficit in the tails. Jua EPT-2.1 Europa leads all-conditions wind (+8.4%), while Jua EPT-2 HRRR leads temperature overall (+12.1%) and in the heat regime (+19.6 +/- 2.2%). EPT-2.1 Europa and DWD ICON Global lead at gale-force wind. Jua EPT-2.1 Helios leads solar overall (+10.2 +/- 1.7%), in overcast conditions (+16.4 +/- 3.4%), and in the clear-sky tail (+24.8 +/- 5.4%). For precipitation, three Jua models gain 14-15% at moderate intensity and 9-11% at P75-P95; EPT-2 Reasoning remains ahead above P95 (+1.7 +/- 0.5%). Failures are model-specific: ECMWF AIFS loses 4.9 +/- 2.0% in the heat tail, while NOAA GFS loses 22.8 +/- 2.0% there. Every model, including numerical weather prediction systems, shows a shared conditional bias toward the centre of the observed distribution, with an inter-model spread several times smaller than the shared signal. Missing relative skill at extremes is therefore not a property of AI weather models as a class, but of particular AI and physical models.
机器学习
[LG-0] Hierarchical Empirical-Bayes Naive Bayes: Minimax Smoothing and Calibration with AODE Extension
链接: https://arxiv.org/abs/2608.11162
作者: Nguyen Thai Anh,Truong Viet Vu,Tran Thien Thanh,Vo Nguyen Quoc Bao,Ngo Hoang Tu
类目: Machine Learning (cs.LG)
*备注: This manuscript has been submitted to the Knowledge-Based Systems
Abstract:The Naive Bayes (NB) classifier remains a standard choice for categorical data, yet its widely used smoothing rules, such as Laplace, Lidstone, Krichevsky-Trofimov, and the m -estimate, all prescribe a fixed smoothing strength that ignores feature cardinality, sample size, and class imbalance, inducing a non-vanishing bias on modern high-cardinality tabular data. We propose hierarchical empirical-Bayes Naive Bayes (HEB-NB), in which each class-feature conditional probability is smoothed by a Dirichlet prior whose concentration is learned data-adaptively via Type-II maximum likelihood, enabling principled information sharing across classes while retaining closed-form inference. We further introduce HEB average one-dependence estimators (HEB-AODE), showing that the adaptive smoothing transfers cleanly to structural relaxations of NB. Theoretically, we establish a non-asymptotic \ell_1 error bound for HEB-NB matching the empirical-distribution minimax rate plus a vanishing data-adaptive bias, together with a matching Laplace-tight lower bound that yields a finite-sample, risk-level strict separation from Laplace. We further derive a plug-in excess Bayes-risk bound via total-variation tensorization and a population top-1 expected calibration error (ECE) corollary. Empirically, across 31 UCI and OpenML benchmarks, HEB-NB attains the best average Friedman rank on probabilistic metrics, with up to 22.1% log-loss reductions on high-cardinality datasets and consistent improvements of HEB-AODE over vanilla AODE. Combining HEB-NB with mutual-information weighting reduces top-1 ECE by 41%-70%, demonstrating substantial gains in probabilistic accuracy and calibration.
[LG-1] DACRI: Decision-Aware Causal Intervention Ranking for Critical Supply Chains CEC
链接: https://arxiv.org/abs/2608.11154
作者: Shiqi Huang,Jiani He,Dingyan Shang,Yihua Xu,Jize Li,Yan Lyu,Lashimi Muraleedharan Nair
类目: Machine Learning (cs.LG)
*备注: Accepted for presentation at the International Conference on Electrical, Computer, Communications and Mechatronics Engineering (ICECCME 2026), 15–17 October 2026, Bali, Indonesia. 5 tables; no figures. Benchmark and code: this https URL
Abstract:Detecting or attributing a supply-chain disruption is not the same as selecting the intervention that maximizes recoverable net value. We present CriticalSCM-Bench v1, a controlled synthetic benchmark with causal ground truth, paired factual/counterfactual rollouts, and an explicit net-value objective. Relative to a full-information train-selected static benchmark, LambdaMART improves median normalized net value by 5.7–16.2%, with paired statistical support on the semiconductor and critical-material archetypes but not on digital infrastructure. On digital infrastructure, a domain-informed constant-buffer policy remains stronger, showing that greater model complexity is not uniformly justified. Across partial and delayed settings, LambdaMART retains 33–75% of full-clamp value. Stress tests further show that intervention fidelity, timing, cost, and held-out disruptions can alter policy ordering. Critical materials show the weakest out-of-distribution retention. Separately, a guarded explanation study over 540 generations preserves every fixed intervention decision after deterministic validation and template fallback, although exact wording remains unstable. Within this controlled setting, the results identify regimes in which adaptive ranking adds value and those in which simpler structural policies remain preferable.
[LG-2] Scheduling Mixed RL Rollouts Beyond Prefix Locality
链接: https://arxiv.org/abs/2608.11152
作者: Zetao Hong,Song Yuan,Yuanhao Ding,Yibo Zhu,Daxin Jiang,Zhibin Wang,Chen Tian
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG)
*备注:
Abstract:Modern reinforcement learning (RL) post-training pipelines for large language models (LLMs) increasingly combine rollout workloads across multiple domains and feedback paradigms. Prefix-aware routing improves inference efficiency through cache reuse and load balancing, but it does not control how heterogeneous rollout sessions compete for KV-cache capacity. When reinforcement learning with verifiable rewards (RLVR), reinforcement learning from human feedback (RLHF), and agentic rollouts share an asynchronous inference service, their distinct sequence structures, interaction patterns, and KV-residency times create substantially different serving demands. Rollout scheduling must account for this heterogeneity without distorting the workload mixture specified by the trainer. We present MISA-T, a routing-layer admission policy for mixed rollout serving. MISA-T combines adaptive session admission, workload-aware KV-capacity allocation, and residency-time-aware KV accounting. In rollout-only ablations on Step3.7 and Qwen3.6-35B-A3B, MISA-T improves rollout throughput over a sweep-tuned cache-aware vLLM Router by 53.3% and 43.6%, respectively, while maintaining high prefix-cache hit rates. In a matched 50-iteration Step3.7 experiment, it increases rollout throughput by 35.6% and reduces mean iteration time by 22.8%, while keeping the consumed workload mixture close to the trainer target and achieving comparable task scores.
[LG-3] A Recommendation System Approach for Interference-Robust Sensor Subset Selection
链接: https://arxiv.org/abs/2608.11143
作者: Kaan Buyukkalayci,Kyle Pak,Merve Karakas,Christina Fragouli
类目: Machine Learning (cs.LG)
*备注:
Abstract:This paper develops a method for sensor-subset selection for tracking. Prior work showed that low-cost acoustic Received Signal Strength Indicator (RSSI) measurements can be used to recommend subsets of sensor nodes whose expensive sensing modalities, such as cameras, can achieve high tracking accuracy. While efficient, RSSI-based approaches are challenged by acoustic interference. We propose a recommendation-system-inspired framework that instead leverages frequency-band acoustic features and a Two-Tower Multi-Layer Perceptron (MLP) architecture to efficiently score candidate sensor subsets. Experimental results on outdoor vehicle-tracking deployments show that the proposed method can improve accuracy by around 20% over the RSSI baseline while maintaining the low computational overhead required for real-time selective sensing.
[LG-4] A Systematic Sample Size Analysis of ML-Based Path Loss Prediction for LPWAN
链接: https://arxiv.org/abs/2608.11083
作者: Robert Bitterling,Christian Nettersheim,Jörn Hees,Michael Rademacher
类目: Networking and Internet Architecture (cs.NI); Machine Learning (cs.LG)
*备注: Accepted at IEEE Conference on Local Computer Networks (LCN)
Abstract:Low Power Wide Area Networks like LoRa are increasingly deployed for smart city applications, requiring accurate path loss prediction for effective network planning. Traditional (empirical) propagation models often exhibit limited accuracy in these scenarios. We investigate machine learning models for LoRa path loss prediction, systematically analyzing how prediction accuracy scales with training set size using real-world measurements from an urban deployment. Our approach employs a Random Forest with LiDAR-derived terrain features and k-Nearest Neighbors with coordinate data, comparing their performance against established empirical models and specialized LPWAN models. Under random pooled splits, both ML models consistently outperform the considered baseline models across the evaluated training-set sizes. At maximum training size, they achieve RMSE values below 6.5 dB compared to 9.7 dB for the best baseline, indicating accurate within-deployment interpolation. A leave-one-gateway-out check qualifies this result: RF shows placement-dependent transfer to held-out gateways, with moderate degradation for several gateways but larger errors for others, whereas coordinate-only k-NN degrades substantially when the gateway location is unseen
[LG-5] Batch Size or Negatives? A Selection Rule for Memory-Constrained Recommender Training
链接: https://arxiv.org/abs/2608.11061
作者: Artyom Sabitov,Daniil Volkov,Alexey Zaytsev
类目: Machine Learning (cs.LG)
*备注:
Abstract:Large-scale neural recommender systems are typically trained with a softmax cross-entropy objective over the full item vocabulary. For a typical large number of possible items K , the final classification layer dominates memory, requiring O(nK) logits and gradients to materialize for a batch of n examples. Sampled softmax reduces this cost by restricting the objective to only k \ll K candidate negative items, resulting in an O(nk) memory. However, for a fixed budget B = n k , it remains unclear whether one should prioritize larger batches or the inclusion of more negative items. We address this question by analyzing sampled-softmax training under a fixed memory constraint. Under standard smoothness and variance assumptions, our theoretical evidence suggests that the fastest convergence arises from an n \sim B, k \sim 1 allocation. So, an actionable rule is to include as many objects as possible given computational constraints. Our theory is supported by controlled synthetic and synthetic and four real sequential recommendation benchmarks, including MovieLens-20M. The suggested configuration achieve faster convergence and better final recommendation quality than imbalanced alternatives within the same memory constraint. These findings provide a theoretical and empirical foundation for configuring memory during the training of recommender systems. Code, reproducibility materials, and all scripts for generating figures are available at this https URL Subjects: Machine Learning (cs.LG) Cite as: arXiv:2608.11061 [cs.LG] (or arXiv:2608.11061v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.11061 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-6] Uncertainty-Aware Deep Learning for Genomics Applications: Insights from an Empirical Study
链接: https://arxiv.org/abs/2608.11054
作者: Sepideh Saran,Mahsa Ghanbari,Uwe Ohler
类目: Machine Learning (cs.LG)
*备注: 21 main pages, 42 total pages, 12 main figures, 13 supplementary figures
Abstract:Deep learning models have emerged as the standard computational tool for a wide range of applications in genomics. Yet, uncertainty quantification (UQ) – and more specifically, the reliability of different uncertainty estimates in this domain – has received little systematic attention. This work presents an empirical analysis of UQ in deep learning models, focusing on genomics applications. In a series of experiments, we contrast Deep Ensembles, Bayesian Neural Networks, and Monte Carlo-dropout methods. We assess their ability to quantify uncertainty in different scenarios, accounting for common dataset characteristics in two genomic application areas and modalities: sequence-to-activity models, and single-cell expression analysis. Our systematic comparison framework provides guidelines for the applicability and reliability of UQ methods in genomics, highlighting their strengths and limitations in different scenarios. We show that Bayesian Neural Networks are better at capturing uncertainty caused by strong class imbalance and out-of-distribution data in genomics, despite their computational disadvantages. Moreover, we show how uncertainty scores can be used to select high-quality predictions in protein-RNA interactions.
[LG-7] Efficient Hypergradient Descent for Inverse Reinforcement Learning
链接: https://arxiv.org/abs/2608.11052
作者: Nikita Sevriukov,Anna Barabanova,Uliana Gagarina,Karina Ivanova,Sofiia Kasaeva,Ilya Levin,Marina Sheshukova
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:Inverse reinforcement learning (IRL) aims to recover a reward function under which the resulting policy reproduces the behavior observed in expert demonstrations. A natural approach is to formulate IRL as a bilevel optimization problem, in which the inner level corresponds to policy optimization under the learned reward and the outer level measures the discrepancy between the induced policy and expert data. However, this formulation is computationally challenging in practice because the outer update requires a hypergradient involving an inverse-Hessian-vector product for the inner objective. We address this challenge by showing that, at the inner optimum, the Hessian of the inner objective is proportional to the Fisher information matrix of the policy, yielding a structured Fisher-based hypergradient closely related to Natural Hypergradient Descent. To address the resulting scalability bottleneck associated with large Fisher matrices, we approximate the required inverse-Fisher-vector product using a streaming spectral sketch, avoiding explicit construction of the Fisher matrix. We evaluate our approach against a first-order stochastic bilevel baseline across discrete- and continuous-control environments. The results demonstrate competitive policy performance and strong reward-ranking quality, while Fisher sketching reduces curvature-storage complexity and can improve computational efficiency relative to an explicit Fisher solver.
[LG-8] SCOUT: Symmetric Consensus Outlier Detection for Failure Localization in LLM Pre-Training
链接: https://arxiv.org/abs/2608.11034
作者: Zhuang Wang
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG)
*备注:
Abstract:In LLM pre-training, synchronization propagates rank-local stalls, slowdowns, and numerical errors into job-wide symptoms, obscuring their origin. Existing diagnosis often relies on in-process monitors that cannot report after the trainer blocks or terminates, or on post-mortem logs that preserve only synchronized symptoms; offline health tests lose the workload and operating conditions that triggered the failure. We present SCOUT, a unified runtime failure-localization framework built on one design principle: identify outliers through strict-majority consensus among equivalent replicas. SCOUT aligns replica progress, timing, and numerical evidence, then uses its Consensus Collective Communication (C3) abstraction to identify ranks whose compact signatures disagree with their peers. An out-of-band CPU observer remains responsive when training hangs, whereas in-situ replay exercises recurring stragglers and silent data corruption (SDC) beside the live job with its model state, kernels, allocations, communication path, and thermal and memory pressure present. Collective fingerprints expose rank-local protocol divergence. Clean replay coverage certifies checkpoint numerical integrity, preventing recovery from selecting state corrupted by SDC. SCOUT integrates with PyTorch, TorchTitan, Megatron-Core, and DeepSpeed without training-loop or framework-source modifications. SCOUT is open source at this https URL.
[LG-9] Derivative Computation in PINNs: Automatic Differentiation Finite Differences and Beyond
链接: https://arxiv.org/abs/2608.11020
作者: Maciej J. Mikulski,Tadeusz Uhl
类目: Machine Learning (cs.LG); Numerical Analysis (math.NA); Computational Physics (physics.comp-ph)
*备注: 22 pages, 5 figures
Abstract:We systematically investigate finite-difference (FD) derivative computation in Physics-Informed Neural Networks (PINNs) as an alternative to automatic differentiation (AD). On three benchmark PDEs we show that, with a properly calibrated step size, FD matches AD in accuracy on every problem while running faster across the full tested batch-size range and using substantially less GPU memory, and that a stochastic variant we propose outperforms AD on a stationary problem. We further show that for neural architectures with inter-sample dependencies (e.g. BatchNorm, self-attention) the standard PyTorch autograd idiom is silently incorrect; the correct per-sample alternative is computationally infeasible at PINN-relevant batch sizes, while FD provides a forward-only approximation that is empirically an order of magnitude closer to the true per-sample derivative.
[LG-10] DEFT: Data-Efficient Frequency-domain Top-k Sampling via Inverse Discrete Fourier Transform for Spatiotemporal Dynamical Systems Modeling
链接: https://arxiv.org/abs/2608.11019
作者: Hengbo Xiao,Jiale Liu,Jiahao Song,Guannan He
类目: Machine Learning (cs.LG)
*备注:
Abstract:Modeling spatiotemporal dynamical systems governed by partial differential equations (PDEs) poses two major challenges: it either requires expensive physics-based simulators that entail iterative numerical solving at high computational cost, or it depends on abundant training data, yet purely data-driven models often generalize poorly to downstream dynamic operating conditions. We propose DEFT, a frequency-domain data sampling method that identifies the dominant Fourier modes of a physical system and systematically varies the corresponding amplitudes and phases to generate physically consistent training data via the inverse discrete Fourier transform. In addition, we derive a generalization bound of this method. We note that it also provides a theoretically principled criterion for selecting K . We evaluate the proposed method through three sets of experiments, each targeting a distinct aspect of its utility. First, we validate the framework on canonical PDEs solving demonstrating that it outperforms traditional methods when the system is dominated by a few prominent frequency components. Second, we employ DEFT as a data-value filter on the diffusion–sorption and Burgers equations of PDEBench, showing that it reduces data requirements by 40% while sacrificing less than 2% in predictive accuracy. Third, to evaluate DEFT for more challenging and practically relevant problems, we validate it in the battery degradation PDE system, achieving consistently high predictive accuracy across various test datasets with R^2 values exceeding 0.99 . Moreover, the learned frequency-domain features transfer to other battery chemistries with only 20% of the fine-tuning data. These results demonstrate that DEFT is an effective data-sampling method for efficient operator learning.
[LG-11] Information Bottleneck under Perfect Privacy
链接: https://arxiv.org/abs/2608.11003
作者: Junle Zhong,Mohamad Assaad,Sreejith Sreekumar
类目: Information Theory (cs.IT); Machine Learning (cs.LG)
*备注:
Abstract:In this work, we study the information bottleneck under perfect privacy, with particular emphasis on the active-rate regime, where the representation-rate constraint is binding and directly limits the achievable utility. The goal is to construct a representation that preserves utility-relevant information while remaining statistically independent of a sensitive variable. This exact independence requirement introduces an additional constraint beyond the classical rate-relevance tradeoff and must be explicitly incorporated into the optimization. To this end, we develop an alternating direction method of multipliers (ADMM)-based method tailored to the resulting problem structure. Under suitable regularity conditions, we establish global convergence of the generated sequence, characterize its convergence rate through the Kurdyka-Lojasiewicz exponent, and extend the analysis to inexact block updates.
[LG-12] GARLIC: Graph Attention-based Relational Learning of Multivariate Time Series in Intensive Care
链接: https://arxiv.org/abs/2608.10969
作者: Ruirui Wang,Yanke Li,Manuel Günther,Diego Paez-Granados
类目: Machine Learning (cs.LG)
*备注: *Equal contribution
Abstract:Healthcare data, such as Intensive Care Unit (ICU) records, comprise heterogeneous multivariate time series sampled at irregular intervals with pervasive missingness. However, clinical applications demand predictive models that are both accurate and interpretable. We present our Graph Attention-based Relational Learning for Intensive Care (GARLIC) model, a novel neural network architecture that imputes missing data through a learnable exponential-decay encoder, captures inter-sensor dependencies via time-lagged summary graphs, and fuses global patterns with cross-dimensional sequential attention. All attention weights and graph edges are learned end-to-end to serve as built-in observation-, signal-, and edge-level explanations. To reconcile auxiliary reconstruction and primary classification objectives, we developed an alternating decoupled optimization scheme that stabilizes training. On three ICU benchmarks (PhysioNet 2012 2019, MIMIC-III), GARLIC sets the new state of the art in outcome prediction, significantly improving AUROC and AUPRC over best-performing baselines at comparable computational cost. Ablation studies confirm the contribution of each module, and feature-removal trials validate the fidelity of importance attribution through a monotonic performance drop (full top 50% random 50% bottom 50%). Real-time case studies demonstrate actionable risk warnings with transparent explanations, marking a significant advance toward accurate, explainable deep learning for irregularly sampled ICU time series data. Moreover, we demonstrated \proposed’s superiority in data imputation and classification on various time-series datasets beyond the ICU domain, showing its generalizability and applicability to broader tasks.
[LG-13] Physics-informed Diffusion Generative Model for Time-Series Data Synthesis in Dynamic Systems
链接: https://arxiv.org/abs/2608.10941
作者: Haiteng Wang,Yunfei Zhu,Tao Wang,Yikang Li,Jiabao Dong,Xiaoge Zhang,Lei Ren
类目: Machine Learning (cs.LG)
*备注: 27 pages; 5 main figures and 4 extended data figures
Abstract:Industrial time-series signals, such as turbine temperature and rotational speed in aero-engines, are essential for monitoring the health and operational status of complex dynamical systems. However, collecting such data is often limited by harsh environments (e.g., high temperature and high pressure) and the high cost of experimental testing. To address this challenge, we introduce PhysDGM, a stepwise physics-embedded diffusion generative model for synthesizing time-series data that are consistent with the underlying physical laws of dynamical systems. PhysDGM embeds physical laws directly into each reverse diffusion step of the generative process, ensuring trajectory-level physical consistency, rather than enforcing constraints only at the final output. A large-scale AI-synthetic dataset (4.4 million samples, 20x scale-up) constructed by PhysDGM demonstrates strong fidelity across 34 datasets spanning turbofan engines, aero-engines, batteries, and chemical processes. After incorporating the synthetic data, the downstream task performance substantially surpassed that using real data alone by 48% for remaining useful life prediction, 15% for health indicator estimation, 22% for state-of-health assessment, and 20% for fault diagnosis. Moreover, it requires 10-20x less training data than existing approaches, substantially reducing the high cost of data collection in dynamical systems. We further demonstrate PhysDGM’s potential in identifying early-stage faults in aero-engines by incorporating AI-synthesized data. In summary, PhysDGM provides a solid foundation for generating physically consistent industrial time-series, paving the way for expanding physics-guided AI into diverse data-scarce environments, including both industrial machinery and complex chemical reaction dynamics.
[LG-14] ReOrder-OPD:Reliability-Aware Prompt Ordering for On-Policy Distillation
链接: https://arxiv.org/abs/2608.10905
作者: Ximo Zhu,Ruiqi Liu,Rong Wang,Ping Wu,Xiang Zheng,Wenzhuo Xu,Xubin Yao,Zhiyuan Yan,Bo Li,Jun Gao,Xiaolei Lv
类目: Machine Learning (cs.LG)
*备注:
Abstract:On-policy distillation (OPD) applies token-level teacher supervision to student-generated trajectories, but this supervision is not always reliable. Existing methods use local confidence or teacher-student agreement to weight, filter, or truncate the sampled trajectory. These signals do not directly determine whether the teacher can continue a student prefix to a correct answer, and trajectory-level interventions can conflate one rollout’s unreliability with low expected training value of its prompt. We define prompt-level teacher continuation reliability R as the teacher’s probability of reaching a correct answer from a student prefix, averaged over prefixes and trajectories induced by the current student. Oracle experiments show that high- R prompts yield larger OPD gains and that descending- R training outperforms random and ascending orders on a fixed prompt pool. Because estimating R requires many teacher continuations, we use the maximum ROUGE-5 F1 between one independent student rollout and verifier-correct same-prompt teacher trajectories. Across ten equal-frequency bins of this actual score, mean R rises monotonically, showing that the proxy separates coarse reliability levels. ReOrder-OPD sorts prompts by the proxy, then draws independent on-policy training trajectories for vanilla OPD. It improves every matched aggregate comparison across Qwen3 and Gemma4 mathematics settings and Qwen3 code settings. Gains in all six FiRe-OPD and ExOPD settings show that prompt ordering complements within-trajectory supervision.
[LG-15] Partially Observable Learning for Multi-Platform Dispatch Optimization
链接: https://arxiv.org/abs/2608.10897
作者: Fengming Yao,Man Luo
类目: Machine Learning (cs.LG)
*备注: 11 pages, 8 figures, 8 tables
Abstract:Instant delivery platforms have become a critical component of urban logistics, increasingly relying on crowdsourced couriers to fulfill highly dynamic orders. In real-world systems, couriers are not exclusive to a single platform and may concurrently serve multiple platforms, while each platform can only observe its own orders and couriers’ interactions due to privacy and operational constraints. This results in a multi-platform dispatch environment with inherent partial observability. However, most existing works on dispatch optimization assume full courier observability and mandatory assignment acceptance, causing substantial performance degradation when deployed in realistic multi-platform settings. In this paper, we propose POLO, a partially observable multi-agent reinforcement learning framework for dispatching optimization in multi-platform instant delivery systems. POLO firstly models each platform-grid pair as an independent agent that learns dispatch policies solely from platform-local observations, aligning the learning process with real-world privacy and operational constraints. To support effective decision-making under incomplete and heterogeneous courier information, POLO introduces a novel attention-based policy representation that selectively aggregates inter-courier information. Moreover, we design a counterfactual reward shaping mechanism to mitigate the non-stationarity induced by joint actions across grids, leading to more stable and scalable learning. We develop a high-fidelity simulator to evaluate dispatch performance under varying numbers of platforms and system scales. Extensive experiments demonstrate that POLO consistently outperforms strong baselines in terms of platform revenue and courier travel efficiency, highlighting its robustness and effectiveness in realistic multi-platform settings.
[LG-16] Benchmarking Time Series Generation Methods for Privacy-Preserving Forecasting
链接: https://arxiv.org/abs/2608.10891
作者: Luis Amorim,Vitor Cerqueira,Moises Santos,Paulo J. Azevedo,Carlos Soares
类目: Machine Learning (cs.LG)
*备注:
Abstract:Time series forecasting in privacy-sensitive domains often requires training models on released data rather than original observations. Synthetic time series generation has been developed primarily for data augmentation, where generated series supplement the original training set. How well these methods perform when fully replacing the original data - and how much privacy risk the released series carry - remains underexplored. We address this gap through a benchmark evaluating synthetic generation methods and noise-based anonymization baselines under a Train on Synthetic, Test on Real (TSTR) protocol. We jointly assess forecasting performance and distance-based empirical privacy risk across seven datasets, characterizing the trade-off between these objectives. We also introduce Grasynda-P, a privacy-motivated extension of the graph-based generator Grasynda, incorporating matrix ensembling and kernel density estimation. Our results show that: (1) no generation method fully substitutes for original training data; (2) noise-based anonymization yields the strongest privacy but the worst forecasting performance; (3) simple transformation-based generators outperform deep generative models for forecasting in this setting; and (4) Grasynda-P lies on the Pareto frontier, achieving competitive forecasting with stronger privacy separation than other generators. This benchmark establishes a reference point for evaluating and developing new privacy-aware synthetic time series generation methods.
[LG-17] Optimistic Rates for Multiclass PAC Learning
链接: https://arxiv.org/abs/2608.10869
作者: Xiaoyu Li,Andi Han,Jiaojiao Jiang,Junbin Gao
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: The main theorems are machine-checked in Lean 4; Section D records what is verified and in which form, and the development is available at this https URL
Abstract:Worst-case multiclass bounds do not become smaller when the best classifier is already nearly correct: what is missing is an optimistic rate, a guarantee whose fluctuation scales with the oracle risk itself. For a class of Natarajan dimension d_N and Daniely-Shalev-Shwartz dimension d_DS , the optimal excess risk is known at the two endpoints ( d_DS/n realizable, \sqrtd_N/n+d_DS/n agnostic [HMZ24, CEH+26, Pab26]) and open in between. We close the gap: at every fixed oracle risk L^\star , the optimal excess risk is \widetilde\Theta(\sqrtL^\star d_N/n+d_DS/n) , uniformly in the alphabet size, attained by a learner that knows neither L^\star nor the confidence level. The upper bound composes the cover-menu-compression architecture of [CEH+26], at the realizable rate of [Pab26], with a new comparator-facing relative compression theorem: a size- k compression rule that empirically dominates a comparator h has population risk at most L(h)+O(\sqrtL(h)\Gamma+\Gamma) with \Gamma=(k\log n+\log(1/\delta))/n , without stability; this transfers the comparison principle of the sharp binary theory [MQZ26] while discarding its Boolean-cube geometry, which does not lift to multiclass labels. The lower bound forces both terms using one class and one distribution at every fixed L^\star , by a pair-Assouad scheme calibrated to L^\star and a fiber argument on the pseudo-cubes underlying the Natarajan-versus-DS separation of [BCD+22]. Both theorems extend to list learning: against the best r -tuple of hypotheses, the same architecture and the same two engines yield an optimistic rate and a lower bound of the same shape, forcing the fluctuation term that [Pab26] expected to be necessary against list comparators, and removing the factor r from the known realizable list lower bound.
[LG-18] Can Bayesian Optimization Efficiently Find a Strong Single Expert in Neural Thickets?
链接: https://arxiv.org/abs/2608.10867
作者: Nigel Bastian Cendra,Abdelhamid Ezzerg,Fernando Julio Cendra,Jeremias Knoblauch,Jakob Zeitler
类目: Machine Learning (cs.LG)
*备注:
Abstract:Gradient-free post-training has emerged as a compelling alternative to gradient-based optimization for large language models (LLMs), but existing approaches remain costly. We ask whether structured search can identify a strong single expert under a modest evaluation budget. Motivated by evidence that useful weight updates lie in low-dimensional subspaces, we apply Bayesian optimization within a random linear embedding of weight space. Our method requires no backpropagation and uses a Gaussian process surrogate to guide candidate evaluations efficiently. Across several reasoning benchmarks with Qwen2.5-Instruct models from 0.5B to 3B parameters, Bayesian optimization using five times less candidate evaluations matches or exceeds RandOpt. These results show that surrogate-guided search can substantially reduce the evaluation cost of gradient-free post-training while producing stronger deployable single experts.
[LG-19] FiGuRO: Intrinsic Dimension Estimation for Multi-Modal Data
链接: https://arxiv.org/abs/2608.10857
作者: Viktoria Schuster,Sana Tonekaboni,Caroline Uhler
类目: Machine Learning (cs.LG)
*备注:
Abstract:Determining the complexity, or Intrinsic Dimension (ID), of data is fundamental to efficient and interpretable representation learning. This is particularly challenging in multi-modal settings when trying to learn disentangled representations for shared and private information. Existing techniques leave a critical gap: they are often static, uni-modal, or in the case of contrastive methods, adapt only to the shared ID implicitly. We introduce Fidelity-Guided Rank Optimization (FiGuRO), a framework for approximating the ID of uni- and multi-modal data under constraints of model capacity and hyperparameters. FiGuRO learns the dimensions of low-rank projections using truncated singular value decomposition and an algorithm that determines when to reduce or increase dimension and in which latent space. Disentanglement of shared and private information arises as an emergent property of this optimization, eliminating the need for complex auxiliary loss functions. We demonstrate that FiGuRO outperforms existing ID estimation techniques and is more robust to hyperparameter changes. Across simulations and real-world data, FiGuRO captures distinct ID scales and varying subspace ratios, and decomposes shared and private information successfully. Furthermore, we show that FiGuRO can be applied to modern uni-modal pretrained models, enabling efficient, post-hoc disentanglement of multi-modal representations.
[LG-20] Diffract: Spectral View of LLM Domain Adaptation ICML2026
链接: https://arxiv.org/abs/2608.10850
作者: Nikita Borodin,Maria Krylova,Artem Zabolotnyi,Dmitry Aspisov,Egor Shikov,Nikita Tyuplyaev,Oleg Travkin,Roman Alferov,Dmitry Vinichenko
类目: Machine Learning (cs.LG)
*备注: Accepted at ICML 2026. Code: this https URL
Abstract:We study continual pre-training (CPT) as a mechanism for adapting general-purpose large language models to specialized domains: mathematics, instruction, code, and natural text. Using singular value decomposition of weight matrices, we find that CPT leaves singular value spectra largely invariant, with adaptation driven mainly by changes in singular vectors. An analysis of attention-head projection matrices reveals strong, domain-dependent head heterogeneity, which we exploit to define a head importance criterion: up to 60% of head updates can be removed without measurable quality loss. Selectively rewinding low-importance heads to their pre-trained state improves benchmark accuracy by up to 4% versus the fully trained baseline. Finally, we identify domain connectivity - linear interpolation between CPT checkpoints yields smooth domain-quality interpolation without notable degradation on either domain - and release Diffract, an open-source toolkit for scalable spectral analysis of billion-parameter models.
[LG-21] MoE Proxy Models for Low-Cost Failure Reproduction and Diagnosis in LLM RL Post-Training
链接: https://arxiv.org/abs/2608.10823
作者: Yikai Wang,Chuansai Zhou,Yuhang Zhou,Weiqiang Wu,Cong Wu,Yue Deng,Ben Feng,Mingming Zhu,Beirong Zhou,Zhibin Wang,Sheng Zhong,Chen Tian,Wangze Zhang
类目: Machine Learning (cs.LG)
*备注:
Abstract:Reinforcement learning (RL) post-training of large language models (LLMs) is computationally intensive and involves complex system pipelines with substantial debugging overhead. In practice, factors such as framework adaptation, numerical precision, and operator implementation can cause failures, including gradient overflow and loss divergence. Reproducing such failures directly on large models requires considerable time and computational resources. This paper systematically analyzes failures encountered during large-scale RL training on the Huawei Ascend platform, summarizes representative failure types, and identifies three model-side factors relevant to fault reproduction. Based on these factors, we propose a proxy-model construction method for low-cost fault investigation and auxiliary diagnosis. It employs structure-preserving, clustering-based expert pruning to select representative experts while retaining the model’s backbone architecture, routing mechanism, and basic task capabilities. Our experimental results show that the proxy models reduce accelerator requirements by 50%-87.5% and achieve up to a 33.3x reduction in per-step NPU-hour cost, while preserving major training dynamics and reproducing fault responses consistent with the original models. Overall, the proxy models can serve as low-cost surrogates for fault reproduction, targeted validation, and auxiliary diagnosis in RL post-training.
[LG-22] Path Integral Value Matching for Linear Quadratic Stochastic Optimal Control
链接: https://arxiv.org/abs/2608.10777
作者: Bangyan Liao,Chenglei Yu,Yuchen Yang,Chuanrui Wang,Zhisheng Song,Peidong Liu,Tailin Wu
类目: Machine Learning (cs.LG); Optimization and Control (math.OC); Machine Learning (stat.ML)
*备注: Project Page: this https URL
Abstract:Linear Quadratic Stochastic Optimal Control (LQ-SOC) establishes a fundamental framework for steering noisy dynamical systems and has recently gained renewed interest in the machine learning community. However, current state-of-the-art policy-based methods suffer from prohibitive computational costs and instability due to their heavy reliance on full-trajectory simulation. To overcome these limitations, we propose a paradigm shift toward a value-based approach by revisiting Path Integral Control (PIC). Although standard PIC suffers from the same high-variance bottleneck as policy-based methods, we discover that by truncating and marginalizing the original path integral formulation, we can derive a temporal recursive form of the value function. Building upon this theoretical foundation, we propose the Path Integral Value Matching (PI-VM) algorithm. Specifically, we employ temporal-difference learning to approximate the recursive value dynamics, and further integrate the Girsanov theorem with experience replay to enable off-policy training. We benchmark PI-VM against SOTA policy-based methods across various SOC benchmarks and sampling tasks. Empirical results demonstrate that PI-VM matches SOTA precision with an order-of-magnitude efficiency gain in low-dimensional settings, while effectively mitigating mode collapse in high-dimensional scenarios. Consequently, PI-VM offers a scalable solution for solving complex SOC problems.
[LG-23] Long-Time Trajectory Approximation via SA-NODEs: Model Predictive and Floquet Strategies
链接: https://arxiv.org/abs/2608.10738
作者: Ziqian Li,Nikolaos M. Matzakos
类目: Machine Learning (cs.LG); Numerical Analysis (math.NA)
*备注:
Abstract:We study the approximation of dynamical systems by semi-autonomous neural ordinary differential equations (SA-NODEs) over long time horizons. For a single network trained on the whole horizon, the available error bound deteriorates double exponentially in the horizon length. We develop two training strategies that avoid this barrier, each built on a reset of the state. The model predictive strategy partitions the horizon adaptively and restarts every window from observed data: when training meets a prescribed tolerance on every window, the composite model meets it uniformly in time, with a parameter budget linear in the horizon for targets with a bounded, uniformly regular reachable tube. The Floquet strategy addresses autonomous targets with a stable limit cycle and uses no data at deployment: a certified contraction of the learned return map confines the error to linear growth in the number of elapsed periods. For the time-periodic architecture we deploy, the scalar certificate degenerates; we prove instead a uniform-in-time orbital guarantee whose hypotheses are measured on the trained model, and an obstruction showing that, for an exactly periodic learned field, small one-period error and a contracting stroboscopic map cannot hold at once. Numerical experiments on four benchmarks confirm the predicted error laws and measure the hypotheses of every guarantee.
[LG-24] SQuaT: Self-Supervised Knowledge Distillation via Student-Aware Quantized Teacher Features AISTATS2026
链接: https://arxiv.org/abs/2608.10709
作者: HyeonJun Lee,Hyeonsik Jo,Jinwoo Chung,Jangho Kim
类目: Machine Learning (cs.LG)
*备注: Accepted at AISTATS 2026
Abstract:Quantization-Aware Training (QAT) enables the deployment of quantized models with minimal accuracy degradation. However, in practical scenarios, training labels are often unavailable due to privacy, copyright, or cost constraints. Knowledge Distillation (KD) is a common approach to address this challenge, but we observe that prior work combining QAT with KD suffers from a fundamental limitation: during distillation, the range mismatch between the teacher and the quantized student model induces an unattainable residual, resulting in an irreducible lower bound on the distillation loss. Motivated by this observation, we propose SQuaT (Student-Aware Quantized Teacher Features), a label-free QAT framework with KD that theoretically eliminates this lower bound by applying the student’s quantization parameters to quantize the teacher’s features during distillation. Through comprehensive experiments across diverse settings, we demonstrate that SQuaT consistently outperforms strong baselines, with particularly pronounced gains in extreme low-bit (e.g., 1- and 2-bit) settings. Furthermore, extensive evaluations across various model design choices show that our approach does not rely on specific architectural assumptions, making it broadly applicable across diverse architectures and quantization settings. The source code is available at this https URL.
[LG-25] IADD-TR: Intervention-Aware Dynamics Decoupling with Targeted Regularization for Model-Based Reinforcement Learning
链接: https://arxiv.org/abs/2608.10634
作者: Zefeng Liang,Jie Qiao,Ruichu Cai,Weilin Chen,Zhifeng Hao
类目: Machine Learning (cs.LG)
*备注:
Abstract:Model-based reinforcement learning (MBRL), which learns environment dynamics to generate synthetic experience, is a promising approach to sample-efficient decision making. Numerous methods have been developed to improve dynamics prediction and policy optimization for MBRL through uncertainty estimation, model regularization, and conservative value learning. However, these methods typically treat the transition model and critic as monolithic predictors, overlooking the policy-induced data bias. Consequently, action can become entangled with environmental evolution, while uneven action coverage may distort the counterfactual value estimates used for policy improvement. To address this, we propose IADD-TR, a unified framework combining Intervention-Aware Dynamics Decoupling (IADD) and Targeted Regularization (TR). IADD factorizes transitions into an action-intervention stage and an action-free natural evolution stage, using a zero-action anchor to resolve the non-uniqueness of this two-stage factorization for robust generalization. Its latent and state-aligned components are identifiable up to an invertible within-block transformation and pointwise, respectively. For policy learning, we derive TR from the efficient influence function of a replay-state policy-gradient functional. TR augments the critic with an action-density-scaled residual correction and optimizes a targeted loss, yielding doubly robust policy-gradient estimation when either the critic or the replay action density is consistently specified. Extensive experiments on five MuJoCo tasks show that IADD-TR achieves competitive returns with improved sample efficiency.
[LG-26] ProbGuard: Calibrated Safety Risk Estimation from LLM Output Distributions
链接: https://arxiv.org/abs/2608.10621
作者: Xinzhe Huang,Biwu Yao,Kedong Xiu,Mengnan Zhao,Di Wang,Puning Zhao,Tianhang Zheng
类目: Machine Learning (cs.LG)
*备注:
Abstract:Recent research on Large Language Model (LLM) safety has widely adopted guardrails to identify unsafe LLM outputs. Existing guardrails typically formulate safety assessment as a deterministic classification task, mapping a discrete token sequence to a discrete safety label. However, this paradigm has two limitations: First, safety assessment is inherently an uncertain problem, particularly during the early generation state. Second, relying solely on discrete token sequences discards the rich probabilistic information embedded in the LLM output distribution. To address these limitations, we propose the first completely probabilistic architecture-agnostic guardrail \textscProbGuard to leverage the LLM early output distributional signals for estimating and calibrating the safety probability, thereby enabling early stopping of unsafe ongoing outputs. Specifically, given an LLM’s generated prefix distribution, we formulate the safety risk as the unsafe probability of its continued generation dynamics and estimate this risk by Monte-Carlo sampling. Through post-training on the distributional signals and calibrated safety risk, \textscProbGuard achieves the best calibration performance across all nine model–dataset combination settings, reducing the average Brier score and ECE by 79.6% and 71.9%, respectively, over the best baseline. \textscProbGuard further limits the attack success rate to at most 1% across six representative jailbreak attacks after observing the LLM early output distributions from only the first ten decoding steps.
[LG-27] Pair-Centric Graph Rewiring for Over-Squashing via Optimal Transport-Guided Communication Alignment
链接: https://arxiv.org/abs/2608.10619
作者: Yan Wang,Chuan-Xian Ren
类目: Machine Learning (cs.LG)
*备注:
Abstract:Message-passing neural networks (MPNNs) often struggle when task-relevant information is distributed across distant regions of a graph, since local propagation must compress remote signals through limited structural interfaces. Graph rewiring provides a structural response to over-squashing. Most existing methods rely on edge-level bottleneck scores or graph-level connectivity surrogates. With a limited rewiring budget, the key question is which pairwise communications most need structural support. This paper proposes PairAlign, a pair-centric graph rewiring framework that makes this question explicit through demand-support shortage. Specifically, PairAlign combines original-graph structural demand with current-graph finite-hop propagation support; their ratio highlights interactions whose communication demand is poorly supported by topology, and our theory shows that this score provides a computable proxy for the corresponding Jacobian-based shortage with a pair-level interpretation of over-squashing. Our theory reveals a two-sided effect of edge insertion: a new edge can create useful walks and simultaneously dilute existing normalized transition mass. Guided by this observation, PairAlign optimizes shortage to favor edge additions that alleviate over-squashing. Beyond selecting useful additions, PairAlign further introduces an Optimal Transport-guided rewiring mechanism to coordinate the finite edge budget for pair-level structural compatibility and shortage-target coverage. It formulates communication alignment between the candidate edge budget and the shortage targets, and the theory shows that this allocation covers shortage targets more broadly and effectively than a greedy-local assignment. Experiments on standard graph benchmarks show PairAlign’s improvement across message-passing backbones, validating pair-level repair as an effective route for alleviating over-squashing.
[LG-28] β-VAEs as Effective Theories: Tolerance-Dependent Dimension
链接: https://arxiv.org/abs/2608.10599
作者: Johannes Hirn
类目: Machine Learning (cs.LG)
*备注:
Abstract:In a \beta -VAE, increasing the regularization strength acts as a spectral cutoff by collapsing low-utility latent coordinates. In the linear Gaussian VAE, the collapse order matches the ranking of reconstruction utilities exactly, because both are set by the PCA spectrum. We ask which parts of this picture survive in fully connected nonlinear VAEs trained on WorldClim. We find that nonlinear interactions shift and broaden collapse onsets, so thresholds no longer coincide exactly with utilities. However, the common ordering is preserved over the resolved ranks, so the spectral cutoff still acts as a utility cutoff and the effective-description logic carries through. The resulting effective-dimension curves reveal a head–tail tradeoff: increasing depth concentrates utility into the first few coordinates but worsens tail fidelity. Subjects: Machine Learning (cs.LG) Cite as: arXiv:2608.10599 [cs.LG] (or arXiv:2608.10599v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.10599 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Johannes Hirn [view email] [v1] Tue, 11 Aug 2026 07:34:18 UTC (147 KB)
[LG-29] BREAD: Baseline-Referenced Explanations for Anomaly Diagnosis
链接: https://arxiv.org/abs/2608.10587
作者: Jiaqi Qiu,Rob Goedhart,Jannis Kurtz,Inez M. Zwetsloot
类目: Machine Learning (cs.LG)
*备注:
Abstract:Artificial Intelligence (AI)-based prospective anomaly detection methods are increasingly deployed in high-dimensional and nonlinear settings. Among these approaches, AI-based statistical process monitoring (SPM) is widely used, providing a structured framework for prospective monitoring. Once an anomaly is detected, a diagnosis method is needed to identify the features driving the flagged observation away from normal behaviour. Traditional SPM diagnosis methods are typically designed for specific detection models and cannot be directly applied to AI-based methods. Model-agnostic explainable AI (XAI) offers a general framework for feature relevance explanation. However, existing methods suffer from scalability limitations or assign relevance to noise features, reducing diagnosis accuracy. We propose a scalable, baseline-referenced diagnosis method that uses both the anomalous observation and normal baseline information. We provide mathematical guarantees that under a mean-shift anomaly setting, the proposed method achieves higher faithfulness in detecting the features causing the anomaly compared to LIME. Simulation studies and a real-world case study validate the effectiveness of the proposed method and show that it generates more faithful and accurate diagnosis results for AI-based prospective anomaly detection methods.
[LG-30] MARCO: Click-Intent Decomposition for Calibrated Ads Conversion Prediction
链接: https://arxiv.org/abs/2608.10562
作者: Shiwen Shen,Xiru Huang,Liang Luo,Jianbo Sun,He Lyu,Zihang Fu,Ivonne Xu,Zhizhuo Li,Zhengyu Zhang,Pei-Ju Sung,Yunmiao Wang,Zixuan Wang,Zhengli Zhao,Qiang Jin,Mike Jermann,Mingda Li,Yang Xiao,Bhavana Challa,Brooke Bian,Yang Li,Ashish Chamoli,Bibek Bhusal,Danning Di,Yuan Jin,Meet Raval,Zhiwen Chen,Boyao Sun,Shuguang Wang,Yunlong He,Yantao Yao,Sagar Chordia,Wenlin Chen,Santanu Kolay,Qin Huang,Ellie Wen
类目: Machine Learning (cs.LG)
*备注:
Abstract:Not all clicks are equal. Industrial ads ranking decouples conversion probability into click-through rate (CTR) and post-click conversion rate (CVR), yet treats every click as the same event. In reality, users provide a free, self-generated signal of intent through their physical UI interactions. Different click types on the same ad exhibit a 4-fold difference in actual conversion rates. By conflating these signals, the standard CVR model under-predicts high-intent clicks and over-predicts low-intent ones, which is a bias masked by near-perfect aggregate calibration. We propose MARCO (Multi-intent Ads Ranking Composition Optimization), a framework that resolves this bias by decomposing each click by intent. Using the logged click type as a free behavioral label, MARCO trains per-intent CVR heads on homogeneous populations, and at serving time composes their per-intent CVR estimates under a predicted distribution over intents. Theoretically, we prove that decomposition never raises population risk, give the exact headroom under squared loss and non-negativity under the deployed loss, and show through a routing-efficiency dial how much of it reaches serving. Because the population-optimal score is unchanged, any gain is a finite-capacity estimation and calibration effect that we validated both offline and online. For deployment at scale, we further cast multi-impression, multi-click attribution as credit assignment with a bias-variance tradeoff analogous to RL return estimation, showing last-impression, first-click attribution is the low-bias, low-variance, deterministic choice under production constraints, and derive three consistency conditions enforced end-to-end at scale. Deployed at binary intent granularity, MARCO corrects per-intent calibration to approximately 100%, lifts conversions per click by +2.80%, and drives +0.98% cumulative improvement in topline metrics.
[LG-31] Benchmarking LLM -Guided Control-Plane Policies for Backend Fault Isolation in HAProxy
链接: https://arxiv.org/abs/2608.10532
作者: Aman Chauhan,Vishnu Pendyala
类目: Networking and Internet Architecture (cs.NI); Machine Learning (cs.LG)
*备注: 43 pages, 6 figures, 15 tables. Submitted to Journal of Network and Computer Applications (Elsevier)
Abstract:Static load balancers cannot mitigate a backend that is degraded rather than down: round-robin and least-connections keep routing traffic to a server returning HTTP 500s until an operator intervenes. We ask whether a Large Language Model can replace the static routing policy itself, reading HAProxy and Prometheus telemetry every 10 seconds and isolating faulty servers through guardrailed calls to the HAProxy Data Plane API. On a reproducible benchmark with a persistent structural fault built into roughly one-third of a heterogeneous fleet, we sweep 15 open-weight models across five families (0.35B to 35B total parameters; dense, mixture-of-experts, and efficient-sparse architectures), reasoning modes, fleet scales of 3 to 9 backends, and two routing algorithms, totaling 240 runs. We find a capability threshold near 3B active parameters. Below it, LLM policies are typically unreliable and sometimes worse than no policy; above it, every model, regardless of architecture, saturates near an 88% reduction in client-perceived 5xx errors over the static baseline. The threshold is approximate: Gemma 4 E2B clears it with 2B active parameters, while the dense 3B Granite 4.0 Micro does not. The availability gain has costs. Draining concentrates load onto surviving servers, inflating tail latency 2.6 to 2.8 times, and enabling reasoning multiplies token spend roughly tenfold, overrunning the control interval and degrading effectiveness. The efficient operating point is a supra-threshold model in its cheapest non-reasoning mode, wrapped inside deterministic guardrails.
[LG-32] Link-adaptive digital twin for robust physical-layer modeling in hybrid-amplified ultra-wideband optical networks
链接: https://arxiv.org/abs/2608.10517
作者: Xiaoxuan Gao,Rentao Gu,Yingchun Wang,Xinyi Liu,Junshi Gao,Yuefeng Ji
类目: Networking and Internet Architecture (cs.NI); Machine Learning (cs.LG); Data Analysis, Statistics and Probability (physics.data-an); Optics (physics.optics)
*备注:
Abstract:Accurate physical-layer modeling is increasingly essential for reliable ultra-wideband operation and capacity optimization, especially under the intensified inter-channel stimulated Raman scattering (ISRS) effect. This paper proposes the link-adaptive digital twin (LA-DT) for hybrid-amplified ultra-wideband links to overcome the generalization and speed limitations of existing methods, achieving accurate modeling and robust generalized signal-to-noise ratio (GSNR) estimation across diverse links. First, to address EDFA heterogeneity, the GSNR modeling task is decomposed into three key power predictions: ASE, NLI, and signal powers before EDFA entry. Second, to enhance cross-scenario generalization, three dedicated DT models are developed using a novel neural architecture with linear modulation layers (LMLs). Third, for rapid adaptation to unseen scenarios with limited data, three domain discriminators guide few-shot fine-tuning of the LMLs. Fourth, the LA-DT explicitly accounts for Raman amplifier (RA) insertion loss, improving practical deployment reliability. Results across 35 scenarios show that LA-DT reduces RMSE for NLI, ASE, and signal power predictions to 0.151, 0.111, and 0.113 dBm with improvements of 56.0%, 58.4%, and 52.7% over the baseline,and achieves an average GSNR estimation RMSE of 0.114 dBm (55.8% improvement). For 12 unseen scenarios, the LA-DT maintains high accuracy through few-shot fine-tuning with only 20 samples per scenario, achieving an average GSNR RMSE of 0.159 dB and demonstrating strong adaptability and robustness.
[LG-33] CARB: A Characterization-Guided Framework for CNN Inference Cost Prediction and Deployment Screening
链接: https://arxiv.org/abs/2608.10506
作者: Linh Nguyen,Zhixin Pan
类目: Hardware Architecture (cs.AR); Machine Learning (cs.LG); Performance (cs.PF)
*备注:
Abstract:Accurate pre-deployment estimation of CNN inference cost–energy, latency, and peak memory–is increasingly critical as models are deployed on resource-constrained GPU platforms. Existing approaches rely on FLOPs, latency measurements, or single-device profiling as energy proxies, overlooking the non-linear interactions between architectural design and hardware load. We present a workload characterization study of 13 419 CNN configurations on two GPU platforms (RTX 5090 and RTX 3080) under GPU telemetry, revealing that energy, latency, and memory exhibit fundamentally distinct scaling behaviors: energy and latency diverge by 3x under high computational demand, and cross-GPU transferability differs by target–energy and latency require platform-specific models while memory transfers well across the two tested platforms. Building on these characterization findings, we develop CARB, a cascade-blended ensemble that jointly predicts all three targets with R2 ~0.99, and a two-stage deployment screening workflow that eliminates over 90% of candidates in seconds, reducing large design spaces to a Pareto-prioritized shortlist validated against real hardware.
[LG-34] A Joint-Distribution Route to Fair Representations with Continuous Sensitive Attributes
链接: https://arxiv.org/abs/2608.10470
作者: Yijin Ni,Xiaoming Huo
类目: Machine Learning (cs.LG); Applications (stat.AP)
*备注:
Abstract:Fair representation learning with a continuous sensitive attribute S requires a representation Z that is statistically independent of S . Existing criteria, including generalized demographic parity, the expectation of integral probability metrics (EIPM), and mutual information, enforce this independence by averaging a per-value discrepancy between the conditional law P_Z \mid S=s and the marginal P_Z over the law of S . This approach requires a nonparametric surrogate for the conditional law at each sensitive value. We propose evaluating independence through a single joint discrepancy d\left(P_Z, S, P_Z \otimes P_S\right) between the joint law and the product of its marginals. We establish a disintegration identity; on decomposable witness classes it equals the conditional-integral functional that EIPM and generalized demographic parity instantiate. By reaching the same target without the conditional law, this discrepancy can be estimated directly from samples via a dependence statistic rather than conditional smoothing. We take the Hilbert-Schmidt independence criterion (HSIC) as an instance of the joint discrepancy d to investigate the statistical efficiency of replacing the conditional formulation. The HSIC estimator is a closed-form O\left(n^2\right) statistic that converges at the O\left(n^-1 / 2\right) rate, in contrast to the nonparametric O\left(n^-2 / 5\right) rate of the conditional-route estimators. We prove this instance is equivalent to the conditional maximum mean discrepancy (MMD) integral up to an explicit spectral tail. The corresponding algorithmic implementation, i.e., FRHSIC, attains fairness-accuracy tradeoffs comparable to conditional-route basel es while reducing per-epoch training time.
[LG-35] Do Time-Series Forecasters Use the Right History: Recoverability Recovery and Functional Use of Temporal Delays
链接: https://arxiv.org/abs/2608.10433
作者: Qipeng Qian,Yuntao Qian
类目: Machine Learning (cs.LG)
*备注:
Abstract:Forecast accuracy does not tell us which past inputs produced a prediction. We separate three questions for time-series models with known delay structure: can the true delay be recovered from the observed data, does the model report it, and does the forecast actually use the same history? We first derive input-conditioned recoverability measures that separate intrinsic ambiguity from model error. We then prove that a delay report can become arbitrarily reliable while forecast risk approaches the oracle even though the predictor still uses the wrong lag. This failure also appears in finite samples on the point-delay task: among forecasts with a correct delay report and normalized excess risk within 10% of the oracle, the reported history is functionally unused under our matched masking test in 55.4% of N-HiTS cases and 92.7% of TCN cases. Finally, we show that routing the prediction through the reported history removes off-report bypass paths; a hard one-hot control achieves exact fixed-report alignment. The main conclusion is simple: a good forecast, even with a correct delay report, does not show that the model used the right history.
[LG-36] deRL: Boosting Agent ic RL Goodput with Readiness-Aware Scheduling
链接: https://arxiv.org/abs/2608.10402
作者: Yanyu Ren,Xizheng Wang,Xiao Liu,Bowen Lv,Hanchen Zhang,Shudan Zhang,Hanyu Lai,Shuai Wang,Li Chen,Dan Li,Jie Tang
类目: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC)
*备注:
Abstract:Reinforcement learning (RL) for large language models is moving toward multi-turn agentic workloads, where rollout tasks repeatedly pause for external environments, resume with growing contexts, and finish at highly variable times. In this setting, RL training goodput, measured by training throughput, matters more than raw GPU occupancy: GPU waiting and repeated prefill recomputation are pure overhead. We present TideRL, a readiness-aware elastic RL system with Continuous Task Batching, Resource-Aware Ref-Actor Pipelining, and Elastic Resource Scaling. CTB preserves useful rollout state, \textrmRA^2\textrmP selects between decoupled streaming and colocated aggregation from the ready backlog and arrival interval, and ERS moves ranks between rollout and training using the same readiness signals. Across text-only and multi-modal agentic workloads, TideRL improves RL training goodput by up to 5.6 \times over synchronous baselines and over 33% over asynchronous baselines, while reaching similar task performance. It also improves KV cache hit rate by 1.58 \times , reduces per-step training time by up to 44.3%, and cuts total waiting time by up to 77.6%.
[LG-37] Do Judges Behave Like Algorithms?
链接: https://arxiv.org/abs/2608.10400
作者: Riya Manchanda,Eric Chen,Chloe Zhu,Cynthia Rudin,Brandon Garrett,Songman Kang
类目: Machine Learning (cs.LG)
*备注: Upcoming Publication, AIES 2026
Abstract:What if judges already behave like algorithms? As artificial intelligence and algorithms are deployed in many settings, including the judicial system, many have debated whether judges should be allowed to rely on them. Instead, we ask whether judges follow predictable, algorithmic-like rules already. If judges already follow consistent, formula-like rules based on discrete and static factors such as criminal history, age, and charge type, then judicial behavior may be improved. However, if judges rely on individualized information that cannot be identified through court data, then standards-based decision-making may be more challenging to understand or improve. This work explores these questions by studying judicial decision-making in misdemeanor bail hearings in Harris County, Texas. Using available court data, we investigate whether magistrate judges follow what resembles an algorithm; whether they consider the same variables in their decision-making; and whether they are consistent with themselves and with each other. To do this, we train machine learning models for each judge, measure variable importance metrics to determine important variables for each judge’s decision-making, and analyze outcomes of similar cases for judges. Our results reveal that these judges generally behave algorithmically: their decisions can be captured by small, interpretable formulas. However, in some cases, judges differ substantially, leading to surprising inconsistency and unequal treatment across similar defendants. Identifying cases where algorithms do not explain judicial decision-making can improve the justice system by focusing attention on decisions where individualized standards, rather than rules, better explains outcomes.
[LG-38] Efficient Weak-Entropy PINN for Solving Hyperbolic Conservation Laws
链接: https://arxiv.org/abs/2608.10389
作者: Qi Gao,Kuang Huang,Xuan Di
类目: Numerical Analysis (math.NA); Machine Learning (cs.LG)
*备注: 27 pages, 7 figures
Abstract:In recent years, neural networks have significantly advanced numerical solutions of partial differential equations (PDEs). However, solving PDEs with discontinuous solutions, such as hyperbolic conservation laws, remains challenging for neural network-based methods such as physics-informed neural networks (PINNs). Existing methods often rely on strong prior assumptions such as knowledge of discontinuity locations, or they introduce artificial smoothing terms that degrade accuracy. However, accurately solving these conservation laws and predicting the formation and propagation of discontinuities in solutions is crucial in many practical applications, including gas dynamics and traffic flow modeling. In this paper, we introduce a novel Weak-Entropy PINN (WEPINN) framework for hyperbolic conservation laws with discontinuous solutions. The method enforces the governing equations in their weak (integral) formulation and incorporates the entropy condition to select the physically admissible solution, while employing the discrete fast Fourier transform (DFFT) for efficient numerical integration. Our method is tested through extensive numerical experiments on a variety of scalar conservation laws and systems of conservation laws in one and two dimensional spaces. These experiments demonstrate that our method can accurately resolve sharp discontinuities while effectively capturing interactions between multiple shock and rarefaction waves.
[LG-39] Dreamer-SAC: Off-Policy Learning in Latent World Models for Sample-Efficient Autonomous Driving
链接: https://arxiv.org/abs/2608.10386
作者: Jiazhuo Li,Linjiang Cao,Qi Liu,Xi Xiong
类目: Machine Learning (cs.LG); Robotics (cs.RO)
*备注: 13 pages, 6 figures
Abstract:Sample-efficient reinforcement learning for autonomous driving is often limited by the trade-off between data efficiency and model bias. While world models reduce the reliance on costly environment interactions, policy optimization over learned dynamics remains sensitive to prediction errors. This paper proposes the Dreamer-SAC framework, which integrates a recurrent state-space world model with an off-policy soft actor-critic algorithm trained directly in latent space. The framework uses a combination of real interactions and short-horizon generated trajectories with n-step target estimation and multi-objective supervision. Evaluated in autonomous driving scenarios with objectives encompassing driving efficiency and safety, the proposed framework consistently outperforms representative reinforcement learning baselines, including DreamerV3, SAC, and PPO, while achieving improved performance with substantially fewer real environment interactions. Experiments reveal an inverted-U relationship between rollout horizon and policy performance, where short-horizon latent rollouts achieve the best trade-off between additional training signals and accumulated model bias. Furthermore, n-step target estimation demonstrates more effectiveness over one-step temporal-difference targets in exploiting predicted experience for value learning.
[LG-40] Generator-Guided Inverse Sampling for Lévy-Driven Generative Models
链接: https://arxiv.org/abs/2608.10384
作者: Tianfu Qi,Jun Wang,Jun Zhang
类目: Machine Learning (cs.LG)
*备注:
Abstract:This paper studies inverse sampling for Lévy-driven generative models from the perspective of Markov generators. Unlike conventional diffusion models, Lévy-driven dynamics involve infinite jump activities, which makes their reverse process nonlocal and difficult to characterize using score information alone. We address this challenge by analyzing the forward and reversed generators. It is derived that the reversed jump component generally becomes a state-dependent Markov jump process governed by a nonlocal density ratio. This observation motivates a structured reverse sampler that decomposes the dynamics into diffusion, small jump, and large jump components. Based on this characterization, we develop a computationally tractable sampler for a class of isotropic linear Lévy SDEs with symmetric \alpha -stable jump components. For the jump component, the neural network is used only to amortize the rate of large jump activities, while jump amplitudes are generated from analytically derived conditional distributions, which improves interpretability and controllability. Efficient implementation techniques are further introduced under this setting to avoid expensive high-dimensional integration and sampling. The sampler is further adapted to approximate observation-guided sampling and applied to OFDM-SISO channel estimation under mixed Gaussian and impulsive noise. Simulations show robust estimation performance with a favorable tradeoff between complexity and performance.
[LG-41] Fisher8: Stabilizing Neural Heteroscedastic Regression via Output-Layer Fisher Geometry
链接: https://arxiv.org/abs/2608.10374
作者: Sumedh Vemuganti,Nickvash Kani
类目: Machine Learning (cs.LG)
*备注:
Abstract:Training neural networks to jointly predict mean and uncertainty estimates from noisy observations can be unstable, prompting a series of independent stabilization efforts. We argue that these interventions highlight a common underlying issue where gradient steps are poorly aligned with the geometry of the loss landscape. To better align updates with local curvature, we derive Fisher8, an output-layer gradient correction that reorients and rescales updates using Fisher geometry rather than Euclidean geometry. Unlike past stabilizers, Fisher8 introduces no data-dependent hyperparameters beyond learning rate and admits an approximate KL trust radius between successive predictive distributions. We show that prior stabilizers converge on overlapping components of this geometric correction. Across multidimensional regression and representation-learning tasks, Fisher8 obtains superior likelihood–error tradeoffs, predicts calibrated uncertainty estimates, and learns rich uncertainty-aware feature spaces.
[LG-42] Invertible Logits Transformation for Accuracy-Preserving Post-Hoc Uncertainty Calibration
链接: https://arxiv.org/abs/2608.10372
作者: Lening Zhao,Qipeng Zhan,Li Shen
类目: Machine Learning (cs.LG)
*备注:
Abstract:Post-hoc calibration aligns a classifier’s predicted confidences with its empirical accuracy without retraining. An ideal calibrator should correct nonlinear miscalibration, scale gracefully to large label spaces, and preserve the original predictions; existing methods typically violate at least one of these properties—temperature scaling lacks expressivity, more flexible parametric alternatives introduce parameters that grow with the number of classes C , and other expressive methods do not preserve the rank ordering of class scores and may alter the predicted class. We propose \textbfInvertible Logits Transformation (InvLT), which applies a learned scalar MLP f:\mathbbR\to\mathbbR element-wise to the pre-softmax logits. Sharing f across all logit dimensions makes the parameter count independent of C . Monotonicity of f —and hence preservation of the argmax prediction—is softly encouraged via a paired inverse network rather than enforced through the numerical integration required by prior monotone calibrators; this avoids their computational overhead while empirically preserving the original classification accuracy in every setting we evaluate. Across standard image classification benchmarks and a range of architectures, InvLT consistently outperforms a broad set of post-hoc baselines on standard calibration metrics.
[LG-43] Accelerated Learning of High Dimensional Functions with a Tensor-Featured Training Network
链接: https://arxiv.org/abs/2608.10351
作者: Karl Pierce,Yuehaw Khoo,Haizhao Yang
类目: Machine Learning (cs.LG)
*备注:
Abstract:In this work we present a method to accelerate the optimization of learning high dimensional functions using deep neural network (DNN). This optimization procedure introduces contextual features into the first layer of a DNN. The parameters of DNN are optimized via standard gradient descent while keeping the input-feature basis fixed. After optimization of the DNN parameters, the feature layer is provided a chance to update and change before DNN optimization resumes. The feature layer has two types of functions: those that can be evaluated quickly in a matrix-free way on the domain (i.e. rank-1 features) and more complex features that must first be decomposed using tensor network (TN) decomposition strategies (tensor features). In particular, we study the effect of adding features which distill pretrained DNN into TNs using a discretize and decompose strategy. To efficiently decompose high-dimensional functions constructed from discretized DNN, we leverage a randomized tensor decomposition strategy. Using randomization, we are able to reduce the storage cost of decomposing high dimensional functions by at least 8 orders of magnitude. Using this approach, we are able to efficiently train models between 5 and 40 dimensions.
[LG-44] Beyond Detection Accuracy: Measuring Explanation Cost Stability and Utility for Resource-Aware IoT Intrusion Detection
链接: https://arxiv.org/abs/2608.10349
作者: Abdurrahman Tolay
类目: Cryptography and Security (cs.CR); Machine Learning (cs.LG); Networking and Internet Architecture (cs.NI)
*备注: 43 pages, 3 figures, 15 tables. Submitted to Internet of Things. Reproducibility materials are publicly archived on Zenodo
Abstract:Machine-learning intrusion-detection studies commonly emphasize predictive accuracy while treating explanation generation as a computationally free post-processing step. This study jointly evaluates predictive effectiveness, explanation cost, local explanation stability, and selective explanation for binary Internet of Things (IoT) intrusion detection. A leakage-safe CICIoT2023 corpus was constructed using exact 39-feature hashes, non-finite-value handling, exact-feature deduplication, conservative label-collision removal, and deterministic hash-level partitioning. Logistic Regression, Decision Tree, Random Forest, and XGBoost were evaluated on natural and balanced test distributions. TreeSHAP cost was measured, stability was assessed under prediction-preserving perturbations, and validation-calibrated policies were used to allocate explanation workload. XGBoost provided the strongest overall predictive profile, while Random Forest produced the lowest false-positive rate. At 5,000 samples, TreeSHAP required 700.759 s for Random Forest and 1.471 s for XGBoost. Random Forest showed the strongest overall base-level explanation stability; XGBoost retained high rank and directional consistency but showed greater top-feature turnover and attribution-magnitude drift. On the balanced test, about 90% false-negative explanation coverage permitted 28-32% compute savings, while about 95% coverage permitted 15-23% savings. Savings were much smaller under the attack-heavy natural prevalence. These results show that operationally useful explainable IoT intrusion detection depends on predictive quality, explanation cost, local stability, workload prevalence, and selective invocation rather than detection accuracy alone.
[LG-45] MERA: Model Evolution and Routing with Skill Adaptation for Agent ic Systems at Scale
链接: https://arxiv.org/abs/2608.10333
作者: Yuhang Yao,Zeyu Wang,Wanyi Chen,Tongyun Yang,Yuhang Han,Jie Xiao,Chengke Bao,Tianyi Zhao,Lynn Ai,Eric Yang,Tianyu Shi
类目: Machine Learning (cs.LG)
*备注: Preliminary version in CAIS RL-Eval
Abstract:LLM agents execute heterogeneous sequences of model calls within a single task: some invocations require careful reasoning, while others are structured steps such as formatting or tool-argument construction. Prior routing methods exploit this asymmetry by assigning easy invocations to a cheaper small model and difficult ones to a large model. Such policies reduce inference cost, but they leave the small model’s capability unchanged, so attainable savings remain bounded by the work the student can already solve. MERA instead improves the small model itself, using a single model invocation as the unit of adaptation. In each cycle, MERA replays failed student invocations to obtain execution-verified teacher demonstrations, distills recurring procedures into an iteratively updated SkillBook, and fine-tunes a student LoRA adapter via supervised learning and optional GRPO. Routing serves as supporting machinery for deployment: the improved student is served behind a cost-calibrated router with verifier-backed fallback, and a candidate SkillBook, adapter, or router is admitted only when joint replay preserves task quality. Empirically, four-cycle adaptation raises Qwen2.5-Coder-1.5B from 28.7% to 49.7% pass on held-out HumanEval+MBPP. Under verifier-backed fallback, the deployed policy retains 88.3% pass at 60.8% of always-Luna cost. On TAU-2, a fine-tuned Qwen3.5-2B improves from 14/35 to 18/35 and matches an unadapted 4B model. These results indicate that verifier-backed multi-cycle adaptation can increase small-model capability, rather than only routing around a fixed student.
[LG-46] opological Feasibility Guarantees for Differentiable Predictive Control
链接: https://arxiv.org/abs/2608.10332
作者: Guangyu Wu,Ján Drgoňa
类目: ystems and Control (eess.SY); Machine Learning (cs.LG)
*备注: 18 pages, 13 figures
Abstract:Differentiable predictive control (DPC), a self-supervised learning approach for approximating explicit model predictive control (MPC) policies, offers significant computational advantages over online optimization-based MPC. However, feasibility guarantees, a core requirement for safe control, are currently provided either probabilistically or via online safety filters. The lack of rigorous feasibility guarantees for offline policy optimization remains an open problem. This paper establishes deterministic feasibility guarantees for DPC using a novel topological analysis of the induced reachable safe set, without requiring online safety filters. By exploiting the inherent model-based nature of DPC, in which differentiable system dynamics are embedded directly into the computational graph, we analyze the properties of the learned control policies and the corresponding system states from topological and geometric perspectives. Inspired by our theoretical analysis, we propose a novel self-supervised offline policy learning strategy that utilizes a proxy loss with Control Barrier Functions (CBFs). Crucially, these properties not only significantly improve policy training but also enable the derivation of strict, deterministic feasibility guarantees from a finite number of training samples. Extensive closed-loop simulations validate our theoretical findings, demonstrating that the empirical constraint violations monotonically decrease to zero as the training sample size increases. Ultimately, this work illustrates that DPC policy optimization yields formal safety certificates that are structurally unattainable with conventional black-box methods, e.g., reinforcement learning (RL) or supervised learning-based approximate MPC, thereby providing a new perspective on feasibility guarantees in learning-based control.
[LG-47] CRHT: A Continuous Regression Hybrid Transformer for Vessel Trajectory Prediction with Online Cluster Sampling
链接: https://arxiv.org/abs/2608.10256
作者: Alexander Schiøtz,Bertram Hage,Christian Rand,Felix Thomsen,Peder Heiselberg
类目: Machine Learning (cs.LG)
*备注: Accepted at IGARSS 2026
Abstract:Accurate vessel trajectory prediction is critical for maritime safety and anomaly detection, yet existing models often struggle with geographic bias and navigational realism. We propose the Continuous Regression Hybrid Transformer (CRHT), a deep learning framework designed to forecast vessel motion using Automatic Identification System (AIS) data. To mitigate spatial data imbalance, we introduce an online K-means cluster sampling strategy that ensures diverse exposure to rare maneuvers during training. Our hybrid architecture integrates 1D convolutional layers for local kinematic feature extraction with a multi-head attention mechanism for global temporal context. CRHT demonstrates superior performance in short-term forecasting, achieving the lowest errors at the 1-hour horizon. The results demonstrate that while discrete models provide high navigational stability over long horizons, CRHT offers an optimal balance of precision and maneuver tracking for real-time maritime surveillance.
[LG-48] STCAD: Scalable Trajectory Clustering and Anomaly Detection on Terabyte-Scale AIS Data
链接: https://arxiv.org/abs/2608.10249
作者: Bertram Hage,Alexander Schiøtz,Felix Thomsen,Christian Rand,Peder Heiselberg
类目: Machine Learning (cs.LG)
*备注: Accepted at IGARSS 2026
Abstract:We present a scalable framework for unsupervised clustering of maritime trajectories derived from terabyte-scale Automatic Identification System (AIS) archives. Variable-length trajectories are encoded with a custom BERT-based model trained via masked token modeling and clustered using CURE hierarchical clustering, producing physically interpretable trajectory groups without requiring a predefined number of clusters. An intrinsic unsupervised anomaly detection method based on reconstruction loss and clustering noise assignment identifies irregular navigation patterns. The framework is demonstrated on a national-scale AIS dataset comprising billions of messages spanning one year, yielding stable trajectory clusters and a clear separation between nominal and anomalous vessel behavior.
[LG-49] A Graph Neural Network–Guided Genetic Algorithm for Physical Internet Supply Chain Optimization under Cost Uncertainty
链接: https://arxiv.org/abs/2608.10245
作者: Faezeh Ardali,Gerald M. Knapp
类目: Neural and Evolutionary Computing (cs.NE); Machine Learning (cs.LG)
*备注: 6 pages
Abstract:Inventory and distribution planning in Physical Internet networks requires coordinating factory-hub assignments, factory supply, lateral transshipment among collaborative hubs, retailer deliveries, and shortages. The problem combines discrete assignment decisions with interdependent continuous flows, while uncertain operating costs make robust planning more difficult. This study formulates deterministic and min-max regret models for a three-echelon network of factories, hubs, and retailers and develops a graph neural network-guided genetic algorithm (GNN-GA) for the assignment decisions. The GNN estimates hub-specific factory-selection probabilities that are used to construct the initial GA population and adapt mutation according to prediction uncertainty. Each previously unseen candidate assignment is evaluated by solving the remaining continuous-flow problem to LP optimality. Simulated annealing, a standard GA, and GNN-GA are compared on 15 instances using matched random seeds and fixed limits on distinct assignment evaluations. Because the evaluation budgets for test Instances 13-15 are smaller than the nominal population size, these experiments primarily assess the quality of learned initialization rather than multi-generation evolutionary search. A separate 400-evaluation experiment on exact test Instance 13 permits three complete offspring generations and a partial fourth pass, with GNN-GA outperforming GA in all 10 matched runs. Three independently generated exact-solvable instances provide a separate test of transfer. Ablation results show that learned initialization provides most of the improvement, while entropy-guided mutation has a smaller, instance-dependent effect. Per-instance solution times include GNN inference and search but exclude model training and one-time model setup.
[LG-50] A matched-integrator evaluation of Hamiltonian neural networks on pendulum and Kepler dynamics
链接: https://arxiv.org/abs/2608.10235
作者: Lenick Kemunto Nyabuto,Yae Ulrich Gaba,Birahim Tewe
类目: Machine Learning (cs.LG)
*备注: 14 pages, 10 figures, 7 tables
Abstract:Hamiltonian Neural Networks (HNNs) parameterize conservative dynamics through a learned scalar Hamiltonian, providing an architectural prior that is absent from generic vector-field neural networks. We evaluate this prior under a controlled protocol in which an HNN and a parameter-matched feedforward baseline are trained on the same RK4-generated trajectories, use the same central-difference derivative targets and optimization settings, and are integrated at inference with the same RK4 scheme. Results are reported over five independent training seeds. On the nonlinear pendulum, the HNN reduces mean energy drift by 42-fold and mean trajectory MSE by 15.8-fold at T = 100, approximately 16 pendulum periods. Its energy drift also remains bounded and exhibits substantially lower seed-to-seed variability than the standard-network baseline. An energy-stratified analysis shows that the difference becomes more pronounced as trajectories explore more nonlinear regions of phase space. As an additional diagnostic, we examine an explicit Störmer–Verlet-style rollout of the learned HNN. Because the learned Hamiltonian is not constrained to the separable form H(q,p) = T§ + V(q), the standard symplecticity guarantee of velocity Verlet does not directly apply. We further apply the same matched-integrator protocol to the three-dimensional Kepler two-body problem. The HNN again exhibits lower trajectory, energy, and angular-momentum drift than the parameter-matched baseline. These experiments provide a controlled study of how Hamiltonian parameterization affects long-horizon prediction and physical consistency across two conservative dynamical systems. Comments: 14 pages, 10 figures, 7 tables Subjects: Machine Learning (cs.LG) Cite as: arXiv:2608.10235 [cs.LG] (or arXiv:2608.10235v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.10235 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Lenick Nyabuto Kemunto [view email] [v1] Mon, 10 Aug 2026 21:12:52 UTC (444 KB)
[LG-51] he Kuramoto Neural Operator: Learning to Solve PDEs via Coupled Oscillator Dynamics
链接: https://arxiv.org/abs/2608.10234
作者: Petr Badolia,Leonid Obukhov,Dmitry Bylinkin,Aleksandr Beznosikov
类目: Computational Engineering, Finance, and Science (cs.CE); Machine Learning (cs.LG)
*备注: 29 pages, 9 figures, 17 tables. Introduces the Kuramoto Neural Operator (KNO), which models PDE solution operators via the evolution of a latent field of coupled spherical oscillators. Includes ablations and appendices with additional robustness and resolution-transfer experiments
Abstract:Operator learning is a rapidly advancing area of computational science. It is particularly well suited to problems where a partial differential equation (PDE) must be solved repeatedly under varying physical configurations. Most existing architectures represent the solution operator in a fixed basis. While this assumption is well aligned with global structures, it is less suitable for phenomena governed by local interactions in physical space. We explore an alternative perspective motivated by the observation that the continuum limit of coupled oscillator systems can describe a broad class of PDEs. Building on this idea, we introduce the Kuramoto Neural Operator (KNO), which represents the solution through the evolution of a latent field of interacting oscillators. Across a diverse collection of PDE benchmarks, KNO achieves strong predictive performance, with improvements over competing approaches. Our experimental evaluation also includes an extensive ablation study that quantifies the contribution of each architectural component incorporated into KNO. Furthermore, we show that the model’s prediction error is closely linked to the collective dynamics of the latent oscillators. It varies systematically with their degree of synchronization, providing insights into the underlying mechanisms.
[LG-52] Boundary-Seeking Policy Gradient for Safe Reinforcement Learning
链接: https://arxiv.org/abs/2608.10204
作者: Chenhua Fan,Jiahui Zhu,Yuhang Zhang,Honghao Wei
类目: Machine Learning (cs.LG)
*备注: CDC 2026
Abstract:Safe reinforcement learning maximizes reward subject to safety constraints. For Constrained Markov Decision Processes, the linear-programming view over occupancy measures implies that whenever the constraint is active at optimality, the optimal policy lies exactly on the constraint boundary, yet standard gradient-based methods do not exploit this structure and often settle in the feasible interior. We introduce Boundary-Seeking Policy Gradient (BSPG), a first-order method whose update combines a tangential component that improves reward while preserving cost to first order with a signed, residual-driven normal component that regulates the policy toward the active boundary from either side; the combined direction admits an algebraic Lagrangian form with an induced coefficient and no learned dual variable. Under exact gradients and stated regularity conditions, the constraint residual converges to zero from either side with a finite-horizon O(1/\sqrtT) bound, the tangential component is a reward-ascent direction on the boundary, and any convergent parameter sequence is stationary on the active constraint set, satisfying the KKT conditions when the limit is also a local maximizer over the feasible set. This complements existing analyses, which certify feasibility but do not characterize the constraint value at convergence. On a standard Safety-Gymnasium navigation task, BSPG attains higher reward while tracking the boundary more tightly than the compared baselines.
[LG-53] Intrinsic Structure: Spectral Identifiability for Mechanistic Interpretability
链接: https://arxiv.org/abs/2608.10172
作者: Ashim Dhor,Pin-Yu Chen
类目: Machine Learning (cs.LG)
*备注:
Abstract:Mechanistic interpretability explains models by identifying circuits inside them, but has no way to tell whether a circuit is a property of the model or an artifact of the method that found it. Sparse autoencoders illustrate the problem: different seeds and widths recover materially different features from the same activations, and no theory says whether that variability is incidental or structural. We put dictionary learning for interpretability on an identifiability footing. Treating the forward pass as a controlled dynamical system with depth as time and lifting it with the Koopman operator yields a finite linear realisation whose \emphspectrum is a coordinate-free property of the model. We prove the spectrum is recoverable from M calibration samples at rate M^-1/2 up to permutation - to our knowledge the first identifiability theorem for a mechanistic-interpretability primitive, with a matching minimax lower bound, a median-of-means variant for heavy-tailed activations, and a dissociation theorem: whenever the realisation is non-normal, the directions carrying activation variance and the directions carrying information across depth cannot coincide. The identifiable object and the legible object are not the same object. On GPT-2 small, Gemma-2-2B and Qwen3-8B-Base the spectrum converges everywhere and attains the predicted exponent on Qwen3-8B-Base ( 0.506 \pm 0.031 ); shortfalls collapse onto one curve against each cell’s sample threshold. Koopman modes beat random directions but lose to principal components on indirect-object identification, with the gap decaying 4.1\times in depth-distance, as the theorem predicts. The Koopman spectrum is an identifiable, model-intrinsic fingerprint with a stated error bar, not a legible decomposition.
[LG-54] REATS: LLM Reasoning -based Ensemble Learning for Adaptive Time Series Forecasting
链接: https://arxiv.org/abs/2608.10149
作者: Xu Zhang,Chang Xu,Hui Sun,Nan Ma,Zijian Zhang,Peng Wang,Wei Wang,Li Zhao
类目: Machine Learning (cs.LG)
*备注:
Abstract:Due to the diversity of real-world time series, no single forecasting model consistently dominates across all samples. Ensemble learning addresses this by combining complementary model strengths, yet existing methods rely on fixed rules or black-box models based solely on numerical inputs, failing to leverage LLM reasoning for interpretable weighting decisions. We propose REATS, which leverages LLM reasoning capabilities as an intelligent ensemble router that jointly processes textual temporal pattern descriptions and numerical features to produce interpretable, sample-adaptive ensemble weights through chain-of-thought reasoning. To enable effective LLM-based ensembling, we study its key design choices and propose: (i) a structured input pipeline that transforms raw time series into hybrid textual–numerical representations with fixed token cost, enabling rule-based chain-of-thought construction without API dependency, augmented with retrieved similar-sample priors; (ii) a diverse multi-row weight supervision scheme coupled with a token-efficient percentage-table format that reduces numerical complexity and mitigates LLM hallucinations; and (iii) a two-stage fine-tuning framework combining SFT with GRPO, where a reciprocal reward mapping transforms the continuous unbounded MSE gap into bounded signals with amplified near-oracle sensitivity, addressing the uniform sensitivity and outlier-dominated advantage compression inherent in naive reward designs for regression-based GRPO. Experiments on eight benchmarks demonstrate that REATS outperforms competitive ensemble baselines while providing natural language explanations and demonstrating strong transfer learning and out-of-domain generalization to unseen candidate models.
[LG-55] he Evaluation Protocol Determines the Result: An Independent Reproduction of LeWorldModel on TwoRoom
链接: https://arxiv.org/abs/2608.10145
作者: Joyjeet Singh
类目: Machine Learning (cs.LG)
*备注: Independent reproduction of arXiv:2603.19312 - this https URL
Abstract:LeWorldModel trains a latent world model with a prediction loss and a single anti-collapse regulariser, and reports approximately 87% of goals reached on TwoRoom, its simplest diagnostic environment. We reproduce that result by independent reimplementation on roughly 25 of rented compute, with all evaluation on one laptop CPU. We reach 94.0% at the repository’s evaluation goal offset, against 84.0% for the authors’ own released checkpoint measured under our protocol on identical episodes, and we reproduce the reported representation result directly (position probe Pearson r = 0.9988 against a reported 0.996). Reaching that point required four conventions that determine the outcome and appear in no released configuration file: dense action gathering across a frameskip block, a programmatically-set action-encoder width, ImageNet pixel normalisation, and action z-scoring. A reproducer following the released configurations alone obtains a model whose predictor cannot converge. The evaluation protocol is itself contested by the released material. The paper’s appendix and the repository’s configuration specify different goal offsets and step budgets; on the authors’ own weights these yield 14.0% and 84.0%, and only the configuration’s values reproduce the reported figure. On fifty identical episodes, changing nothing but how the goal is constructed moves that checkpoint from 84.0% to 8.0%. Two findings generalise. One-step prediction accuracy does not predict long-horizon planning success: across three checkpoints spanning a sevenfold range in prediction error, including the authors’ own, it orders short-horizon success monotonically and fails to order long-horizon success at all. And a batch normalisation layer inflated our reported validation loss by up to a factor of 300, concealing a training loss that was flat throughout. Comments: Independent reproduction of arXiv:2603.19312 - this https URL Subjects: Machine Learning (cs.LG) Cite as: arXiv:2608.10145 [cs.LG] (or arXiv:2608.10145v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.10145 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Joyjeet Singh [view email] [v1] Mon, 10 Aug 2026 19:00:51 UTC (396 KB)
[LG-56] SeFoRA: Sketch-Aggregated Federated Low-Rank Adaptation with Heterogeneous Client Ranks
链接: https://arxiv.org/abs/2608.10144
作者: Yue Xia,Tayyebeh Jahani-Nezhad,Mayank Bakshi,Rawad Bitar
类目: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC)
*备注:
Abstract:We consider federated parameter efficient fine-tuning of large neural networks with low-rank adaptation (LoRA,~Hu et al.\ 2022). Combining LoRA with federated PEFT introduces challenges absent from either setting alone: clients may use different LoRA ranks, making their factor matrices dimension-incompatible, and factor-wise averaging suffers from a bilinear mismatch. We propose SeFoRA, a sketch-aggregated federated LoRA algorithm in which each client transmits a linear sketch of its local updates, enabling direct aggregation at the federator. As a result, SeFoRA alleviates the bilinear mismatch, and allows for aggregation in a small subspace of the full model. We introduce a rank-homogeneous version called SeFoRA-Ho which allows for direct adapter aggregation in this setting. We prove convergence to a neighborhood of the first-order stationary point at rate \cO(1/T) for the rank-homogeneous setting. Numerical experiments on fine-tuning RoBERTa-Large on GLUE datasets show how our algorithms outperform the state-of-the-art.
[LG-57] ChronoSSM: Training for Temporally Aware Representations in Autoregressive State Space Models
链接: https://arxiv.org/abs/2608.10120
作者: Adrien Schoen,Nachiketa Ratnakar Patil,Arjun Bhagoji,Francesco Bronzino
类目: Machine Learning (cs.LG); Networking and Internet Architecture (cs.NI)
*备注:
Abstract:Modern sequence models, from Transformers to State Space Models, have enabled powerful generative modeling across diverse domains, yet they are typically trained to predict what happens while treating when it happens as a secondary concern. In data-mining settings where events are associated with explicit timing information, this separation can limit temporal reasoning, anomaly detection, and faithful reconstruction of event chronology. A common strategy is to treat timing as an auxiliary signal, training a separate timing model using representations learned solely for event prediction. However, this two-stage approach implicitly assumes that representations optimized for event prediction already contain sufficient temporal structure. We introduce ChronoSSM, an autoregressive State Space Model (SSM) that jointly models events and timestamps with a shared backbone trained using combined token and temporal generation objectives. We compare the joint regime, where temporal supervision updates the backbone, with the two-stage regime, where timing is learned only using the frozen event representations. Across four domains spanning dense and partial timestamp supervision, joint training consistently makes inter-arrival information more recoverable from frozen representations without any systematic degradation in content-generation quality overall. Our results show that temporal supervision can produce more temporally informative representations without materially degrading autoregressive event modeling. Subjects: Machine Learning (cs.LG); Networking and Internet Architecture (cs.NI) Cite as: arXiv:2608.10120 [cs.LG] (or arXiv:2608.10120v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.10120 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-58] Observational Policy Ranking for SMB Financial Guidance from Multi-Action Accounting Logs
链接: https://arxiv.org/abs/2608.10050
作者: Shrutendra Harsola,Vignesh Subrahmaniam,Vikas Raturi,Kamalika Das,Xiang Gao,Kratika Gupta,Ruocheng Guo,Padmaja Jonnalagedda,Ananya Pramod,Sricharan Kumar
类目: Machine Learning (cs.LG)
*备注:
Abstract:Small and medium-sized businesses need timely financial guidance, yet historical accounting logs record self-selected and often co-occurring business changes rather than randomized recommendations. We formulate this setting as observational policy ranking: from pre-decision financial information, a policy selects one of 34 ledger-derived business-change categories for a target financial KPI. Using 85,078 company-month observations from 7,505 firms, we introduce Covariate-Adjusted Residual Policy Learning (CAR-PL), an action-wise R-learner that operates directly on multi-hot logs and regularizes selection by observational support. We compare CAR-PL with an uplift T-Learner, a conservative contextual value model, a zero-shot LLM, and non-personalized references on company-disjoint held-out firms under a shared model-assisted scoring rule. CAR-PL has the highest Gross Profit point estimate (0.084), the T-Learner has the highest Revenue point estimate (0.085), and the contextual value model has the highest Quick Ratio point estimate (0.062). CAR-PL and the T-Learner are not statistically separated on either growth KPI in matched company-clustered comparisons, while CAR-PL selects 33-34 categories and produces less concentrated selections across the catalog. Outcome-model-only scoring retains the same KPI-level point-estimate leader or top pair, and category rankings remain similar when the all-zero treatment reference is replaced by the most common training co-action pattern. These findings support objective-specific ranking of SMB financial guidance from multi-action accounting logs.
[LG-59] Detecting Soft Skills in ML Engineering Roles CVs
链接: https://arxiv.org/abs/2608.10046
作者: Aidin Azamnouri,Nouran Ayad,Justus Bogner,Stefan Wagner
类目: Machine Learning (cs.LG); Computers and Society (cs.CY)
*备注:
Abstract:Soft skills shape collaboration among ML engineers, data scientists, and software engineers building ML-enabled systems, yet what we know about them comes almost entirely from the demand side. Job advertisements, surveys, and hiring manager interviews capture what employers ask for. How candidates themselves articulate these competencies has not been studied, and existing CV-mining work is both keyword-based, so it cannot see skills conveyed through narrative, and descriptive, reporting frequency rankings without testing whether group differences exceed sampling variation. We close both gaps. Using a balanced corpus of 300 curated CVs spanning the three roles, we extract explicitly listed and implicitly narrated soft skills with an LLM-based pipeline validated against a human-annotated ground truth, a distinction that existing extractors were not designed to make. We then convert the demand-side literature’s claims into 13 falsifiable hypotheses about role signatures, seniority progression, and disclosure style, and test them with effect sizes under family-wise error control, so that candidate-side data can corroborate or contradict the demand-side account rather than merely illustrate it. Eleven hypotheses are supported, one partially, and one refuted. Candidates disclose soft skills through narrative rather than keyword lists by roughly three to one, and most so for the competencies employers value most: leadership, coordination, and mentoring (88-96% narrative). Seniority nearly triples the odds of articulating leadership. That competency, assumed universal in prior work, is articulated by software engineers at half the rate of their peers. Technical candidates do articulate soft skills, but a keyword-based screening systematically misses them.
[LG-60] FlowScout: From Execution Feedback to Reliable Tool-Using Agent Workflows
链接: https://arxiv.org/abs/2608.10039
作者: Shuo Hao,You Lu,Bihuan Chen,Xin Peng
类目: Machine Learning (cs.LG)
*备注:
Abstract:Agentic workflows have become an important abstraction for building reliable LLM-based automation systems by organizing large language models (LLMs), tools, and control logic into explicit execution structures. However, constructing high-quality agentic workflows remains largely manual and requires substantial domain expertise. Recent studies have explored automatic agentic workflow generation from historical task-solving records, but they mainly produce LLM-centric workflows, where real tool executions are abstracted and simulated by LLM nodes, limiting the usability and stability of generated workflows. To address these limitations, we propose FlowScout, an execution-guided framework for generating tool-integrated agentic workflows from historical task-solving records. Specifically, FlowScout represents an agentic workflow as a directed graph composed of LLM nodes, tool-calling nodes, and dependency edges. It first mines a common tool coordination skeleton from historical records to construct an initial workflow, and then refines the workflow topology through Monte Carlo tree search guided by execution feedback. We evaluate FlowScout on four representative task domains and compare it with three baselines, i.e., PM4Py, ReAct and AFlow. Experimental results show that agentic workflows generated by FlowScout improve tool invocation correctness by at least 92.69% and execution quality by at least 17.66% over the baselines, while achieving lower performance variation across repeated runs.
[LG-61] CurveFP: Rational-Radix Logarithmic Datatypes with Closed Products for Language Models
链接: https://arxiv.org/abs/2608.10010
作者: Ye Qiao
类目: Machine Learning (cs.LG)
*备注:
Abstract:Low-precision datatypes reduce language-model cost, but most formats optimize scalar fidelity while leaving the arithmetic induced by their products unchanged. We introduce CurveFP, a closed-product codebook family that distributes quantized magnitudes across interleaved logarithmic curves under compact block scales. A rational radix tunes dynamic range against local resolution, while uniform curve indices make every nonzero product algebraically closed. Product formation becomes an exact sign XOR and integer-index update, and a derived finite phase count determines the accumulation schedule. We instantiate this algebra as CurveFP eight E4C3/E5C2 for training and CurveFP seven E3C3 for compact deployment. In evaluation, CurveFP seven beats tensor-wise FP8 perplexity on four 7B–9B models with one fewer element bit and stays within 1.32% of native quality. CurveFP eight lowers operand NMSE in all 36 paired forward and backward GEMM comparisons. Across three matched 128.3M-parameter triplets, every mode completes 3B-token pretraining per seed; CurveFP eight reaches mean BF16-inference perplexity 22.5366 versus 22.5407 for FP8 and incurs a lower format-induced penalty in all three seeds. A 36-cell downstream matrix finds lower WikiText-103 perplexity for the CurveFP eight-trained checkpoints in all 12 seed-format comparisons, with mixed PG-19 and task deltas. Together, these results establish CurveFP as an arithmetic co-design that combines FP8-class numerical behavior, seven-bit inference, and a substantially simpler product path.
[LG-62] HyperShape: Hyperelasticity Across Diverse Shapes
链接: https://arxiv.org/abs/2608.09938
作者: Leo Widmer,Sidaty El Hadramy,Stéphane Cotin,Philippe Claude Cattin
类目: Computational Engineering, Finance, and Science (cs.CE); Machine Learning (cs.LG)
*备注: 17 pages, 10 figures
Abstract:Hyperelastic deformations are highly sensitive to domain geometry and boundary conditions, making generalization across both a critical capability for neural operators applied to these problems. However, existing benchmarks for neural operators on hyperelasticity rely on simple or few geometries, which makes it difficult to assess this capability rigorously. To address this gap, we introduce HyperShape, an extensible framework designed to generate synthetic shapes and their corresponding hyperelastic simulation data, producing a suite of 2D and 3D datasets with adjustable complexity and controllable shape variations. This design enables systematic assessment of generalization across in-distribution, out-of-distribution, and synthetic-to-real transfer settings. Using this framework, we evaluated the performance of several state-of-the-art neural operators over diverse shape distributions. Our findings reveal that neural operators perform well on simple shapes but struggle as shape complexity, geometric diversity, and boundary condition variability increase, requiring large amounts of training data in such regimes. Performance degrades consistently and predictably with geometric complexity highlighting the need for further model development. As an open and extensible benchmark, HyperShape is designed to grow alongside the field: new geometries, material models, loading conditions, and evaluation settings can be easily incorporated to validate hyperelastic surrogate models.
[LG-63] Wrong Design Intent Is Worse Than Never Conditioning: A Derangement-Control Diagnosis of Header Conditioning in CAD Program Completion
链接: https://arxiv.org/abs/2607.23191
作者: Yang Xiao
类目: Machine Learning (cs.LG); Graphics (cs.GR)
*备注: 47 pages, 4 figures. v4: referee-response revision, no evaluation re-run. Causal claim narrowed to correct-to-wrong content sensitivity; the control’s offset vs the unconditioned baseline is arm-invariant and now reported. Cluster-level analysis over 298 distinct inputs is now the headline, with bootstrap intervals. Wrong-target uptake added, control as placebo
Abstract:Fine-tuned code LLMs are routinely conditioned on a design-intent specification, but the correctness axis of such a signal – a wrong intent rather than an absent one – has not been tested, and the benefit of conditioning is usually scored with the same detector that defines the signal. We study CADCON, a five-feature design-intent header prepended to CadQuery-style programs during LoRA fine-tuning of Qwen2.5-Coder-1.5B, scoring adherence with executable geometric assertions that share no code with the header-defining extractor. On a pre-registered sample of 400 deduplicated held-out programs stratified over eleven intent profiles, at 40% prefix and three seeds, a semantically wrong header degrades adherence below the never-header-trained baseline on 3/3 seeds under both tokenizations at the program level, and on 3/3 token and 2/3 text seeds at the 298 distinct model inputs they present. Wrong-header executability is not depressed relative to that baseline. A derangement control, retrained so every program receives another program’s header – holding the header marginal fixed while destroying its correlation with the program – saw the same programs, indices and wrong headers. Its correct-to-wrong change is -0.006/+0.016/-0.003 against 0.124/0.241/0.230 for the standard model, and the interaction is significant on 3/3 seeds (p = 5.9e-7), so the model’s sensitivity to whether the header is right or wrong requires the learned mapping. The control sits below the baseline by the same margin under a correct as under a wrong header, so we claim that sensitivity and not the below-baseline level. On features the true intent lacks, the standard model realizes a feature far more often when the wrong header names it; the control does not. Ground truth itself scores only 0.567 here, the scale on which arm levels should be read. Wrong design intent is not inert: it actively misdirects generation.
[LG-64] A Quantum Roadmap for Softmax Attention: Exact Born-Rule Analogs for Softmax Attention on the Probability Simplex
链接: https://arxiv.org/abs/2608.11173
作者: Eric A. F. Reinhardt,Adam J. Hauser
类目: Quantum Physics (quant-ph); Machine Learning (cs.LG)
*备注: 33 pages, 10 figures
Abstract:The attention mechanism forms the foundation of many modern AI models such as the Transformer. In one subclass of problems where attention is used, inputs and outputs are bound to the probability simplex so that all outputs sum to one. In this setting, softmax attention admits an exact, component-by-component quantum realization. Attention scores are Hadamard-test statistics on block-encoded projections of amplitude-encoded inputs. The exponential softmax is the interior of a cosine-squared family generated by Born-rule measurement under an exact bijection, whose boundary expresses sparse attention with exact zeros at finite parameter values. The softmax temperature is a repetition count where post-selected measurement rounds realize discretized inverse temperature exactly. Value aggregation is a deterministic column-loading channel that dilates the column-stochastic value matrix. The gated residual is the preparation angle of a single ancilla, with the additive identity at a mixing angle of \pi/2. Every learnable parameter is a rotation-gate angle. The composed layer is exact in the infinite-shot limit with one measure-and-reload step per attention score; a fully-coherent variant is \epsilon-approximate via quantum singular value transformation in the infinite depth limit. The algebraic core is machine-checked in Lean 4.
[LG-65] Conditional Independence Tests for Constraint-Based Causal Discovery: A Survey
链接: https://arxiv.org/abs/2608.11156
作者: Pavel Averin,Theodoros Moysiadis,Ioannis Katakis
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注: 33 pages. Published in Transactions on Machine Learning Research (07/2026). this https URL
Abstract:Conditional Independence (CI) tests are the statistical engine of constraint-based causal discovery: in algorithms such as PC (Peter-Clark) and FCI (Fast Causal Inference), skeleton pruning and key orientations follow directly from CI decisions. This survey reviews CI testing with emphasis on assumptions, robustness, and scalability in high-dimensional and mixed-type settings common in biomedical domains. The survey organizes widely used CI methods into six families: partial-correlation, contingency-table, regression, nearest-neighbor, kernel, and machine-learning-based. Special emphasis is provided on the robustness layers that address the limitations of these families. For each family, the survey examines when CI decisions reflect the data-generating distribution and when they fail. By this, we link test-level properties, including power decay with conditioning set size and asymmetric type I/II error consequences, to graph-level errors in skeleton recovery and v-structure orientation. The survey also compares adoption across major R and Python libraries and summarizes open challenges, including mixed-type CI testing without discretization, small-sample error control, and strategies for improving scalability of CI-testing.
[LG-66] Gromov-Wasserstein Quantization and Clustering: Structure Rates and Algorithms
链接: https://arxiv.org/abs/2608.11016
作者: Florian Beier,Stephan Eckstein
类目: Optimization and Control (math.OC); Computational Geometry (cs.CG); Machine Learning (cs.LG); Probability (math.PR)
*备注:
Abstract:Clustering is a fundamental class of data analysis techniques with the most important representatives being centroid-based methods like k -means. Such methods are strongly connected to quantization problems, which aim to approximate general probability measures with discrete ones. For example, k -means corresponds to quantization with respect to the Wasserstein distance. While Wasserstein quantization clusters points within a fixed space, this paper studies Gromov-Wasserstein (GW) quantization, which additionally aims at clustering the ambient geometry of the space. We show existence of solutions to the GW quantization problem and give a characterization that justifies an analogue to the k -means algorithm (Lloyd’s algorithm) to approximate them numerically. We further calculate the quantization rate for usual Euclidean geometries that are used in the GW context, and relate it to standard Wasserstein quantization rates. Finally, numerical experiments show that GW quantization opens up many modeling possibilities beyond normal clustering methods (e.g., for geodesic distances of 3D shapes or structured pruning of neural networks) and that the introduced algorithm leads to useful numerical solutions with approximation quality often in line with theoretically optimal rates.
[LG-67] hreshold Structure of Optimal Policies in Restart POMDPs
链接: https://arxiv.org/abs/2608.10936
作者: Konstantin Avrachenkov,Alexey Piunovskiy,Yi Zhang
类目: Optimization and Control (math.OC); Machine Learning (cs.LG); Probability (math.PR)
*备注: 8 pages, the paper has been accepted to IEEE CDC 2026
Abstract:We study a Restart POMDP (Partially Observable Markov Decision Process) on a general Borel state space, where the controller either lets the hidden state evolve unobserved or restarts the system and observes the new state. Exploiting a sufficient-statistic representation consisting of the last observed state and the elapsed time since restart, we reduce the problem to a fully observed MDP. Under a natural one-step cost deterioration condition, we prove that optimal policies have a threshold structure in the elapsed time for both the discounted and total undiscounted cost criteria. When the state space is partially ordered and the kernel is stochastically monotone, we further show that the optimal threshold is nonincreasing in the state. For the average cost criterion, under additional assumptions of geometric ergodicity and domination of the transient gain, we establish analogous threshold results via the vanishing discount approach, after showing the uniform boundedness of the optimal thresholds and relative value functions.
[LG-68] Self-Normalized Inference for Constant-Stepsize Temporal-Difference Learning under Markovian Sampling
链接: https://arxiv.org/abs/2608.10896
作者: Min Zeng,Yichen Zhang,Xiaofeng Shao
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:Constant-stepsize temporal-difference (TD) learning is attractive for policy evaluation, but inference from a single Markov trajectory must account for serial dependence and a stepsize-dependent stationary target. For fixed-stepsize linear TD, we establish a functional central limit theorem whose covariance retains the multiplicative component induced by the random TD matrix and the stationary iterate error. We then derive a joint functional limit for parallel Richardson–Romberg (RR) recursions driven by the same trajectory. A Brownian-bridge self-normalizer yields asymptotically pivotal confidence regions for prespecified state-value contrasts without estimating the long-run covariance or selecting a bandwidth or batch length. For such a contrast, the procedure admits a one-pass implementation whose memory does not grow with the trajectory length. At a fixed stepsize, the inferential center is the RR stationary target. We also study horizon-indexed designs in which the stepsize remains constant within each run and decreases across longer horizons. Under an explicit RR-dependent rate window, the residual RR target shift, multiplicative remainder, and initialization effect are negligible at the root- n scale, yielding inference for the projected Bellman solution. Experiments on FrozenLake and Garnet illustrate stationary-target coverage, RR target correction, and the finite-sample behavior of the horizon-indexed design.
[LG-69] Spectral Embeddings of Degree-α Laplacians in Random Dot Product Graphs
链接: https://arxiv.org/abs/2608.10845
作者: John Park,Ning Hao
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:Spectral clustering methods for network data are commonly based on a few matrix representations, such as the adjacency matrix and the symmetric Laplacian. We study a continuum of degree-normalized spectral embeddings that includes these commonly used choices as special cases. Under a random dot product graph model, we establish a row-wise central limit theorem for this family of embeddings. The result provides an explicit description of how degree normalization affects both population geometry and the local uncertainty of embedded nodes. We use the limiting distributions to compare different normalizations in two-community stochastic block models through a projected-Gaussian Bayes-error diagnostic. These comparisons show that no single normalization is uniformly preferred. Instead, the favored normalization depends on network density, community imbalance, and block-probability structure. Typically, stronger normalization is favored in lower-density or more imbalanced settings. These results provide a unified distributional understanding of when and why alternative normalizations may improve spectral clustering.
[LG-70] A lower bound for stepsize-based acceleration of gradient descent
链接: https://arxiv.org/abs/2608.10418
作者: Jianhao Ma,Yuxin Chen
类目: Optimization and Control (math.OC); Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:Recent work has shown that, for smooth convex optimization, plain gradient descent can be accelerated from its textbook convergence rate of O(T^-1) (where T denotes the number of iterations) to O\big(T^-\log_2(1+\sqrt2)\big) using carefully designed stepsize schedules alone, without resorting to momentum or other algorithmic modifications. Despite this progress, however, little was known about lower bounds for such methods beyond the classical \Omega(T^-2) benchmark for general first-order methods. In this work, we present a new lower bound of \Omega(T^-1.9319) for the last-iterate convergence rate of gradient descent with predetermined nonnegative stepsize schedules. This result provides rigorous evidence that stepsize schedules alone cannot accelerate plain GD to the optimal O(T^-2) convergence rate. The proof was developed by GPT-5.6 Sol Pro under the authors’ guidance.
[LG-71] On the Importance of Geometric Nonlinearity and Temperature-Dependent Properties in Multi-Material Thermo-Mechanical Topology Optimization
链接: https://arxiv.org/abs/2608.10344
作者: Shirin Hosseinmardi,Xiangyu Sun,Ramin Bostanabad
类目: Materials Science (cond-mat.mtrl-sci); Computational Engineering, Finance, and Science (cs.CE); Machine Learning (cs.LG)
*备注:
Abstract:Thermo-mechanical compliant devices are commonly designed with small-strain linear elasticity and temperature-independent material properties, even though they might operate hundreds of kelvin above ambient where both assumptions are questionable. In this work, we quantify the effect and cost of each assumption in multi-material topology optimization of thermally actuated compliant devices. To this end, we introduce a physics-informed, simultaneous analysis-and-design framework with (i) a finite-strain quadratic-Hencky (logarithmic-strain) constitutive model whose isotropic thermal eigenstrain admits an exact additive split in log-strain space, and (ii) temperature-dependent conductivity, thermal expansion, and elastic moduli for a titanium–copper–steel material system. We optimize a thermal actuator and a thermal gripper at three design temperatures under both a baseline model and the full physics, subject to mass and manufacturability constraints. Every converged design is re-evaluated by verified nonlinear finite element solvers in the full factorial of constitutive law and property model. The comparison between the two factors reveals that the constitutive law is the decisive modeling choice: These devices work as linkages where linear kinematics mistakes rotation for compressive strain; its error therefore grows with the design temperature and concentrates on the very layouts that exploit rotation best. Because a linear optimizer also steers away from the rotation-rich mechanisms that would expose this bias, the model can deceptively appear trustworthy when validated against its own designs. Designing with the full physics yields consistently stronger and more temperature-robust devices at a modest increase in design-time cost.
[LG-72] Stochastic Emulation of a Fully Coupled Preindustrial E3SMv3 Simulation
链接: https://arxiv.org/abs/2608.10277
作者: Elynn Wu,James P. C. Duncan,Troy Arcomano,Jeremy McGibbon,Oliver Watt-Meyer,Christopher S. Bretherton,Naser Mahfouz,Claudia Tebaldi,Luke Van Roekel,Andrew Roberts,Wuyin Lin,Finn Rebassoo,Jean-Christophe Golaz,Peter M. Caldwell
类目: Atmospheric and Oceanic Physics (physics.ao-ph); Machine Learning (cs.LG)
*备注:
Abstract:We present a stochastic coupled emulator of E3SM version 3, built on the SamudrACE framework, which couples an atmosphere emulator (ACE2) with a full-depth ocean emulator (Samudra). We replace the deterministic atmosphere emulator with its stochastic counterpart, ACE2S, and fine-tune the coupled system with a probabilistic objective, so that the atmosphere acts as a source of internal variability for the ocean. Trained on 105 years of a pre-industrial control simulation and evaluated on an independent 400 years, the emulator reproduces E3SMv3’s mean climate state with biases much smaller than existing model-to-observation differences. Relative to a deterministic baseline, stochastic training maintains internal variability across timescales, most notably in the ENSO power spectrum, eddy-rich SST anomalies, and sea ice variability in the marginal ice zone. The emulator captures daily precipitation accurately up to the 99.99th percentile, but underestimates the rarest tropical extremes. These results show that stochastic coupled emulators can reproduce long-timescale variability with high fidelity, while extrapolation to unseen extremes remains a key challenge.
[LG-73] BreastMammo and DenseMammo: Benchmarks for Mammography Domain Generalization MICCAI2026
链接: https://arxiv.org/abs/2608.10271
作者: Hongyi Pan,Gorkem Durak,Halil Ertugrul Aktas,Andrea Mia Bejar,Mustafa Ege Seker,Nebile Alibeyoglu,Rumeysa Guclu,Rana Gunoz Comert Bozkurt,Sibel Ozkan Gurdal,Neslihan Cabioglu,Beyza Ozcinar,Ravza Yilmaz,Vahit Ozmen,Erkin Aribal,Sukru Mehmet Erturk,Yalda Zafari,Mohamed Mabrok,Kayhan Batmanghelich,Mohammad Yaqub,Ziyue Xu,Ulas Bagci
类目: Image and Video Processing (eess.IV); Machine Learning (cs.LG)
*备注: This paper was accepted to the MICCAI 2026 workshop Deep-Brea3th
Abstract:Breast density classification is a critical component of breast cancer risk assessment, yet AI models often struggle to generalize across clinical sites due to vendor-specific acquisition styles. In this work, we introduce two new datasets, BreastMammo and DenseMammo, to facilitate robust multi-view mammography research. We propose a domain generalization framework that utilizes a foreground-only histogram matching protocol to resolve the domain shift issue arising from disparate clinical sources. Internal evaluation using a 5-fold cross-validation protocol demonstrates the efficacy of our approach, with the Swin Transformer backbone achieving a peak AUC of 98.32% for density classification. External evaluation on the TNMammo and LUMINA datasets demonstrates that the proposed approach consistently reduces domain shift, significantly outperforming prominent domain generalization paradigms, including MixStyle and Discrete-Fourier-Transform-based frameworks.
[LG-74] Deciding When to Switch: E-Processes for Adaptive Minimax Training for Generative Adversarial Nets
链接: https://arxiv.org/abs/2608.10096
作者: Hyunjoo Kim,Sicheng Wu,Agastya Venkatraman,Guang Lin,Sehwan Kim
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:Modern data science increasingly gives rise to hypothesis-testing problems that are not naturally formulated in terms of parameters within prespecified statistical models. One important example is the dynamic evaluation of optimization algorithms, where decisions must be made during training about whether further updates remain beneficial or the algorithm should switch to a different phase. This issue is particularly relevant in stochastic min-max optimization. Generative adversarial networks (GANs) provide a canonical example, as their training requires repeated decisions about when to switch between discriminator and generator updates, yet existing methods typically rely on fixed update ratios or heuristic criteria. We formulate this switching problem as sequential hypothesis testing and develop an e-process-based adaptive training procedure. During discriminator updates, one e-process tests the null that the discriminator-induced separation between the empirical data distribution and the generator law remains below a target level. During generator updates, with the discriminator fixed, a second e-process tests the reverse null that this separation remains above a refresh level. Conditional on the observed training sample, we prove that fresh empirical indices and latent draws yield conditional e-values that can be accumulated into e-processes, providing anytime-valid Type I error control under adaptive model updates and data-dependent switching. Across multimodal synthetic distributions and image benchmark datasets, the proposed method matches or outperforms the best fixed-ratio baselines under several widely used GAN objectives.
[LG-75] Deep Learning-Based Statistical Downscaling of Sea Surface Temperature Using a Residual Corrective Neural Network
链接: https://arxiv.org/abs/2608.10022
作者: Onkar Jadhav,Tim French,Ivica Janekovic,Nicole L. Jones,Matthew Rayson
类目: Atmospheric and Oceanic Physics (physics.ao-ph); Machine Learning (cs.LG)
*备注:
Abstract:The large-scale oceanic and atmospheric forecasts provided by global climate models typically lack sufficient resolution to accurately capture the response of the coastal ocean to atmospheric forcing and coastal circulation that drive fine-scale SST variability. Dynamical downscaling is computationally prohibitive, when applied to extensive coastlines, predictive ensembles, or long time periods. Therefore, this work presents a statistical downscaling of sea surface temperature (SST) from the seasonal coupled ocean-atmosphere forecast system (ACCESS-S2) using machine learning techniques. This study proposes a novel deep learning framework that uses a U-Net to generate an initial high-resolution SST estimate, which is subsequently refined using a residual corrective approach. The target SST fields are derived from the Regional Ocean Modeling System (ROMS). This two step approach called Residual Corrective Neural Network (RCNN) progressively refines initial U-Net predictions by incorporating dynamically scaled residuals at each step, enabling accurate capture of broad patterns and fine-grained features such as eddies and fronts. We also introduce a custom loss-assisted RCNN variant to improve performance during extreme events, which may be absent from training data due to climate-driven shifts in SST extremes. The framework efficiently downscales SST along the west coast of Australia. A 2011 marine heatwave case study shows that the RCNN improves ACCESS-S2 SST predictions by increasing horizontal resolution from 25 km to 2 km, enabling identification of fine-scale anomalies unresolved in the ACCESS-S2 dataset. This balance between computational efficiency and accuracy supports applications in coastal impact assessment and marine ecosystem studies.
[LG-76] HIPNO: Symmetry-Aware Physics-Informed Neural Operators for Noninvasive Hemodynamic Inference
链接: https://arxiv.org/abs/2608.10011
作者: Yunbei Pan,Jiahang Sha,Simon A. Lee,Maxime Cannesson,Wei Wang,Jeffrey N. Chiang
类目: Quantitative Methods (q-bio.QM); Machine Learning (cs.LG); Signal Processing (eess.SP)
*备注: 12 pages, 3 figures
Abstract:Continuous hemodynamic monitoring guides treatment decisions in surgery and intensive care. However, gold-standard signals are only measured in severe cases due to risks associated with invasive measurement. In this work, we introduce HIPNO (Hemodynamic Inference via Physics-informed Neural Operators) to recover hemodynamic state from ubiquitous, non-invasive signals and expand access to advanced monitoring. HIPNO addresses a problem of scale symmetry in physics-informed hemodynamic inference, where different combinations of flow, resistance, and compliance can generate the same observed pressure. We identify the symmetry group of the observation model and parameterize the network in its quotient space. For the 3-element Windkessel model, the quotient coordinates are the compliance-normalized flow U=Q/C , the decay time constant \tau_WK=R_2 C , and the characteristic-impedance coordinate \kappa=R_1 C . Across 945499 intraoperative windows from 2562 patients, HIPNO predicts \tau_wave , a proxy for vascular decay derived from pressure, with 32% lower error on the log scale than a population baseline while preserving mean arterial pressure accuracy. Because vascular decay and flow drive occupy separate coordinates, counterfactual perturbations produce the expected directional responses in at least 90% of windows in almost all prespecified scenarios, a separation unavailable to pressure-only baselines. The coordinates are also used as inputs to a calibration model for monitored cardiac output. Finally, the formulation identifies the external compliance or flow reference required to recover absolute physical scale.
[LG-77] Projected climate memory and inherited warm-tail risk in accelerated European summer warming
链接: https://arxiv.org/abs/2608.09966
作者: Mauricio Herrera-Marín,Alex Godoy-Faúndez,Diego Rivera
类目: Atmospheric and Oceanic Physics (physics.ao-ph); Machine Learning (cs.LG); Applications (stat.AP)
*备注:
Abstract:European summer warming reflects interactions among background change, persistent ocean–land–circulation states, and same-season variability. We develop an empirical reduced-dynamics framework that decomposes regional summer indicators into inherited slow-state memory, its predictable component, and contemporaneous innovation. Projection-operator theory motivates the decomposition, implemented with finite causal filters, ridge-regularised prediction, and logistic risk models. Using ERA5-derived summer indicators for 28 IPCC AR6 European sub-regions over 1950–2024, with validation on 2006–2024, we find that Mediterranean-state memory improves mean summer-temperature prediction relative to trend-only and ARX baselines. The gain over ARX is modest, while moving-average, exponentially weighted, and tempered filters contain similar annual information, indicating that the data identify useful slow-state memory more robustly than a unique kernel shape. Predictable-state reconstruction is ridge-sensitive and therefore treated diagnostically rather than as a forecasting model. The strongest result concerns warm-tail risk. In parsimonious logistic models, high accumulated Mediterranean memory raises predicted upper-tail event probability by about 8–11 percentage points for annual maximum summer temperature, warm-day frequency, and warm-spell duration at 1-, 3-, and 5-year horizons. Regional bootstrap intervals remain positive for all targets and horizons. Circular-shift placebos yield one-sided probabilities of approximately 0.05–0.14 and do not survive strict family-wise correction across nine tests, so the evidence is moderate rather than decisive. Overall, annual projected memory is not a universal short-horizon predictor, but a physically interpretable inherited risk-loading variable identifying years and regions predisposed to warm-tail outcomes.
[LG-78] AIFS-TC: A simple correction competitive with the operational frontier for tropical cyclone intensity forecasting
链接: https://arxiv.org/abs/2608.09959
作者: Anna Allen,Wessel P. Bruinsma,Michael Maier-Gerber,Harrison Cook,Matthew Chantry,Richard E. Turner
类目: Atmospheric and Oceanic Physics (physics.ao-ph); Machine Learning (cs.LG)
*备注: 6 pages, 5 figures, 2 tables
Abstract:AI weather models are in the process of revolutionising weather forecasting. While these models have been shown to achieve superior performance to physics-based NWP in forecasting tropical cyclone (TC) tracks, they dramatically underestimate intensity. Here we present AIFS-TC, a simple correction to the AIFS-Single model that is competitive with the operational state-of-the-art for forecasting maximum wind speed and minimum central pressure at lead times of 12 h to seven days. This performance also holds for rapid intensification events. Notably, the entire system was autonomously designed and built by a large language model (Claude Fable 5) in a few hours, directed through a small number of natural-language prompts by a single domain scientist. That the operational frontier can be reached with an open-source AI forecast model (AIFS-Single) and relatively simple, cheap post-processing is significant for TC science, and points to agentic coding as a route to rapid exploration and progress in life-saving early-warning systems in other domains.
[LG-79] An adaptive and evolvable deep reinforcement learning framework for weather prediction
链接: https://arxiv.org/abs/2608.09948
作者: Qiang Wu,Han Li,Jianping Huang
类目: Atmospheric and Oceanic Physics (physics.ao-ph); Machine Learning (cs.LG)
*备注:
Abstract:No single AI weather model excels at all variables, pressure levels, and lead times. Rather than building yet another architecture, we reframe the forecasting problem as one of coordination. Here we present Feitian Adaptive Ensemble Weather (FTAE-Weather), a lightweight framework that learns, through deep reinforcement learning, when and where to trust each member of an open pool of pretrained forecasters. A tactical Weight-Agent reads the current atmospheric state and assigns variable- and horizon-specific fusion weights, while a strategic Evolve-Agent periodically prunes underperforming models and absorbs newly released ones. Asynchronous prediction caching keeps training cost independent of the slowest constituent model. Adding fewer than 0.01 percent extra parameters, FTAE-Weather reduces RMSE by from 17.2 percent to 78.3 percent over the best individual model in 10 atmospheric variables and outperforms conventional ensemble baselines across lead times from 72 to 360 hours. The framework thus converts a growing, fragmented inventory of specialist models into a single prediction system that strengthens as the field of AI weather forecasting releases new architectures-turning model diversity from a coordination challenge into a compounding scientific advantage.
[LG-80] Optimized Sequential Testing for Binary Ensemble Classifiers
链接: https://arxiv.org/abs/2606.15237
作者: Joseph Kalman,Amit Moscovich
类目: Methodology (stat.ME); Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 33 pages, 5 figures
Abstract:Ensemble classifiers are predictive models that combine the results of simpler base models, often by majority vote. A classic example is random forests, which combine the predictions of decision trees. Ensembles that use more base models can be more accurate but also more costly to train and run. In this paper, we consider strategies for reducing the computational cost of binary classification using an approach from the field of sequential testing. Rather than evaluating all the base models and taking a majority vote, we evaluate the base models sequentially and stop execution when a clear majority emerges. We consider three different notions of optimality for early-stopping strategies that minimize the number of base models executed while controlling the rate of disagreement with the full ensemble. For each notion of optimality and allowable disagreement rate, we show that a linear program can be constructed and solved efficiently to find the optimal stopping strategy. We tested these methods on real-world datasets taken from the UC Irvine Machine Learning repository, and on the benchmark datasets proposed by Grinsztajn et al. We found that on most datasets, these methods provide speed-ups of 4x or more while controlling disagreement at 0.1%
[LG-81] Quantifying the noise sensitivity of the Wasserstein metric for images
链接: https://arxiv.org/abs/2510.01015
作者: Erik Lager,Gilles Mordant,Amit Moscovich
类目: atistics Theory (math.ST); Machine Learning (cs.LG); Image and Video Processing (eess.IV)
*备注:
Abstract:Wasserstein metrics are increasingly adopted as similarity scores for images. We consider the sensitivity of Wasserstein metrics with respect to pixel-wise additive noise when the images are treated as discrete measures on the pixel grid. We derive finite-sample expectation bounds for a Gaussian noise model. Among other results, we prove that the error in the signed 2-Wasserstein discrepancy scales with the square root of the noise standard deviation. This is favorable compared to the Euclidean metric that scales linearly, and thus provides a theoretical basis for the benefits of optimal transport distances in noisy settings. We present experiments that support our theoretical findings and point to a peculiar phenomenon where increasing the level of noise can decrease the Wasserstein distance. A case study on cryo-electron microscopy images demonstrates that the Wasserstein metric can capture the geometry of the data manifold in high noise settings even when the Euclidean metric fails.
附件下载
