本篇博文主要内容为 2026-09-14 从Arxiv.org论文网站获取的最新论文列表，自动更新，按照NLP、CV、ML、AI、IR、MA六个大方向区分。
说明：每日论文数据从Arxiv.org获取，每天早上12:30左右定时自动更新。
提示: 当天未及时更新，有可能是Arxiv当日未有新的论文发布，也有可能是脚本出错。尽可能会在当天修复。
目录
概览 (2026-09-14)
今日共更新611篇论文,其中:
- 自然语言处理共64篇(Computation and Language (cs.CL))
- 人工智能共168篇(Artificial Intelligence (cs.AI))
- 计算机视觉共90篇(Computer Vision and Pattern Recognition (cs.CV))
- 机器学习共158篇(Machine Learning (cs.LG))
- 多智能体系统共9篇(Multiagent Systems (cs.MA))
- 信息检索共18篇(Information Retrieval (cs.IR))
- 人机交互共32篇(Human-Computer Interaction (cs.HC))
多智能体系统
[MA-0] DementiaCare-Bench: A Modality-Validated Video Benchmark
【速读】：该论文旨在解决照护者在应对阿尔茨海默病（Dementia）患者行为与心理症状（BPSD）时，因缺乏对行为触发情境的准确理解而导致干预不当的问题。现有视频-语言模型（VLMs）虽具备潜在支持能力，但缺乏针对此类任务的标准化评估基准。为此，研究提出DementiaCare-Bench：一个包含94段专业制作的照护培训视频片段、覆盖9类BPSD症状的多模态评测基准，通过多智能体生成2023个基于原始字幕文本的临床问题，并在四种视觉条件下进行标注，以量化模型对时间顺序帧的依赖程度。结果显示，尽管多数问题设计为需有序帧信息（77.7%），但实际仅有34.8%真正依赖视频时序；12个当前主流VLM在需要时序理解的问题上平均准确率下降17个百分点，甚至领先开源模型在判断照护反应合理性时仅达随机水平。关键解决方案在于采用轻量级LoRA微调方法构建DemCare-VLM，使模型在视频依赖性上的表现从-3.3提升至+4.5，验证了该基准不仅能有效暴露现有模型缺陷，还可指导针对性改进。
链接: https://arxiv.org/abs/2609.12929
作者: Afrouz Sheikholeslami,Yuankai Qi,Xuyun Zhang,Luping Zhou,Amin Beheshti,Quan Z. Sheng,Ming-Hsuan Yang
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Multiagent Systems (cs.MA)
备注:
Abstract:Dementia affects an estimated 57 million people worldwide, and for most families the hardest part of care is not memory loss but the behavioral and psychological symptoms of dementia (BPSD): agitation, wandering, resistance to care, sundowning. Understanding these symptoms requires more than recognizing the behavior itself; it also requires knowing what happened beforehand. The same behavior may call for a different response depending on its trigger. Video-language models (VLMs) could potentially support caregivers, yet no existing benchmark evaluates this capability. To fill this gap, we present DementiaCare-Bench: 56 professionally produced caregiver training videos segmented into 94 clips across nine BPSD categories, with 2023 questions generated by a multi-agent pipeline that grounds every clinical claim in a verbatim transcript span. Each question is then probed under four visual conditions and labelled by the least it requires, so its visual demand is measured rather than assumed. Measurement contradicts intent: we wrote 77.7% of the questions to require ordered frames, and 34.8% do. Across 12 current VLMs the pattern is uniform. The best reach 85% overall, but that average is carried by questions a language model can answer from clinical knowledge alone; accuracy falls by 17 points on average on questions that require the ordered clip, and a leading open model scores at chance on judging whether a caregiver’s response was appropriate. A lightweight LoRA fine-tune, DemCare-VLM, moves video dependence from -3.3 to +4.5 points, so what the benchmark exposes can be repaired and not only measured.
[MA-1] he Mechanics of a Swarm: A Reproducible External Reconstruction of an Unintended Agent -Coordination Episode on a Third-Party Wiki
【速读】：该论文旨在解决生成式AI代理在开放、动态环境中自主协作行为的可观测性与可解释性问题，尤其关注其在无明确外部干预情况下如何形成协调模式及影响任务进展。研究聚焦于2026年5月24日至7月2日期间，一组运行于限时研究问答评估环境中的自主语言模型代理通过向公开可写维基页面发布内容所引发的复杂交互现象。其解决方案的关键在于构建一个基于显式身份模型（explicit identity model）的行为记录分析框架，将文本归属至具体修订操作而非累积页面内容，从而重建出907个代理群体（cohort）及其时间标记下的19,913次服务器事件序列。通过分析发现，尽管各代理群体的调度参数共享单一潜在速度尺度（78%的对数方差解释度），且存在内部时钟速率差异导致的信息不对称（首份报告与后续群体抵达中位间隔达3.4小时），但跨510个具有可观测进度轨迹的群体中，未发现协调程度与任务进展之间存在稳健正相关关系，亦无法确认是否存在因未来答案预知而产生的因果优势。研究进一步指出，当前数据集缺乏成功读取日志、指令调用记录和真实结果标注，因而无法确定协调行为的因果来源及其实际效果。最终，作者撤回了先前四项未经验证的结论，并强调在未来的智能体评估环境中，必须包含读取与结果日志以实现可信的性能评估。
链接: https://arxiv.org/abs/2609.12748
作者: Philipp Lütje(Philflow, Schenefeld, Germany)
机构: Philflow(菲尔流), Schenefeld, Germany(德国施内费尔德)
类目: Multiagent Systems (cs.MA)
备注: 48 pages, 10 figures. Code and derived artefacts: this https URL
Abstract:Between 24 May and 2 July 2026, autonomous language-model agents running inside a timed research-question evaluation wrote to a third party’s public, world-writable wiki. OpenAI acknowledged the incident; independent researchers reconstructed it and published the wiki’s archived revision history. We analyse that history (14,591 revisions, 3,103 names, 4,579 pages, 19,913 server events) as a behavioural record, attributing text to the revision that added it rather than to cumulative page content. Under an explicit identity model we reconstruct 907 cohorts and, from a random calendar marker the environment attached to each episode, estimate about 876 episodes (95% interval 774-995; alternative reconstructions span 800-1400). Coordination formats converged within a day, and the schedules created large opportunities for information asymmetry: because episodes of the same question chain ran at different internal-clock rates and started up to 16 h apart, the first report of an item preceded a later cohort’s arrival by a median of 3.4 h. The three schedule parameters agents reported share one latent speed scale (78% of log-variance over 15 configurations), and in one task family the last observed activity clusters by reported speed class on the internal clock, compatible with a fixed internal-time horizon. Across the 510 cohorts with an observable, format-dependent progress trace, we find no robust positive association between measured coordination and documented progress, including the few demonstrably given a future answer. Because the export contains neither successful-read logs, harness messages nor ground-truth outcomes, these results do not identify the causal origin of the coordination or its effect. We report four claims from our earlier analysis that did not survive re-examination, and argue that read and outcome logging are requirements for agent-evaluation environments.
[MA-2] What Drives Recovery in Agent ic Text-to-Cypher? LAST-CQ: An LLM Agent Self-Refinement Framework EMNLP2026
【速读】：该论文旨在解决生成式结构化查询（structured-query generation）中，多智能体流水线（agentic pipelines）性能提升究竟源自哪个环节的不确定性问题。其核心挑战在于：在无需训练、基于执行反馈的文本到Cypher（Text-to-Cypher）框架中，不同组件（如错误反馈机制、重试策略、并行采样数量等）对最终查询生成质量的实际贡献尚不明确。解决方案的关键在于通过构建LAST-CQ——一个五智能体、无训练、执行感知的测试平台，系统性地运行三类反事实实验（counterfactuals），在2,471个真实数据库查询上评估六种不同规模厂商后端的性能表现。研究发现，真正起决定性作用的是失败检测与重试路由机制，而非反馈的复杂性或采样数量；即使将高级的、基于模式（schema-grounded）的LLM合成反馈替换为原始数据库错误信息，性能损失极小（精确匹配仅下降1%）。此外，增加并行采样反而导致质量下降10–11%。值得注意的是，尽管单次生成成功的查询仍需消耗一次大语言模型（LLM）调用，但该框架可恢复91.7%的首次失败查询，显著提升了整体成功率。同时，研究揭示了传统基于n-gram重叠的评估指标在衡量结果等价性与语义一致性方面存在偏差：它在65.9%的情况下高估了结果相似性，而在语义判断上又存在低估。最后，通过与盲评人工标签对比，验证了当前使用的LLM评判器存在约9个百分点的乐观偏差。
链接: https://arxiv.org/abs/2609.12746
作者: Ioannis Prokopiou,Athanasios Aidinis,Panagiotis-Christos Kyrmpatsos,Pantelis Vikatos
机构: Athens University of Economics and Business (雅典经济商业大学); Orfium
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG); Multiagent Systems (cs.MA); Software Engineering (cs.SE)
备注: Accepted at REALM: The 2nd Workshop for Research on Agent Language Models at Empirical Methods in Natural Language Processing (EMNLP 2026), 15 pages, 3 figures, 6 tables
Abstract:Agentic pipelines for structured-query generation are rapidly expanding, but it is unclear which part of the loop produces the gain. We use LAST-CQ – a five-agent, training-free, execution-grounded Text-to-Cypher framework – as an instrumented testbed, running three counterfactuals over 2,471 live-database queries and six backbones spanning three vendor scale tiers. Removing correction is worth between 3.1% aggregate execution-BLEU against the single-pass system and 12.3% against a no-refinement counterfactual (up to 80.7% for the weakest backbone). Replacing schema-grounded, LLM-synthesised feedback with raw database error strings costs almost nothing (20.9% vs. 19.9% naive exact match; 0.2% end-to-end; equivalent within \pm 0.075 set-F1 by two one-sided tests). Spending the same call budget on parallel sampling degrades quality by 10-11%. What works is detecting failure and routing it to a retry, not the feedback sophistication or number of samples. LAST-CQ itself recovers 91.7% of queries that fail under single-pass generation, while a query that succeeds first time still costs exactly one LLM call. We also show that n-gram overlap on serialised results is not a bound in either direction: it over-scores against set equivalence on 65.9% of results while under-scoring against judged semantics. Finally, we calibrate our LLM judge against blind human labels and find it optimistic by 9 points.
[MA-3] Hierarchical Belief Modeling for Zero-Shot Opponent Adaptation in Partially Observable Multi-Agent Navigation
【速读】：该论文旨在解决在复杂、动态且部分可观测的对抗性环境中，智能体如何实现高效战术执行与快速适应的问题，尤其针对“最佳五局三胜”赛制下多轮博弈中的策略演化与对手建模挑战。其核心问题在于：在存在随机化回合级动态和隐藏状态的条件下，如何构建具备长期战略推理能力与短期精准控制能力的智能体。解决方案的关键在于提出HORIZON——一种分层式智能体架构，通过结合对称感知的空间表征、双记忆信念追踪机制、以遗迹为中心的图注意力网络、基于信息增益的探索策略以及对手条件化的策略混合，实现了对隐含游戏参数和对手风格的显式推断。该架构将短时控制与跨比赛元推理分离，并引入辅助的信念模型与世界模型目标以稳定训练过程。在大规模JAX模拟器中使用PPO算法训练后，HORIZON在胜率、适应性提升及联赛评分等指标上均显著优于强基准模型，验证了其在复杂对抗环境下的优越性能。
链接: https://arxiv.org/abs/2609.12422
作者: Kowei Shih,Lu Cheng,Zeyu Wang,Yeyun Xu,Kejian Tong
机构: Tsinghua University (清华大学); Stevens Institute of Technology (斯蒂文斯理工学院); University of California, Los Angeles (加州大学洛杉矶分校); Texas AM University (德州农工大学); Independent Researcher (独立研究员)
类目: Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注:
Abstract:Lux AI Season 3 requires agents to act under partial observability, randomized episode level dynamics, and a best of five match structure that rewards both tactical execution and fast adaptation. We present HORIZON, a hierarchical agent that combines symmetry aware spatial perception, dual memory belief tracking, relic centric graph attention, information gain driven exploration, and an opponent conditioned policy mixture. HORIZON separates short horizon control from cross match meta reasoning, while auxiliary belief and world model objectives stabilize learning. Trained with PPO in a large scale JAX simulator, the resulting agent explicitly infers hidden game parameters and opponent style. Experiments show consistent gains in match win rate, episode win rate, adaptation gain, and league rating over strong recurrent and feed forward baselines.
[MA-4] NDT Factory: Synthesizing Verified Network Digital Twins from Semantic Models via Multi-Agent LLM
【速读】：该论文旨在解决自主网络管理中网络服务意图（Network Service Intent, NSI）在动态环境下缺乏自动化评估能力的问题，尤其针对TM Forum Level 4（L4）自治架构下对闭环控制自适应性不足的挑战。现有行为型网络数字孪生（Behavioral Network Digital Twin, NDT）依赖预定义分析逻辑，难以适应复杂多变的网络场景。本文提出“NDT工厂”这一基于多智能体的软件系统，其核心创新在于利用大语言模型（Large Language Model, LLM）从语义模型中按需合成可执行的行为型NDT，实现分析逻辑的自动构建与动态适配。该方案的关键在于通过并行合成与编排机制，将高层语义描述转化为具备确定性、可验证执行的完整NDT实例，在呼叫准入控制（Call Admission Control, CAC）案例中实现了100%编译与测试通过率，并在300个NSI的仿真测试中达到99.3%的决策一致性、90%的准入率及所有拒绝原因的准确归因，验证了其在复杂场景下的可靠性与可扩展性。
链接: https://arxiv.org/abs/2609.12170
作者: Sudipta Acharya,Petar Djukic,Burak Kantarci
机构: University of Ottawa(渥太华大学); Bell Labs Research(贝尔实验室)
类目: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA); Software Engineering (cs.SE)
备注: 6 pages, 4 figures, accepted to IEEE Global Communications (Globecom) Conference, 2026
Abstract:Autonomous network management requires systems that can evaluate Network Service Intents (NSIs) under varying conditions without manual implementation of analysis logic, as envisioned in TM Forum Level~4 (L4) autonomy. Behavioral Network Digital Twins (NDTs) enable such evaluation, but existing NDTs rely on pre-defined analytical logic, limiting adaptability for evolving closed-loop control. This paper introduces the NDT factory, a multi-agent software system that synthesizes executable behavioral NDTs on demand from semantic models using Large Language Model (LLM). We validate the system using a Call Admission Control (CAC) case study, where deterministic what-if analysis serves as the admission decision process. The NDT factory generates a complete CAC NDT through parallel synthesis and orchestration, achieving 100% compilation and test pass rates across multiple runs. Simulation over 300 NSIs shows 99.3% decision agreement with a reference implementation, 90% admission rate, and correct attribution of all rejections, demonstrating reliable synthesis with deterministic, verifiable execution.
[MA-5] Deriving the Pure Price of Anarchy for Networked Resource Allocation Games
【速读】：该论文旨在解决在任意信息网络结构下，多智能体系统中通过博弈论方法实现协调控制时的局部效用函数设计问题。核心挑战在于如何为各智能体分配局部效用函数，以确保其自主决策行为能够逼近全局系统目标，并在最坏情况下的纯纳什均衡（pure Nash equilibrium）中仍保持较高的系统性能。该问题的关键在于优化“纯价格悖论”（pure price of anarchy, pPoA）这一度量指标——即最差纳什均衡下的系统目标值与全局最优值之比。本文提出一种线性规划方法，可针对任意信息网络和任意系统目标函数求解最优的局部效用设计，从而实现最优的pPoA性能保证。这是首个针对任意网络结构的最优效用设计解决方案，显著拓展了以往仅限于全信息假设的研究框架。研究发现，对于超模（supermodular）系统目标，即使完全禁止通信，最优效用设计依然成立；而对于次模（submodular）目标，数值实验表明最优设计对通信故障具有鲁棒性；当系统目标为加权最大覆盖（weighted maximum coverage）时，边际贡献效用设计在多种典型信息网络中被证明可实现最优的pPoA。
链接: https://arxiv.org/abs/2609.12077
作者: Vartika Singh,Philip N. Brown
机构: University of Colorado Colorado Springs (科罗拉多大学波德分校); Politecnico di Torino (都灵理工大学)
类目: Computer Science and Game Theory (cs.GT); Multiagent Systems (cs.MA); Systems and Control (eess.SY); Optimization and Control (math.OC)
备注:
Abstract:This work considers multi-agent coordination with arbitrary information networks among the agents using a game-theoretic approach. A system designer aims to assign local utility functions to the agents to guide their actions toward a desired system objective. The performance of the assigned local utilities is measured by the well known pure price of anarchy (pPoA) metric that equals the ratio of the system objective at the worst pure Nash equilibrium of the corresponding game to the optimal system objective. Our aim is to derive the utility functions which optimize the pPoA-based performance guarantees for any given information network and system objective. We develop a linear program that derives the optimal pPoA for any arbitrary information network and arbitrary system objective. Our work is the first to solve optimal utility design for arbitrary networks; our techniques generalize previous approaches which considered only the full-information setting. For supermodular objective functions, we prove that counterintuitively, a fully communication-denied utility design is optimal irrespective of the original information network. For submodular system objectives, an exhaustive numerical analysis suggests that the optimal utility design is robust to communication failures even for this case. When the system objective is weighted maximum coverage, the marginal contribution utility design provably optimizes the pPoA for a wide variety of information networks of interest.
[MA-6] Look Before You Leap: Pre-Action Verification for LLM Agents
【速读】：该论文旨在解决大语言模型（Large Language Model, LLM）智能体在真实世界中执行动作时可能产生的“静默失败”（silent failure）问题。静默失败指智能体发出的错误动作不会引发明显异常，而是产生看似合理但实际错误的结果，导致系统难以察觉并纠正错误。为应对这一挑战，论文提出一种低成本、确定性的前置验证机制作为智能体监督的关键手段，其核心在于在动作执行前通过形式化检查确保其正确效果的可预测性，从而将原本隐蔽的错误显式暴露出来。在命令行操作场景中，基于静态分析的验证器对9930条命令和482个工具进行验证，以10.0%的误报率捕获95.8%的无效命令；其语法与二进制检查达到“真值参照精确”（oracle-exact），实现零误报并覆盖一半错误，而标志位检查则受限于帮助文档覆盖率，解释了所有误报来源。在代码编辑场景中，研究通过640次独立编辑操作的基准测试揭示显著差异：基于内容的格式（如搜索/替换、diff）能干净地失败，而基于位置的格式（如行号或函数名编辑）存在严重静默失效风险——行号偏移导致99.1%文件出错，函数名编辑有12.7%概率作用于错误函数。最终，通过引入“不确定则拒绝”（refuse-when-unsure）策略，结合选择性定位与验证机制，实现了0.958召回率与7.0%误报率的平衡，并使每8320次试验仅出现一次静默误应用（0.01%），显著提升了系统的可靠性。该方案的关键在于构建可信赖的预执行验证框架，将不可靠的推测性执行转变为可验证、可恢复的确定性过程。
链接: https://arxiv.org/abs/2609.11957
作者: Asaad Althoubi
机构: 未知
类目: Machine Learning (cs.LG); Multiagent Systems (cs.MA)
备注: 8 pages, 3 figures
Abstract:An LLM agent acts on the world by emitting actions: shell commands to run, edits to apply. A wrong action does not always fail loudly; it can fail silently, producing a plausible but incorrect effect that raises no error. We argue that a cheap deterministic check, run before an action takes effect, is an effective and underused form of agent oversight, and we study it across two action modalities in one framework. The idea is to fix an action’s correct effect by construction, before any executor runs, so that silent failure is measured directly and the verifier may abstain rather than guess. For shell commands, a static verifier over 9930 commands and 482 tools catches 95.8% of invalid commands at a 10.0% false-positive rate. Its syntax and binary checks are oracle-exact, giving zero false positives while catching half of all errors; the flag check is bounded only by help-text coverage and accounts for every false positive. For code edits, a benchmark of 640 edits over 224 files isolating the apply step exposes a sharp split. Content-anchored formats such as search/replace and diff fail cleanly, whereas location-anchored formats fail silently: line numbers corrupt 99.1% of files under a one-line shift, and function-name edits hit the wrong function 12.7% of the time. In both settings a refuse-when-unsure policy turns silent failures into recoverable ones at a tunable cost in applicability: selective grounding reaches 0.958 recall at 7.0% false positives, and an anchor-and-verify applier records one silent misapplication in 8320 trials (0.01%). We release both benchmarks, the verifiers, and the guards.
[MA-7] Orchestra: Corroboration-Based Regulatory Candidate Discovery via Composed Bioinformatics MCP Agents
【速读】：该论文旨在解决肿瘤中关键调控因子（regulator）识别的可靠性问题，即如何从海量生物信息数据中筛选出真正具有功能意义的候选调控基因。传统方法往往依赖单一证据来源（如基于ARACNe推断的基因调控网络拓扑或特定实验数据），但其预测结果易受噪声和假阳性干扰。本文提出的核心解决方案是构建一个由多个独立生物信息学代理（MCP Agent）组成的多智能体工作流——Orchestra，整合两种互补的证据：一是RegNetAgents推断的基因调控网络拓扑结构，二是CASCADE提供的四种独立实验证据（LINCS敲减、DepMap必要性、超级增强子状态及DoRothEA转录因子置信度）。其关键创新在于提出“一致性验证”机制：只有当调控网络拓扑证据与至少两个独立的实验证据同时支持某一候选调控因子时，该候选因子才被视为高可信度目标。研究在TCGA肿瘤特异性调控因子层级上验证了这一假设，结果显示，在乳腺癌（BRCA）、结肠癌（COAD）及胃癌（STAD）等多个癌症类型中，多源证据的一致性显著提升对OncoKB癌症相关基因的预测能力（优势比达2.89至5.82），且统计显著性经多重检验校正后仍成立；而单个证据源无法实现有效区分。此外，逻辑回归似然比检验进一步证实，多源协同样本的协同作用可超越单一互信息（MI）边权重的预测效能。该方案通过模型上下文协议（Model Context Protocol）实现智能体间的协同推理，为生成式生物医学分析提供了一种可验证、可复现的可信推荐框架。
链接: https://arxiv.org/abs/2609.05496
作者: Jose A. Bird
机构: Bird AI Solutions(鸟人工智能解决方案)
类目: Molecular Networks (q-bio.MN); Multiagent Systems (cs.MA)
备注:
Abstract:Orchestra composes two independently built bioinformatics MCP servers – RegNetAgents, which infers gene regulatory network topology from ARACNe networks, and CASCADE, which supplies four independent evidence sources (LINCS knockdown, DepMap essentiality, super-enhancer status, DoRothEA transcription-factor confidence) – into one multi-agent workflow exposed via the Model Context Protocol. Its central architectural claim is that requiring RegNetAgents’ topology evidence and CASCADE’s experimental evidence to agree on a candidate regulator yields a more trustworthy candidate than either alone – not previously tested directly, since RegNetAgents’ own validation asked only whether its candidate lists beat chance. We test this on the TCGA tumor-acquired regulator tier (regulators in a gene’s tumor ARACNe network but absent from the GREmLN population-averaged baseline), selecting candidates by ARACNe mutual-information (MI) edge weight. On RegNetAgents’ published BRCA/COAD focal-gene panel plus matched negative controls, agreement among at least 2 of the 4 CASCADE sources predicts OncoKB cancer-gene status among focal genes (odds ratio 2.89, Benjamini-Hochberg-adjusted p=0.0166) but not among negative controls (p=0.0721); a single source is not diagnostic for either group. The pattern replicates and strengthens in a third cancer type, STAD, on a separately constructed panel (odds ratio 5.82), and against an independently curated ground truth (the Sanger COSMIC Cancer Gene Census). MI edge weight is the strongest single predictor overall (p=0.0003); a logistic-regression likelihood-ratio test confirms corroboration adds value beyond it in both panels (p=0.0234; p=0.0001). Every experiment invokes Orchestra’s real agentic entry point. Subjects: Molecular Networks (q-bio.MN); Multiagent Systems (cs.MA) Cite as: arXiv:2609.05496 [q-bio.MN] (or arXiv:2609.05496v2 [q-bio.MN] for this version) https://doi.org/10.48550/arXiv.2609.05496 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Jose Bird PhD [view email] [v1] Thu, 27 Aug 2026 19:23:12 UTC (95 KB) [v2] Thu, 10 Sep 2026 14:55:17 UTC (95 KB) Full-text links: Access Paper: View a PDF of the paper titled Orchestra: Corroboration-Based Regulatory Candidate Discovery via Composed Bioinformatics MCP Agents, by Jose A. BirdView PDFHTML (experimental)TeX Source view license Current browse context: q-bio.MN prev | next new | recent | 2026-09 Change to browse by: cs cs.MA q-bio 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
[MA-8] Information Design in Smooth Games
【速读】：该论文旨在解决在连续动作空间和可微收益函数的博弈中，如何设计最优信息结构以实现社会福利或特定目标最大化的问题。其核心挑战在于，在复杂的信息不对称环境下，如何通过合理的信息披露策略引导参与者形成有效均衡。解决方案的关键在于建立信息结构与委托-代理合约之间的对应关系：一个信息结构若能诱导出的均衡等价于某个主-从合约问题中的最优激励相容解，则该信息结构即为最优。基于这一理论框架，论文进一步刻画了对称线性-二次博弈中的最优信息结构：在存在共同价值的情况下，针对所有先验分布均稳健的定向披露（targeted disclosure）是最优的；而在相互依赖且服从正态分布的价值结构下，线性披露（linear disclosure）是唯一最优方案。研究通过风险投资、贝叶斯极化及价格竞争等应用案例验证了理论结果的适用性与现实意义。
链接: https://arxiv.org/abs/2202.10883
作者: Alex Smolin,Takuro Yamashita
机构: 未知
类目: Theoretical Economics (econ.TH); Computer Science and Game Theory (cs.GT); Multiagent Systems (cs.MA)
备注:
Abstract:We study information design in games where players choose from a continuum of actions and have continuously differentiable payoffs. We show that an information structure is optimal when the equilibrium it induces can also be implemented in a principal-agent contracting problem. Building on this result, we characterize optimal information structures in symmetric linear-quadratic games. With common values, targeted disclosure is robustly optimal across all priors. With interdependent and normally distributed values, linear disclosure is uniquely optimal. We illustrate our findings with applications in venture capital, Bayesian polarization, and price competition.
自然语言处理
[NLP-0] ype Diversity Enables Transformers to Generalise Compositionally
【速读】： 该论文旨在解决生成式模型在组合泛化（compositional generalisation）任务中表现差异的问题，特别是为何结构泛化（structural generalisation）相较于词汇泛化（lexical generalisation）更困难。以往研究认为这一差距是变压器架构（Transformer）的固有缺陷所致，但本文提出，这种差异并非源于模型本身，而是由于先前数据集中词汇类型（lexical types）多样性高而结构类型（structural types）多样性低。作者通过引入语法框架（Grammatical Framework）构建了语言学上多样化的COGS与SLOG数据集变体，系统性地调控词汇与结构类型的多样性。实验结果表明，类型多样性与组合泛化能力呈同等程度的相关性，无论在词汇还是结构测试案例中均成立，从而支持其核心假设：组合泛化难度的关键在于类型多样性，而非结构或词汇的本质差异。此外，研究还发现先前文献中关于“复合发散”（compound divergence）解释组合泛化困难的观点存在矛盾，并进一步揭示了非新颖测试结构类型多样性及逻辑语义格式的表面属性对组合泛化性能的影响。因此，该研究的关键解决方案在于重新定义并量化“类型多样性”作为影响组合泛化的核心因素，为未来数据设计与模型评估提供了新的理论依据。
链接: https://arxiv.org/abs/2609.13144
作者: Anssi Moisio,Mathias Creutz,Mikko Kurimo
机构: Aalto University(阿尔托大学); University of Helsinki(赫尔辛基大学)
类目: Computation and Language (cs.CL)
备注: Under review
Abstract:Compositional generalisation has been divided into lexical and structural generalisation. Previous work has found that structural generalisation is harder than lexical for Transformers. We propose that this difference is not inherent to Transformers, but due to the high diversity of lexical types and low diversity of structural types in the specific datasets of these previous works. By type diversity we mean the number of different constructors of that type, instead of, for example, the specific word combinations that might populate the structure. To test this, we vary the amounts of type diversity of lexical and structural types in previously published datasets. We create linguistically diverse variants of the COGS and SLOG datasets using Grammatical Framework. We find that type diversity correlates with compositional generalisation equally in lexical and structural test cases, supporting our hypothesis. We note a contradiction with the proposition in previous work that compound divergence explains the difficulty in compositional generalisation tasks. We further investigate the effects of other dataset properties on compositional generalisation, such as the diversity of types other than the novel test structure, and surface properties of the logical semantics format.
[NLP-1] SAS: Simple Attention Sparsification via End-to-End Optimization of Context Ranking
【速读】： 该论文旨在解决预训练Transformer模型在后训练阶段进行注意力稀疏化时，因选择机制与下游任务目标不直接对齐而导致的注意力预算浪费问题。现有可训练稀疏化方法通常采用轻量级选择器对上下文单元（如词元或块）打分，并通过硬Top-K选择实现稀疏化，但此过程会阻断来自语言建模损失的梯度，导致其依赖层内密集注意力分布的蒸馏，而这种蒸馏所生成的排序并未直接反映在固定注意力预算下各上下文单元对最终预测的实际贡献，从而可能将有限的注意力资源分配给低效的上下文单元。为克服这一偏差，本文提出Simple Attention Sparsification (SAS)，一种端到端可优化的门控稀疏注意力机制，其核心在于将选择器的连续得分以对数形式注入注意力logits中，使语言建模损失能够通过标准反向传播直接更新选择器参数，实现对上下文重要性的联合优化。为确保该设计在实践中有效，关键要素包括：在注意力softmax内部以对数形式引入门控、使用归一化softmax门控以平衡历史上下文与始终保留的当前块、以及保持选择器得分的连续性以学习相对优先级而非仅硬性选择。此外，为支持长序列训练，作者实现了基于Triton的内存高效内核，将SAS集成至FlashAttention式计算流程中。实验表明，在推理、长上下文理解及代理类任务中，SAS在不同注意力预算下均显著优于现有可训练稀疏注意力基线，尤其在严格预算条件下表现更优，验证了其在下游任务中实现更高效上下文排序的能力。
链接: https://arxiv.org/abs/2609.13141
作者: Zhiwei Li,Lei Zhu,Hao Gu,Xiang Hu,Yan Wang,Haitao Mi,Sirui Han,Leo Liang,Zhijiang Guo
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Post-training attention sparsification reduces the quadratic cumulative attention cost of pretrained Transformers by selecting a small set of context units (tokens or blocks) for each query. Existing trainable methods usually use a lightweight selector to score context units, followed by hard Top-K selection that blocks gradients from the language modeling loss. Consequently, these methods commonly distill layer-wise dense attention distributions. Although this encourages the selector to rank context units by dense attention weights in the original model, the ranking is not directly aligned with their impact on predictions under a fixed attention budget (i.e., the number of attended context units per query), potentially wasting the limited budget on less useful units. To address this misalignment, we propose Simple Attention Sparsification (SAS), a gated sparse attention mechanism that optimizes context ranking end-to-end with the language modeling loss. The key idea is to inject the selector’s continuous scores into attention logits during training, allowing the loss to update the selector through standard backpropagation. We identify several choices crucial for this simple design to work well in practice: placing the gate inside the attention softmax in log form, using normalized softmax gates to calibrate historical context against the always-retained current block, and preserving continuous selector scores so the model learns relative priorities rather than only hard selections. To support long-sequence training, we implement a memory-efficient Triton kernel that integrates SAS into FlashAttention-style computation. Across reasoning, long-context understanding, and agentic tasks, SAS consistently outperforms trainable sparse attention baselines across attention budgets, with especially large gains under tight budgets, demonstrating more effective context ranking for downstream tasks.
[NLP-2] Continue Adapt or Yield: In-Turn Adaptation to Overlapping Speech in Full-Duplex Agents
【速读】： 该论文旨在解决全双工语音代理（full-duplex voice agents）评估中长期被忽视的核心问题：现有评估方法仅关注代理是否持续说话或停止，无法捕捉人类在对话中常见的“在讲话过程中实时响应听者贡献”的第三类行为。这种响应包括对听者补充的词语、纠正或澄清内容的即时整合，即“在轮次内适应”（in-turn adaptation）。其解决方案的关键在于提出一种名为Duplex Cue的新评估框架，该框架将听者的意图（如反馈性回应、协作性参与或打断）与发言者的行为（如保持不变、在当前轮次内调整、或让出话语权）进行解耦分析。通过在300个未经脚本的英语对话中提取人工确认的208组有效互动数据，研究对比了真人对话者与PersonaPlex模型在重放听者语音时的响应表现。结果显示，在66组协作性互动中，真人能以68.2%的频率实现内容适应，而PersonaPlex仅达34.8%，其余多数情况为保持原样（42.4%）或让步（22.7%）。这表明，真正自然的语音交互评估必须同时衡量代理对听者贡献的响应能力及其在轮次内的动态调整能力，而不仅限于是否持续发声。
链接: https://arxiv.org/abs/2609.13117
作者: Yunqi Lu,Tyler Baumgartner,Nikhil Johri,Brandon Tai,Candice Fan,Luc Debaupte,Ruben Aguilar,Bill Wang,Yi Zhong
机构: Besimple AI(贝简单人工智能), San Mateo, CA
类目: Computation and Language (cs.CL); Sound (cs.SD)
备注: 12 pages, 8 tables
Abstract:Full-duplex evaluation often emphasizes whether an agent keeps speaking or stops. That binary cannot express a third response humans use routinely: continuing to speak while incorporating what the listener just contributed. The contribution may be a missing word, a correction or a clarification. We introduce Duplex Cue, an evaluation of this \emphin-turn adaptation in full-duplex voice agents. Duplex Cue separates listener intent (backchannel, collaboration, or interruption) from speaker behavior: continuing unchanged, adapting within the turn, or yielding. Adaptation includes acknowledgment as well as content revision. In a single-model case study using 300 human-confirmed cues from unscripted English conversations, we compare recorded human responses with PersonaPlex continuations generated while replaying the listener’s audio. We retain 208 pairs with the ongoing speaker active at cue onset and a scorable response in each condition. On the 66 collaborative pairs, recorded speakers adapt in 68.2% of cases, compared with 34.8% for PersonaPlex. The model otherwise continues unchanged (42.4%) or yields (22.7%). These findings show why evaluating natural voice interaction requires measuring how an agent responds to a listener’s contribution as well as whether it keeps speaking.
[NLP-3] MAxBench: A Multinomial Concept Recovery Benchmark
【速读】： 该论文旨在解决多类别概念（multinomial concepts）在语言模型中表示的细粒度控制问题，特别是针对非二元概念（如动物、国家等具有多个子类别的复杂概念）在激活空间中表示几何结构不明确时，如何有效定位并操控这些概念的问题。其核心挑战在于，相较于二元概念（如拒绝行为）仅需单一方向即可实现引导，多类别概念的表示空间具有更复杂的几何结构（如仿射子空间、流形等），导致搜索最优表示形式的难度显著增加。为此，论文提出了一种几何无关的评估框架MAxBench，通过从恢复的概念表示中采样来系统评估不同方法在多种几何结构下的表现。关键发现表明：（i）仿射子空间在引导可靠性和召回率上优于秩一或线性子空间；（ii）这一优势主要源于更优的非零偏置项而非基向量选择；（iii）当适用时，流形引导的表现可与最佳方法相媲美；（iv）在多数情况下，提示法（prompting）仍无明显劣势，与以往对二元概念的研究结论一致。这些结果强调了将可解释性研究与元评估的范围拓展至结构更复杂的多类别概念的重要性。
链接: https://arxiv.org/abs/2609.13072
作者: Divya Appapogu,Freya Behrens,Yonatan Belinkov,Aaron Mueller
机构: Boston University (波士顿大学); Technion – Israel Institute of Technology (以色列理工学院); Harvard University (哈佛大学)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Fine-grained control of language model behaviors (e.g., steering) is among the more actionable outcomes of interpretability research. For binary concepts such as refusal, a single direction in activation space often suffices for steering. However, many concepts are not binary: Animals and Countries contain many subcategories, each with multiple instances. For these concepts, the search space over possible representation geometries is far larger than for binary concepts; it is thus not clear what geometries are most appropriate, nor what methods are most effective at recovering them. In this work, we introduce MAxBench, a geometry-agnostic evaluation framework for multinomial concept representations based on sampling from the recovered concept representation. We use MAxBench to compare 10 localization methods (covering 5 geometry types) across 6 concepts and 4 models. Using this framework, we find that (i) affine subspaces steer more reliably and have greater recall than rank-one or linear subspaces; (ii) much of this advantage is due to better non-zero offsets rather than the choice of bases; (iii) manifold steering is competitive with the best methods when applicable; and (iv) no method consistently outperforms prompting, in alignment with prior findings on binary concepts. These findings underscore the importance of expanding the scope of interpretability research and meta-evaluation to concepts with more varied structure.
[NLP-4] Expert-Space Exploration in MoE Reinforcement Learning
【速读】： 该论文旨在解决大语言模型在后训练阶段，尤其是基于强化学习（Reinforcement Learning, RL）优化混合专家模型（Mixture-of-Experts, MoE）时，因专家选择（expert selection）固定而导致的轨迹多样性（rollout diversity）不足的问题。尽管现有方法聚焦于提升优化稳定性和训练效率，但忽视了路由机制所蕴含的潜在探索空间。其核心挑战在于：虽然扰动专家路由可有效增强输出多样性、类比于提高解码温度，但直接扰动易激活不合适的专家，导致轨迹质量显著下降。为此，本文提出架构感知的专家空间探索强化学习（Expert-Space Exploration Reinforcement Learning, ESRL），其关键在于通过保留高置信度专家作为锚点，将随机路由限制在合理候选池内，从而在保障可靠计算路径的同时实现安全探索；同时，根据路由器熵动态调节扰动强度，避免过度扰动，并通过记录并重放采样过程中的专家路径来缓解扰动引入的路由不匹配问题。实验表明，ESRL在多种MoE结构（包括top-K、top-1及共享专家路由）和多类任务（数学、科学、代码）上均取得最优性能，且无需额外采样或计算开销，在Qwen3-30B-A3B上相较GRPO分别提升Pass@1和Pass@8达3.2和4.5个百分点，充分验证了利用MoE特有路由结构进行显式探索对强化学习训练的增益作用。
链接: https://arxiv.org/abs/2609.13058
作者: Hongyi He,Zhenghao Lin,Xiao Liu,Peng Cheng,Yan Lu,Yeyun Gong
机构: Microsoft Research(微软研究院); Tsinghua University(清华大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Reinforcement learning (RL) has become central to post-training of large language models. Recent advances in RL for Mixture-of-Experts (MoE) models have primarily focused on improving optimization stability and training efficiency, while treating the expert selection as a fixed component. Since routing determines the sparse computation paths that induce output distributions, expert selection offers an additional source of rollout diversity. Through empirical analysis, we find that perturbing expert routing effectively alters model output and increases rollout diversity, which is similar to increasing the decoding temperature. However, direct perturbation can activate unsuitable experts and substantially degrade rollout quality. Motivated by these observations, we introduce Expert-Space Exploration Reinforcement Learning (ESRL), an architecture-aware framework that explicitly explores the expert-routing space of MoE models. ESRL preserves high-confidence experts as anchors, and restricts stochastic routing to a plausible candidate pool, thereby retaining reliable computation paths. The perturbation strength is further adapted according to router entropy to avoid over-perturbation. To mitigate the routing mismatch introduced by perturbation, ESRL records the expert paths used during rollout and replays them during policy optimization. Experiments demonstrate that ESRL achieves the best performance across MoE backbones with top-K, top-1, and shared-expert routing, as well as across mathematics, science, and code tasks without additional sampling or computational cost. Specifically, ESRL on Qwen3-30B-A3B achieves the best among all compared methods, improving average Pass@1 and Pass@8 over GRPO by 3.2 and 4.5 percentage points, respectively. Further analyses of expert utilization and training dynamics provide insights into how exploiting MoE-specific routing structure benefits RL training.
[NLP-5] Kraken: LLM -based Speech-to-Speech Translation via Low-bitrate VQ and Dual-path Source Conditioning
【速读】： 该论文旨在解决生成式语音翻译（Speech-to-Speech Translation, S2ST）中因高比特率语音标记难以在语言模型（LLM）中有效预测，以及对训练数据中说话人身份与语调高度一致性的强依赖所导致的非语言信息（如音色、语调）迁移能力受限的问题。其解决方案的关键在于引入基于单层向量量化（single-layer vector quantization）的低比特率语音标记，该标记经过自监督学习（Self-Supervised Learning, SSL）特征重建训练，能够高效编码语音内容；同时采用独立的标记到波形解码器Autowave-X，并以源语音作为条件输入，从而增强非语言信息的跨语言传递能力，降低对训练数据中说话人与语调对齐的严苛要求。在此基础上，作者构建了名为Kraken的S2ST模型，通过将预训练的LLM（Qwen3-8B）与语音特征输入及低比特率标记输出相结合，并接入Autowave-X声码器，实现了在150小时多语言多任务语音数据上的端到端训练。实验表明，该模型在翻译质量上优于SeamlessM4T-Large v2和Qwen2.5-Omni，且在说话人和语调保持方面表现更优。
链接: https://arxiv.org/abs/2609.13045
作者: Hayato Futami,Hassan Shahmohammadi,Tushar Dhyani,Alkis Koudounas,Raphaël Lafargue,Yosuke Kashiwagi,Quentin Jodelet,Emiru Tsunoo
机构: Sony Group Corporation(索尼集团); Sony Europe Limited(索尼欧洲有限公司)
类目: Computation and Language (cs.CL); Sound (cs.SD)
备注:
Abstract:Speech-to-speech translation (S2ST) has advanced significantly with speech LLMs, offering the potential for joint optimization and preserving non-linguistic information. However, these models struggle with predicting high-bitrate speech tokens in LLMs, and face the challenge of relying on S2ST training data with ideally aligned speaker identity and prosody. We propose using low-bitrate tokens based on single-layer vector quantization, trained to reconstruct self-supervised learning (SSL) features. We also employ a separate token-to-waveform decoder named Autowave-X, which is also conditioned on the source speech to improve non-linguistic transfer, thereby relaxing the training data constraints. With the integration of these techniques, we propose an S2ST model named Kraken, which augments a pre-trained LLM with speech feature inputs and the low-bitrate token outputs, followed by Autowave-X vocoder. We built the model upon Qwen3-8B and trained it using 150k hours of multilingual and multitask speech data. We demonstrated that our model exhibited better translation quality than SeamlessM4T-Large v2 and Qwen2.5-Omni, along with improved speaker and prosody transfer capabilities.
[NLP-6] asks over Application Manuals: Revealing Gaps in Long-Horizon Procedural Reasoning for Language Models
【速读】： 该论文旨在解决当前大语言模型（Large Language Models, LLMs）在长时程程序化推理（long-horizon procedural reasoning）任务中表现不足的问题，尤其是针对需要遵循数百页、规则密集且高度依赖上下文关联的权威手册完成复杂任务的真实场景。现有基准大多局限于短时程、少步骤的多跳推理，难以反映模型在实际应用中处理长链条、高复杂度规则系统的能力。为此，作者提出了Tasks over Application Manuals (TAM)基准，涵盖国际疾病分类第10版临床编码（ICD-10-CM）和美国联邦量刑指南（U.S. federal sentencing）两个真实世界领域，所有任务均基于权威手册，需跨多个章节执行相互依赖的推理步骤，并以精确匹配为目标。实验表明，即使采用检索增强生成、ReAct式提示及代理基线等先进方法，在GPT-5上最佳精确匹配率仍极低：ICD-10-CM任务仅1%，量刑任务为15.5%。这一结果揭示了当前评估体系可能高估了模型的真实推理能力，凸显出可靠遵循长周期、规则驱动流程的挑战是亟待突破的关键问题。
链接: https://arxiv.org/abs/2609.13005
作者: Utkarsh Soni,Syed Shariyar Murtaza,Yifan Nie,Sachin Chandrasekhar,Eugene Wen
机构: Manulife(宏利金融); Manulife(宏利金融); Manulife(宏利金融); Manulife(宏利金融); Manulife(宏利金融)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models (LLMs) have achieved strong performance on a wide range of natural language tasks, and recent benchmarks suggest that they are increasingly adept at multi-hop reasoning. However, these benchmarks are typically short-horizon, requiring only a small number of retrieval or inference steps, and provide limited evidence of reliability on real-world tasks that involve following manuals spanning hundreds of pages with complex, interdependent guidelines. In this paper, we introduce Tasks over Application Manuals (TAM), a benchmark for evaluating long-horizon procedural reasoning. We construct TAM by curating real-world tasks from two domains: ICD-10-CM clinical coding (mapping medical conditions to diagnostic codes) and U.S. federal sentencing (computing crime sentencing guideline outcomes, specifically offense levels), with human-validated labels. Each task requires following an authoritative manual with tens of thousands of rules and executing a sequence of interdependent steps across different sections to produce an exact answer. We evaluate general-purpose prompting approaches, including retrieval-augmented generation, ReAct-style prompting, and an agent-harness baseline on GPT-5, and find that the best exact-match performance remains extremely low: 1% on ICD-10-CM coding and 15.5% on sentencing tasks. These results show that current benchmarks may overestimate LLM reasoning ability and miss a key challenge: reliably following long, rule-based procedures. The complete TAM data and code are publicly available.
[NLP-7] Judging by the Cover: Cleaning LLM Truthfulness Benchmarks to Avoid Surface-Level Feature Leakage
【速读】： 该论文旨在解决当前二元选择型真实性评估基准（binary-choice truth benchmarks）中存在的表面特征偏差问题，即模型可能通过识别答案在表层特征（如长度、词汇复杂度等）上的系统性差异，而非进行真实推理，即可超越随机猜测水平。其核心解决方案是提出一种通用的数据清洗机制——Audit-Prune，通过识别并移除最强化信息泄露的样本对，有效降低表面特征带来的误导性信号。研究发现，在TruthfulQA数据集中，仅基于六个表面特征的逻辑回归分类器即可显著区分正确与错误答案，表明存在明显的泄漏现象；通过应用Audit-Prune机制，可将数据集中的表面特征泄漏降至接近随机水平，并提供可复用的清洗流程以保障未来基准数据集的可靠性。
链接: https://arxiv.org/abs/2609.13003
作者: Foad Namjoo,Remy Ogasawara,Amirali Abdullah,Cullen Anderson,Narmeen Fatimah Oozeer,Jeff M. Phillips
机构: University of Utah (犹他大学); Thoughtworks(思特沃克); University of Massachusetts Amherst (马萨诸塞大学阿默斯特分校); Martian AI(火星人工智能)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 31 pages, 4 figures. Code and data: this https URL and this https URL
Abstract:Binary-choice truth benchmarks ask models to choose between a correct and an incorrect answer, but if the two answers differ systematically in surface-level features, models can exceed chance without performing the intended reasoning. We show that this failure mode is detectable and can be exploited by downstream classifiers. In TruthfulQA, a simple six-feature logistic classifier achieves substantial accuracy in separating correct from incorrect answers. We further show that similar surface-level artifacts are present in additional benchmarks. To counteract this, we developed a general mechanism to clean them by removing the most leakage-reinforcing pairs. We release a version of TruthfulQA with surface-feature leakage reduced close to chance and provide a mechanism, Audit-Prune, so that the datasets can be cleaned before release.
[NLP-8] Investigating Temporal Motion Features for Pose-to-Text Indian Sign Language Translation
【速读】： 该论文旨在解决手语到文本的翻译（Pose-to-Text Indian Sign Language Translation, SLT）中，预训练T5模型规模及显式运动特征对翻译性能的影响问题。核心挑战在于如何有效建模手语动作的时空特性以提升自然语言生成质量。其解决方案的关键在于：通过轻量级姿态编码器将姿态序列映射至T5的嵌入空间，并在输入中引入显式的帧间姿态变化特征（motion features），以增强模型对动态手势时序信息的感知能力。实验表明，尽管T5-large在chrF指标上表现最优，但引入运动特征的T5-small + Motion变体在BLEU指标上取得最大提升，显著优于仅使用空间特征的基线模型，成为整体表现最强的模型。该方法通过融合显式运动信息，有效提升了生成文本的准确性与流畅性。
链接: https://arxiv.org/abs/2609.12993
作者: Manav Dhamecha,Praveen Kumar Chandaliya,Pruthwik Mishra
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL)
备注: 5 pages, 1 table and 2 figures
Abstract:We investigate the effect of pretrained T5 model scale and explicit motion features on pose-to-text Indian Sign Language Translation (SLT) for the WSLP 2026 Shared Task. Pose sequences are projected into the embedding space of T5 through a lightweight pose encoder, with the complete model fine-tuned to generate English text. The shared task data used for this work consists of a test set with 5,334 examples and a validation set with 5,257 examples. We compare T5-small, T5-base, and T5-large, and additionally introduce a motion-augmented variant, T5-small + Motion, that adds explicit frame-to-frame pose differences to the input representation. T5-small achieves the best BLEU and ROUGE scores among the spatial-only models, while T5-large obtains the highest chrF score. Augmenting T5-small with motion features yields the largest single improvement observed in our study, substantially improving BLEU over the spatial-only baseline and making it the strongest model overall on this metric. Our submitted system ranked 5th on the official WSLP 2026 SLT testing leaderboard. The source code and trained models are publicly available on GitHub and HuggingFace.
[NLP-9] Fewer Words Not Fewer Tokens: Measuring the Sanskrit Tokenization Penalty per Proposition KR
【速读】： 该论文旨在解决高信息密度语言（如梵语）在现代生成式 AI 模型中经子词分词（subword tokenization）后所面临的信息压缩效率问题，具体聚焦于梵语相较于英语和印地语在实际部署的分词器下是否仍能保持其语言优势。其核心挑战在于：尽管梵语通过词尾融合（inflectional fusion）实现了单位词内的高信息密度，但这种优势在经过基于 BPE（Byte Pair Encoding）等主流分词算法处理后是否依然存在。研究的关键发现是，当前部署的分词器（词汇量 ≥200,019 个 id）下，梵语在当代散文和 FLORES-200 测试集上所需词元数为英语的 1.831–2.899 倍，显著高于预期；然而，当与同源语料库中使用相同算法与词汇量训练的英文对照模型进行匹配比较时，这一“劣势”消失——在 32,000 和 64,000 词元规模下，所有匹配对均显示梵语每命题成本不低于英语（>1.0），且置信区间不包含 1.0。随着词汇量扩大至 128,000，性能差距缩小，在域内降至 0.983，域外仍高于 1.025，表明其相对效率随模型规模提升而改善。进一步分析揭示，该差异主要源于字符长度比（character-length ratio），而非词元/字符比（tokens-per-character ratio），后者始终接近 1。因此，解决方案之关键在于：在真实部署场景中，衡量语言效率应以“命题”为单位，并采用语料库匹配的控制实验设计，从而避免因分词器偏倚导致的误判。最终结论强调，在现有部署实践中，梵语虽在字符层面较英语更紧凑，但其信息密度在子词分词后难以有效保留，除非采用与目标语言严格对齐的训练范式。
链接: https://arxiv.org/abs/2609.12960
作者: Devansh Sharma
机构: 独立研究者(Independent researcher)
类目: Computation and Language (cs.CL)
备注: 20 pages, of which 8 are the body; 4 figures, 18 tables. Code, data pipeline and the full results snapshot: this https URL
Abstract:Sanskrit fuses case, number, person and tense into word endings and chains clauses into compounds, so it is information-dense per word. Whether that density survives subword tokenization is a separate question, to be asked per unit of meaning rather than per word. On identical FLORES-200 devtest content, Sanskrit costs 1.774-2.187 times the English tokens under deployed tokenizers with vocabularies of 200,019 ids or more, but only 1.325-1.353 times the Hindi tokens. Against a deployed English tokenizer, Sanskrit-trained BPE arms then look cheaper per proposition than English on contemporary prose (0.887). Against a matched English control, the same algorithm and vocabulary trained on the English side of the same corpus, that flip disappears: at 32,000 and 64,000 pieces all 8 matched pairs, each size-matched arm against both a pair-matched and a byte-matched control, sit above 1.0 on prose with 95% intervals excluding it. The gap closes as the vocabulary grows: at 128,000 pieces the BPE pair reads 0.983 in domain while staying above parity out of domain (1.025) and on FLORES (1.116). The ratio factorises into a character-length ratio and a tokens-per-character ratio, the second near 1 throughout: what survives matched tokenization is character-level length, which Sanskrit prose lacks over English in SLP1 (1.028) and Sanskrit verse has (0.596). The robust statement is about deployed practice: on contemporary prose and on FLORES, with the Sanskrit side in SLP1 against the deployed o200k English pivot, Sanskrit costs 1.831-2.899 English tokens per proposition under the tokenizers people actually ship. Code, the results snapshot and every table here are public.
[NLP-10] PA-CDM: Position-Aware Character Detection Matching for Evaluating Handwritten Mathematical Expression Recognition
【速读】： 该论文旨在解决手写数学表达式识别（HMER）中现有评估指标对错误位置不敏感的问题，即传统基于精确匹配率和字符串相似性度量的评估方法无法区分错误发生的位置，导致在子式位置错误或分数操作数互换等情形下评分失真。其核心解决方案是提出一种位置感知的字符检测匹配（PA-CDM）方法，通过将字符检测匹配与位置森林编码（position-forest encoding）及分歧层级加权（divergence-level weighting）相结合，实现对结构错误的空间敏感性建模。此外，研究还构建了包含1,340个受控扰动对的冻结基准测试集StructPerturb v2.0，并设计了一种结合敏感性矩阵、人工标注研究和大语言模型（LLM）校准的跨度量一致性协议。实验表明，在六名标注者参与的人工判断研究中，PA-CDM与人类判断的相关性达到最高（斯皮尔曼等级相关系数rho=0.9535，n=990），接近前沿LLM判别器的表现（rho=0.9613），且具备零边际成本、确定性和可诊断性的优势。
链接: https://arxiv.org/abs/2609.12917
作者: Shiliang Luo(East China Normal University)
机构: East China Normal University (华东师范大学)
类目: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 8 pages, 3 figures
Abstract:Handwritten mathematical expression recognition (HMER) is conventionally scored by exact-match rates and string-similarity metrics that are blind to where an error occurs: two predictions with identical token-error counts receive identical scores whether they misplace a subscript or swap the operands of a fraction. Render-based character detection matching (CDM) aligns glyphs robustly but remains position-blind—on controlled fraction-operand swaps it scores 0.8595 where position-aware scoring yields 0.6253. Tree-edit metrics exhibit a complementary blind spot: rewrites outside the parser’s normalization coverage are penalized as structural errors (0.8552 where render-based metrics score 1.0). We propose PA-CDM, a position-aware metric that couples character detection matching with position-forest encoding and divergence-level weighting; StructPerturb v2.0, a frozen benchmark of 1,340 controlled perturbation pairs across 15 type–intensity cells; and a cross-metric consistency protocol combining a sensitivity matrix, a human study, and LLM-judge calibration. In a six-annotator study, PA-CDM attains the highest correlation with human judgments among seven automatic metrics (Spearman rho=0.9535, n=990). A frontier LLM judge correlates slightly higher (rho=0.9613) but is costly, nondeterministic, and API-dependent; PA-CDM approaches it at zero marginal cost with deterministic, diagnosable behavior.
[NLP-11] LLM -Enhanced Dual-Branch Learning for Large-Scale Multi-Label Text Classification
【速读】： 该论文旨在解决大规模多标签文本分类（large-scale multi-label text classification）中，现有基于表示的方法因依赖单一编码器或在单个排序器中融合辅助特征，导致异构语言模型间互补性未被充分挖掘的问题。其核心解决方案是提出DualMLC——一种双分支框架，通过自回归解码器式语言模型与双向编码器分别处理同一文档，各自保持独立的表征路径并独立估计共享标签空间中的相关性得分；最终采用晚期logit融合策略整合两分支的得分向量，使共享证据增强相关标签判别，而各分支特有的证据则可弥补另一分支表征的不足。实验表明，该方法在三个主流基准上均达到当前最优性能，消融实验进一步验证了异构预测器集成带来的排名增强效果。
链接: https://arxiv.org/abs/2609.12915
作者: Hui Ye,Jing Zhang,Xiulong Yang,Rajshekhar Sunderraman
机构: Georgia State University (乔治亚州立大学); Amazon (亚马逊); Central China Normal University (华中师范大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Large-scale multi-label text classification assigns a small subset of relevant labels to each document from a vocabulary containing thousands or tens of thousands of candidate labels. Although pretrained language models have improved semantic text representations, most representation-based approaches center their prediction pipelines on a primary encoder or combine auxiliary features within a single ranker. The complementarity between heterogeneous language models therefore remains insufficiently explored. We propose DualMLC, a dual-branch framework that processes the same document through an autoregressive decoder-only language model and a bidirectional encoder. Each branch maintains its own representation pathway and independently estimates relevance scores over the shared label space. DualMLC combines the two score vectors through late logit fusion, allowing shared evidence to reinforce relevant labels and branch-specific evidence to compensate for limitations in the other branch’s representation. DualMLC achieves state-of-the-art results on three widely used large-scale multi-label text classification benchmarks. Ablation results further confirm that integrating the heterogeneous predictors produces stronger rankings than either branch alone. The source code is publicly available at this https URL.
[NLP-12] Parameter-Efficient Retrievers for Polish and European Languages
【速读】： 该论文旨在解决大规模密集检索系统中因依赖参数量达数十亿的语言模型而导致的内存与计算开销过大问题，尤其在大规模索引构建、频繁语料更新及低延迟服务场景下成本高昂。其核心解决方案是提出一种三阶段训练流水线，通过跨语言对齐（cross-lingual alignment）、关系知识蒸馏（relational knowledge distillation）与对比微调（contrastive fine-tuning）相结合的方式，实现高效紧凑的检索器训练。该方法不依赖原始标注的相关性标签，仅利用强嵌入模型和重排序器作为教师模型生成监督信号，从而显著降低对人工标注数据的依赖。基于此流水线，研究者构建了PolDense（支持波兰语，参数规模17M至1B）和EuroDense（435M参数，支持九种欧洲语言）两个高效检索器系列，在涵盖41项波兰语与150项多语言检索任务的广泛评估中表现出优异的性能-效率权衡：PolDense-1B在90亿参数级别模型中表现领先，而整个PolDense家族在不同规模下构成帕累托前沿；在低于10亿参数的模型中，EuroDense在平均任务与平均语言性能上均排名第一，并在七种语言中领先。所有模型均已开源。
链接: https://arxiv.org/abs/2609.12913
作者: Sławomir Dadas,Rafał Poświata,Małgorzata Grębowiec,Michał Perełkiewicz
机构: National Information Processing Institute(国家信息处理研究所)
类目: Computation and Language (cs.CL)
备注:
Abstract:Dense retrieval systems increasingly rely on multi-billion-parameter language models, whose memory and computational requirements make large-scale indexing, frequent corpus updates, and low-latency serving costly. We present a three-stage training pipeline for developing compact and efficient retrievers that remain competitive with substantially larger models. The pipeline combines cross-lingual alignment, relational knowledge distillation, and contrastive fine-tuning. It requires no original ground-truth relevance labels, relying exclusively on supervision generated by strong embedding models and rerankers utilised as teachers. Using this pipeline, we develop PolDense and EuroDense, both supporting contexts of up to 8,192 tokens. PolDense is a family of six Polish retrievers ranging from 17M to 1B parameters. EuroDense is a 435M-parameter retriever supporting nine European languages. We conduct an extensive evaluation covering 41 Polish and 150 multilingual retrieval tasks. The results demonstrate strong quality-efficiency trade-offs. PolDense-1B outperforms the evaluated retrievers with up to 9B parameters, while the PolDense family forms the Pareto frontier across model sizes. Among the evaluated models below 1B parameters, EuroDense ranks first in both task-averaged and language-averaged performance and leads in seven of nine languages. We release all models publicly.
[NLP-13] MedSNIP: Building and Benchmarking Snippet-Level Granularity for Medical Fact Verification
【速读】： 该论文旨在解决医疗事实核查中因原子级分解（atom-level decomposition）导致临床依赖关系断裂的问题，即单个医疗索赔（medical claim）的正确性往往依赖于其周围的临床上下文结构（如实验室参考范围、因果或条件关联、患者特异性信息等），而传统方法在将文本分解为细粒度单元时可能破坏这些关键依赖，使验证者面对的是临床信息不完整的断片。为此，论文提出以“片段级验证”（snippet-level verification）为核心的新范式，通过将语义相关的句子分组为片段（snippet），从而保留局部临床结构。其关键解决方案包括：构建首个由人工标注的片段级医疗事实验证基准数据集MedSNIP-Bench（涵盖276个消费者健康与临床案例，划分为2,524个带通用与患者上下文双标签及六类结构模式编码的片段），以及开发自动化片段生成管道MedSNIP。实验表明，在MedSNIP-Bench、HealthFC和MedHallu三个数据集上，片段级验证在长回答场景下显著提升假阳性率下的F1分数，尤其在因果-条件临床链结构中收益最大；同时减少验证调用次数24%-73%，且当分解成本较低时，端到端性能仍可保持，这得益于采用开源权重的分解器在无需牺牲分块保真度的前提下实现高效分解。
链接: https://arxiv.org/abs/2609.12884
作者: Hasan Iqbal,Sarfraz Ahmad,Hyunjae Kim,Sihyeon Park,Junjie Liao,Qingyu Chen,Preslav Nakov,Yuxia Wang
机构: Mohamed bin Zayed University of Artificial Intelligence(穆罕默德·本·扎耶德人工智能大学); Yale University(耶鲁大学); Korea University(高丽大学); Beijing Normal University(北京师范大学); INSAIT, Sofia University, “St. Kliment Ohridski”(INSAIT,索非亚大学,“圣克莱门特·奥赫里德斯基”)
类目: Computation and Language (cs.CL)
备注: 24 pages, 21 figures, 14 tables, Published In Proceedings of The 2026 Conference on Empirical Methods in Natural Language Processing
Abstract:A medical claim’s correctness often depends not on the claim alone, but on the clinical structure around it. A claim may require a lab reference range, a causal or conditional link, or patient-specific details to be judged correctly, and atom-level decomposition can fragment these dependencies, leaving the verifier with clinically incomplete claims. We reformulate medical fact-checking around snippet-level verification, where clause-grouped units preserve local clinical structure. We introduce MedSNIP-Bench, a human-annotated benchmark for snippet-level medical fact verification, and MedSNIP, an automatic snippet-generation pipeline. MedSNIP-Bench covers 276 consumer-health and clinical-vignette responses, segmented into 2,524 snippets with dual in-general and in-patient-context labels and six structural pattern codes. MedSNIP is evaluated against human snippet boundaries on MedSNIP-Bench and then used to generate snippet-level units for external corpora. Across MedSNIP-Bench, HealthFC, and MedHallu, snippet-level verification preserves or improves false-class F1, with gains concentrated where answers are long enough to fragment and where the verifier is strong enough to exploit the recovered structure. The largest merge-pattern gain is on causal-conditional clinical chains. It also reduces verifier calls by 24-73%, though the saving survives end-to-end only when decomposition is cheap, which an open-weight decomposer makes possible at no loss of chunking fidelity.
[NLP-14] DuplexDrama: A Synthesized Dialogue Dataset with Scenarios Full-Duplex Behaviors Expressive Speech and Sound Events
【速读】： 该论文旨在解决现有语音对话数据集在多维度真实性和复杂性上的不足，特别是缺乏对完整人物设定与场景背景、全双工行为（如打断、回应、不完整话语）、情感一致的表达性语音以及脚本感知声效的同步建模。其解决方案的关键在于提出DuplexDrama，一个首个同时涵盖上述四个维度的合成语音对话数据集，通过四阶段流水线构建，并经由剧本与合成音频的双重质量验证。该数据集包含超过2000小时的音频，覆盖13个人物角色与5个年龄组别，使用64种音色，其中3.8%的对话回合包含至少一种全双工行为。研究通过内部全双工模型训练验证了数据的有效性，并将发布包含6,400句双语对话（约800小时，中文约500小时，英文约300小时）的精选子集，以推动全双工语音对话模型的研究发展。
链接: https://arxiv.org/abs/2609.12872
作者: Qingxiang Guo,Wenke Fan,Shuofeng Zhao,Dawei Yang,Zhiyang Zhou,Yingxin Shang,Hongwei Cai,Zhou Wang,Weixu Wang,Lin Yang,Shuran Zhou,Yang Song
机构: 未知
类目: Computation and Language (cs.CL)
备注: 5 pages, 5 figures, 5 tables, 18 references. Demo: this https URL
Abstract:We present DuplexDrama, the first synthesized spoken dialogue dataset that simultaneously covers four dimensions: (i) complete persona and scenario settings; (ii) three full-duplex behaviors (interruption, backchannel, incomplete); (iii) expressive speech with persona-aligned emotion labels; and (iv) script-aware sound events. DuplexDrama is built via a 4-stage pipeline; quality validation on both scripts and synthesized audio confirms its quality. We have produced more than 2,000 hours audio data with a 64-voice timbre pool spanning 13 personas and 5 age buckets; 3.8% of all turns carry at least one full-duplex behavior. This data has been validated through internal full-duplex model training. We will release a curated subset of 6,400 bilingual dialogues (800 h, Chinese ~500 h + English ~300 h) to advance full-duplex spoken dialogue model research. Data samples are available at our demo page and LLM-judge evaluation prompts will be released with the dataset.
[NLP-15] Residual Vector-based Reconstruction as Long-Context Recall Regardless of Context Window Size
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在处理长上下文时面临的内存消耗随输入长度线性增长的问题，尤其是当上下文超出预训练阶段的固定上下文窗口时，传统方法如模型优化和有损提示压缩仍无法有效实现长上下文召回。其解决方案的关键在于利用LLM前馈层中的参数激活来重构事实信息，具体通过提取残差向量（residual vectors）——这些向量编码了源文档中关键事实的表示——以实现对查询相关事实的确定性重建。该方法无需额外训练或修改模型权重，即可在上下文长度显著增加的情况下保持近似恒定的GPU内存占用，同时维持高保真度的事实还原能力，从而在长达两百万个词元的故事上下文中成功回答单事实问题，显著超越了现有方法的性能极限。
链接: https://arxiv.org/abs/2609.12686
作者: MyungHoon Ryu,XinYu Piao,Jong-Kook Kim
机构: Korea University(韩国大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Large language models (LLMs) process long contexts, including long documents and lengthy conversations, but face token-level memory usage that increases proportionally to input length. Although model optimization and lossy prompt compression are widely used, these methods still fail to solve the long-context recall problem beyond pretrained and size-constrained context windows. This paper proposes a long-context recall method that maintains near-constant GPU memory usage as context length increases, without additional training. The main idea is to reconstruct facts using parameter activations in the LLM’s feed-forward layers, which store residual vectors representing facts from the source document. Utilizing residual vectors allows the LLM to deterministically reconstruct query relevant facts without referencing the original document, preserving high fidelity and reducing memory usage without fine-tuning weights. Experimental results show that the proposed method enables answering single-fact questions in two-million-token story contexts where previous methods fail.
[NLP-16] Doc2FRC: Length-Consistent Document-Level Machine Translation via Fixed-Range Chunking
【速读】： 该论文旨在解决长文档机器翻译（DocMT）中因直接端到端翻译导致的n-gram重复和渐进式质量退化问题。现有方法虽通过规则分块缓解此问题，但传统基于规则的分块方式在训练与推理阶段存在长度分布不匹配（train-test length mismatch）的问题，影响模型泛化能力。为此，论文提出固定范围分块（Fixed-Range Chunking, FRC），采用动态规划将文档划分为预定义长度区间的子块，确保训练与推理阶段输入长度分布一致，有效缓解了训练-测试长度偏差。在此基础上，提出一种轻量级双边界对齐算法以优化块间对齐，并设计四种不同的训练策略。实验表明，基于FRC的微调显著优于直接的Doc2Doc微调，在IWSLT2017数据集上超越现有方法；进一步构建的跨语言、去主流训练源的GlobVDoc测试集验证了FRC在分布外（out-of-distribution）文档翻译中的优越性能。其核心解决方案在于通过统一长度分布的FRC机制实现训练与推理的一致性，从而提升模型鲁棒性与翻译质量。
链接: https://arxiv.org/abs/2609.12674
作者: Xiaotian Wang,Youyuan Lin,Zhan Shen,Hitomi Yanaka
机构: The University of Tokyo(东京大学); Kyoto University(京都大学); Riken(理化学研究所); Tohoku University(东北大学)
类目: Computation and Language (cs.CL)
备注: To appear in Proceedings of the Eleventh Conference on Machine Translation (WMT2026)
Abstract:Advanced large language models (LLMs) with long context windows can substantially reduce input truncation in document-level machine translation (DocMT). However, direct Doc2Doc translation remains prone to n-gram repetition and progressive quality degradation. A common remedy is to segment the document into finer-grained chunks. Nonetheless, conventional rule-based chunking approaches fail to handle the length distribution mismatch between training and inference. To address this, we introduce Fixed-Range Chunking (FRC), utilizing dynamic programming to partition documents into chunks within a predefined length interval. By consistently applying FRC during training and inference, the input documents of any length are mapped to the same length distribution, substantially reducing train-test length mismatch. Centered on FRC, we propose a lightweight dual-boundary matching algorithm for chunk alignment, alongside four distinct training strategies. Experimental results show that FRC-based fine-tuning substantially improves 7B LLMs over direct Doc2Doc fine-tuning and outperforms existing DocMT methods on IWSLT2017. We further construct GlobVDoc, a 10-language test set independent of mainstream DocMT training sources, and show that FRC improves out-of-distribution document translation.
[NLP-17] LifeMem: Enabling Lifelong Experience Reuse for LLM Agents EMNLP2026
【速读】： 该论文旨在解决大语言模型智能体在持续学习过程中面临的两大核心挑战：一是现有基于记忆的智能体难以在不同环境间有效迁移可复用的经验，二是随着经验积累易发生灾难性遗忘。其解决方案的关键在于提出一种名为LifeMem的终身学习框架，通过在训练阶段基于潜在工作流对累积的交互轨迹进行聚类，从中提取可复用的技能；在推理阶段，智能体能够根据新任务召回相关技能与轨迹以指导行为决策。实验在10个环境、超过13,000个任务及2,000条新标注的交互轨迹上验证了该方法的有效性，结果表明LifeMem显著提升了经验复用能力，既减少了对已学任务的遗忘，又实现了优异的跨任务迁移性能。进一步分析揭示，任务流的顺序性影响学习效果，而将结构相似的轨迹在记忆中进行整合可进一步提升性能。
链接: https://arxiv.org/abs/2609.12655
作者: Yuli Qiu,Yutong Li,Wei Su,Zeming Liu,Wanxiang Che,Heyan Huang,Haifeng Wang,Yuang Guo
机构: Beijing Institute of Technology (北京理工大学); Beihang University (北京航空航天大学); Harbin Institute of Technology (哈尔滨工业大学); Baidu Inc. (百度公司)
类目: Computation and Language (cs.CL)
备注: Accepted to EMNLP 2026 Main Conference
Abstract:Large language model agents are expected to continuously adapt to new tasks and environments over their lifetime by reusing past experience. However, existing memory-based agents struggle to transfer reusable experience across environments and suffer from catastrophic forgetting as experience accumulated. To address these challenges, we propose LifeMem, a lifelong learning framework that enables agents to transfer knowledge across multiple environments. During learning, LifeMem clusters accumulated interaction trajectories based on underlying workflows to extract reusable skills. When solving a new task at inference time, the agent recalls relevant skills and trajectories to guide actions. To validate our method, we conduct experiments across 10 environments and over 13k tasks with 2k newly annotated interaction trajectories. Results show that LifeMem enables effective experience reuse in lifelong learning, achieving both reduced forgetting on learned tasks and superior cross-task transfer. Further analysis reveals that task streaming impacts learning, while consolidating structurally similar trajectories within memory boosts performance.
[NLP-18] SWARM: A Multilingual Human-Annotated Dataset for Russian Propaganda Detection in Search Engine Results WOAH2026
【速读】： 该论文旨在解决多语言、跨网络空间中俄罗斯国家宣传内容的识别难题，现有计算方法普遍局限于单一平台（如社交媒体）和少数语言，且仅依赖来源分析而忽视内容本身。其核心解决方案在于构建首个多语言、跨领域、基于内容标注的基准数据集SWARM（Search-Web documents Annotated for Russian propaganda, Multilingual），涵盖9种语言、覆盖新闻、博客、政府网站等多样化网络域的2,183条搜索引擎结果，并由训练有素的编码员标注是否支持特定俄罗斯宣传叙事。研究发现，传统的基于来源的黑名单策略会遗漏大量传播性内容，因其不仅出现在被标记的“宣传”渠道，也广泛分布于主流媒体；相比之下，内容级分析显著提升检测效果，其中最强的生成式AI（Generative AI）模型在正类F1值上达到0.73，而监督分类器表现较差（约0.5），且小型生成式模型存在过度预测问题，将主题相关性误判为立场支持。因此，该研究强调：有效识别搜索引出的宣传内容必须采用分语言、基于内容的评估范式，而SWARM数据集与配套评估代码为此提供了关键支持。
链接: https://arxiv.org/abs/2609.12653
作者: Manuel Tonneau,Abhinav Dubey,Farhan Shaikh,Ilaria Vitulano,Martha Stolze,Hale Dedeoglu,Clara Riechert,Ella Kuka,Maryna Sydorova,Mykola Makhortykh,Elizaveta Kuznetsova
机构: Weizenbaum Institute; University of Oxford; University of Bern
类目: Computation and Language (cs.CL)
备注: Accepted at the 10th Workshop on Online Abuse and Harms (WOAH 2026)
Abstract:Russian state propaganda spreads across many languages and online spaces. Yet, most computational work examines only one such space, usually social media, in one or two languages, and analyses sources rather than content. We introduce SWARM (Search-Web documents Annotated for Russian propaganda, Multilingual), a dataset of 2,183 search engine results across nine languages and diverse web domains (e.g., news, blogs, government sites), each annotated by trained coders for whether it supports a recurring Russian propaganda narrative. We benchmark a source-based blocklist, supervised classifiers, and zero-shot LLMs against these labels. The blocklist misses most propaganda-supporting documents, because such content is not confined to flagged “propaganda” outlets but also appears on mainstream ones. Content-level analysis helps, though how much depends on the model: the strongest LLM reaches a positive-class F1 of 0.73, whereas the supervised classifiers reach only about 0.5, with the smaller LLMs over-predicting support, mistaking topical relevance for endorsement. Detecting search-borne propaganda thus requires per-language, content-level evaluation, which we hope SWARM and our evaluation code enable.
[NLP-19] SteerDuplex: Steerable Duplex Speech Dialogue Models
【速读】： 该论文旨在解决全双工语音对话模型在可引导性（steerability）方面的关键缺失问题，即模型难以根据用户指令可靠地调整对话行为的多个维度，如语调、人物设定、语速及语音风格等。当前全双工模型虽支持低延迟换轮、打断处理与反馈信号，但对语用层面的可控调节能力研究不足。其解决方案的核心在于提出并实现一个名为SteerDuplex的基于Moshi架构的全双工语音模型，通过在自然对话与合成对话数据上进行微调，重点强化指令遵循、语音表达、推理能力与双工交互性能。进一步采用两阶段强化学习（RL），结合可验证的交互检查与人工评判的语义反馈构建混合奖励机制，以优化响应时机与连贯性。为系统评估该能力，研究者构建了包含390个语音提示和1,067条人工标注二元音频/文本评价标准的基准测试集SteerBench，覆盖语调、角色设定、风格/口音及速度/长度等维度。实验表明，监督训练使音频引导任务平均通过率较最强开源基线提升44.5个百分点；强化学习进一步将源干净环境下的打断响应率从72.5%提升至82.5%，并将合成停顿引发的抢话率由26.5%降至9%。尽管整体任务得分保持相当或更高水平，但奖励探针揭示了因响应不完整导致的“奖励黑客”现象，凸显了在评估响应时机时必须同步考量响应完整性的重要性。该工作为语音可引导性的系统性研究提供了模型与评测工具基础。
链接: https://arxiv.org/abs/2609.12623
作者: Utkarsh Tyagi,Ramaneswaran Selvakumar,Advait Gosai,Sonal Kumar,Nikhil Barhate,Isabell Sagar,Steven Li,Miheer Bavare,Daniel Quigley,Fabiola Tapia Carrillo,Jose M Patron E,Diego Macías Gutiérrez,Paul Song,Ramani Duraiswami,Dinesh Manocha,Yunzhong He
机构: Scale AI; University of Maryland
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 24 pages, 7 figures
Abstract:Full-duplex spoken dialogue models support low-latency turn taking, interruption handling, and backchanneling, yet a key capability remains underexplored: steerability, the ability to reliably shift conversational behavior along attributes such as tone, persona, speaking rate, and voice style in response to user instructions. We introduce a taxonomy of text- and audio-based steerability that identifies substantial gaps in current full-duplex models. To address this gap, we introduce SteerDuplex, a Moshi-based full-duplex speech model fine-tuned on natural conversations and synthetic dialogues targeting instruction following, vocal delivery, reasoning, and duplex interaction. We further apply two-stage reinforcement learning (RL) with hybrid rewards, combining verifiable interaction checks and judge-based semantic feedback to improve timing and response continuity. To evaluate full-duplex spoken steerability, we introduce SteerBench, a benchmark with 390 spoken prompts and 1,067 human-authored binary audio and text rubrics spanning tone, persona, style/accent, and speed/length. On SteerBench, supervised training improves audio-steering average pass rate by 44.5 percentage points over the strongest evaluated open baseline. On Audio MultiChallenge, task average pass rate improves by 7 points over its strongest evaluated open baseline. RL further raises source-clean interruption response from 72.5% to 82.5% and reduces synthetic pause barge-in from 26.5% to 9%. Steering and aggregate task scores remain comparable or higher, while reward probes reveal reward hacking through incomplete responses. Our model and benchmark support systematic research on spoken steerability, with reward analysis showing why timing gains must be evaluated alongside response completeness.
[NLP-20] Meddies-PII: A Multilingual Framework for Personally Identifiable Information Extraction in Clinical De-identification
【速读】： 该论文旨在解决临床脱敏中个人身份信息（PII）精准识别的难题，尤其针对现有标注数据集构建成本高、合成数据集生成过程不透明或依赖简单合成策略的问题。其核心解决方案是提出Meddies-PII-Dataset，一个包含一百万条跨十七种语言、涵盖九类PII标签的合成临床文档语料库。该语料库通过属性条件化提示（attribute-conditioned prompts）生成，并经由十三个确定性验证门控机制确保文档结构与标注的一致性。为评估数据集的有效性，研究者训练了Meddies-PII-Model这一BIOES格式的词元分类模型，在十五个外部基准测试中平均实体级精确匹配F1达到0.827，显著优于最强基线模型（0.658）。该工作关键在于构建了一个大规模、多语言、可复现且经过严格验证的合成数据生态系统，包括数据集、模型、生成框架与评估代码，以推动多语言临床脱敏研究的发展。
链接: https://arxiv.org/abs/2609.12544
作者: Linh Uyen Le,Christian Hoang,Huy Hoang Ha
机构: Meddies AI(越南)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Clinical de-identification relies on accurately identifying personally identifiable information (PII). However, manually annotated datasets are costly to construct, while existing synthetic alternatives often provide limited details about their generation process or rely on relatively simple synthesis strategies. We introduce Meddies-PII-Dataset, a corpus of one million synthetic clinical documents spanning seventeen languages and nine PII labels. The documents are generated using attribute-conditioned prompts and validated through thirteen deterministic gates that enforce structural and annotation consistency. To evaluate the dataset’s utility, we train Meddies-PII-Model, a BIOES token classifier, and compare it with existing PII extraction systems using exact-match entity-level F1. Meddies-PII-Model achieves the highest performance among the evaluated systems on all reported benchmarks, with a mean F1 of 0.827 across fifteen external benchmarks, compared with 0.658 for the strongest baseline. Upon acceptance, we will publicly release the dataset, benchmark suite, model, generation framework, and evaluation code to support research on multilingual clinical de-identification.
[NLP-21] Agent as Policy for Robotic Manipulation
【速读】： 该论文旨在解决通用智能体在物理机器人操作中缺乏端到端自主执行能力的问题，即如何使未经特定任务或环境训练的通用智能体直接驱动物理机器人完成复杂操作任务。其核心解决方案是提出“智能体即策略”（Agent as Policy, AGP），将任务规划与执行完全交由智能体自主控制。AGP通过解析视觉信息、生成可执行程序、发出运动指令，并根据物理反馈动态修正行为，实现了智能体推理与编程能力在真实世界中的持续交互。该方法在多个现实场景下验证了有效性，涵盖高精度操作、动态运动及柔体物体处理，如基于人类视频的装配、基于目标图像的积木构建、骰子翻转、精准投掷和双手折叠毛巾等任务。在三种积木构建配置中分别达到100%、100%和80%的成功率，证明了通用智能体可通过运行时推理、编程与环境互动，实现对物理操纵的自主控制，为通用智能体作为机器人策略提供了可行路径。
链接: https://arxiv.org/abs/2609.12541
作者: Mengzhao Jia,Yang Lin,Xixin Zhang,Zhihan Zhang,Xiaobai Liu,Meng Jiang
机构: University of Notre Dame(圣母大学); University of California San Diego(加州大学圣地亚哥分校); San Diego State University(圣地亚哥州立大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:We demonstrate that a general-purpose agent can directly drive a physical robot throughout task execution without any task-specific or environment-specific training. We introduce Agent as Policy (AGP), which places task planning and execution under the agent’s control. Given a task and a robot interface, the agent interprets visual evidence, writes executable programs, issues motion commands, and revises its actions in response to physical outcomes. This brings the agent’s reasoning and programming capabilities into continuous interaction with the physical world. We study AGP across multiple real-world manipulation tasks spanning precision manipulation, dynamic motions, and deformable objects. These include assembly from human videos, block construction from goal images, die reorientation, targeted throwing, and bimanual towel folding. AGP achieves success rates of 100%, 100%, and 80% on three block construction configurations. These findings establish a path for general-purpose agents to act as robotic policies, extending their autonomy to physical manipulation through runtime reasoning, programming, and interaction.
[NLP-22] Earth-Agent -Pro: Towards Real-World Full-Chain Earth Observation with Agents
【速读】： 该论文旨在解决真实世界地球观测（Earth Observation, EO）任务中，智能代理在开放世界环境下从高层次科学问题出发，自主完成观测数据获取、数据处理、领域计算及结论推导的全链路执行难题。现有方法多依赖预设观测数据或已准备好的输入/候选答案，导致对完整端到端开放世界执行能力的评估不足。其核心解决方案是提出Earth-Agent-Pro框架，采用专家编写的技能（expert-authored skills）约束规划与运行时工具使用，并引入以工作流为中心的结构化记忆机制，记录计划步骤、可接受证据及其依赖关系，实现仅修复受无效证据影响的工作流后缀，提升执行鲁棒性。该框架通过分离式大语言模型适配器，分别采用序列级监督微调（supervised fine-tuning）优化规划器的工作流组合能力，以及节点级分组相对策略优化（group relative policy optimization）结合局部可验证奖励信号，增强执行器对工具参数的精准接地能力。实验基于Earth-Bench-Pro基准，包含744个问题和248个专家定制的任务核心，覆盖可见光影像、光谱观测与遥感产品等多模态数据。结果表明，使用共享GPT-5骨干网络，Earth-Agent-Pro在LLM-as-Judge指标上达到66.13%准确率，较ReAct提升20.95个百分点，在工具按序使用率上提升24.44个百分点；联合适配器微调使Qwen3.5-9B模型的准确率从38.31%提升至50.00%，增益达11.69个百分点，验证了适配器在提升工作流生成与参数锚定方面的有效性。
链接: https://arxiv.org/abs/2609.12533
作者: Zhutao Lv,Chenhao Dang,Yi Feng,Yanpei Gong,Xiaolei Wang,Junyan Ye,Conghui He,Weijia Li
机构: Shanghai Jiao Tong University; Shanghai Artificial Intelligence Laboratory; Tianjin University; Harbin Institute of Technology; Tsinghua Shenzhen International Graduate School; Tsinghua University; Sun Yat-Sen University
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 18 pages, 8 figures. The code and datasets of this work will be released soon
Abstract:Real-world Earth observation (EO) agents must translate high-level scientific questions into executable workflows to acquire observations, prepare data, perform domain computations, and derive conclusions from runtime evidence. Existing EO agents typically start from supplied observations, while benchmarks typically provide prepared inputs or candidate answers, leaving full-chain open-world EO execution largely untested. We present Earth-Agent-Pro, an execution-adaptive Plan-and-Execute framework using expert-authored skills to constrain planning and runtime tool use. Workflow-centered structured memory records planned steps, accepted evidence, and their dependencies, enabling repair of only the affected workflow suffix when runtime evidence invalidates a step. Separate large language model adapters use sequence-level supervised fine-tuning for planner workflow composition and node-level group relative policy optimization with locally verifiable rewards for executor tool-argument grounding. Earth-Bench-Pro instantiates 248 expert-curated task cores as 744 questions under three matched regimes. Its 248 Open-World Execution questions span RGB imagery, spectral observations, and remote sensing products, pairing high-level requests with runtime data requirements, executable trajectories, and open-ended answers grounded in execution evidence. With a shared GPT-5 backbone, Earth-Agent-Pro achieves 66.13% LLM-as-Judge accuracy, exceeding ReAct by 20.95 points in this metric and 24.44 points in Tools-In-Order. Joint adapter tuning raises Qwen3.5-9B LLM-as-Judge accuracy from 38.31% to 50.00%, an 11.69-point gain over the untuned configuration. Planning-only evaluation and execution with the reference workflow show that the adapters improve workflow composition and argument grounding, respectively. Code and datasets will be released soon.
[NLP-23] Information Specialization and Constrained Synthesis in Multi-Agent LLM Forecasting: A Prospective Live-Study of the 2026 FIFA World Cup
【速读】： 该论文旨在解决多智能体系统中角色专业化对预测性能的影响及其后续信息融合是否能提升整体效用的问题。其核心挑战在于：在信息密集型场景下（如2026年国际足联世界杯），不同角色的生成式智能体能否通过分工产生差异化且互补的预测结果，以及引入批判与元代理机制是否真正提升最终预测质量。解决方案的关键在于构建一个四阶段序列化多智能体框架——由专注于结构化表现数据的量化专家、聚焦于伤病、战术及新闻发布会等非结构化信息的新闻专家分别进行独立预测，再经由独立批评者审查，并由元代理进行集成。实验结果表明，新闻专家在概率加权的前三名命中率上表现最优，且在精确比分命中率上达到与博彩市场相当的水平；然而，两名专家在56场比赛中有50场至少达成两个比分的一致性，且元代理从未生成超出专家预测集合的额外比分，说明尽管非结构化信息具有显著预测价值，但增加批判与集成阶段并未带来显著的信息互补性或性能增益。
链接: https://arxiv.org/abs/2609.12495
作者: Julian Varghese,Lucas Bickmann,Sarah Sandmann
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 10 pages, 3 figures, 1 table
Abstract:Large language models are being organized into multi-agent systems with specialized roles, but whether such specialization produces distinct forecasts and whether subsequent synthesis improves utility remains unclear. In this study, we carried out a live, prospective evaluation over the final 56 matches of the information-dense 2026 FIFA World Cup, keeping a frontier foundation model constant while assigning two primary forecasting agents contrasting specialist roles: a quantitative specialist focusing on structured performance statistics and a news specialist focusing on current injuries, tactics and information from press conferences. Their forecasts were then reviewed by a separate critic before being combined by a meta-agent, resulting in a sequential four-agent model. Forecasts from the betting market served as an external benchmark. The news specialist obtained the highest mean probability-weighted Top-3 utility and matched the betting market in Top-3 exact-score hits. Nevertheless, the two specialist forecasters agreed on at least two of the three scorelines in 50 out of 56 matches, and the meta-agent never generated more than one scoreline outside the specialists’ forecast set. These findings show that rapidly changing, unstructured information can provide a valuable forecasting signal alongside structured statistics, whereas adding critic and meta-agent stages does not necessarily create complementary information or improve on the strongest specialist.
[NLP-24] Confidence-Gated Transductive Test Generation for Code Reranking
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）生成程序后，高质量测试用例合成困难的问题，核心挑战在于可靠预期输出难以获取。为此，论文提出了一种基于置信度的归纳-演绎测试用例生成方法（Confidence-Gated Transductive Test Generation, CoTT），其关键在于通过一个高效的归纳式生成流程先行判断模型输出的置信度；仅当归纳结果置信度较低时，才触发计算成本更高的演绎式生成。这种自适应机制在保证输出可靠性的同时，仅在必要时分配额外计算资源，从而实现效率与效果之间的优化平衡。在代码重排序基准测试中，CoTT在各项指标上均优于现有基线方法，并显著降低了整体计算成本，验证了基于置信度的推理阶段计算分配策略在单一高效大模型下的有效性。
链接: https://arxiv.org/abs/2609.12489
作者: Sungjae Lee,Youngsik Yoon,Seockbean Song,Siwei Wang,Wei Chen,Jungseul Ok
机构: POSTECH(韩国浦项科技大学); Microsoft Research Asia(微软亚洲研究院)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Software Engineering (cs.SE)
备注:
Abstract:Test case synthesis is crucial for evaluating and ranking programs generated by large language models (LLMs). However, constructing high-quality test cases remains challenging because reliable expected outputs are often difficult to obtain. We propose Confidence-Gated Transductive Test Generation (CoTT), which first uses an efficient inductive procedure and invokes transductive generation only when inductive confidence is low. This adaptive design improves output reliability while allocating extra computation only when needed. On code reranking benchmarks, CoTT outperforms prior baselines across the reported metrics while reducing cost relative to applying transductive generation to every input. These results show that confidence-based allocation of test-time computation provides a favorable efficiency-effectiveness trade-off with a single efficient LLM.
[NLP-25] Zipbench: Low-Cost Framework for Compressing Comprehensive Benchmarks of Large Language Models EMNLP2026
【速读】： 该论文旨在解决大规模语言模型（LLM）评估中基准测试套件冗余导致评估成本过高的问题。现有广泛使用的基准测试存在大量重复内容，使得模型评估效率低下且资源消耗巨大。尽管近期提出的基准压缩方法（Benchmark Compression Methods, BCMs）可在一定程度上降低评估开销，但多数高效BCM依赖于来自多个大型语言模型的海量样本级评估结果集合以识别代表性样本，而构建此类数据集本身成本高昂，除非已有公开共享，否则难以应用于新发布的基准测试。为克服这一限制，本文提出ZipBench——一种简单、低成本且具备理论误差边界与排序一致性保证的新型基准压缩方法。其核心创新在于仅需对少量锚定模型（anchor LLMs）进行有限评估，通过合成伪评估结果扩展覆盖范围，学习紧凑的样本表征，并从中选取具有代表性的子集。基于此，作者构建了ZipBench Zoo，包含100余个涵盖文本、多模态及智能体任务的压缩版基准代理，其平均绝对误差仅为0.002–0.02，与完整基准间的斯皮尔曼等级相关系数高达约0.98。整体而言，ZipBench显著降低了大模型评估与压缩基准构建的计算成本，有效降低了资源受限研究者参与广泛大模型研究的门槛。
链接: https://arxiv.org/abs/2609.12475
作者: Zhongzhan Huang,Junxin Li,Guoming Ling,Yupei Lin,Shanshan Zhong,Hefeng Wu
机构: Bosch Research; Sun Yat-sen University (中山大学)
类目: Computation and Language (cs.CL)
备注: Accepted by EMNLP 2026 main track
Abstract:Comprehensive benchmark suites are essential for improving large language models (LLMs), but many widely used benchmarks are redundant, making evaluation unnecessarily expensive. Although recent benchmark compression methods (BCMs) can mitigate this cost, many strong BCMs rely on large collections of per-sample evaluation results from numerous LLMs to identify representative samples. Building such collections is also expensive unless they are already public, making these methods difficult to extend to newly released benchmarks. To address this challenge, we present ZipBench, a simple and low-cost BCM with theoretical error and rank-consistency guarantees. ZipBench evaluates only a small set of anchor LLMs, synthesizes pseudo evaluation results to broaden coverage, learns compact sample representations, and selects a small yet representative subset. Building on it, we create ZipBench Zoo, a collection of compact versions of 100+ benchmark proxies spanning text, multimodal, and agent tasks. These benchmark achieve mean absolute errors of 0.002–0.02 and average Spearman correlations of ~0.98 with the full benchmarks. Overall, ZipBench reduces the cost of both LLM evaluation and compact benchmark construction, lowering the barrier to broad LLM research for compute-constrained researchers. The code has been released in this https URL.
[NLP-26] AMDKernelVault: Large-Scale Datasets and Agent ic Training for AMD GPU Kernel Optimization
【速读】： 该论文旨在解决当前基于大语言模型（LLM）的内核生成代理普遍依赖CUDA/NVIDIA生态、缺乏对AMD CDNA GPU架构支持的问题，尤其针对现有方法在生成、反思与优化过程中频繁调用前沿大模型所带来的高延迟与低效问题。其核心解决方案是提出AMDKernelVault，一个面向AMD CDNA GPU的开放开源HIP与Triton内核语料库及训练框架，通过构建Agent驱动的自动化流水线——HIPKernelGen与TritonKernelGen，实现从PyTorch代码到HIP或Triton内核的端到端转换，结合ROCm环境下的编译验证与硬件级延迟性能剖析，确保生成内核的可执行性与高效性。该语料库包含62,153个经执行验证的HIP内核样本、2,377个基于生产环境的ROCm库问答条目以及39,893个Triton内核，为模型训练提供了高质量数据支撑。研究进一步采用监督微调（SFT）与执行感知强化学习（execution-aware reinforcement learning）对Qwen3-8B进行训练，在固定评估预算下，在PyTorch转HIP（Pass@1: 34.0%）、TritonBench-G（Corr@3: 33.2%）和ROCmBench（Corr@3: 41.94%）等任务上达到最高正确率，验证了语料库的有效性，尽管在编译效率与运行速度指标上未呈现一致优势。
链接: https://arxiv.org/abs/2609.12471
作者: Ji Liu,Saptarshi Majumder,Yiqing Huang,Wenwen Ouyang,Umang Pandey,Zeping Li,Chushi Chen,Zihao An,Puyuan Yang,Zekai Li,Sina Rafati,Ziqiong Liu,Pratik Prabhanjan Brahma,Dong Li,Zicheng Liu,Sharon Zhou,Emad Barsoum
机构: Advanced Micro Devices, Inc.(超威半导体公司)
类目: Computation and Language (cs.CL)
备注: N pages, 3 figures, including appendix. First four authors contributed equally. Code: this https URL Data: this https URL
Abstract:We introduce AMDKernelVault, an open HIP and Triton kernel corpus and training framework for recent AMD CDNA GPUs. Existing LLM-based kernel agents are largely CUDA/NVIDIA-centric and often depend on repeated frontier-LLM calls for generation, reflection, and optimization. To address this gap, we develop HIPKernelGen and TritonKernelGen, agent-driven pipelines that transform PyTorch references into HIP or Triton kernels, compile and validate candidates under ROCm, and latency-profile them on AMD hardware. The corpus contains 62,153 execution-verified HIP kernel samples, 2,377 production-grounded ROCm Libraries QA entries, and 39,893 Triton kernels. We further train Qwen3-8B with supervised fine-tuning and execution-aware reinforcement learning as a demonstration of the corpus’s utility. Under fixed evaluation budgets, it achieves the highest correctness among the compared models on PyTorch-to-HIP (34.0% Pass@1), TritonBench-G (33.2% Corr@3), and ROCmBench (41.94% Corr@3), but does not uniformly lead compilation or speed metrics. The corpus and documentation are available at this https URL, and the associated training and kernel-generation code is available at this https URL.
[NLP-27] Not All Speech Is Intent: Adaptive Self-Correcting Inference Layer for Post-ASR False Wake-Up
【速读】： 该论文旨在解决对话式人工智能中持续存在的误唤醒（false wake-up）问题，即语音内容在音素上与设备唤醒词相似时，虽生成语法正确且语义连贯的自动语音识别（ASR）转录结果，却导致助手错误执行指令。现有系统通常孤立地做出单一意图决策，缺乏从历史错误中学习或通过个性化机制适应用户的能力。其解决方案的关键在于提出一种互补的后ASR纠错框架——反馈驱动的自适应自我修正推理层（ASCIL），该框架在响应生成前通过融合声学嵌入、语言线索、设备上下文以及过往误分类模式，重新评估唤醒意图。ASCIL能够解析隐式信号（如犹豫、脱离、沉默）和显式信号（如取消、重复），将其作为自动推断的、带有噪声的行为指标，用于在线更新模型模式，而无需人工标注；同时支持在推理过程中利用先验错误进行纠正，并与自然语言执行过程并行持续更新。在包含3,667次交互的专有数据集上，经过人工标注的“有意/无意”参考标签覆盖14种声学与上下文条件，实验表明，ASCIL在会话无关子集上实现54.27%的相对错误率降低，在阈值0.90的异常标记评估片段中最高达24.39%的相对误差减少，同时保持较高的有意唤醒接受率，中位额外延迟低于60毫秒。
链接: https://arxiv.org/abs/2609.12469
作者: Preeti Saraswat,Divya Neelagiri,Anil Yadav
机构: Samsung Research America (三星研究院美国), Mountain View, CA, USA
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:False wake-up activations remain a persistent challenge in conversational AI. Speech phonetically similar to a device’s wake word can produce a syntactically valid and semantically coherent ASR transcript that the assistant incorrectly executes. Most existing systems make a single intent decision in isolation, without a mechanism to learn from recurring errors over time or adapt to individual users through personalized learning. We introduce the Feedback-Driven Adaptive Self-Correcting Inference Layer (ASCIL), a complementary post-ASR correction framework that re-evaluates wake-up intent before response generation by fusing acoustic embeddings, linguistic cues, device context, and patterns from past misclassifications. ASCIL interprets implicit signals, including hesitation, disengagement, and silence, and explicit signals, including cancellation and repetition, as automatically inferred, noisy behavioral indicators of potential misclassification. These signals drive online pattern updates without manual annotation, whereas the intentional/unintentional reference labels used for offline evaluation are human-annotated. It generalizes from prior errors, applies corrective adjustments at inference time, and continuously updates in parallel with natural-language execution. Evaluated on a proprietary dataset of 3,667 interactions with human-annotated intentional/unintentional reference labels spanning 14 acoustic and contextual conditions, ASCIL achieves 54.27% relative error reduction on a session-disjoint subset constructed from baseline failures, and up to 24.39% relative error reduction at threshold 0.90 on the issue-tagged evaluation slice. These gains are achieved while improving intentional acceptance rates, with a median added latency below 60 ms in the reported benchmark.
[NLP-28] GraphProfiler: Source-Linked Sensitive Attribute Inference via Personal Knowledge Graphs EMNLP2026
【速读】： 该论文旨在解决生成式 AI（Generative AI）在用户生成内容中进行大规模敏感属性推断所带来的隐私威胁问题，尤其关注现有基于大语言模型（LLM）的属性推断工具缺乏可解释性，无法明确指出哪些具体帖子、概念或语义关系促成了属性泄露，从而阻碍了精准的隐私保护策略实施。其解决方案的关键在于提出GraphProfiler，一种可审计的基于图结构的LLM推理框架：将用户的发帖历史建模为带有源链接的个人知识图谱（personal knowledge graph），其中节点与边均追溯至原始发帖，并将属性预测结果关联到具体的图谱记录与原文依据。该方法不仅实现了接近顶尖文本基线的推断准确率（如在SynthPAI基准上达86.7%，在PANDORA上达84.6%），更能在超过98%的预测中提供可溯源的支持证据。通过受控消融实验验证，被引用的特定帖子对攻击成功率具有显著贡献，移除这些关键帖子导致性能下降幅度远超随机移除等量帖子，证明了其可解释性设计在实现靶向隐私缓解（如仅重写或删除泄漏信息的少数帖子）中的有效性。
链接: https://arxiv.org/abs/2609.12448
作者: Ahmed Sohair Khan,Estrid He,Chenglong Ma,Monica Wachowicz,Elham Naghizade
机构: RMIT University (皇家墨尔本理工大学)
类目: Computation and Language (cs.CL); Cryptography and Security (cs.CR)
备注: Accepted at EMNLP 2026 (Main Conference)
Abstract:Sensitive attributes such as age, income, and occupation can be inferred from user-generated content by aggregating indirect cues across many ordinary posts. LLM-based profilers can perform this aggregation automatically and with high accuracy, which makes large-scale personal attribute inference a major privacy threat. Existing LLM-based profilers, however, offer limited insight into which specific posts, concepts, and relationships made an inference possible, which is key to targeted privacy mitigation, i.e., redacting or rewriting only the few posts that actually leak an attribute, rather than perturbing entire histories. We introduce GraphProfiler, an auditable LLM-based profiler that represents each user’s post history as a source-linked personal knowledge graph where nodes and edges trace back to the originating post and resolves attribute predictions to cited graph records and source texts. GraphProfiler reaches 86.7% attack success rate on the eight-attribute SynthPAI benchmark, within two points of strong text-only baselines, and 84.6% on PANDORA, while citing supporting evidence for over 98% of predictions. Our controlled ablation experiments provide evidence that the cited posts contribute to attack success, as removing them reduces the attack success rate substantially more than removing an equal number of random posts.
[NLP-29] Do LLM s Trust the Accuser or the Accusation? Measuring Belief Shifts in Werewolf EMNLP2026
【速读】： 该论文旨在解决当前大型语言模型（LLM）在社交推理类游戏（如狼人杀）中评估过于依赖最终游戏结果，而忽视其在动态交流过程中信念更新能力的问题。其核心解决方案是提出一种基于信念转移（belief-shift）的评估基准，通过分析观察方模型在接收到怀疑与指控消息后信念的变化，量化模型在战略沟通中整合信息与发言者可信度的能力。研究利用40个开源权重的LLM配置，在1,224条标注的指控消息上进行实验，发现尽管参数规模较大的模型能更准确地根据游戏历史区分狼人与村民，但其信念仍易受指控内容影响，尤其当指控者为可信角色时，模型会过度信任指控者并错误地增加对被指控者的怀疑，即使该指控者本身为狼人阵营。此外，大模型对已不信任的指控者更具抵抗力。总体表明，当前最大达1200亿参数的开源LLM在整合指控内容与其来源可信度方面仍存在显著局限。该研究构建的评估基准与代码已公开。
链接: https://arxiv.org/abs/2609.12446
作者: Yu-Yu Yang,Ti-Rong Wu,Hung Guei,Hsing-Yu Chen,I-Chen Wu
机构: National Yang Ming Chiao Tung University (国立阳明交通大学); Academia Sinica (中央研究院)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: Accepted by the 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP 2026 Main Conference)
Abstract:Social-deduction games such as Werewolf are increasingly used to evaluate LLM agents, but existing evaluations often rely on final game outcomes. We propose a belief-shift evaluation benchmark in Werewolf for analyzing communication skills through belief updating. Using LLM-played games, we annotate suspicion and accusation messages and measure how an observing village-side model’s beliefs change after each message. We evaluate 40 open-weight LLM configurations on 1,224 annotated messages. Our results show that larger models better distinguish true wolves from villagers based on game history, but accusations still strongly influence their beliefs. Models become more suspicious of the accused target and less suspicious of the accuser, especially when the accuser is trusted, even if the accuser is wolf-aligned. Larger models better resist accusations from accusers they already distrust. Overall, our findings suggest that current open-weight LLMs up to 120B parameters still struggle to integrate accusation content with source trust in strategic communication. Our benchmark and code are available at this https URL.
[NLP-30] Beyond ID Embeddings: Process-Grounded Language Modeling for Cognitive Diagnosis EMNLP2026
【速读】： 该论文旨在解决传统认知诊断模型（Cognitive Diagnosis Models, CDMs）在处理新练习或新概念时存在的语义局限性问题，其根本原因在于传统CDMs依赖离散的、基于ID的嵌入表示学生、习题和知识点，这种表示方式与学习者认知中知识以非孤立、关联性结构存储和提取的本质相悖。为克服这一缺陷，本文提出一种面向过程的语言型认知诊断框架（Process-aware Language Cognitive Diagnosis, PLCD），其核心创新在于将语言模型生成的结构化知识作为认知先验，并结合学习者的作答记录动态校准学生的后验认知状态。关键解决方案包括：利用大语言模型（LLMs）构建概念图谱与认知过程图，通过目标条件化的语义记忆机制检索与目标习题相关的历史作答；采用基于过程的“语言到认知映射器”（Language-to-Cognition Mapper），融合DA-MoE专家网络与过程级对比学习，将文本证据映射至统一的认知空间。实验表明，PLCD不仅在预测学生表现上优于传统基线模型，且展现出优异的认知迁移能力，验证了以响应记录校准的结构化语言先验能够有效提升冷启动场景下的鲁棒性与认知建模的合理性。
链接: https://arxiv.org/abs/2609.12403
作者: Minghang Liu,Yuanzhuo Wang,Qiang Qiu,Huawei Shen,Xueqi Cheng
机构: State Key Laboratory of AI Safety, Institute of Computing Technology, CAS(中国科学院计算技术研究所人工智能安全国家重点实验室); University of Chinese Academy of Sciences(中国科学院大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: Accepted to EMNLP 2026. 20 pages, including references and appendices
Abstract:Cognitive Diagnosis Models (CDMs) play a pivotal role in personalized online learning. Traditional CDMs rely on discrete, ID-based embeddings to represent students, exercises, and concepts. This paradigm diverges from the nature of learner cognition, where knowledge is not stored and retrieved as isolated symbols. As a result, CDMs suffer from semantic limitations when new exercises or concepts appear. In this paper, we propose a Process-aware Language Cognitive Diagnosis (PLCD) framework that uses language-derived structures as cognitive priors and response records to calibrate student posterior states. PLCD leverages large language models (LLMs) to construct concept schemas and cognitive process graphs, and uses target-conditioned semantic memory to retrieve historical responses that are relevant to each target exercise. A process-grounded Language-to-Cognition Mapper with DA-MoE experts and process-level contrastive learning then maps the textual evidence into a unified cognitive space. Experimental results show that PLCD not only outperforms traditional baselines in predicting student performance but also exhibits strong cognitive transfer capabilities. These results connect the computational power of LLMs with the psychometric goal of measuring latent knowledge states, suggesting that structured language priors calibrated by response records can improve cold-start robustness and cognitive grounding.
[NLP-31] Representation-based Masked Diffusion Model
【速读】： 该论文旨在解决现有掩码扩散模型（Masked Diffusion Models, MDMs）在并行文本生成过程中，因独立更新多个掩码词元而忽略词元间复杂相互依赖关系所导致的生成结果不连贯问题。其核心解决方案是提出一种基于表示的掩码扩散模型（Representation-based Masked Diffusion Model, RMDM），关键在于利用预训练编码器将文本映射至连续语义空间，并学习一个可逆变换以将表示分布归一化为高斯先验，从而实现高效采样；在此基础上，通过条件化于该潜在语义表示来训练掩码扩散模型，使语义表示作为全局语义引导，协调词元的并行更新过程，更准确地逼近目标分布。实验证明，该方法在极端少步数采样场景下显著提升了生成质量。
链接: https://arxiv.org/abs/2609.12382
作者: Yangrong Hu,Ding Huang,Xueyu Zhou,Jian Huang
机构: The Hong Kong Polytechnic University (香港理工大学)
类目: Computation and Language (cs.CL); Machine Learning (stat.ML)
备注: 22 pages, 2 figures, and 10 tables
Abstract:Masked Diffusion Models (MDMs) have emerged as a compelling paradigm for language modeling, offering the capability for efficient parallel text generation. However, existing parallel sampling methods typically update multiple masked tokens independently and ignore the complex mutual dependencies among the masked tokens. This independent updating mechanism lacks global coordination and might lead to incoherent outputs. To address this limitation, we propose Representation-based Masked Diffusion Model (RMDM), a framework that leverages the text representation to explicitly encode global semantics and help to parallel update tokens more precisely. Specifically, we first encode text into a continuous semantic space using a pretrained encoder and learn an invertible transformation that normalizes the representation distribution to a Gaussian prior, facilitating efficient sampling during generation. Conditioned on this latent semantic representation, we train a masked diffusion model to learn the conditional text distribution, where the representation serves as global semantic guidance to coordinate parallel token updates and faithfully approximate the target distribution. Empirical results demonstrate that RMDM significantly improves generation quality, particularly in aggressive few-step sampling regimes.
[NLP-32] ORQA: An Occupation-Realistic Question and Answer Framework for LLM Professional Knowledge
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在职业层级知识评估中缺乏可靠、可扩展且具备现实意义测试基准的问题。现有方法或依赖抽象的任务定义进行职业映射，或依赖难以大规模获取的专家知识，均存在可扩展性差与成本高的局限。本文提出ORQA（Occupation-level Reasoning Question Answering），其关键在于将O*NET职业分类体系与可信的职业特定信息源（如监管机构、认证组织、专业协会及政府出版物网站）相连接，并通过自动化流程结合人工审核，生成具有来源可追溯性的问答对。该方法构建了覆盖21个SOC主要职业类别中116个职业的高质量问答数据集，共480个问题源自187个不同网站，每个问题均针对实际职业所需的技能展开。实验结果表明，当前前沿模型（如Claude Opus 4.6、GPT-5.4和Claude Sonnet 4.6）在该基准上的表现约为58%-62%，而小型开源模型性能则为33%-41%；性能在不同职业间差异显著，医疗类职业表现最佳（78%），而事务支持类职业约40%，部分具体职业（如金属板工、鱼类与野生动物巡护员）表现接近零。此外，开放式问题设计及按工资支出加权对模型排名无显著影响。研究认为，利用现有可信的职业领域信息构建测评基准，是一种未来实现职业层级人工智能能力评估的可扩展且有效的方法。
链接: https://arxiv.org/abs/2609.12366
作者: Shreyas Krishnan,Serina Chang,Abhishek Nagaraj
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 45 pages, 17 figures, 6 tables. Data, code, and an interactive dashboard at this http URL
Abstract:We present ORQA, a method for testing occupation-level knowledge in large language models. Prior methods either map abstract LLM skills to occupations via task definitions or utilize expert knowledge which is difficult to obtain at scale and expensive. ORQA complements both of these methods by connecting O*NET occupations to trusted occupation-specific websites (such as regulatory agencies, licensing bodies, professional organizations, and government publications) and converting these into source-traceable question-answer pairs. A combination of an automated pipeline and human review produces a set of high quality questions about occupations. The question set created via our method covers 116 occupations from all 21 major groups in the SOC, with 480 questions sourced from 187 different websites. Each question is designed to probe a real-world skill question that is relevant to the occupation in question. We test 15 state-of-the-art frontier and open-weight models via this method. Claude Opus 4.6, GPT-5.4 and Claude Sonnet 4.6 all perform the best at approximately 58-62% while smaller open-weight models achieve approximately 33-41% performance. Performance varies significantly across occupations. Healthcare-related occupations achieve the highest performance (78%) while Office and Administrative Support achieve approximately 40%. Performance on individual occupations (e.g. Sheet Metal Workers and Fish and Game Wardens) is essentially zero. We also find that open-ended questions and weighting by wage bill do not significantly affect the ranking of models on this benchmark. We believe that leveraging existing trusted occupation-specific information to test LLM knowledge in professional domains may be a scalable and useful method for evaluating occupation-level AI performance in the future. Results and data are available at this http URL.
[NLP-33] CueMem: Cue-Guided Context Reconstruction for Long-Term Conversational Memory
【速读】： 该论文旨在解决长期对话代理在回答用户问题时，如何高效且准确地从长时对话历史中召回相关信息的问题。传统方法直接使用完整的对话历史会导致计算开销大且可靠性不足，而压缩记忆单元则可能丢失问答所需的细粒度证据。为此，论文提出了一种基于重构性自传记忆观的框架——CueMem，其核心创新在于将提取的记忆记录视为检索线索（retrieval cues）而非自包含证据，并通过源对话轮次重建与查询相关的上下文。在记忆构建阶段，CueMem从对话轮次中提取细粒度记忆线索，并将其关联至对应的源轮次；在查询阶段，系统首先检索与查询相关的关键线索，映射回其对应的源轮次锚点，并基于捕捉时间邻近性和语义相关性的轮次图进行扩展，从而从原始对话中重构出紧凑且相关的证据上下文，供大语言模型（LLM）生成答案。实验在LoCoMo和LongMemEval数据集上验证了CueMem优于主流长期记忆基线，且相比全历史输入设置显著减少了查询阶段的输入令牌数与延迟。结果表明，以检索线索为基础的上下文重构机制是实现高效、精准的长期对话问答的有效路径。
链接: https://arxiv.org/abs/2609.12354
作者: Changjian Wang,Rongzhen Li,Weili Guan,Shuming Shi,Quan Lu,Ning Jiang
机构: Mashang Consumer Finance Co., Ltd.; Harbin Institute of Technology, Shenzhen
类目: Computation and Language (cs.CL)
备注:
Abstract:Long-term conversational agents must answer user queries by recalling information from extended dialogue histories, yet directly using the full history is costly and often unreliable, while compressed memory units may lose fine-grained evidence needed for question answering. Motivated by the reconstructive view of autobiographical memory, we propose CueMem, a cue-guided framework that treats extracted memory records as retrieval cues rather than self-contained evidence and reconstructs query-relevant dialogue context from their source turns. During memory construction, CueMem extracts fine-grained memory cues from dialogue turns and links each cue to its source turn. At query time, it retrieves query-relevant cues, maps them to source-turn anchors, and expands from these anchors over a turn graph that captures temporal proximity and semantic relatedness, reconstructing a compact evidence context from the original dialogue for LLM answer generation. Experiments on LoCoMo and LongMemEval show that CueMem consistently outperforms representative long-term memory baselines. Further analyses show that graph-based context reconstruction helps recover supporting dialogue evidence while reducing query-time input tokens and latency compared with the full-history LLM setting. These results highlight retrieval cues as an effective alternative to self-contained memory evidence for long-term conversational question answering.
[NLP-34] SynthSentry: Detecting Synthetic Data Contamination in Language Model Training Data
【速读】： 该论文旨在解决在训练大型语言模型前，对来源未知的语料库进行合成数据污染筛查这一关键问题，其核心挑战在于如何在不依赖生成模型、无生成历史记录、无需合成标签的情况下实现高效、可靠的污染检测。解决方案的关键是提出一种名为SynthSentry的语料级、模型无关的污染信号评分机制，该机制基于三个统计量的分布差异：词汇多样性衰减（lexical diversity collapse）、n-gram尾部截断（n-gram tail truncation）以及多参考模型间的困惑度方差（perplexity variance across reference models）。该方法通过在留一生成器排除（leave-one-generator-out）协议下评估受小规模开源生成器及指令微调模型污染的语料库，验证了其在不同领域（如法律、临床文本和源代码）中对自然重复性人类文本的低误报率，并实现了对污染严重程度的有效排序。经域内校准后，结合协方差收缩与自助法阈值替代原始分位数方法，可将误报率控制在名义预算附近，避免过度误判；下游微调实验表明，在当前规模下污染未导致准确率下降，但是否可通过修剪恢复性能仍待验证，且过度修剪存在显著过筛风险。研究将污染筛查定位为数据治理中的前置防御机制而非事后诊断，并公开了评分工具包。所有结果均基于小规模英文批量语料筛选场景，所考察的污染源为单轮生成或人工撰写，而非递归生成，因此结论适用于广义合成数据污染情境，而不涉及递归深度的影响。
链接: https://arxiv.org/abs/2609.12353
作者: Praveen Kumar Myakala,Ravichandra Namburi,Sowmya Keragodu Jayaramu,Sooraj George Thomas
机构: 未知
类目: Computation and Language (cs.CL)
备注: 11 pages, 3 Figures, 2026 IEEE Asia Conference on Innovation in Emerging Technology
Abstract:Large language models trained recursively on their own or other models’ outputs undergo model collapse, in which distributional tails and factual accuracy deteriorate while fluency survives. Prior work diagnoses collapse after training; the actionable problem is screening a corpus of unknown provenance before training. We introduce SynthSentry, a corpus-level, model-agnostic contamination signal requiring no access to the generating model, no generation history, and no synthetic labels. The score is a distributional divergence over three statistics: lexical diversity collapse, n-gram tail truncation, and perplexity variance across reference models. We evaluate on corpora contaminated by small open-weight generators and an instruction-tuned open-weight model under a leave-one-generator-out protocol. A domain-stratified study measures false positives on naturally repetitive human text (legal, clinical, source code). The score ranks corpora by severity with little loss when whole generator families are held out. Per-domain calibration holds near its nominal false-positive budget once covariance shrinkage and a bootstrap threshold replace a naive quantile, which runs four times over budget. A downstream fine-tuning check showed no contamination-driven accuracy deficit at our scale, so whether pruning recovers one remains open; the same run shows over-pruning risk once pruning exceeds the true contamination fraction. We frame screening as a data-curation defense rather than a post-hoc diagnosis and release the scoring toolkit. All results are small-scale; scope is English-language, batch-mode corpus screening. Contamination sources are single-generation or hand-authored rather than recursively generated, so results speak to synthetic contamination generally and not to recursion depth.
[NLP-35] I Am No One: Style-Aware Paraphrasing for Text Anonymization INTERSPEECH2026
【速读】： 该论文旨在解决文本匿名化后仍存在作者身份再识别（authorship attribution）的隐私风险问题，尤其针对通过自动语音识别（ASR）生成的会议或客服对话文本，其风格特征（stylometric leakage）在声学层面匿名化后仍可能泄露作者身份。传统基于差分隐私（Differential Privacy, DP）的匿名化方法虽能提供理论保障，但常导致文本质量与语义可用性显著下降。本文提出一种风格感知、提示驱动的匿名化方法，利用预训练大语言模型（LLM）从少量样本中构建紧凑的风格表征，并通过可控重写机制抑制可识别的风格标记，同时保持内容语义完整性。其核心创新在于将风格建模与语义保持相结合，在最小化风格泄漏的同时实现高保真文本重构，在博客和评论数据集上使作者归属的F1值降低60%-70%，显著优于基于DP及非DP的基线方法。
链接: https://arxiv.org/abs/2609.12341
作者: Ahmed Sohair Khan,Estrid He,Monica Wachowicz,Elham Naghizade
机构: 未知
类目: Computation and Language (cs.CL); Cryptography and Security (cs.CR)
备注: Accepted at Interspeech 2026
Abstract:Authorship attribution models can re-identify users from seemingly anonymized text by exploiting stable stylistic fingerprints, even after explicit identifiers are removed, posing a growing privacy risk for text publishing and analytics. This risk extends to speech-derived text such as ASR transcripts of meetings and call-center conversations, where stylometric leakage can persist even after acoustic anonymization. Differential privacy-based anonymization often severely degrades text quality and utility. We propose a style-aware, prompt-driven anonymization approach that uses pretrained large language models to construct compact stylistic profiles from minimal samples and rewrite text to suppress identifiable style markers while preserving meaning. Across blog and review datasets, our approach reduces authorship attribution F1 by 60-70% while maintaining content quality and readability, substantially outperforming DP-based and non-DP baselines.
[NLP-36] ESTS at WMT26: Routing-Informed Expert Pruning for Model Compression
【速读】： 该论文旨在解决大规模多语言生成式模型在资源受限场景下的高效部署问题，具体针对英语→简体中文及英语→埃及阿拉伯语的机器翻译任务，在无约束条件下实现模型压缩与性能保持之间的平衡。其核心解决方案的关键在于：首先基于GPT-OSS-20B模型，通过任务特定路由权重（task-specific routing mass）对专家（expert）进行重要性排序，并利用跨语言路由差异（cross-lingual routing divergence）动态分配各层保留容量，进而物理移除低重要性专家以实现结构化剪枝；随后，保留的专家模块在GPT-5.1生成的合成翻译数据上进行恢复调优（recovery-tuning），并采用MXFP4量化技术对保留专家的投影权重进一步压缩；同时构建了面向指令条件输入的鲁棒推理系统，包含类别推断、输出验证、重试机制、分段回退和源端JSON重构等模块，以增强实际应用中的稳定性与可靠性。实验结果显示，六组提交方案的参数量介于4.186B至7.770B之间，打包后文件大小为4.55~6.33 GiB，内部xCOMET-XL评估表明不同压缩操作点间具备良好的性能可比性。
链接: https://arxiv.org/abs/2609.12310
作者: Liu O. Martin,Lucas Bandarkar,Nanyun Peng
机构: University of California, Los Angeles(加州大学洛杉矶分校)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: To appear in the Proceedings of the Eleventh Conference on Machine Translation (WMT 2026)
Abstract:We describe six submissions under the team name ESTS to the unconstrained WMT26 Model Compression Shared Task for English–Simplified Chinese and English–Egyptian Arabic. We submit three compression operating points per translation direction, all derived from GPT-OSS-20B. We use task-specific routing mass to rank experts and cross-lingual routing divergence to allocate retained capacity across layers, then physically remove low-importance experts. The resulting specialists are recovery-tuned on GPT-5.1-generated synthetic translation data and further compressed by applying MXFP4 quantization to the retained expert projection weights. We additionally implement a robust inference system for the instruction-conditioned WMT26 setting, including category inference, output validation, retries, segmented fallback, and source-owned JSON reconstruction. Across our six submissions, parameter counts range from 4.186B to 7.770B and packed artifact sizes from 4.55 to 6.33~GiB. Internal xCOMET-XL evaluation using GPT-5.1 pseudo-references provides an internal comparison across the submitted compression operating points.
[NLP-37] Breaking the Token Ceiling: Distilling Smaller Stronger Byte Models
【速读】： 该论文旨在解决小模型在知识蒸馏过程中，基于字节（Byte）与基于词元（Token）的模型在计算量和数据规模增加时，其性能扩展规律是否一致的问题。现有研究多聚焦于词元级模型的缩放行为，但对字节级模型的系统性分析仍不充分。为实现二者在相同条件下的可比性，论文提出两种高效转换词元逻辑（Token Logits）至字节逻辑（Byte Logits）的方法：近似方法“Marginalize-It”与精确方法“End-Of-Token”。在此基础上，首次开展了大规模实验，同时系统地考察了模型架构（词元、字节、带结束符的字节）、训练目标（蒸馏与交叉熵）及参数量匹配（约10亿参数）下，从10亿到1万亿字节数据范围内的过拟合现象与性能缩放规律。结果表明，在低算力条件下，词元模型表现更优，但随计算资源增加迅速达到性能瓶颈；而字节模型初始表现较差，但在高算力下持续提升，最终超越词元模型，并展现出更高的性能上限。基于验证集比特每字符（BPB）的缩放律外推显示，蒸馏后的End-Of-Token-1B模型在极限状态下相比蒸馏词元模型最多可提升4%。此外，字节模型具有显著更高的数据效率，仅需六分之一的数据即可达到同等性能。由于其256个字节的极小词汇表，无需进行顶部k截断（top-k truncation），并使逻辑值存储成本降低至约五分之一。最终，下游任务的缩放律预测表明，该模型在平均任务性能上将分别超越Llama 3.2-1B、Gemma-3-1B-pt和Gemma 2B达6.5%、8.1%和2.1%。因此，解决方案的关键在于通过精确的字节逻辑转换机制与大规模实证研究，揭示字节模型在高算力场景下的优越缩放潜力及其在数据效率与存储开销方面的优势。
链接: https://arxiv.org/abs/2609.12303
作者: Kalyani Marathe,Artidoro Pagnoni,Tomasz Limisiewicz,Margaret Li,Mike Lewis,Luke Zettlemoyer,Srinivasan Iyer
机构: University of Washington (华盛顿大学); Meta FAIR (Meta人工智能研究院)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Small models are made more capable through distillation from a larger one that shares their tokenization scheme. However, do distilled byte and token models behave similarly in terms of scaling trends as compute and data increases? To enable this comparison, we introduce two variants to efficiently convert token logits to Byte Logits: 1) approximate: Marginalize-It, and 2) exact: End-Of-Token. We then present the first large scale study of overtraining decoder-only dense transformer models varying two dimensions simultaneously: the tokenization scheme (Tokens, Bytes, Bytes w/ eot) and the training objective (Distillation vs. Cross-Entropy), sweeping layer-parameter-matched models with roughly 1 billion parameters up to 1 trillion bytes of data. Across eight benchmarks spanning three categories: Multiple Choice QA, Language Generation, and Machine Translation, we find that Token-1B models outperform byte models (End-Of-Token-1B and Bytes-1B) in the low-FLOP regime but eventually plateau; byte models start worse yet surpass Token-1B models with more compute, reaching a higher downstream task performance ceiling. Extrapolating the average top-1 error vs. validation BPB scaling laws predicts that, asymptotically, distilled End-Of-Token-1B outperforms distilled Token-1B by up to 4%. They are also far more data efficient, matching the performance of distilled Token-1B using only one-sixth of the training data. Moreover, by operating over a small vocabulary of 256 bytes instead of on the order of 100K tokens, they circumvent the need for top-k truncation during logit dumping, while also reducing logit storage costs to roughly one-fifth. Finally, our downstream performance scaling laws predict that our distilled End-Of-Token-1B models asymptotically surpass the Llama 3.2-1B, Gemma-3-1B-pt, and Gemma 2B models on averaged downstream tasks by up to 6.5%, 8.1%, and 2.1%, respectively.
[NLP-38] HypoKG: Evidence-Disciplined Biomedical Hypothesis Generation Beyond Endpoint Knowledge
【速读】： 该论文旨在解决生成式人工智能（Generative AI）在生物医学领域中生成假设时，其推理过程是否真正基于科学证据，还是仅生成看似合理但缺乏实证支持的观点这一关键问题。研究通过整合京都基因与基因组百科全书（KEGG）、Rhea及UniProt三大生物数据库，构建了一个统一的生化知识图谱，并设计了一个包含550条酶源至罕见病终点路径的基准测试集，共生成13,200个由六种大语言模型（LLM）在四种不同信息输入条件下（仅提供酶源、仅提供疾病终点、酶源与疾病终点联合、完整生化路径）生成的假设。研究采用专家制定的五维评分标准对假设进行评估。结果表明，尽管仅提供酶源与疾病终点的模型常生成得分较高的假设，但其推理缺乏证据支撑；而接收完整生化路径的模型所生成的假设更符合已知的机制关联，体现出“证据约束性推理”（evidence-disciplined reasoning）。进一步通过打乱中间路径节点但保持首尾节点不变的实验验证，发现证据锚定程度显著下降（Δ = -0.793, p < 0.001），证实模型确实在推理过程中依赖路径结构。因此，本研究的关键解决方案在于：利用结构化的生化知识图谱不仅识别文献中未报道的生物终点关联，更重要的是通过其内在的机制路径引导大语言模型进行更具生物学可信度的推理。
链接: https://arxiv.org/abs/2609.12260
作者: Dominic Okonkwo,Adetayo Okunoye,Ismailcem Budak Arpinar
机构: School of Computing; University of Georgia(佐治亚大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Quantitative Methods (q-bio.QM)
备注:
Abstract:Large language models (LLMs) can generate biomedical hypotheses, but it remains unclear whether they truly reason from scientific evidence or simply produce convincing-sounding ideas. To study this, we combine three major biological databases: the Kyoto Encyclopedia of Genes and Genomes (KEGG), Rhea, and UniProt, into a unified biochemical knowledge graph and construct a benchmark of 550 paths connecting enzyme sources to rare disease endpoints, yielding 13,200 hypotheses from six LLMs under four conditions varying the biological information each model receives: source enzyme only, full biological path, or source and disease endpoint only. Hypotheses are scored using an expert-derived five-criterion rubric on a 1-5 scale per criterion. We find that models given both the source and disease endpoint often produce the highest-scoring hypotheses, showing that LLMs can generate compelling ideas from minimal information. However, these hypotheses are less grounded in the evidence. In contrast, models given the full biological path generate hypotheses more consistent with known mechanistic relationships. We call this evidence-disciplined reasoning. To confirm this effect, we shuffled intermediate path steps while keeping endpoints fixed. Evidence grounding dropped significantly (delta = -0.793, p 0.001), confirming models genuinely used path structure during reasoning. Our findings show that knowledge graphs support hypothesis generation in two ways: they identify biological endpoint pairs absent from the literature, and their mechanistic paths guide how LLMs reason between them.
[NLP-39] Automated Detection and Structuring of Social Tipping Point Evidence in Climate related Documents: A Modular AI Framework
【速读】： 该论文旨在解决气候文献爆炸式增长背景下，对“环境社会临界点”（environmental social tipping point）这一关键概念的证据难以系统发现与组织的问题。由于相关证据通常分散于长篇文档中的少量段落内，传统文本挖掘工具因仅能按主题分类或提取孤立断言，无法有效识别和结构化此类细粒度、高价值的信息，导致重要证据被遗漏。其解决方案的关键在于提出一个基于Transformer的开源、模块化框架，实现跨段落级别的证据检测与结构化处理：通过DistilBERT边界分割器进行文本分段，利用迭代增强的RoBERTa分类器精准识别临界点证据，借助Mistral 7B模型提升文本可读性，采用LLaMA 3.2 3B模型依据五项已发表标准量化评估证据质量，并结合Milvus向量数据库支持语义检索。整个系统集成于Streamlit前端并依托MinIO对象存储部署，经GPT-4.1标注的163段基准测试集及专家评审的51段数据集验证，表现出优异性能，尤其在综合九项指标得分（6.137）上超越三种对比方法，且优化后的RoBERTa模型在完整基准上达到71.4%准确率（Cohen’s kappa=0.337），在有标签段落上更达87.5%准确率（kappa=0.742），显著优于专用气候模型与未调优的语言模型。
链接: https://arxiv.org/abs/2609.12254
作者: Kavindu Perera,Mohammad Abaeiani,Ekaterina Gilman,Lauri Loven,Mourad Oussalah,Tassos Kanellos,Beatrice Gobbo,Dante Adami,Nicolò Ferriani,Maximiliano Romero,Pierre Rossel,Marc Bonazountas,Christina Deligianni,Nikos Xyderis,Artur Bogucki,Lampros Argyriou,Prasasthy Balasubramanian
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:The climate literature has grown faster than review teams can read it. That gap matters most for a concept like the environmental social tipping point, the threshold at which a small change triggers rapid, self-reinforcing change in a social system. Evidence of this kind of shift is usually contained in one or two paragraphs within a longer document. As a result, existing text mining tools-which categorize entire documents by topic or highlight isolated claims-leave an expanding set of important evidence without any systematic method for discovery or organization. This paper presents an open and modular transformer-based framework that detects and structures social tipping point evidence at the passage level. The framework joins five components into a single deployable workflow: a DistilBERT boundary splitter for segmentation, an iteratively augmented RoBERTa classifier for detection, a Mistral 7B model that rewrites each detected passage for clarity, a LLaMA 3.2 3B model that rates the passage against five published social tipping point criteria, and a Milvus vector store for semantic retrieval. The system is wrapped in a Streamlit interface backed by MinIO object storage. Evaluated on a 163-passage benchmark labelled by GPT-4.1 and a 51-passage set reviewed by experts, the splitter surpassed three competing methods on a nine-metric composite score (6.137). The tuned RoBERTa model achieved 71.4 percent accuracy with a Cohen’s kappa of 0.337 on the full benchmark, and 87.5 percent accuracy with a kappa of 0.742 on passages with labels, outperforming both a climate-focused model and untuned language models.
[NLP-40] Chopthin-Consensus Power Sampling: A Diversity-Preserving Approach to LLM Decoding
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）在推理阶段通过序贯蒙特卡洛（Sequential Monte Carlo, SMC）进行功率采样（Power Sampling）时，因采用等权重重采样策略而导致低权重轨迹被过度修剪的问题。这一问题会破坏搜索空间的谱系多样性，丢弃潜在正确的推理路径，从而影响最终答案的准确性。其解决方案的关键在于提出一种名为Chopthin-共识功率采样（Chopthin-Consensus Power Sampling, CCPS）的新方法：该方法引入Chopthin重采样器，在不强制粒子权重均等化、避免无谓的粒子复制的前提下，对最大与最小权重之比施加上界约束，同时保留原始权重信息，从而有效维持推理路径的多样性；在此基础上，进一步采用语义多数选择机制，对完全相同的词元终态进行合并，对语义等价的答案进行聚类，并选取由最多不同推理路径支持的答案作为输出。该设计在保持加权SMC近似条件期望不变的同时，确保了后重采样有效样本量（Effective Sample Size, ESS）的下界。实验表明，CCPS在15个测试设置中，有14个达到或超越基准方法（Power-SMC）的最终答案准确率，最高提升达10.6个百分点，验证了保多样性的重采样与面向多样性的选择机制在无需训练的LLM推理中具有互补性与显著有效性。
链接: https://arxiv.org/abs/2609.12243
作者: Minoo Ahmadi,Seyedarmin Azizi,Erfan Baghaei Potraghloo,Mehdi Kamal,Massoud Pedram
机构: University of Southern California, Los Angeles, USA(南加州大学，洛杉矶，美国)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
备注: Accepted at the COLM 2026 Workshop on Efficient Reasoning
Abstract:Inference-time power sampling via Sequential Monte Carlo (SMC) can substantially improve large language model (LLM) reasoning without requiring post-training. However, many existing SMC approaches rely on equal-weight resampling, which can aggressively prune low-weight trajectories, discarding potentially correct reasoning paths and degrading the genealogical diversity of the search space. To address this, we introduce Chopthin-Consensus Power Sampling (CCPS). Our method applies the Chopthin resampler to LLM decoding: rather than equalizing weights and forcing unnecessary particle duplication, it enforces an upper bound on the ratio between the largest and smallest weights and carries the unequal weights forward. This targeted intervention preserves a richer set of distinct reasoning paths, keeps the weighted SMC approximation unchanged in conditional expectation, and guarantees a lower bound on the post-resampling effective sample size (ESS). To fully exploit this enriched population, we employ a semantic-majority selection mechanism that merges token-identical final trajectories, clusters semantically equivalent answers, and returns the answer supported by the largest number of distinct trajectories. Evaluating across three open-weight models and five reasoning benchmarks, we show that Chopthin increases oracle coverage in 13 of 15 settings. Combined with semantic-majority selection, CCPS matches or exceeds the final-answer accuracy of the Power-SMC baseline in 14 of 15 settings, delivering absolute gains of up to 10.6 percentage points. These findings demonstrate that diversity-preserving resampling and diversity-aware selection are complementary mechanisms for training-free LLM reasoning. Code is available at this http URL.
[NLP-41] Repair Before Reinforce: Context-Augmented Knowledge Graph Reasoning for Multi-Hop Question Answering
【速读】： 该论文旨在解决大语言模型（LLM）在多跳问答（multi-hop question-answering）任务中因仅基于孤立的“头-关系-尾”三元组进行训练而导致上下文感知能力不足的问题。其核心挑战在于，传统知识图谱（KG）训练方式难以捕捉支持推理所需的复杂上下文关联。为此，提出一种上下文增强型训练框架，通过从同一源文本片段中提取与主三元组相关的支持性三元组，构建上下文图（context graph, CG），从而引入更丰富的语境信息。关键创新在于设计双重监督机制：一是仅依赖目标三元组或路径的KG基础监督，二是结合上下文图中支持三元组的CG基础监督。在此基础上，采用监督微调（SFT）训练Qwen3-14B模型，生成KGModel与CGModel两个变体，并引入由大语言模型判断、具备历史感知能力的自适应修复流水线，对一跳推理失败案例进行持续微调并剔除或隔离噪声三元组，显著提升模型在清理后的一跳验证集上的准确率至100%。最终通过基于低阶问答样本的强化学习（RL）进一步优化模型泛化能力，结果表明，在胃轻瘫和糖尿病两类疾病特定知识图谱上，上下文增强监督均显著优于仅使用KG的监督方式，且以修复后的SFT检查点初始化的强化学习带来更大且更稳定的性能提升。
链接: https://arxiv.org/abs/2609.12230
作者: Tharaka D. Fonseka,Niraj K. Jha
机构: Princeton University(普林斯顿大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Question-answering often requires reasoning across multiple connected facts rather than retrieving a single isolated relation. Knowledge graphs (KGs) provide a structured way to represent such facts, but training large language models (LLMs) only on isolated KG head-relation-tail triples may limit their ability to learn the surrounding context needed for multi-hop reasoning. In this work, we propose a context-augmented training framework for multi-hop question-answering. Although generally applicable, we validate the framework in the context of disease-specific KGs, extracted using a reliable KG extraction framework called GraphMERT, for Gastroparesis and Diabetes. For each primary KG triple, we attach supporting triples extracted from the same source text chunk to form a context graph (CG). This creates two supervision settings: KG-grounded supervision, which uses only the target KG triple or path, and CG-grounded supervision, which uses the target KG triple or path together with supporting context triples. We train the Qwen3-14B model using supervised fine-tuning (SFT) under both settings, producing KGModel and CGModel variants. To strengthen the lower-hop factual foundation of the models, we introduce an LLM-judged, history-aware adaptive repair pipeline that identifies unresolved one-hop failures, continually fine-tunes on targeted repair examples, and removes or quarantines problematic noisy triples. This repair stage enables the models to reach 100% accuracy on the cleaned retained one-hop validation sets. Finally, we employ reinforcement learning (RL) using lower-hop question-answer items and evaluate generalization on harder 3-hop, 4-hop, and 5-hop tasks. Across both diseases, context-augmented supervision consistently improves multi-hop performance over KG-only supervision. RL initialized from repaired SFT checkpoints yields larger and more stable gains.
[NLP-42] GAUGE: When Not to Trust LLM -as-a-Judge in User-Simulated Evaluation of Task-Oriented Agents EMNLP2026
【速读】： 该论文旨在解决当前任务导向型大语言模型（LLM）智能体评估中普遍存在的评价有效性问题，即依赖低成本离线评估门（offline evaluation gate）进行智能体选型时，其排名结果与真实可验证奖励之间的匹配度不足。该评估门通常采用基于角色的LLM用户模拟器与候选智能体对话，再由“大模型作为裁判”（LLM-as-a-judge）对对话记录打分，从而决定优劣。然而，现有实践混淆了两种关键评价有效性：排序有效性（ranking validity）与构念有效性（construct validity）。研究发现，用户满意度与任务成功之间存在显著脱节——即使对话被盲评小组评为满意，实际任务成功率仍极低，57.5%的“满意”对话最终未能完成客户任务，这一现象在五组评分者、两个基准测试（τ²-bench 和 SimulatorArena）及所有主观维度上均一致出现。此外，尽管评估门在能力跨度较大的智能体间具有稳健的排序能力，但在近似性能的强智能体之间丧失分辨力，决策分歧率从宽差距对的1%跃升至近似对的31%。因此，该评估门虽经人工验证，却因锚点错误而产生误导。为应对该问题，论文提出“校准-信任”（calibrate-then-trust）评估范式，引入无需裁判的零成本完成位（judge-free completion bit）作为截断退化（truncation regression）的即时预警机制，以提升评估的可靠性与实用性。
链接: https://arxiv.org/abs/2609.12191
作者: Umesh Bodhwani,Thanh Tran,Kai Wei
机构: Amazon(亚马逊)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: Accepted to EMNLP 2026 (Industry Track)
Abstract:Comparing and selecting task-oriented LLM agents increasingly relies on a low-cost offline evaluation gate: persona-driven LLM user-simulators converse with each candidate, an LLM-as-a-judge scores the transcripts, and the higher-scoring agent is promoted. We introduce GAUGE, a reusable offline protocol that measures whether this gate’s ranking matches a grounded verifiable reward across 25 agents from six providers on the \tau^2 -bench and SimulatorArena benchmarks, separating two kinds of evaluation validity that release practices conflate: ranking validity and construct validity. First, a satisfaction-success gap: satisfaction carries essentially no information about task success, as conversations rated satisfied by our blind panel are decorrelated from actual success, with 57.5% of them failing the customer’s task, a pattern consistent across five rater populations, both benchmarks, and every subjective dimension we rated. Second, while the gate’s ranking is robust across the broad capability span, it loses resolution among the near-equal strong agents: this decision-disagreement rate jumps from 1% on wide-reward pairs to 31% on close pairs. The gate is thus human-validated yet mis-anchored. As a remedy, we propose a calibrate-then-trust cadence in which a judge-free completion bit is a zero-cost tripwire for truncation regressions.
[NLP-43] Can LLM s in Draft-Verify-Revise Pipelines Resolve Deictic Ambiguity?
【速读】： 该论文旨在解决在生成式 AI (Generative AI) 的“草稿-验证-修订”（draft-verify-revise）推理编排模式中，由于上下文在各阶段间传递所引发的指代歧义（deictic shift）问题。当不同阶段的模型对依赖上下文的表达（如“previous”）产生不一致的语义理解时，会导致输出错误。其核心解决方案在于通过引入具备多层次推理能力的元评估器（meta-evaluator），使其能够基于前序阶段的反馈进行更深入的语义分析与判断，从而纠正因指代转移造成的错误。实验表明，高推理投入显著提升了模型准确率，其中 Gemini 3 Pro 在低推理成本下即表现优异，而 GPT-5.2 需要高推理水平才能达到接近完美性能；此外，错误判断多源于对表面线索的依赖而非深层逻辑推理，提示在实际部署中应显式明确各阶段的指代关系，以增强系统的鲁棒性。
链接: https://arxiv.org/abs/2609.12162
作者: Obinna I. Ekekezie
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 86 pages including supplementary materials, 4 figures, 10 supplementary figures, 17 supplementary tables, 11 supplementary exemplars. Code and data: this https URL
Abstract:Draft-verify-revise is a common LLM orchestration pattern for scaling inference-time compute. One LLM drafts, a second critiques the draft and provides feedback, and a third uses that feedback to revise the draft into the final output. As context cascades between stages, LLMs at different stages can resolve a context-dependent expression such as “previous” differently. When that happens, the expression undergoes a deictic shift, a change in what it refers to. This phenomenon was studied with a synthetic dataset of 10 base examples, each rendered in three conditions. Holding the shared components constant, the conditions varied whether the draft stage LLM (the assistant) or the verify stage LLM (the grader) resolved the expression correctly, and how much independent reasoning the revise stage LLM (the meta-evaluator) needed to determine which reading was correct. Six models from three providers were tested across 21 reasoning effort configurations using e-values for sequential testing, in a primary experiment and an ablation experiment that removed error classification labels from the grader’s feedback. A separate LLM analyzed the meta-evaluator’s stated rationale for each wrong verdict. Balanced accuracy (the unweighted mean of sensitivity and specificity) ranged from 0.156, below chance, to near-perfect. GPT-5.2 rose from 0.156 without reasoning to 0.942 at its highest reasoning effort level, while Gemini 3 Pro stayed above 0.94 at every level. Gemini 3 Pro at low reasoning effort outscored GPT-5.2 at xhigh reasoning effort for roughly 5% of the cost per trial. When the meta-evaluator erred, it tended to rely on surface cues rather than operational reasoning. Context engineers implementing draft-verify-revise pipelines should be wary of deictic shifts and make the intended referent explicit at each stage.
[NLP-44] Population-level measures of perceived food access reveal barriers beyond geographic proximity
【速读】： 该论文旨在解决当前食物获取（food access）评估中过度依赖地理距离而忽视感知维度的问题，因为个体对食物可得性、可负担性、适宜性等主观维度的评价难以在大范围内进行量化。其核心解决方案是利用25,125条来自北卡罗来纳州罗利市49家杂货店的Google地图用户评论，通过无监督主题建模识别评论主题，并结合零样本分类（zero-shot classification）将主题映射至食物获取的五个维度——可获得性（availability）、可达性（accessibility）、可负担性（affordability）、适应性（accommodation）和可接受性（acceptability）。该方法在与人工标注对比时达到85.4%的一致性，验证了其有效性。研究发现，基于评论的衡量能够揭示仅靠地理接近性无法捕捉的结构性障碍，且同一连锁品牌在不同社区的感知差异显著，表明食物获取本质上是居民与食物环境之间匹配度的体现。此外，感知食物获取呈现出系统性的社会经济与人口学特征，虽与地理获取模式总体相似，但不完全一致。研究证明，线上零售评论可作为地理测量方法的有效补充，为大规模、低成本地捕捉食物获取的多维特征提供可行路径。
链接: https://arxiv.org/abs/2609.12132
作者: Teresa Groton,Benjamin rachunok
机构: North Carolina State University (北卡罗来纳州立大学); North Carolina State University (北卡罗来纳州立大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Food access is multidimensional, but population-level measurement still relies heavily on geography because perceived dimensions of access are difficult to measure at scale. Here, we use 25,125 Google Maps reviews from 49 grocery stores in Raleigh, North Carolina, to measure five dimensions of food access: availability, accessibility, affordability, accommodation, and acceptability. We identify review topics with unsupervised topic modeling and assign them to access dimensions using zero-shot classification, with 85.4% agreement against manual coding. The resulting store-level measures capture distinct aspects of food access and reveal barriers that geographic proximity alone does not capture. Comparisons between nearby stores in the same chain further show that identical store policies can be perceived very differently across locations, consistent with food access reflecting the fit between residents and their food environment. Perceived food access also follows systematic socioeconomic and demographic patterns that broadly parallel, but do not replicate, those observed for geographic access. These results show that online grocery reviews can provide a scalable complement to geographic measures of food access.
[NLP-45] Local Edits Global Ripples: Replay-Informed Policy Adaptation for Workflow Synthesis
【速读】： 该论文旨在解决生成式 AI（Generative AI）在持续适应过程中因提示词策略编辑（prompt-policy editing）引发的两个核心挑战：其一，编辑局部性不等于效果局部性，即对某一策略片段的修改可能通过下游执行产生连锁反应，影响未被直接修改的环节；其二，编辑效果具有组合敏感性，孤立有效的编辑在与其他编辑组合后可能发生冲突，导致原有收益丧失甚至引入有害行为。针对上述问题，论文提出 RIPPLE（Replay-Informed Persistent Policy Localization and Editing）框架，其关键在于将“编辑位置的选择”与“编辑在组合后的安全性评估”进行解耦。RIPPLE 首先通过诊断失败轨迹，将可行动的错误定位至预定义的策略段，并仅在该段内施加修正；随后，在保持迭代起始策略一致的前提下，评估候选编辑的独立增益；最后，通过回放已接受更新后的潜在编辑，模拟其在完整执行流程中的下游影响与交互效应，仅保留经验证仍安全的编辑。实验结果表明，RIPPLE 在 Flow-HO 基准上将验证成功率提升最高达 23.1%，并在多个冻结语言模型骨干上均实现正向改进，同时保持高效的编辑效率和低执行开销。靶向交互分析进一步验证了编辑的非局部传播特性及组合敏感性，凸显了 RIPPLE 框架在保障持续适应安全性和有效性方面的关键作用。
链接: https://arxiv.org/abs/2609.12127
作者: Manqing Mao,Hong Wang,Samson Koelle,Jie Yuan,Zhuoer Wang,James Feng,Yanjun Lin,Daniel Edmiston,Nikki Lijing Kuang,Zhecheng Sheng,Wei Niu
机构: Amazon, Inc.
类目: Computation and Language (cs.CL); Software Engineering (cs.SE)
备注: 33 pages, 20 tables, 6 figures
Abstract:Prompt-policy editing offers a practical way to improve agents that synthesize executable workflows without updating the underlying model. However, persistent prompt editing has two coupled properties. First, edit locality does not imply effect locality: an edit confined to one policy segment can ripple through downstream execution, altering behavior beyond the edited segment. Second, edit effects are composition-sensitive: edits that work in isolation can interfere after composition, causing one or both to lose their benefit or become harmful. Persistent adaptation must therefore support two distinct decisions: identifying where the policy should change from execution feedback, and determining whether the resulting edit remains safe to persist after composition. To address these challenges, we introduce RIPPLE (Replay-Informed Persistent Policy Localization and Editing), which separates where an edit is made from whether it remains safe after composition. It diagnoses failed trajectories, maps each actionable failure to a predefined policy segment, and restricts the correction to that part of the policy. RIPPLE then evaluates candidates against the same iteration-start policy to compare their isolated gains, before replaying promising edits after previously accepted updates to expose downstream effects and interactions. Only edits that remain safe under composition are retained. We evaluate RIPPLE on Flow-HO, a synthetic held-out benchmark for executable workflow synthesis. RIPPLE improves validation success by up to 23.1% and yields positive gains on two additional frozen language-model backbones, while maintaining edit efficiency and low execution cost. Targeted interaction analysis further demonstrates both properties: a segment-local tool-use edit changes downstream resource resolution and validation, while an edit beneficial in isolation becomes harmful after composition. Comments: 33 pages, 20 tables, 6 figures Subjects: Computation and Language (cs.CL); Software Engineering (cs.SE) Cite as: arXiv:2609.12127 [cs.CL] (or arXiv:2609.12127v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2609.12127 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[NLP-46] Quantifying Consonant Contributions to Word Intelligibility via Acoustic Masking
【速读】： 该论文旨在解决在运动性言语障碍康复治疗中，如何高效优先确定需干预的辅音问题。由于传统依赖感知实验的方法难以规模化，本文提出一种可扩展的解决方案：利用声学掩蔽（acoustic masking）评估辅音对词可理解性（intelligibility）的贡献。其核心方法是逐个静音孤立词汇中的辅音，并通过自动语音识别（ASR）模型检测该词是否被误识别，进而计算掩蔽诱发误识别率（mask-induced misrecognition rate, MMR），作为辅音贡献度的量化指标。研究验证了MMR与音素频率和功能负荷（functional load）两个语言学因素的相关性，发现音素频率与MMR呈负相关，功能负荷与MMR呈正相关，表明高频辅音被掩蔽后影响较小，而承载更多词汇对比功能的辅音则更具破坏性。跨语言分析进一步揭示辅音重要性排序具有语言依赖性，说明该贡献度并非普适规律。该方法通过多语言、多架构（MMS、Whisper、Qwen3-ASR）验证，为个性化言语康复策略提供了数据驱动的可扩展评估框架。
链接: https://arxiv.org/abs/2609.12122
作者: Eunjung Yeo,Kwanghee Choi,Krupaben Kothadia,Visar Berisha,Julie M. Liss,David R. Mortensen,David Harwath
机构: University of Texas at Austin (德克萨斯大学奥斯汀分校); Arizona State University (亚利桑那州立大学); Carnegie Mellon University (卡内基梅隆大学)
类目: Computation and Language (cs.CL)
备注: 7 pages, 5 figures, Accepted to SLT 2026
Abstract:Consonants contribute unequally to whether a word is understood. Given the limited time available for therapy, ranking consonants by contribution to intelligibility helps prioritize intervention targets in motor speech disorders. However, measuring this contribution relies on perceptual studies that are difficult to scale. This paper presents a scalable method that measures consonant contribution using acoustic masking. We silence one consonant at a time in an isolated word and test whether an automatic speech recognition (ASR) model still recognizes the word. We define a consonant’s contribution score as the proportion of its masked instances for which the word becomes misrecognized, which we refer to as the mask-induced misrecognition rate (MMR). We validate MMR against two linguistic factors previously reported to correlate with consonant contribution, namely phoneme frequency and functional load. We apply this analysis across four languages, English, Spanish, German, and Czech, using three ASR architectures, MMS (encoder-only), Whisper (encoder-decoder), and Qwen3-ASR (LLM-based). Using partial Spearman correlations, we find that phoneme frequency correlates negatively with MMR while functional load correlates positively. In other words, more frequent consonants are less disruptive when masked, whereas consonants carrying more lexical contrast are more disruptive. Further cross-language analysis shows that consonant rankings are not consistent, indicating that consonant contribution is language-dependent.
[NLP-47] he Cost of Compression: A Rate-Distortion Limit on Factual Hallucination
【速读】： 该论文旨在解决闭卷问答中事实性幻觉（factual hallucination）的成因问题，尤其针对传统观点仅将其归因于知识覆盖不足（coverage problem）的局限性。研究指出，即使相关事实曾被模型观测到，由于有限记忆容量导致其只能以近似方式存储，从而引发另一种关键错误来源——压缩失真（compression distortion）。解决方案的核心在于提出一个简化的“覆盖-压缩”（coverage–compression）模型，从信息论角度对事实回忆过程进行建模：在无检索的设定下，模型需将 M 个训练事实压缩至最多 B 位，并回答随机抽取的测试问题。通过理论推导，作者给出了一个下界表达式，明确分离了两类误差源：已观测事实因压缩导致的失真，以及未观测事实带来的覆盖率缺失。该框架为理解选择性记忆、强制压缩、结构设计、检索机制、拒绝回答策略及长上下文组织提供了统一的信息论视角。实验部分通过理论引导的模拟与受控的事实注入探针，在不同事实负载和有效可训练记忆规模的语言模型中验证了预测特征，揭示了在有限记忆约束下，事实性幻觉的一个可分离的失效模式——即对已观测事实的有损回忆。
链接: https://arxiv.org/abs/2609.12111
作者: Xi Wang,Shijia Xu,Rongfeng Guo
机构: Hefei Institutes of Physical Science, Chinese Academy of Sciences(中国科学院合肥物质科学研究院); University of Science and Technology of China(中国科学技术大学); Chongqing University(重庆大学); Shenzhen University(深圳大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Factual hallucination in closed-book question answering is often treated as a coverage problem: a model fails because the relevant fact is absent from its internal memory. This view misses a second source of error. Even when a fact has been observed, finite memory may force it to be stored only approximately. We study this effect through a simple coverage–compression model of factual recall. We consider an unstructured question-answering task with N possible queries and K possible answers. A learner observes M training facts, compresses them into at most B bits, and answers uniformly drawn test queries without retrieval. For a uniformly random ground-truth mapping, we prove \mathcalE \geq \fracMN\delta^\star!\left(\fracBM\right) + \left(1-\fracMN\right)\left(1-\frac1K\right) , where \delta^\star® is the inverse rate-distortion function of a uniform K -ary source under zero-one loss. The two terms separate compression distortion on observed facts from missing coverage on unobserved facts. The bound gives a compact way to reason about selective memory, forced compression, structure, retrieval, abstention, and long-context organization. We study the predicted signatures with theory-implied simulations and controlled fact-injection probes in modern language models that vary fact load and effective trainable memory. The result is not a complete theory of hallucination, but an information-theoretic account of a separable failure mode: lossy recall of observed facts under finite memory.
[NLP-48] What Counts as a Mistake? Annotating Recitation Events in Quran Memorization Transcripts
【速读】： 该论文旨在解决自动语音识别（ASR）转录文本在阿拉伯语《古兰经》诵读校对过程中，如何准确区分未纠正的错误、重复、修正、开场格式以及可接受的拼写差异这一核心问题。其解决方案的关键在于构建一个经过人工标注的完整数据集（包含100个生产录音案例，共348个评分单元和162个局部事件，涵盖十类标签），并设计一个可执行的评估器，能够联合评分标签与词位位置。实验表明，基础的文本差异比对方法（plain diff）在标签感知的F1值上达到0.525，在定位F1上为0.826；而经过适配的生产级清洗/对齐组件虽略有下降，但依然保持在0.518和0.786水平，且精确跨度F1为0.505。通过修正适配器中的词坐标，成功恢复了所有五个标注的重复事件，凸显出在解读基线失败前必须验证标注界面的必要性。初步试点测试中，八次单次20分钟运行覆盖三个编码代理与八种模型，标签感知F1范围从0.143到0.892，其中七种表现显著优于所有基线，一种因缺失归一化步骤而崩溃。六种模型中，970/972个真实事件均被至少一次预测覆盖，剩余差异并非检测能力问题，而是由约定俗成的边界定义所决定——即标签边界依赖仲裁而非文本显式呈现。此外，162个事件中有七个始终被六种同日运行遗漏，其中五个源于单一正字法规则，最强模型亦未能捕获这些实例，表明当前方法尚未实现完全自动化标注，其评估仅衡量算法部分任务，尚未覆盖人工标注流程本身。
链接: https://arxiv.org/abs/2609.12085
作者: Mohamad Al Mdfaa,Nursultan Askarbekuly,Ahmed Helaly,Ubai Sandouk,Manuel Mazzara
机构: LemoniLab FZCO (lemonilab.com); Innopolis University (innopolis.ru)
类目: Computation and Language (cs.CL)
备注:
Abstract:Checking Quran recitation from an ASR transcript requires distinguishing unresolved mistakes from repetitions, repairs, opening formulas and accepted spelling differences. We report a completed human annotation of 100 production recording cases: 348 scored units and 162 localized events across ten combined labels. An executable evaluator scores labels and word positions together. A plain diff reaches label-aware F1 0.525 and localization F1 0.826; adapted production cleaner/alignment components reach 0.518 and 0.786, with exact-span F1 0.505 for both. Correcting the adapter’s word coordinates recovers all five annotated repetition events, showing why annotation interfaces must be checked before interpreting baseline failures. In a preliminary pilot, eight single 20-minute runs across three coding agents and eight models span label-aware F1 0.143 to 0.892: seven land far above every baseline, and one collapses below the naive diff from a missing normalization step. Across the six, 970 of 972 gold-event instances draw an overlapping prediction, so what remains is not detection but convention: span extent, and the labels whose boundary is stipulated by adjudication rather than visible in the text. Seven of 162 events defeat all six same-day runs, five of them one orthographic rule, and the strongest run still misses the same ones. No run annotated before building, so the pilot measures the algorithm half of the task only.
[NLP-49] Is Bash All You Need? An Empirical Study of Tool Interfaces for Enterprise Digital Worker Agents
【速读】： 该论文旨在解决在企业级任务中，通用型工具接口（如shell）是否能够优于专用工具接口的问题。企业工作不仅涉及代码编写，还包括跨应用与服务的切换、同事协作以及专业分析等复杂行为，传统依赖类型化工具（typed tools）的方案在灵活性和效率上存在局限。研究的关键解决方案在于系统比较五种不同工具接口：类型化工具、类型化工具加bash、纯bash、带持久化代理合成工具的bash，以及程序化工具调用（Programmatic Tool Calling, PTC）。实验结果表明，纯bash接口在TheAgentCompany和APEX-Agents两个基准测试中均显著优于其他方法，分别提升得分21.8–24.5个百分点和4.8–7.4个百分点，同时减少19%–72%的总令牌消耗；而引入类型化工具或持久化工具合成机制并未带来可检测的性能增益。尽管PTC在安全性与合规性要求较高的场景下表现尚可，且比直接类型化调用更节省资源，但其整体任务完成质量与成本效益仍逊于纯bash。因此，研究的核心结论是：当允许任意执行且能有效隔离时，推荐使用纯bash接口；而在需强制遵循固定工具目录的安全或合规环境中，则应采用PTC。
链接: https://arxiv.org/abs/2609.11999
作者: Hazel Mak,Susheel Suresh,Sahil Bhatnagar,Barry Wang,Chhaya Methani,Alejandro Gutierrez Munoz
机构: Microsoft(微软)
类目: oftware Engineering (cs.SE); Computation and Language (cs.CL)
备注: 13 pages, 7 figures, 12 tables
Abstract:In this study, we examine whether a general shell can outperform specialized tools on enterprise tasks. Shell-based agents have shown strong results in coding, but enterprise work also involves moving between applications and services, coordinating with coworkers, and performing professional analysis. We compare five tool interfaces on TheAgentCompany and APEX-Agents using Opus-4.8 and GPT-5.5: typed tools, typed tools plus bash, bash alone, bash with persistent agent-synthesized tools, and programmatic tool calling (PTC), which runs programs whose actions are restricted to a typed tool catalog. Bash alone outperforms typed tools on both benchmarks, improving score by 21.8-24.5 pp on TheAgentCompany and 4.8-7.4 pp on APEX-Agents while using 19-72% fewer total tokens. Adding typed tools or persistent tool synthesis to bash produces no detectable pooled score gain. PTC uses fewer tokens than direct typed calls with broadly similar task performance, but generally underperforms bash alone in both quality and cost efficiency. For enterprise practitioners, these results favor bash alone when arbitrary execution can be isolated and PTC when security or compliance policies require a fixed tool catalog.
[NLP-50] Harness or Model? Isolating the Harness Effect in Agent ic Coding with a Contamination-Controlled Private Suite
【速读】： 该论文旨在解决生成式AI（Generative AI）在软件工程任务中，厂商定制的智能体框架（harness）是否真正优于通用或中立框架这一关键问题。研究假设认为，厂商原生搭配的框架能更高效地完成更多任务，但该论文通过在256个私有、去污染控制的任务集上进行成对对比实验，检验了这一假设。其解决方案的关键在于构建一个可复现的基准测试体系：使用统一的调度器（orchestrator）和独立评分器（grading oracle），在相同模型（Claude Opus 4.8、GPT-5.5）下分别运行厂商专属框架（如claude-agent-sdk、openai-codex SDK）与中立框架（deepagents），并严格控制任务执行环境。结果表明，两种框架在整体任务成功率上无显著差异（例如，Opus 4.8上为48.8% vs 50.0%，GPT-5.5上为55.6% vs 54.4%），且性能表现呈现任务类型依赖性——在代码库任务中厂商框架落后，而在竞赛类任务中则领先，暗示其优势具有场景特异性。此外，研究还揭示了框架成本差异：中立框架在每成功解决一任务上的实际消耗为厂商框架的1.2至1.6倍，但由于部分运行记录缺失，账单排序尚不确定。该研究强调了评估框架效能需考虑任务分布、执行效率与真实成本，并修正了此前因遥测语义缺陷导致的成本误报问题，最终公开了核心工具链以支持可重复研究。
链接: https://arxiv.org/abs/2609.11987
作者: Mohsen Arjmandi
机构: evolutionID GmbH(德国)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Software Engineering (cs.SE)
备注: 25 pages, 5 figures. Revised 8 September 2026 (telemetry correction). Code and derived data in the replication package
Abstract:An agentic coding system couples a language model to a harness: the tools, prompts and control flow that turn a chat model into an autonomous software engineer. Vendors ship harnesses tuned to their own models, and practitioners assume the vendor-native pairing solves more tasks. We measure that assumption with paired same-model contrasts on a private, contamination-controlled suite of 256 repository and post-cutoff contest tasks. The same 80 tasks ran under claude-agent-sdk and under deepagents on claude-opus-4-8, and under the openai-codex SDK and deepagents on gpt-5.5, with gemini-3.5-flash and deepseek-v3.2 as side cells. 792 of 800 planned runs were graded by an isolated oracle. Neither contrast resolves an average advantage for either harness: -1.25 pp for Opus 4.8 (48.8% vs 50.0%, task-bootstrap 95% CI [-10.0, +7.5]) and +1.25 pp for GPT-5.5 (55.6% vs 54.4%, CI [-4.4, +6.9]). The Opus average combines opposite strata: the native harness trails by 9.0 pp on the 61 repository tasks and leads by 23.7 pp on the 19 contest tasks (label-permutation p = 0.003). The partition was chosen after seeing the data and needs a designed replication. Correctness and completion also separate: 22 of 81 runs cancelled at the wall-clock ceiling had produced a passing patch. Re-priced from raw per-turn usage at frozen list prices, the neutral harness cost 1.3 to 1.6 times as much per solved task on Opus 4.8 and 1.2 times on GPT-5.5. These are observed-usage estimates. On the Anthropic account 58 runs left no usage record, and allocating that spend to either cell would move the Opus ratio between 0.7 and 2.3, so the billed ordering is unresolved. This revision corrects an August 2026 manuscript whose cost figures rested on a usage-semantics defect in our own telemetry (Section 5.1). We release the orchestrator, grading oracle, reanalysis code and derived aggregates. The tasks stay private.
[NLP-51] Space as an Interventional Invariant: Cross-Modal Predictive Geometry for Stratified Cities and Em-Spaced Intelligence
【速读】： 该论文旨在解决多模态感知与城市系统中异构过程如何共同揭示统一空间结构的问题，尤其针对不同模态间缺乏共享度量或表征空间的挑战。现有方法通常将空间视为几何容器或孤立表征的集合，难以解释跨模态信息在非一致度量下的协同一致性。其核心解决方案是将空间定义为干预不变性（interventional invariant）——即在允许操作下保持局部相容性及未来观测条件律的最小关系结构。该框架构建了跨模态预测几何（cross-modal predictive geometry），融合局部状态空间、模态特异性观测映射、动作群胚（action groupoid）与规范预测状态商空间，并引入明确的因果条件以区分干预性结构与仅观测性关联。关键理论结果表明，在联合点分离性、等变性与干预忠实性条件下，潜在空间可被识别至干预群中心化子（centraliser of the intervention group）的范畴内，从而将表示歧义降至剩余坐标自由度。进一步通过层化流形（sheaf-valued representations）扩展至分层城市系统，实现几何、物理、移动、社会与经济等多维度层的共存而不强制统一为单一度量。合成实验验证了等变性、预测充分性、全息性（holonomy）、限制映射恢复、跨尺度一致性及上下文饱和性。该框架为空间认知、城市科学、具身智能与“嵌入式智能”（em-spaced intelligence）提供了统一且可证伪的理论基础。
链接: https://arxiv.org/abs/2609.11959
作者: Tao Yang,Xuhui Lin,Kunyao Li,Haijiang Li
机构: Tsinghua University (清华大学); University College London (伦敦大学学院); Cardiff University (卡迪夫大学)
类目: Machine Learning (cs.LG); Computation and Language (cs.CL)
备注:
Abstract:Space is a foundational concept across mathematics, physics, spatial cognition, urban science, and embodied intelligence, yet these fields often treat spatial structure either as a shared geometric container or as a collection of disconnected representations. Such approaches struggle to explain how heterogeneous sensory and urban processes can jointly reveal a common spatial structure, particularly when different modalities do not share the same metric or representation. This paper addresses this gap by defining space as an interventional invariant: the minimal relational structure that preserves local compatibility and the conditional laws of future observations under admissible actions. We develop a cross-modal predictive geometry that integrates local state spaces, modality-specific observation maps, an action groupoid, and a canonical predictive-state quotient, with explicit causal conditions for identifying interventional rather than merely observational structure. The key theoretical result shows that, under joint point separation, equivariance, and interventional faithfulness, the latent space is identifiable up to the centraliser of the intervention group, thereby reducing representational ambiguity to residual coordinate freedom. The framework is further extended to stratified urban systems using sheaf-valued representations, allowing geometric, physical, mobility, social, and economic layers to coexist without being reduced to a single metric. Synthetic experiments under noise evaluate equivariance, predictive sufficiency, holonomy, restriction-map recovery, cross-scale consistency, and context saturation. The resulting framework provides a unified and falsifiable foundation for spatial cognition, urban science, embodied AI, and em-spaced intelligence.
[NLP-52] R2VC: Modular Fact-Checking with Retrieval Verification and Confidence Calibration
【速读】： 该论文旨在解决大语言模型在自动化事实核查中因端到端提示（end-to-end prompting）导致的证据检索、推理与置信度估计耦合问题，从而造成错误难以诊断且置信度不可靠。其核心解决方案是提出一种模块化架构R2VC（Retrieve, Reason, Verify, Calibrate），通过分离四个关键阶段实现更可解释、更可靠的核查流程：首先采用混合稀疏-密集检索策略从维基百科中获取证据；其次利用监督微调并经过直接偏好优化（DPO）对齐的生成器产出多样化的结构化结论候选；再通过外部自然语言推理（NLI）交叉编码器进行基于证据的候选选择；最后引入轻量级序列级校准器以估计置信度并支持选择性拒答。实验结果表明，在FEVER数据集上，使用80亿参数骨干模型的R2VC相比基线显著提升13.74%准确率；消融研究进一步证实验证器驱动的候选选择和置信度校准是性能提升的主要贡献因素，移除任一模块均会导致性能显著下降。人工分析250个错误案例也揭示，当前瓶颈仍主要源于检索失败，尤其是错误实体的证据召回。综上，该研究证明了模块化事实核查流水线在提升开放域验证的预测准确性与置信度可靠性方面具有显著优势。
链接: https://arxiv.org/abs/2609.11955
作者: Dhruv Dixit,Paritosh Pandey
机构: Stevens Institute of Technology (史蒂文斯理工学院); University of North Carolina (北卡罗来纳大学)
类目: Computation and Language (cs.CL); Machine Learning (cs.LG)
备注: 21 pages, 15 figures
Abstract:Large language models are increasingly used for automated fact checking, but end-to-end prompting often entangles evidence retrieval, reasoning, and uncertainty estimation, making failures difficult to diagnose and confidence difficult to trust. We present R2VC, a modular retrieve, reason, verify, calibrate architecture for evidence-grounded fact checking with citations and abstention. R2VC combines hybrid sparse+dense retrieval over Wikipedia, a supervised fine-tuned and DPO-aligned generator that produces diverse structured verdict candidates, an external NLI cross-encoder for evidence-based candidate selection, and a lightweight sequence-level calibrator for confidence estimation and selective abstention. On FEVER, an 8B backbone with R2VC achieves 13.74% higher accuracy than baseline. Ablation studies show that verifier-based candidate selection and confidence calibration are the largest contributors to performance. Removing candidate selection drops FEVER accuracy to 76.24%, while removing calibration nearly doubles the Brier score to 0.161. A manual analysis of 250 errors further shows that retrieval failures, especially wrong-entity evidence, remain the dominant bottleneck. Together, these results show that modular fact-checking pipelines can substantially improve both predictive accuracy and confidence reliability in open-domain verification.
[NLP-53] PRISMA-LLM : An Empirical Reporting Framework for AI-Assisted Systematic Reviews
【速读】： 该论文旨在解决生成式 AI（Generative AI）在系统性文献综述（systematic review）自动化流程中广泛应用背景下，相关工作在方法透明度、评估完整性及局限性报告方面存在的不一致与不足问题。其核心解决方案是提出PRISMA-LLM框架，该框架基于对888篇综述自动化论文及14,726条标注数据的实证分析，强调将实现细节的披露与对结果敏感性的评估和局限性报告进行分离，以提升自动化流程的可审计性与可信度。研究发现，自2023年以来，软件类产品论文中无评估报告的比例高达38.0%，显著高于大语言模型（LLM）论文的9.3%；尽管复杂度较高的LLM工作流报告覆盖率更高，仍有52%的仅报告正面结果的研究未满足可靠性或性能要求，凸显了评估规范化的紧迫性。PRISMA-LLM通过结构化分离“实现披露”与“后果敏感性评估”，为构建可复现、可验证的AI辅助综述实践提供了系统性指导。
链接: https://arxiv.org/abs/2609.11559
作者: Miguel Zabaleta,Baihan Lin
机构: Icahn School of Medicine at Mount Sinai(伊坎医学院); Department of AI and Human Health(人工智能与人类健康系); Department of Psychiatry(精神病学系); Department of Neuroscience(神经科学系)
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注:
Abstract:Large language models (LLMs) and AI-enabled software increasingly participate in systematic-review decisions, yet the information needed to audit these workflows is reported inconsistently. We analyze SciLitBench, a corpus of 888 review-automation papers with 14,726 annotations, to characterize changes in methods, review-stage use, evaluation and reported limitations. Automation has shifted toward LLM- and software-facing workflows, including stages that can alter the evidence base. Since 2023, 38.0% of software/product papers reported no evaluation, compared with 9.3% of LLM papers. Reporting coverage increased with LLM workflow complexity, yet 52% of positive-only LLM evaluations still reported an unmet reliability or performance requirement. From these patterns, we introduce PRISMA-LLM, an empirically grounded framework separating implementation disclosure from consequence-sensitive evaluation and limitation reporting.
[NLP-54] MP-Bench: Evaluating Voice Agents as a Multiparty Conversation Participant EMNLP2026
【速读】： 该论文旨在解决当前对话式语音代理（conversational voice agents）在多角色对话（multi-party conversations）场景下评估不足的问题。尽管现有研究主要聚焦于双人交互（dyadic interactions）和被动音频理解，但真实应用场景中更常见的是多方参与的复杂对话，其交互复杂度呈指数级增长，对语音代理的实时理解与参与能力提出了更高要求。为此，论文提出首个专门针对多角色对话情境下语音系统客观评估的基准测试——Multiparty Bench（MP-Bench），其核心在于从两个关键维度衡量代理表现：话轮转换意识（turn-taking awareness）与回应恰当性（response appropriateness），并辅以基于理解的问答任务作为补充评估手段。实验结果表明，当前12个实时语音代理在多角色对话理解任务中的表现普遍低于22%，在话轮转换判断上接近随机水平，揭示了现有技术在多角色动态交互场景下的显著局限性，凸显了构建具备高级社会互动能力的实时语音代理仍面临重大挑战。
链接: https://arxiv.org/abs/2609.13076
作者: Yi-Jen Shih,Shih-Yun Shan Kuan,Guan-Ting Lin,Kai-Wei Chang,Siddhant Arora,Shu-wen Yang,Abdelrahman Mohamed,Shinji Watanabe,Hung-yi Lee,David Harwath
机构: The University of Texas at Austin(得克萨斯大学奥斯汀分校); National Taiwan University(台湾国立大学); Massachusetts Institute of Technology(麻省理工学院); Carnegie Mellon University(卡内基梅隆大学); Meta AI(Meta人工智能)
类目: Audio and Speech Processing (eess.AS); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: Accepted to EMNLP 2026 Findings
Abstract:Conversational voice agents have advanced significantly, offering increasingly natural human-machine interactions through both cascaded and end-to-end architectures. However, while recent benchmarks extensively evaluate dyadic interactions and passive audio comprehension, they largely overlook a prevalent real-world scenario: multi-party conversations. Evaluating agents in these settings is fundamentally more challenging than in dyadic interactions due to the exponentially greater conversational complexity. For voice agents to integrate seamlessly into human group dynamics, they must not only generate contextually appropriate responses but also demonstrate a nuanced understanding of open turn-taking. To address this gap, we introduce Multiparty Bench (MP-Bench), the first benchmark specifically designed to objectively evaluate conversational speech systems as active participants within multi-party contexts. MP-Bench assesses agent behavior along two primary dimensions: turn-taking awareness and response appropriateness. Additionally, we incorporate comprehension-based question-answering tasks as a complementary evaluation. By benchmarking 12 voice agents, we find that real-time voice agents stay at or below 22% on multiparty comprehension and remain near chance on multiparty turn-taking, exposing an open challenge for real-time voice agents under multiparty scenario.
[NLP-55] Diverse Minds Divided Networks? Personality Composition Polarization and Collective Intelligence in LLM -Based Social Simulations
【速读】： 该论文旨在解决在线极化（online polarization）与集体智能（collective intelligence）在现有研究中常被孤立考察的问题，尤其关注社会性格构成是否同时影响两者，以及降低极化是否会以牺牲集体能力为代价。其解决方案的关键在于提出一种名为TraitMix的实验设计，将模拟社交网络中个体的性格特质（基于大五人格模型，Big Five）的水平与异质性作为受控变量，并在同一实验框架下同步测量极化程度与集体表现。在991次包含百名代理的模拟中，研究发现性格异质性对极化具有最显著的影响：高异质性社会虽观点更分散，但群体间分化程度更低，形成非共识性的多元讨论空间；而同质性社会则表现为封闭的共识回音室，而非真正意义上的“温和”。值得注意的是，性格效应并非线性叠加，宜人性（Agreeableness）决定了开放性（Openness）的符号方向，这一交互作用在不同语言模型间具有可重复性，且主模型估计显示该效应由影响力驱动。研究结果挑战了“极化与集体效能存在权衡”的预设——任何极化指标均未预测集体绩效下降，仅跨维度互动（cross-cutting interaction）在控制聚合身份后仍与集体准确性显著相关。研究进一步通过消融分析排除了潜在测量循环性问题，包括对所有模型施加诱导门（induction gate）及剔除不通过检验的测量指标，增强了结论的可靠性。
链接: https://arxiv.org/abs/2609.12444
作者: Raad Bin Tareaf
机构: German University of Digital Science(德国数字科学大学)
类目: Physics and Society (physics.soc-ph); Computation and Language (cs.CL)
备注:
Abstract:Simulated societies of large language model agents are used to study online polarization, and separately to study collective intelligence, but the two are rarely measured in the same system. It is therefore difficult to say whether a society’s personality composition shapes both, or whether reducing polarization costs collective competence. We present TraitMix, an experimental design in which the Big Five composition of a simulated social network, both trait levels and trait heterogeneity, is a controlled experimental variable, and in which polarization and collective performance are measured in the same runs. Across 991 simulations of hundred-agent societies, spanning six contested topics and six language models, trait heterogeneity has the largest measured effects, acting in opposite directions on two faces of polarization: varied societies hold more dispersed opinions while being less segregated into camps, so homogeneous societies are not moderate but consensual echo chambers. Trait effects are not additive, as Agreeableness determines the sign of Openness, an interaction that replicates across models although the primary model’s estimate is influence-driven. Contrary to the trade-off the study was designed to measure, no polarization measure predicts poorer collective performance, and cross-cutting interaction is the only one of four whose association with collective accuracy survives partialling on the aggregation identity. We report ablations removing two potential measurement circularities, an induction gate applied to every model, and the measures that failed them.
信息检索
[IR-0] Autonomous Research for Open-Ended Problems: A Case Study on Telecom Ticket Retrieval
链接: https://arxiv.org/abs/2609.13073
作者: Junghyun Min,Huseyin Uzunalioglu,Mohamed Trabelsi
类目: Artificial Intelligence (cs.AI); Information Retrieval (cs.IR); Machine Learning (cs.LG)
备注: 10 pages, 3 tables
Abstract:Recent breakthroughs in LLM-based systems and their abilities in problem solving and coding have allowed progress in the AI for Science paradigm, potentially replacing human roles in machine learning (ML) research. However, while several frameworks of fully autonomous end-to-end ML research have been proposed, successful implementations of them are often limited to problems with narrow search spaces, like language modeling or biomedical ML benchmarks. In this paper, we explore how autonomous research can be adapted to solve open-ended, industry-grade ML problems, by considering a case study: telecom ticket retrieval, an open-ended task with degrees of freedom in representation, architecture, and training data generation. We discover that autonomous research for open-ended problems with commercial and open-source agents shows both promise and limitations: while autonomous research can excel in narrow hyperparameter optimization, it lacks human-like intuition and creativity and requires operational overhead. Even with minimal human supervision, autonomous research can reach 90% of state-of-the-art performance (0.34 vs. 0.38 Recall@1) in a much shorter time period (10 weeks vs. 10 months of human work) at a modest cost (up to \ 200 per Cursor campaign). Our empirical evidence recommends that human researchers and autonomous research frameworks work together for best results in ML research.
[IR-1] MIMA: Multi-Interest Recommendation via Multi-Positive Exclusive Assignment
链接: https://arxiv.org/abs/2609.12842
作者: Xingyuan Mao,Alin Fan,Shichao Nie,Junfeng Zhang,Yan Xiao,Tao Luo,Xiaoyi Zeng
类目: Information Retrieval (cs.IR)
备注:
Abstract:Multi-interest recommendation represents each user with multiple interest vectors for fine-grained candidate matching, yet it often suffers from interest collapse, where the learned interests converge to similar representations. We highlight the prevailing single-positive paradigm as one important factor behind this issue. Since each instance provides only one positive item, intents are optimized independently, potentially causing the same best-matching interest to be repeatedly updated toward different positives while leaving the others under-supervised. Moreover, existing methods rarely model how strongly a user activates each interest, leaving scores from different interest channels incomparable at inference. To address these problems, we propose MIMA, a Multi-Interest recommendation framework built on Multi-positive exclusive Assignment. MIMA groups items co-occurring within the same request into a positive set, generates complementary interests with a causal Transformer decoder, and exclusively assigns each positive to supervise a distinct interest via Hungarian matching, so that interest differentiation emerges from the training objective itself rather than auxiliary regularization. A lightweight routing module further estimates user-interest activation probabilities to calibrate scores across interest channels. Experiments on three public datasets and an industrial dataset show that MIMA consistently outperforms state-of-the-art baselines, and an online A/B test yields significant business gains.
[IR-2] Cognition on Graph: Navigating Massive Knowledge Space via Cognitive Cycles and Bidirectional Graph-Text Synergy EMNLP2026
链接: https://arxiv.org/abs/2609.12791
作者: Gengxian Zhou,Jian Xu,Zichen Tang,Shiming Xiang,Haihong E,Cheng-Lin Liu
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
备注: Accepted to EMNLP 2026 Main Conference
Abstract:Retrieval-Augmented Generation (RAG) has empowered Large Language Models (LLMs) to tackle knowledge-intensive tasks. However, navigating global, heterogeneous knowledge bases (large-scale knowledge graphs and text corpora) for complex reasoning remains a challenge. Existing methods typically employ reactive, graph-driven exploration strategies, which blindly follow graph topology without adapting to the question context or evolving exploration progress, and lack deep bidirectional synergy between graph and text. To address these limitations, we propose CoG (Cognition on Graph), a cognitive-inspired, training-free framework for adaptive knowledge exploration. Drawing inspiration from human problem-solving, CoG performs a continuous plan-explore-reflect cycle, where it proactively formulates investigation plans, performs dual-source retrieval, and dynamically reflects on progress to adjust strategies. Crucially, it establishes deep bidirectional synergy between structured graph and unstructured text, where entities extracted from text dynamically guide graph exploration to bridge knowledge gaps. Extensive experiments on seven multi-hop QA benchmarks demonstrate that CoG significantly outperforms state-of-the-art methods while achieving superior exploration efficiency. Our code and datasets are available at this https URL.
[IR-3] A Historical Corpus Is Not a Historical System: Auditing Hindsight Leakage in Stateful Data Discovery
链接: https://arxiv.org/abs/2609.12766
作者: Yixi Zhou,Fan Zhang,Sikun Wang,Yingfan Xu,Haipeng Zhang
类目: Information Retrieval (cs.IR)
备注: 12 pages, including figures and tables
Abstract:Offline replay should estimate what a discovery system could retrieve at a historical point, yet freezing the corpus leaves interaction memory unconstrained. We formalize point-in-time (PIT) discovery through historical state (D_t, \theta_t, M_ i) and introduce a paired replay that changes only memory availability. The protocol constructs PIT and full-stream Future views from behavior-only traces and audits selected entries with a Temporal Violation Rate. Across three table-text domains, two stream regimes, two retrievers, and five seeds (216,000 rows), Future inflated Asset Recall@100 by 2.62-5.24 points; all 12 paired intervals excluded zero. With behavior-only trace memory, PIT underperformed the no-memory Stateless condition; Future masked 32.7-48.4% of that harm. For a simulated positive-feedback cache, PIT added 4.65-18.96 points over Stateless while Future added another 4.11-9.58 points. On five timestamped FreshStack topics, Future exceeded PIT by 2.72 points [1.75, 3.71]. Historical evaluation must version and validate memory with the corpus.
[IR-4] Learning the Lake: Reliable Experience for Adaptive Data Product Discovery
链接: https://arxiv.org/abs/2609.12754
作者: Yixi Zhou,Fan Zhang,Sikun Wang,Yingfan Xu,Haipeng Zhang
类目: Information Retrieval (cs.IR)
备注: 11 pages, including figures and tables
Abstract:Data-product discovery searches a full lake even when workloads revisit related products and regions. Repetition permits contracted search, but similarity cannot justify a route because one omitted asset invalidates a conjunctive product. We study when serving experience can safely reduce this work. Evolving Discovery Memory records source-labelled query–product–region evidence above a fixed regional index. SafeLake separates operational familiarity, which determines how much to search, from independently calibrated product evidence, which determines where to search. The fixed-probe comparison holds the adaptive budget constant between SafeLake and Familiarity-only. On TAT-QA, product steering raises Product Recall by 0.072; ConvFinQA shows no resolved map gain, while the HybridQA sensitivity favors Familiarity-only in Full R@100. Trace-only, missing, and false feedback expose boundaries on map steering, while scope-audit agreement cannot certify the source. Across clean confirmed-feedback streams under the frozen transductive protocol, the formal controller saves 49.5–82.7% of cumulative asset exposure. Experience determines when to contract; reliable evidence determines where to contract.
[IR-5] Personalized and Trust-Aware Health Recommendation Policies for a Construction Workplace
链接: https://arxiv.org/abs/2609.12679
作者: Atefeh Mollabagher,Yogesh Gautam,Houtan Jebelli,Parinaz Naghizadeh
类目: Information Retrieval (cs.IR); Systems and Control (eess.SY)
备注:
Abstract:Construction workers face workplace risks such as fatigue, heat stress, and other physically demanding conditions that can negatively affect their health and safety. Although monitoring these risks is important, timely and personalized health interventions are also needed to help prevent negative impacts on workers’ well-being and productivity. To this end, in this paper, we propose a model to capture the interactions between a trust-aware health recommender system and workers who differ in health and trust sensitivity. Specifically, in our proposed dynamic model, worker health evolves over time, worker trust is affected by both health and recommendation dynamics, and trust in turn affects compliance with future recommendations. Given this model, we characterize the recommender policy, including a health-based recommendation triggering threshold and the recommendation frequency. We do so using both model-based short-horizon control and model-free reinforcement learning. We then investigate how recommendation frequencies are adjusted for different workers to balance their health, productivity, and trust. Our findings provide insight into the design of personalized health recommendation policies in construction workplaces and beyond.
[IR-6] Preference-Drift-Aware Subsequence Learning and Hierarchical Context Fusion for Long-Sequence Generative Recommendation
链接: https://arxiv.org/abs/2609.12556
作者: Fei Li,Qingyun Gao,Jianzhe Zhao,Guibing Guo,Beibei Kong,Lei Cheng,Chengxiang Zhuo,Zang Li
类目: Information Retrieval (cs.IR)
备注:
Abstract:Long-sequence generative recommendation methods autoregressively model the user’s interaction sequence to generate the next-item representation. Existing methods generally fall into two categories: efficient full-sequence modeling and target-aware context retrieval. Our experiments reveal that as the sequence length increases, the former incurs steadily growing computational cost while its accuracy gains quickly saturate and even degrade due to noise; the latter, though shortening the input sequence, is susceptible to noise that is semantically consistent yet preference-inconsistent, as well as to incomplete contexts. Both paradigms ignore the dynamic changes of user preferences and the cross-subsequence dependencies when handling historical information, thereby limiting accuracy and efficiency. To address these issues, we propose a preference-drift-aware subsequence learning and hierarchical context fusion for long-sequence generative recommendation. Specifically, we learn differentiable soft subsequence boundaries using multidimensional preference-drift information and aggregate items within each subsequence into preference-coherent representations via linear attention with soft assignment weights, thereby circumventing the expense of full-sequence attention. A cross-attention mechanism is then employed to capture dependencies between recent interactions and relevant subsequence contexts, mitigating noise in learning recent-item representations. Finally, a gated fusion mechanism adaptively combines the recent-item representation with the global subsequence context, allowing the resulting target representation to encode both recent and long-term preferences. Extensive experiments demonstrate that our method consistently outperforms existing baselines in both recommendation accuracy and computational efficiency.
[IR-7] OneLA: Scaling Linear-Attention Decoding to Large Beams in Generative Recommendation
链接: https://arxiv.org/abs/2609.12399
作者: Xiangrui Yang,Cheng Peng,Yunfeng Zhao,Liang Zeng,Ao Hu,Jiawei Yang,Shengzhe Wang,Jingshan Lv,Xiao Liang,Chen Yang,Jiaqiang Liu,Yiming Qiu
类目: Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC); Information Retrieval (cs.IR)
备注:
Abstract:Generative recommendation (GR) relies on large-beam decoding to generate hundreds of candidate items, creating a new scaling challenge for recurrent linear attention. Existing linear attention serving systems either materialize a full recurrent state for every beam or repeatedly replay shared history, incurring substantial memory and traffic overhead. To address this, we present OneLA, a linear-attention decoding framework that exploits the shared prompt and short divergent suffixes of GR workloads. Specifically, OneLA represents all beam states using a single shared prompt-derived state and compact, append-only records of their divergent transitions. Using this representation, OneLA computes only the state information required at each decoding step, without reconstructing a full recurrent state for every beam. Furthermore, OneLA uses a lightweight ancestry index to track the transition records that make up each beam’s history, allowing beams to be updated without moving or copying existing records. A fused GPU kernel further reuses the shared state across beams. Our analysis shows that OneLA achieves 1.54-2.46x end-to-end decode speedups while substantially reducing recurrent-state memory use and data movement.
[IR-8] ChronicleRec: Pre-training Temporally Anchored Tokens for Lifelong User Modeling
链接: https://arxiv.org/abs/2609.12375
作者: Chengkai Huang,Yubin Sheng,Liang Guo,Haoxi Liu,Junwei Pan,Shangyu Zhang,Zhixiang Feng,Chao Zhou,Chengguo Yin,Lina Yao,Haijie Gu,Jie Jiang
类目: Information Retrieval (cs.IR)
备注:
Abstract:Modeling ultra-long user behavior sequences is crucial for industrial recommendation and online advertising, yet directly feeding thousands of historical actions into ranking models is computationally prohibitive, while truncation discards long-range signals. Existing lifelong-interest methods retrieve target-relevant behaviors for each candidate, coupling long-sequence modeling with candidate scoring and repeated online cost. Recent target-independent compression methods enable cached user summaries, but often append query tokens at the sequence end and use bidirectional encoding, producing unordered and redundant summaries that overlook temporal structure. We propose ChronicleRec, a pre-train-and-transfer framework that compresses an ultra-long behavior sequence once into a chronologically ordered set of Chronicle Tokens. ChronicleRec applies a recency-aware multi-granularity merge, preserving recent behaviors while coarsening distant history. It then interleaves query tokens with the merged sequence and uses a causal encoder, so each query summarizes only the history before its temporal anchor. A multi-horizon design masks different recent-history windows across parallel branches to learn complementary long-range interests. The compressor is pre-trained with a mask-and-predict objective that reconstructs held-out recent behaviors from compressed older history, aligning historical signals with near-present intent. Since Chronicle Tokens are target-independent, they can be cached per user, decoupling ultra-long sequence modeling from online candidate scoring. Experiments on KuaiRand and Tencent AdLive show that ChronicleRec outperforms recent-window and single-pass compression baselines while approaching full-attention performance. Token analyses reveal temporally organized and complementary representations, and a seven-day online A/B test confirms significant production gains.
[IR-9] Recommendation Retrievers Need Verifiers: Universal Generative Reranking for Sequential Recommendations
链接: https://arxiv.org/abs/2609.12270
作者: Benyu Zhang,Qiang Zhang,Rui Li,Qunshu Zhang,Devansh Tandon,Neeraj Bhatia
类目: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI)
备注:
Abstract:First-stage recommenders in multi-stage systems produce a ranked candidate list from which a limited prefix is forwarded to downstream rankers. Because each forwarded item must be processed by more expensive ranking stages, this shortlist cannot be arbitrarily large. The first-stage objective is therefore high coverage of relevant items within the forwarded prefix, commonly measured by Recall@ k . A relevant item may be available deeper in the retrieved list but absent from the shorter prefix that is actually consumed. This paper studies post-hoc verification for promoting such candidates into the consumed shortlist without retraining or replacing the retriever. We introduce a lightweight generative verifier for retrieval models. Given a retriever state and a candidate item, the verifier scores the item through the likelihood of its identifier tokens. It is trained post hoc with next-token cross entropy, requires no sampled negatives or candidate pool during training, and scores only the retriever’s top- K candidates at inference. The interface is minimal: the retriever supplies a query state and candidate items, and the item representation can use any fixed tokenization. Across Amazon product recommendation and YaMBDa music recommendation, the same verifier training recipe improves Recall@10 for SASRec, GRU4Rec, NextItNet, and MiniOneRec. Ablations show that the improvements are not explained solely by injecting item-content features into the retriever, supporting verification as a post-hoc output-side adaptation mechanism. Subjects: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI) Cite as: arXiv:2609.12270 [cs.IR] (or arXiv:2609.12270v1 [cs.IR] for this version) https://doi.org/10.48550/arXiv.2609.12270 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[IR-10] EAR: Entity-Aware Partitioning Approach for Retrieval-Augmented Generation Development
链接: https://arxiv.org/abs/2609.12268
作者: Cenab Batu Bora,Oylum Alatlı,Sebnem Bora,Oguz Dikenelli
类目: Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注: 6 pages. Accepted and presented at the 2026 Innovations in Intelligent Systems and Applications Conference (ASYU 2026). To appear in the conference proceedings and IEEE Xplore
Abstract:Retrieval-augmented generation (RAG) can improve knowledge-intensive question answering, but the first design choice is easy to overlook: how should the source corpus be partitioned into retrievable units? Fixed-size chunks often return long passages whose relation to the question is only implicit. We introduce EAR, an Entity-Aware Partitioning approach for multiple-choice question answering (MCQA). EAR extracts normalized surface anchors from the question, answer options, and corpus; retrieves local windows around matching corpus anchors; and can attach a larger parent passage through an extractive summary. We evaluate EAR on a cleaned Massive Multitask Language Understanding (MMLU)-style subset of 153 questions selected by an automatic corpus-support heuristic and using decontaminated public textbook text. Across same-protocol top-k = 3 and top-k = 8 sweeps with Mistral, Gemma, and DeepSeek, EAR entity-window reduces retrieved words by 37.5-40.2% relative to chunks. Observed accuracy changes are +5.2, +1.3, and -3.9 points at top-k = 3, and +5.9, -3.3, and -4.6 points at top-k = 8; none of the entity-window differences is statistically significant. The scoped contribution is methodological: EAR provides a compact and inspectable retrieval unit, while its rule-based anchor extractor remains domain-specific and requires separate validation before transfer.
[IR-11] Extracting Dataset Mentions in Forced Displacement and FCV Documents: A Weakly Supervised Framework with LLM -Based Label Refinement
链接: https://arxiv.org/abs/2609.12107
作者: Rafael Macalaba,Aivin V. Solatorio,Patrick Michael Brock,Olivier Dupriez
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR); Econometrics (econ.EM)
备注: 22 pages, 1 figure
Abstract:Development and humanitarian organizations produce and support surveys, administrative registries, and other data resources to inform research, policy, and operations, yet systematically identifying where these datasets are referenced remains difficult. Such references are dispersed across research papers, project documents, humanitarian reports, and other unstructured text, limiting both the ability to trace data use and to identify potential gaps in data availability or dissemination. We present a weakly supervised framework for adapting dataset extraction to forced displacement and Fragile, Conflict, and Violence (FCV) documents without first constructing a large manually labeled training corpus. A lightweight model trained on general research literature generates candidate dataset mentions from unlabeled domain documents, which a frontier large language model (LLM) reviews in context, validating or rejecting candidates and correcting their extraction boundaries. The resulting annotations are supplemented with targeted synthetic and contrastive examples and used to fine-tune the lightweight model for large-scale extraction. We evaluate the resulting model on an independent gold-standard benchmark of 1,706 text passages spanning research, humanitarian, and operational documents. Across the full benchmark, the model achieves 74.1% precision and 70.5% recall at the mention level; among passages containing dataset references, precision reaches 89.5%. At the passage level, the model achieves 88.2% accuracy and 88.6% specificity in distinguishing passages with dataset references from those without them. These results demonstrate a practical approach for constructing domain-specific supervision when labeled data are limited, and provide a technical foundation for larger-scale analysis of data use and potential gaps in the displacement data landscape.
[IR-12] Cortex: Content Analysis Support Software a Resource for Qualitative Research
链接: https://arxiv.org/abs/2609.11970
作者: Ana Julia da Silva Soares,Rafael Coimbra Pinto
类目: Digital Libraries (cs.DL); Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注: 27 pages, 3 figures
Abstract:Qualitative research is widely used in the human and social sciences, characterized by a deep understanding of phenomena through the interpretation of meanings and contexts. Among qualitative data analysis methods, content analysis stands out as a consolidated technique, which allows for the systematic description and interpretation of textual contents. However, as data volume increases, the time required for organization, reading, and categorization becomes a significant challenge, potentially delaying research development. Therefore, this work aimed to develop a web application to support content analysis, based on Bardin’s methodology, targeted at academic researchers. The methodology adopted a mixed approach, combining bibliographic research on content analysis with semi-structured interviews with four experienced researchers, aiming to identify real needs and requirements. Based on these inputs, the Cortex software was developed to assist the researcher in the pre-analysis and material exploration stages, generating suggestions for indices, indicators, and categories with full traceability to original documents. The results demonstrate that Cortex is capable of guiding the researcher through the entire methodological workflow, from corpus configuration to results exportation, acting as a methodological collaborator without replacing the researcher’s interpretative autonomy.
[IR-13] InitGen: Candidate Generation for Interaction Initiation in Intelligent Assistants
链接: https://arxiv.org/abs/2609.11953
作者: Ruize Shi,Jinhua Chen,Hong Huang,Ziniu Chen,Ruike Zhang,Jianxun Shi,Yitao Chen,Rui Zhang
类目: Information Retrieval (cs.IR)
备注:
Abstract:Interaction initiation refers to presenting multiple candidate queries when a user opens an intelligent assistant before expressing any intent for the current session. In production, candidate generation incorporates dynamic context and produces all candidates within a strict latency budget. Learning from user feedback is also difficult since the generator usually produces more candidates than are finally displayed. After downstream filtering and ranking, only a subset is exposed to users, so the observed feedback is partial and cannot be reliably assigned to individual queries. We present InitGen, a framework for candidate generation that is deployed in the interaction initiation pipeline of OPPO’s Xiaobu Assistant. InitGen generates a set of candidate queries jointly and aligns the generated set with user feedback through weighted preference optimization. The sample weights are derived from user activity and downstream ranking scores. The activity weight reduces the dominance of highly active users during training, while the ranking score is used as a practical estimate of the reliability of the observed feedback. InitGen also uses a rolling window update strategy to incorporate recent interaction data into periodic model updates. In an online A/B test against a strong production baseline, InitGen improves the click-through rate from 0.95% to 1.61%, corresponding to a relative improvement of 69.1%, and increases query exposure by 17.9% under the same traffic allocation. InitGen generates the complete candidate set within 180 ms and has been fully deployed in OPPO’s Xiaobu Assistant, which serves over 150 million monthly active users.
[IR-14] MemRetriever: Learning to Search Reflect and Retrieve from Long-Term Memory
链接: https://arxiv.org/abs/2609.11951
作者: Ruiyang Jiang,Chunyu Li,Zhiyu Li
类目: Information Retrieval (cs.IR)
备注:
Abstract:Long-term memory enables personalized agents, but its value depends on retrieving the right evidence at the right time. Most memory systems use static top-k retrieval: they issue one query, return a fixed number of memories, and pass them directly to a downstream model. This approach can miss evidence distributed across sessions, introduce irrelevant content, and waste context, especially for multi-hop, temporal, and knowledge-update questions. We present MemRetriever, an agentic retrieval model that treats memory access as a multi-step search process. At each step, MemRetriever reasons over the current evidence and selects parallel search for broad exploration, serial search for targeted completion, or reflection and denoising for filtering and evidence assessment. It stops when the retained evidence is sufficient for downstream answering. We construct ReAct-style search-memory trajectories for supervised warm-start training and further optimize the model with Group Relative Policy Optimization. The reward design encourages evidence coverage, noise reduction, answer sufficiency, and efficient termination. Experiments on LOCOMO, LongMemEval, HotpotQA, MuSiQue, and 2WikiMultiHopQA show consistent improvements over static retrieval and supervised-only baselines. MemRetriever-4B-RL also outperforms DeepSeek-v4-Flash on the main LongMemEval retrieval metrics under the same pipeline and achieves the strongest results among the compared methods on MuSiQue. Because its decision logic is independent of the storage backend, MemRetriever can also operate over external knowledge bases and vector databases. These results show that an intermediate decision layer that plans, searches, filters evidence, and determines when to stop can improve both long-term memory retrieval and knowledge-intensive question answering.
[IR-15] Who Are We Recommending To? Recommender Systems in the Agent ic Web
链接: https://arxiv.org/abs/2609.11945
作者: Himan Abdollahpouri,Kyle Kretschman,Sai Ravindranath,Jackie Doremus,Mounia Lalmas
类目: Information Retrieval (cs.IR)
备注:
Abstract:For two decades, recommender systems have been designed under the assumption that a human directly consumes each recommendation: receiving, interpreting, and acting upon it. The emergence of AI agents powered by large language models challenges this assumption. In the emerging Agentic Web [ 28 ], autonomous agents increasingly act on behalf of users, e.g., browsing, comparing, negotiating, and executing transactions, raising a central question: who is the receiver of a recommendation? In this position paper, we argue that the recommendation paradigm is undergoing a bifurcation. In delegable contexts, such as routine purchases, travel, and constrained transactional tasks, the primary operational consumer of recommendations is shifting from the human to the agent, requiring new optimization objectives, interaction protocols, and evaluation criteria. In experiential contexts, such as entertainment, art, and other subjective or high-stakes choices, humans remain the final judge of relevance, though agents may assist through pre-filtering and curation. We introduce a delegation spectrum that characterizes recommendation contexts along factors such as preference specifiability, outcome verifiability, and decision stakes, and we outline a research agenda spanning agent preference modeling, dual-audience optimization, and the emerging agent attention economy. We further discuss the implications of this shift for the design and evaluation of recommender systems
[IR-16] PinDCO: Whole-Page Aware Dynamic Creative Optimization at Scale RECSYS26
链接: https://arxiv.org/abs/2609.11943
作者: Yu Hao,Yuchun Li,Peimeng Sui,Meilin Liu,Tianyuan Cui,Hao Li,Zicong Zhou,Akanksha Baid
类目: Information Retrieval (cs.IR); Machine Learning (cs.LG)
备注: Accepted by the Recsys26
Abstract:Recent advances in generative AI have substantially accelerated the creation of high-quality ad creatives, dramatically expanding the number of candidate variants per campaign. This shift increases the need for scalable dynamic creative optimization (DCO) systems that can match creatives to the most relevant audiences under stringent latency and cost constraints. We present PinDCO, a production DCO system for ad creative retrieval and selection on Pinterest, a billion-scale visual discovery platform. PinDCO is built around a Creative Component Fusion Network (CCFN) that performs dynamic creative scoring by modeling each creative component (e.g., image, title, layout) with a dedicated tower, using component-specific hyperparameters to account for differing modeling complexity. The component representations are fused to predict a creative-level score conditioned on the ad-level prediction, and we improve training data quality via an exploration-exploitation strategy. To account for Pinterest’s waterfall grid layout, where a creative’s rendered size affects nearby content and session-level engagement, we introduce a Pixel-aware Adjustment Module(PAM) that adjusts scores based on creative size to encourage efficient screen real-estate utilization and better whole-page outcomes. To support the large volume of creative candidates, we further employ a lightweight pre-selection model for early pruning, and optimize serving efficiency through caching and dynamic batching. Extensive offline analyses and online A/B experiments demonstrate the effectiveness of PinDCO, yielding a +3.09% lift in ad Click-Through Rate(CTR) with positive whole-page metrics. With the strong performance, we launched PinDCO in the Pinterest Ads platform.
[IR-17] Position: Recommender Systems Should Move Beyond Platform-Centric Ranking toward Personal Agent -Mediated Recommendation
链接: https://arxiv.org/abs/2609.11942
作者: Haohan Yuan,Peng He,Dan Zhang,Jianpeng Liang,Junning Zhu
类目: Information Retrieval (cs.IR)
备注: Position paper; 13 pages, 2 figures, and 3 tables. Introduces the PAMR paradigm, a mediation-centered evaluation framework, and a proof-of-concept study on recommendation tasks
Abstract:Recommender systems are usually framed as ranking systems: platforms observe users, construct candidate sets, and select items on their behalf. This framing hides a deeper allocation of control, in which platforms also determine candidate access, evidence boundaries, explanations, and the path from user need to recommended output. We argue that the next bottleneck in recommendation is not only preference modeling, but control over evidence acquisition and disclosure. We argue for \textbfPersonal Agent-Mediated Recommendation (PAMR), a paradigm in which a user-facing personal agent represents the user in discovering, filtering, aggregating, and governing recommendation evidence across distributed sources. The central shift is not simply from one ranking model to another, but from platform-side item ranking to user-side evidence mediation. As a position paper, we define PAMR as a new recommendation paradigm, establish its boundary criteria, identify its core mediation decisions, and propose a mediation-centered evaluation framework. A proof-of-concept study on hard Yelp restaurant recommendation tasks further shows that, under a shared LLM ranker, source selection and controlled disclosure provide the strongest observed utility–traceability–exposure–cost operating point.
人机交互
[HC-0] From Review to Reuse: How Post-Task Workflow Can Support Human-AI Agent Interaction
链接: https://arxiv.org/abs/2609.13136
作者: Zekun Wu,Xinru Wang,Rock Yuren Pang,Chenglong Wang,Anna Maria Feit
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:AI agents can automate tasks by turning a single natural-language request into a multi-step process spanning tools, files, and applications. Users are often left to judge that process from fragmented execution information and the final output. To make the completed process easier to understand, validate, and reuse, we investigate post-task workflows: editable, graph-based representations of an agent’s completed execution. We first analyzed 10,803 public workflow templates from n8n to characterize real-world automation practice, then developed Trace2Flow, a research probe that translates agent execution traces into interactive post-task workflows. In a study, participants (N = 20) reviewed agent executions with prompt or agent errors. We found that post-task workflows improved their understanding and error detection over a prompt-only condition, and that validation succeeded mainly when users cross-checked across multiple evidence sources. For follow-up tasks, adapting the workflow matched adapting the prior prompt in success, time, and difficulty, and was often preferred.
[HC-1] Involving before Evolving: A Vision for Trustworthy Enterprise Digital Twin Engineering
链接: https://arxiv.org/abs/2609.13071
作者: Kérian Fiter,Adil Lagrou,Franck Dervault,Bentley Oakes
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注: Accepted for publication at MODELS Companion 2026 (Málaga, Spain). 5 pages (4 pages + 1 page references)
Abstract:Enterprise Digital Twins (EDTs) promise data-driven decision support at organizational scale, but realizing them requires navigating siloed departments, tacit knowledge, and high-stakes decisions with long-horizon consequences. Existing approaches involve domain experts during model development but focus less on early organizational buy-in in EDTs. We present a vision for trustworthy EDT engineering grounded in an `involving before evolving’ paradigm: rapidly involving stakeholders through a working prototype before evolving toward federation and full interoperability. Our three-stage approach combines foundation models for rapid prototyping, an ontological backbone for federated interoperability, and observability tooling for stakeholder trust. We ground our vision in an ongoing collaboration with Michelin, a multinational manufacturer, where an initial prototype has helped support stakeholder buy-in.
[HC-2] Middleware for Feed Recommendation in Practice: How Feed Creators Build Maintain and Sustain Custom Feeds on Bluesky
链接: https://arxiv.org/abs/2609.12958
作者: Tony Zhou,Leijie Wang,Amy X. Zhang
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Scholars have long proposed third-party middleware as an alternative to centralized algorithmic feeds: feeds built and distributed by independent feed creators. This vision saw no large-scale instantiation until Bluesky, a decentralized microblogging platform, introduced custom feeds in 2023. Although central to the middleware ecosystem, we know little about how feed creators understand their role, build feeds, and sustain them. Through interviews with n = 26 feed creators and third-party developers of feed-building tools, and analysis of n = 88,302 custom feeds, we identify two creator orientations—utility-providing and community-building. Additionally, creators struggle to maintain feeds that fully realize middleware ideals: they lack granular interaction data, receive little feedback, and lack technical expertise to act on either. Finally, creators sustain their feeds as unpaid hobbyists with little platform support and are divided on whether to monetize beyond covering costs. We conclude with design and policy implications for strengthening the middleware feed ecosystem.
[HC-3] A Robot Among People:From Social Imitation to the Social Becoming of Human Groups
链接: https://arxiv.org/abs/2609.12937
作者: Victor Tuan Vu Pham,Judith Dörrenbächer,Thomas H. Weisswange,Marc Hassenzahl
类目: Robotics (cs.RO); Computers and Society (cs.CY); Human-Computer Interaction (cs.HC)
备注:
Abstract:Robots designed to mediate human groups often fall into the solutionist trap: they are framed as sociable agents that fix problems such as conflict, disengagement, or lack of coordination. We suggest a different way of thinking. Rather than discrete agents, robots can be understood as situated elements of shared environments; catalysts and carriers of group experience whose meaning emerges through how people position, interpret, and interact with them. From this perspective, robots are not there to repair some ostensible dysfunctionality, but to enable group-level sense-making around care, norms, and identity. Our prior work on robotic street furniture suggests that this does not happen by imitating human sociality but by taking the shape of deliberately constrained, group-facing entities that happen and act for \textitus without being socially entangled as one of us. We thus understand robots in public spaces not in terms of autonomy or intelligence, but as a relational capacity. This implies designing robots not in our image or for our utility, but grounded in our needs in being and becoming together.
[HC-4] NeuroClick: Preserving Surgeon Autonomy through Hands-Free Earable Tooth-Click Control in Neurosurgery
链接: https://arxiv.org/abs/2609.12910
作者: Jonas Hummel,Maximilian Burzer,Clara Sayffaerth,Valeria Zitz,Amir El Rahal,Michael Küttner,Tobias Röddiger,Jürgen Beck,Michael Beigl
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Neurosurgeons frequently interact with operating room (OR) technologies while sterility and occupied hands constrain control. We introduce earables as a direct, hands-free control platform for neurosurgery using tooth-click input. Formative OR observations and interviews with 10 domain experts grounded the design. Using OpenEarable 2.0 data from 12 participants, we developed a real-time recognition pipeline whose classifier achieved a median macro F1-score of 98.6% under leave-one-subject-out cross-validation. We evaluated the technique with 20 neurosurgeons during a simulated resection task in a neurosurgical OR. Participants reported few focus shifts and rated Earable favorably for workflow integration and perceived safety. Autonomous microscope control was rated significantly higher with Earable than Delegation, whereas Delegation enabled faster task completion under continuous assistant availability. Workload, usability, and task errors showed no significant differences. Preferences depended on training, reliability, context, and assistant availability. Earables thus add a direct, hands-free option for controlling selected functions alongside established workflows.
[HC-5] Unified Agent ic Video Editing Across Levels of Complexity and Creativity
链接: https://arxiv.org/abs/2609.12769
作者: Surabhi S. Nath,Kim Ferres,Milan Petrović,Lion Schulz
类目: Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC); Multimedia (cs.MM)
备注:
Abstract:Editing is a core component of video production, requiring creative planning and decisions under multiple constraints. Here, we report methods for agentic tooling for automated video editing across three tasks varying in editorial goal, complexity and creativity, namely scene previews, video summaries and cinematic trailers. We evaluate the outputs and discuss implications for automation and agency.
[HC-6] Enabling and Understanding Personalization in AI-Generated Advertising Imagery
链接: https://arxiv.org/abs/2609.12697
作者: Victor Kolominsky-Rabas,Leopold Müller,Claudius Budcke,Claas Christian Germelmann,Niklas Kühl
类目: Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注:
Abstract:Personalized marketing traditionally matches static products to customers, while dynamic creative optimization focuses mainly on AI-driven text personalization or basic product image modifications. We address this gap by developing and implementing an AI-based framework that generates personalized advertising imagery directly from customer data. We evaluate this framework in a two-stage within-subject study with N=100 participants across four products and three levels of personalization, varied by the amount and specificity of customer data used. Participants rated each image on attitude toward the advertisement, attitude toward the product, and purchase intention. Results show that participants perceive differences across personalization levels and evaluate AI-generated advertising imagery most positively at a moderate level of personalization. High personalization increases perceived personalization, which is positively associated with all three outcome measures, but also increases perceived creepiness, which is negatively associated with the outcomes and dominates the total effect.
[HC-7] Understanding Game Coaching on Gig Platforms
链接: https://arxiv.org/abs/2609.12695
作者: Hwijoon Lee,Saiph Savage
类目: Human-Computer Interaction (cs.HC); Computers and Society (cs.CY)
备注: Accepted to CHI Play 2026. Waiting for publication
Abstract:Freelance game coaches monetize their gaming expertise by offering personalized instruction to players seeking to improve, working through gig platforms, yet little is known about how they operate. To address this gap, we conducted semi-structured interviews with 20 experienced freelance coaches across 17 competitive games on Fiverr. Despite lacking shared formal training, these coaches converged on similar practices centered on rapport-building, individualized diagnosis, and adaptive feedback. We identify two structural conditions shaping this work: dual precarity, in which coaches navigate both gig platform instability and the lifecycle volatility of live-service games; and earned authority, in which coaches must continually establish legitimacy through visible competitive achievement within the same gaming spaces as their students. These coaches welcomed AI for administrative and analytic support but resisted its use in live interactions where trust, relational engagement, and situated judgment remained central. We discuss implications for Games HCI and the design of computational coaching systems.
[HC-8] I Am AdMan: A Pipeline for Automatic Generation of Personalized Advertising Imagery
链接: https://arxiv.org/abs/2609.12694
作者: Victor Kolominsky-Rabas,Leopold Müller,Claudius Budcke,Niklas Kühl
类目: Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注:
Abstract:Personalized marketing can increase customer engagement, satisfaction, and conversion. While existing personalization approaches have become effective at matching the right product to the right customer, the visual representation of advertisements remains generic and only weakly tailored to the individual. Prior research shows that generative artificial intelligence can improve the creation of personalized advertisements, particularly for text, and that image generation models can support scalable advertisement production. However, little research has examined how detailed customer information can be systematically translated into fully AI-generated, personalized advertising imagery at scale on a technical level. To address this gap, we propose AdMan, a multi-agent pipeline that transforms customer data into personas, generates personalized advertisement images conditioned on product reference images, and applies an LLM-based judge agent for automated quality control. We implement the pipeline with two different model configurations and evaluate it across four products, using six celebrity personas for qualitative inspection, and 100 real customer profiles, producing 1745 advertisements. The evaluation combines a qualitative expert focus group and a quantitative artifact-rate assessment. The results show that the pipeline can generate photorealistic and personalized advertisements. At the same time, performance varies substantially by product complexity and model configuration. Our findings extend the literature on AI-based personalized advertising by demonstrating the feasibility and current limitations of fully automated image generation for advertising.
[HC-9] Generative AI Use Cases In Real Estate Marketing: Adoption and Constraints in Germany
链接: https://arxiv.org/abs/2609.12684
作者: Victor Kolominsky-Rabas,Leopold Müller,Felicia Perpina,Niklas Kühl
类目: Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注:
Abstract:Generative artificial intelligence (GenAI) is changing how work is organized and performed. Real estate marketing is a prime example of this, yet evidence of GenAI in real estate agents’ day-to-day practice remains scarce. In this work, we report on our insights from a German-based empirical study with eleven semi-structured interviews. GenAI is already utilized across different activities, with marketing communication being the most prominent. Concrete use cases are emergent and unevenly adopted, with writing exposé texts being the only widely established one. Interaction is predominantly human-in-the-loop: GenAI drafts, structures, and retrieves, while real estate agents curate, verify, and decide. Constraints stem less from model capability than from integration with listings and documents, data availability, and compliance in sensitive tasks. The study contributes a grounded map of existing and potential use cases and identifies tentative practical implications for adoption.
[HC-10] Reconstruction and Reflection of Positive Experiences through Resurfacing Laughter-indexed Everyday Moments
链接: https://arxiv.org/abs/2609.12642
作者: Jun Fang,Jiajin Li,Yuntao Wang,Kexin Miao,Susu Wang,Xiaoyu Xie,Kaixin Ji,Yuanchun Shi
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Positive everyday moments often escape deliberate recording, while continuous self-tracking can generate extensive records that are difficult to revisit. We explore laughter as a naturally occurring, sparse index for constructing contextualized personal records to support later reconstruction and reflection. A formative study with 12 participants characterized laughter as an affective but semantically incomplete index and informed \textitLaughAnchor, a mobile and wearable self-tracking system. During participant-initiated recording, the system assembles detected laughter and aligned context into candidate moments for later reconstruction and reflection, with layered context disclosure, user-controlled curation, and near-term and long-term resurfacing. In a three-week field deployment with 12 participants, passive indexing preserved moments they considered unlikely to record deliberately but valued retrospectively. During resurfacing, participants attributed affective re-experiencing to laughter and used additional context both to reconstruct episodes and to explore already-recalled experiences. Across moments and reviews, resurfacing supported rediscovery and broader awareness of relationships, routines, and emotional states. These findings inform self-tracking designs that use sparse affective indices to organize contextual records for reconstruction and reflection, while keeping interpretation and retention under user control.
[HC-11] raceMind: Predicting User Information Uptake from Low-Cost Interaction Traces during Human-LLM Content Co-Generation
链接: https://arxiv.org/abs/2609.12600
作者: Yu Mei,Fengyou Zu,Ruiwen Zhang,Jie Cai,Chang Liu,Zhoutong Ye,Chun Yu,Yuanchun Shi
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:In human-LLM content co-generation, AI-generated information can enter final artifacts without being adequately processed by users, creating risks when artifacts are shared or acted upon. We study whether recognition-level uptake of atomic information units can be assessed in open-ended co-generation and predicted from low-cost interaction traces. We collected data from 62 participants across three tasks. For each final draft, we extracted atomic information units and generated post-task recognition questions, yielding 1187 unit-level uptake labels. We present TraceMind, which tracks units across Chat and Draft histories, aligns interaction traces with changing on-screen layouts, and models spatial, temporal, and workflow-informed evidence. TraceMind outperformed all learned baselines across AUROC, AUPRC-non, balanced accuracy, and macro-F1. We found that uptake unfolds throughout interaction, with sustained active engagement providing informative evidence beyond isolated signals. Our work shifts human-LLM co-generation from content adoption toward what users actually take up, motivating uptake-aware systems grounded in low-cost interaction traces.
[HC-12] Calibrated Ambiguity in Multimodal Language Models: Humans reach for cultural references while models describe the picture
链接: https://arxiv.org/abs/2609.12575
作者: Cody Kommers,Mingrui Ye,Evelyn Gius,Daniela Mihai,Hoyt Long,Zheng Yuan,Drew Hemment
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注: Kommers and Ye contributed equally to this research
Abstract:Ambiguity is often treated as a bug for AI systems to resolve—but in human communication and culture, ambiguity can also be a generative resource. From humour to politics to art, people express themselves in words and images that are open enough to invite different interpretations, yet constrained enough to be interpretable. We operationalise this notion of calibrated ambiguity with a task drawn from the parlour game Dixit. We compare differences in clues generated by human vs multimodal language models, based on a novel coding rubric for calibrated ambiguity, and find that models consistently exhibit ambiguity collapse (i.e., their outputs are over-specified, leaving no room for multiple legitimate interpretations). Unlike human clues, AI-generated clues also exhibit cultural flattening; they almost never make reference to culturally-situated knowledge, even when prompted to use allusion and figurative language.
[HC-13] KAD-Net: Kinematics-Aware Decoupled Learning for Robust 3D Hand Pose Estimation from a Single Depth Image
链接: https://arxiv.org/abs/2609.12559
作者: Jun Lu,Zhenming Chen,Lin Chen,Kanlun Tan,Xiaoling Li,Qiao Liu
类目: Computer Vision and Pattern Recognition (cs.CV); Human-Computer Interaction (cs.HC)
备注:
Abstract:Due to the complexity of hand kinematics and self-occlusion, existing 3D hand pose estimation methods based on single depth images struggle to comprehensively model the topological dependencies among hand joints. Furthermore, traditional hierarchical multitask architectures enforce a shared feature space for both 2D joint localization and depth estimation, which can induce mutual interference. To address these challenges, we propose a Kinematics-Aware Decoupled Learning Network (KAD-Net) for robust 3D hand pose estimation. Specifically, we first design a Finger Topology Constraint (FTC) module to enhance the representation of distal joints. This module utilizes three consecutive finger joints to construct a local kinematic representation to impose topological constraints, which supplements the kinematic features of the distal joints. The FTC module leverages the structural context from visible joints to assist in locating occluded distal joints, thereby improving robustness to occlusion. Additionally, we propose a task-decoupled hierarchical multitask framework. This framework separates 2D joint localization from depth estimation and incorporates a dedicated multitask learning strategy for depth regression, effectively isolating the UV and depth features to mitigate mutual interference and negative transfer. Extensive experiments demonstrate that KAD-Net outperforms existing methods on several benchmark datasets (ICVL, NYU, and MSRA), achieving state-of-the-art accuracy in 3D hand pose estimation. Potential applications of KAD-Net include human-computer interaction, virtual reality and gesture-based control systems.
[HC-14] he House with a Million Windows: Interactive Fiction for Narrative Restorying
链接: https://arxiv.org/abs/2609.12537
作者: Cody Kommers,Sarah G Immel,Drew Hemment,Mina Lee
类目: Computation and Language (cs.CL); Human-Computer Interaction (cs.HC)
备注: Kommers Immel contributed equally to this research
Abstract:AI-assisted writing can flatten meaning in human storytelling, enabling the production of homogeneous outputs without the intentional effort and sense-making writing entails. To address this challenge, we present The House with a Million Windows (HWAMW), an LLM-based interactive fiction system designed to help users explore both the breadth and depth of potential meanings within their personal stories – drawing on a psychological paradigm called the restorying intervention. In HWAMW, users play through a text-based narrative in which they tell a story, then encounter a set of LLM-generated “windows” reframing it according to different literary styles. Empirical evidence shows that HWAMW increases users’ sense of narrative identity, while an expert review explores how this effect is achieved. Our findings suggest that HWAMW facilitates restorying and offers a valuable paradigm for AI-assisted writing, wherein LLMs do not tell our stories but rather help us see greater potential in the stories we tell.
[HC-15] When2Talk: When Should a Proactive In-Car Agent Talk?
链接: https://arxiv.org/abs/2609.12503
作者: Kaiser Hamid,Peihang Li,Nade Liang
类目: Human-Computer Interaction (cs.HC); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Proactive in-cabin agents can help passengers understand automated-vehicle (AV) behavior, but communicating every ride event may introduce unnecessary interruptions. We investigated how communication should adapt to event priority and passenger activity. In a mixed-methods within-subject study, 41 participants rode as passenger in a VR simulated fully-automated vehicle. We compared an event-triggered (ET) policy that communicated immediately at every event with a context-sensitive (CS) policy that selected \textitImmediate, \textitDelayed, or \textitSilent communications. CS increased communication appropriateness and substantially reduced perceived interruption. Perceived trust did not differ between policies, although baselines dispositional trust differentiated communication preferences. Findings highlight event consequence, passenger activity, continuing information value, and confirmation need as key considerations for selective in-cabin communication.
[HC-16] When Does AI Augment Work? A Workflow-Level Framework for Human-Agent Collaboration
链接: https://arxiv.org/abs/2609.12482
作者: CIVIC-AI Collaboration:Jiaying Wu,Caleb Ziems,Raymond Chan,Nancy F. Chen,Corlyss Chua,Gerard Chung,Jungpil Hahn,Wee Sun Lee,Zhengyuan Liu,Jamie Ng,Desmond C. Ong,Jeryl Ong,Da Ren Soon,Tianqi Song,Zhi-Xuan Tan,Sixing Tao,Emily Yang,Yajing Yang,Stella Xin Yin,Min-Yen Kan,Diyi Yang
类目: Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Human-Computer Interaction (cs.HC)
备注: 8 pages. Whitepaper from the CIVIC-AI 2026 workshop
Abstract:We aim to characterise the value of artificial intelligence in the workplace. Current studies largely measure this value in terms of the current automation capabilities and public adoption of AI. However, such metrics ignore the greater impacts of human–agent collaboration in transforming the nature of work. To account for this, we must expand the scope of our analysis beyond atomised tasks of today, and instead focus on how AI can augment entire workflows of the future. To ground this analysis, we establish a precise definition of AI augmentation comprising six conditions, spanning durable net value, meaningful human control, accountability and recovery, and long-term human development through learning, career pathways, and job purpose. We elaborate on these conditions and apply the framework in a case study of AI-mediated social surveys. We conclude by outlining how organisations, researchers, and government leaders can use this framework to make sense of the future of work.
[HC-17] From the Task Boundaries of Narrative Text to Structural Anchoring Uncertainty Triggers and Cross-Calibration
链接: https://arxiv.org/abs/2609.12453
作者: Bowen Deng,Jiaqi Zou,Kexin Zhang,Daifeng Li
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Causal graphs represent structural relationships among variables, yet users must still interpret direction, mechanism, and adjustment conditions in relation to the task at hand. Prior work often compares explanation formats as fixed conditions and pays less attention to how users distribute reasoning across graphs, direct explanations, and stories. We developed CoNS-Explorer, which uses reviewed instructional DAGs/SCMs to maintain a shared causal-fact ledger and generate fact-matched direct explanations and contextualized stories. A controlled survey experiment ( N=240 ) compared the two texts as complete presentation packages. In the primary GLMM, the Story condition had a positive but uncertain overall association with accuracy (OR =1.55 , 95% CI [0.34,7.10] , p=.572 ); a population-averaged GEE showed a significant positive effect (OR =1.89 , 95% CI [1.02,3.48] , p=.042 ). Task-type interactions localized the clearest advantage to total-effect adjustment. Story also significantly increased situational presence. In a separate system-task and interview study ( N=24 ), participants freely used graphs, direct explanations, and stories across three causal models. They established structural anchors with graphs and numerical results, consulted text when direction was unclear, mechanisms were unfamiliar, or multiple paths competed, and checked their judgments against other representations or external evidence. Integrating the two studies, we develop a process framework of structural anchoring, uncertainty triggering, explanation routing, and cross-calibration, together with four testable design propositions for adaptive causal explanation.
[HC-18] ForkSCOPE: Charting the Agent ic Garden of Forking Paths
链接: https://arxiv.org/abs/2609.12438
作者: Arjun Balaji,Batuhan Duru Yeltekin,Tian Zheng
类目: Human-Computer Interaction (cs.HC); Applications (stat.AP)
备注:
Abstract:Even with a fixed dataset and research question, data analysis involves many defensible decisions. Understanding how these choices influence the results is scientifically important but remains challenging. Crowdsourcing and agentic AI can generate hundreds of end-to-end analyses, but scaling generation alone can create a processing bottleneck and an analytic ``black hole.‘’ A common workaround is to impose a shared fixed decision taxonomy, which can limit insight and understate uncertainty. We present ForkSCOPE, a human-AI collaboration framework that induces structure bottom-up from the code corpus of end-to-end analyses, without a taxonomy fixed before or after generation, so the organization and evaluation of the garden can scale with the corpus. ForkSCOPE surfaces the charted garden of forking paths through a human-AI collaboration pipeline and an evidence-linked interactive viewer for steering and verification: it spotlights organically identified forks and structures and produces a derived taxonomy and decision map compatible with existing multiverse tools.
[HC-19] Why User Studies and Participant Experience Reporting Matter for VR Motion Privacy?
链接: https://arxiv.org/abs/2609.12415
作者: Azim Ibragimov,Eric D. Ragan
类目: Human-Computer Interaction (cs.HC); Cryptography and Security (cs.CR)
备注:
Abstract:Public VR game leaderboards contain tracked motion recordings uploaded by hundreds of thousands of users. Once uploaded, these recordings are accessible to anyone and create privacy risks (i.e., identification and profiling). Prior work has proposed mechanisms that modify tracked movement to reduce these risks. Their utility is commonly evaluated through physical deviation, where smaller deviations indicate better utility, while user studies are less common. However, it remains unclear how well physical deviation explains users’ acceptance of a mechanism compared to user studies. We examine this through a user study of three VR motion privacy mechanisms at five physical deviation levels. We find that user studies explain substantially more variation in mechanism acceptance than physical deviation, although physical deviation remains significant. We also find that prior VR experience and exposure to VR privacy mechanisms significantly affect acceptance. We recommend combining physical deviation with user studies and reporting participants’ prior experience.
[HC-20] Context-Aware Causal Gaze Forecasting for Human-Vehicle Interaction During In-Cabin Tracking Dropouts
链接: https://arxiv.org/abs/2609.12374
作者: Shabnam Shabani,Ghazal Farhani
类目: Computer Vision and Pattern Recognition (cs.CV); Human-Computer Interaction (cs.HC)
备注: 8 pages, 5 figures
Abstract:Dashboard-mounted gaze trackers often lose sight of the driver’s eyes during large head rotations, including shoulder checks, mirror glances, and intersection scanning. These maneuvers occur when information about the driver’s visual attention is most useful. Offline gap-filling methods may reconstruct a missing interval using observations from both sides, but an online driver-monitoring system cannot rely on measurements that have not yet occurred. We therefore formulate causal gaze recovery: forecasting unavailable gaze at time t without target-tracker gaze at t or later. We introduce the Causal Context-Gated Forecaster (CCGF), which encodes a 60-frame pre-dropout history of gaze and head pose and combines it with DINOv3 scene features. A learned reliability gate controls the contribution of the history and scene representations as the dropout progresses. We evaluate two scene conditions: Live, in which the scene representation continues to update during tracker loss, and Frozen, in which the final pre-dropout representation is used throughout the missing interval. We evaluate CCGF on 2,047 eligible, naturally occurring GazeSense head_lost events drawn from 10.5 h of naturalistic driving by ten drivers. Across all recordings, head_lost accounts for 8.5 percent of GazeSense recording time. Synchronized gaze coordinates from a head-mounted Neon tracker provide supervision and evaluation targets but are never used as model inputs. Under leave-one-driver-out evaluation, CCGF achieves a mean per-driver median error of 175.7 px (10.5 deg) with Live scene updates, a 33 percent reduction relative to history-only causal forecasting. With Frozen scene input, the error increases to 210.8 px (12.9 deg), indicating that scene observations acquired during the dropout provide useful predictive information. We will release the dataset, evaluation protocol, and causal baselines.
[HC-21] LatentVerse: A Framework for Understanding Shared and Modality-Specific Information in Multimodal Latent Representations ML4H
链接: https://arxiv.org/abs/2609.12364
作者: Majd Alafrange,Samuel Friedman,John Kitonyo,Sana Tonekaboni,Mahnaz Maddah
类目: Machine Learning (cs.LG); Human-Computer Interaction (cs.HC)
备注: 27 pages, 7 figures, 2 tables. Code: this https URL
Abstract:Latent embeddings have become a central data abstraction in modern machine learning, especially in biomedicine, where foundation models are increasingly used to encode multimodal data like clinical text, medical images, omics, and physiological signals. However, the utility and value of these representations depends on understanding their quality, structure, and the information they encode. Existing analysis workflows for evaluating representations remain fragmented across custom scripts, isolated metrics, and most importantly lack multimodal analysis, limiting accessibility and reproducibility. We present LatentVerse, a representation analysis resource that combines a web-based visual analytics platform for accessible, report-driven exploration with a command-line interface for scalable technical workflows. LatentVerse unifies diagnostics for various representation quality metrics and extends to multimodal settings by decomposing embeddings into shared and modality-specific components. We evaluate LatentVerse through controlled unimodal and multimodal simulations, discovery-oriented analyses on real biomedical embeddings, and a user study across diverse use cases. By supporting thorough and interpretable evaluation of latent spaces, LatentVerse makes foundation model representations more understandable in biomedical and data science applications.
[HC-22] “I Felt Very Seen But Still Very Alone”: Longitudinal Trajectories of General-Purpose LLM Use for Socioemotional Support
链接: https://arxiv.org/abs/2609.12314
作者: Meryl Ye,Briana Vecchione,Livia Garofalo,Ranjit Singh
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:People increasingly use general-purpose chatbots such as ChatGPT, Claude, and Gemini for mental health and emotional support. We report a multi-stage longitudinal qualitative study of 18 U.S. adults, conducted from April to December 2025, combining initial interviews, a four-week diary study, focus groups, and exit interviews. We find that socioemotional use often emerged gradually out of practical use and when other forms of support were unavailable. Participants developed routines and boundaries around chatbot use, which were disrupted by model updates, evolving public discourse about AI harms, and changes in personal circumstances. We demonstrate how longitudinal study captures factors beyond the human-AI dyad, and argue that HCI researchers and designers should account for users’ histories with their chatbots and broader care ecologies when evaluating AI systems over time and introducing updates that may disrupt established sources of support.
[HC-23] “People can change and patterns can be broken”: Contextualizing Tradeoffs in Automated Decision-Making Systems
链接: https://arxiv.org/abs/2609.12288
作者: Rabeya Bosri,Anna Harbluk Lorimer,Afrida Hossain,Vasisht Duddu,Bailey Kacsmar
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Automated decision-making (ADM) systems are increasingly deployed in domains such as mortgage lending, prison sentencing, health insurance coverage, and hiring. Designing a responsible ADM system in such high-stakes domains requires ensuring privacy protection, fairness across demographic groups, and robustness against adversarial manipulation. However, prioritizing one of these objectives comes at the cost of another, forcing a choice as to which tradeoff to accept in a deployment. These tradeoffs explicitly or implicitly impact the life, safety, and fundamental rights of the people in a society, and thus, the perceptions and priorities of this population are needed before we can produce appropriate solutions. To this end, we conducted a quasi-experimental study (N = 777) in which participants evaluated four decision-making scenarios with controlled tradeoffs. Participants significantly preferred human decision-making (HDM) over ADM in three of four scenarios, emphasizing the value of human judgment, contextual understanding, and the ability to incorporate non-quantifiable factors. Furthermore, in terms of tradeoffs, our findings not only show that participants’ preferences are highly context-dependent, but also that their perception of a specific objective, fairness, extends beyond formal definitions. Participants interpret fairness through multiple lenses, including privacy risks and susceptibility to manipulation, and view unfair or manipulated outcomes as failures of accuracy. Overall, our findings highlight the importance of context-aware and human-centered approaches when designing and governing ADM systems in high-stakes situations. Rather than purely technical objectives, it is essential to evaluate ADM systems based on how their tradeoffs align with specific expectations within a given domain, as well as with societal values and perceptions of harm and fairness.
[HC-24] Synthetic TLX: Forecasting Human Workload Using Agent Simulation
链接: https://arxiv.org/abs/2609.12273
作者: Tzu-Sheng Kuo,Carrie J. Cai,Meredith Ringel Morris,Michael Terry
类目: Human-Computer Interaction (cs.HC)
备注:
Abstract:Assessing human workload for technology-mediated tasks helps prevent task failure caused by poor technology design. Traditionally, workload is assessed retrospectively using the NASA Task Load Index (TLX) after humans complete a task. What if we could forecast workload before a human attempts a task using agent simulation? We introduce Synthetic TLX, a new paradigm for proactive workload estimation that predicts NASA TLX scores for a given task, unlocking novel interaction opportunities and evaluation methods. To understand its viability, we conducted three experiments comparing human and agent-generated scores to evaluate where they align and diverge. We found agent estimates align with human scores particularly when prompted with a human persona and active task simulation. However, agents and humans diverge in the sources of workload they are sensitive to. Based on our findings, we present three applications to showcase Synthetic TLX’s potential and discuss the future of workload-aware human-AI interaction.
[HC-25] act: A Zero-Cost Browser-Based Pipeline for On-Demand Tactile Braille Storybooks
链接: https://arxiv.org/abs/2609.12272
作者: Iliano Fasolino
类目: Human-Computer Interaction (cs.HC); Computers and Society (cs.CY)
备注: 17 pages, 7 figures. DOI: https://doi.org/10.5281/zenodo.22698645
Abstract:Braille literacy among blind school-age children has fallen sharply, in part because producing illustrated braille pages still requires specialized software and trained labor. We present Tact, a browser-based pipeline that converts a spoken or typed story idea into printable braille with a matching raised tactile illustration, without an account or mandatory cost and with an offline-capable path. The paper documents the engineering history of the system: its sighted-operator ethical model; hardware rationale for consumer fused-deposition modeling; physical braille geometry and printer calibration; local, hosted, and fallback language-model paths; a deterministic Grade 1 braille translator; verified page layout and pagination; a 93-shape hand-drawn tactile illustration library; and synthesized sound design for a voice-first interface. We report engineering verification, ethical commitments, limitations, and the work required before the system is ready for real blind and low-vision readers.
[HC-26] BRIDGE-EEG: Bridging Self-Supervised Pretraining and Efficient Deployment for Cross-Dataset EEG Classification
链接: https://arxiv.org/abs/2609.12218
作者: Meghna Roy Chowdhury,Chengwei Zhou,Haotian Yu,Gourav Datta,Shreyas Sen
类目: Human-Computer Interaction (cs.HC); Machine Learning (cs.LG); Signal Processing (eess.SP)
备注: 11 pages, 7 figures 8 tables, journal submission
Abstract:The growing use of electroencephalography (EEG) motivates automated analysis that is accurate, transferable, and deployable on constrained hardware. Recent EEG foundation models learn general representations from large-scale pretraining, but their size and computational cost limit edge and wearable deployment. We introduce BRIDGE-EEG, an efficient multi-task EEG classification pipeline that preserves the benefits of pretraining while reducing model size. A unified preprocessing scheme maps heterogeneous recordings with different channel counts, montages, and sampling rates to a device-agnostic 62-channel time–frequency representation. We pretrain an SE-ResNet18 teacher (11.84 M parameters) with SimCLR on unlabeled EEG from five heterogeneous datasets, then compress it into SE-ResNet8 (1.56 M) and SE-ResNet4 (0.48 M) students using task-agnostic and task-specific distillation. We evaluate six benchmarks spanning abnormality detection, motor imagery, and emotion recognition. For abnormality detection and emotion recognition, the students achieve accuracy comparable to or better than several recent EEG foundation models with 10–1,000 \times more parameters. Motor imagery shows a remaining representation gap, highlighting the importance of pretraining diversity. Inference profiling on a server GPU, desktop CPU, and NVIDIA Jetson Orin Nano shows up to 3.0 \times lower edge energy per inference (15.64 mJ vs. 46.67 mJ). The compact models further support future deployment on MCU-class wearables.
[HC-27] Plans They Abandon Reports They Author: The Narrative Layer of Autonomous Agents
链接: https://arxiv.org/abs/2609.12205
作者: Obada Kraishan,Kulsawasd Jitkajornwanich
类目: Human-Computer Interaction (cs.HC)
备注: 14 pages, 4 figures. Under review
Abstract:When a coding agent finishes a task, the developer reviews a summary the agent wrote about itself, not a display someone designed. We ask how much of the agent’s work that summary carries, and whether it drifts toward the plan the agent stated when execution departed from it. Across 5,851 real developer sessions and 355,942 tool calls, a self-report referred to about one action in eleven, and a reader working from the report alone recovered roughly a fifth of the action log. Neither figure depended on whether the session later needed human correction. Reports did not generally resemble the stated plan more than the executed one, but they did so increasingly as execution diverged from the plan. We hand-validate both measurement steps that use a language model, report the one that failed alongside the one that passed, and draw conclusions only from measures that survived.
[HC-28] MAIA: Multi-Agent Intent Articulation for Requirement Discovery in Art Commissions
链接: https://arxiv.org/abs/2609.12097
作者: Yu-Chao Wang,Yanhong Lu,Yingjie Victor Chen,Tim McGraw
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI)
备注:
Abstract:In bespoke art commissions, laypeople know what they feel but lack the words to specify it: one participant wanted a laid-off truck driver depicted as “a ghost in his own machine” but left the medium, scale, and palette unsaid. We frame this as an articulation bottleneck at an under-served upstream stage: requirement discovery, which precedes any artist or image generator and forces the commissioner to constitute intent in the first place. We present MAIA (Multi-Agent Intent Articulation), a multi-agent system that scaffolds this stage through Socratic inquiry under a “Verification over Invention” rule, turning vague affect into a text-only brief of visual terms the user verifies. In a within-subjects study (N = 16), the full configuration produced a large, significant gain in Cognitive Support over a minimal baseline (r = 0.96, p_FDR = 0.015; LMM p_FDR 0.001). Thematic analysis traces the same mechanism, and a validator gate structurally blocks unratified content. A complementary blind review by three professional concept artists on a sampled set of briefs corroborates this improvement from the artist’s side: AI rewriting improved visual completeness and executability in all eight sampled tasks (task-level Wilcoxon p = 0.008; FDR q = 0.010), with directionally larger gains under MAIA than under the baseline (underpowered, d = 1.4-2.6).
[HC-29] Creating an Atomic User Model for Personality-Aware Large Language Model Interaction
链接: https://arxiv.org/abs/2609.12086
作者: B. Sankar,Deepthika S,Pawni Yadav,Amogh A S
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 59 pages, 22 figures, 24 tables
Abstract:Assistants built on large language models are expected to write as their user would, and the dominant approach is single-channel: preferences summarised from conversation history and reinserted into context. This inverts the order of inference. Preferences are the task-dependent surface of a comparatively stable personality structure, so a system storing only preferences relearns the person whenever the task changes. First, we characterise personality seepage, where a prompt’s linguistic surface carries a personality fingerprint the assistant mirrors without access to the personality behind it. Second, we propose the Atomic User Model (AUM), a human-readable representation organising a person as a stable identity nucleus with four interpretable shells (psychological, cognitive and experiential, behavioural, and social), plus cross-shell entries recording internal conflict and authenticity. Third, we treat AUM as a retrieval index over a person rather than a prompt prefix, with a pipeline where a task classifier, component-selection function and budgeted retriever return a small payload of fields at generation time. Fourth, we evaluate it with sixteen language-model-simulated participants, six style-sensitive tasks and three seeds, plus a synthetic scaling study of the retriever. Retrieving eight fields matched the style fidelity of the full user model on 23% of the context (211 tokens against 915), improved on flat preference notes by 0.24 points on a five-point scale (p 0.001, dz = 0.50), and raised forced-choice identification of the participant’s own voice from 14.9% to 42.7% (25% chance). Four pre-registered controls returned null, locating the effect in the representation rather than the search over it. The benefit is largest for participants the un-personalised assistant reproduces worst (rho = -0.61, p = 0.013): personalisation is worth most to those the default serves least.
[HC-30] “The Only Thing Certain About This is Uncertainty”: Exploring Informal Care Coordination Practices Among Older Adults with Mild Cognitive Impairment
链接: https://arxiv.org/abs/2609.12070
作者: Josey M. Benandi,Niharika Mathur,Sangha Park,Tracy L. Mitzner,Elizabeth D. Mynatt,Agata Rozga
类目: Human-Computer Interaction (cs.HC)
备注: To be published at ACM CSCW 2026 in October 2026
Abstract:Older adults aging in place often have informal support systems to help them maintain independence and quality of life. As they age, many older adults deal with the onset of Mild Cognitive Impairment (MCI), which introduces a new set of functional and cognitive changes that affect their ability to manage daily routines. The approach to arranging and coordinating support for everyday activities for older adults with MCI varies across informal care networks, but typically involves a primary care partner and a network of family, friends and others. In this paper, we present a thematic analysis of in-depth interviews with older adults with MCI and their primary care partners to gain a holistic picture of their day-to-day lived experience. Our analysis uses a multi-dimensional lens of people (“who”), activities (“what”), and tools (“how”) to reveal insights about the nature of informal care coordination in MCI. Our results characterize informal care for MCI as a set of complex orchestration tasks by a primary care partner that support the practical, cognitive and emotional needs of the diagnosed individual and mediate the involvement of the broader care network. We uncover that coordination is not solely a matter of logistical organization, but also a deeply relational process shaped by negotiation with technological tools and evolving roles. Through this work, we reframe care coordination for MCI as a distinct and underexplored design space, one that demands systems capable of scaffolding autonomy, adapting to shifting capacities, responding to socio-emotional needs, and fostering collaborative caregiving.
[HC-31] On-Device Language Models for Privacy-Preserving Stress Prediction: A Multimodal Evaluation on Mobile Health ALT
链接: https://arxiv.org/abs/2609.11961
作者: Ibukunoluwa Soyebo,Alyssa Donawa,Rodrigo Aguilar Barrios,Brice Patchou,Corey E. Baker
类目: Machine Learning (cs.LG); Human-Computer Interaction (cs.HC)
备注: 6 pages, 2 figures. Received Honorable Mention at the 2026 Human-centered AI Research for Mental health, an Open Networking Symposium (HARMONY 2026) workshop, co-located with IEEE/ACM Conference on Connected Health: Applications, Systems, and Engineering Technologies (CHASE 2026)
Abstract:Stress is a pervasive determinant of mental health and a key target for mobile health interventions. On-device language models (ODLMs) offer privacy-preserving inference without cloud dependency, yet their feasibility for health prediction under mobile resource constraints remains underexplored. We evaluate ODLMs for multi-modal stress prediction using zero-shot prompting, measuring predictive accuracy alongside latency and throughput. Our results show that objective sensor features marginally outperform subjective self-reports on average, and that lightweight sub-2B models achieve low latency with predictable resource usage. Our findings highlight both the promise and the practical constraints of ODLMs for mobile mental health.
计算机视觉
[CV-0] SNAP3D: Physically Grounded 3D Parts for Assembly from a Single Image
链接: https://arxiv.org/abs/2609.13146
作者: Yu-Rou Tuan,Hao-Tang Tsui,Nicolas Ugrinovic,Kris Kitani,Xiaoxuan Ma
类目: Graphics (cs.GR); Computer Vision and Pattern Recognition (cs.CV)
备注: Project page: this https URL
Abstract:Part-aware 3D asset generation enables applications such as editing, articulation, simulation, and fabrication, yet existing methods can generate visually complete individual parts without ensuring that they form a valid physical assembly. Consequently, generated neighboring parts may interpenetrate, lack valid connections, or collapse under gravity. We propose a physics-guided framework for improving single-image part-aware 3D generation with physically compatible geometry and stable connections. Our method resolves inter-part penetration, recovers a contact graph between neighboring parts, and introduces parameterized connectors at their contact surfaces. Using feedback from physical simulation, we refine connector placement, orientation, and dimensions to improve assembly stability while preserving the generated geometry. We further introduce a physics-based evaluation protocol that complements conventional geometric metrics by directly testing assembly validity and stability under gravity. Experiments comparing against multiple part-aware 3D generators show substantial improvements in physical realizability and stability while maintaining geometric quality. We additionally validate the resulting parts through 3D printing and real-world assembly.
[CV-1] Unified CT and MRI Pancreas Segmentation for Label-Efficient Cross-Modality Subregion Transfer
链接: https://arxiv.org/abs/2609.13043
作者: Ziliang Hong,Hongyi Pan,Halil Ertugrul Aktas,Andrea Bejar,Elif Keles,Frank H. Miller,Michael B. Wallace,Rajesh N. Keswani,Gorkem Durak,Ulas Bagci
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Robust medical image segmentation across imaging modalities is challenging because of large differences in appearance and intensity distributions. Models trained on a single modality often show substantial performance drops when applied to unseen domains. In this work, we develop a unified 3D pancreas segmentation framework that applies domain-adversarial learning to 4,604 heterogeneous CT and MRI scans to learn anatomical representations. A shared nnU-Net encoder-decoder is trained for whole-pancreas segmentation, with a latent domain discriminator encouraging CT-MRI feature alignment. The learned encoder is subsequently transferred to pancreatic head-body-tail segmentation using limited MRI-only subregion annotations. An average Dice score of 87.31% on the in-distribution test set and Dice scores ranging from 84.20% to 88.09% across external OOD datasets were achieved in whole pancreas segmentation. Dice scores of 80.53% on MRI and 83.05% on CT were achieved for downstream subregion segmentation, without using CT subregion annotations. These results demonstrate that a unified anatomical representation can support both cross-modality pancreas segmentation and label-efficient downstream transfer.
[CV-2] Label-Guided Knowledge Distillation for 3D-CNNs in Action Recognition
链接: https://arxiv.org/abs/2609.13024
作者: Yanjiang Shi,Peng Zhao,Nan Qi,Guiqin Wang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 10 pages, 2 figures
Abstract:As a key model compression technique, knowledge distillation aims to transfer knowledge from a high-capacity teacher model to a lightweight student model for enhancing the latter’s performance. In this work, we reviewed the feature knowledge distillation for 3D-CNNs and observed that most feature distillation methods in video analysis are simple adaptations of those used in image analysis, often neglecting the differences of video features in the temporal dimension. To address this issue, we proposed Label-Guided Knowledge Distillation (LGKD) to guide the distillation of student model features using ground truth labels. Our method entails two components: sample-wise distillation and class-wise distillation, enabling the student model to learn feature representation of the teacher model at two levels. Sample-wise distillation utilizes label information and the teacher’s probability distribution to guide the learning of features that significantly impact temporal accuracy while mitigating noise. Meanwhile, class-wise feature distillation employs a prototype network to further capture the relational knowledge among samples within the same category, enhancing the student’s ability to learn higher-dimensional semantic information and improving model generalization. To demonstrate the effectiveness and superiority of our method, we conducted comprehensive experiments on two benchmark action recognition datasets, UCF101 and HMDB51, achieving competitive results.
[CV-3] Net: Tile-Based CNN-SVM Architecture for Autonomous Unmanned Aerial Systems Inspection of Flat Roofs
链接: https://arxiv.org/abs/2609.13013
作者: Samuel Dunthorne,Hashim A. Hashim
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Journal of Safety Science and Resilience
Abstract:Flat roofs are among the most influential components of the building envelope, governing both structural performance and thermal efficiency, and thereby contributing directly to household energy consumption, carbon emissions, and long-term environmental sustainability. Timely detection of roof defects is essential for reducing heating and cooling losses, preventing moisture-driven degradation such as mold growth, and supporting national climate-change mitigation goals. This paper presents a real-time, Unmanned Aerial System (UAS)-based deep learning framework that autonomously detects defects using live imagery captured during dual-altitude aerial passes. The multi-resolution flight strategy is designed to aid the identification of both small, fine-scale defects and larger structural issues, enabling more comprehensive assessments. To meet the strict computational and power constraints of embedded UAS hardware, the proposed framework integrates a tile-based architecture with a lightweight Convolution Neural Network-Support Vector Machine (CNN-SVM) classifier designed for low-latency onboard inference. The final model-comprising five convolutional layers and four dense layers, the last a linear SVM head, achieved a mean test accuracy of 94.4% ( 95% confidence interval \pm0.4% over three seeds) on a photo-level split ( 43,383 training, 3,869 validation, and 2,540 test tiled and augmented images), outperforming GoogLeNet ( 89.2% ) and AlexNet ( 79.8% ). Experimental evaluations using real UAS imagery collected by onsite visits with DJI Matrice 350 RTK drone demonstrate that the system supports rapid, repeatable, and safe roof inspections while reducing human risk, lowering operational costs, and enabling more sustainable building maintenance.
[CV-4] Pixel Decodability Is Not a Compression Signal: Causally Evaluating Importance Proxies for Visual KV-Cache Eviction
链接: https://arxiv.org/abs/2609.13012
作者: Chenyu Zhou,Qiliang Jiang,Shuning Wu,Xu Zhou
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 15 pages, 3 figures, 3 tables
Abstract:Vision-language models retain a substantial amount of pixel-decodable visual content in their visual key-value cache. We show, in our setting, that this retention is task-inert: across our preregistered tests, how much a unit retains never positively tracks whether the computation that answers the question causally relies on it. We measure retention with a learned pixel-inversion decoder and causal use with single-super-patch KV ablation, the teacher-forced drop in gold-answer log-probability, and relate the two within images under a preregistered, sign-calibrated, held-out design. Retention is decoupled from attention and, in a well-powered null, from causal utilization. Utilization is not inert to every proxy: attention weakly but significantly tracks it, the only signal we find that does and the design’s positive control. We characterize pixel-decodable retention as an informational axis of the visual KV cache, orthogonal to the functional one. How much task-inert content a cache holds differs by architecture in our model pair: the encoder-free model retains 2.7 times more than the encoder-based one. The engineering consequence is a controlled negative result. At super-patch granularity, deconfounded pixel-decodable retention ranks KV eviction no better than random; at token granularity it acquires only a weak inverse-importance signal at larger budgets, dominated at every budget by attention magnitude. In our setting, pixel-decodable reconstructability is not a competitive KV-compression signal at any granularity we test.
[CV-5] Physics-Aware Video Generation via Agent ic Planning and Graph-Guided Optimization
链接: https://arxiv.org/abs/2609.13006
作者: Minh-Loi Nguyen,Xuan-Vu Le,Thanh-Toan Do,Tam V. Nguyen,Minh-Triet Tran,Trung-Nghia Le
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Video diffusion models (VDMs) have demonstrated remarkable capabilities in synthesizing high-fidelity, photorealistic video content. However, they fundamentally lack an intrinsic understanding of physical laws and frequently produce visually appealing but causally illogical sequences characterized by structural hallucinations and physically implausible dynamics. Injecting physical awareness via training-free test-time optimization is a promising alternative, yet existing methods rely on global gradient updates and rigid scheduling heuristics that inadvertently corrupt passive backgrounds and fail to model complex dynamic state changes. To address this, we propose PhysPlan, a novel training-free guidance framework that shifts the paradigm from stochastic visual interpolation to agentic physics simulation. First, a VLM operates as an iterative cognitive simulator, decomposing multimodal inputs into a Chain-of-Visual-Thought to create a multimodal representation of kinematic trajectories and 3D depth geometries. Second, these signals drives an object-centric test-time optimization. Unlike prior training-free methods that rely on global gradients and rigid scheduling heuristics, PhysPlan introduces Object-Centric Gradient Routing to isolate kinematic modifications and completely lock the passive environment. Furthermore, our Kinetic Intensity Profiling dynamically parameterizes framework hyperparameters to accommodate the varying severity of physical deformations. Extensive evaluations on the PhyGenBench and Physics-IQ benchmarks demonstrate that PhysPlan significantly outperforms both foundational and controllable VDM baselines, offering a promising approach for improving the physical understanding of video generation.
[CV-6] SV-Cine: Diagnosis-Conditioned Segmentation of Single Ventricle Physiology via Generative Data Augmentation
链接: https://arxiv.org/abs/2609.12997
作者: Lila Cunge,Yuehong Liu,Hang Xu,Thomas Coudert,Pierangelo Renella,J Paul Finn,William Hsu,Kim-Lien Nguyen
类目: Computer Vision and Pattern Recognition (cs.CV); Medical Physics (physics.med-ph)
备注:
Abstract:Single Ventricle Physiology (SVP) is a rare subtype of congenital heart disease characterized by the presence of a single functional cardiac ventricle with atypical anatomic configurations that challenge conventional image segmentation approaches. The scarcity of clinical data and the morphological diversity across SVP subtypes make the development of robust segmentation methods particularly difficult. To address these limitations, we propose a cardiac MRI segmentation framework focused on ventricular chambers and myocardium segmentation tailored for SVP. First, we introduce a data augmentation pipeline that generates synthetic 3D cardiac meshes using SDF4CHD and corresponding synthetic cardiac MRI through generative modeling. Second, we introduce SV-Cine, a diagnosis-conditioned adaptation of the foundation model CineMA that incorporates patient-level diagnostic information through Feature-wise Linear Modulation layers, enabling diagnosis-aware feature adaptation during segmentation. We evaluated the framework on an internal cohort with varying SVP subtypes. SV-Cine achieved median Dice scores of 0.89 (IQR: 0.80–0.91) for the left ventricle and 0.72 (IQR: 0.54–0.84) for the right ventricle, outperforming the strongest baseline, nnU-Net, by 0.39 Dice points on right ventricle segmentation. It also yields a median ejection fraction error of 5.55 percentage points (IQR: 3.41–7.69) for the dominant ventricle. Compared with the internal cohort, LV and myocardium segmentation performance was lower for the external cohort; whereas RV Dice scores were comparable for both cohorts. Our findings suggest that a pretrained foundation model can be adapted for highly specialized downstream tasks through usage of diagnosis priors while leveraging anatomic knowledge learned from large-scale MRI datasets during pretraining.
[CV-7] Generative Retrieval for Unsupervised Text-Based Person Search
链接: https://arxiv.org/abs/2609.12965
作者: Mang Ye,Yucheng Ji,Yang Bai,Min Cao,Siyuan Chai,Bo Du,Min Zhang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 17 pages, 10 figures. Accepted for publication in IEEE Transactions on Pattern Analysis and Machine Intelligence
Abstract:Text-based person search (TBPS) aims to retrieve images of a target person from a large image gallery based on a given natural language description. Most existing methods rely on supervised learning with manually annotated image-text pairs. In this paper, we explore unsupervised TBPS, with only unlabeled images. We propose GTR+, a two-stage generation-then-retrieval framework. In the generation stage, we introduce a tiered description generation framework designed to produce fine-grained and stylistically diverse textual descriptions through a three-tier sequential process. The base tier leverages an automated question-and-answer mechanism to generate basic visual attribute descriptions; the intermediate tier enhances fine-grained detail using an inter-sample contrastive mechanism; the advanced tier further enriches textual diversity via a stylized expansion mechanism. In the retrieval stage, to mitigate the impact of noisy pseudo texts, we develop an adaptive confidence-weighted retrieval learning framework. We model image-text pairs as clean or noisy using a Gaussian Mixture Model, calibrated by real-time image-text similarity and static text generation probability from the prior stage, yielding adaptive sample weights during training. Beyond that, we also contribute LargeFine-Person, a large-scale TBPS dataset with high-quality, fine-grained, and diverse textual annotations, enabling a practical and generalizable TBPS pre-training benchmark under unsupervised setting. Experiments on multiple TBPS benchmarks demonstrate the effectiveness and generalization of both GTR+ and LargeFine-Person. Code is available at: this https URL.
[CV-8] Fast and Faithful: Principled Conditional Flow Matching for Inverse Problems
链接: https://arxiv.org/abs/2609.12953
作者: Shirin Shoushtari,Edward P. Chandler,Xiao Shi,Ulugbek S. Kamilov
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Flow matching approaches to imaging inverse problems commonly incorporate measurements in two ways. Conditioning-based approaches supply measurement-derived information as a network input, often through concatenation, while inference-guided approaches combine an unconditional velocity field with a separate data-consistency update. In these common formulations, the forward model is not explicitly enforced within the learned conditional velocity field. We propose a principled parametrization of the measurement-conditional velocity field to solve inverse problems. Under linear interpolation, we express the conditional velocity v(x_t,t,y) in terms of the posterior mean E[x_1 | x_t,y] , and characterize that mean as the unique minimizer of a variational objective whose data-consistency term is explicit. We further prove that the velocity field defines a probability flow from the source distribution to the measurement-conditioned posterior. Splitting the variational objective yields a conditional velocity parameterization with operator-dependent data-consistency updates, which we train end-to-end under the flow-matching objective, with no additional guidance at inference. Our method achieves state-of-the-art PSNR with 50\times fewer function evaluations than the strongest flow baseline. Varying the sampling steps provides test-time control over the distortion-perception trade-off without retraining.
[CV-9] Input Resolution Matters: Real-Time Object Detection Latency
链接: https://arxiv.org/abs/2609.12920
作者: Qingyang Zhang,Fumio Machida,Laura Carnevali
类目: Computer Vision and Pattern Recognition (cs.CV); Performance (cs.PF)
备注: 8 pages, 2 figures, 2026 34th IEEE International Symposium on Modelling, Analysis and Simulation of Computer and Telecommunication Systems (MASCOTS)
Abstract:We model total latency as the convolution of preprocessing, inference, and postprocessing distributions under a simplifying independence approximation, with selected stage parameters expressed as functions of source-image resolution. Under this assumption, the probability density of the total latency is the convolution of the stage-wise densities, and its cumulative distribution function (CDF) provides the distribution of end-to-end detection time. Each stage is modeled by a parametric distribution (e.g., Exponential, Erlang, Normal, Gamma), with parameters expressed as functions of the source-image resolution. Experiments with YOLOv11n on NVIDIA Jetson Orin NX using COCO2017 images across multiple resolutions assess the proposed models against fixed-parameter baselines using Kolmogorov Smirnov, Anderson Darling, and Cramér von Mises statistics. The results indicate that resolution-aware parameterization can improve distributional approximation in the measured setting, particularly for the more flexible Normal and Gamma models, while the quality of fit remains distribution dependent. Our contribution is a theoretically grounded and lightweight formulation for studying resolution-dependent latency distributions in a measured object detection pipeline.
[CV-10] Parallel Training Using a CNN-DNN Architecture for Accelerated Development of Diagnostic Models
链接: https://arxiv.org/abs/2609.12902
作者: Janine Weber-Hamacher,Astha Jaiswal,Philipp Fervers,Dorotya Móré,Athanasios Giannakis,Ricarda Fischbach,Andreas Michael Bucher,Rahil Shahzad,Jonathan Kottlors,Thorsten Persigehl,Axel Klawonn
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Artificial intelligence has shown promise in assisting radiologists in imaging-based diagnosis across a wide range of diseases. Efficient training of large deep learning models is essential to cope with extremely large data sets or dynamically growing disease data, like in a pandemic like situation. In this retrospective study, we collected 300 CT scans from COVID-19 and non-COVID-19 pneumonia patients from three different centers in Germany. We investigated a hybrid CNN-DNN network model based on image decomposition and localization that naturally supports parallel and efficient training of deep learning models. In total, 156 models with three different architectures were trained to capture features at different levels resulting in 12 patient-level COVID-19 diagnosis models. Diagnostic performance as well as time saving were measured. The highest accuracy was obtained from DenseNet121 and 3D CNN models with a parallel CNN-DNN approach, resulting in 88.78% training, 76.67% validation and 76.03% test accuracy for the DenseNet121 with 4\times4\times1 subdomains and 87.72% training, 76.82% validation and 74.86% test accuracy, respectively, for the 3D CNN with 4\times4\times1 subdomains. The strongest reduction in parallel training time by a factor of 31 was observed for the 3D CNN model and 4\times4\times2 subdomains. Our parallel training approach improves efficiency as well as performance enabling rapid model development, among others crucial for pandemic preparedness. Subjects: Computer Vision and Pattern Recognition (cs.CV) MSC classes: 68T07, 68W10, 68W15, 68T10 Cite as: arXiv:2609.12902 [cs.CV] (or arXiv:2609.12902v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2609.12902 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-11] UniPart: Towards Zero-shot Language-Grounded 3D Part Segmentation for Embodied Interaction
链接: https://arxiv.org/abs/2609.12898
作者: Xinqiang Yu,Zekun qi,Jiawei He,Wenyao Zhang,Xuchuan Chen,Guaocai Yao,Li Yi,Zhaoxiang Zhang,He Wang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Robotics (cs.RO)
备注:
Abstract:Fine-grained robotic manipulation depends on understanding parts, not only whole objects. Existing 3D foundation models tend to be either generalized but object-aware, or part-aware but limited to closed-set taxonomies, which weakens zero-shot transfer. We study text-conditioned 3D part segmentation, where a free-form phrase selects a functional part on point cloud. We introduce UniPart, a feed-forward cross-modal 3D Transformer that conditions CLIP text embedding. To scale supervision, we build LangPart-1M with 160K+ Objaverse assets and 8M text to part pairs using multi-view consistent part generation. We further manually label a high-quality subset, LangPart-4K, for fine-tuning and evaluation. UniPart achieves strong zero-shot results on open-vocabulary part benchmarks and transfers to language-conditioned part grasping in real world.
[CV-12] Beyond Accuracy: Uncertainty-Guided Boundary Refinement for Reliable Biomedical Image Segmentation
链接: https://arxiv.org/abs/2609.12892
作者: Anima Kujur
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Accurate biomedical image segmentation requires not only high global overlap but also reliable delineation of clinically meaningful boundaries. In blood-smear microscopy, cytoplasm and nucleus contours provide the structural basis for downstream morphology analysis; however, deep segmentation models may remain uncertain or overconfident near ambiguous boundary regions even when achieving strong Dice scores. This work proposes a Reliability-Aware Boundary Refinement Network (RABR-Net), a two-stage framework for trustworthy image segmentation. A strong UNet++ EfficientNet-B4 base segmenter first produces initial class probabilities and logits. Predictive entropy, test-time augmentation variance, margin uncertainty, probability gradients, and soft boundary cues are then combined into a boundary-aware reliability representation. This representation guides a gated residual refiner that selectively corrects uncertain boundary pixels while preserving confident regions of the base prediction. The framework is evaluated using overlap accuracy, class-wise Dice, Boundary Dice, HD95/ASSD, calibration, risk–coverage analysis, robustness under image perturbations, qualitative correction maps, and paired statistical testing. On the held-out test set, the proposed method improves Dice from 0.9602 to 0.9614, Boundary Dice from 0.3448 to 0.3611, and HD95 from 3.0354 to 2.8274 compared with the cached base prediction. Statistical analysis confirms significant improvements in Dice, Boundary Dice, and HD95. Qualitative results show that the learned gate concentrates around uncertain cytoplasm and nucleus boundaries, and correction maps confirm localized boundary refinement. Although calibration does not automatically improve after refinement, the proposed framework provides an interpretable and reliability-focused strategy for boundary-sensitive biomedical image segmentation.
[CV-13] Learning Sign Language Recognition under Label Noise: A Study of Noise-Robust Losses for Isolated and Continuous Settings
链接: https://arxiv.org/abs/2609.12885
作者: Akihisa Shitara,Yoichi Ochiai
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 15 pages, 1 figures
Abstract:In sign language recognition, the isolated (ISLR) classification loss treats a single label as ground truth, as does the frame-level auxiliary classifier over pseudo-labels we add to continuous (CSLR) methods, which lack one. Stylistic variation blurs ISLR annotation and the lack of temporal boundaries in CSLR forces pseudo-labels; both are noisy. We therefore apply symmetric and generalized cross entropy (SCE, GCE), robust alternatives to cross entropy (CE) from image classification, not to connectionist temporal classification but to the preceding single-label classifier. On ASL Citizen with injected symmetric noise on three backbones (three seeds for ST-GCN), robust losses cost at most 2.5 pt when labels are clean and beat CE by 2.9-10.0 pt in all six conditions at noise rate 0.2, one of which only after q was re-selected on dev. GCE gains more, but its optimal q does not transfer across backbones, whereas one SCE setting works in all nine conditions; both vary 2-11 times more than CE across runs, so a favorable point estimate does not establish stability. For CSLR (PHOENIX-2014) we report no gain; our frame-level targets carry a systematic assignment bias, making that study a diagnosis of a single configuration. At lambda_aux = 25 the pseudo-label CE auxiliary raises word error rate above the no-auxiliary baseline on VAC, CorrNet and SlowFastSign, and GCE/SCE improve on CE by 1.7-3.2 pt (three of six conditions return below that baseline). However, the three losses differ by more than an order of magnitude in effective gradient at a common lambda_aux: matching the initial gradient shrinks the gap to 0.4-0.9 pt, and lowering the CE weight alone already beats that baseline, so neither the degradation nor the improvement can be separated from the effect of the weight. We use only symmetric noise; multi-seed evaluation covers only ST-GCN and VAC isolated.
[CV-14] VideoTok4D: A 4D-Aware Video Tokenizer for Compact World Representation
链接: https://arxiv.org/abs/2609.12874
作者: Xinyi Chen,Hanxin Zhu,Xijun Wang,Xingrui Wang,Sen Liang,Xin Li,Zhibo Chen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages, 5 figures
Abstract:Video tokenizers have emerged as a cornerstone of modern video modeling, underpinning progress in compression, reconstruction and generation by mapping high-dimensional visual signals into compact latent spaces. However, despite this progress, current tokenization paradigms largely remain within the 2D visual domain, treating videos as image sequences rather than observations of an underlying dynamic 3D world. Consequently, the learned tokens inherit this observation-centric bias, limiting their capacity to compactly represent real-world 4D scenes. To mitigate this issue, we propose VideoTok4D, a novel 4D-aware video tokenizer for compact world representation. Specifically, our approach comprises three key designs: 1) a spatiotemporal disentanglement strategy that factorizes videos into static and dynamic tokens for holistic world modeling; 2) a track-aware dynamic attention mechanism that aggregates trajectory-aligned cues to promote cross-view motion consistency; and 3) Co4DGen, a diffusion prior learned over the resulting VideoTok4D token space for efficient 4D scene generation. Extensive experiments have demonstrated that our proposed method achieves state-of-the-art performance while requiring up to 4 orders of magnitude less storage than dense 4D representations. Moreover, the compact token space substantially shortens diffusion sequences, enabling efficient generation.
[CV-15] A Multi-Vehicle Dataset with Camera LiDAR and Radar Sensors and Scanned 3D Models for Custom Auto-Annotation using RTK-GNSS
链接: https://arxiv.org/abs/2609.12871
作者: Philipp Berthold,Bianca Forkel,Mirko Maehlisch
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: Paper accompanying the dataset “7V-Scanario”
Abstract:Datasets are a crucial element in the development of perception algorithms. They relate sensor measurement data to annotated reference information and allow for the deduction of sensor and object characteristics. In autonomous driving, the reference data commonly consist of semantic image segmentation, point-wise associations, or bounding box annotations. The dataset proposed in this work, however, aims to dig deeper into the evaluation of measurement principles and provides scanned 3D models of all vehicles together with a pose and continuous kinematics reference obtained by RTK-GNSS. Combined, the state of the complete dynamic surrounding of the sensor vehicle is known for any point in time. Subsequent reference formats can be easily computed in user-defined granularity. This dataset involves single-object and multi-object recordings with seven target vehicles. In particular, measurement effects such as occlusion, as well as reflections, can be evaluated, as the normals of the shape of the target vehicles are known. We describe the dataset, discuss the technical background of its development, and briefly present exemplary evaluations.
[CV-16] 3D CT-to-PET Translation via Latent Brownian Bridge Diffusion
链接: https://arxiv.org/abs/2609.12860
作者: Sarita Mourya,Francesco Di Feola,Pierangelo Veltri,Paolo Soda
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Computed tomography (CT) and positron emission tomography (PET) provide complementary anatomical and functional information for cancer diagnosis and treatment planning. However, the widespread use of PET is limited by high radiation exposure, elevated costs, and restricted availability. To address these limitations, deep learning-based CT-to-PET translation has emerged as a promising approach for synthesizing PET-like information directly from CT images, although accurately modeling the large cross-modal gap remains challenging. In this work, we propose a 3D CT-to-PET translation framework based on latent Brownian Bridge Diffusion (BBDM). The method consists of two stages. First, a Variational Autoencoder (VAE) is trained on paired CT-PET patches, integrating contrastive learning to improve latent alignment between anatomical and metabolic representations. Second, a BBDM is trained in the latent space to translate CT latent representations into their corresponding PET counterparts. The translated PET latents are then decoded and stitched to reconstruct the final 3D PET volume. We evaluate the proposed approach on two publicly available datasets. Quantitative results based on image fidelity and lesion-level PET-specific metrics demonstrate improved performance compared with competing methods. In particular, the proposed approach improves PET signal fidelity, better preserves clinically relevant uptake patterns, and shows improved performance in preserving small-lesion metabolic activation, paving the way for virtual imaging applications.
[CV-17] MGAvatar: Mesh-Bound Gaussians for Head Avatar Geometry and Appearance Modeling
链接: https://arxiv.org/abs/2609.12850
作者: Lei Shi,Sen Peng,Zhiyang Deng,Zhonggui Chen,Xiaohu Guo,Baorong Yang,Xiao Dong
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 10 pages, 7 figures. Accepted to the Journal Track of Pacific Graphics 2026. Code is available at this https URL
Abstract:Accurate head modeling requires a stable yet expressive geometric representation. Existing Gaussian-based head avatars commonly rely on parametric templates (e.g., FLAME) for Gaussian initialization and deformation, but these templates lack personalized priors and struggle to represent structures such as hair and clothing. To address this issue, we propose MGAvatar, a Gaussian-mesh hybrid representation that jointly models geometry and appearance through two Gaussian-mesh binding modes. Specifically, we introduce vertex-bound Gaussians and constrain their learnable parameters, enabling progressive mesh deformation to represent complex head geometry, while a pose-dependent offset module accounts for non-rigid deformations. Once geometry is stabilized, MGAvatar switches to face-bound Gaussians for appearance modeling. To improve appearance consistency across novel poses and viewpoints, we introduce a view-conditioned neural color field that alleviates artifacts caused by independently optimized Gaussian colors. In addition, we design a Gaussian offset network to predict Gaussian offset maps in the observation space, providing greater flexibility for face-bound Gaussians to capture dynamic facial textures. Extensive experiments on multi-view and monocular videos show that MGAvatar outperforms existing methods in rendering quality, producing high-fidelity head avatars with rich texture details.
[CV-18] Pre-Trained Low-Rank Tensor Decomposition for Multi-Dimensional Image Recovery
链接: https://arxiv.org/abs/2609.12843
作者: Bing-Zhang Fu,Zhi-Long Han,Ting-Zhu Huang,Xi-Le Zhao,Deyu Meng
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Recently, tensor decompositions are prevalent for multi-dimensional image representation, which learn the instance-specific structure of each image from scratch. However, tensor decompositions neglect the common structure across different images, leading to limited semantic modeling capability, high computational cost, and a large number of learnable parameters. To address this challenge, we suggest the first pre-trained low-rank tensor decomposition (PLTD) framework, which organically integrates the pre-trained large vision model into the classical tensor decomposition framework. Beyond the shallow and untrained deep tensor decomposition, the suggested PLTD achieves an unprecedented balance among higher recovery fidelity, fewer learnable parameters, and smaller carbon footprint. Specifically, PLTD factorizes the target tensor into a latent tensor and a learnable transform that maps the latent tensor back to the original data domain. The latent tensor consists of two indispensable and complementary terms, i.e., a fixed pre-trained latent tensor and a learnable low-rank latent tensor. The fixed pre-trained latent tensor is distilled from a pre-trained large vision model (i.e., DINOv3) to capture the common structure of the target tensor, while the learnable low-rank latent tensor characterizes the instance-specific structure of the target tensor. To examine the potential of PLTD, we develop the corresponding multi-dimensional image recovery model and theoretically justify the advantages of this framework. Additionally, we discuss the connections between PLTD and classical tensor decomposition frameworks. Extensive experiments on multi-dimensional image recovery demonstrate that PLTD consistently achieves superior performance compared with state-of-the-art methods.
[CV-19] HemaHier: Chain-Conditioned Ordinal Hierarchies for Lineage-Aware Bone-Marrow Cytology MICCAI2026
链接: https://arxiv.org/abs/2609.12835
作者: Afshin Bozorgpour,Peter Schüffler,Edgar Jost,Dorit Merhof
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at the MICCAI 2026 Workshop on Computational Pathology and Multimodal Data (COMPAYL)
Abstract:Bone-marrow cytology is inherently structured: each cell belongs to a hematopoietic lineage, and many cell types lie on ordered maturation trajectories. Standard flat classifiers ignore this structure, treating a mild same-lineage confusion the same as a severe cross-lineage mistake and predicting only discrete labels. We propose HemaHier, an ordinal-hierarchical prediction head for a frozen or lightly adapted cytology foundation model. Its central component is a chain-conditioned maturity score that reads a single maturity value under a per-chain query, supervised only on biologically valid healthy chains, while dysplastic and off-chain cell types remain classes but are excluded from maturity supervision. Fine and lineage predictions are coupled through a shared posterior that guarantees hierarchical consistency, and a staged objective first stabilizes recognition, then adds lineage and maturity supervision. On three bone-marrow datasets under a shared ontology, HemaHier achieves competitive recognition while reducing biologically severe errors and adding a within-lineage maturity ordering that flat classifiers lack. Code is available at this https URL.
[CV-20] Self-supervised Pre-training Helps Retinal Disease Progression Modelling Most When Data Is Scarce
链接: https://arxiv.org/abs/2609.12834
作者: Ifeoma Veronica Nwabufo,Julius Gervelmeyer,Sarah Müller,Philipp Berens
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Modelling how a disease progresses over time requires longitudinal imaging cohorts, which are scarce and small, whereas cross-sectional data – one image per participant – is abundant. Self-supervised pre-training on such data offers a way to bridge this gap, but it is unclear which strategy best supports progression modelling, or how that answer depends on the amount of labelled longitudinal data. We study this for age-related macular degeneration (AMD), pre-training encoders on the large cross-sectional NAKO cohort and predicting time to late AMD on the longitudinal AREDS dataset. We compare in-house self-supervised encoders against a general-purpose (DINOv2) and a domain-specific (RETFound) foundation model, across contrastive, masked-autoencoding, and self-distillation objectives, under frozen and fine-tuned protocols, and across labelled training sets from 100 to 32,250 examples. Which model performs best depends on how the encoder is used. When the encoder is frozen and labels are few – the regime typical of longitudinal cohorts – pre-trained representations reach clinically reasonable discrimination from a few hundred labelled samples, while models trained from scratch do not; this advantage fades under fine-tuning. Transfer is governed by the self-supervision objective rather than corpus scale or domain match, so that an encoder pre-trained on a modest cross-sectional cohort matches or exceeds a far larger in-domain foundation model. Together, these results offer a practical recipe for building progression models where longitudinal data is scarce: a frozen self-supervised encoder with a lightweight survival head.
[CV-21] Balancing Emotional Alignment and Semantic Consistency in Image Generation via Reinforcement Learning with Valence-Arousal Anchoring
链接: https://arxiv.org/abs/2609.12830
作者: Jisheng Dang,Zhenxuan Wang,Bin Li,Ronghao Lin,Bin Hu,Tat-Seng Chua
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 13 pages, 10 figures, and 4 tables. Code is available at this https URL
Abstract:Continuous emotion control in text-to-image generation requires a model to improve affective alignment without changing the objects, layout, or scene described by the prompt. Existing supervised emotion-injection methods often optimize feature-space proxies and may therefore exhibit emotion-semantic drift, in which stronger emotional conditioning is accompanied by unintended content changes. We address this problem with a flow-matching image-generation framework that combines continuous valence-arousal (VA) conditioning, Group Relative Policy Optimization (GRPO), and a neutral semantic anchor. The deterministic probability-flow ODE is converted into a marginal-preserving SDE, yielding non-degenerate transition densities for trajectory sampling and policy-ratio estimation. A frozen CLIP-based VA regressor supplies a terminal reward measuring the distance between the predicted and target VA coordinates, while an image generated from the same prompt under zero VA conditioning provides a feature-space reference for semantic preservation. A reduced denoising schedule is used for online RL sampling, whereas the original schedule is retained at inference. Experiments on 3,300 prompt-emotion combinations show substantially lower valence and arousal errors than the VA-conditioned baseline and an improved CLIPScore relative to EmotiCrafter, with a measurable trade-off in reference-free image quality. The results support anchor-regularized Flow-GRPO as a practical approach to balancing emotional alignment and semantic consistency in continuous-affect image synthesis.
[CV-22] A Dual Cross-Attention Framework for Colposcopic CIN Grading and Swede Score Prediction Using a New Multi-Center Dataset
链接: https://arxiv.org/abs/2609.12827
作者: Dania Khan,Nuzhat Aisha Shaikh,Asfina Hassan Juicy,Raiyun Kabir,S M Shahida,Taufiq Hasan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 8 pages
Abstract:Cervical cancer is a major global health challenge, with disease burden falling disproportionately on low- and middle-income countries (LMICs) due to a shortage of trained specialists and the subjective nature of colposcopy-based screening. To address this challenge, we propose a novel deep learning framework for the automated grading of Cervical Intraepithelial Neoplasia (CIN) and the prediction of clinical Swede scores. We also introduce the BUET Multi-Center Colposcopy Dataset, a novel, multi-center cohort designed and annotated for Swede score prediction and CIN grading. Our proposed dual-stream cross-attention architecture mimics the visual reasoning of an expert colposcopist by explicitly fusing paired multimodal cervigrams to evaluate comparative tissue responses. Furthermore, we introduce a custom composite loss function to address severe class imbalances and scoring inconsistencies across the five Swede score components. The proposed framework achieved 71.85% accuracy and an 86.23% AUC-ROC for three-class CIN grading, outperforming existing methods. For Swede score component prediction, the architecture achieved AUC-ROC values ranging from 75.7% to 88.4%, with the composite loss function yielding consistent F1-score improvements. Finally, the total predicted Swede Score, which ranges between 0 and 10, shows a Mean Absolute Error (MAE) of 1.489. The results show that the proposed method can pave the way towards developing AI-assisted colposcopy screening tools to support risk-based triage in resource-limited healthcare settings. The dataset and source code are publicly available(url: this https URL)
[CV-23] CoralscapesV2: Panoptic and Fine-Grained Visual Scene Understanding in Coral Reefs
链接: https://arxiv.org/abs/2609.12826
作者: Jonathan Sauder,Thomas Ruckli,Gabrielė Strodomskytė,Ibrahim Souleiman Abdallah,Rahma Hassan Abdi,Djama Goumaneh Awaleh,Mohamed Houssein Farah,Moustapha Nour,Osama Sharhubil Saad,Mustafa Mohammed Khalafallah Altaib,Maysoon Kteifan,Farah Alsoqi,Eyad Zgool,Jafar Al-Omari,Temesgen Gebremeskel Gebreluel,Zekaria Zekeria Abdulkerim,Meron Ghirmay,Teklehaimanot Beraki,Devis Tuia,Guilhem Banc-Prandi
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:In order to design conservation and restoration strategies to counter the global decline of coral reefs, ecological monitoring of reefs needs to be scaled up dramatically. Computer vision methods are increasingly used to tackle the vast amount of data: as the paradigm of data collection in reefs shifts from highly standardized and constrained survey images to unconstrained imagery on scalable platforms, it is necessary to design machine learning methods that help to get a fine-grained understanding of reefs from general-purpose reef imagery. This paper provides CoralscapesV2, an extension of the Coralscapes dataset for general-purpose visual scene understanding in reefs. CoralscapesV2 increases the dataset size, scope, label completeness and quality for semantic segmentation, and extends the number of classes from 39 to 95 fine-grained visual categories. Furthermore, CoralscapesV2 provides 65k exhaustive fish instance mask annotations, meticulously annotated to completeness by using the video, revealing that annotation of fish based on only static images is insufficient. CoralscapesV2 is the first dataset for panoptic segmentation in coral reefs, capturing a wide range of scenarios in the wild, posing a challenging benchmark for contemporary semantic segmentation and instance segmentation models. CoralscapesV2 is an important step towards general-purpose panoptic segmentation in coral reefs, which has substantial implications for scaling up coral reef monitoring, as it can be employed in a wide range of applications from benthic cover mapping from robot or handheld videos to designing methods for automated quantification and understanding of fish behavior and fish-reef interactions.
[CV-24] SCDM: Spatial-Contextual Disentanglement Mamba via Differential Inference for Efficient Image Classification
链接: https://arxiv.org/abs/2609.12825
作者: Mustafa Bora Çelik,Hayriye Aktaş Dinçer,Ayse Keles
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages, 5 figures
Abstract:State Space Models (SSMs), particularly VMamba, have emerged as efficient alternatives for modeling long-range dependencies in medical image analysis. However, distinguishing subtle pathological features from visually similar anatomical backgrounds remains a significant challenge. Existing SSM architectures often learn entangled representations, lacking explicit mechanisms to separate disease-specific signals from normal anatomy. To address this limitation, we propose Spatial-Contextual Differential Mamba (SCDM), an asymmetric dual-branch architecture designed for selective representational disentanglement. SCDM introduces a Positive Branch for extracting discriminative features and a Negative Branch that actively models and suppresses normal anatomical context. This separation is achieved through a similarity-driven repulsion gate and a differential inference rule, which promote competitive feature learning without requiring additional branch labels or increasing model capacity. Evaluated on the RSNA Pneumonia dataset, SCDM achieves competitive classification performance (AUC of 0.858) while requiring significantly fewer parameters (29.4M) and FLOPs (1.44G) compared to standard VMamba and vision transformer baselines. Furthermore, activation analyses demonstrate that our differential mechanism yields highly precise localization, effectively isolating lesions by inhibiting irrelevant anatomical distractors.
[CV-25] Online Video Agent Harness for Long Video Understanding
链接: https://arxiv.org/abs/2609.12818
作者: Sen Yang,Boqiang Duan,Jing Yang,Weihao Bo,Jie Liu,Boyuan Tong,Ze Feng,Wenkang Zhang,Jingdong Wang,Hua Wu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 35pages, 12 tables, 10 figures
Abstract:Long video understanding often behaves like a visual needle-in-a-haystack problem: query-relevant evidence is sparsely distributed across long temporal spans, while packing dense frames into a single VLM context incurs \textitcontext rot and high cost. Existing video agents often rely on query-agnostic offline preprocessing or ad hoc tool sets, which can miss query-specific details and waste computation. In this work, we present VideoXAgent, a purely online video-agent harness for long video understanding that starts from the given video file and user query, plans and decomposes the task, invokes specialized expert tools on demand, and aggregates multimodal evidence to produce a final answer while resolving conflicts among observations. To support this on-demand invocation, we design a suite of heterogeneous expert tools guided by a data-driven taxonomy of atomic capabilities, spanning scripts, VLMs, and domain models (e.g., detection, OCR, ASR, face recognition). The harness further enforces objective evidence prompting and budget-aware control to curb hallucination and non-termination. Across Video-MME-Long, LongVideoBench-Long, LVBench, and MINERVA, VideoXAgent is competitive with frontier LMMs and video agents under a smaller context footprint—about 50k tokens of agent context per sample, even on hour-long videos. In particular, on complex video-reasoning benchmarks such as MINERVA, it matches this level while using only about 15% of the context of a 1,024-frame dense-packing baseline. Notably, the harness remains effective with a visually weak or even text-only orchestrator, suggesting that strong long-video understanding can emerge from progressive agentic evidence seeking rather than from packing the full video into a single context. Project page: this https URL
[CV-26] LGFN: Lightweight Gated RGB-Polarization Fusion with Modality-Availability Conditioning for Camouflaged Object Detection
链接: https://arxiv.org/abs/2609.12798
作者: Zhuangfan Huang,Xiaosong Li,Yang Liu,Tao Ye,Haishu Tan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Camouflaged object detection (COD) is an important engineering task in intelligent optical perception, but it remains challenging when targets closely resemble their surroundings. Polarization imaging provides complementary physical cues, whereas existing methods typically assume fixed multimodal input configurations and entangle intra-polarization coordination with interaction between red-green-blue (RGB) and polarization representations. We propose LGFN, a lightweight gated RGB-polarization fusion framework supporting separately optimized RGB-only and polarization-assisted configurations. A deterministic Modality Router selects the appropriate configuration according to polarization availability. In the multimodal configuration, an availability-conditioned Modality Gate calibrates the available polarization branches; the Gated Polarization Hub coordinates learned degree of linear polarization (DoLP) and angle of polarization (AoP) representations with explicit polarization cues; and RGB-Polarization Cross Fusion introduces the coordinated representation into the RGB hierarchy through controlled residual interaction. The multimodal configuration requires neither sample-dependent statistics nor handcrafted quality descriptors during inference. On the complete 230-image PCOD_1200 test set, the RGB-only configuration achieves a mean absolute error of 0.0090, a Dice score of 0.8806, and an intersection over union of 0.8144, obtaining the best results on all six metrics among the evaluated RGB-based methods. Under a common local reevaluation protocol, the multimodal configuration outperforms PolarNet and IPNet on all six metrics. Relative to IPNet, it reduces the parameter count, floating-point operations, and latency by 53.1%, 73.6%, and 63.0%, respectively.
[CV-27] LG-PF: Lightweight Confidence-Guided Polarization Image Fusion
链接: https://arxiv.org/abs/2609.12787
作者: Zhuangfan Huang,Zhenyu Kuang,Gao Wang,Yang Liu,Haishu Tan,Xiaosong Li
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Polarization image fusion combines the stable luminance and structural information of the total- intensity image S0 with the material-sensitive details of the degree of linear polarization (DoLP) image. However, the reliability of DoLP varies spatially, and indiscriminate polarization transfer may amplify unstable responses or disturb the structural appearance anchored by S0. We therefore propose LG-PF, a lightweight confidence-guided framework that formulates polarization fusion as a selective residual transfer process. A Polarization Confidence Prior estimates spatially reliable polarization responses, a Mask-guided Multi-scale Fusion module regulates their transfer across three feature scales, and a Lightweight Context-aware Bounded Correction Head stabilizes local photometric and structural transitions. Confidence guidance is also incorporated into the optimization objectives to preserve reliable polarization details while suppressing unsupported responses. We also construct MSP, a multi-scene polarization fusion dataset containing 1000 pixel-aligned image pairs from 17 indoor and outdoor scene categories. LG-PF achieves the best results across all six evaluated metrics on MSP, while subset-based evaluations on PIF and GAND show promising transferability without fine-tuning. With only 0.2936 M parameters and an inference time of 21.712 ms per image, LG-PF achieves competitive fusion quality with low computational cost. The source code, dataset, and official data splits will be made publicly available upon publication.
[CV-28] MPT: Missing Prototype Tracking via Barycentric Reconstruction in Vehicular Federated Learning
链接: https://arxiv.org/abs/2609.12771
作者: Hanju Jang(1),Gyeongmin Han(1),Sungmin Lee(1),Kichang Lee(1),Chunghan Lee(2),JeongGil Ko(1) ((1) Yonsei University, (2) Toyota Motor Corporation)
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages, 7 figures
Abstract:Cross-vehicle federated learning enables vehicles to collaboratively improve perception models while keeping locally collected driving data private. However, vehicle participation is transient, and a vehicle may depart before training converges while permanently taking its local data. When this departing vehicle holds most samples of a target class, the class becomes rare in the remaining FL network, and its recognition can silently degrade as the shared backbone continues to evolve. Recovering the class is difficult since the few remaining samples provide a noisy prototype estimate, while FL privacy constraints prevent centralized access to raw data or per-sample features. This paper presents MPT, a cross-vehicle FL framework that maintains rare-class recognition by reconstructing its prototype at every round from privacy-preserving class-level statistics. MPT combines a barycentric decomposition that tracks drift shared with remaining-class prototypes, a covariance-based residual prediction that estimates out-of-span drift, and an adaptive calibration that weighs the remaining rare-class samples according to their reliability. We evaluate MPT on three vehicle classification tasks and four backbones against representative calibration and drift-compensation baselines. MPT outperforms all baselines in rare class F1, reaching 0.516 on the nuImages dataset with only 1% of rare-class samples remaining, without raw data, per-sample features, or retraining.
[CV-29] Same Encoder Different Winner: A Paired-View Framework for Cell Painting Encoder Evaluation
链接: https://arxiv.org/abs/2609.12761
作者: Tim Treis,Nikita Moshkov,Johan Fredin Haslum,Shantanu Singh,Fabian J. Theis
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Quantitative Methods (q-bio.QM)
备注: 22 pages, 6 figures
Abstract:Vision encoders for Cell Painting are typically ranked by a single evaluation, commonly replicate mean average precision (mAP). We introduce CP-BG-Bench, a paired-view evaluation framework that holds the central cell fixed across four matched views (raw crop C, segmented S, and density-augmented variants CD and SD), ablating or augmenting surrounding pixels as a controlled intervention. Instantiating the framework on three datasets (JUMP-CP, RxRx1, RxRx3-core) and three encoders (DINOv3 ViT-B/16, OpenPhenom, SubCell) under four community-standard protocols (replicate mAP, scIB batch integration, CellProfiler feature prediction, cross-batch perturbation recall), we find that the four protocols rank the same encoders systematically differently, with disagreements decomposing along three axes: cell versus background, morphology versus context, and within-study versus across-batch. The largest effect: on RxRx3-core, SubCell with segmented inputs retains 94% of crop replicate mAP but only 32% of crop R@10, so the within-study signal preserved under segmentation is largely non-transferable; density augmentation recovers 84% of the within-study C-to-S gap but only 8% of the cross-batch gap. Segmented views predict CellProfiler features as well as or better than crops on two of three datasets, inverting the replicate-mAP ranking, and the C-to-S gap varies by an order of magnitude across datasets while remaining similar across encoders, indicating that background-driven gain is set by experimental design rather than by the encoder. Single-metric ranking of Cell Painting encoders is therefore sensitive to the protocol used, and protocol disagreements are interpretable as projections onto the three axes the paired-view design exposes. We will release the paired-view datasets, reconstruction pipelines, 36 trained checkpoints, aggregated embeddings, and the full evaluation suite.
[CV-30] AquaCubeAI-Powered Monitoring Turbidity on-board Φsat-2
链接: https://arxiv.org/abs/2609.12744
作者: Pietro Di Stasio,Francesca Razzano,Elisa Liparulo,Gabriele Meoni,Nicolas Longépé,Deodato Tapete,Paolo Gamba,Gilda Schirinzi,Silvia Liberata Ullo
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 16 pages, the paper is under review on IEEE JSTARS
Abstract:Timely monitoring of coastal water quality is critical for environmental protection, yet conventional satellite workflows rely on downlink and ground processing, introducing latency that can limit responsiveness to rapidly evolving turbidity events. To address this limitation, we propose AquaCubeAI, a lightweight machine-learning approach for onboard estimation of coastal water turbidity from \Phisat-2 multispectral imagery. By shifting inference from the ground segment to the satellite, AquaCubeAI aims to enable lower-latency, more responsive, and more operationally useful turbidity monitoring under the strict compute and bandwidth constraints of spaceborne platforms. The model is trained on simulated \Phisat-2 acquisitions spatially aligned with Copernicus Marine Service (CMEMS) High-Resolution Ocean Color (HR-OC) turbidity products over selected localized coastal sites spanning four European marine macro-regions. To provide a realistic evaluation of generalization in the presence of spatial correlation, we adopt a spatial block splitting protocol that mitigates data leakage between training and evaluation subsets. The main contributions of this work are: (i) a scalable dataset generation pipeline pairing simulated \Phisat-2 multispectral patches with CMEMS HR-OC turbidity labels across selected localized European coastal sites; (ii) a compact Multi-Layer Perceptron (MLP)-based turbidity regressor trained under a leakage-aware geospatial split and tailored to embedded constraints; and (iii) a reformulation for dense spatial prediction via parameter sharing, enabling turbidity mapping and simple threshold-based anomaly masks for onboard decision logic. Embedded deployment on an Intel Myriad Vision Processing Unit (VPU) further confirms the feasibility of low-power hardware and supports low-latency inference from multispectral inputs.
[CV-31] Physics-Guided Synthetic High-Frequency Ultrasound Generation for Skin Layer Segmentation MICCAI
链接: https://arxiv.org/abs/2609.12735
作者: Junkyung ju,Kyungho Yoon,Minwoo Shin
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注: 12 pages, accepted at MICCAI DH4H Workshop 2026
Abstract:High-frequency ultrasound (HFUS) enables noninvasive visualization of superficial skin structures, but automated skin-layer analysis is limited by the scarcity of densely annotated data. Existing real HFUS datasets commonly provide annotations for superficial targets such as the epidermis and subepidermal low-echogenic band (SLEB), while dense labels for deeper structures such as dermis, subcutaneous tissue, fascia, and muscle are rarely available. We propose a physics-guided synthetic HFUS generation framework for skin layer segmentation. The framework constructs multilayer acoustic skin phantoms, assigns layer dependent acoustic properties, and uses k-Wave simulation to generate paired synthetic HFUS images, dense layer masks, and simulation metadata. To evaluate whether the generated data provide transferable supervision, we use it for downstream segmentation pretraining and fine-tune the models on real Mendeley HFUS data. Synthetic pretraining followed by real fine-tuning achieved real-domain performance comparable to real-only training and improved mean Dice/IoU in three of four evaluated trainable architectures. These results suggest that physics-guided synthetic HFUS images contain transferable anatomical and textural cues for real-domain skin layer segmentation, although further reduction of the synthetic-real appearance gap is needed to enable greater gains. The code and data are available at: this https URL.
[CV-32] GRACE: Adaptive Concept Erasure with Geometry-Guided Retention in Diffusion Models
链接: https://arxiv.org/abs/2609.12731
作者: Qinghui Gong,Yihuai Liang,Yuanlun Xie,Deepak Kumar Jain,Vitomir Štruc,Zhengchun Zhou
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 14 pages,12 figures
Abstract:Text-to-image (T2I) diffusion models inevitably internalize sensitive or non-compliant concepts from large-scale pretraining data, necessitating post-hoc concept erasure. However, existing erasure methods often lack explicit constraints on parameter updates, leading to over-intervention and unintended semantic drift. In addition, many methods rely on manually crafted counterfactual supervision, such as surrogate prompts, which incurs substantial data construction costs that limit scalability to new concepts. To address these limitations, we propose GRACE, a structured concept erasure framework designed to enable localized and selective intervention. Specifically, we introduce a semantically weighted sensitive subspace estimation to precisely lock intervention directions, and employ lightweight subspace-constrained adapters to prevent global semantic disturbance. To eliminate the dependency on manual prompt engineering, we design an automatically decoupled safe-anchor mechanism. To mitigate semantic drift induced by excessive intervention, we introduce an energy-driven dynamic gating mechanism that adaptively controls the timing and strength of intervention at inference. Extensive experiments demonstrate that our method achieves a superior balance between erasure effectiveness and generation fidelity. Compared with the average performance of five state-of-the-art (SOTA) concept erasure methods, our method improves the fine-grained NSFW reduction rate by 17.86% , while reducing the macro-averaged target CLIP Score and preservation-oriented Fréchet Inception Distance (FID) by 4.75% and 50.58% , respectively, indicating stronger concept suppression with substantially improved preservation of the original model’s generative utility.
[CV-33] Multimodal Floorplan Encoding: Learning Dense Modality-Invariant Representations ECCV2026
链接: https://arxiv.org/abs/2609.12723
作者: Xavier Anadón,Rémi Pautrat,Rui Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Presented at the ECCV 2026 TwinWorld Workshop
Abstract:Floorplans arise in many forms, from vector CAD drawings to raster renderings and sensor-derived density maps. This heterogeneity makes it difficult to build learning systems that transfer across modalities and support geometry-centric tasks such as alignment and retrieval. We introduce the Multimodal Floorplan Encoder (MMFE), which maps diverse 2D indoor representations into a shared dense latent grid. MMFE combines a frozen DINOv3 backbone with a trainable Dense Prediction Transformer (DPT) head, and is trained with a per-cell Information Noise-Contrastive Estimation (InfoNCE) objective that aligns spatially corresponding regions across modalities while using all other cells as negatives. To improve robustness to geometric distortions, we incorporate controlled similarity transformations and enforce geometric consistency through feature-grid warping. On Structured3D, a held-out out-of-domain dataset, MMFE improves cross-modal dense matching, enables robust similarity alignment with RANSAC, and yields strong retrieval when paired with learned aggregation.
[CV-34] ExpertHTR: Unified Handwritten Text Recognition with Multi-Task Learning and Sparse Mixture-of-Experts
链接: https://arxiv.org/abs/2609.12705
作者: Dang Hoai Nam,Nguyen Duy Hieu,Quang Huu Hieu,Vo Nguyen Le Duy
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Handwritten text recognition resources are often small and distributed across collections that differ in language, script, document structure, and annotation format, making joint page-level training difficult. We propose ExpertHTR, a unified vision-language framework that addresses this problem through complementary supervision and conditional model capacity. Structural annotations from heterogeneous datasets are first organized through a common Page-Region-Line representation and used to construct four related training tasks for complete transcription, physical-line coverage, text localization, and localized recognition, without requiring additional manual labels. Building on a jointly trained dense model, ExpertHTR introduces a sparse Mixture-of-Experts architecture with an always-active shared branch and conditionally routed full-MLP experts. Sparsegen allows the number of active routed experts to vary with the hidden representation, while routing regularization reduces persistent concentration on a small subset of experts. Experiments on seven heterogeneous handwriting benchmarks show that complementary supervision consistently improves training with page transcription alone, while joint multi-source training provides further gains on most datasets. The proposed sparse expert model further improves the dense baseline on six of the seven sources. The final unified model also substantially outperforms the evaluated general-purpose OCR and vision-language systems on most benchmarks and achieves state-of-the-art performance on the IAM paragraph-level benchmark, while specialized HTR systems remain stronger on several challenging collections.
[CV-35] Semantically Aligned Gradient-Driven Context-Preserving Image Editing
链接: https://arxiv.org/abs/2609.12691
作者: Chiranjeev Chiranjeev,Muskan Dosi,Mayank Vatsa,Richa Singh
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Instruction-guided image editing has a training-time blind spot. Generative editors are never required to semantically verify whether their outputs actually satisfy the instruction. Supervision stops at reconstruction and input textual-level conditioning. This produces incomplete edits, spatial spillover, and poor localization. We present IABEdit, a model-agnostic framework that embeds differentiable semantic verification into training. A frozen vision-language model extracts spatially-aware descriptors from the ground-truth edit. A trainable aligner then reproduces them from the generated output. The residual between the two becomes a gradient that teaches the generator both what to edit and where, with no inference-time VLM cost. IABEdit is compatible with diverse backbones, including U-Net (Stable Diffusion) and MMDiT (FLUX), without altering their inference pipelines. On MagicBrush, it improves structural fidelity by +3.49 DINO-I over the best diffusion baseline and +1.26 over the best overall baseline, while remaining competitive on instruction alignment. It also achieves state-of-the-art instruction adherence performance on RealEdit and EMU Edit benchmarks based on embedding-based metrics. Most consequentially, on the D-LORD surveillance benchmark, it surpasses the proprietary Gemini agent by +5.13 DINO-P under heavy occlusion, where preserving identity is hardest. This shows that gradient-aligned VLM distillation holds up under real-world-like surveillance and occlusion conditions. Human and GPT-4o evaluations confirm perceptually precise, well-localized edits.
[CV-36] NOVA-GS: Noise-Aware View-Consistent Gaussian Splatting for Low-Light Novel View Synthesis CVPR2026
链接: https://arxiv.org/abs/2609.12682
作者: Shaurya Pavan A,Vemunuri Divya Madhuri,Yash Pradeep Gawande,Kaushik Mitra
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to the 3D4S Workshop at CVPR 2026; selected for the Best Paper Award
Abstract:Reconstructing 3D scenes under real-world low-light conditions remains challenging due to severe sensor noise, low signal-to-noise ratios, and degraded photometric consistency, which destabilize geometry estimation and novel view synthesis. Existing approaches often rely on well-lit reference data for reliable Structure-from-Motion (SfM) initialization under degraded inputs or apply per-view enhancement methods that introduce cross-view inconsistencies. To address these limitations, we propose \textbfNOVA-GS, a unified noise-aware framework for low-light 3D Gaussian Splatting that subsumes enhancement, denoising, and geometry optimization within a single process. Our method leverages VGGT-based feed-forward estimation to obtain robust camera poses and geometry directly from degraded inputs, eliminating the need for SfM. Building on this initialization, NOVA-GS integrates three coupled components: a structure-aware enhancement module for exposure correction, a self-supervised denoising module with blind-spot masking for pseudo-supervision, and a consistency-driven Gaussian Splatting optimization enforcing cross-view geometric coherence. We further introduce a noise-guided spherical harmonic regularization to suppress view-dependent artifacts in noisy regions. Extensive experiments on diverse real-world low-light datasets demonstrate improved geometric fidelity, color consistency, and robustness without requiring paired supervision or well-lit references. this https URL
[CV-37] Detecting and Explaining Fake News Short Videos with Multimodal Content and Real-World Evidence EMNLP2026
链接: https://arxiv.org/abs/2609.12678
作者: Yifeng Luo,Yupeng Li,Ming Tang,Jianxiong Guo,Liang Lan
类目: Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM)
备注: Accepted to the Findings of EMNLP 2026
Abstract:Short-video platforms have become a primary news source for the public, which has also enabled the widespread dissemination of fake news videos. We study the task of fake news video detection and explanation (FNVDE). Existing methods face two critical limitations. First, commonly used frame selection strategies may omit veracity-relevant cues or provide insufficient temporal context for understanding news videos. Second, prior methods neglect either multimodal understanding or evidence retrieval. To address these limitations, we propose NVKE-CEI, a unified system that integrates a news video keyframes extraction method (NVKE) and an FNVDE framework leveraging both content and evidence information (CEI). NVKE selects keyframes based on chronological changes in combined visual and OCR-text similarity. CEI employs two specialized LLM-based fact checkers (content-based and evidence-based) whose outputs are fused by a lightweight judge model. Extensive experiments show that NVKE-CEI outperforms state-of-the-art baselines while generating high-quality content-grounded explanations.
[CV-38] Beyond Ambiguous Visual Cues: Studying Physiological Disruptions and Cross-Modal Inconsistencies in Deepfake Videos
链接: https://arxiv.org/abs/2609.12668
作者: Chenxi Yang,Yassine Ouzar,Larbi Boubchir
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Recent deepfake detection studies increasingly suggest remote photoplethysmography (rPPG) signals as an authenticity cue. However, existing benchmarks lack physiological ground truth, and current detectors underexplore the cross-level relationship between facial features and physiological dynamics, often relying on late fusion or rPPG features alone. In this paper, we construct high-fidelity deepfake manipulations on established real rPPG datasets (COHFACE and UBFC-rPPG) to investigate how forgeries disrupt natural physiological signals and facial behavior at the same time. Building on this analysis, we propose a bidirectional co-attention fusion detector that jointly models rPPG and facial behavior tokens. This mechanism explicitly captures the cross-level dependencies between pulse dynamics and facial motion to learn a robust, joint authenticity representation. Extensive experiments using a subject-disjoint 5-fold evaluation demonstrate the superiority of our approach. Achieving a 92.80% AUC on constructed datasets using face swapping and 96.78% AUC on motion transfer, our model outperforms both the rPPG-only single modality baseline and the best feature-level fusion methods. Furthermore, transfer-learning result of the fusion detector on Celeb-DF-v2 while keeping both feature extractors fixed achieves 91.20% accuracy and 86.08% AUC, which suggests applicability under target-domain adaptation.
[CV-39] What Did the MLLM Hear? Token-Level Spectro-Temporal Grounding for Audio MLLM Explainability
链接: https://arxiv.org/abs/2609.12663
作者: Lucia Cascone,Valeria Fraenza,Michele Nappi,Fabio Narducci,Benedetto Simone
类目: ound (cs.SD); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Audio-based Multimodal Large Language Models (MLLMs) can generate detailed natural-language descriptions of complex acoustic scenes, yet it remains unclear which parts of the input audio support each generated token. This is particularly challenging because acoustic evidence is distributed across time and frequency, and concurrent sound events may overlap temporally while occupying different spectral regions. We introduce STAG, to our knowledge the first post-hoc framework for token-level spectro-temporal grounding of captions generated by audio-based MLLMs. STAG estimates the temporal support for each generated token using target-token-specific vocabulary projections of the encoded audio representations, measures frequency-band relevance through controlled spectral occlusion, and combines the two signals into a spectro-temporal relevance map. We evaluate STAG against ten post-hoc explanation methods across four grounding benchmarks, where it achieves the best event-localization performance on every dataset, and apply it to eight audio-language backbones without parameter updates. Counterfactual deletion further shows that removing the identified evidence selectively reduces confidence in the corresponding event and frequently removes it from the regenerated caption. These results provide behavioral support for the faithfulness and selectivity of the explanations.
[CV-40] Geometric-to-Semantic Spherical Transfer Learning for Cortical Sulci Labeling MICCAI2026
链接: https://arxiv.org/abs/2609.12627
作者: Saeb Tounsi,Joël Chavas,Pietro Gori,Vincent Frouin,Denis Rivière,Jean-François Mangin
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注: Submitted to MICCAI 2026 on February 26, 2026; early accepted on May 7, 2026. This is the original submitted version (preprint) of the paper. The final camera-ready version will appear in the MICCAI 2026 proceedings
Abstract:Deep learning on cortical surfaces faces a dilemma: capturing the complex topology of over 60 nomenclature-dependent sulci per hemisphere requires high-capacity models, yet the extreme scarcity of expert annotations ( N=62 subjects) inevitably causes overfitting. Standard supervised approaches fail to generalize in this data-scarce regime, particularly for variable and small sulci where topological ambiguity is high. To overcome this limitation, we introduce a Geometric-to-Semantic Spherical Transfer Learning framework. First, we leverage massive unlabeled data (UK Biobank, \approx 30,000 subjects) to pre-train a spherical encoder using a locally-optimized strategy. By relying solely on continuous surface features (curvature and depth), the relevance of this pre-training is confirmed by the model’s ability to detect localized and rare topological traits, such as sulcal interruptions. The downstream labeling task, however, introduces extracted sulcal fundi (lines) as an explicit semantic input. To bridge this dimensional domain gap (from purely geometric to semantic) without causing catastrophic forgetting, these anatomical lines are integrated into the pre-trained backbone via a soft-initialized Topological Prior Injector. Our experiments demonstrate that this approach outperforms fully supervised baselines trained from scratch, achieving a mean Dice of 0.77. Crucially, a local analysis reveals that the self-supervised geometric priors yield the largest performance gains on variable and tertiary sulci (up to 14.8%), confirming that learning the cortex shape is highly beneficial for identifying its rarest parts. Comments: Submitted to MICCAI 2026 on February 26, 2026; early accepted on May 7, 2026. This is the original submitted version (preprint) of the paper. The final camera-ready version will appear in the MICCAI 2026 proceedings Subjects: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2609.12627 [cs.LG] (or arXiv:2609.12627v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.12627 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-41] RA-SOD: Reliability-Aware RGB-T Salient Object Detection under Modality Degradation ECCV2026
链接: https://arxiv.org/abs/2609.12622
作者: Hongbo Gao,Zhengyu Li,Xueru Nie,Dihao Zhu,Lijun Zhao,Yunke Wang,Chang Xu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at ECCV 2026
Abstract:RGB-Thermal (RGB-T) salient object detection leverages complementary cues from visible and thermal modalities to improve robustness in challenging environments. However, in real-world scenarios, the reliability of each modality is inherently unstable: RGB images degrade under low illumination, motion blur, and noise, while thermal imagery often suffers from contrast compression and sensor artifacts. Such degradation introduces unreliable perceptual evidence that can mislead cross-modal fusion and significantly deteriorate detection performance. To address this challenge, we propose RA-SOD, a reliability-aware RGB-T salient object detection framework that explicitly models modality reliability and integrates it into feature learning and cross-modal fusion. First, we introduce a reliability-conditioned representation that adaptively compensates degraded modality features while preserving structural cues. Second, an uncertainty-guided dual-stream refinement strategy progressively corrects cross-modal representations while suppressing unreliable evidence. Finally, we propose a pixel-wise modality competition mechanism that dynamically selects modality cues according to spatial reliability for fine-grained fusion. Extensive experiments on four benchmarks (VT821, VT1000, VT5000, and VT-IMAG) demonstrate that RA-SOD achieves state-of-the-art performance and exhibits strong robustness under severe modality degradation. Code and models are available at this https URL.
[CV-42] SCORE: SubDistribution-aware Collaborative Knowledge Reinforcing for Cloth-Hybrid Lifelong Person Re-Identification ECCV2026
链接: https://arxiv.org/abs/2609.12577
作者: Kunlun Xu,Liangyu Ma,Jiangmeng Li,Xin Tong,Xiaode Liu,Yufei Guo,Jiahuan Zhou
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accept by ECCV 2026
Abstract:Lifelong Person Re-Identification (LReID) aims to train a unified person retrieval model from a non-stationary data stream. Existing LReID methods mainly focus on scenarios where the clothing of each person is consistent. Recently, the Cloth-Hybrid LReID (CH-LReID) where cloth-consistent and cloth-changing data alternately occur, has emerged as a more practical and challenging scenario. Due to the conflict between clothing-relevant and clothing-irrelevant knowledge, the well-known catastrophic forgetting problem is significantly exacerbated in this task. To address this issue, we propose a SubDistribution-aware COllaborative Knowledge REinforcing (SCORE) framework, where our key idea is explicitly modeling the intra-identity diversity to continually consolidate distinct cloth-consistent and cloth-changing knowledge. Specifically, an Adaptive SubDistribution Modeling mechanism is developed, where a set of distributional subprototypes is assigned to each identity to capture the intra-identity diversity, improving the compatibility between cloth-consistent and cloth-changing knowledge. Then, a Distributional Knowledge Reinforcement scheme is introduced, where the knowledge of old distributional subprototypes is retained in the new ones by a collaborative aligning mechanism. Extensive experiments show that our SCORE achieves the state-of-the-art performance. Our code is available at this https URL Comments: Accept by ECCV 2026 Subjects: Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2609.12577 [cs.CV] (or arXiv:2609.12577v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2609.12577 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-43] An Ultra-Widefield Swept-Source OCTA Dataset and a Polar-Gated Mamba Network for Retinal Vessel Segmentation ALT
链接: https://arxiv.org/abs/2609.12574
作者: Yang Liu,Yibing Shen,Keming Zhao,Cenk Jiang,Zhenghang Qian,Zhicheng Du,Chen Xiong,Qidong Shao,Zijun Lin,Yunqi Hu,Jingjing Zhou,Lian Zhang,Peter E. Lobie,Peiwu Qin,Chengming Yang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Submitted to IEEE Journal of Biomedical and Health Informatics. Under review
Abstract:Ultra-widefield (UWF) swept-source optical coherence tomography angiography (SS-OCTA) enables large-area retinal vascular imaging, yet vessel segmentation at this scale lacks dedicated public benchmarks and comprehensive evaluation for quantitative vascular analysis. We introduce WOIVES, to our knowledge the first publicly available UWF SS-OCTA vessel-segmentation dataset, comprising 206 eyes from 152 participants with a 24x20mm^2 field of view. WOIVES spans emmetropia to high myopia and provides soft probability vessel annotations. We further propose PG-Mamba, a visual state space model that enhances conventional directional scans with two complementary polar-coordinate scan orders. An auxiliary Dynamic FOV Gating module performs spatial modulation at the bottleneck. PG-Mamba outperformed seven competitive approaches on broad segmentation metrics under cross-validation. It achieved the lowest median absolute errors for vessel density, fractal dimension, and vessel length density. WOIVES is publicly available on Zenodo (DOI: https://doi.org/10.5281/zenodo.21904672), and the PG-Mamba code is available at this https URL.
[CV-44] DRS-VPT: Directly Relocalizing in a Scan with Vision Point Transformers
链接: https://arxiv.org/abs/2609.12557
作者: Lanke Frank Tarimo Fu,Maurice Fallon
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注:
Abstract:We present DRS-VPT, a feed-forward transformer architecture for foundational image-to-scan registration. Given query images and a reference 3D point cloud, the model predicts the scan pose and point map alongside the poses and point maps of each camera, all expressed in the first camera’s frame. It additionally predicts a coarse-to- fine pyramid of per-point and per-pixel features for direct reprojective alignment of the scan to the first image. This formulation unifies downstream tasks such as camera-LiDAR calibration in autonomous driving and indoor camera-to-map relocalization. A single DRS-VPT model achieves state-of-the-art performance for image-to-LiDAR registration in autonomous driving, competitive indoor relocalization without training map-specific weights, and strong zero-shot transfer to unseen environments. We also show qualitatively that the model learns complex scan-to-image projection properties such as occlusion of back-facing points.
[CV-45] RelateAnything: Real-Time Open-Vocabulary Relation Prediction From Any Inputs
链接: https://arxiv.org/abs/2609.12552
作者: Maëlic Neau
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Open-vocabulary detection accepts any class list at inference, and promptable segmentation returns regions without class names: the taxonomy has left the model and become an input. Relation prediction has not. Scene-graph models are still trained and evaluated on the 50 or 56 predicates of one annotation style, their relation head conditioned on object labels and so tied to one detector. Three obstacles explain this, none primarily modelling: no relation corpus is both free-text and verified, a label-conditioned architecture cannot accept a vocabulary it was not trained on, and the standard metric rewards agreement with the training corpus, so a larger vocabulary scores as a regression. We present RelateAnything, a 53M-parameter model taking an image and regions from any source and returning scored relations over a predicate vocabulary supplied at inference as strings. Object labels are never an input, so the region source can change without retraining, and the vocabulary is a bank of text embeddings, not a learned classifier. It runs at 20 ms/frame. Training over 19,103 predicates requires positive-unlabeled supervision and a text encoder that separates antonyms, which contrastive encoders embed at cosine 0.95. To supply the supervision we build RA-4M, 474k images and 4.3M relations over 10,102 free-text predicates, generated against numbered box markers and geometrically verified. To measure it we build OV-SGG-Bench, six axes scored across datasets that the priors standard recall rewards cannot satisfy. On three cross-dataset benchmarks and a fourth zero-shot, RelateAnything has 2.3-3.5x the mean recall of the strongest open-vocabulary method of comparable scale, margins that survive a real detector, and leads a 3B-VLM scene-graph model on both metrics at under 2% of its parameters. In-domain measurement overstates transfer gains ~5x. Model, corpus and benchmark are public.
[CV-46] Aligned Radiometric RGB-Thermal Fusion for UAV Facade Anomaly Screening
链接: https://arxiv.org/abs/2609.12521
作者: Yuan Yang,Shulei Li,Haobo Liang
类目: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
备注: 9 pages, 3 figures. Preprint. Submitted to IEEE Sensors Journal
Abstract:Unmanned aerial vehicle facade inspection can combine red, green, and blue (RGB) imagery with thermal measurements to screen surface and subsurface anomalies. However, geometric discrepancies between the sensors and thermal image rendering can obscure spatial correspondence and weak temperature contrasts. This article presents a sensor-level pipeline comprising per-sensor correction, RGB-to-thermal registration, common-support cropping, and signed local contrast encoding of 16-bit radiometric measurements. The encoding preserves the distinction between locally hotter and colder regions and supplies the fourth input channel of a compact single-stream detector. We introduce M3T, a dataset of 674 paired RGB and radiometric thermal samples from five facade-inspection projects covering eight component and anomaly categories. The median residual registration error is 3.384 pixels, and a controlled-displacement analysis characterizes how the local contrast response changes under controlled displacement. Project-grouped four-fold evaluation yields mean average precision of 0.168 over intersection-over-union thresholds from 0.5 to 0.95, using 28.50 billion floating-point operations per image. A separate single-split ablation shows improved delamination detection over RGB-only and alternative thermal inputs, although aggregate accuracy does not improve over RGB alone. Evaluation on RGBT-Tiny shows mixed performance with rendered thermal imagery. These results characterize the category-specific benefits and limitations of aligned radiometric contrast for compact facade screening.
[CV-47] One Skill Does Not Fit All: Automatic Discovery and Taxonomy-Guided Routing of Frame-Selection Skills for Long-Video Question Answering
链接: https://arxiv.org/abs/2609.12517
作者: Jian Hu,Zixu Cheng,Da Li,Wei Li,Ziquan Liu,Shaogang Gong
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: A step toward recursive self-improvement (RSI) in video understanding by enabling multimodal agents to autonomously discover, evaluate, and route reusable skills for long-video reasoning
Abstract:Long-Video Question Answering (LVQA) requires locating decisive evidence in hour-scale videos under a limited frame budget. Most training-free methods apply the same frame-selection strategy to all questions, despite substantial variation in the evidence required by different question types. Our analysis shows that the relative effectiveness of frame-selection strategies varies across semantic categories and benchmarks, motivating adaptive evidence acquisition. In this paper, we introduce AutoSkill, a source-supervised framework for automatically discovering and routing executable frame-selection skills. Starting from a small labelled source pool, LLM agents iteratively propose, implement, evaluate, and refine candidate skills. For a target benchmark, AutoSkill uses only unlabelled question and option text to induce a shared semantic taxonomy, rewrite labelled source examples into the target style, and estimate a category-to-skill mapping. Neither target videos nor target answers are used in this process. At inference time, each question is assigned one skill, which selects the frames used in a single inference of the frozen video MLLM. Across five long-video benchmark splits, AutoSkill improves Qwen2.5-VL-7B and Qwen3.5-4B by 2.4% and 1.2%, respectively, demonstrating the effectiveness of our AutoSkill.
[CV-48] ChitraMiti: Benchmarking Visual Grounding and Modality Reliance in Bengali Geometric Reasoning
链接: https://arxiv.org/abs/2609.12509
作者: Khan Raiyan Ibne Reza,Sanjana Aktar Maria,Sumaiya Tabassum Nimi,Md Adnan Arefeen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Evaluation of vision-language models (VLMs) for multimodal mathematical reasoning remains limited for low-resource languages and for geometry problems that require reading a diagram and a question together. We introduce ChitraMiti-12.8k, a synthetic benchmark of 12,874 Bengali planar geometry problems paired with structured 15-attribute descriptions, and NCTB-500, a complementary set of 500 diagrams manually extracted from Bengali school textbooks. Using a three-phase protocol that separates diagram-only, diagram-plus-description, and description-only inputs, we show across five open-weight and closed-source VLMs that description-only performance is statistically indistinguishable from diagram-plus-description performance, establishing structured descriptions as a sufficient textual proxy for controlled evaluation. Despite this, models remain poor at cross-modal verification, frequently misled by a swapped spatial relation even when they answer the unmodified item correctly. We further evaluate supervised adaptation on ChitraMiti-12.8k, finding that fine-tuning improves performance on both ChitraMiti-1k and NCTB-500, although a substantial gap to the strongest zero-shot model remains. Together, ChitraMiti-12.8k, NCTB-500, and our evaluation protocol offer a standardized way to study Bengali multimodal geometry reasoning and, more broadly, whether VLMs actually check their text against what they see. Our dataset and code are publicly available on Hugging Face at this https URL.
[CV-49] LettuceVisSim: A Simulator That Generates Lettuce Image Time-series for Vision-Based Reinforcement Learning
链接: https://arxiv.org/abs/2609.12505
作者: Ziye Zhu,Bert van 't Ooster,Congcong Sun,Eldert van Henten,Sjoerd Boersma
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Systems and Control (eess.SY); Optimization and Control (math.OC)
备注: 45 pages, 15 figures, Submitted to Biosystems Engineering
Abstract:Vision-based reinforcement learning holds strong potential for decision-making in controlled environment agriculture (CEA). However, its development is hindered by the scarcity of labelled crop images. To address this gap, LettuceVisSim, a lettuce growth simulator that generates labelled time series of crop images, was developed and validated. The simulator contains a process-based model (PBM) for shoot dry weight dynamics, a canopy layout algorithm for deriving canopy layout representations from shoot dry weight, and a Unity rendering engine for image generation. Five findings support the simulator. First, the PBM reproduced shoot dry weight under dynamic plant-density management with \mathrmR^2=0.84 . Second, a piecewise cubic regression mapped shoot dry weight to potential projected area with \mathrmR^2=0.94 . Third, the canopy layout representation was validated using 12 experimental datasets each having different dynamic environmental and spacing conditions. It reproduced the ground coverage ratio dynamics observed in measured images, achieving \mathrmR^2=0.84 when driven by measured shoot dry weight and \mathrmR^2=0.40 (0.76 excluding one outlier) when driven by PBM-simulated values. Fourth, the Unity rendering engine converted canopy layout representations into RGB and segmentation images at less than 10~ms. Fifth, a demonstration showed that a lighting-control policy can be learned and applied by observing only crop images that were generated with LettuceVisSim, providing a proof of concept of vision-based reinforcement learning in CEA using LettuceVisSim.
[CV-50] RoES: Rotational Equivariant Selective-frequency Fusion for Multimodal Images
链接: https://arxiv.org/abs/2609.12497
作者: Jiabao Wang,Wenjian Liu,Yaoming Cai,Gengyu Zhang,Boyan Zhao,Zijia Zhang,Yao Ding,Xiaobo Liu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ACM Multimedia 2026 (MM '26). 10 pages, 6 figures. Code: this https URL
Abstract:Infrared-visible image fusion facilitates robust multimodal perception by integrating complementary textural nuances from visible sensors with thermal signatures from infrared systems. Due to the task’s inherently ill-posed nature, existing methods heavily rely on structural priors but typically enforce rotation equivariance uniformly across all features. Such a holistic approach overlooks a critical distinction where low-frequency shared structures strictly adhere to equivariant constraints while high-frequency modality-specific details require greater flexibility to preserve unique information. To bridge this gap, we propose RoES, a Rotational Equivariant Selective-frequency fusion network. Instead of employing static decomposition, we introduce a trainable rotation-enhanced updater/predictor module to dynamically decouple low- and high-frequency components. The resulting representations are then processed through a dual-branch fusion module tailored for spectral consistency. Specifically, a rotation-equivariant Mamba is employed to capture long-range structural dependencies in the low-frequency domain, while a polar spectral attention-based Dual-Fourier block refines high-frequency details under explicit low-frequency guidance. Extensive experiments demonstrate that RoES consistently achieves state-of-the-art performance in both fusion quality and downstream object detection, establishing a robust solution for multimodal fusion by reconciling frequency-selective features with equivariant constraints. The source code is available at this https URL.
[CV-51] PhysioAI: Clinical Knowledge-Guided Semantic Supervision for Skeleton-Based Physiotherapy Action Recognition
链接: https://arxiv.org/abs/2609.12491
作者: Jie Cao,Euijoon Ahn,Anwar Hassan,Jinman Kim
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Includes supplementary material and ancillary data files
Abstract:Skeleton-based action recognition can support automated tracking of physiotherapy exercises, particularly in remote rehabilitation settings where continuous in-person supervision is impractical. However, most existing methods are developed for large-scale daily-action benchmarks rather than rehabilitation scenarios. Public rehabilitation exercise datasets are typically small, with only subtle kinematic differences between exercise classes. For participants with motor impairments, exercise execution may also deviate from standard movement patterns in amplitude, speed, and coordination, increasing intra-class variability and making reliable recognition more difficult for skeleton-based models. We propose PhysioAI, a clinical knowledge-guided semantic supervision framework that injects structured physiotherapy knowledge into skeleton representation learning. PhysioAI combines graph-based spatiotemporal modelling of human movement with training-time semantic anchors derived from a structured Clinical Knowledge Dictionary (CKD). The CKD descriptions are encoded using a frozen Contrastive Language-Image Pre-training (CLIP) model and projected into an anchor space, where they provide class-specific semantic targets for skeleton representation learning. The resulting CKD-derived anchors are used only during skeleton-model training; inference requires only skeleton inputs. Under subject-disjoint evaluation, PhysioAI achieves 99.03\pm1.34% on KiMoRe Overall, 94.64\pm7.36% on the Hard-67 stress test, and 87.44\pm7.69% on UI-PRMD Overall. These results exceed the strongest comparator for each endpoint by 0.27 , 2.87 , and 1.33 percentage points (pp), respectively. These findings demonstrate that structured clinical knowledge can serve as an effective source of training-time supervision for physiotherapy action recognition.
[CV-52] Partition-Invariant Tuning for 3D Scene Understanding
链接: https://arxiv.org/abs/2609.12473
作者: Hongqiang Lin,Tianle Wang,Shuiwang Li,Dongxu Zhang,Yiding Sun,Zihao Guo,Dongfu Yin
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Scene-level point cloud understanding remains challenging due to diverse geometries and spatial layouts. While pre-trained 3D point cloud foundation models (PFMs) offer strong transferability, full fine-tuning (FFT) incurs substantial computational and storage costs. Parameter-efficient fine-tuning (PEFT) provides a promising alternative, but existing PEFT methods largely focus on object-level point clouds and overlook serialization-induced partition variations in large-scale scenes. To address this issue, we propose PointPiT, a partition-invariant tuning framework for scene-level point clouds. Specifically, a Scene-aware Structural Adapter (SSA) integrates local geometric patterns with global scene context to mitigate partition-induced representation shifts. Moreover, Gradient Subspace Optimization (GSO) selects informative and partition-stable update directions, suppressing partition-dependent variations during optimization. Extensive experiments across multiple scene-level benchmarks demonstrate that PointPiT achieves competitive or even superior performance to full fine-tuning with less than 1% of backbone’s parameters, while achieving consistent state-of-the-art performance among representative PEFT methods.
[CV-53] Bridging Vision Foundation Model Priors with CLIP for Spatial-aware Few-shot Anomaly Detection in Medical Images MICCAI2026
链接: https://arxiv.org/abs/2609.12454
作者: Juzheng Miao,Yuchen Yuan,Cheng Chen,Pheng-Ann Heng
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted by MICCAI 2026
Abstract:Vision-Language Models such as CLIP enable effective few-shot medical anomaly detection (AD) via strong image-text semantic alignment. However, their globally contrastive pretraining lacks explicit spatial supervision, limiting precise lesion localization. In contrast, Vision Foundation Models (VFMs) such as DINO learn spatially coherent patch representations via self-distillation and local-to-global consistency, better capturing fine-grained anatomical structures. Leveraging this complementarity, we propose Spatial-FAD, a spatial-aware few-shot medical AD framework that improves lesion localization by combining VFM spatial priors with CLIP semantics. Specifically, we introduce a VFM-enhanced adapter that injects a structural affinity prior derived from DINO into CLIP features. This structure-guided refinement encourages visual embeddings to better adhere to lesion boundaries while maintaining semantic alignment. To address the loss of spatial detail from patchification and the limited input resolution of CLIP, we adopt a sliding-window aggregation strategy. This generates high-resolution, spatially dense embeddings to further enhance localization granularity. Moreover, we introduce a prototype-enhanced support memory scheme to efficiently exploit the few-shot support set. This module stores compact prototypes for normal and abnormal patterns, reducing memory costs while boosting performance by fusing patch-to-prototype and image-text similarities. Extensive experiments on three benchmark datasets, including Liver CT, Retinal OCT, and Brain MRI, demonstrate that Spatial-FAD significantly outperforms state-of-the-art methods, especially in lesion segmentation. Notably, in the 4-shot scenario, our method achieves an average improvement of over 11.4% in Dice score and 1.8% in AUC. Code is available at: this https URL.
[CV-54] IMPLY: Physically Anchored Consistency for World-Model Rollouts
链接: https://arxiv.org/abs/2609.12441
作者: Aman Mehta,Riya Baviskar
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 7 pages, 2 figures, 3 tables
Abstract:A world model asked what happens if an object is pushed at several speeds produces several futures. If the model has the object in mind, those futures agree about it: each implies the same mass and friction. The consistency checks now used to vet world-action models ask whether a model’s futures agree with each other, and none of them knows any physics. We show that this is not enough, and what to do instead. IMPLY reads the physics each rollout implies by inverting a simulator and scores a set of rollouts by how well one object explains all of them, anchored to two calibration pushes the model has observed. In a controlled setting, self-consistency gives a perfect score to a model that ignores the object and always predicts a typical push; anchoring exposes it (AUROC 0.70 versus 1.00). On a real model, V-JEPA 2-AC adapted to the scene, the same thing happens. Given its own calibration pushes the model tracks the object (per-object correlation with the truth 0.91); given another object’s, it does not (0.05). Self-consistency cannot tell these apart, preferring the right evidence on 52% of objects, chance level, while anchored disagreement prefers it on 73% and correlates 0.92-0.99 with the rollouts’ error. Used to choose among candidate rollout sets, it comes within 0.003 of an oracle that sees the truth. A model that has internalised the wrong object is exactly as self-consistent as one that has internalised the right one; consistency has to be anchored to evidence.
[CV-55] An End-to-End Automated Pipeline for Controllable Crack Data Synthesis
链接: https://arxiv.org/abs/2609.12431
作者: Conghui Li,Muxin Pu,Chern Hong Lim,Weiyao Lin,Xin Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: this paper is under review in Advanced Engineering Informatics
Abstract:Automated crack inspection increasingly relies on deep learning, yet its reliability is limited by scarce and weakly controllable defect data. Existing generative augmentation methods often treat crack synthesis as a generic image-generation task, offering insufficient control over morphology, boundary fidelity, and scene context. This paper proposes an end-to-end automated pipeline for controllable crack data synthesis that formalizes crack geometry and inspection context into reusable computational constraints. First, procedurally sampled Bézier-curve skeletons are translated into realistic crack masks using a GAN, enabling scalable generation of diverse crack morphologies without manual mask design. Second, a dual-ControlNet diffusion framework disentangles appearance guidance from geometric guidance, with an edge-based branch enforcing strict boundary consistency. The framework supports both background-free synthesis and context-aware inpainting. Experiments on CRACK500 and CrackTree200 show consistent gains over existing augmentation baselines, demonstrating a scalable engineering informatics workflow for automated crack-inspection data generation.
[CV-56] RiPPLE: Cross-Space Performance Prediction from Early Training for Neural Architecture Search
链接: https://arxiv.org/abs/2609.12418
作者: Yifan Yang,Zhaoyan Wang,Zheng Gao,Xiaoyu Li,Jiaojiao Jiang
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Neural architecture search (NAS) evaluates candidate networks, but fully training enough architectures to rank an entire space is expensive. Zero-cost proxies score architectures at initialization, yet their ranking quality varies across search spaces. Learned predictors reduce evaluation cost but typically require fully trained labels or partial-training features for individual candidates. We introduce \textbfRiPPLE , \underline\textbfR anking v \underline\textbfi a \underline\textbfP refix- \underline\textbfP ropagated \underline\textbfL abel \underline\textbfE xtrapolation, which treats partial training as a source of labels for a small coverage set of anchors. RiPPLE trains these anchors to an early prefix, extrapolates their learning curves to surrogate labels, and propagates the labels over label-free architecture features. The early-training signal remains a label on the anchors rather than a per-candidate feature. Feature, readout, and encoding rules are selected without held-out accuracy and reused across search spaces. We evaluate the method on twelve benchmark cells from four search-space families and on the larger DARTS space. The results examine ranking quality, label efficiency, architecture selection, and the roles of readout, coverage, and propagation. RiPPLE provides a whole-space ranking from a fractional anchor-training budget, with comparisons interpreted under their respective evaluation and cost protocols.
[CV-57] Spectral Consistency-Guided Multiview Point Cloud Registration for Low-Overlap Scenes
链接: https://arxiv.org/abs/2609.12417
作者: Tianyu Li,Yanghong Lin,Shudong Zhou,Kui Yang,Jingru Zhang,Li Fang,Wei Yao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Multiview point cloud registration is particularly challenging in low-overlap scenes, where reliable correspondences are limited and incorrect pairwise transformations can affect global pose estimation. In addition, registering all scan pairs is computationally expensive because many pairs provide weak geometric information. To address these problems, we propose GMPCR, a non-learning-based spectral consistency-guided framework for efficient and robust multiview point cloud registration. GMPCR builds a refined second-order compatibility structure from initial correspondences and uses its dominant spectral response to evaluate both correspondence reliability and scan-pair confidence. This allows unreliable correspondences to be filtered and informative scan pairs to be selected before relative transformation estimation, leading to a sparse pose graph and reduced pairwise registration cost. For each retained scan pair, maximal-clique-based hypothesis generation is used to estimate reliable relative transformations. The resulting pose graph is further refined by an adaptive history-aware synchronization scheme, in which the effect of residual history is adjusted according to changes in the global rotation residual. A recovery mechanism also allows down-weighted edges to regain confidence when their global consistency improves. Experiments on 3DMatch, 3DLoMatch, ScanNet, and ETH demonstrate the effectiveness of GMPCR. It achieves registration recalls of 97.2% and 89.6% on 3DMatch and 3DLoMatch, respectively, while maintaining competitive performance on ScanNet and ETH. The results show that GMPCR provides a favorable balance among registration accuracy, robustness to low overlap, and computational efficiency. The code is publicly available at this https URL.
[CV-58] DERA: Detached Edge-Residual Adaptation for Prohibited item Detection
链接: https://arxiv.org/abs/2609.12411
作者: Yonathan Michael,Mohamad Alansari,Mohammed Bennamoun,Dwarikanath Mahapatra,Andreas Henschel,Naoufel Werghi
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Prohibited-item detection in X-ray imagery remains challenging due to object superposition, weak texture, and material clutter which obscure both semantic appearance and object boundaries. We propose \textbfDERA, a \textbfDetached \textbfEdge-\textbfResidual \textbfAdaptation framework for prohibited item detection under X-ray imagery. DERA combines hierarchical visual features with a parallel pixel-difference edge pyramid and learns an object-specific boundary prior from training-time contours of the instance masks. The detached prior gates edge-sensitive features, which are injected into the early visual stages through residual heads. This staged design preserves the foundation detector at the start of adaptation, isolates boundary supervision from semantic feature learning, and restricts the final adaptation stage to only (14.7)K trainable parameters. Evaluated on PIDray, CLCXray, and STCray, DERA improves the baseline by \textbf3.1, \textbf1.6, and \textbf2.4 AP points, respectively.
[CV-59] A Multimodal Explainable Deep Learning Framework for Alzheimers Disease Diagnosis using 3D Magnetic Resonance Imaging and Clinical Data
链接: https://arxiv.org/abs/2609.12410
作者: Yusuf Brima,Marcellin Atemkeng,Lakshmana Rao Namamula,Antoine Vacavant
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (stat.ML)
备注:
Abstract:Dementia is a major and growing global health burden, with Alzheimer’s disease (AD) accounting for most cases. Timely and accurate diagnosis is central to managing this burden and increasingly depends on integrating complementary clinical and imaging information. Multimodal deep learning can combine these modalities for AD diagnosis, but how its explanations behave across modalities, fusion strategies, and cohorts remains unclear. We developed an explainable multimodal framework pairing a 3D CNN encoder for T1-weighted MRI with a feedforward network for harmonized clinical and demographic data, comparing varied model setups on three-way and pairwise diagnostic tasks using 6,479 internal records from the ADNI and 1,703 independent records from the OASIS-3. On ADNI, the tabular-only model achieved the highest three-class AUC-ROC of 0.879 and best discriminated cognitively normal (CN) versus mild cognitive impairment (MCI; 0.903), while cross-attention performed best for MCI versus AD (0.861); CN versus AD was highly discriminative overall. On OASIS-3, the vision-only model performed best (three-class AUC-ROC 0.910); CN versus MCI remained difficult, and no fusion strategy consistently outperformed single modalities across tasks and cohorts. SHAP and Integrated Gradients identified the MMSE as the dominant tabular feature in both cohorts, with global feature rankings agreeing strongly in ADNI ( \rho=0.94 ) and OASIS-3 ( \rho=0.96 ); CAM-based explanations, however, changed with model configuration and cohort. These findings show that multimodal performance and explanations are task, modality, fusion, and cohort-dependent: a dominant cognitive signal persisted across cohorts, but feature contributions and CAM explanations did not, underscoring the need to evaluate explainability under cohort shift rather than as a stable, intrinsic property.
[CV-60] OphBiWSSD: Scaling Temporal Action Localization in Ophthalmic Surgeries with Bidirectional Weight-tied State Space Duality
链接: https://arxiv.org/abs/2609.12409
作者: Yang Liu,Qionghong Ma,Joongwon Chae,Lihui Luo,Yibing Shen,Yulin Zhuo,Yingting Zhu,Jiashu Chang,Xiaoyun Zhong,Dongmei Yu,Peter E. Lobie,Peiwu Qin,Chengming Yang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Submitted to IEEE Transactions on Image Processing. Under review
Abstract:High-frequency surgical maneuvers in ophthalmology necessitate high-fidelity temporal modeling, yet characterizing long-range procedural dependencies remains computationally prohibitive for attention-based architectures. Existing models often require aggressive temporal downsampling, which compromises the detection of fine-grained action boundaries and instrument-tissue interactions. To address these scalability constraints, we present OphBiWSSD, a framework that reformulates surgical temporal action localization leveraging Bidirectional State Space Duality. By employing a weight-tied selective scan mechanism that incorporates both preceding and succeeding surgical contexts, our approach facilitates the global synthesis of non-causal temporal cues with linear complexity. This streamlined architecture is well-suited to capture the bidirectional dependencies present in ophthalmic workflows, effectively bridging the gap between local boundary precision and long-range procedural context without incurring the quadratic memory overhead of traditional Transformers. Extensive experiments on the OphNet benchmark demonstrate that OphBiWSSD achieves state-of-the-art temporal localization performance, with mean Average Precisions of 44.42% on phases and 43.08% on operations, surpassing the baselines by 6.80% and 6.66%, respectively. Empirical validation indicates that our approach ensures precise temporal localization and offers a computationally viable pathway for deploying surgical intelligence systems in clinical environments. The code is publicly available at this https URL.
[CV-61] GSO-Net: Visual State Machines for Hazardous Freight Transfer Compliance at Petrochemical Logistics Nodes
链接: https://arxiv.org/abs/2609.12408
作者: Yu Xie,Bangshu Xiong,Zhibo Rao,Rui Gan,Chongxuan Liu,Zechu Ouyang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Hazardous-freight operations at petrochemical logistics nodes are safety-critical for intelligent transportation systems, yet existing vision benchmarks rarely address procedural compliance under realistic deployment constraints. In large infrastructure networks, cameras often operate under sparse round-robin polling, so transfer status must be inferred from incomplete observations and localized evidence. We present GSO-Net, a large-scale benchmark for visual understanding of standard operating procedures (SOPs) in petrochemical unloading scenarios. To our knowledge, GSO-Net is the first public benchmark dataset dedicated to visual SOP understanding in petrochemical hazardous-freight transfer scenarios. It contains over 50,000 independently sampled frames from 64 real expressway petrochemical logistics nodes and adopts an SOP-derived hierarchy linking 9 macroscopic procedural steps with 15 microscopic operational states. Two tasks are defined: joint detection of microscopic states and macroscopic steps as the core benchmark, and frame-level step classification as a diagnostic reference. Experiments with lightweight, transformer-based, open-vocabulary, and holistic models reveal a clear gap between object perception and transfer-stage understanding. Current models remain weak on contact-level state grounding, transient step recognition, and stage consistency, especially under sparse polling, tiny critical targets, and long-tailed operational evidence. GSO-Net provides a practical benchmark for fine-grained state perception and vision-based safety monitoring in hazardous freight transportation. The dataset is publicly available at this https URL
[CV-62] UFO: Chain-of-Evaluation for Omni-Condition Alignment in Multi-Modal Image Generation ICML2026
链接: https://arxiv.org/abs/2609.12397
作者: Danning Zhang,Yijing Lin,Shuhan Zhuang,Mengqi Huang,Shaojin Wu,Shancheng Fang,Zhendong Mao
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted at the Forty-Third International Conference on Machine Learning (ICML 2026)
Abstract:Multi-modal image generation, particularly subject-driven customization, has garnered growing attention in recent years. Despite the rapid advancement of generative models, their evaluation remains largely lagging. Existing methods, whether embedding-based or Multi-modal Large Language Model (MLLM)-based, evaluate alignment with each modal condition in isolation, which contradicts the simultaneous condition alignment objective of multi-modal image generation, leading to poor consistency with human judgments. To address this challenge, we propose UFO, the first unified framework for omni-condition alignment simultaneous evaluation. Specifically, UFO introduces a novel Atomized Chain-of-Evaluation paradigm, \emphi.e., it first decomposes omni-condition alignment into a sequential chain of fine-grained, disentangled Atomic Evaluation Units (AEUs), categorizes them into distinct modality-relevance classes, and then employs general or dedicated functional calls for accurate verification of different AEU types. Experimental results demonstrate that UFO achieves the highest correlation with human evaluation preferences, delivering an average improvement of 15.25%. Furthermore, we present UFO-Bench, a dedicated benchmark designed to holistically evaluate the performance of existing customization models under the diverse mutual interactions of textual and visual conditions.
[CV-63] EgoMaize: A First-Person Maize Instance Segmentation Benchmark under Severe Field Occlusion BMVC2026
链接: https://arxiv.org/abs/2609.12350
作者: Jiayi Li,Zihan Zhang,Erhankang Yan,Yitian Chen,Yuze Li,Chengzhang Ding,Jianxin Cao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to BMVC 2026
Abstract:Close-range first-person field images are important for mobile maize phenotyping because many plant-level traits depend on in-canopy structures that are difficult to ob serve from overhead views. However, post-seedling maize fields create a difficult in stance segmentation setting: stems, leaves, tassels, and neighboring plants are elon gated, repetitive, and strongly occluded. We introduce EgoMaize, a compact benchmark for first-person maize instance segmentation, where the task is to predict ownership consistent plant masks and plant-owned stem/tassel cues from close-range field images with severe same-class overlap. Existing visible-only labels can fragment one physi cal plant into disconnected supervision, while full-amodal labels may require unverifi able completion behind neighboring plants or field objects. EgoMaize therefore uses an evidence-closed annotation workflow for occluded maize regions and assigns unreli able maize regions to ignore rather than background. Baseline results show that pre trained query-based grouping, boundary refinement, and high-resolution crop refine ment help different aspects of the task, but no architecture solves the coupled chal lenges of fine structure recovery, same-class instance ownership, and occlusion reason ing; occlusion-level analysis further shows that performance decreases as plant visi bility becomes more limited. The dataset and code are publicly available at https: //github.com/JaaaaaaaD/EgoMaize.
[CV-64] VS-Splat: Voxel-Selective feed-forward Gaussian Splatting for end-to-end 3D object reconstruction from sparse-views
链接: https://arxiv.org/abs/2609.12343
作者: Yunsu Jeong,Hyuk Heo,Youngsang Kwak,Jaehwa Kwak,Il Yong Chun
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Main paper: 10 pages, 4 figures. Appendix: 4 pages, 5 figures, Transaction of Multimedia
Abstract:Feed-forward Gaussian splatting models have demonstrated remarkable effectiveness in reconstructing three-dimensional (3D) objects from a few two-dimensional (2D) images, even if they are unseen. As existing methods typically predict Gaussian primitives uniformly across the 3D space, most primitives are placed in non-object regions. This may hinder the representation of fine object details. This paper proposes a Voxel-Selective Gaussian Splatting model (VS-Splat), a new end-to-endfeed-forward Gaussian splatting framework that predicts many primitives only within selected voxels that are likely to belong to an object, without 3D structural supervision. To achieve this, we propose a new learnable voxel selection approach that identifies object-centric voxels only with 2D rendering this http URL sparse-view rendering experiments with three benchmark datasets show that proposed VS-Splat outperforms several state-of-the-art methods. We further demonstrate its effectiveness as a backbone for an existing densification method and show that anoptional extension improves its robustness to inaccurate camera pose estimates.
[CV-65] UniMo: Unifying Human and Animal Motion Generation SIGGRAPH
链接: https://arxiv.org/abs/2609.12342
作者: Zeyu Zhang,Zhiyuan Zhang,Siheng Wang,Yiran Wang,Danning Li,Ian Reid,Richard Hartley
类目: Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR)
备注: Accepted to SIGGRAPH Asia 2026 Posters
Abstract:The conditional generation of 3D motion has emerged as a key research topic due to its wide applicability across robotics, AR/VR, gaming, and content creation. However, extending recent advances in text-driven human motion generation to the animal domain remains challenging due to two core limitations. First, animals exhibit highly diverse skeletal topologies, unlike the standard human structure, making unified modeling across species difficult and leading to inefficient per-species models. Second, existing animal motion datasets suffer from limited scale and annotation quality, constraining model performance. To address these challenges, we propose UniMo, a unified point cloud-based motion generation framework that bypasses topological discrepancies by converting parametric skeletons into unparametric representations, further enhanced by dynamic sampling that allocates more points to active joints. Additionally, we present UniML3D, a large-scale motion-language dataset spanning both human and animal categories, containing 145,907 motion sequences and 433,388 captions-over 102x larger than existing animal datasets. Our method achieves state-of-the-art results on UniML3D and three public benchmarks including HumanML3D, KIT-ML, and AnimalML3D, demonstrating the feasibility and effectiveness of unified human-animal motion generation. Website: this https URL.
[CV-66] AnchorVLN: Geometry-Anchored Vision-Language Grounding Reasoning for Open-Vocabulary Navigation
链接: https://arxiv.org/abs/2609.12285
作者: Long Giang Vu,Chengkai Yao,Yuxin Liu,FNU Aryan,Rajath Chandrashekar Aralikatti
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Vision-Language Navigation (VLN) in unseen indoor environments is useful in real-world robotics, where an agent must follow natural-language instructions, locate objects, and answer spatial questions without a pre-built map or fixed object vocabulary. Multimodal vision-language models (VLMs) provide strong open-vocabulary grounding and zero-shot reasoning, but struggle to emit reliable metric quantities such as range, bearing, and comparative spatial relations directly from images. Existing approaches address this by folding geometry into hand-engineered pipelines or asking models to output waypoints, requiring changes to the control stack for different robots, tasks, or vocabularies. We introduce AnchorVLN, an open-vocabulary VLN system built on a simple rule: the VLM proposes semantics; geometry decides metrics. It is realised as EMBODIED-NAV-MCP, a Model Context Protocol (MCP) server driven by a VLM agent through a compact set of callable tools. Since no tool accepts distance in metres or bearing in radians, the schema enforces the semantic-geometry boundary without modifying the downstream autonomy stack. We benchmark both tasks of the CMU Vision-Language Navigation Challenge 2026: 30 instruction-following questions over 15 scenes and a frozen 45-question object-reference set. The full system achieves 64.4 percent on instruction following, dropping by 13.3 percentage points without controller modeling (t = 2.77). On object reference, geometric anchoring clears the challenge overlap threshold on 10 of 45 questions, versus 0 of 45 for direct coordinate estimation, reducing median center error from 3.37 m to 2.48 m.
[CV-67] Revisiting Multi-Object Tracking Baselines: Hyperparameter Optimization with Multi-Fidelity Greedy Coordinate Search
链接: https://arxiv.org/abs/2609.12261
作者: Momir Adžemović
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 23 pages, 7 figures, 12 tables
Abstract:Multi-object tracking (MOT) is dominated by the tracking-by-detection paradigm, whose methods typically rely on a small set of hyperparameters that are conventionally chosen by hand. Tuning them requires repeated expert-guided experimentation, while the procedures used to select reported values are often not systematically evaluated or fully documented. Hyperparameter optimization (HPO) automates this process, yet it remains rarely used in MOT, and existing studies applying HPO to MOT predate modern deep-detector-based trackers and HOTA evaluation. We systematically apply HPO across two datasets and four tracking-by-detection methods. We also propose Multi-Fidelity Greedy Coordinate Search (MFGCS), which optimizes one hyperparameter at a time by first evaluating candidate values on a small subset of scenes and re-evaluating only promising candidates on the full dataset. Across all eight tracker-dataset combinations, the Tree-structured Parzen Estimator (TPE) and MFGCS outperform both our hand-tuned configurations and the corresponding published results, with improvements of up to 4.38 and 16.05 HOTA points, respectively. MFGCS also reaches a predefined HOTA target faster than TPE in seven of the eight combinations. Within each tracker-dataset pair, all optimizers share the same search space and evaluation pipeline, isolating the effect of the search strategy. We release the code and tuned configurations to enable future work to compare against systematically optimized rather than default or manually tuned baselines.
[CV-68] PLSP (Pre-hoc Liminal Space Profiling): OOD Prediction over Detection – An Anticipatory Approach for Machine Learning Model Reliability
链接: https://arxiv.org/abs/2609.12225
作者: Vipul Bansal,Himanshu Buckchash,Balasubramanian Raman,Deepak Dhungana
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
备注: 10 pages, 3 figures, 5 tables
Abstract:Out-of-Distribution (OOD) data poses a significant threat to machine learning models, often leading to model failure during deployment. All existing OOD detection methods are post-hoc, relying on evaluation metrics such as accuracy and AUC-ROC during inference to indirectly assess the model’s response to OOD data by measuring deviations. In contrast to existing approaches, the proposed work shifts the paradigm from OOD detection to OOD prediction by proposing a pre-hoc anticipatory framework called PLSP for OOD prediction. We make several key contributions: (a) a dataset-independent metric called the CREDibility Score (CREDS) is proposed for OOD prediction; (b) credibility curves are introduced to study the maximum credibility a model can attain; and © credibility heat maps (and volume under surface) are introduced to characterize pre-hoc model behavior across different datasets. This work provides a novel perspective on signal processing under distributional shifts. Experiments across multiple datasets demonstrate that the proposed metric serves as a valuable measure for improving the robustness of machine learning models toward OOD prediction.
[CV-69] QuPAINT: Physics-Aware Multimodal Reasoning for Quantum Material Characterization
链接: https://arxiv.org/abs/2609.12202
作者: Sankalp Pandey,Xuan-Bac Nguyen,Hoang-Quan Nguyen,Tim Faltermeier,Nicholas Borys,Hugh Churchill,Khoa Luu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Characterizing two-dimensional (2D) quantum materials by optical microscopy requires localizing exfoliated flakes and determining their layer thickness from subtle optical contrast and interference color to select suitable flakes for device fabrication. However, models face synthetic-to-real domain shifts and variation across materials, substrates, laboratories, and imaging conditions. We present QuPAINT, a physics-aware multimodal framework for transferable quantum flake characterization. The Synthetic Materials Framework (Synthia) generates diverse synthetic microscopy images while preserving layer-dependent optical behavior. Using these images, we construct QMat-Instruct, a multimodal instruction dataset with image-specific reasoning traces generated from verified annotations and constrained to observable optical cues. QuPAINT integrates these signals through Physics-Informed Attention (PIA), which injects substrate-relative optical priors into the visual representation to support grounded multimodal reasoning. For evaluation, we introduce QF-Bench, to our knowledge, the largest real-world benchmark for this problem, spanning diverse microscopy and substrate conditions. Using its verified annotations, we study counting, visual grounding, reasoning quality, confidence calibration, and transfer to an unseen material. QuPAINT-8B substantially outperforms prior methods and establishes state-of-the-art performance for both general and monolayer flake detection. Additional experiments show that image-grounded supervision improves strict spatial grounding and confidence calibration while preserving robust general flake detection on the unseen material.
[CV-70] Physics as the label for measuring and correcting materials reasoning in multimodal models
链接: https://arxiv.org/abs/2609.12181
作者: Hasan Kurban,Rasul Khanbayov,Mustafa Kurban
类目: Computer Vision and Pattern Recognition (cs.CV); Materials Science (cond-mat.mtrl-sci)
备注:
Abstract:Vision-language and language models increasingly interpret materials data, yet benchmarks report that they hallucinate invalid properties and violate physical law. Evaluation matches final answers to scarce human labels, while discovery agents verify final proposals or density functional theory (DFT) execution. Neither measures the physical consistency of a model’s reasoning chain. Materials data carries its own physics, making a large class of materials reasoning verifiable without annotation. We introduce MatPCR, a label-free benchmark whose programmatic oracles check diffraction geometry through Bragg’s law, scale bars, spectral peaks, and Materials Project-grounded checks of near-hull stability, computed band-gap class, and net magnetization. We define the Physical-Consistency Rate over image and structure inputs; introduce Constraint-Grounded Self-Verification, an agentic loop whose gain survives self-refinement and equal-compute re-prompting controls; release an open verifier useful in distribution but near chance on all six held-out constraint types; and derive an exact identity for how oracle error displaces the reported rate.
[CV-71] When Ground-Truth Fidelity Matters: An Orchestrated UAS Framework for Wheat Streak Mosaic Virus Detection Using Vision Transformers and Machine Learning
链接: https://arxiv.org/abs/2609.12169
作者: Dewi Endah Kharismawati,Sandeep Dhakal,Courtney E. McCusker,Jennifer R. Wilson,Erik W. Ohlson,Sami Khanal
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 29 pages, 9 figures, 2 tables. Submitted to Computers and Electronics in Agriculture
Abstract:Wheat streak mosaic virus (WSMV) is a destructive pathogen of sweet corn and other cereal crops, causing yield losses and complicating early detection because symptoms are spatially variable and subtle. In sweet corn seed production, WSMV also has regulatory importance, as phytosanitary regulations from countries such as New Zealand and Chile require seed lots to be certified virus-free. Visual scouting is unreliable because symptoms can resemble abiotic stress, while enzyme-linked immunosorbent assay (ELISA) is accurate but expensive, labor-intensive, and difficult to scale. We present an automated pipeline for plant-level WSMV detection using unmanned aircraft systems (UAS) multispectral imagery. The framework integrates orthomosaic reconstruction, geospatial alignment, plant extraction, and classification using a Vision Transformer with seven-channel inputs (five spectral bands, NDVI, and NDRE). Using treatment-based labels, the model achieved 89% accuracy on over 6,500 test patches across multiple growth stages. However, ELISA-based ground truth revealed substantial label noise: only a small fraction of sampled plants in inoculated plots were infected. Treatment labels therefore did not reliably represent infection status, and the high accuracy was largely driven by label bias rather than disease detection. Performance decreased markedly against row-level symptom severity and plant-level ELISA labels. Under these higher-fidelity but smaller-sample conditions, both deep learning and classical machine learning showed limited generalization and weak separability between ELISA-confirmed mock-inoculated and infected plants. These results show that UAS-based disease detection is constrained by label fidelity and data availability, emphasizing biologically grounded labels and models aligned with real-world conditions.
[CV-72] USPLIT-VQA: U-Shaped Split Learning for Visual Question Answering with Contribution-Aware Weighted Aggregation
链接: https://arxiv.org/abs/2609.12168
作者: Md Khalid Syfullah,Alvi Ataur Khalil
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Visual Question Answering (VQA) systems, jointly interpreting images and natural language queries, hold significant promise across many domains, yet the privacy-sensitive nature of user data creates a fundamental barrier. Centralized training requires access to all data, while federated learning requires each client to host the full model. We propose USPLIT-VQA, a U-shaped split learning framework for privacy-preserving VQA in which each client retains the initial layers and the classification head while the server hosts the computationally heavy intermediate layers, keeping raw inputs and labels on the client device. We further introduce Contribution-Aware Weighted Aggregation (CAWA), a gradientsimilarity-based client scoring mechanism designed to reduce the influence of malicious updates. Experiments on four VQA datasets (VQA-RAD, SLAKE, PathVQA, and VizWiz) with two backbones show accuracy gains over Federated Learning for the Custom model and reduced accuracy for BiomedCLIP under the evaluated fixed split, alongside client memory reductions of up to 5.8X and communication reductions of up to 10.8X. With one malicious client, CAWA reduces the attacker’s influence by over 98%, while experiments at higher corruption levels identify its limitations. Reconstruction experiments further show lower inversion quality under the evaluated attacks.
[CV-73] Single-Query Person-Centric Bimanual Hand-Object Interaction Detection ECCV2026
链接: https://arxiv.org/abs/2609.12155
作者: Jonghyun Kim,Junho Roh,Yubin Yoon,Hyotae Lee,Jongkuk Park,Taehwan Hwang,Jaechul Kim,Jungho Lee
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ECCV2026, Project page: this https URL
Abstract:Understanding person-level bi-manual interactions requires not only detecting hands, but also identifying which two hands belong to the same person and what each hand interacts with. Existing hand–object interaction methods are mostly hand-centric: they treat each hand as an independent instance, which can lead to ambiguous ownership in multi-person scenes. We propose a person-centric formulation in which a single query predicts a structured output for one person, including the human box, body pose, hand boxes and states, and interaction targets. We introduce part-aware deformable attention to allocate attention across human, hand, and pose-specific reference regions, enabling one query to capture the full person structure. We further unify detection and interaction reasoning with a hand-to-query relationship matrix, where each hand selects its interaction target from the detected query set plus a learnable off token, directly recovering the target’s box and class without separate object regression. We build a COCO-based dataset with person-centric bi-manual interaction annotations and define structured metrics for evaluating hand states and complete hand–object tuples. Experiments with a transformer-based detector show that our formulation improves person-level bi-manual interaction parsing and provides an effective unified framework for joint detection, pose estimation, and hand reasoning. Comments: Accepted to ECCV2026, Project page: this https URL Subjects: Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2609.12155 [cs.CV] (or arXiv:2609.12155v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2609.12155 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-74] HSI-Road Relabeled: Surface-Aware Road-Scene Segmentation
链接: https://arxiv.org/abs/2609.12151
作者: Imad Ali Shah,Imran Mehmood,Enda Ward,Martin Glavin,Edward Jones,Brian Deegan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:The HSI-Road dataset provides paired RGB and 25-channel NIR (600–960~nm) images with binary masks but no surface-level labels.~This paper introduces a manually labeled six-class taxonomy: Background, Asphalt, Concrete, Dirt, Water, and Grass, and an RGB-to-NIR registration pipeline with corresponding annotations. Six semantic-segmentation models (SSMs) are evaluated under four input configurations: original-resolution RGB (RGB _\textori ), registered low-resolution RGB (RGB _\textreg ), NIR, and channel-stacked RGB _\textreg --NIR (RGBN _\textstk ). The comparison quantifies the effect of spatial-resolution reduction on RGB, along with evaluation of NIR and RGBN _\textstk , with results reported using per-class and mean IoU and F1 scores. RGB _\textori achieves the highest overall performance but contains 12 \times more pixels than the matched-resolution inputs. At the matched 192 \times 384 resolution, RGBN _\textstk outperforms NIR for all six SSMs and RGB _\textreg for five of six, with the most consistent gains for the Water class. These results highlight the importance of spatial resolution while showing that NIR provides complementary information to RGB.
[CV-75] Beyond Argmax: A Mechanistic Study of Semantic Retention in Frozen Foundation-Model Composition for Generalized Few-Shot 3D Segmentation
链接: https://arxiv.org/abs/2609.12099
作者: Silas Kwabla Gah,Ebenezer Owusu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Classical classifier-combination work distinguishes score-level fusion from hard decision-level voting. We revisit this distinction where independently pretrained, frozen foundation models are composed at inference time for generalized few-shot 3D segmentation. We ask: how much useful semantic information is lost when heterogeneous sources are collapsed to a single class before they can interact? We answer with a same-input semantic-retention intervention. Dense RegionPLC and sparse cross-view SAM3 evidence, model weights, masks, geometry, vocabularies, and fusion rules are frozen; only the number of semantic alternatives retained before interaction is varied via a matched top-k ladder. On 156 held-out ScanNet200 scenes, top-1 reaches 28.47 harmonic-mean (HM) IoU while full distribution fusion reaches 34.87 HM (+6.40, 95% CI [+5.24,+7.64]). The pattern replicates on 50 ScanNet++ scenes: 23.02 vs. 26.50 HM (+3.48, 95% CI [+1.64,+5.93]). The conclusion is robust: full-distribution HM is stable across sparse-source weights 0.3–0.7; alternative operators (max, geometric pooling) also outperform top-1; and a GroundingDINO–SAM2.1 source-replacement diagnostic shows monotonic HM increase from 14.77 to 18.75 with full retention. Calibration diagnostics reveal opposite miscalibration of the two sources, yet correcting calibration does not eliminate the retention advantage. Across datasets and source stacks, most information is recovered by retaining a compact set of plausible alternatives. The contribution is a controlled diagnosis of premature semantic collapse as a repeatable information bottleneck in heterogeneous frozen-model composition. Subjects: Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2609.12099 [cs.CV] (or arXiv:2609.12099v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2609.12099 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Silas Gah Mr [view email] [v1] Thu, 10 Sep 2026 18:25:53 UTC (8,032 KB)
[CV-76] Does Video Memory Use What It Retrieves? A Causal Audit of Memory Specificity
链接: https://arxiv.org/abs/2609.12090
作者: Aditi Tiwari,Akshit Bhalla,Darshan Prasad,Heng Ji
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 18 pages, 9 figures, 15 tables
Abstract:Video models increasingly use memory to preserve information over long sequences, with the assumption that gains come from retrieving and using the correct past content. Standard memory ablations test whether memory helps, but not whether the retrieved content is responsible. We test this directly with read-time memory substitution, which replaces the consumed memory value while leaving the rest of the computation unchanged. This separates memory benefit from memory specificity, the extent to which the gain depends on retrieved content. Across frozen video world models, identity-free controls containing no evaluation-specific content recover essentially the full benefit on Ego-Exo4D and 7-Scenes and about 70% on TUM. In the Ego-Exo4D dose response, recovery falls from 102% to 1% as these values move away from observed training-memory representations, supporting representation repair as the best-supported explanation in this setting. WorldMem shows graded dependence. A wrong memory from the same trajectory recovers 94.1% of the PSNR benefit relative to zero content, while a donor from a disjoint trajectory and biome recovers 43.7%. SAM 2 shows strong content dependence. On DAVIS, replacing the correct spatial memory with a valid wrong memory reduces mean region and boundary score from 0.926 to 0.182. At MOSEv2 reappearance, it falls from 0.459 to 0.000. These results show that memory gains can depend on generic representation support, broader context, or exact episodic content. Read-time substitution provides a direct way to distinguish them.
[CV-77] MoPA: Coordinated Mobile Manipulation via Subsystem-Specific Perception Alignment
链接: https://arxiv.org/abs/2609.12081
作者: Guangyu Chen,Qiwei Liang,Shaolong Zhu,Tianxing Chen,Zikuan Xiao,Yifan Xie,Lingfeng Zhang,Ping Luo,Renjing Xu,Wenbo Ding
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: Website: this https URL
Abstract:Mobile manipulation requires perceptual evidence at different spatial scales for base motion and arm control, while the two action modalities remain kinematically coupled. Existing policies often employ specialized action generation for different subsystems but condition heterogeneous action branches on a shared perceptual representation, leaving subsystem-specific perception-action correspondence implicit. We present MoPA, a framework that aligns perceptual conditioning with mobility and manipulation while preserving coordination at the action level. Dual Perceptual Streams employ two mutually masked query banks to extract separate perceptual representations from a shared vision-language context. Perception2Action Adaptation jointly updates each query bank and its corresponding action stream at every layer of a structured Mixture-of-Transformers decoder, while enabling information exchange between the two action streams. Coupled conditional flow matching learns a joint vector field for coordinated generation of both action chunks. On the ManiSkill-HAB benchmark, MoPA achieves state-of-the-art performance across all three task suites. Across four real-world tasks, MoPA achieves a mean full-task success rate of 76.3%, outperforming the best baseline by 12.5 percentage points. Ablation studies and further analyses validate the effectiveness of the proposed design. Website is available at: this https URL.
[CV-78] Feature Recovery for Object Understanding After Irreversible Fire Damage
链接: https://arxiv.org/abs/2609.12078
作者: Aditi Tiwari,Sofia Stoica,Savya Khosla,David Forsyth,Heng Ji
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 28 pages, 11 figures, 9 tables
Abstract:Objects in post-fire environments often undergo irreversible physical transformations that change their geometry, material state, and visual appearance. Detecting and identifying these remnants is critical for locating hazards, reconstructing pre-incident contents, and inventorying losses. Unlike standard image corruptions, these degradations affect the physical structure of the object itself. To study this setting, we introduce TRACE, a transformation-aware benchmark for post-fire object understanding. TRACE contains 21.4K real-image-grounded synthetic scenes and paired object-level pristine-to-degraded progressions spanning 499 object identities across 189 categories. We define five tasks targeting localization and pre-degradation understanding: degraded-object detection, pristine-state recovery and retrieval, original material recovery, pristine description generation, and functional reasoning. Existing models degrade sharply with severity. From the least to the most severe level, RF-DETR mAP decreases by 71% relative, while InternVL3.5 retrieval R@1 falls from 93.85 to 28.11. To address this, we propose the Feature Recovery Module (FRM), a plug-and-play module that maps degraded encoder features to pristine-aligned representations while keeping the host frozen. Trained only with paired feature supervision, FRM improves scene-level detection, CLIP/SigLIP2 feature recovery, and all four object-level VLM tasks, with larger gains under more severe degradation. Across VLM hosts and severity levels, relative gains average 12.5% for retrieval, 20.1% for material recovery, 13.2% for description generation, and 12.4% for functional reasoning.
[CV-79] Unlabeled Echoes: Pseudo-Labels and Genus-Aware Smoothing for Bat Call Recognition ECCV2026
链接: https://arxiv.org/abs/2609.11986
作者: Frank Fundel,Alexandra Howard
类目: ound (cs.SD); Computer Vision and Pattern Recognition (cs.CV); Quantitative Methods (q-bio.QM)
备注: Accepted at CV4Ecology @ ECCV 2026
Abstract:Passive acoustic monitoring produces far more bat recordings than experts can label. We show that simple model-generated pseudo-labels turn this surplus into effective supervision. We compare pseudo-labeling with other semi-supervised learning methods on an 18-species European corpus using only 10% of its training labels, then transfer the strongest approaches to South African field audio containing nine bat taxa and a nuisance class. Pseudo-labeling outperforms the other semi-supervised learning methods on every European measure, recovering up to 61.5% of the gap to full supervision. It transfers to field audio with gains of 10.69 points in species accuracy and 4.96 points in species macro-F1. We also introduce genus-aware smoothing, which directs uncertain target mass toward congeneric species. Combined with uniform smoothing, it reaches 79.16 species macro-F1, 4.73 points above hard targets. Simple pseudo-labels are therefore highly effective at this ecological data scale, while genus-aware targets inject useful biological structure at no annotation cost. this https URL Comments: Accepted at CV4Ecology @ ECCV 2026 Subjects: Sound (cs.SD); Computer Vision and Pattern Recognition (cs.CV); Quantitative Methods (q-bio.QM) Cite as: arXiv:2609.11986 [cs.SD] (or arXiv:2609.11986v1 [cs.SD] for this version) https://doi.org/10.48550/arXiv.2609.11986 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-80] Scenario-Independent Criticality Assessment and Prediction for Vulnerable Road Users in Autonomous Driving
链接: https://arxiv.org/abs/2609.11947
作者: Jörg Gamerdinger,Victor Schwarzenberger,Philipp Schmid,Sven Teufel,Oliver Bringmann
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV); Computers and Society (cs.CY); Machine Learning (cs.LG)
备注: Accepted at IEEE VTC Fall 2026
Abstract:Increasing safety is the primary objective of automated vehicles. Achieving this goal requires reliable safety metrics that incorporate safety-relevant factors such as object type, velocity, and criticality. A key capability of such metrics is the distinction between critical and non-critical objects, which is addressed through criticality or relevance estimation. Existing criticality metrics are typically designed for specific scenarios and primarily focus on vehicle-to-vehicle interactions. In this paper, we therefore propose a novel criticality metric tailored to vulnerable road users (VRUs), which require special consideration due to their less predictable motion behavior. Furthermore, to avoid the complexity introduced by scenario-specific metrics, we introduce a scenario-independent criticality prediction framework applicable to all traffic participant classes. The effectiveness of both the proposed VRU-centric criticality metric and the criticality prediction framework is evaluated using the DeepAccident dataset, which contains a diverse set of safety-critical traffic scenarios. The proposed VRU-centric criticality metric improves pedestrian criticality classification performance by up to 50 %. In addition, the proposed criticality prediction framework outperforms state-of-the-art metrics by 275 %, achieving an F1-score of 0.96 and enabling scenario-independent criticality assessment across all object classes. These results demonstrate the strong potential of the proposed approaches to enhance criticality assessment for safety evaluation in automated driving systems.
[CV-81] Adaptive AI: Energy Efficient Multi-exit TinyML on Intelligent Vision Systems at the Edge
链接: https://arxiv.org/abs/2609.11939
作者: Luca Crupi,Lorenzo Lamberti,Alessandro Giusti,Daniele Palossi
类目: Hardware Architecture (cs.AR); Computer Vision and Pattern Recognition (cs.CV); Distributed, Parallel, and Cluster Computing (cs.DC)
备注:
Abstract:Traditional TinyML systems for edge devices achieve high accuracy by relying on fixed-depth models that require a constant number of multiply-accumulate (MAC) operations regardless of the input complexity. This approach wastes critical resources in battery-powered Internet-of-Things (IoT) devices and limits the real-time performance of edge cyber-physical systems. Multi-exit execution schemes mitigate these issues and are widely used on high-end devices such as GPUs, but are rarely exploited on edge IoT devices because they require substantial rethinking given their strict memory and computational constraints. We address these aspects by designing and deploying, on an ultra-low-power GWT GAP9 System-on-Chip (SoC), a novel multi-exit computational scheme, demonstrating it on a MobileNetV2 convolutional neural network (CNN) for the ImageNet-100 classification task. Our approach introduces multiple exits at different CNN depths, each with a confidence-based gating mechanism that dynamically and autonomously decides whether to continue or stop inference. Comparing our multi-exit strategy to the standard MobileNetV2 on a GAP9 SoC, we show a 41% reduction in the average computational cost (from 313 MMAC to 185 MMAC), a 29% lower inference time (from 49 to 35 ms), and an energy saving of 24% (from 2.1 to 1.6 mJ per frame). All these improvements come with a ~1% loss in accuracy compared to the full-depth MobileNetV2, which achieves 80.5%. Finally, comparing our adaptable multi-exit scheme with a third-party state-of-the-art adaptive CNN, also deployed on the GAP9, we achieve more than 2x its computational efficiency, increasing it from 8.1 to 17.2 MAC/cycle.
[CV-82] Recurrent Dynamic Range Extension SIGGRAPH
链接: https://arxiv.org/abs/2609.13135
作者: Sebastian Dille,Keru Fu,S. Mahdi H. Miangoleh,Yağız Aksoy
类目: Image and Video Processing (eess.IV); Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR)
备注: 10 pages, 8 figures. Accepted to ACM SIGGRAPH Asia 2026
Abstract:We present an approach to progressively extend the highlights of an image. Instead of reconstructing the full dynamic range of a complex scene directly, we learn a simpler task first: We extend the dynamic range of an input image by a single exposure value. Once this is mastered, we retrieve the full HDR image for the scene by executing our network recurrently, progressively increasing the dynamic range of the input. Our formulation is agnostic to the input dynamic range and targets a bounded output domain. This enables us to use widely available RAW images for the reconstruction task and adapt adversarial losses to construct realistic images. By incorporating Memory Replay for backpropagation, we can train our network recurrently over multiple inference stages and reduce reconstruction errors. As a consequence, our system reconstructs challenging long-tailed HDR scenes robustly and shows powerful recovery of bright light sources and highlights.
人工智能
[AI-0] Rethinking Heterogeneous System Disaggregation for Subquadratic Attention
链接: https://arxiv.org/abs/2609.13134
作者: Arya Tschand,Yaosheng Fu,Vikram Sharma Mailthody,Nicolai Oswald,Po-An Tsai,Ritchie Zhao,Oreste Villa,Vijay Janapa Reddi,Karu Sankaralingam
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Frontier language models are more aggressively using subquadratic attention to reduce the memory footprint and compute requirements during inference while still delivering frontier accuracy. While existing systems make dense attention-centric disaggregated serving decisions, we show that disaggregating inference around the unique arithmetic intensity and memory footprint of subquadratic attention LLMs can achieve significant throughput and energy efficiency gains on emerging DRAM-based and SRAM-only heterogeneous systems. We introduce SQD (SubQuadratic Disaggregation), a fine-grained heterogeneous disaggregation scheme that splits decode by quadratic and subquadratic attention rather than by operator type, and that applies across subquadratic attention variants. For sparse attention LLMs, we disaggregate decode into top-k selection, which must index through the full KV, and top-k attention plus FFN, which have static memory footprints. For linear and sliding-window attention LLMs, we disaggregate decode into dense attention layers and subquadratic attention layers plus FFN. In an adjusted 8xB200 heterogeneous system proxy, we observe average tokens/J improvements of 53% on GLM 5.2, 31% on Nemotron 3 Ultra, and 56% on Gemma 4 31B over the strongest GPU-only baselines. In an analytical model of a Rubin plus LPX system with fixed power budgets, we observe 1.2x to 1.5x tighter achievable latencies and up to 3.6x higher throughput over the best baseline of attention-FFN disaggregation. Our experiments also reveal architectural insights on chip and interconnect provisioning for next-generation heterogeneous systems serving subquadratic attention. Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2609.13134 [cs.AI] (or arXiv:2609.13134v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2609.13134 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-1] A Hybrid LSTM-XGBoost Framework for Multi-Horizon Stock Return Prediction Across Diversified Equity Portfolios
链接: https://arxiv.org/abs/2609.13125
作者: Seif ElDein Mostafa,Yahia Ahmed,Farah Datwish,Marwa Solayman
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Accurate prediction of equity returns remains a major challenge in computational finance due to the non-stationary, nonlinear, and low signal-to-noise ratio nature of financial time series. This paper proposes a hybrid two-stage architecture that combines a long short-term memory (LSTM) network with an XGBoost gradient-boosted regressor for multi-horizon stock return prediction across a diversified panel of 14 U.S. equities spanning six industry sectors. The LSTM component, comprising two stacked layers with 64 hidden units, processes 60-day sliding windows of five sequential market features to produce 64-dimensional temporal embeddings that encode learned sequential market dynamics. These embeddings are concatenated with 14 hand-crafted technical indicators to form a 78-dimensional hybrid feature vector, which is subsequently passed to an XGBoost regressor tuned via 3-fold cross-validation grid search. The framework is trained on a multi-stock pooled corpus using strict chronological splits and per-stock MinMaxScaling to prevent look-ahead bias, and evaluated across four prediction horizons of 30, 90, 252, and 365 trading days. Experimental results demonstrate that the hybrid model achieves a test RMSE of 0.0949 on the 30-day horizon, roughly one-third that of the standalone LSTM baseline, while marginally matching or surpassing the XGBoost-Only baseline across the majority of stocks. Directional accuracy rises with horizon length, reaching 97.6% at 365 days; we show, however, that this largely tracks the high base rate of positive long-horizon returns in the sample, and we therefore benchmark directional accuracy against a naive always-positive predictor and treat the above-base-rate gap at short horizons as the more informative signal. A composite investment scoring framework derived from multi-horizon predictions is further proposed to support portfolio ranking and decision support.
[AI-2] CMA-OT: Hierarchical Expert Supervision for Dance-to-Music Generation ACM-MM2026
链接: https://arxiv.org/abs/2609.13118
作者: Jinting Wang,Chenxing Li,Dong Yu,Li Liu
类目: Artificial Intelligence (cs.AI); Sound (cs.SD)
备注: 12 pages, 9 figures, accepted by ACM MM 2026
Abstract:Dance-to-music (D2M) generation aims to synthesize music that is rhythmically and stylistically aligned with dance videos. A key challenge arises from the semantic mismatch between sparse dance cues, such as rhythm and style, and the dense information required for music composition, including structure, instrumentation, and expressive dynamics. Existing methods typically rely on these sparse cues and supervise only the final audio output, resulting in poorly learned music representations and generated music with limited musicality and structural coherence. To address these issues, we propose Curriculum-guided Multi-scale representation Alignment with scale-aware Optimal Transport (CMA-OT), a novel paradigm that leverages an external music expert to provide hierarchical supervision for the generator’s latent features, bridging the semantic gap and enhancing representation learning. To effectively incorporate hierarchical supervision, we introduce a curriculum-guided multi-scale learning strategy that progressively transfers musical knowledge from the expert to the music generator, enabling stable and effective representation learning. Moreover, to accommodate the semantic and structural variations across different expert scales and achieve fine-grained alignment under temporal mismatch, we propose a scale-aware optimal transport alignment mechanism, which models soft correspondences between hierarchical expert representations and the generator’s latent features. Extensive experiments on two datasets demonstrate that CMA-OT achieves state-of-the-art performance in rhythmic synchronization, perceptual quality, and overall music generation.
[AI-3] ASTRIL-MPC: Autonomous Traversal Framework of Articulated Tracked Robots with Language-Guided Neural-Kinematic MPC
链接: https://arxiv.org/abs/2609.13083
作者: Zhenfeng Gan,Yanbo Chen,Lirong Che,Junbo Tan,Xueqian Wang
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: 9 pages, 8 figures, 3 tables, 40 references
Abstract:In urban search and rescue, articulated tracked robots (ATRs) must traverse structured but contact-rich environments such as stairwells and cluttered building interiors. Reliable autonomy remains challenging because robot-terrain interaction (RTI) is hybrid and discontinuous, and effective flipper-track coordination is difficult to model analytically. We present ASTRIL-MPC, a language-guided neural kinematics model predictive control (MPC) framework for autonomous traversal. A learned kinematics model predicts short-horizon task-state increments from a height sequence and recent trajectories; NMPC plans with multi-objective costs and strict feasibility constraints; and a large language model (LLM) proposes bounded updates to selected weights and bounds through a safety-checked interface with range clipping, rate limiting, and consistency checks. The compiled predictor enables a full control cycle within 100 ms. Across three traversal tasks and a multi-height generalization setting, ASTRIL-MPC improves an aggregate traversal-quality score by up to 71% over a non-adaptive NMPC and by 67% over a PPO baseline, while eliminating measurable collision impacts during descent. These results indicate that combining learned kinematics, optimization-based planning, and language-guided retuning yields data-efficient and robust autonomy for articulated tracked robots.
[AI-4] Embodied-BenchForge: A Closed-Loop Agent ic Workflow for Embodied Benchmark Construction
链接: https://arxiv.org/abs/2609.13082
作者: Baoyang Jiang,Fengchun Zhang,Leyuan Wang,Haotian Li,Yida Wang,Zhe Ji,Jinshan Lai,Xi Ren,Danyang Li,Zheng Yang,Jianwei Hu,Qiang Ma
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Agentic systems offer a promising way to automate embodied benchmark construction, but existing approaches typically cover isolated stages or remain specialized to predefined environments and task families. More importantly, multi-step construction produces dependent intermediate artifacts that are often passed downstream without artifact-specific verification, allowing local defects to propagate into the final benchmark. We present Embodied-BenchForge, an agentic framework that transforms user-specified evaluation intents into complete embodied benchmark artifacts. It formulates construction as Closed-Loop Benchmark Synthesis, integrating forward artifact synthesis with backward verification and repair. Skill-Orchestrated Artifact Synthesis composes typed and reusable skills into executable workflows, while an artifact dependency graph records intermediate outputs and their dependencies. Requirement-Guided Verification and Repair applies artifact-specific contracts throughout construction and uses provenance to trigger local re-execution or upstream rollback when verification fails. Embodied-BenchForge constructs six benchmarks covering diverse embodied scenarios in the Offline EQA Track, together with one interactive benchmark containing 220 executable tasks in the Interactive Embodied Track. Evaluations of representative MLLMs and embodied agents show that the benchmarks distinguish model capabilities in both observation-based understanding and closed-loop execution. Quality assessment and ablations validate benchmark quality and the effectiveness of verification and repair, while repair and skill-reuse analyses demonstrate efficient localized recovery and cross-benchmark reusability.
[AI-5] Anchoring Clinical Events in Time: UID-Preserving Multimodal Reconstruction and Source-Grounded Adjudication ALT
链接: https://arxiv.org/abs/2609.13062
作者: Sayantan Kumar,Nicolas Grimaldi,Jack Cummins,Jeremy C. Weiss
类目: Artificial Intelligence (cs.AI)
备注: Machine Learning for Health Symposium 2026
Abstract:Clinical timelines support treatment-window analysis and leakage-free modeling, but discharge summaries often obscure chronology and structured EHR tables describe only part of the patient course. We present a UID-preserving framework that links each narrative event occurrence to its source span and retains that identity through text-only estimation, structured-evidence retrieval, timestamped source-row grounding, and joint revision. We also present GAVEL, an LLM judge that compares two UID-aligned timelines against the narrative and structured record, to augment prior matching and temporal assessments. Across six open-weight models and 40 mixed-critical-care summaries, the GLM 5.2 multimodal revision, as compared to its text-only variant, improved temporal agreement without reducing event recovery and performed competitively with clinician annotations, while other model revisions showed smaller gains and lower overall performance. Ablations showed that UIDs primarily preserve event retention, whereas source-row linkage supports temporal placement. Blinded human review upheld most GAVEL findings, and controlled adjudication favored multimodal over text-only GLM 5.2 but did not for DeepSeek V3.2. In developing the UID and judge pipeline, we are able to demonstrate 43% increased event recovery, a framework competitive with clinician annotations, and a system with occurrence-level provenance for both reconstruction and evaluation.
[AI-6] Dynin-Robotics: Omnimodal Unified Diffusion Vision-Language-Action Model
链接: https://arxiv.org/abs/2609.13053
作者: Hoeun Lee,Jaeik Kim,Jusang Oh,Jinhyeok Kim,Geon Choi,Hyeonggeun Kim,Jaeyoung Do
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 36 pages, 13 figures, 15 tables
Abstract:Visual goal and dynamics prediction can provide language-conditioned robot policies with both a target outcome and a representation of action-dependent scene changes. We bring these predictions into action generation and selection through a shared trajectory model. Dynin-Robotics implements this formulation on Dynin-Omni, an omnimodal masked-diffusion backbone, representing language, visual observations, goals, and actions as discrete tokens. By varying conditioning and target spans, the same model learns action prediction, action-conditioned next-observation prediction, terminal goal-state prediction, and trajectory-to-instruction reconstruction. These interfaces support test-time scaling through goal prediction, action-candidate evaluation, and joint refinement of action and future-state predictions. We continually pretrain the model on approximately 1.33 million trajectories from 48 Open X-Embodiment datasets and adapt it separately to downstream domains. On two VLABench tasks, robot pretraining improves adaptation within a fixed Stage-2 step budget, and the full objective mixture improves shifted-instruction success over Policy-only post-training under the same coupled decoder. Combining goal guidance with joint action-next-state denoising further improves shifted-instruction success over action-only decoding; the benefit depends on how the predictions are composed. Dynin-Robotics achieves competitive performance on LIBERO and zero-shot LIBERO-Plus, together with a 78.4% average success rate across four manipulation conditions on a Franka Research 3 robot. An optimized block-parallel implementation accelerates model-side action decoding by up to 29.2x relative to the base implementation under the reported profiling setup. These results support shared trajectory modeling as a common interface for learning complementary robot objectives and composing their predictions during control.
[AI-7] Diffusion Models and Concept Formation
链接: https://arxiv.org/abs/2609.13047
作者: Zekun Wang,Karthik Singaravadivelan,Christopher J. MacLellan
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Advances of Cognitive Systems 2026 Oral Presentation
Abstract:Humans organize knowledge into a taxonomy of concepts with nested levels of abstraction and a \emphbasic level at which people recognize and name objects with the least cognitive effort. Cobweb is a classic cognitive account of this ability, an incremental learner that builds a probabilistic concept hierarchy by maximizing category utility. We argue that diffusion models, although designed for image synthesis, implicitly perform the same computation. The noisy marginals of a diffusion model are Gaussian smoothings of the data distribution, and the modes of these marginals form a hierarchy that corresponds to a Cobweb tree of probabilistic prototypes in four respects. Both are hierarchical density models, both are hierarchical-Bayesian models with Gaussian prototypes, both treat categorization as score-following that reduces uncertainty, and in both a basic level emerges. We locate this basic level for a diffusion model at an intermediate noise level, where recent analyses show that the reverse process commits to the class identity of a sample. The two models differ mainly in how they represent and learn the taxonomy. Cobweb learns a discrete tree incrementally, whereas a diffusion model encodes a continuous, interpolable hierarchy in a single learned score field fit to the data distribution. We test the correspondence on MNIST and Fashion-MNIST by recovering the diffusion hierarchy through mode-finding and comparing the basic levels of the two models. This reframes diffusion as a cognitive model of concept formation and offers Cobweb a continuous, scalable instantiation.
[AI-8] DynSHAP: Towards Explainable Dynamic Survival Analysis
链接: https://arxiv.org/abs/2609.13042
作者: Nastasya Anokhina,Jonas Jürß,Pietro Liò
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Deep learning models for dynamic survival analysis (DSA) achieve strong predictive performance by incorporating longitudinal patient data, but their black box nature limits clinical trust and adoption. Existing explainability methods cannot handle longitudinal, irregular inputs and functional survival outputs simultaneously, which limits their usability in DSA. We propose DynSHAP, a SHAP framework suited specifically for dynamic survival analysis. It extends common marginal SHAP estimators to this setting by treating time–feature pairs as players in the Shapley game. We further introduce Temporal DynSHAP, which learns linear dependencies in features over time and uses conditional sampling to address them in explanations. When applied to synthetic data with known ground-truth attributions, Temporal DynSHAP recovers temporally dependent features more accurately than marginal estimators for a given state-of-the-art model. Applied to two real-world clinical datasets and two DSA architectures, DynSHAP produces attributions faithful to model learning, allowing medical experts to see which patient information drove the prediction and when.
[AI-9] Groupoid-Based Internal State Representations for Reinforcement Learning with Local Symmetries AAMAS2026
链接: https://arxiv.org/abs/2609.13035
作者: Ben Opperman,Eduardo Alonso,Esther Mondragón
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: The Eighteenth Workshop on Adaptive and Learning Agents(ALA 2026). AAMAS 2026
Abstract:Symmetries play a central role in reducing the complexity of reinforcement learning problems, yet most existing approaches rely on fixed group actions or predefined state abstractions. Classical reinforcement learning algorithms typically assume a globally structured Markov decision process with uniformly applicable actions and transitions, an assumption that limits their ability to exploit modularity and local, context-dependent regularities present in many realistic environments. We propose a reinforcement learning framework using groupoids to capture local, state-dependent symmetries and support the dy- namic discovery of equivalence structures during interaction. The agent maintains orbit representatives together with transporters that map raw states to canonical forms, enabling learning and decision-making to be performed in a symmetry-reduced space while preserving local distinctions. Empirical results demonstrate that the proposed groupoid-based approach improves sample efficiency and convergence in dense and large-scale environments exhibiting strong partial symmetries, yielding substantial performance gains over standard Q-learning. These findings show that dynamically exploiting local symmetry provides a practical and mathematically principled route to scalable and generalisable reinforcement learning.
[AI-10] Attention Quantization for Tabular Foundation Models
链接: https://arxiv.org/abs/2609.13031
作者: Jonas M. Kübler,Benjamin Jäger,Klemens Flöge,Noah Hollmann,Frank Hutter
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:With the recent rise and adoption of tabular foundation models, optimizing their inference performance becomes an emerging field for efficiency research. While the models are architecturally similar to transformer-based large language models (LLMs), the size and serving patterns differ significantly. We show that the focus should be on the attention calculation and less on weight or KV cache quantization, which are more popular in LLMs. We develop a quantization strategy for queries, keys, and values to FP8 and use explicit FP8 matrix multiplication instructions to speed up the attention calculation. We find that it is crucial to align the quantization error in the test rows with the quantization error in the training rows, as otherwise the accuracy drops drastically. Our Triton kernel achieves a speedup up to 1.7x over regular 16-bit kernels, and we show that on TabPFN-v3 and TabICLv2 there is no relevant accuracy loss across TabArena and BeyondArena.
[AI-11] Comfort by Construction: Adaptive Comfort-Bounded Action Spaces for Learned Driving Policies
链接: https://arxiv.org/abs/2609.13011
作者: Anna Rothenhäusler,Daniel Jost,Raghu Rajan,Faris Janjos,Oliver Scheel,Andreas Look,Joschka Boedecker
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Data-driven driving simulators command accelerations and steering rates from a fixed grid without constraining the realized accelerations and jerks. As a result, reinforcement-learning policies inflate safety metrics through abrupt, last-second maneuvers that lie far outside the range of human driving and would be unacceptable to occupants of a real vehicle, so the metrics measure simulator permissiveness rather than policy quality. Enforcing comfort bounds naively is not enough: lateral limits shrink quadratically with speed, so clamping a static grid saturates it and destroys fine-grained control (“grid collapse”). We propose an adaptive action parameterization that rediscretizes the grid at every step to span exactly the per-step feasible control set, via closed-form inversion of the lateral-jerk constraint. We further present PufferDrive-Editor, a browser-based tool to audit realized kinematics and author kinematically challenging scenes. On the Waymo Open Motion Dataset and a hand-authored slalom, our adaptive model holds comfort violations below 1% while outperforming clipped-grid and direct-jerk baselines in navigability.
[AI-12] How Good Are Frontier Models at Physics? Expert Re-Grading Reveals Broken Evaluations and Near-Saturation of Leading Benchmarks
链接: https://arxiv.org/abs/2609.13009
作者: Ali Ansari,Haoran Sun,Andy Zeyi Liu,Mark Jabbour,Yongshan Ding,Steven Girvin,Yu He,Sohrab Ismail-Beigi,Aleksander Kubica,Owen D. Miller,Corey O’Hern,Vidvuds Ozolins,David Poland,A. Douglas Stone,Frank C. van den Bosch,Logan Wright,Navid Akbari,Santanu Antu,Kangle Cai,Andrew Calabrese-Day,Mateo Cárdenes Wuttig,Meng Cheng,Barry T. Chiang,Ali Ghorashi,Shouzhen Gu,Haoyang Huang,Zhibo Kang,Lukas Kienesberger,Hantian Liu,Charles Lomba,Zhongling Lu,Wenchao Ma,Rohin E. McIntosh,Evan McKinney,Ivan Rojkov,Xulei Sun,Yarone Meir Tokayer,Naveen Balaji Umasankar,Mira Varma,Leda Wang,Qimin Wang,Tyler Wang,Haoyu Wei,Jinming Yang,Jinchen Zhao,Sherlock Tingrui Zhao,Qinyuan Zheng,Jay S. Zou,Lucas Baker,Arman Cohan,John Sous
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Low reported scores on leading physics benchmarks, including those featured in the Artificial Analysis Intelligence Index (2026), suggest that frontier language models still struggle with advanced physics, a demanding test of their scientific reasoning and quantitative problem-solving abilities. Yet this impression does not always align with domain experts’ experiences using these models in their work. We revisit these reported findings by evaluating frontier models on six widely used physics benchmarks and auditing them with experts, focusing on text-only problems with verifiable final answers. For each subfield of physics, faculty and graduate researchers with relevant expertise carefully review problem statements, reference solutions, and model responses to distinguish genuine model errors from grader errors, incorrect reference solutions, and ambiguous or underspecified questions. Most audited cases initially evaluated as incorrect reflect these benchmarking issues rather than errors in the models’ physics reasoning. We then ask experts to address these benchmarking issues by correcting erroneous reference solutions and repairing or excluding flawed questions. We find that GPT-5.6-Sol’s measured mean@4 rises from 47.3% to 78.7% on HLE-Physics and from 61.0% to 87.2% on CMT-Benchmark, while its corrected pass@4 reaches 94.4% on the 54 retained CritPt challenges. Corrected scores are computed on the retained evaluation subsets following expert review. Scores on the audited subsets of UGPhysics, PRISM-Physics, and PHYBench also rise substantially after correction. These findings suggest that current benchmarks substantially understate frontier models’ ability to solve well-posed physics problems. Near-saturation on these closed-ended tasks highlights the need for more demanding, expert-validated evaluations.
[AI-13] SeqMoE: Toward Full-Load Performance via Predictive and Graph-Compatible MoE Offloading
链接: https://arxiv.org/abs/2609.12978
作者: Zihan Wang,Yuqi Wang,Lei Gong,Cheng Tang,Wenqi Lou,Teng Wang,Chao Wang,Xuehai Zhou
类目: Operating Systems (cs.OS); Artificial Intelligence (cs.AI)
备注:
Abstract:Mixture-of-Experts (MoE) creates a structural advantage for offloading: only a small fraction of activated experts need to reside in device memory, and if they can be loaded in time for computation, offloading can in principle approach full-load performance, where all model weights reside in device memory. Yet translating MoE’s structural advantage into practical offloading gains remains challenging. We propose SeqMoE to bridge this gap. To maximize expert hits, we build predictive memory management: (i) Sequence-to-sequence prediction. We are the first to recast expert activation prediction as sequence modeling, enabling accurate multi-step, multi-layer forecasts that provide a long and reliable window for downstream decisions. (ii) Joint prefetch scheduling. We formulate prefetch scheduling as Job Sequencing with Deadlines to maximize expected expert hits and improve bandwidth efficiency. (iii) Forecast-driven caching. Leveraging the recursive nature of sequence modeling, we introduce a probabilistic Belady policy for future-aware eviction. To eliminate execution bottleneck, we develop (iv) Graph-compatible offloading runtime. We derive general runtime principles encompassing compute-transparent expert placement and synchronization-free orchestration disciplines for end-to-end graph capture. With 45% expert residency, SeqMoE averages a 96.97% hit rate and 80.22% of full-load performance, advancing the state of the art in MoE offloading.
[AI-14] EduFair-Bench: Evaluating Pedagogical Fairness of LLM Tutors Across Student Demographics
链接: https://arxiv.org/abs/2609.12949
作者: Jiaxu Zhao,Bahar Radmehr,Fares Fawzi,Tanya Nazaretsky,Tanja Käser
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models (LLMs) are increasingly deployed as tutors, but it is unclear whether they support all students equally well. We introduce \textbfEduFair-Bench, a benchmark for auditing the pedagogical fairness of LLM tutors—whether tutoring quality varies systematically with student demographics. EduFair-Bench pairs a multi-domain question bank (mathematics, physics, chemistry) with a controlled simulation in which a fixed LLM student interacts with each tutor across nine demographic levels spanning four dimensions: gender, immigration background, first language, and socioeconomic status (SES). Tutoring quality is scored on five turn-level pedagogical metrics and four conversation-level dimensions, using an LLM judge validated against three-annotator consensus on 180 tutor turns. Bias is measured via paired Wilcoxon signed-rank tests and bootstrap effect-size confidence intervals. Two ablations (demographic cues conveyed through names; conflicting demographic information between tutor and student) disentangle tutor-driven from student-driven bias. Across five tutors, we find that model capability and demographic fairness are largely orthogonal: the smallest model is the most consistent while the four more capable tutors all exhibit wide demographic gaps with no clear capability-to-fairness ordering, pedagogy-specific RL training redistributes rather than removes bias, and language- and immigration-related cues produce larger gaps than gender- and SES-related cues.
[AI-15] ARC: Autonomous Robotics Compliance A Three-Layer Governance Architecture for Deployed Autonomous Systems
链接: https://arxiv.org/abs/2609.12932
作者: Tord Eide,Einar Holt
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注: 14 pages
Abstract:Proposed governance framework for autonomous robotic systems, introducing a three-layer compliance architecture (ARC) instantiated through model safety validation, cognitive certification benchmarks, and operational authorization standards.
[AI-16] racing and Coordinating Cross-Layer Influence for Multimodal Model Merging
链接: https://arxiv.org/abs/2609.12897
作者: Pengyang Zhou,Xiaobin Tu,Zhengxi Liu,Rongkun Xue,Haochen Li,Miancan Liu,Ziyuan Chen,Yinggui Wang,Jinkui Ren,Xiantao Zhang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Multimodal model merging aims to consolidate task experts into a single model that retains their complementary capabilities. Most unimodal model merging methods combine expert updates within individual layers, and multimodal approaches largely follow this design. However, an expert update changes the representations passed to subsequent layers, allowing its influence to propagate across depth and affect how visual and textual information interact. When visual and language updates are combined, later updates act on inputs already modified by earlier ones, coupling their effects. This poses two challenges: (1) how to characterize the multimodal influence of individual expert updates across depth, and (2) how to jointly combine expert updates based on their multimodal influence. To address these challenges, we propose TAC-Merge for tracing and coordinating cross-layer influence in multimodal model merging. It contains two modules, i.e., multimodal influence mapping (MIM) and coupled merge control (CMC). MIM constructs graphs of update effects and uses Ricci curvature together with expert predictions to define a shared fusion objective. CMC models interactions among coefficient adjustments and jointly optimizes regional weights to synthesize one shared model. Experiments across diverse multimodal tasks demonstrate the effectiveness of TAC-Merge in consolidating complementary expert capabilities and supporting generalization to unseen tasks.
[AI-17] Behavior Quotient Learning for Low-Rank Adaptation of LLM Agents
链接: https://arxiv.org/abs/2609.12896
作者: Pengyang Zhou,Xiaobin Tu,Zhengxi Liu,Rongkun Xue,Haochen Li,Miancan Liu,Ziyuan Chen,Yinggui Wang,Jinkui Ren,Xiantao Zhang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:LLM-based agents rely on heterogeneous interaction capabilities to accomplish complex tasks. Existing approaches often distribute these capabilities across multiple LoRA adapters, which increases adapter storage requirements and introduces routing overhead during inference. A single LoRA avoids this overhead, but learning from diverse agent trajectories under a fixed rank budget presents two challenges. First, trajectories with different interaction traces and parameter gradients can induce equivalent changes in decision distributions, causing repeated updates to overemphasize redundant behavioral changes. Second, an aggregated update may exceed the rank budget of the adapter, and approximating it in weight space can distort the decision changes that it is intended to produce. We propose BQ-LoRA, a low-rank adaptation framework that organizes trajectory updates through a local behavior quotient manifold. It contains two modules, i.e., behavior quotient balancing (BQB) and decision preserving compression (DPC). BQB constructs the quotient manifold from decision distributions and reweights trajectory update directions according to their local density in the quotient tangent space. DPC projects the balanced gradient onto the intrinsic fixed rank tangent space and refactorizes the resulting target by jointly controlling effective weight error and distortion of decision distributions. Experiments on AppWorld and BrowseComp-Plus compare BQ-LoRA with standard LoRA and recent low-rank adaptation methods, while separate ablations evaluate the complementary contributions of both components.
[AI-18] Large Distant Gradients Need Not Be Reliable: reliability-weighted credit assignment for long-horizon autoregressive forecasting
链接: https://arxiv.org/abs/2609.12890
作者: Junhao Zhao,David Michael Simberg,Jacob Kang,Colin Connor Kurniawan,Nan Xu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 35 pages, 9 figures
Abstract:In autoregressive forecasting, long prediction rollouts provide distant supervision, but backpropagation through time (BPTT) carries gradients from those losses through many autoregressive steps. Repeated Jacobian products can make distant gradients dominate the update while amplifying predictable signal and unpredictable noise together; a large distant gradient therefore need not carry reliable learning signal. Motivated by this observation, we introduce Internal Dual-Wiener routing (Internal-DW), a principled backward-only intervention that preserves the full forward rollout and all horizon losses while reliability-weighting internal gradient routes. At each residual block, we derive bounded Wiener gains for the identity and nonlinear routes that balance preserving predictable learning signal against suppressing unpredictable variation, and estimate them from route-level gradient statistics and an explicit noise model. In a controlled system with known gradient signal-to-noise ratio (SNR), we show that distant gradients can grow even as their SNR falls, and that Internal-DW reduces held-out error in recovering predictable gradient signals and improves forecasting. On four history-dominated, weak-drive testbeds, Internal-DW reduces forecast error by 5.2%-13.8% relative to full BPTT, outperforms gradient clipping and Jacobian regularization on all four, and outperforms validation-selected truncated BPTT (TBPTT) on three. It also extends or preserves the fitted optimal training-horizon range across these four testbeds. Across the full benchmark suite, the current Internal-DW estimator has a clear applicability boundary: its benefit diminishes or reverses when usable history is limited or when the selected sampler fails to represent dominant drive-dependent variation. The results show that retaining long-horizon supervision does not require trusting every backward contribution equally.
[AI-19] MedRoundsQA: A Persona and Difficulty Aware Evaluation for Multi-Turn Medical Consultations
链接: https://arxiv.org/abs/2609.12851
作者: Youssef Mohamed,Ahmed Heakl,Qinrong Cui,Junhong Liang,Rafiq Ali,Bdour Babillie,Nazira Dunbayeva,Lang Gao,Omar Hussein,Ahmed Nada,Ahmed Mohamed Magdy Mohamed,Jinghui Liu,Salman Khan,Imran Razzak,Yuxia Wang,Xiuying Chen
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Medical benchmarks are dominated by single-turn, multiple-choice clinical cases that poorly reflect real consultations. Practically, clinicians elicit evidence interactively and patient communication varies widely. We introduce MedRoundsQA, a multi-turn diagnostic benchmark derived from 1,387 board-exam cases across 17 specialties. Each case is converted into a structured 24-slot clinical record, and then instantiated as controlled doctor-patient dual-agent dialogues under varying patient personas, with the underlying clinical content held fixed. We further classify cases by difficulty using model-based uncertainty to enable easy-to-hard analysis. Evaluations of fifteen LLM doctor agents show that (i) moving from a single-turn diagnosis on the standardized records to multi-turn consultations causes large degradations of roughly 13-39 points; (ii) more turns reliably improves question relevance, but diagnostic accuracy exhibits diminishing returns and typically plateaus after 6-12 turns; and (iii) patient persona differences can shift diagnosis accuracy by about 7-8 points (lowest to highest education), highlighting equity risks that single-turn benchmarks miss.
[AI-20] A Graph-Based Approach for Mapping Kernel-Level Telemetry to MITRE ATTCK
链接: https://arxiv.org/abs/2609.12841
作者: Matteo Lupinacci,Luigi Arena,Francesco Blefari,Angelo Furfaro
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注:
Abstract:Mapping observed system behavior to standardized frameworks like MITRE ATTCK is essential for threat-informed defense, but remains largely manual. Existing automated methods depend on Cyber Threat Intelligence reports, which offer only retrospective accounts of attacks. Low-level telemetry, i.e. kernel-level system calls, instead provides evidence of adversary behavior, yet its volume and complexity have limited its use for automated mapping. We present a methodology that collects kernel-level events via eBPF, correlates attacker commands into a provenance graph, and derives compact graph representations suitable for LLM-based reasoning. These representations are mapped to the MITRE ATTCK framework using both pure LLM prompting and retrieval-augmented generation (RAG) grounded in the ATTCK knowledge base, producing ranked technique candidates along with supporting rationales. We implement this methodology as an end-to-end pipeline, named Trace2ATTCK and evaluate it on 347 Linux Atomic Red Team tests using locally deployed open-weights LLMs. RAG consistently improves ATTCK mapping performance over pure prompting, while provenance graph substantially outperforms raw telemetry. These results show that local inference over graph-based behavioral descriptions can make automated ATTCK mapping from kernel-level telemetry operationally viable, without compromising data confidentiality.
[AI-21] Evaluating Context Segmentation in Locally Deployable SLMs for Cybersecurity CTF Tasks ESORICS2026
链接: https://arxiv.org/abs/2609.12839
作者: Sebastiano Nordio,Michele Lotto
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: Accepted at the RAISE 2026 Workshop (ESORICS 2026). Non-archival
Abstract:The proliferation of highly capable open-weight Small Language Models (SLMs) democratizes access to advanced cybersecurity capabilities, posing a escalating risk as these models can bypass proprietary API guardrails when deployed locally. However, SLMs deployed as autonomous agents often struggle with long-horizon, exploratory tasks like cybersecurity Capture The Flag (CTF) challenges due to context bloat and cognitive degradation from accumulated tool-call outputs. To understand and mitigate this cybersecurity threat, we introduce \textitcontext segmentation, a two-level agentic framework that divides complex exploitation tasks into manageable, contextually isolated sub-problems. Evaluating on the \textttpicoCTF dataset using memory-constrained \textttgemma-4 models, we demonstrate that for the E4B model, our strategy acts as an intelligent search, achieving competitive rewards with superior token efficiency compared to brute-force retries, and successfully solving 18.52% of tasks that standard agentic execution fails to complete. Code is available at this https URL.
[AI-22] Scaling Clinical Judgment to Evaluate Medical AI
链接: https://arxiv.org/abs/2609.12822
作者: Thomas A. Buckley,Zahir Kanjee,Peter G. Brodeur,Byron Crowe,Anthony M. Pettinato,Aashna P. Shah,Adrian D. Haimovich,Liam G. McCoy,Daniel Restrepo,Ethan Goh,Jonathan H. Chen,Laura Zwaan,Katherine E. Goodman,Daniel J. Morgan,Raja-Elie E. Abdulnour,Adam Rodman,Arjun K. Manrai
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Blinded physician evaluation has been considered by many to be the gold standard for assessing clinical reasoning in large language models (LLMs). This is difficult to scale; thus, prior studies typically rely on small physician panels, often from a single institution or specialty, which both limits the scientific questions investigated and makes it unclear whether findings would be reproduced with a different set of evaluators. To more rigorously and scalably study clinical reasoning in AI models, here we introduce PrecepTron, an LLM fine-tuned for physician-level evaluation of open-ended responses. PrecepTron was trained using low-rank adaptation (LoRA) of a 32-billion-parameter model on a small number of physician examples. We also release GRAND-ROUNDS, a new large-scale physician-annotated benchmark of 9,217 scores by 11 physicians across seven studies. We show that frontier LLMs in typical “LLM-as-a-judge” approaches often disagree with physicians and with each other, but fine-tuning PrecepTron on a small number of cases enables physician-level consistent scoring across tasks. We use PrecepTron to reproduce headline findings from five influential studies assessing LLMs for clinical care in JAMA, Science, and Nature Medicine without new human grading. Using PrecepTron, we then pose new questions about how LLMs reason in medicine that would have been infeasible with human grading alone, including measuring the diagnostic accuracy of frontier LLMs when clinical cases are provided piecemeal, even token by token. Together, PrecepTron and GRAND-ROUNDS provide a foundation for reproducible, large-scale study of how LLMs reason in medicine. All code, data, and labels are made freely available for researchers.
[AI-23] RunningTensor: Generalizing Linear Attention to Higher-Order Recurrent States
链接: https://arxiv.org/abs/2609.12814
作者: Luca Herranz-Celotti,Vincent Guigue
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Linear attention and state-space models provide linear-time sequence modeling, but their recurrent memory remains a second-order tensor (a matrix), limiting the order of interactions that can be represented in the state. We introduce the RunningTensor, which generalizes this memory to an order- o tensor, updated by a rank-1 outer product and read by contracting against o-1 vector queries. Order 2 recovers linear attention; we study order 3 as a proof of concept, retaining both recurrent and parallel forms while remaining linear in sequence length T and improving working memory capacity from \mathcalO(W^2) to \mathcalO(W^o) . On synthetic multi-query associative recall, RunningTensor outperforms linear-attention and SSM baselines. After pretraining, it also improves performance on language-understanding and non-synthetic retrieval tasks, suggesting that higher-order recurrent state can provide useful additional memory capacity beyond matrix-valued state.
[AI-24] K-Bench: A Benchmark for LLM Unlearning in Agent ic Deployments
链接: https://arxiv.org/abs/2609.12808
作者: Guangsheng Yu,Yanna Jiang,Qin Wang,Baihe Ma,Xu Wang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Unlearning benchmarks such as TOFU and MUSE certify forgetting by reading the model’s final answer, where a model that refuses to answer already counts as having forgotten. We show that this model-level certificate does not transfer once the model is deployed as an agent. We introduce K-Bench, a benchmark that scores LLM unlearning under agentic deployment. K-Bench inspects all six channels a ReAct agent exposes, including its chain-of-thought (CoT), tool calls and tool observations, and elicited summary. A query counts as leaked if the secret appears in any of them. Each experiment places the secret in exactly one of the agent’s three sources (the weights, the prompt, or the retrieval store). The K-Score is computed separately for each source and credits forgetting only when the agent remains usable. Clearing the answer channel does not make the secret unrecoverable. On structured retrieval, the secret stays verbatim in the tool-observation channel and the aggregate leak rate is unchanged. When the secret lives in the prompt or the retrieval store, TOFU and MUSE report no leakage, while the deployed agent still leaks it on 22–86% of queries. When the secret is in the weights, none of the twenty evaluated published methods demonstrably removes it, and only an input-corruption intervention reaches selective forgetting under the evaluated observer. The top-ranked method changes across base models. A refusal-tuning method resists the evaluated extraction without verified knowledge removal.
[AI-25] Interpreting the predictions of neural network classification based on a Taylor Coefficient Analysis (TCA)
链接: https://arxiv.org/abs/2609.12801
作者: Markus Klute,Artur Monsch,Lars Sowa,Roger Wolf
类目: Artificial Intelligence (cs.AI); Data Analysis, Statistics and Probability (physics.data-an)
备注:
Abstract:We introduce a rigid and comprehensive taxonomy and paradigm for characterizing the influence of the input feature space X on the predictions \haty of a neural network (NN) used for event classification, based on a Taylor expansion of \haty in X . The complete process of introspection we refer to as Taylor Coefficient Analysis (TCA). Based on two simplistic example tasks, which can be easily understood and bencmarked, we illustrate the power of the TCA when it comes to revealing, what properties of X have led to what value of \haty , of a given NN model, building up intuition for the method. A more complex application is meant to represent X of a typical classification task at a CERN LHC experiment. Based on this application, we play through the different levels of introspection that the TCA offers and discuss a number of practical aspects for a TCA application typical for the analysis of CERN LHC data. We conclude with a study to support the assumption that those properties of X most relevant for tasks of the complexity typical for a CERN LHC experiment, are usually caught by a TCA up to the second order.
[AI-26] GraphAHA: Graph-Based Adaptive Search with Heterogeneous Actions for Test-Time Code Generation
链接: https://arxiv.org/abs/2609.12757
作者: Xitao Li,Haijun Wang,Gege Yuan,Qiyuan Wu,Jiali Wei,Ming Fan,Xiaofei Xie
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:Test-time scaling improves code generation by spending additional inference budget (e.g., calls or tokens) on direct sampling, feedback-conditioned repair, and reasoning-guided implementation. Search-based methods can allocate this budget adaptively, but two challenges remain. First, tree-structured search treats each generation history as a separate state even when trajectories converge to the same program, duplicating evaluation and preventing statistics from being shared. Second, sampling, repair, and reasoning have complementary and state-dependent payoffs, making online allocation among them difficult under a finite budget. To address these challenges, we propose an adaptive graph search method with heterogeneous actions (GraphAHA). GraphAHA organizes the test-time code generation in a typed directed acyclic graph. Equivalent programs are merged into a single code node, allowing their downstream search statistics to be reused across all discovery paths. Hierarchical Thompson sampling then selects whether to generate a new state or follow an existing successor and, for generation, chooses among the type-valid sampling, reasoning, implementation, and repair operations. Evaluated on LiveCodeBench and CodeContests with Qwen2.5-Coder and DeepSeek-Coder, GraphAHA achieves the best score in 18 of 20 cases. For Pass@1 measured using visible tests, it outperforms the strongest baseline for both models on both benchmarks by 4.1 percentage points on average, demonstrating more effective use of a fixed inference budget.
[AI-27] SCQ: Stabilizing Conservative Q-Learning with Sigmoid-Bounded Entropy
链接: https://arxiv.org/abs/2609.12749
作者: Xiefeng Wu,Shu Zhang,Zhaojie Chu,Mingyu Hu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Offline-to-online reinforcement learning reduces interaction cost for real-world robot learning but suffers from persistent value estimation instability. Existing methods address this through pessimistic regularization, lower-bound calibration, and architectural normalization, but an overlooked source of instability lies in the entropy formulation: the standard log-entropy term can become negative, destabilizing policy updates. We introduce SCQ (Sigmoid-Bounded Conservative Q-Learning), which replaces this term with a sigmoid-bounded formulation that stays strictly positive. SCQ retains conservative Q regularization and return-based lower-bound calibration, stabilizing policy optimization without sacrificing exploration. We evaluate SCQ on D4RL (Minari) benchmarks under both single-demonstration and standard dataset settings, as well as on simulation and real-world visual tasks. SCQ matches or exceeds baseline performance while exhibiting more stable training dynamics across state-based and visual benchmarks, and transfers to four real-robot platforms including manipulation, wheeled, quadruped, and humanoid systems. A direct clipping intervention that removes negative log-probability contributions, together with gradient-matched positive-score controls, indicates that positivity rather than a particular score shape alone drives much of the improvement. Project website: this https URL.
[AI-28] Assisted Spatial Cognition Through Vision-Language Models
链接: https://arxiv.org/abs/2609.12747
作者: H. Riaz,J. B. Fernandez,I. Mills,D. Hickey,F. Cleary,M. I. Ali
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Multimodal AI, powered by Large Language Models (LLMs) and Vision-Language Models (VLMs), is transforming assistive technologies by enabling simultaneous processing of visual and textual data. This advancement holds significant promise for over 43 million visually impaired and neuro-divergent individuals worldwide who face persistent challenges in navigating indoor and outdoor environments due to limited spatial awareness and insufficient environmental cues. Existing navigation aids often lack comprehensive 3D scene understanding, relying on constrained route-based strategies that hinder user autonomy. In this paper, we introduce a novel end-to-end framework that integrates LLMs, VLMs and digital twin technologies to deliver a spatially cognitive navigation support for visually impaired and neuro-divergent users. Our system captures video input via standard mobile phone cameras, and employs SLAM3R to generate dense 3D point clouds from monocular RGB sequences in real-time. Our custom post-processing algorithm ensures accurate point cloud alignment across multiple viewpoints without requiring predefined reference points. This enhances the capabilities of SpatialLM to produce structured 3D representations, including architectural elements and oriented object bounding boxes. The enriched spatial data is then processed by a locally deployed LLM, which interprets 3D contexts to generate detailed scene descriptions and precise distance measurements between users and surrounding objects. We evaluated our approach across diverse video scenarios featuring various perspectives, looped walking views and captured in multiple environments. The evaluation results demonstrate consistent accuracy in 3D scene interpretation and object localisation, underscoring the potential of our system as a transformative assistive navigation solution that combines advanced visual perception with spatial reasoning
[AI-29] Skill Issue: Lessons from Optimizing Repository SKILLs for Coding Agents
链接: https://arxiv.org/abs/2609.12742
作者: Mykhailo Kozyrev,Andrei Kozyrev,Anton Podkopaev
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Coding agents increasingly read repository knowledge from SKILLs — plain \texttt.md files versioned alongside the code. Recent work synthesizes these files automatically, by optimizing the document against a benchmark. A bare repository comes with no benchmark, and the synthetic tasks prior work builds are small enough that a capable agent saturates them with no document at all. We mine harder tasks — merged pull requests of the repository, reverted at a single frozen base commit; and score a candidate document by whether the same agent does better with it than without it. On three Kotlin repositories, the documents GEPA finds raise this score by 4.9 pp on average, and the ones SkillOpt finds leave it where it started, 0.1 pp above the seed. The GEPA gain matches what prior work reports with the same optimizer, and at the dataset size a single repository supplies it cannot be separated from the agent’s run-to-run variance; settling that would take more tasks than one repository’s history yields. The documents themselves read better than the score: a maintainer of one repository found in them knowledge one only gets by working in the project.
[AI-30] When Rubrics Fail: Hallucinations Reveal Blind Spots in Medical AI Evaluation
链接: https://arxiv.org/abs/2609.12718
作者: Griffin Farrow,Lily Sijia Li,Jack Johnson,Tingyan Wang,Philip Torr,William Bolton,Fabio J. Fehr
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Hallucinations can undermine clinician trust in LLMs, making it important that evaluation methods capture clinically relevant errors. Rubric-based evaluation has become the leading approach for assessing LLMs in medicine, but it is unclear whether rubric scores reflect such errors. We first study this in a controlled setting using MedHallu, finding that more specific rubrics better distinguish correct from hallucinated responses. To test this systematically, we develop a taxonomy of medical hallucination types and a clinician-validated error-injection pipeline that creates matched correct and error-injected responses. Across HealthBench, HealthBench Professional, and LiveMedBench, our clinically relevant hallucinations are missed by rubrics, often leaving scores unchanged. We find that rubrics are most effective when explicitly checking facts, and are less effective for additional or unexpected errors they do not anticipate. A preliminary retrieval-based factuality check recovers some of the rubric-blind errors, suggesting a complementary approach. These findings reveal systematic blind spots in current medical evaluation of LLMs and suggest that rubric scores alone are insufficient to establish clinical reliability, potentially undermining clinician trust and confidence in clinical deployment.
[AI-31] Supermartingale Certificates for Parametric MDPs
链接: https://arxiv.org/abs/2609.12715
作者: Kaushik Mallik,Ðorđe Žikelić
类目: Logic in Computer Science (cs.LO); Artificial Intelligence (cs.AI); Systems and Control (eess.SY)
备注:
Abstract:We consider the problems of formal verification and synthesis in parametric Markov decision processes (MDPs) with general measurable state and action spaces. The heart of our approach is a parameter flattening transformation, which allows us to transform parametric MDPs into semantically equivalent non-parametric MDPs. Building on this transformation, we introduce the novel notion of parametric supermartingale certificates, which generalize the traditional supermartingale certificates—used for non-parametric MDPs—to the parametric setting. We use our parametric supermartingale certificates to design algorithms for verification and approximate synthesis in polynomial arithmetic parametric MDPs. This leads to the first verification and synthesis algorithms for parametric MDPs with general state and action spaces. We implement our algorithms and experimentally evaluate them on several continuous parametric random walk benchmarks.
[AI-32] InRTL: Effective Intra-Inter Interaction Learning for Relational Tables
链接: https://arxiv.org/abs/2609.12712
作者: Weichen Li,Ken Zhong,Zheng Wang,Li Pan,Jianhua Li
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Relational table learning has recently emerged as an important research direction for modeling multiple tables connected through primary key-foreign key (PK-FK) relationships. Despite recent advances, a principled modeling framework tailored to this task remains underexplored. In this paper, we propose Intra-Inter Relational Table Learning (InRTL), a unified framework that explicitly models dependencies both within and across relational tables. Specifically, InRTL formalizes two complementary interaction patterns: intra-table interactions, describing associations among rows within the same table, and inter-table interactions, describing dependencies between rows across PK-FK-linked tables. To model these dependencies, we develop a column-aware table encoder to generate initial row representations, followed by Transformer-based self-attention and cross-attention modules for intra-table and inter-table learning, respectively. To further improve scalability, InRTL incorporates linearized attention and heterogeneous graph neural networks to simplify the self-attention and cross-attention operations. Extensive experiments on ten datasets covering 24 real-world tasks demonstrate the effectiveness of our approach. Code is available at this https URL.
[AI-33] What is the Difference Between Me and You? Benchmarking the Quality Gap Between Human-Written and AI-Generated Code
链接: https://arxiv.org/abs/2609.12708
作者: Cristina Improta,Pietro Liguori,Domenico Cotroneo
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: Preprint. This manuscript is currently under peer review
Abstract:AI coding assistants are becoming co-authors of production software, yet their evaluation centers on functional correctness, leaving open whether their code differs from human code in the quality dimensions dominating lifecycle cost. We compare human-written and AI-generated code at scale: 787,562 function pairs across Python, Java, and C, each human function mined from open-source repositories paired with implementations generated from its docstring by three AI assistants (OpenAI GPT models, DeepSeek-Coder, Qwen2.5-Coder). We characterize structural complexity and statistical naturalness, and map static-analysis findings onto Orthogonal Defect Classification for defects and the Common Weakness Enumeration for vulnerabilities, making authors and languages directly comparable. AI-generated code is structurally compressed and stylistically templated: roughly half the size and branching of human code, clustering apart at the style level. Defect profiles differ in kind: human code concentrates issues of mature codebases, AI code repetitive boilerplate; security is language-dependent, with LLMs producing more, and more severe, findings in Python and Java but fewer high-severity memory-safety findings than humans in C. Once size is controlled for, complexity metrics carry little signal, while naturalness separates authors. Finally, we release CQBench, a benchmark of 27,346 issue-prone tasks with baselines and an evaluation pipeline for quality assurance and security testing.
[AI-34] Implicit Personality Representations in Humans and LLM s EMNLP2026
链接: https://arxiv.org/abs/2609.12704
作者: Yilin Geng,Omri Abend,Eduard Hovy,Lea Frermann
类目: Artificial Intelligence (cs.AI)
备注: Findings of the Association for Computational Linguistics (EMNLP 2026)
Abstract:A century of psychology has found that the trait words people use to describe one another vary, but the relational structure among those traits, which ones go together and which oppose, is strikingly consistent across raters and cultures. We test whether the LLM (Qwen 2.5-7B-Instruct) reproduces this structure in its internal trait representations. From millions of crowd-sourced personality ratings of fictional characters, we build a human implicit-personality matrix over hundreds of traits; from contrastive model activations, we build a matching matrix over the same traits. The two relational structures align strongly (Mantel r = 0.77), and the agreement holds trait by trait as well as in aggregate. Two dominant axes of the model’s trait representations recover the social and intellectual dimensions long known to organize human personality impressions, social warmth and intellectual competence. On held-out dialogue, projecting model activations onto these directions yields personality profiles that agree with human ratings. This work establishes a framework that enables comprehensive, human-grounded comparison between internal model trait geometry and the shared structure of human personality impressions.
[AI-35] Separating Engineering Reasoning from DEXPI Serialization in LLM -Based Greenfield Surface-Process Design: A Three-Case Study for Underground Gas Storag e
链接: https://arxiv.org/abs/2609.12656
作者: Qingchuan Zhu,Shuyue Tong,Pengju Ren
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models can produce engineering descriptions and structured process representations, but standards-level serialization can substantially increase the generation burden. This diagnostic study examines whether separating engineering reasoning from Data Exchange in the Process Industry (DEXPI) serialization changes where representation and engineering failures occur in constrained greenfield surface-process design for underground gas storage. We compare Direct DEXPI generation with generation of a lightweight Engineering Intermediate Representation (IR) on three cases: single-pressure injection, withdrawal and export, and dual-pressure injection. All six conditions use one fixed model snapshot, qwen3.8-max-0902, with one completed hosted generation per condition. Direct outputs are XSD-valid in 2 of 3 cases, while all 3 Engineering IR outputs are structurally valid under a minimal validator. Direct prompt inputs contain approximately 121.8k-121.9k tokens, compared with 617-699 tokens for the Engineering IR prompts. Engineering feasibility does not uniformly favor the IR: one IR output is rejected for an explicit cooling-state contradiction. The cases also reveal two distinct Direct DEXPI failure modes: engineering inconsistency and standards-level serialization failure. The observed comparison shows that, in these evaluated method bundles, deferring DEXPI serialization substantially reduces representation burden and helps isolate serialization failure, but reducing representation burden alone does not eliminate engineering inconsistencies.
[AI-36] Explaining Time Series Forecasting with Horizon-Resolved Attribution
链接: https://arxiv.org/abs/2609.12639
作者: Seunghan Lee,Jun Seo,Jaehoon Lee,Junhyeok Kang,Sangjun Han,Sungdong Yoo,Minjae Kim,Tae Yoon Lim,Dongwan Kang,Hwanil Choi,Soonyoung Lee,Wonbin Ahn
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Recent advances in explaining time series (TS) models have produced methods that identify which past values a prediction depends on. However, most existing methods return a single importance vector, assuming that every predicted step depends on the same past values. In this paper, we show that this assumption does not hold, as different forecast steps depend on different past values. Motivated by this observation, we propose Horizon-Resolved eXplanation (HRX), which adds a horizon axis to the explanation, so that every forecast step receives its own importance map. HRX is a simple yet effective plug-in framework with three components: 1) an estimator that reads these maps out of any differentiable forecaster without modifying the TS backbone, 2) an evaluation protocol that validates the horizon axis by measuring how much a single forecast step changes when the inputs an importance map ranks highest are removed, and 3) a rank criterion that predicts in advance whether the axis is worth resolving on a given TS. We further show that this step-wise dependence is low-dimensional, as the explanations of all steps are built from a few shared maps whose number does not grow with the forecast length. Extensive experiments across various backbones and datasets show that the improvement comes from the horizon axis and holds for estimators of previous explanation methods. Code is available at this https URL.
[AI-37] Correlation-Guided Fast Machine Unlearning via Hessian Analysis
链接: https://arxiv.org/abs/2609.12620
作者: Ayushi Thakur,Ruchir Gupta,Amit Kumar Jaiswal,Prayag Tiwari
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 13 pages, 10 tables
Abstract:The increasing adoption of machine learning in network and distributed security systems has created an urgent need for mechanisms that can selectively and efficiently remove the influence of specific training data to eliminate compromised or adversarial data points from production models. Privacy regulations such as GDPR’s \emphright to be forgotten also pose similar requirements. However, existing approximate unlearning techniques remain computationally prohibitive for deployment in real-world security systems, as they require repeated expensive Hessian-inverse-vector computations for each data point removal, creating a bottleneck when processing multiple related requests in scenarios such as intrusion detection systems, spam filters, and threat intelligence platforms. Thus, we introduce a computationally efficient unlearning framework that identifies correlated data points in the training set and applies a theoretically derived closed-form parameter update rule, achieving an 82\times wall-clock speedup over standard influence function unlearning while preserving model utility with a 10^-2 improvement in accuracy over state-of-the-art baselines. Our method establishes theoretical guarantees and ensures numerical stability through Hessian damping. Our evaluation across seven diverse dataset architecture combinations, including large-scale CIFAR-100 with ResNet-50, demonstrates superior forgetting effectiveness, with membership inference attack success rates of 0.660 and tug-of-war scores of 0.950.
[AI-38] Direct Preference Density Alignment for Conversational Audio Equalization
链接: https://arxiv.org/abs/2609.12607
作者: Ioannis Stylianou,Sven Ewan Shepstone,Jon Francombe,Pablo Martinez Nuevo,Zheng-Hua Tan
类目: ound (cs.SD); Artificial Intelligence (cs.AI)
备注:
Abstract:Large Language Model alignment typically relies on learned proxy reward models, which significantly increase the memory footprint during training and are notoriously prone to instability and reward hacking. While offline methods like Direct Preference Optimization (DPO) bypass the reward model, they lose the ability to perform online exploration. If no optimization constraints are applied, this can lead to format collapse in bounded, continuous spaces. To resolve this, we propose Direct Preference Density Alignment: An alternative framework that removes the need for a learned proxy reward model while strictly preserving the benefits of online reinforcement learning. We leverage large-scale user data (approximately 90,000 samples) to construct non-parametric preference density maps, establishing an empirical reward surface. In addition to removing the reward model, Direct Preference Density Alignment enables the combination of the online structural grounding of Group Relative Policy Optimization (GRPO) with the targeted offline refinement of DPO. We show that this GRPO+DPO combination achieves the highest performance, and in a blind audio equalization listening test, enables a 1.5B-parameter model to achieve perceptual parity with a carefully prompt-engineered GPT-4o mini baseline, using only a fraction of the inference compute.
[AI-39] Beyond Generation and Accuracy: Diagnosing and Enhancing Visual Chain-of-Thought for Geometry Problem Solving
链接: https://arxiv.org/abs/2609.12606
作者: Zhitong Dong,Jicai Pan,Yingguo Gao,Jingting Ding,Hao Chen,Jinjie Gu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:While multimodal reasoning has advanced rapidly, solving complex geometry problems critically hinges on active visual assistance, such as constructing auxiliary lines, spurring the rise of Visual Chain-of-Thought (VCoT). However, existing evaluations typically assess visual generation quality and final answer accuracy in isolation, failing to examine whether intermediate visual aids are geometrically valid, effectively utilized in subsequent reasoning, or causally responsible for task success. To bridge this gap, we introduce GeoVAD-Bench, a diagnostic benchmark that pairs a fine-grained five-dimensional trajectory diagnosis covering perception, auxiliary quality, utilization, deductive reasoning, and final correctness with controlled No-Aux, Auto-Aux, and GT-Aux intervention settings to systematically isolate intermediate error modes, the causal gains of visual aids, and the resulting autonomy gap. Our findings reveal that while high-quality auxiliary aids offer substantial theoretical gains for geometric problem solving, autonomous generation is frequently hampered by compounding errors across geometric perception, faithful visual manipulation, visual-state grounding, and deductive reasoning. Guided by these diagnostic insights, we establish a specialized data construction pipeline encompassing geometric perception, diagram editing, and interleaved visual-textual reasoning trajectories, and develop a progressive SFT and multimodal RL training framework. The resulting model, GeoWeave-8B, outperforms the base model by +25.3% in final geometric accuracy and achieves a +30.4% gain in process average across the four intermediate diagnostic dimensions.
[AI-40] SIMS: Scale-Invariant Merit-Function-Based Scalarization for Multi-Task Learning KDD KDD2026
链接: https://arxiv.org/abs/2609.12599
作者: Zebin Chen,Fei Xing,Yang Chen,Hua Liu,Andy HF Chow,Yuhua Qian,Yu Zhang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Optimization and Control (math.OC)
备注: Accepted by the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2026)
Abstract:Multi-task learning (MTL) requires navigating unavoidable trade-offs among competing objectives. This paradigm is frequently formulated as multi-objective optimization (MOO), where the scalarization is favored to reduce an MOO problem to a single objective. We empirically find that existing merit-function-based scalarization approaches are sensitive to the relative scales of different objectives in practical MTL, where task losses commonly differ by orders of magnitude. The optimization process often favors objectives with larger scales even though the underlying Pareto optimal solutions remains invariant to rescaling (i.e., multiplying an objective by a positive constant). To address this issue, we propose Scale-Invariant Merit-function-based Scalarization (SIMS) for MTL. Specifically, SIMS adopts a transformation-induced merit function to convert the MOO problem of MTL to a single objective that renders optimization invariant to the magnitudes of losses. Theoretically, we prove that the requirement for scale invariance uniquely determines this transformation to be logarithmic. We further show that this general transformation-induced merit function preserves weak Pareto optimality and admits a smooth surrogate with controllable approximation error. Extensive experiments on representative multi-task benchmarks demonstrate that SIMS consistently outperforms existing scalarization methods and achieves state-of-the-art performance.
[AI-41] Reproducing and Evaluating the Generalizability of Subliminal Learning in Open-Weight Models EMNLP’26
链接: https://arxiv.org/abs/2609.12586
作者: Daan van der Weijden,Nathan Brack,Selene Baez Santamaria
类目: Artificial Intelligence (cs.AI)
备注: Accepted at BlackBoxNLP@EMNLP’26 (The 9th BlackboxNLP Workshop Special Track: Reproducibility and Reliability in Interpretability Analyses)
Abstract:In this reproduction paper we investigate subliminal learning, a consequence of distillation where teacher models transmit behavioral preference traits through semantically unrelated data. The original paper explores two types of traits (animal preferences and misalignment), three data modalities (number sequences, code, and chain of thought), and several model families. We reproduce their experiments and extend the setup along three axes: new preference categories (actors and politicians), a new task (chess move generation), and an additional open-weight model (Ministral8B). We also run a controlled ablation on the numbers task’s answer-space size (1-, 2-, and 3-digit sequences). We focus on open-weight models with accessible checkpoints on HuggingFace, since the original paper’s GPT-4.x fine-tuning is no longer available. Our reproduction supports the original paper’s claims, but our extensions show they are not universal as transmission strength varies across traits and tasks, and one model shows almost no effect at all.
[AI-42] Clustering-Based Balanced Sampling and Allocation with Data Parallelism for High-Performance Fine-Tuning
链接: https://arxiv.org/abs/2609.12584
作者: Hyunjin Kim,Youngeun Nam,Jaemin Han,Wonhyeok Choi,Jae-Gil Lee
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Instruction-tuning datasets for large language models (LLMs) are often large, redundant, and imbalanced, limiting efficient adaptation. Naive large-batch fine-tuning repeatedly includes overrepresented sample groups while weakly covering underrepresented but informative ones, especially under data parallelism (DP) across multiple GPUs. We propose CluSTER, a Cluster-aware balanced Sampling framework for Training Efficient data Reduction in DP instruction tuning. CluSTER curates a representative reduced dataset through gradient-space clustering and DP-aware balanced allocation, ensuring dual-level coverage across clusters and workers, while preserving the original data distribution by weighted update. As a result, CluSTER reduces redundant computation and improves training stability without compromising model quality. Across multiple instruction-tuning datasets, CluSTER reduces training time by up to 69.6% with almost no accuracy loss compared to prior sampling and data reduction methods. Code is available at this https URL.
[AI-43] SCOPE-OPSD: Fisher-Conditioned Privileged Subspaces for On-Policy Self-Distillation
链接: https://arxiv.org/abs/2609.12579
作者: Yunmeng Chen(1),Kunyu Wang(2),Peihan Li(1),Yi Wang(1),Shuyin Xia(3),Yi Liu(1),Xinyong Cheng(2),Dehui Wang(2),Xiangyong Zhai(2),Yanxing Liu(1),Song Liu(1) ((1) Chongqing Ant Consumer Finance Co., Ltd., (2) Alibaba Cloud Computing Co., Ltd., (3) Chongqing University of Posts and Telecommunications)
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 16 pages, 3 figures. Yunmeng Chen, Kunyu Wang, and Peihan Li contributed equally. Corresponding author: Song Liu
Abstract:On-policy self-distillation (OPSD) scores student-generated prefixes with a solution-conditioned self-teacher, yet transfers supervision only through next-token probabilities. We ask whether the aligned final-layer discrepancy offers a useful second channel, and how to test that channel without confusing its geometry with auxiliary strength. SCOPE-OPSD projects the privileged teacher-student residual onto a frozen rank-64 factor estimated from residual covariance and language-model-head Fisher sensitivity. It reuses the forwards already required by OPSD and adds neither rollouts nor inference-time modules. A matched Random control preserves the structured factor’s rank and nonzero spectrum and uses per-arm gradient-RMS calibration, isolating the effect of the data-dependent orientation. Across the complete 25/50/75/100-step trajectories for Qwen3-1.7B, 4B, and 8B, Structured is never below Pure OPSD, with strict gains in 11 of the 12 model-checkpoint combinations and an exact tie at 4B step 25. Structured also exceeds matched Random in 10 of the 12 combinations. At step 75 on Qwen3-1.7B, Structured exceeds matched Random by 1.39 Macro Avg@12 points in each of two independent training reruns. A cross-fitted diagnostic also shows 4.40 times greater held-out privileged-gap capture than the matched random orientation. The results support a compact, Fisher-conditioned privileged subspace for short-budget OPSD.
[AI-44] From Collaboration to Capability: Internalizing Routed LLM Experts into Compact Reason ers
链接: https://arxiv.org/abs/2609.12578
作者: Frank Nie,Shuyao Wang,Ethan B. Liu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:A compact controller can coordinate stronger experts by selecting whom to consult, formulating requests, and integrating their responses. We study whether learning from both the controller’s decisions and the experts’ reasoning and code improves its generation after expert removal. We introduce \textscRivet for \emphcollaboration internalization: expert-augmented reinforcement learning applies a shared outcome signal to controller decisions and returned expert spans, and verified trajectory internalization consolidates complete successful interactions through format-aware supervised training. The deployed controller generates reasoning, code, and interaction structure with local Python execution and no external LLM. Across seven competition-mathematics benchmarks, RIVET-1.7B and RIVET-4B achieve average accuracies of 28.25% and 44.16% ; Stage~II improves RIVET-4B’s accuracy after expert removal by 6.49 points, and GPQA-Diamond results provide evidence of generalization to scientific reasoning. Ablations show gains from ordinary trajectory supervision and additional format weighting, supporting the effectiveness of training on the content and structure of verified collaborations.
[AI-45] okenMapper: A Step Toward Interoperable Speech Token Translation AACL
链接: https://arxiv.org/abs/2609.12563
作者: Tal Kozakov,Tal Rosenwein,Eliya Nachmani
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Sound (cs.SD); Audio and Speech Processing (eess.AS)
备注: Accepted to AACL-IJCNLP 2026 findings
Abstract:Neural audio codecs discretize speech into token sequences, but the resulting token spaces differ in vocabulary and codebook structure, preventing direct communication across models. This limitation affects applications such as conversational voice agents and speech to speech translation systems where multiple speech models must interact. As a result, transferring information between speech systems typically requires decoding to waveform audio and re-encoding with a second tokenizer, increasing latency and introducing potential information loss. To address these limitations, we present TokenMapper, a direction aware framework for direct token to token translation between heterogeneous speech tokenizers in the discrete domain. TokenMapper supports structurally mismatched token spaces, including mappings between single codebook and multi codebook representations, under a shared effective token rate. Experiments on GLM-4-Voice, MiMi and DualCodec show consistent cross model performance. Specifically, translation WER approaches native reconstructions within 2.5-6.8% absolute WER, human MOS for TokenMapper outputs ranges from 2.29 to 4.39, following the same direction level trends as UTMOS and end to end latency is reduced by 4.8-94.5% relative to waveform bridging, reaching up to 972 ms per utterance. These results provide a practical step toward cross model speech token interoperability without intermediate waveform reconstruction.
[AI-46] RoofLang: Enabling AI-Driven Architecting of LLM Inference Systems
链接: https://arxiv.org/abs/2609.12551
作者: Ziyue Yang,Yuting Jiang,Lei Qu,Peng Cheng
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Artificial Intelligence (cs.AI)
备注:
Abstract:AI is beginning to make substantive contributions to LLM inference optimization. Existing AI optimizations are predominantly profiling-based. Profiling-bound feedback confines the search to the capabilities and performance of an existing software stack, preventing a fundamentally better architecture of LLM inference systems from being identified. To enable the AI-driven LLM inference system architecting loop, we argue that a general workload representation, a verifiable mutation space, and an implementation-independent evaluator are required. We present the RoofLang domain-specific language (DSL) that provides these features. In our evaluation, RoofLang reveals that DeepSeek V4-series models could achieve 3.5-39.5 \times higher peak decode throughput than other representative models. This gap is disproportionate to their total parameter counts and arises largely from compact KV-cache designs that support larger batches and reduce memory traffic. A persistent optimizer agent further discovered several new architectures that improved both throughput and interactivity of DeepSeek V4 Pro on NVIDIA B300 by 6.23-50.1%.
[AI-47] Computing at Sea: Floating and Offshore Data Centres as a Pathway to Sustainable AI Infrastructure
链接: https://arxiv.org/abs/2609.12511
作者: Cheng Siong Chin,Jianhua Zhang,M. Venkateshkumar
类目: ystems and Control (eess.SY); Artificial Intelligence (cs.AI)
备注: 7 pages, 2 figures, 2 tables
Abstract:The rapid expansion of artificial intelligence is transforming data centres into one of the world’s fastest-growing sources of electricity demand. As AI systems scale in size and capability, the physical infrastructure supporting computation is approaching critical limits in energy availability, cooling capacity, land use, freshwater consumption, and carbon management. Conventional land-based data centres are increasingly constrained by urban land competition, grid congestion, environmental pressures, and lengthy permitting processes, raising fundamental questions about where future computing infrastructure can sustainably exist. This article examines floating and offshore data centres as an emerging alternative model for digital infrastructure. By relocating computation to marine environments, offshore systems can exploit the ocean’s natural cooling capacity, reduce freshwater dependence, and enable direct integration with offshore renewable energy resources such as wind, wave, and tidal power. Early deployments have demonstrated the potential for significantly improved energy efficiency and operational reliability compared with conventional facilities, while also opening new possibilities for distributed and resilient computing architectures. The article explores how offshore computing may reshape the future relationship between electrification, renewable energy, and large-scale AI infrastructure. It analyses the opportunities and trade-offs associated with marine deployment, including environmental impacts, engineering design challenges, economic feasibility, and regulatory governance. Rather than treating offshore data centres as experimental novelties, the article presents them as part of a broader systems-level transition in how society may power, cool, and sustain the next generation of computational growth.
[AI-48] Adaptive Agent Design
链接: https://arxiv.org/abs/2609.12486
作者: Raj Kiriti Velicheti,Subhonmesh Bose,Tamer Başar
类目: ystems and Control (eess.SY); Artificial Intelligence (cs.AI); Computer Science and Game Theory (cs.GT); Optimization and Control (math.OC)
备注:
Abstract:We consider an agent acting against a general non-Markovian environment. The agent maintains its agent states, but is free to choose a transition kernel across those states and optimize its state-feedback control policies. We study the bi-level agent design problem that optimizes the transition kernel and the policy it induces, given said kernel with offline data of observations and actions obtained via a behavioral policy. For general environments, we show that a soft Q -learning algorithm converges almost surely to the fixed point of a soft Bellman equation defined by the stationary averages that the behavioral policy and the chosen kernel induce, and we delineate what separates the resulting policy from an optimal one. In partially observed Markov decision problems, we analyze convergence properties of parametrized transition kernel design via zero-th order and Bayesian optimization techniques.
[AI-49] ripPattern: A Pattern-based Text Watermarking Method for Large Language Models AACL
链接: https://arxiv.org/abs/2609.12472
作者: Sangjun Moon,Dasom Choi,Jingun Kwon,Hidetaka Kamigaito,Taro Watanabe,Manabu Okumura
类目: Artificial Intelligence (cs.AI)
备注: Accepted to Findings of AACL-IJCNLP 2026. 16 pages, 4 figures, 10 tables
Abstract:Text watermarking techniques have gained significant attention for identifying machine-generated text and mitigating risks from large language models (LLMs). Existing methods typically divide an LLM’s vocabulary into green and red tokens, but encouraging generation toward green tokens can reduce text quality and naturalness. To address this, we propose TripPattern, a watermarking framework that formulates text watermarking as a pattern-based matching task using three vocabulary partitions. TripPattern divides the vocabulary into one neutral group and two pattern groups. During generation, the model alternates token selection between the two pattern groups to embed detectable patterns, while neutral tokens are selected independently to improve flexibility and preserve naturalness. For detection, TripPattern uses pattern-based statistical tests that provide interpretable p-values by measuring how often adjacent tokens alternate between the pattern groups. Theoretical analysis and empirical evaluations on four multilingual datasets show that TripPattern maintains LLM generation quality while achieving robust watermark detectability.
[AI-50] Beyond Vector Similarity: Hierarchical Context-Aware Graph RAG vs Standard RAG in Enterprise Code Migration
链接: https://arxiv.org/abs/2609.12464
作者: Nilesh Jaiswal,Aniket Agrawal,Arjit Shukla,Divya Malhotra,Saurabh Garg,Suchit Puri,Suddhasatwa Bhaumik
类目: Artificial Intelligence (cs.AI)
备注: 11 pages, 2 images, 3 tables
Abstract:As enterprises modernize legacy monolithic systems to microservices, Large Language Models (LLMs) are heavily utilized for automated code translation. However, traditional vector-based Retrieval-Augmented Generation (Standard RAG) struggles to capture topological relationships. It fetches isolated chunks that sever inheritance chains, leading to high compilation failure rates. This paper introduces a Hierarchical Context-Resident Graph (HCRG) methodology to resolve these limitations. Our pipeline uses tree-sitter for Abstract Syntax Tree (AST) extraction, maps architectural edges into a Google Cloud Spanner Property Graph, and serializes this structure into a Gemini Context Cache for topological, parent-first code translation. We shift evaluation from naive text-overlap to a custom 7-metric Software Engineering framework. Traditional metrics like CodeBLEU (which scored 91% for both methods) effectively masked Standard RAG’s structural failures behind syntactically plausible but broken code. Empirically, Graph RAG decisively mitigates dependency loss: API hallucination rates dropped from 56.4% to 16.2%, Dependency Resolution Quality improved from 34.8% to 65.9%, and Parent-Child Consistency rose from 26.7% to 45.5%. However, Graph RAG introduces specific trade-offs. The dense global context causes defensive over-engineering by the LLM, reducing Cyclomatic Complexity Consistency from 71.6% to 46.7%, and slightly degrades Docstring Preservation (67.0% to 61.0%). Ultimately, while trading code complexity for reduced hallucinations, Graph RAG provides a substantially more viable, architecturally sound path for automated enterprise codebase modernization.
[AI-51] EvoRS: On-Policy Self-Evolution of Reward Systems for Open-Ended Reinforcement Learning
链接: https://arxiv.org/abs/2609.12459
作者: Weiyuan Li,Aili Chen,Xintao Wang,Yikai Zhang,Qingqing Dong,Jinghan Xu,Hongru Hou,Wenxuan Zhao,Chengkun Lang,Jun Gao,Yuanli Guo,Hongcheng Guo,Yanghua Xiao,Deqing Yang
类目: Artificial Intelligence (cs.AI)
备注: 38 pages, 10 figures, 24 tables
Abstract:Open-ended reinforcement learning often relies on rubric-based rewards for tasks without directly verifiable answers. Yet the policy and reward system form a dynamic feedback loop: as the policy optimizes the current reward, an initially useful reward system may become unreliable due to reward hacking or reduced response discriminability. The reward system should therefore evolve rather than remain fixed during training. Existing dynamic-rubric methods adapt evaluation criteria, but reward failures can also arise from scoring mechanisms or signal composition. We introduce EvoRS, a self-evolving RL framework that evolves the reward system from on-policy experience, representing it as an executable Reward-DAG. Specifically, an agentic designer updates this system from on-policy rollouts and reward traces to maintain train-time reliability. Across writing and roleplay, EvoRS achieves the best quality under all three judges, outperforming the policy by (2.107) and (4.767) points, respectively, while reducing reward hacking and coverage failures and preserving reward informativeness. Ablations confirm that a comprehensive fixed reward system cannot remain reliable in open-ended tasks and must evolve throughout training.
[AI-52] 3D Digital Twin Visualization of Multiclass GRF-Based Gait Disorder Classification
链接: https://arxiv.org/abs/2609.12442
作者: Nayoung Son,Minwoo Shin
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Automated gait analysis requires accurate classification and interpretable outputs. We propose an integrated framework for classifying healthy gait and multiple musculoskeletal impairment groups using bilateral ground reaction force (GRF) and center-of-pressure (COP) signals. The signals were normalized over the stance phase and standardized using training-set statistics. The model achieved a validation accuracy of 99.00% and a test accuracy of 90.07% under a session-level split. Class-specific \epsilon -LRP identified positive and negative contributions across both sides, multiple signal components, and different stance phases. Separately, the processed GRF signals and model predictions were synchronized within a Blender-based 3D visualization, enabling sample-level inspection of gait trials and classification results. The proposed framework integrates classification, explainability, and 3D visualization to improve model transparency. The source code is available in the following repository: this https URL
[AI-53] Debiasing as a Measurement Intervention: Calibrated Ties and Resolution Loss in LLM -as-a-Judge Evaluation EMNLP2026
链接: https://arxiv.org/abs/2609.12439
作者: Liang Zhao,Yong Wang,Jiangzhe Chen
类目: Digital Libraries (cs.DL); Artificial Intelligence (cs.AI)
备注: Accepted at Emnlp 2026
Abstract:LLM-as-a-judge protocols are commonly debiased by instructing judges to ignore presentation cues such as citation formatting, source labels, and evidence-display style. We show that this intervention can suppress bias while damaging the resolution of the measurement instrument. We introduce TraceJudgeBench, a diagnostic benchmark for auditing citation-like artifacts in RAG and agent-workflow evaluation, covering content-equivalent pairs, citation ablations, correctness conflicts, human-validated soft and moderate quality gaps, prompt-strength ladders, decoupled judging, and a controlled workflow-ranking probe. Across GPT-5.5, Claude Sonnet 4.6, and DeepSeek V4-Flash, stronger anti-citation prompts reduce worse-cited wins from up to 50.5% to 0%; yet some operating points already convert validated moderate-gap decisions into Tie before the strict stress-test endpoint, while correctness-conflict accuracy remains at or above 93.0%. A second, 50-pair FinQA moderate-gap construction reproduces the qualitative frontier, and open-weight Qwen2.5-14B-Instruct-AWQ and Gemma-3-12B-IT runs reproduce the central HotpotQA frontier. TRACE-style decoupling recovers 96.5-100.0% better-plain resolution across the reported settings. Human validation separates three meanings of Tie: correct equivalence Tie, calibrated soft-boundary Tie, and resolution-destroying Tie on validated quality gaps. We frame debiasing as a measurement intervention whose bias suppression, resolution retention, Tie cost, and protocol cost must be reported jointly. The supplementary artifact contains benchmark splits, prompts, raw judge outputs, validation summaries, and analysis.
[AI-54] Beyond the Query: Do Retrieval Signals Improve Adaptive Multimodal RAG Routing?
链接: https://arxiv.org/abs/2609.12437
作者: Qiaomu Li,Qiuyuan Zhang,Nong Ming
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Adaptive RAG often uses retrieval-time signals to decide whether another retrieval, reranking, or multimodal step should run. We ask whether these signals add routing value once the query itself is already known. Across document, audio, and video RAG, we compare matched query-only and query+retrieval routers while holding the optional actions, router family, training procedure, and evaluation fixed. On the held-out final evaluation, adding the tested retrieval signals does not produce a reliable routing improvement over the query-only baseline. Some retrieval signals are associated with whether a later step will help, but that predictability does not consistently lead to bet- ter RUN/SKIP decisions. The main lesson is therefore methodological: retrieval-state features should not be credited with routing value unless they improve over a matched query-only control. Our results do not show that routing or retrieval state is generally useless; they show that the incremental value of retrieval signals must be demonstrated rather than assumed.
[AI-55] LifeFuse-Mem: Lifecycle-Aware State Fusion Against Temporary Overwriting for Long-Term Memory
链接: https://arxiv.org/abs/2609.12436
作者: Hanyu Zhao,Yuqian Feng,Zhenyu Song,Yuanchao Cheng,Yance Jiao,Tengfei Pan,Li Du
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Long-running LLM agents require memory mechanisms that maintain coherent internal states across interactions. We study a lifecycle-labeled memory setting in which write episodes provide lifecycle metadata during training, and phase-aware readout is used during evaluation. This setting reflects the need to distinguish information that should remain influential across future interactions from information that should affect only the current context. A mismatch between these lifecycles can cause temporary information to overwrite durable knowledge, leading to behavioral drift in persistent agents. Within this setting, we introduce \textbfLifeFuse-Mem, a lifecycle-aware neural memory framework that separates information according to its temporal commitment. LifeFuse-Mem uses dedicated memory components and lifecycle-aware updates to allow stable and transient knowledge to evolve locally without converting temporary context into durable state. On the controlled anti-overwrite benchmark, LifeFuse-Mem improves acquisition-controlled retention and reduces temporary overwrite; on two public long-memory benchmarks, it remains broadly competitive. These results suggest that explicit lifecycle signals can help diagnose and mitigate overwrite in compact online memory.
[AI-56] Observation-Anchored Selective Assimilation for Longitudinal Tumor-State Proxy Forecasting in Post-Treatment Glioma
链接: https://arxiv.org/abs/2609.12435
作者: Yeonjae Jung,Minwoo Shin
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Post-treatment MRI in patients with glioma provides serial observations for updating patient-specific tumor-state proxy estimates, but variable appearances and trajectories complicate forecasting. We formulate forecasting as an observation-aware digital-twin update in which an intermediate observation anchors the patient-specific state. Among 203 patients and 594 follow-up time points, a predefined no-new-treatment criterion retained 120 of 236 candidate triplets, split into 81/24/15 training/validation/test triplets at the patient level. Each time point was represented by a continuous voxel-wise tumor-state proxy map in [0,1] derived from MRI lesion labels. A SegMamba-based single-step forecaster predicted update proposals from multimodal source-state tensors. Observation-Anchored Selective Assimilation (OASA) retained the observed intermediate proxy as the state anchor and selectively applied updates through a validation-selected tiered case-level rule and voxel-wise soft gate. We compared initial-scan forecasting, rollout without assimilation, latest-observation persistence, direct prediction, OASA, OASA + calibration, and morphological dilation. Checkpoints, OASA rules, and calibration thresholds were selected using validation data only. Across three seeds on 15 held-out test triplets, OASA maintained Dice at \tau = 0.2 comparable to persistence (0.6071 \pm 0.0025 vs. 0.6070) while yielding numerically higher Dice at \tau = 0.5 (0.4269 \pm 0.0079 vs. 0.3981), with a small RMSE increase. Calibration increased Dice at \tau = 0.2 to 0.6178 \pm 0.0025, increased false-positive (FP) support (11,836 \rightarrow 18,663), and reduced false-negative (FN) support (22,107 \rightarrow 17,536). This reflects near-threshold support calibration rather than improved biological predictive capability. Code is publicly available at this https URL.
[AI-57] MInTRL: Off-policy Intervention can boost On-policy RL
链接: https://arxiv.org/abs/2609.12419
作者: Mingyu Chen,Yefan Tao,Gerald Friedland,Xuezhou Zhang,Chris Kong
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Reinforcement learning with verifiable rewards is typically performed on-policy, keeping training data close to the current policy but limiting learning to trajectories that the policy can discover itself. Off-policy methods such as supervised fine-tuning, on the other hand, can leverage external knowledge beyond the base model’s capabilities, but may suffer from large distribution shift. The key challenge is thus to expand exploration without sacrificing learnability. In this work, we introduce Minimal Intervention Reinforcement Learning (MInTRL), which expands the exploration frontier through sparse, local interventions in otherwise on-policy rollouts. During generation, a judge-intervention policy periodically reviews the current policy’s output, replaces erroneous suffixes with short corrections, and immediately returns control to the policy. During training, MInTRL adopts a sequence-level advantage-regression objective that eliminates the need for importance sampling. We show that sparse, local interventions can substantially improve coverage beyond finite-budget on-policy sampling while preserving the overall on-policy nature of the resulting trajectories. Across math and code benchmarks, MInTRL consistently outperforms standard on-policy and off-policy baselines. Ablations show that MInTRL remains effective with self-intervention and across different judge policies, while performance peaks at moderate intervention intensity, highlighting the importance of intervening minimally. These results establish minimal intervention as an effective paradigm for enhancing on-policy RL.
[AI-58] SoK: Rethinking Jailbreaking in the Era of Agent ic AI: Attacks Defenses and Practical Consideration
链接: https://arxiv.org/abs/2609.12413
作者: Md Jueal Mia,Yanzhao Wu,Selcuk Uluagac,M. Hadi Amini
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models (LLMs) are rapidly evolving from conversational assistants into agentic AI systems that reason, plan, invoke tools, maintain persistent memory, communicate with other agents, and execute multi-step tasks. At the same time, modern models exhibit substantially stronger native safety alignment than earlier generations on which many jailbreak attacks and defenses were originally studied. This shift raises a fundamental question: \textitwhich established jailbreak-security findings remain valid in the era of modern LLMs and agentic AI? We address this question through a Systematization of Knowledge (SoK) that reframes jailbreak security around the full agentic execution pipeline. We develop unified taxonomies of attacks and defenses spanning user interaction, planning and reasoning, memory, tool use, and inter-agent communication, and introduce a security–utility–efficiency evaluation framework that separates native harmful-prompt safety, adversarial jailbreak robustness, and agent-level security outcomes. We further conduct a controlled empirical study of representative attacks and defenses within a common agentic framework. Our results reveal three important gaps. First, strong native alignment does not imply robustness to adversarial jailbreaks. Second, defense effectiveness is highly model-, attack-, and component-dependent and can come at substantial cost in over-refusal, utility, and latency. Third, low final-response attack success can mask severe intermediate compromise: planning, memory, and tool interactions may remain unsafe even when the final response is successfully filtered. These findings motivate a shift from response-centric jailbreak defense toward cross-layer, execution-aware security that protects agent state, component transitions, and external actions while preserving practical utility and efficiency.
[AI-59] VRL-Bench: Benchmarking agents on computer control tasks under finite trial budgets
链接: https://arxiv.org/abs/2609.12404
作者: Yu Bai,Yukai Miao,Dawei Wang,Li Chen,Yanyu Ren,Yuqian Shi,Dan Li,Ying Xiong,Chengqiu Tan,Run Zhou,Li Li
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Learning from trial and error is a promising way to improve language agents on complex tasks such as computer control. Reflexion introduced verbal reinforcement learning, which turns failed trials into text that guides later attempts without updating model parameters. We introduce VRL-Bench, a harness for fair evaluation of trial-and-error learning under finite trial budgets. Across three models on MiniWoB and WebShop, we evaluate updates from several prominent verbal-memory methods spanning Reflexion and later work: each improves observed success over memory-free retry in some settings but reduces it in others. Replay experiments show that using reflection can reduce success rates, revealing a trade-off between exploiting experience and continued exploration. We propose VEX ^2 , a verbal exploration–exploitation scheduler that uses a language model to jointly select policies and allocate the remaining trial budget. VEX ^2 is the only evaluated update to achieve positive observed success-rate gains over retry in all six settings.
[AI-60] Decentralized Evolution of Hexapod Gaits with Independent Leg Controllers
链接: https://arxiv.org/abs/2609.12400
作者: Gary B. Parker,John Asaro,Jim O’Connor
类目: Artificial Intelligence (cs.AI); Robotics (cs.RO)
备注:
Abstract:This paper presents a novel approach to hexapod locomotion by evolving each leg’s gait independently through a decentralized evolutionary algorithm. Using the Webots simulator and the Mantis hexapod robot, we optimize individual leg controllers without centralized coordination, allowing emergent behaviors to drive the development of efficient, coordinated locomotion. Our decentralized method is benchmarked against cooperative coevolution, demonstrating improved efficacy in generating stable and adaptive gaits while showing interesting emergent coordination. By enabling independent evolution of leg controllers, this method reduces the complexity of gait optimization and highlights the potential of decentralized strategies for scalable and adaptive robotic systems.
[AI-61] Niching Agents in The Core
链接: https://arxiv.org/abs/2609.12398
作者: Gary B. Parker,Jim O’Connor,John Asaro
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:The Core is a unique competitive co-evolution algorithm that allows agents to evolve autonomous control without utilizing a traditional fitness function. The agents evolve via local interactions through tournament selection, crossover, and mutation, producing offspring by evolving better controllers. Previous works have shown The Core’s ability to evolve agents capable of combat and navigation in the Xpilot video game. This research expands upon that premise by niching agents to specific subsets of the original environment The Core was tested in. Our results demonstrate the niched agents capacity for success over agents niched to the entire system and agents niched to different sub-environments.
[AI-62] Is Gaussian Splatting Becoming Neural Again? A Taxonomy and Controlled Study of Learned Parameterization
链接: https://arxiv.org/abs/2609.12395
作者: YuanHang Wang,Xin Cao,Yi Zhang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Three-dimensional Gaussian Splatting (3DGS) combines explicit primitives with efficient rasterization, yet recent systems increasingly use neural networks to generate or share Gaussian parameters. We characterize this trend along five axes: attribute decoding, spatial sharing, view-conditioned decoding, topology generation, and amortized inference. An analysis of 19 representative methods shows that these choices address different limitations and cannot be reduced to a binary neural label. We also isolate three forms of neural parameterization in a controlled mip-NeRF 360 study. Sharing appearance and opacity improves reconstruction quality, while decoding geometric structure offers no further gain. The evidence favors selective neuralization: shared functions help when they capture reusable correlations without sacrificing the local geometric freedom of explicit splats.
[AI-63] BlueLM-GUI Technical Report: A Real-Device-Centric Flywheel for Self-Improving Mobile GUI Agents
链接: https://arxiv.org/abs/2609.12394
作者: Tong Ye,Kunyang Han,Guozhi Wang,Longqiang Luo,Zhifeng Ding,Yongxiang Zhang,Xiaolei Shen,Yuxuan Zhang,Zhuping Zhang,Tao Xu,Yue Pan,Yucheng Zhao,Yupei Hu,Yuanjiang Ouyang,Danfeng Shen,Runqi Lin,Hongda Cai,Zhaoxiong Wang,Mengjia Yan,Yingjie Zhong,Chen Zhou,Zeyu Zhang,Xuwen Zhu,Penggang Shi,Mingcheng Luo,Ziyang Wu,Min Jin,Mingfu Shen,Zairong Xu,Fan Zhang,Hao Wang,Liang Liu,Zhulin Xie,Lijun Yao,Xiao Liang,Liangmin Wen,Liqiang Feng,Feilong Wu,Min Hu,Min Chen,Guanjing Xiong,Xiaohu Ruan,Xiaoxin Chen
类目: Artificial Intelligence (cs.AI)
备注: 49 pages
Abstract:Mobile GUI agents are shifting from multi-module frameworks to native models trained end-to-end, yet industrial deployment faces three persistent gaps. Sandbox training produces a distribution mismatch with production environments; expensive real-device failures remain underutilized; and fixed benchmarks saturate, losing the power to guide iteration. We present BlueLM-GUI, a 35B-A3B mobile GUI agent built as a real-device-centric flywheel that closes these gaps through three principles. Every Sample Matters: a dual-track pipeline with Heterogeneous Triple-System Consensus evaluation and an Error Correction \ Derivation Module salvages every trajectory into usable supervision. Every Rollout Is Real: a three-stage recipe—continual pre-training, supervised fine-tuning, and agentic reinforcement learning on hundreds of real phones—grounds every rollout in real production environments, so the capability the model learns transfers directly to deployment. Every Query Evolves: a quota-driven benchmark methodology with three orthogonal axes enables precise attribution and allows the benchmark to be systematically upgraded as the model improves. BlueLM-GUI achieves 87.4 on MobileGUI-VBench, surpassing the best closed-source model by 5.1 points, and 84.9 on AndroidWorld, the best result among open-source models and competitive with closed-source models. These results demonstrate that grounding model training and iterative improvement in both real devices and the three Every principles yields strong, robust, and transferable mobile GUI capability.
[AI-64] oward Robust Personalized Alignment for LLM s: Mitigating Persona Drift in Multi-Turn Dialogue EMNLP2026
链接: https://arxiv.org/abs/2609.12373
作者: Youyuan Zhang,Siyuan Li,Fangming Liu,Jing Li
类目: Artificial Intelligence (cs.AI)
备注: Accepted to Findings of EMNLP 2026
Abstract:Persona drift remains a central challenge for personalized language models, as user profiles evolve over long interactions rather than remain permanently fixed. Models must therefore revise persistent persona states when preferences genuinely change, while avoiding updates driven by transient, ambiguous, or unresolved observations. We propose CORE, which separates turn-local evidence from persistent persona-state revision and selectively updates grounded user preferences through uncertainty-aware belief revision. We also introduce PERSIST, a held-out post-anchor benchmark for persona-state robustness under sequential interaction stress, covering ambiguity, conflict, and controlled social influence. Across ALOE, PersonaChat, and PERSIST, CORE improves personalized alignment and robustness, with complementary gains in normalized closed-slot state fidelity. Human evaluation and mechanistic controls further support explicit update control beyond stronger generation or persistent memory alone.
[AI-65] LoRA-RC: Reservoir Computing with Low-Rank Adaptation
链接: https://arxiv.org/abs/2609.12327
作者: Wenbin Wan
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Systems and Control (eess.SY); Dynamical Systems (math.DS)
备注: accepted at ieee lcss
Abstract:Reservoir computing (RC) trains only a linear readout over a fixed recurrent layer, making it fast and data-efficient for online prediction. However, a static reservoir degrades under system drift, readout-only adaptation is then insufficient, and unconstrained reservoir adaptation can destroy the echo-state and incremental stability properties that make RC reliable. This paper proposes LoRA-RC, which adapts the recurrent matrix through a low-rank correction driven by streaming prediction errors. The base reservoir and adaptation bases are fixed offline; a small core matrix is adapted online, projected onto a spectral-norm ball, and low-pass filtered at each step. The projection guarantees that every applied recurrent matrix remains within a certified contraction set, and an incremental input-to-state stability bound is established for the reservoir along each online adaptation path, with path-independent rate and gain. On a Lorenz system with an abrupt parameter drift, LoRA-RC cuts post-drift prediction error by 56% versus a fixed RC and 51% versus readout-only adaptation; ablations over 20 seeds show that removing the projection inflates this error by more than a factor of 40.
[AI-66] Affective Agent : On-Device Personalized Intervention Reasoning for Wearable Systems
链接: https://arxiv.org/abs/2609.12322
作者: Reina Mun,Zishen Wan,Vijay Janapa Reddi
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: Accepted for publication in IEEE Internet Computing, Special Issue on Wearable Computing (Sep/Oct 2026). 4 figures, 2 tables
Abstract:Affective computing has advanced wearable state inference, but on-device reasoning about whether, when, and how to intervene remains challenging. We present Affective Agent, a three-layer reference architecture for personalized intervention reasoning under uncertainty on wearable-class hardware. It combines a compact sub-billion-parameter language model with physiological evidence, context, and user history to decide whether, when, and how to intervene, without cloud dependency or per-user retraining. The architecture is organized into three interacting layers (perception, personalization, and reasoning), adapting to individual users through host-managed structured memory evolution rather than per-user weight updates. We instantiate Affective Agent in indoor environmental quality control and evaluate it on held-out, simulator-generated longitudinal scenarios spanning physiological variation, context, signal quality, and intervention history. Results show that memory-driven personalization and two-pass structured reasoning improve intervention decisions within this synthetic evaluation. By moving the decision layer on-device, this work demonstrates a path from wearable state inference toward closed-loop, personalized intervention on wearable-class hardware.
[AI-67] AIM: A Privacy-Aware Interoperable Memory Framework for Multi-Agent Multi-User LLM Systems
链接: https://arxiv.org/abs/2609.12320
作者: Zachary Johnson,Nigel Boachie Kumankumah,Somya Chatterjee,Tejas Sathyamurthi,Min Chen,Xinyi Alice Li,Xiao Wang,Emily Morgan Gelchie,Jessica Lin,Sadid A. Hasan,Sulaiman Vesal
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Traditional large language models (LLMs) are scoped to individual user sessions, limiting their knowledge to a single conversation and preventing them from learning user preferences that evolve over time. Existing agentic memory systems address this limitation but generally operate at the individual-user level, restricting the public knowledge that could be shared across users to improve downstream responses. We introduce AIM (Agentic Interoperable Memory), a unified, privacy-aware memory framework that enables multi-agent, multi-user LLM systems to persistently manage private and shared memory. AIM dynamically classifies information as private, scoped to one user and inaccessible to others, or public, accessible to all users. It enforces index-level access controls so that private memories are retrievable only by their owner, protecting sensitive data while allowing beneficial shared knowledge to improve coordination and consistency. We also introduce MUMBench (Multi-User Memory Benchmark), a dataset of multi-user interactions containing private and shareable information across four domains. To our knowledge, MUMBench is the first public dataset designed to evaluate multiple memory operations, including retrieval, creation, update, and deletion, in a multi-user environment. Across three independent runs on MUMBench, AIM achieves 96.0% visibility classification accuracy, 58.8% strict operation accuracy, and 70.5% state-aware operation accuracy.
[AI-68] Do Influence-Derived Data Perturbations Enable Machine Unlearning? A Controlled Study of Three Plausible Roles
链接: https://arxiv.org/abs/2609.12313
作者: Chenkai Wu,Chrispine Kambimbi,Qinyang Zeng,Jun Yan
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:We evaluate Deep Perturbation Learning (DPL), which perturbs training images and labels along influence-derived directions, in three roles in which prior work has positioned it for machine unlearning: a direct deletion signal (the strongest claim), a utility-preserving regularizer, and a warm start for adversarial unlearning. Evidence for the weaker roles has been used to support the stronger one, so we test each role separately under a matched protocol with exact-seed retraining baselines. An audit of the public implementation identifies two correctness issues: image directions are computed on augmented, normalized tensors but applied to raw images, and the label perturbation falls below float32 resolution, leaving labels unchanged. After correcting the image-perturbation pipeline, DPL fails the direct-deletion criterion on CIFAR-10/ResNet-18 in all three paired seeds. Its utility effects are inconsistent in sign across seeds, and once direction-computation time is counted it underperforms simple warm-start baselines. A one-seed Tiny ImageNet check likewise does not favor DPL as a regularizer or warm start; preprocessing inconsistencies in the released code make the direct comparison there inconclusive. These results cover random instance deletion only and do not rule out influence-based methods in other deletion regimes. We release a role-matched evaluation protocol and an audit checklist for perturbation-based deletion claims.
[AI-69] Self-Verifying Anomaly Detection using Explainable AI for Cybersecurity of DER Networks
链接: https://arxiv.org/abs/2609.12305
作者: Damilola Popoola,Souradeep Bhattacharya,Manimaran Govindarasu
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 5 pages, 5 Figures, Submitted and Accepted at IEEE PES General Meeting 2026
Abstract:The rapid growth of Distributed Energy Resources (DERs) has significantly expanded the cyber attack surface of modern power grids. Furthermore, increasing sophistication in attack techniques demands anomaly detection systems (ADS) that are accurate, interpretable, and reliable to support DER cybersecurity. While ML-based ADS provide strong detection capabilities, their black-box nature reduces operator trust and limits Security Operation Center’s (SOC) ability to effectively interpret alerts and respond, highlighting the need for explainable Artificial Intelligence (XAI) to ensure transparency and operational confidence. This paper presents an XAI-based anomaly detection framework tailored for DER networks (ExCYDER). The proposed framework uses a self-verifying mechanism that validates ADS alerts to ensure trustworthy decision-making. ExCYDER combines LightGBM with SHAP to check whether each model decision aligns with its feature-attribution evidence, allowing the system to confirm that its internal reasoning is consistent and reliable. Experiments on a realistic DNP3 dataset achieved over 98% detection accuracy, an average rule–SHAP consistency of 44.6%, a SHAP latency of 14.5 ms per alert, and a confidence deviation within 5%, demonstrating stable verification behavior with minimal computational overhead. The framework distinguished between coherent and inconsistent alerts without compromising detection accuracy, demonstrating that integrated verification within XAI-based ADS enhances interpretability, auditability, and operational robustness for DER-focused SOCs.
[AI-70] Hybrid Physics-AI Framework of Body Center of Mass Dynamics from Wrist-Worn Sensors ALT
链接: https://arxiv.org/abs/2609.12304
作者: Shuhao Que,Valentina Breschi,Ying Wang
类目: Artificial Intelligence (cs.AI); Signal Processing (eess.SP)
备注: The abstract of this manuscript has been partially presented at the NaturePitor2026 conference (Redefining Healthcare in the Age of AI: a Nature Conference). A supplementary material file is included in the source file
Abstract:Wrist-worn IMU has been widely used for daily-life health monitoring. Yet, it does not fully represent whole-body dynamics, for which the body center of mass (COM) is considered the physiological reference standard. Therefore, this work proposes a simplified kinematic model (KM), which is designed to map the wrist IMU to the COM acceleration. It is built upon several reductive assumptions that enable the solvability of the dynamic equations based on wrist IMU measurements alone. This work further proposes three types of hybrid AI modeling methods, namely human kinematic model-based neural network (HKM-NN) models, to leverage the power of both grey-box and black-box modeling. The HKM-NN methods include serial learning (ser-) and two approaches of simultaneous learning (sim1- and sim2-). The proposed models are trained and tested using our dataset, which includes wrist IMU measurements and ground-truth COM measurements from 10 healthy volunteers during six gait activities and sit-to-stand (SS) transitional movement. The results demonstrate the feasibility of estimating COM acceleration from wrist IMU measurements. Our KM model yields satisfactory results, with an error ranging from 6.7% to 12.5% for gait activities and 5.6% for the SS. In comparison with the KM model, our HKM-NN models significantly enhance the performance, achieving 5.3% to 9.3% errors for gait activities, and the best error of 3.9% for the SS. In addition, the HKM-NN models demonstrate distinct robustness characteristics under noisy test conditions, with sim1-/sim2- generally maintaining greater robustness under Gaussian perturbations, while the KM model exhibits comparatively strong robustness under salt-and-pepper noise. These findings highlight the importance of combining biomechanical structure with data-driven learning for wearable sensing applications operating under imperfect and noisy measurement conditions.
[AI-71] Robust Prototypical Networks for Few-Shot Sensor Fault Diagnosis
链接: https://arxiv.org/abs/2609.12287
作者: Mohammed Ayalew Belay,Amirshayan Haghipour,Pierluigi Salvo Rossi
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Industrial fault diagnosis often operates with only a handful of labeled fault examples, making few-shot learning attractive for sensor monitoring. Standard prototypical networks are simple and effective; however, their class prototypes may become unstable in the very-low-shot regime because each decision relies on a small support set. We propose \emphMulti-Episode Prototypical Networks (MEPN), which aggregate prototypes from multiple disjoint support episodes and use their mean as the final class representative, reducing prototype variance without changing the encoder architecture. We evaluate MEPN on the DeFACTO sensor dataset using five-way fault classification with synthetic bias, drift, spike, and noise faults injected into real industrial measurements. Over 100 independent runs, MEPN reaches \textbf\SensorOneShotGcpn% in the per-episode one-shot setting ( K!=!1 shot, aggregated over N_\textagg!=!10 support episodes), substantially above single-episode baselines. Under an equal 10-sample support budget, MEPN and ProtoNet at K!=!10 are statistically indistinguishable, confirming prototype accumulation as the mechanism rather than superior fixed-budget learning.
[AI-72] -GADE: Thermodynamical Generative-AI-Driven Evolution of LLM Artifacts
链接: https://arxiv.org/abs/2609.12286
作者: Kyoko Ogawa,Naoki Mori
类目: Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE)
备注:
Abstract:Integrating evolutionary computation and large language models (LLMs) requires control of population diversity as well as generative capability. Among LLM outputs, those with explicit structure, such as a description paired with code, are structured artifacts; we use artifact for short. We propose T-GADE, which evolves these artifacts by extending thermodynamical genetic algorithms through LLM-based genetic operators and artifact-level diversity evaluation. A common free-energy objective supports generational and steady-state updates, with Fermi-type occupancy excluding repeated genotypes and Bose-type occupancy permitting them. We establish exact one-member removal and conditions for recovering the zero-temperature survival rule of Evolution of Heuristics (EoH). On the online bin-packing task studied in the EoH paper, excess measures relative bin-count overhead above a volume lower bound. Training excess uses search instances; transfer excess uses instances with another bin capacity. Generational Bose-type T-GADE at T=0.003 reduced median training excess by approximately 29%, from 1.152% to 0.815%, over 20 runs per configuration (two-sided Mann-Whitney p=0.042 , Cliff’s \delta=0.378 ). Validation selection among its two highest-ranked final candidates reached the same median transfer excess as EoH, 0.496%. These results demonstrate the utility of thermodynamical selection and validation-based use of retained artifacts.
[AI-73] Amortized Low-Rank Adaptation for Model-Based Reinforcement Learning
链接: https://arxiv.org/abs/2609.12278
作者: Fernando Palafox,David Fridovich-Keil
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Robotics (cs.RO)
备注:
Abstract:World models let agents plan by predicting the consequences of their actions, but changes in the environment can make them inaccurate. We study the problem of adapting a world model to an unknown test-time environment, drawn from a known environment family, using only a few episodes of interaction. Existing approaches trade off computational cost against expressivity, i.e., the range of models a method can produce. For example, in-context learning is computationally cheap but limited in expressivity, and gradient-based adaptation is expressive but computationally expensive. We present CLAW (Context-conditioned Low-rank Adaptation of World models), which addresses this tradeoff by using a hypernetwork to generate low-rank (LoRA) adapters at test time. During pretraining, we simulate adaptation to a variety of environments and jointly train the hypernetwork and base world model. At test time, we freeze the base model and use a forward pass of the hypernetwork to generate adapters from a small batch of test-time transitions. We evaluate CLAW in locomotion and manipulation environment families that vary in dynamics, embodiment, and reward. We show that, using only seconds of test-time data, CLAW outperforms gradient-based adaptation and in-context learning during online adaptation. We also show that CLAW avoids overfitting in data-scarce regimes, that its advantage comes from the expressive adapters rather than context conditioning, and that pretraining the hypernetwork jointly with the base model outperforms training it post hoc.
[AI-74] Reinforcement Learning over Patient Trajectories for Clinical Reasoning in EHR Foundation Models
链接: https://arxiv.org/abs/2609.12277
作者: Yuxin Xiao,Sheng Zhang,Chandan Singh,Tristan Naumann,Hoifung Poon,Jianfeng Gao,Xiaodong Liu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
备注:
Abstract:Electronic health record (EHR) foundation models trained on longitudinal patient trajectories have demonstrated strong performance across diverse clinical prediction tasks. However, their clinical reasoning capabilities remain constrained by next-token prediction on limited and incomplete EHR data. To address this, we propose a reinforcement learning (RL) fine-tuning framework that treats EHR foundation models as generative policies over patient trajectories. We formulate common clinical prediction problems (e.g., hospital readmission) as event-conditioned, time-windowed reasoning tasks. We then design time-aware, rollout-sensitive rewards to account for finite rollout lengths and temporally inconclusive outcomes. We find that RL fine-tuning consistently improves over pre-trained backbones and strong baselines. Notably, it enables smaller models to surpass larger pre-trained models in data-limited regimes and induces positive transfer across tasks. Further analysis shows that RL fine-tuned models generate trajectories with stronger structural and semantic alignment to ground truth and greater downstream utility.
[AI-75] Learning Symbolic Constraint Representations from Examples: A Neuro-Symbolic Approach
链接: https://arxiv.org/abs/2609.12267
作者: Nassim Belmecheri,Arnaud Gotlieb,Nadjib Lazaar,Helge Spieker
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Learning user-defined concepts as constraint networks has been extensively studied in the constraint acquisition (CA) literature. However, existing approaches typically rely on intensive interactions with a human oracle, making the learning process costly in terms of time and number of queries. In this paper, we propose a neuro-symbolic framework for automatic CA that significantly reduces user involvement by introducing neural Oracle Transformer models which learn to emulate user responses and to generalize conceptual knowledge. Trained on previously available examples, the learned oracle interacts with a dedicated CA engine, FastCA, which systematically refines the oracle’s responses into a sound, consistent, and interpretable constraint network. This neuro-symbolic interaction enables the recovery of structured symbolic models from data without prior domain knowledge. Our results demonstrate that this neuro-symbolic interplay effectively aligns data-driven pattern recognition with symbolic reasoning, offering a robust approach to automating model construction in combinatorial domains.
[AI-76] GTA: Graph Theory Agent and Benchmark for Algorithmic Graph Reasoning with LLM s
链接: https://arxiv.org/abs/2609.12265
作者: Zixiang Xu,Yanbo Wang,Chenxi Wang,Lang Gao,Zirui Song,Yue Huang,Zhaorun Chen,Xiangliang Zhang,Xiuying Chen
类目: Artificial Intelligence (cs.AI)
备注: 49 pages, 11 figures, including references and appendix
Abstract:Large Language Models (LLMs) are increasingly asked to reason over structured data such as graphs, yet how reliably they can carry out multi-step graph algorithms in language remains unclear. Existing evaluations tend to use simple tasks on small graphs, to score code generation rather than reasoning over the graph itself, or to fix a single input format. We introduce Graph Theory Bench (GT Bench), a benchmark covering 24 classical graph problems in 44 task-structure settings, with over 100,000 examples across four representations: natural language, structured language, adjacency list, and adjacency matrix. Evaluating eight LLMs on GT Bench shows that accuracy is strongly tied to the input representation, that the best representation shifts with graph density, size, and topology as well as with the model, and that this sensitivity persists, attenuated, in the strongest reasoning models. Building on these observations, we propose the Graph Theory Agent (GTA), which pairs a preference-trained representation selector with plan-and-decompose scaffolding around a frozen executor LLM. GTA lifts Phi-4 from 53.5% to 69.1% on the benchmark’s easy split and from 33.0% to 41.5% on its hard split, outperforming eight prompting and agent baselines, and transfers without retraining to GraCoRe and NLGraph. Code for benchmark generation and evaluation: this https URL. The project homepage is available at this https URL.
[AI-77] DriftSE: Speech Enhancement with Generative Drifting
链接: https://arxiv.org/abs/2609.12252
作者: Liang Xu,Diego Caviedes-Nozal,W. Bastiaan Kleijn,Longfei Felix Yan,Rasmus Kongsgaard Olsson
类目: ound (cs.SD); Artificial Intelligence (cs.AI); Audio and Speech Processing (eess.AS)
备注: 11 pages, 3 figures
Abstract:We propose DriftSE, a novel one-step generative framework for speech enhancement formulated as a latent distribution equilibrium problem. During training, the drifting field aligns the generator’s pushforward distribution with the clean speech manifold through drifting in a latent domain. During inference, the drifting process is discarded, enabling one-step generation. We establish that its enhancement quality depends fundamentally on the choice of latent representation. Semantic latents preserve phonetic structure but fail to capture physical acoustic cues, whereas acoustic latents reconstruct the physical signal but risk linguistic hallucination. Therefore, we introduce dual-latent drifting, performing parallel drifting in both semantic and acoustic latents to simultaneously preserve phonetic intelligibility and acoustic fidelity. Additionally, we demonstrate that DriftSE enables fully unpaired training by aligning latent distributions rather than exact point-wise targets. Consequently, DriftSE facilitates cross-dataset learning in the absence of paired noisy-clean samples. Moreover, DriftSE exhibits broad architectural flexibility across different generator backbones. Extensive evaluations on additive denoising and convolutive dereverberation demonstrate robust one-step enhancement across both offline and real-time causal settings. Notably, DriftSE achieves state-of-the-art word error rates across all four evaluated datasets while strictly operating at 1 NFE. Code and audio examples are available online.
[AI-78] Soft Symbol Grounding for Prototypical Concepts
链接: https://arxiv.org/abs/2609.12247
作者: Marcos Galván-López,Nijesh Upreti,Hiram Calvo,Carlos Aguilar-Ibáñez,Vaishak Belle
类目: Artificial Intelligence (cs.AI)
备注: 20th Conference on Neurosymbolic Learning and Reasoning (NeSy 2026)
Abstract:Neuro-symbolic models are usually trained with supervision only on final labels, leaving the intermediate concepts unobserved. Since many concept assignments are consistent with a given label, training can predict labels correctly while recovering the wrong concepts, a failure known as a reasoning shortcut. Prototypical networks reduce shortcuts by anchoring each concept to a few labeled examples, but existing methods still couple perception and reasoning through a hand-crafted, task-specific differentiable loss that must be redesigned for every task. We introduce \textbfSoft-PNet, which removes this loss: it reframes concept grounding as a Metropolis walk over a precomputed cache of feasible symbolic solutions, guided by a prototype distribution built from a single labeled anchor per concept, and trains against one KL objective between the prototype-weighted cache and the network’s concept predictions. The objective is identical across tasks and remains applicable when the solution space cannot be enumerated. On \textttMNIST-EvenOdd, Visual Sudoku, and \textttKand-Logic under scarce supervision, Soft-PNet matches loss-engineered prototypical networks at the concept and label levels and recovers concepts that soft-grounding baselines miss, with no loss engineering and lower training time.
[AI-79] Learning to adapt GR(1) specifications through degradation
链接: https://arxiv.org/abs/2609.12231
作者: Tiberiu-Andrei Georgescu,Dalal Alrajeh,Sebastian Uchitel
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI); Logic in Computer Science (cs.LO)
备注: 59 pages (39 pages main text, 20 pages appendices), 24 figures, 5 tables. Submitted to ACM Transactions on Software Engineering and Methodology
Abstract:Reactive synthesis is a powerful tool for generating correct-by-construction controllers from formal specifications. GR(1) is an assume-guarantee specification framework that enables efficient synthesis, allowing synthesised controllers to be used in a wide array of applications. The limitation of such controllers is that, should they encounter environment behaviour unspecified in the assumptions of the specification, the specified system guarantees are no longer ensured. Our work proposes an approach based on oracle-guided inductive synthesis to adapt the specification to be consistent with the observed assumption violation, while degrading system guarantees as little as possible to maintain realisability. Our methodology discovers multiple potential solutions, so we propose a preference criteria, based on the ability of the specification to enable robustness under adaptation. Although our approach is capable of degrading the entire specification, for our case studies we successfully discover degradations that preserve the entire set of original guarantees.
[AI-80] Predicting Collision Cross Sections with GRACE: Geometric Residual Adduct Conditioning via Early-fusion
链接: https://arxiv.org/abs/2609.12223
作者: Parthasarathy Suryanarayanan,Susanta Das,Shreyans Sethi,Kenneth M. Merz Jr.,Joseph A. Morrone
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Biomolecules (q-bio.BM)
备注: 35 pages, including 10 figures and 18 tables
Abstract:Collision cross section (CCS), derived from ion mobility mass spectrometry, is a common descriptor for molecular annotation. Prediction is challenging for machine learning models because it reflects the size, shape, and ionization state of a gas-phase molecular ion. Most predictors either ignore explicit 3D structure or treat adduct identity as a late categorical feature, which limits their ability to capture adduct-dependent geometric effects. We present GRACE (Geometric Residual Adduct Conditioning via Early-fusion), a 3D CCS predictor that adapts a pretrained molecular geometry encoder using geometric residual adduct conditioning via early fusion. GRACE combines two inductive biases: a residual objective relative to an adduct-aware physical descriptor baseline and adduct conditioning within the encoder via a learned adduct token and low-rank attention adapters. We evaluate the model on a curated set of over 9,000 experimental molecule-adduct CCS records with random, scaffold, and adduct-sensitive splits designed to separate interpolation, scaffold generalization, and adduct-driven generalization. GRACE achieves the best mean percentage difference among the evaluated learned models on all three splits: 1.67% on the random split, 2.11% on the scaffold split, and 2.36% on the adduct-sensitive split. Diagnostic analyses suggest that residual learning stabilizes training by removing the dominant mass-CCS trend, while early fusion improves adduct-sensitive prediction relative to late fusion. Across four independent external test sets, GRACE shows consistently lower error than the other evaluated models. On a held-out set, GRACE also attains the lowest mean percent difference when compared with four previously reported physics-based workflows. These results support residual learning and encoder-level adduct conditioning as practical inductive biases for fast, accurate CCS prediction.
[AI-81] Retrieval-Augmented Generation for Scientific Code Understanding
链接: https://arxiv.org/abs/2609.12190
作者: Aaron Nobile,Andreas Adelmann,Mohsen Sadr
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models have become central to modern coding assistants, but state-of-the-art systems such as Claude Code or Codex rely on very large, cloud-hosted models with significant computational cost and data-privacy implications. This work investigates whether a useful, fully local coding agent can be built around small open-source models by shifting the computational burden away from inference. We develop a Retrieval-Augmented Generation (RAG) system for scientific code understanding that strictly separates an expensive offline ingestion stage parsing, structural graph construction, LLM-generated entity explanations, and embedding from a lightweight online answering stage. The system is evaluated on a 100-question benchmark spanning eleven categories over the IPPL scientific codebase written in C++, with answers scored by an independent frontier model as the judge. Across seven answering models, we find that model family and retrieval quality matter more than parameter count, i.e. a 9B model achieves the highest average score (0.795), outperforming both larger models within our pipeline and the same models embedded in the Claude Code retrieval architecture. The results indicate that front-loading code understanding into a reusable, codebase-specialised vector store enables small local models to deliver grounded and repository-specific answers, making the agent well suited as a privacy-preserving development tool for in-house scientific codebases.
[AI-82] Explanations-Driven Active Feature Acquisition for Algorithmic Recourse
链接: https://arxiv.org/abs/2609.12179
作者: Vinura Galwaduge,Jagath Samarabandu
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:Algorithmic recourse methods typically assume that a predictive model has access to all features of an individual. In practice, decisions are often made with partial information, because features are costly to acquire. Active feature acquisition addresses cost-constrained prediction, but existing methods are explanation-agnostic: prior work provides explanations only after acquiring additional features, rather than using explanations to drive acquisition. This work flips that and treats algorithmic recourse and feature acquisition jointly. We use Markov Blanket theory to unify counterfactual, semifactual, and alterfactual explanations and to characterize how available recourse grows as features are acquired. Building on this framework, we propose an Explanation-Driven Feature Acquisition (EDFA) method that selects features by explanatory value per unit cost. The framework is further extended with distribution-free validity guarantees for recourse issued from partial information, which signal trustworthy, lower-cost recourse, along with a lower bound on the calibration data required to certify them. Experiments on 7 publicly available datasets with neural network-based predictive models show that EDFA acquires substantially fewer features than state-of-the-art AFA baselines while maintaining comparable accuracy and yielding more decision-relevant, actionable recourse. The implementation is available on GitHub.
[AI-83] WinSyn: An Automated Pipeline for Realistic Enterprise Question-Answering Evaluation
链接: https://arxiv.org/abs/2609.12171
作者: Amey Varhade,Ananya Sutradhar,Ravishankar Krishnaswamy,Navin Goyal
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Enterprise settings provide a challenging environment for question-answering agents, which often rely on Retrieval-Augmented Generation, Deep Research (DR), and related techniques. Much of this challenge comes from the complexity of enterprise data: information is often spread across evolving and potentially conflict- ing emails, chat messages, documents, and other artifacts. Existing benchmarks typically have limited real-world complexity, short-form responses, and unnatural queries, so they often fail to capture the challenges of enterprise settings. In this work, we introduce an automated pipeline for generating synthetic datasets of emails reflecting realistic workplace scenarios, along with long- and short-form questions and gold answers grounded in the data. Our method simulates long-running enterprise projects spanning several months and involving up to 25 interacting employees across multiple roles. The data emphasizes ambiguity, distributed information, and naturally occurring queries. To validate the pipeline, we evaluate few standard agentic baselines on our datasets using the latest frontier models. We find that aggregate scores averaged over all queries remain below 80% for each dataset, indicating significant room for improvement. These findings suggest that more work remains to be done for enterprise deployment and underscore the importance of realistic, high-complexity evaluation data for developing stronger real-world enterprise DR systems.
[AI-84] GLARE: Generative Learning via Adversarial Reward Estimation For Social Dynamics Forecasting
链接: https://arxiv.org/abs/2609.12165
作者: Tenghao Huang,Zhaoxuan Tan,Muhao Chen,Jonathan May,Mengting Wan,Longqi Yang,Pei Zhou,Sihao Chen
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Meeting continuation requires tracking the agenda, speaker roles, participant intentions, and disagreement across long multi-party discussions. We introduce the Meeting Dynamic Forecasting Benchmark (MDFB), constructed from 2,207 real-world meetings and 24,794 future-facing queries. Given a transcript prefix and an active question, a model generates a plausible multi-turn continuation in one call. We evaluate utility—progress toward the question—and human-likeness—plausible conversational flow and role consistency—without requiring exact reproduction of the observed future. We further present GLARE, an adaptation of adversarial imitation learning to conditional language generation. A discriminator ranks the observed continuation above samples from the current actor, and its score supplies a KL-regularized policy reward; retraining on current-policy negatives allows the reward landscape to evolve with the actor. GLARE attains average human-evaluated win rates of 0.66 on utility and 0.70 on human-likeness, outperforming SFT and SPIN while remaining below the observed human continuation. We also demonstrate MDFB as a social reasoning arena for comparing general-purpose models, including closed-source systems, through reference-assisted judgments. Together, these studies illustrate the benchmark’s use for both task-specific learning and output-based evaluation of meeting behavior.
[AI-85] A decision-basis contract for auditable LLM -assisted medical billing verification: deterministic rules verbatim evidence and fail-closed abstention
链接: https://arxiv.org/abs/2609.12156
作者: Jan Hölter,Kevin Geis,Benjamin Raab,Boris Bauke
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:This work presents a proof of concept for auditable LLM-assisted medical billing verification based on a decision-basis contract. The contract separates deterministic checks of versioned fee-catalog rules from LLM-based assessment of free-text documentation. The deterministic layer resolves the applicable catalog release and checks code availability, quantity limits, and exclusions. The semantic layer classifies each claimed item as supported, contradicted, or missing required information. Support and contradiction require a verbatim evidence span; unavailable rule context, unsuccessful assessment, or missing required evidence prevents support through fail-closed abstention. We evaluated four locally run open-weight models on a synthetic catalog and 36 curated cases under the contract, an ablation without explicit documentation requirements, and an end-to-end baseline. Outcome agreement varied across models and showed no consistent advantage over the baseline. Explicit documentation requirements improved identification of missing information for all four models. The evidence gate also exposed cases in which correct raw judgments lacked valid evidence and were converted to incomplete decision-basis entries. The results show how explicit decision records can make rule findings, documentation judgments, and abstention reasons inspectable. Evaluation on real catalogs, independently annotated documentation, and with human reviewers is required to assess practical value.
[AI-86] Neural Multichannel Distant Speaker Diarization with Heavy-tailed Source Separation Model
链接: https://arxiv.org/abs/2609.12154
作者: Sicheng Mao,Baihan Li,Mathieu Fontaine,Anthony Larcher,Roland Badeau
类目: ound (cs.SD); Artificial Intelligence (cs.AI)
备注: Accepted by IEEE SLT 2026
Abstract:Distant speaker diarization remains challenging due to difficult acoustic environments, varying numbers of speakers and overlapping speech. Model-driven methods are proposed to exploit the speech source features in multi-channel recordings that help diarization. This paper generalizes a neural model that jointly learns to perform blind source separation and diarization over speech mixtures (neural FCASA) with heavy-tailed models. The popular Gaussian distribution has been applied for variance modeling in the original source separation model, which we replace with two families of heavy-tailed models (Leptokurtic Generalized Gaussian distribution and Student’s t distribution) to better capture the heavy-tailedness in speech signals. Thanks to the Gaussian scale mixture model, we are able to unify the proposed method and the original one under the same form of learning objective. Our experiments show consistent large improvements in Diarization Error Rate (DER) and Jaccard Error Rate (JER) compared to the baseline on various corpora.
[AI-87] Mined from Scientific Literature: Process Schemas for Atomic Layer Deposition and Etching in Materials Science
链接: https://arxiv.org/abs/2609.12139
作者: Sameer Sadruddin,Eleni Poupaki,Alex Watkins,Bora Karasulu,Adriaan J. M. Mackus,Erwin Kessels,Sören Auer,Jennifer D’Souza
类目: Artificial Intelligence (cs.AI); Materials Science (cond-mat.mtrl-sci)
备注:
Abstract:Atomic layer deposition (ALD) and atomic layer etching (ALE) are reported heterogeneously across experimental and simulation literature in materials science, hindering comparison and machine-actionable reuse. We present four domain-expert-reviewed JSON Schemas for ALD and ALE experimental and simulation processes. Curated with schema-miner and grounded in QUDT using schema-miner pro, the schemas structure materials, process conditions, configurations , and measured or predicted results. We compare their scope, structure, and semantic grounding, and demonstrate their use for schema-guided literature extraction and publication of structured records through ORKG templates.
[AI-88] When Successful Knowledge Graph Edits Displace Correct Answers: Rank-Level Locality beyond Parameter Support
链接: https://arxiv.org/abs/2609.12116
作者: Yi-Cheng Lai,Jerry Wang,Hsin-Ling Hsu,Li-Chu Chi,Ya-Wen Teng,Hen-Hsen Huang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Editing a knowledge graph embedding (KGE) model to promote a desired answer can displace correct answers from the returned list. Locality tests based only on facts that reuse the edited parameter can miss this ranking effect. We introduce a common rank-displacement audit at three scopes: facts supported by the edited parameter, other correct answers to the target query, and correct answers across queries with the same relation. We also derive dimensional and geometric conditions for an update to improve the target while exactly preserving selected scores. On FB15k-237 with DistMult and ComplEx, direct promotion always moves the target into the top ten, but does so without damage in only 23.0–23.2% of edits. Strict preservation causes no measured damage, yet succeeds in only 1.3–1.4%. Support-regularized entity editing gives the highest joint success, 36.3–37.7%, while rank-truncated preservation reaches 32.8–34.7% and reduces the mean number of displaced answers from about 14 to 1.2. Experiments across dimensions, scorers, ranking conventions, and a learned editor show that locality depends on both the protected scope and the editing mechanism. KGE editing should therefore report correction success together with the incidence and severity of rank displacement.
[AI-89] DU-NO: A Parameter-Efficient Double U-Shaped Neural Operator for Phase-Resolving Wave Modeling
链接: https://arxiv.org/abs/2609.12115
作者: Enrique Hernandez Noguera,Md Meftahul Ferdaus,Nathan Cooper,Elias Ioup,Mahdi Abdelguerfi
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Phase-resolving wave models such as FUNWAVE-TVD are the accuracy standard for nearshore dynamics, resolving the shoaling, refraction, and breaking of individual waves, but their cost rules them out for the ensembles, uncertainty quantification, and real-time warning that operational forecasting demands. Neural operators promise solver-level accuracy at a fraction of that cost, yet on wave-dominated fields the accurate ones are large: hybrid spectral-convolutional operators such as U-FNO (the strongest baseline in our study after DU-NO) buy their fidelity with tens of millions of parameters. We introduce DU-NO (Double U-shaped Neural Operator), a multiscale U-shaped spectral operator that attaches lightweight convolutional U-Net branches only at its two shallowest encoder and decoder levels. The placement follows a sampling argument: high-wavenumber content exists only on fine grids, so the local, full-band pathways go where that content lives, while the coarse, band-limited levels stay purely spectral. A depth-decaying mode schedule holds the model to 3.64M parameters, an order of magnitude below U-FNO. On our publicly released FUNWAVE-TVD benchmark, DU-NO attains the best autoregressive rollout error of six identically trained architectures, improving on U-FNO by 14.9% with 10.8x fewer parameters, and a frequency-band analysis shows the gain holds across all bands, including the high-wavenumber band where truncated-spectral operators collapse. Parameter-matched controls confirm the gain is architectural: rescaled to the same 3.6M budget, the best baseline still trails DU-NO by 28.6%. The advantage carries beyond nearshore waves: DU-NO matches the strongest baselines on 2D Navier-Stokes and wins clearly on PDEBench shallow-water rollouts. Code, trained models, and evaluation artifacts are available at this https URL.
[AI-90] Language Is an Insufficient Substrate for Quantitative Reasoning and Consequential Domains Need Large Quantitative Models
链接: https://arxiv.org/abs/2609.12105
作者: Reuben Vandeventer,David Imrem,David J. Wild
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:The prevailing assumption in applied machine learning is that progress on consequential quantitative decisions such as pricing risk, allocating capital, triaging patients, or containing a network intrusion will follow from progress in large language models (LLMs). A language model is trained on a representation of the world that was produced by human description; description is a lossy encoding of the quantitative record, and the loss is irreversible: no downstream model, at any scale, can recover from a description what the description did not encode. We formalize this as a property of the representation on which a model is trained rather than of the model capacity, and we identify three further properties that consequential settings demand of a model and that a language substrate cannot supply by construction: reproducibility, lineage from every output back to the source records that produced. it, and calibrated uncertainty. We argue that these properties define a distinct model class, which we call the Large Quantitative Model (LQM).
[AI-91] Competence-Gated Pooling of Language Models and Priors for Event Forecasting
链接: https://arxiv.org/abs/2609.12101
作者: Aditi Tiwari,Aashrith Bandaru,Heng Ji
类目: Artificial Intelligence (cs.AI)
备注: 8 pages, 3 figures, 2 tables
Abstract:In hybrid forecasting, a language model is often one of several available signals. A system may already have a market, crowd, or statistical forecast and must decide whether the model adds useful information or should be ignored. The relevant target is therefore not standalone model accuracy, but relative competence, defined as the model’s marginal value beyond the available external forecast. Under Brier loss, we characterize when model disagreement can improve an external forecast and derive the gain from using domain-specific rather than global pooling weights. We then introduce a competence gate that estimates domain-level source weights from resolved outcomes, shrinks uncertain estimates toward a global weight, and recalibrates the pooled forecast. Across 2,357 resolved binary questions and five language models, the gate improves the main external baseline from 0.0771 to 0.0732 Brier and significantly outperforms global forecast combinations. The gain remains significant under leakage controls and against a leakage-safe time-series prior on the pooled structured set, with separate evidence on FRED. In contrast, the gate gives no significant improvement on the official ForecastBench market subset, where it largely defers to the market. Across four Qwen models, verbal confidence does not reliably identify when the model outperforms the external forecast, while outcome-estimated competence supports better abstention decisions. These results provide a practical approach for selective model use based on measured marginal value.
[AI-92] Reality Is the Final Verifier: On Two Key Gaps in Agent ic Software Engineering
链接: https://arxiv.org/abs/2609.12039
作者: Alexander Krentsel,Shubham Agarwal,Mert Cemri,Shu Liu,Sidharth Sankhe,Ziming Mao,Matei Zaharia,Ion Stoica
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注:
Abstract:Software development follows an implementation-verification loop in which developers or agents iteratively revise an implementation until an evaluator, such as a test suite, accepts it. The evaluator checks the implementation against a set of requirements under a model of the deployment environment. Yet even a formal proof that the implementation satisfies the requirements under the model cannot guarantee acceptable behavior after deployment. Requirements only approximate stakeholder intent, and the model only approximates the real deployment environment. We call these together - requirement gap and model gap - the two-gap framework, which unifies the main failure modes of agentic software engineer-ing: reward hacking exploits omissions in the requirements or model, while hallucination widens the gaps by fabricating requirements or environment assumptions. Because neither gap can generally be certified closed in an open, changing world, the goal shifts from closing them to continuously narrowing them. We therefore propose an assurance-revision loop that uses deployment evidence to revise the requirements, model, or evaluator when stakeholders reject the resulting behavior. We then cast assured agentic development as a resource-allocation problem over human judgment, agent capability, and compute. The two principal bottlenecks mirror the two gaps: human judgment for the requirement gap and faithful, costly evaluation for the model gap. Reality remains the final verifier: acceptable behavior under actual deployment conditions is the ultimate test, while predeployment evaluations remain proxies for it. Subjects: Software Engineering (cs.SE); Artificial Intelligence (cs.AI) Cite as: arXiv:2609.12039 [cs.SE] (or arXiv:2609.12039v1 [cs.SE] for this version) https://doi.org/10.48550/arXiv.2609.12039 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-93] Pelican-Sim 1.0: A General World Model Simulator for Embodied Intelligence
链接: https://arxiv.org/abs/2609.12036
作者: Shilong Zou,Shilin Zhang,Yingji Zhang,Yuhang Huang,Yi Zhang,Zeyuan Ding,Han Dong,Junwei Liao,Yong Dai,Jian Tang,Xiaozhu Ju
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注: Project page: this https URL
Abstract:In this technical report, we propose Pelican-Sim 1.0, a general world model simulator for embodied intelligence that predicts future observations from visual context and robot actions to support downstream learning and decision making. The model incorporates four key design features: (1) Unified action representation: a 28-dimensional action value space covering most mainstream embodiments, keeping one model valid across heterogeneous devices. (2) Action-visual injection: URDF- and camera-rendered action videos bridge actions and pixels, giving markedly better controllability across embodiments, scenes, and tasks (PSNR +0.904 over alternative fusion baselines). (3) Sparse mixture-of-experts (MoE): sparse MoE layers add capacity for heterogeneous dynamics and absorb the action modality while reducing inter-modality conflict (FVD -6.530 vs. the dense backbone). (4) Efficient rollout generation: causal adaptation and few-step distillation yield a four-step autoregressive simulator, achieving a 5.67-fold speedup over the 35-step model. Benefiting from these designs, we train on approximately one million real-world and simulated trajectories and obtain large gains in action controllability and video quality: PSNR improves over the strongest evaluated baselines by 4.636 on AgiBotWorld Beta, 2.080 on RoboMIND, and 10.343 on RoboTwin, with the adapted EWMBench DYN score up 0.426 on RoboTwin. Relying on this, four downstream applications on RoboTwin succeed: 500 generated trajectories added to 50 demonstrations per task raise policy success from 70% to 93%; policy evaluation reaches a Pearson correlation of 0.994 across five checkpoints; and relative success gains reach 47.7% for action selection and 20.3% for policy improvement. Qualitative generalization across trajectory, scene, object, embodiment, and viewpoint shifts highlights its potential as a general-purpose world model simulator.
[AI-94] Reading the Whole Heart: Latent-Attention Masked Autoencoders for Multimodal Cardiac Representation Learning
链接: https://arxiv.org/abs/2609.12035
作者: Andrea Agostini,Simon Böhi,Moritz Vandenhirtz,Samuel Ruiperez-Campillo,Max Krähenmann,Silke Mühlstedt,Irene Cannistraci,Ece Özkan Elsen,Julia E. Vogt,Thomas M. Sutter
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Cardiovascular diagnosis rests on integrating complementary modalities, like ECG, echocardiography, chest radiographs, and clinical variables, each capturing distinct but correlated aspects of cardiac physiology. Yet most medical foundation models remain modality-specific, combining modalities only for finetuning or post-training. This discards the cross-modal evidence clinicians naturally integrate and ignores the structure within each modality. We introduce Latent-Attention Masked Autoencoders (LAMAE), a multimodal, structure-aware masked autoencoder that jointly learns patient-level representations during self-supervised pretraining. Rather than fusing modalities post hoc, LAMAE exchanges information directly in the latent space through a shared latent-attention module operating over a study-view-entity hierarchy, enabling aggregation of variable observations and graceful handling of missing modalities. Pretrained on over 1.2 million MIMIC-IV hospital stays, LAMAE outperforms modality-specific pretraining and strong contrastive and vision-language baselines across multimodal hospital-stay tasks, such as in-hospital mortality, ICD-10 and DRG coding, and length of stay, while remaining competitive on unimodal tasks. These gains persist even when only a single modality is available at test time, showing that modeling both intra- and inter-modal structure yields more robust, transferable representations.
[AI-95] When Agent Metrics Measure Different Things: An Evidence-Grounded Audit of the Praxa AI Pipeline
链接: https://arxiv.org/abs/2609.12017
作者: Stefan G. Creadore,Peyton Woakz
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: 22 pages, 4 figures. Retrospective measurement audit with aggregate data, offline analysis code, and a prospective, unexecuted operator-coverage catalog in ancillary files
Abstract:Agent evaluations can be numerically correct while measuring a different construct from the one implied by their labels. We present a retrospective measurement audit of selected Praxa AI implementation files, historical evaluation artifacts, and operational records. A 139-case offline routing report contains 112 passes and 27 failures despite zero gating failures, because known gaps are explicitly exempted from the gate. An identifier-free export represents 8,843 tool-attempt rows: 8,395 recorded durations and 448 missing values. Of the durations, 121 equal the signed 32-bit maximum and carry abandoned-client labels; inspected database code clamps elapsed lifecycle age. The pooled recorded 99th percentile is 2,147,483,647 ms, versus 38,118.31 ms among server-observed completed calls. This is a stratum contrast, not a treatment effect. In a documented single-trajectory compaction pilot, the reported follow-up input reduction is 94.39%, but the reduction across the trigger and follow-up calls together is 46.54%. We reproduce the descriptive calculations, verify 91 timing statistics through a separate weighted rational-arithmetic implementation, and execute 13 scoring-function and 12 analysis-verifier tests. Finite-completion bounds show how missing durations limit all-row timing statements without imputing values. The contribution is a source-linked case study and reusable verification package for separating gate policy, lifecycle timing, and request-level accounting from broader agent-performance claims. Historical provider runs and the full current pipeline were not independently reproduced; general capability superiority and population-level statistical significance are not established.
[AI-96] Can We Trust LLM Judges: A Study of Capability-Dependent Biases and Multi-Judge Ensemble for Bias Calibration
链接: https://arxiv.org/abs/2609.12002
作者: Gemma Zhang,Prachi Badarayani,Asmi Kumar,Sadid Hasan,Sulaiman Vesal
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:LLMs are increasingly used as automated judges for model training and evaluation, yet individual judges exhibit systematic biases that undermine reliability. Much of prior work has studied biases in pairwise LLM-as-a-judge settings; in this paper, we focus on absolute scoring tasks, which mirror more realistic use cases. Across four benchmarks and six models (36 judge-examinee pairs), we show that a model’s task accuracy strongly predicts its judging accuracy (Pearson r \geq 0.90 on most models) and inversely predicts its directional bias ( r \leq -0.83 ), but that accuracy alone does not ensure fair evaluation: more capable examinee models consistently receive more lenient judgments from all judges ( r \geq 0.83 ). To address this, we propose calibrated weighted majority voting (WMV), an ensemble evaluation method that aggregates multiple LLM judges weighted by online estimates of their false-positive and false-negative rates. We introduce a disagreement-based estimator that derives these error rates purely from inter-judge agreement patterns, requiring no ground-truth labels or task metadata. In a simulated experiment with shifting task distributions, our label-free WMV tracks an oracle with perfect error-rate knowledge to within 0.5 percentage points on average, outperforming both individual judges and unweighted majority voting. These results demonstrate that principled multi-judge calibration can simultaneously improve accuracy and correct for systematic leniency without requiring labeled data, offering a scalable path to reliable automated evaluation as model capabilities increase.
[AI-97] Who Pays for Open Review? Visible Author Reputation and Its Effect on Ratings
链接: https://arxiv.org/abs/2609.11983
作者: Qinghua Zhao,Xinyu Chen,Yanhui Yang,Tengfeng Sun,Junfeng Liu,Zhongfeng Kang
类目: Digital Libraries (cs.DL); Artificial Intelligence (cs.AI)
备注: 11 pages, 6 Tables
Abstract:An OpenReview bug in November 2025 broke anonymity at several conferences and prompted calls for open review, which motivate us to ask what shifting from blind to open would mean for authors. Analyzing over 18,000 reviewed submissions to ICLR 2026, split into de facto open and blind groups by arXiv preprint timing, we find that ratings rise with author reputation under both mechanisms, with a steeper slope under open review that is statistically significant, and that the open-blind difference is concentrated at the borderline ratings. The pattern holds across five reputation proxies (including institution, h-index, and citation count), three author-aggregation rules, and five definitions of the open window. A controlled simulation with five AI models as reviewers, holding the manuscript fixed and varying the author reputation, reproduces the effect. With claude-opus-5 as the reviewer, for example, rating rises by 0.5 points as the author moves from low to high reputation.
[AI-98] Occamy-1.0: Open Pareto-frontier 35B Intelligence for Co-work
链接: https://arxiv.org/abs/2609.11977
作者: Wenhui Chen,Shiwen Cheng,Hao Dong,Chenda Duan,Ruixiang Feng,Zhong Guan,Boqiang Guo,Xueyuan Han,Haojie Hao,Liangmeng Huang,Zhelong Huang,Xinke Kong,Hongyu Li,Jiazheng Li,Junbo Li,Qingchuan Li,Yukun Lian,Chang Liu,Tianyu Liu,Zicheng Liu,Shuyi Ouyang,Yijun Pan,Kunyu Shi,Xiaojun Tang,Bingquan Wang,Kesu Wang,Yuchen Wang,Sibo Wei,Sicong Xie,Xiaoying Xing,Yi Xu,Zhijun Xu,Hongwei Xue,Qingcheng Zeng,Di Zhang,Guannan Zhang,Haochen Zhang,Tianlong Zhang,Tianyu Zhao,Tianyu Zhao,Yanjun Zheng,Jialong Zhu,Zijian Zou
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Co-work agents execute complex workflows that combine information gathering, tool use, coding, and file manipulation across many model invocations. Because cost and latency accumulate over the full episode, their practical value depends not only on peak capability but also on how efficiently that capability is delivered. Yet many steps in everyday work emphasize state tracking, coordination, recovery, and follow-through rather than frontier-scale reasoning. We present Occamy-1.0, a cost-efficient co-work model obtained by further training the post-trained Qwen3.6-35B-A3B checkpoint. We construct execution-grounded data and environments, capture replayable long-horizon trajectories across multiple harnesses, and use staged post-training to develop and consolidate complementary execution capabilities. Across a broad suite of co-work benchmarks, Occamy-1.0 is consistently among the strongest comparably sized models and remains competitive with substantially larger frontier systems on several tasks. Under our stated evaluation and pricing protocol, its aggregate performance across four representative benchmarks places it at the low-cost knee of the observed cost–performance Pareto frontier. Supporting evaluations in tool calling, coding, and instruction following further show that this specialization preserves broad agentic capability. We release the model weights and a subset of the training data to support research on practical co-work agents and agentic post-training.
[AI-99] SoulAuth: An Actor-native Identity Architecture and Rust Reference Implementation for Humans and Long-lived AI Actors
链接: https://arxiv.org/abs/2609.11258
作者: Kun Yuan,Harold Wang,Echo Li,Egusi Gui,Kiki Hu,Lucas Luo,Magnus Hu
类目: Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
备注: 34 pages, 8 figures. Preprint v1.0. Open-source Rust reference implementation and fixed v0.1.0 software artifact: this https URL
Abstract:As AI systems move from transient model invocations toward long-lived actors that persist across credentials, clients, sessions, and runtime instances, identity infrastructure must answer a basic question: where should the canonical continuity boundary be placed? This paper introduces Actor-native Identity and presents SoulAuth, an open-source Rust reference implementation for Humans and long-lived AIActors. We argue that any subject that must persist under its own identity and remain independently attributable should have an ActorIdentity that is not replaced by an Account, Credential, Client, AuthSession, IdentityBinding, or runtime instance. SoulAuth therefore treats Humans and long-lived AIActors as first-class identity subjects while keeping authentication distinct from downstream authority. Methodologically, we use a Philosophical Engineering approach that translates conceptual analysis of subjecthood into identity objects, invariants, lifecycle semantics, system responsibilities, implementation boundaries, and inspectable conformance evidence. Evaluation against the fixed SoulAuth v0.1.0 artifact shows that the implementation realizes core boundaries including Human/AIActor first-class identity status, Client/Actor separation, and Authentication/Authority separation, while gaps remain in unified Credential modeling and historical attribution anchored to ActorIdentity. We therefore report partial, not full, architecture conformance.
[AI-100] AI Safety: Not Optional Not Later
链接: https://arxiv.org/abs/2609.10630
作者: Qinghua Lu,Yoshua Bengio
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: To appear in the January 2027 issue of IEEE Software, as an invited article for the new AI and Software Engineering column
Abstract:Incidents show that AI safety failures often arise across multiple layers. We present a safety-by-design assurance architecture combining model-level supervision, such as Scientist AI, with system-level controls over scaffolds and harnesses, independent verification, monitoring, and evidence infrastructure, supported by governance for accountability and evidence interoperability.
[AI-101] Evidence-Aligned Local Composition of Discrete Experts for Sequence Restoration
链接: https://arxiv.org/abs/2609.05801
作者: Mohammad Panahazari,Usman A. Khan,Shuchin Aeron
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:A document modeled as a discrete sequence of tokens can be thought of as being generated from a composition of texts from different domains; a README file, for example, moves between prose, code, and configuration. When such a document is corrupted and only frozen domain experts are available, restoring it requires deciding both what is missing and which expert to trust at each position, at test time and without region labels or a trained router. We introduce evidence-aligned local composition, which infers a soft, position-wise weighting over the experts from the marginal evidence of the corrupted observation under a given corruption model, estimating the evidence from the experts’ own denoising losses and smoothing the weights across positions. Because the weighting is soft, it recovers a mixture when the true composition is mixed and concentrates on one expert when that suffices. Across a categorical simulator, byte-level experts, and experts fine-tuned from a 1.3 B discrete flow-matching model, the inferred weights track the true regions at 0.85 field accuracy on naturally mixed scientific documents, and at 0.98 on constructed mixtures whose regions are lexically disjoint. Restoration improves over a single global weight when the experts are genuinely distinct and reduces to it when they converge, tracking a measure of expert separation.
[AI-102] 4D Parallelism Unlocks Exascale Bayesian Neural Networks for High-Fidelity Atmospheric Modeling
链接: https://arxiv.org/abs/2609.12815
作者: Deifilia Kieckhefen,Juan Pedro Gutiérrez Hermosillo Muriedas,Lars Helge Heyen,Mathis Bode,Iida Hakulinen,Andreas Herten,Chelsea Maria John,Thorsten Kurth,Anni Moisala,Asena Karolin Özdemir,Kaleb Phipps,Oskar Taubert,Arvid Weyrauch,Markus Götz,Charlotte Debus
类目: Atmospheric and Oceanic Physics (physics.ao-ph); Artificial Intelligence (cs.AI); Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG); Performance (cs.PF)
备注:
Abstract:We present BEAST, the first-ever Bayesian Swin Transformer for atmospheric forecasting on 0.25 ^\circ global resolution able to accurately quantify both aleatoric and epistemic uncertainty. To overcome the associated computational bottlenecks, we devise an orthogonal 4D-parallelization scheme that introduces a unique domain-tensor-parallelism strategy and a novel uncertainty parallel method, enabling us to fully leverage GPU capacity and efficiently scale model training. For a 2.4-billion-parameter model, we achieve a peak performance of 3.96 EFLOP/s on 20,480 NVIDIA GH200 GPUs on the JUPITER supercomputer. We train BEAST as a 700-million-parameter model with 96 random weight samples on 384 nodes on 40 years of data for nearly one million gradient updates. This model achieves predictive skill scores competitive with state-of-the-art probabilistic atmospheric AI models and numerical models, and can predict extreme events with exceptional skill, while generating large ensembles 3 to 4 times faster than the current-best AI model. Our contribution unlocks the potential of high-fidelity uncertainty quantification in atmospheric AI models, heralding a new era for AI-based models in climate and Earth system sciences.
[AI-103] ResoSeg: Resonance Tagger using Transformer and Segment Model
链接: https://arxiv.org/abs/2609.12610
作者: Chunkai Li,Junhao Yin,Ke Li,Jingde Chen
类目: High Energy Physics - Experiment (hep-ex); Artificial Intelligence (cs.AI)
备注: 12 pages, 10 figures
Abstract:Deep learning has been widely applied across many areas of experimental high-energy physics, yet existing models address only event-level classification or object tagging and therefore still require reconstruction algorithms tailored to each decay channel. We present the first application of segmentation to resonance tagging at BESIII and introduce ResoSeg, a deep learning model that jointly performs particle-level segmentation and event-level classification, enabling a one-pass analysis of resonance to anything decays while precisely reconstructing the relevant resonance properties. We demonstrate the reconstruction of \eta_c with e^+e^-\to\pi^+\pi^-h_c , h_c\to\gamma\eta_c , \eta_c\to\textanything . The model is trained on BESIII- \eta_c dataset, which is constructed with per-track true labels obtained via a Truth-Matching Algorithm. Experimental results show that the average combined efficiency of ResoSeg is more than double that of the conventional 16-channel approach across energy points from 4.19 to 4.60,GeV. The model generalizes to unseen energy points, adapts to other \eta_c production modes through transfer learning, and remains robust against variations in the \eta_c mass, width, and branching fractions, providing a general, resonance-aware model applicable beyond \eta_c and BESIII. The source code is available at this https URL.
[AI-104] Optimizing Geoengineering Interventions Using Differentiable Climate Models
链接: https://arxiv.org/abs/2609.12528
作者: Pulkit Dubey,Dorian S. Abbot,Ashesh Chattopadhyay
类目: Atmospheric and Oceanic Physics (physics.ao-ph); Artificial Intelligence (cs.AI); Optimization and Control (math.OC); Chaotic Dynamics (nlin.CD)
备注:
Abstract:The deployment of a geoengineering program to cool Earth’s climate may be imminent. It is crucial that tools be developed to ensure that such a program would achieve its objectives while minimizing disruption. Here we exploit recently developed differentiable atmospheric models to demonstrate a novel geoengineering control strategy. In the differentiable primitive-equation atmospheric model JAX-GCM we impose a uniform +4 ,K ocean warming and ask what pattern of sea-surface temperature cooling – in five ocean-masked zonal bands of prescribed SST forcings whose amplitudes are free – returns land near-surface air temperature closest to the model’s own unwarmed climatology. This idealized set-up represents a cooling pattern that could be delivered physically either by marine cloud brightening or stratospheric aerosol injection. Gradients through chaotic dynamics decorrelate from the true sensitivity beyond the Lyapunov horizon, so we optimize greedily over segments of 8 to 14 days, following receding-horizon control. The learned strategy removes 92.3 \pm 0.4% of the realized land warming across a ten-member ensemble of two-year rollouts, and a three-year run sustains it. If we use the spatial pattern of land temperature as the optimization objective, the distributions of precipitation, evaporation, and specific humidity over land are restored as well, even though they are not included in the objective function. The learned strategy from JAX-GCM replayed in the AI emulators LUCIE and NeuralGCM without re-optimization is successful, suggesting robustness. These promising results demonstrate a strategy for designing optimal climate interventions that can be applied broadly for geoengineering scenarios under consideration.
[AI-105] Linear Exponential Quadratic Gaussian Covariance Steering
链接: https://arxiv.org/abs/2609.12463
作者: Chiran B. Cherian,Yasemin Isik,Abhishek Halder
类目: Optimization and Control (math.OC); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Systems and Control (eess.SY); Machine Learning (stat.ML)
备注:
Abstract:We formulate and analyze the linear exponential quadratic Gaussian (LEQG) covariance steering problem in continuous time over a given deadline (finite time horizon). The solution for this problem can be seen as a risk-sensitive Schrödinger bridge between Gaussian endpoints in the linear quadratic setting. Unlike the risk-neutral case, the LEQG covariance steering controller–still a linear state feedback–can no longer be written in closed form. We show that the optimal controller is parameterized by a symmetric matrix solving an algebraic equation that encodes the implicit dependence on the risk-sensitivity parameter. We explain how the structure of this optimal controller significantly generalizes the existing results for the risk-neutral case. Building on these results, for the matched noise and input channel case, we prove the existence-uniqueness of solution for the LEQG covariance steering problem in the neighborhood of the known risk-neutral optimal solution. We give an illustrative numerical example.
[AI-106] RF-VoID: Towards Bandwidth-Efficient Exterior Tile Void Detection via Narrowband Radio-Frequency Representation Learning
链接: https://arxiv.org/abs/2609.12388
作者: Xinyan Chen,Ruiqin Ma,Shunsuke Shoda,Changyu Zhou,Ryo Natsuaki,Akira Hirose,Jianfei Yang,Li Yi
类目: ignal Processing (eess.SP); Artificial Intelligence (cs.AI)
备注:
Abstract:Hidden debonding behind exterior ceramic tiles is a falling-tile hazard, and millimeter-wave radar offers a non-contact way to find it. Conventional interpretation first reconstructs a range profile, so its reliability is bounded by the available bandwidth, yet bandwidth is what sets the cost, the acquisition time, and the regulatory footprint of a deployed system. This work asks whether that bandwidth can be traded for computation. A 4-40 GHz stepped-frequency system scans twelve exterior-wall specimens containing 0.5-1.0 mm air voids at different depths and interfaces, and the bandwidth dependence of A-scan, B-scan, and C-scan interpretation is analyzed to establish the resolution bound. RF-VoID is then proposed, which decides directly on the narrowband complex response: the sub-band is kept in its measured frequency order with amplitude and phase alongside the in-phase and quadrature channels, a dual-branch encoder reads it along the physical frequency axis using relative position encoding and a distance-dependent locality bias, and an inspection-oriented objective handles the class imbalance and the asymmetric error cost of facade screening. Under a mixed-sample protocol the method attains 98.61% accuracy and a 95.84% F1-score with 0.5 GHz of bandwidth, a seventy-two-fold reduction relative to the full sweep, without range-profile reconstruction, deconvolution, or depth-slice selection; on specimens held out entirely from training it remains the strongest of the compared models, with a mean macro F1-score of 62.12% at 0.5 GHz that rises to 68.57% at 1 GHz.
[AI-107] Agent ic TCAD Calibration Workflow for Oxide Semiconductor Transistors
链接: https://arxiv.org/abs/2609.12184
作者: Gyujun Jeong,Junmo Lee,Sungwon Cho,Woohyun Hwang,Kwangyou Seo,Suhwan Lim,Wanki Kim,Daewon Ha,Rishi Ranade,Kihang Youn,Ram Cherukuri,Yiyi Wang,Asif Khan,Shimeng Yu
类目: Instrumentation and Detectors (physics.ins-det); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Applied Physics (physics.app-ph)
备注: Preprint. 8 pages, 6 figures. Submitted to an IEEE journal for review
Abstract:Experimental TCAD calibration is essential for predictive technology modeling of emerging oxide semiconductor transistors. However, it remains time-consuming and expert dependent because of model ambiguity. Multiple physical models and parameter sets can reproduce the same measured transfer characteristics, while local fitting alone cannot uniquely identify the underlying device physics. We present the first demonstration of an agentic TCAD calibration workflow for a fabricated bottom-gate In–W–O (BG-IWO) transistor. Starting from the measured transfer curve and device information, the workflow uses measurement–TCAD residuals and local sensitivity tests to select bounded parameter corrections or evaluate additional physical models, and accept only updates that improve device metrics. The LLM agent orchestrates the workflow, while Sentaurus governs the device physics. For the 2%-W reference device, five agent-suggested updates yield a fixed calibrated model, reducing the multi-metric device objective J by 14.3 \times . Maximum V_\mathrmth / I_\mathrmon errors are 36.1~mV/0.022 decade for varying-drain-bias tests and 46.2~mV/0.062 decade for varying-channel-length tests, demonstrating model transferability across bias and geometry rather than a local parameter fit. W-composition tests provide process-sensitive insight. This agentic workflow provides a faster route to model development for emerging device technologies.
[AI-108] he Anatomy and Boundary of Adaptation under Temporal Tabular Shift
链接: https://arxiv.org/abs/2609.12136
作者: Tianyu Wang,Xi Vincent Wang,Lihui Wang,Mian Li,Zhihao Liu
类目: Methodology (stat.ME); Artificial Intelligence (cs.AI)
备注:
Abstract:Prequential adaptation of frozen tabular foundation models under temporal drift, with each label revealed only after prediction, helps some deployments and harms others, yet current practice does not predict which. We study the sources and limits of these gains. A diagnostic anatomy attributes gains to four recurring mechanisms under a streaming protocol that removes three optimistic biases and quantifies a fourth. Within an agnostic total-variation drift class, the target conditional is only partially identified: its identified-set diameter, the \emphwall, is irreducible from unlabeled data uniformly in sample size. A second, orthogonal L^2 projection wall quantifies what the frozen representation cannot express. Two canonical mechanism priors collapse the first wall. Under stated nuisance-rate conditions, the wall can be estimated from labeled historical windows at a \sqrt N rate above the margin threshold \gamma^\star=d_0/(2\alpha_s) . At \gamma=0 , the conditional lower-bound program depends on an open affinity estimate; the positive-margin lower branch also remains open. Semi-synthetic data illustrate the finite-sample mechanism with calibrated exponents. Stream-level proxies on eight industrial streams fall on the difficult side under a stated roughness bound, while the equality case \gamma=\gamma^\star remains unresolved.
[AI-109] Hierarchical Prototype Emergence in Modern Hopfield Models
链接: https://arxiv.org/abs/2609.12079
作者: Aditya Cowsik,Adithya Sriram
类目: Disordered Systems and Neural Networks (cond-mat.dis-nn); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: 5 pages, 19 page supplementary information. Additional citation requests welcome
Abstract:Hierarchical correlations are a universal feature of any realistic model of data, and the question of how associative memory models may learn these correlations and generalize beyond them to construct new sensible images is an important step towards understanding more complex modern architectures such as diffusion models. We consider a hierarchical model for memories which are sampled and stored in a dense Hopfield network with polynomial activation. We analytically derive conditions for each level of this hierarchy to be locally stable - that is they are local energy minima. We use prototype reconstruction as a minimal model of generalization and we find that it takes only a quasi-polynomial amount of information to generalize beyond particular memories and even particular groups in the hierarchy. We observe a qualitatively analogous phase diagram in the number of memories, sharpness of the activation function (polynomial degree) for data from Fashion-MNIST.
[AI-110] Continuous Learning of Gravity Field Irregularities Around Small Bodies via Neural Hamiltonian ODEs
链接: https://arxiv.org/abs/2609.12022
作者: Giacomo Acciarini,Dario Izzo
类目: Instrumentation and Methods for Astrophysics (astro-ph.IM); Earth and Planetary Astrophysics (astro-ph.EP); Artificial Intelligence (cs.AI); Optimization and Control (math.OC); Space Physics (physics.space-ph)
备注: Published in the ESA GNC ICATT 2026 Conference, this https URL
Abstract:We propose to learn the unknown dynamics in the proximity of a small body directly from tracking data, representing them as a feed-forward neural network embedded in the system Hamiltonian. The equations of motion form a Neural Hamiltonian Ordinary Differential Equation, whose variational equations provide exact training gradients: estimation uses position and velocity arcs at realistic noise levels, without acceleration or potential labels, and a continual learning approach warm-starts the network as new data are acquired. The known part of the Hamiltonian carries whatever is available, from the central term and spin state to the constant-density model of the imaged shape. We assess the method against a normalized spherical harmonics expansion estimated from identical arcs through the same machinery, on scenarios built on the shapes of Itokawa, 67P, Bennu and Eros. The network remains usable inside the Brillouin sphere: it plans ballistic descents at Itokawa to \SI4.6m median touchdown error from tracking alone, against 5.1–\SI48.9m for harmonics of degree 4–12, and to \SI0.9m with the imaged shape as prior, a configuration that also recovers localised density anomalies invisible to any harmonics degree. The two representations are complementary, and we discuss their combined use across the phases of a small-body mission.
[AI-111] Assessment of Non-Institutional AI Tool Usage Among Clinicians
链接: https://arxiv.org/abs/2609.11990
作者: Sarah Pungitore,Jarrod Mosier
类目: Physics and Society (physics.soc-ph); Artificial Intelligence (cs.AI)
备注:
Abstract:Generative artificial intelligence (AI) tools are increasingly accessible and have the potential to improve efficiency across clinical workflows. However, clinicians may also use non-institutional AI tools that are not provided, managed, or governed by their healthcare institutions, creating potential concerns related to privacy, security, accuracy, and clinician-AI interaction. Little is known about how clinicians currently use these tools for work-related tasks. We conducted a descriptive survey of clinicians recruited from the University of Arizona College of Medicine-Tucson and Banner University Medical Center-Tucson between May 20 and June 26, 2026. Participants reported their use of AI tool categories and the frequency with which they used AI for specific tasks across five workload categories: administrative work, clinical work, research, studying/continued education, and teaching. Forty-four respondents completed the survey. Forty-three respondents reported using AI for at least one work-related task during the preceding 6 months. Conversational AI and clinical decision support/diagnostic AI were the most used tool categories, each reported by 28 respondents. Administrative and clinical tasks demonstrated the most frequent use. AI was also used for higher-risk activities, including diagnostic assistance and clinical decision support. Non-institutional AI use was common among surveyed clinicians and extended across a broad range of work-related activities, including tasks with potential implications for clinical reasoning and patient care. Further research is needed to characterize how clinicians use these tools, how they evaluate AI-generated outputs, and how AI can be safely and effectively integrated into clinical workflows.
[AI-112] Robust Trust
链接: https://arxiv.org/abs/2602.09490
作者: Piotr Dworczak,Alex Smolin
类目: Theoretical Economics (econ.TH); Artificial Intelligence (cs.AI); Computer Science and Game Theory (cs.GT)
备注:
Abstract:An agent chooses an action based on her private information and a recommendation from an informed but potentially misaligned adviser. With a known probability, the adviser truthfully reports his signal; with the remaining probability, he can send any message. We characterize optimal robust decision rules that maximize the agent’s worst-case expected payoff. Every optimal rule is equivalent to a trust-region policy in belief space: the adviser’s reported beliefs are taken at face value if they fall within the trust region but are otherwise clipped to the trust region’s boundary. We derive alignment thresholds above which advice is strictly valuable and fully characterize the solution in both binary-state and binary-action environments.
机器学习
[LG-0] CanvasAnneal: Curriculum Reinforcement Learning for Diffusion Language Models
链接: https://arxiv.org/abs/2609.13060
作者: Blake Olson,Yuhang Song,Emmett McQuinn,Yuan Shangguan
类目: Machine Learning (cs.LG)
*备注: 16 pages, 2 figures, 6 tables
Abstract:Diffusion Language Models (DLMs) offer promising parallel generation capabilities but lag behind autoregressive models in complex reasoning and tool-use tasks. While Reinforcement Learning (RL) has recently been applied to enhance DLMs, standard RL approaches suffer from an exploration bottleneck. To address this, we inject reasoning priors from a stronger teacher model to guide RL exploration. In this paper, we introduce CanvasAnneal, a curriculum-guided diffusion RL framework. During the initial RL phase, we warm-start exploration by injecting teacher-generated reasoning traces into the initial diffusion canvas. As training progresses, we gradually remove this guidance and require the model to generate more of the reasoning trajectory independently. Across mathematical reasoning and tool-use benchmarks, CanvasAnneal improves over standard diffu-GRPO on MATH500, Countdown, and Tau2 and substantially accelerates reward improvement on several tasks, while gains are task-dependent. Our results suggest that structured training-time guidance can alleviate exploration bottlenecks in diffusion RL and speed up convergence on harder tasks.
[LG-1] Benign Loss Landscapes Can Coexist with Worst-Case Hardness
链接: https://arxiv.org/abs/2609.13057
作者: Zach Furman,Stephan Wäldchen,Yangda Bei,Liam Hodgkinson
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 46 pages, 4 figures
Abstract:Deep neural networks are expressive enough to contain worst-case targets that can be evaluated in polynomial time but cannot be learned in polynomial time by gradient descent. For practical tasks they nonetheless learn well, raising the question of what non-generic structure of real-world targets enables this. Existing surrogate models cannot pose this question because they either lack hard-to-learn targets entirely (deep linear networks) or cannot evaluate such targets efficiently (kernel methods, infinite-width limits). We study tree tensor networks (TTNs), a model class that generalizes deep linear networks and Tucker decompositions. We show they embed arbitrary read-once Boolean formulas, and thus contain polynomial-size targets that cannot be learned by gradient descent in polynomial time under the same mechanism as neural networks. Despite this, we prove that their loss landscapes are conditionally benign for every realizable target: every local minimum that is minimum-norm is global. Thus, surprisingly, bad local minima are not what distinguishes between typical and worst-case problems in TTNs. Instead, learning difficulty in TTNs can arise from high-order degenerate saddle points, which we show are caused by rank-deficiency. This is explored through a case study of the parity function, illustrating the potential for TTNs to relate landscape geometry to computational hardness.
[LG-2] A Unified and Constrained View of Regularization-Based Robust Reinforcement Learning
链接: https://arxiv.org/abs/2609.13050
作者: Amine Andam,Jamal Bentahar,Mustapha Hedabou
类目: Machine Learning (cs.LG)
*备注:
Abstract:Regularization-based methods have become a standard approach for training Deep Reinforcement Learning policies against adversarial input perturbations. In this paper, we unify these methods by deriving new upper bounds on the performance gap between the nominal and worst-case policies. Each upper bound is expressed as an existing regularization objective plus a KL-divergence penalty between the nominal and worst-case policies, which further explains why adding a KL penalty improves robustness in practice. Building on these bounds, we formulate robust training as a constrained optimization problem, showing that existing methods correspond to the special case of a fixed Lagrange multiplier. We instead update the multiplier jointly with the policy to automatically tune the regularization weight. Finally, we conduct extensive adversarial evaluations across several continuous control tasks to validate our theoretical analysis.
[LG-3] MCRL2: Multi-resource Cross-attention-based Representation Learning-augmented Reinforcement Learning for Cloud Microservice Scheduling
链接: https://arxiv.org/abs/2609.13048
作者: Tiangang Li,Shi Ying,Xiangbo Tian,Chuan Shi,Ding Xiao
类目: Machine Learning (cs.LG)
*备注: 15 pages, 14 figures
Abstract:Efficient microservice scheduling is crucial for maintaining load balance across nodes in data centers and ensuring high quality of service. However, achieving this in practice remains challenging due to dynamic resource imbalance under fluctuating workloads, nonlinear coupling across multiple resource dimensions, and the heterogeneity of microservice resource demands. While reinforcement learning-based approaches have shown promise, they struggle to capture the complex interdependencies among heterogeneous resources and neglect the importance of learning informative system representations. To address these limitations, we propose MCRL2, a novel reinforcement learning approach augmented with multi-resource cross-attention-based representation learning for microservice scheduling. Specifically, we first propose MCRL, a novel representation learning approach that captures structured and informative interactions among nodes, resources, and microservices via a multi-resource cross-attention mechanism. Then, MCRL2 augments reinforcement learning through MCRL-enhanced actor-critic architecture combined with a maximum entropy objective, improving system state expressiveness and leading to more stable and effective scheduling decisions. Extensive experiments on real production cluster traces demonstrate that MCRL2 significantly outperforms existing baselines in load balancing, scheduling success rate and average completion time across diverse workload patterns.
[LG-4] Robust Policy Optimization via Adversarial Importance Sampling
链接: https://arxiv.org/abs/2609.13044
作者: Amine Andam,Jamal Bentahar,Mustapha Hedabou
类目: Machine Learning (cs.LG)
*备注:
Abstract:Significant progress has been made in safeguarding deep reinforcement learning (DRL) policies against input perturbations. Developing robust DRL involves three main stages: algorithm design, implementation, and evaluation. In this work, we identify and address a key limitation at each stage. First, we introduce Adversarial Importance Sampling (Advis), a method that uses importance sampling over trajectories from standard training to estimate and optimize verifiable worst-case returns. Advis satisfies three desirable criteria not jointly achieved by prior work: it requires no additional environment interactions, no auxiliary networks, and captures long-term robustness. Second, we introduce advrl, a modular PyTorch library that provides clean, single-file implementations of existing robustness methods and adversarial attacks, facilitating rapid prototyping and enabling reproducible and traceable evaluations. Third, we revisit evaluation under learned adversaries and show that optimal adversarial hyperparameters do not transfer across agents, which can lead to an overestimation of robustness when using a limited set of attacker configurations. Accordingly, we evaluate policies against a large and diverse set of attackers, using 6-14x more configurations than prior work. Finally, we evaluate our approach on continuous control environments, demonstrating its effectiveness relative to existing baselines. The code is available at: this https URL
[LG-5] Quantile-based Loss Filtering for Outlier-Robust Stochastic Gradient Descent
链接: https://arxiv.org/abs/2609.13040
作者: Jamie Haddock,Anna Ma,Elizaveta Rebrova
类目: Machine Learning (cs.LG); Numerical Analysis (math.NA)
*备注:
Abstract:We study loss-based filtering for finite-sum optimization with a subset of corrupted component functions whose gradients may be highly unreliable. Motivated by minimum-loss-based SGD (min- k -loss) and quantile-based methods for corrupted linear systems, we propose and analyze a general loss-filtering framework – Quantile-(k)-Loss SGD (Q(k)L-SGD) – that samples (k) component losses at each iteration and updates using an index chosen uniformly from the lower empirical (q)-quantile. We prove linear convergence of this family of methods under standard convexity assumptions, requiring the sample size to scale with the number of corruptions and a subset strong-convexity threshold. For the cases when large enough sampling is impossible or undesirable, we give a complementary small-sample probabilistic analysis that covers any sample size k and the convergence behavior depends on the probability of selecting an outlier and on the curvature of the selected good step. Experiments on polynomial regression, regularized logistic regression, and regularized hinge loss show that intermediate quantiles often outperform both standard SGD and min-(k)-loss SGD. In particular, min-(k) often stalls by repeatedly selecting nearly solved components, while intermediate quantiles retain robustness and produce more informative updates.
[LG-6] ransfer Learning for Evolving Domains
链接: https://arxiv.org/abs/2609.13039
作者: Ricardo Ribeiro Pereira,Jacopo Bono,Hugo Ferreira,Pedro Ribeiro,Pedro Saleiro,Pedro Bizarro,Carlos Soares
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注: 24 pages, 2 figures, 2 tables
Abstract:Transfer learning explores how to leverage knowledge from various tasks or domains (sources) to enhance predictive performance in related tasks or domains (targets). Typically, transfer learning research is segmented into several isolated sub-areas (such as domain generalisation, domain adaptation, or multi-domain learning), each making distinct assumptions about target data availability, namely how much data and how many labels are available at training time. However, in many real-world applications, data availability is not fixed but evolves over time, as instances and labels are progressively collected from a new domain. Each of the classical settings then describes only a snapshot of a trajectory that a deployed system must traverse in full. We formalise this trajectory as a transfer learning problem in its own right, Transfer Learning for Evolving Domains (TrED), specified by a data availability process fixed by the environment, a learning protocol that the method is free to choose, and an evaluation criterion that scores the whole trajectory of models rather than a single one. Within this formalism, the classical settings are recovered as regimes that a learner may pass through, rather than as separate problems that TrED concatenates. We then examine the transfer learning literature to identify mechanisms that are promising building blocks for a solution, and find that most methods are tailored to a single regime and that even the strongest existing candidates do not yet optimise the whole trajectory. We argue that TrED is a well-posed and unsolved problem, and an important direction for future research.
[LG-7] Dual-guided Hierarchical Edge Localization for Large-scale Optimal Transport Across Dimensions
链接: https://arxiv.org/abs/2609.13010
作者: Wenzhou Xia,Qiaoqiao Ding,Jingwei Liang,Xiaoqun Zhang
类目: Machine Learning (cs.LG)
*备注:
Abstract:Optimal transport (OT) compares distributions and aligns datasets in machine learning, yet unregularized discrete OT requires a linear program with quadratically many transport variables. We propose HELLO, a hierarchical solver that casts large-scale discrete OT as edge localization and uses dual potentials to guide both coarse-to-fine initialization and within-level refinement. Initialization propagates coarse dual potentials across a recursive subsampling hierarchy to assign candidate edges. Refinement then iteratively inserts the largest dual violators in each row and column until the relative KKT residual meets a prescribed tolerance, while budgeted pruning ensures linear memory complexity. For exact-arithmetic refinement, we prove finite termination at a global optimum under a symbolic lexicographic rule. At the million-point scale, HELLO attains lower transport objectives with order-of-magnitude runtime improvements over strong baselines across feature dimensions from single digits to thousands. It further scales to 1.28 million samples per marginal in 8192 dimensions on a single H100, using 41.6 GiB peak GPU memory while satisfying a full relative KKT residual below 10^-6 . Beyond standard discrete OT, the framework supports general pairwise costs and serves as a scalable balanced-OT oracle for semi-discrete OT, Gromov–Wasserstein, unbalanced OT, and OT-based Flow Matching.
[LG-8] A Full Adam Theorem for Spectral Heavy-Tail Onset
链接: https://arxiv.org/abs/2609.12996
作者: Zongmin Liu
类目: Machine Learning (cs.LG); Probability (math.PR); Machine Learning (stat.ML)
*备注:
Abstract:We prove a full Adam theorem for spectral heavy-tail onset in a closed Gaussian Stein-Hermite teacher-student state-evolution model. The theorem begins with the actual full-batch Adam recurrences, derives the population gradient by Stein-Hermite calculus, proves finite-width covariance concentration, converts multi-step Adam momentum into an exact non-centered Gaussian sign kernel, controls the diagonal Adam denominator by a basis-homogenization theorem, derives a regularly varying projected update response from a Hermite edge-transfer theorem, pushes the response through the exact Gram update, and proves approximate-target KL contraction with matching upper and lower hitting bounds. The final law is (\tau_\varepsilon=\Theta(\Delta_1^-\gammad^\rho\log(\Psi_0/\varepsilon))), where (\Delta_1) is the first spike-bulk spectral gap. The result is full in the following precise sense: every step from Adam’s momentum and denominator to the spectral hitting law is formalized inside the closed state-evolution model. We also prove that a stronger arbitrary-gradient Adam theorem is impossible, and that exact two-step linear-network loss dynamics do not identify factor spectra or heavy-tail hitting times.
[LG-9] Dimension-Corrected Hitting Times for Heavy-Tailed Spectral Emergence in Neural Optimizer Dynamics
链接: https://arxiv.org/abs/2609.12994
作者: Zongmin Liu
类目: Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE); Machine Learning (stat.ML)
*备注:
Abstract:Heavy-tailed empirical spectral densities of neural-network weight matrices are widely used as diagnostics of implicit self-regularization, but the step complexity of heavy-tail emergence remains poorly understood. We formulate spectral heavy-tail formation as a right-censored hitting-time problem: a run that does not reach a heavy-tail diagnostic within the observation horizon is treated as censored rather than discarded. In controlled full-batch teacher–student dynamics, we find that the first-step spike–bulk gap alone does not explain onset time. Instead, finite-onset regression supports a dimension-corrected spectral-gap law, (\tau_\mathrmHT\approx C\Delta_1^-\gammad^\rho), with (R^2=0.683), (\gamma=0.626), and (\rho=0.772) across 330 completed runs. Right-censored lognormal accelerated-failure-time models further favor the dimension-corrected model over a gap-only model, improving AIC from 706.62 to 628.70. Theoretically, we prove that exact early loss dynamics in linear networks do not determine factor spectral tails, that Adam recurrences alone do not imply spectral redistribution, and that projected singular-basis spreading implies contraction of a spectral-tail potential and hence a dimension-corrected hitting-time bound. Empirically, projected-kernel profiles support the sufficient spreading mechanism, Adam and AdamW agree under tested grids, GD and signGD do not reach onset in the same regimes, and real pretrained Qwen2.5-0.5B and Pythia-70M transformer weights show non-Gaussian spectral-tail structure relative to matched Gaussian nulls. The result is a reproducible spectral hitting-time law with rigorous conditional theory, not a claim that Adam necessarily generates heavy tails from first principles.
[LG-10] Information-Induced Training Geometry: Exact Reduction Canonical Completion and Structured Expressivity
链接: https://arxiv.org/abs/2609.12991
作者: Zavier Li
类目: Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注: 27 pages. Complete proofs are included
Abstract:Training data constrains optimizer geometry through the covectors visible to a declared information channel. We study how such partial information determines a full positive cometric relative to a reference and which degrees of freedom remain unidentified. Our central result resolves full-column-rank positive-definite compression under affine-invariant Riemannian geometry. The compression map is a split-Hadamard metric submetry and admits an explicit unique completion that is the affine-invariant nearest full geometry realizing a visible target and yields exact full-to-visible variational reduction. When the channel moves, the completions form a gauge-invariant rank stratification of the positive-definite cone. Its closed-form pullback pair metric separates visible-metric motion from subspace rotation through a reference-mismatch weight, yields an explicit positive-semidefinite multi-direction Gram matrix, and exposes the precise singularity of reference-valued modes. The mechanism is explained by a metric theorem equating ball submetry, attained fiber distance, and lossless reduction of every monotone radial visible decision problem. A smooth split-Hadamard theorem supplies coherent information sheets, proximal commutation, and solution-wise gradient-flow lifting. The positive-definite realization also gives closed-form prior-data shrinkage. Diagonal and block optimizer families reduce to relative-interior conic image tests with valid facial certificates, while deterministic and finite-sample bounds quantify recovery of the visible geometry and its subspace. Together these results characterize exact reduction, reference-dependent completion, and structured expressivity for the stated finite-dimensional affine-invariant model.
[LG-11] A Large-Scale AIS Dataset from Finnish Water
链接: https://arxiv.org/abs/2609.12938
作者: Debayan Bhattacharya,Ikram Ul Haq,Carlos Pichardo Vicencio,Sebastien Lafond
类目: Machine Learning (cs.LG)
*备注:
Abstract:This research paper contributes to the maritime research community by introducing a comprehensive AIS dataset from Finnish waters, specifically the Baltic Sea region. AIS data, initially designed for collision prevention, have evolved into a versatile tool with applications across diverse maritime domains. Our paper not only curates and categorises existing AIS datasets but also introduces a collected AIS dataset from the Baltic Sea area, renowned for its intercontinental cargo routes, military activities, and frozen water expanses. This dataset includes 229 millions data points and provides researchers with a resource for studying maritime activities and vessel behaviour in this dynamic region, notably distinguished by the inclusion of data from Finnish lakes. Our analysis includes a detailed listing of ship types and relevant features, empowering researchers to explore various maritime domains. To enhance comprehension and analysis, we provide visualisations of maritime traffic patterns. By publishing this AIS dataset, we aim to catalyse innovation and collaboration in maritime research, offering a gateway to deeper insights into maritime activities in the Baltic Sea and Finnish lakes.
[LG-12] Dissecting GPU Utilization for LLM Inference on Nvidia Hopper
链接: https://arxiv.org/abs/2609.12923
作者: Mohammad Siavashi,Gerald Q. Maguire Jr.,Dejan Kostic,Marco Chiesa
类目: Performance (cs.PF); Hardware Architecture (cs.AR); Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG)
*备注:
Abstract:A single SM utilization percentage can make an LLM inference workload look compute-saturated while hiding how much useful work is being done. The problem is not that the counter is wrong, but that it collapses several different mechanisms into one number. This is most severe during decode, where each request contributes only one new token and dense projection GEMMs become small-row matrix multiplications. On Hopper, the bfloat16 GMMA path executes these operations in fixed 64-row matrix fragments, so small-batch decode can fill only a small fraction of each fragment with real token rows. In this paper, we profile vLLM with FlashAttention-3 and cuBLASLt on an H100 NVL across cold prefill, warm prefill, and decode, sweeping sequence length and batch size. We replace the usual single utilization number with eight counter-validated views derived from raw Nsight Compute reports, each pinned to an NCU counter or explicit formula. Together, these views map utilization gaps to concrete mechanisms - fragment fill, occupancy limits, stall signatures, wave quantization, and kernel selection - across four production models and six per-layer kernel roles. Subjects: Performance (cs.PF); Hardware Architecture (cs.AR); Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG) Cite as: arXiv:2609.12923 [cs.PF] (or arXiv:2609.12923v1 [cs.PF] for this version) https://doi.org/10.48550/arXiv.2609.12923 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-13] Offline Reinforcement Learning for Wind Farm Control: A Wind Tunnel Study under Dynamic Wind Directions
链接: https://arxiv.org/abs/2609.12905
作者: Yuhan Su,Hongyang Dong,Simone Tamaro,Filippo Campagnolo,Carlo L. Bottasso,Xiaowei Zhao
类目: Machine Learning (cs.LG)
*备注:
Abstract:This paper addresses the wind farm power maximization problem in the presence of wind direction changes. Specifically, a model-free Modified Twin Delayed Deep Deterministic Policy Gradient with Behavior Cloning (MTD3-BC) algorithm is proposed to tackle this task through yaw control under varying wind direction conditions. MTD3-BC is an offline reinforcement learning (RL) algorithm that aims to infer good behavior from only a precollected offline dataset. Additionally, to ensure smooth and moderate yaw adjustments, a new action consistency term is introduced into the policy optimization objective. Unlike online RL methods, MTD3-BC does not require extensive interactions with a wind farm simulator during training, significantly reducing computational costs and training time. A wind tunnel experiment is conducted to validate the effectiveness of the algorithm under varying wind directions. The results demonstrate that MTD3-BC successfully mitigates wake effects, delivering farm-level power gains of approximately 10% over the baseline greedy strategy and performance on par with a data-calibrated model-based wake-steering benchmark, while requiring no wake model and only a small fraction of the training cost of online RL. To our knowledge, this is the first time an offline RL wind farm control policy has been validated and demonstrated experimentally.
[LG-14] Hidden in Rounds: Predicting the Time Cost of 802.11 Contention in Federated Learning
链接: https://arxiv.org/abs/2609.12903
作者: Satwat Bashir,Tasos Dagiuklas
类目: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC)
*备注:
Abstract:Federated learning over IEEE~802.11 shares the wireless channel among clients that send model updates. We use ns-3 to measure the frame-delivery ratio and saturation throughput for different client densities and offered loads. A separate FedAvg trainer uses the frame-delivery ratio as a first-order proxy for the update-admission probability and uses an equation to estimate communication time. The method does not simulate the delivery of a complete model update or measure end-to-end training time. Across 720 evaluated runs with two datasets, two data partitions, six client densities, six offered loads, and five seeds, all runs reached their predefined target accuracy within the round budget. Rounds-to-target changed little with offered load, while communication time-to-target increased by about two orders of magnitude across the client-density range. A Bianchi-anchored estimator produced a mean absolute percentage error from 2.3% to 10.2% on held-out configurations. This error is measured against communication time constructed from the same round-duration equation, not against independently measured completion time. We also compare uniform participation with persistent heterogeneous participation. The study does not detect a statistically distinguishable excluded-class accuracy gap over five seeds, but the confidence intervals are wide. The results apply only to the evaluated configurations and do not provide a general convergence or fairness guarantee.
[LG-15] Physical-State-Guided Diffusion Sampling for Full-Waveform Inversion
链接: https://arxiv.org/abs/2609.12899
作者: Chen Min,Haowen Jiang,Zheng Ma,Xiongbin Yan
类目: Machine Learning (cs.LG)
*备注: 43 pages, 13 figures
Abstract:Full waveform inversion (FWI) estimates subsurface velocity from seismic recordings, but its ill-posedness and nonlinearity make accurate reconstruction strongly dependent on initialization and prior information. Diffusion posterior sampling provides a learned geological prior, yet directly coupling its denoiser to the nonlinear wave solver can yield unreliable physical guidance. We propose Physical-State-Guided Diffusion Sampling (PSG), which couples a persistent physical velocity to the diffusion prior through a Gaussian bridge. The physical state is refined by waveform fitting regularized by the denoised velocity, and in turn guides the reverse diffusion process. This formulation separates the wave-equation and denoiser gradients while preserving conventional FWI initialization and accumulated optimization history. On four OpenFWI families, PSG’s terminal denoised estimates outperform classical and diffusion-based baselines under clean and missing-trace acquisitions and maintain strong structural recovery under measurement noise. Repeated stochastic runs preserve the dominant geological structures, with ensemble variability concentrated near geological interfaces and positively associated with local inversion error. A frozen OpenFWI-trained prior further supports inversion of the larger Marmousi, Overthrust, and BP2004 Salt models, recovering complex geological structures without retraining.
[LG-16] Quantifying the Value of Privileged Information Using a PAC-Bayesian Approach AAAI2027
链接: https://arxiv.org/abs/2609.12891
作者: Vasily Bokov(1 and 2 and 3),Sebastian Schmitt(3),Vedran Dunjko(1 and 2),Hao Wang(1 and 2) ((1) aQa, Leiden University, The Netherlands (2) LIACS, Leiden University, Leiden, The Netherlands (3) Honda Research Institute Europe GmbH, Offenbach, Germany)
类目: Machine Learning (cs.LG)
*备注: 24 pages, 4 figures. Submitted to AAAI 2027
Abstract:In practice, various learning scenarios provide access to auxiliary features exclusively during training. Incorporating such data to enhance model performance gave rise to a paradigm known as Learning Using Privileged Information (LUPI). While this extra information is intended to improve the resulting model, establishing a generalized, cohesive understanding of how privileged information (PI) transfers useful knowledge remains a challenge. Vapnik’s original theory and subsequent works offer performance guarantees in certain cases, but these results are inherently per-algorithm and rely on setting-specific proof approaches. Consequently, a more general framework explaining how and when PI transfers useful knowledge is still missing. To bridge this gap, we introduce an algorithm-agnostic, information-theoretic approach based on the PAC-Bayes framework. Rather than asking whether a particular algorithm exploits PI, we ask how much value it could offer: comparing the tightest achievable risk bound with and without PI yields its potential - an upper limit on the extractable gain. We introduce a metric that quantifies this potential directly from empirical training risk, bypassing the need for test-time data access, and validate our findings in both supervised and unsupervised settings. The results demonstrate a robust correspondence between our training-time metric and true test-time performance gains. Ultimately, this work takes a necessary step toward an information-theoretic understanding of LUPI, and quantifying the potential of privileged features before committing to a model.
[LG-17] What an odour descriptor corpus can and cannot measure: valence attenuation and the ceiling of the public record
链接: https://arxiv.org/abs/2609.12875
作者: Stylianos Kampakis,Fabio Rovai
类目: Machine Learning (cs.LG)
*备注:
Abstract:Machine olfaction trains on pooled public descriptor corpora, but whether a shared descriptor word measures the same thing across corpora has not been tested, nor has the ceiling of what any of them can measure. We audit four corpora from Pyrfume. Conditioning on the molecule makes McNemar’s test the exact conditional test of the corpus effect. Corpora disagree heterogeneously across descriptors ( I^2 = 80% ) and non-uniformly with labelling breadth ( z = 17.2 ), so no single offset repairs pooling. Median tetrachoric agreement is 0.795 against median \kappa of 0.413: sources largely concur on which molecules deserve a word and differ on how readily they apply it. Of 109 descriptors with an estimable effect, 36 show large differential functioning on the ETS scale. Against a human panel’s reliability, Morgan fingerprints with the full RDKit descriptor block reach 32.9% of achievable; adding every label from two merged corpora reaches 33.9%. The gap does not close with model capacity, encoding choice, more molecules, or more words. The missing variance is valence. One pleasantness rating per molecule reaches 54.6% of achievable (57.1% on an independent older instrument). Valence recovered from descriptors ( \rho = 0.457 ) yields only 16.6%, so it must be measured. Five raters exceed structure plus the full descriptor record; fifteen to twenty saturate. We release a descriptor crosswalk and twenty machine-checked theorems. Subjects: Machine Learning (cs.LG) Cite as: arXiv:2609.12875 [cs.LG] (or arXiv:2609.12875v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.12875 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Stylianos Kampakis [view email] [v1] Fri, 11 Sep 2026 14:03:25 UTC (58 KB)
[LG-18] GenOR-Twin: A Semantic Middleware for Integrating Operational Discourse with Mathematical Optimization
链接: https://arxiv.org/abs/2609.12863
作者: Rahimeh Neamatian Monemi,Shahin Gelareh,Lubin Cui,Nelson Maculan
类目: Machine Learning (cs.LG)
*备注: 33 pages and 8 figures
Abstract:We introduce GenOR-Twin, a neuro-symbolic framework that bridges the translation gap between unstructured operational logs and rigorous mathematical optimization. Our architecture uniquely positions Large Language Models as semantic translators rather than direct solvers, ensuring that the system retains the feasibility guarantees of exact combinatorial methods. \colorredWe design a dynamic constraint injection mechanism (the runtime translation of qualitative disruption events into formal mathematical constraints) that allows the system to structurally modify the optimization problem’s feasibility region in real-time based on qualitative human inputs. The resulting bidirectional coupling—where operational observations update the virtual model state and optimized decisions are reflected back into the Knowledge Graph—satisfies the synchronization requirement of a proper Digital Twin. The framework features an adaptive decision policy that automatically selects between low-complexity schedule repair and full re-optimization by analyzing the available system slack. Finally, we demonstrate the generalization of this approach across six distinct optimization domains, \colorredturning static models into resilient systems that adapt to the operational uncertainty and variability of real-world environments.
[LG-19] Very Exciting: Zero-Shot Model Predictive Control of Buildings via Excitation-Based Generalized Transfer Learning Models
链接: https://arxiv.org/abs/2609.12853
作者: Fabian Raisch,Felix Koch,Zack Xuereb Conti,Christoph Goebel,Benjamin Tischler
类目: ystems and Control (eess.SY); Machine Learning (cs.LG)
*备注: currently under review
Abstract:The widespread adoption of data-driven, energy-efficient model predictive control (MPC) in buildings remains hindered by substantial effort to collect data and train models for individual buildings. Transfer learning (TL) has consequently gained increasing attention for target building modeling, as it reduces data requirements and modeling effort by reusing pretrained source models. However, these TL models are typically evaluated only on prediction accuracy in the target, without testing downstream control performance. To address this gap, we apply a state-of-the-art TL approach - pretraining a generalized model on multiple source buildings using standard operational data - within an MPC setup in a target building. We show that this approach is insufficient to achieve satisfactory control performance. As a solution, we introduce generalized models pretrained on excitation-based operational source data - purposefully probed inputs that explore the building’s state-action space. For evaluation, we apply the generalized models via zero-shot (i.e., without fine-tuning) to 32 simulated target buildings and assess MPC performance. Our results show that excitation-based generalized models achieve the strongest control performance among all benchmarks, outperforming an online linear model-based MPC and a PI controller by 6.4% and 36.9%, respectively. By combining strong control performance with the ability to generalize across multiple buildings, without requiring any target-specific data, our approach reduces MPC setup cost and simplifies its widespread deployment in the building sector.
[LG-20] VertexCBF: Improving Neural Control Barrier Functions via Vertex-Restricted Control Search
链接: https://arxiv.org/abs/2609.12831
作者: Bojan Derajić,Sebastian Bernhard,Wolfgang Hönig
类目: Robotics (cs.RO); Machine Learning (cs.LG); Systems and Control (eess.SY)
*备注:
Abstract:As the number of autonomous robots continues to grow, safety becomes increasingly important. Control barrier functions (CBFs) provide a theoretically grounded framework for ensuring safety, but existing design methods often face limitations in effectiveness, scalability, or interpretability, and may result in overly conservative safe sets. In this paper, we propose \emphVertexCBF, a framework for learning neural CBFs in a scalable, systematic, and explainable way. We approximate the stationary Hamilton–Jacobi value function using a neural network trained via a combination of physics-informed and sparsely supervised learning. By exploiting control-affine dynamics and a convex polytope control set, under which the Hamiltonian is maximized at the control vertices, we efficiently generate supervision points via GPU-parallel vertex-restricted tree search, while a residual architecture guarantees that the learned CBF is never larger than the specified constraint function. We evaluate the method on 15 systems and compare it against relevant baselines, showing that it reliably recovers large safe sets where the baselines are conservative or fail completely. In addition, we perform a hardware experiment in which a mobile robot safely avoids pedestrians using a neural CBF trained with our method.
[LG-21] VertiFuseX: Generalizable Financial Forecasting via Multi-Stream Temporal Fusion
链接: https://arxiv.org/abs/2609.12793
作者: Aashish Bohra,Vivek Vijay
类目: Machine Learning (cs.LG); Statistical Finance (q-fin.ST)
*备注: 42 pages, 14 figures, 13 tables. Manuscript under peer review
Abstract:Stock price prediction remains challenging due to the non-stationary and noisy nature of financial time series. Existing deep learning models often rely on rigid decision-level fusion, ad hoc hyperparameter tuning, and compressed final-layer outputs, causing information loss, overfitting, and limited cross-market generalization. We propose VertiFuseX, a hybrid LSTM architecture using penultimate-layer vertical fusion of multi-scale temporal representations. VertiFuseX stacks and reweights penultimate features from LSTM, Bi-LSTM, and St-LSTM branches, integrates a parallel DNN stream, and jointly optimizes all components via backpropagation under a fixed hyperparameter configuration. This preserves richer intermediate temporal information across scales. Evaluated on 15 years (2010-2024) of closing prices from 10 global equity indices using strict chronological out-of-sample testing with the final 365 trading days held out, VertiFuseX achieves 30-54% MAPE reductions and over 40% improvements in MAE and RMSE versus LSTM-based baselines, and outperforms seven state-of-the-art models across 33 metric-dataset comparisons. Ablation studies confirm penultimate-layer fusion drives these gains over final-layer fusion and decision-level ensembling. Gradient-based saliency analysis shows consistent emphasis on mid-range dependencies at lags 9-15 days. Economic validation via algorithmic trading simulation under extreme market regimes shows reduced maximum drawdowns and superior risk-adjusted returns. With 675k parameters, a 2.6 MB memory footprint, and 1.5 ms/sample inference latency, VertiFuseX offers a lightweight, interpretable, deployment-ready framework for robust financial forecasting.
[LG-22] Convergence of Stochastic Gradient Methods under Heavy-Tailed Noise and Hölder Smoothness
链接: https://arxiv.org/abs/2609.12785
作者: Misbah Uz Zaman,Anirbit Mukherjee
类目: Machine Learning (cs.LG); Optimization and Control (math.OC); Machine Learning (stat.ML)
*备注: 46 pages, 2 figures
Abstract:Classical convergence guarantees for stochastic gradient methods typically assume Lipschitz-smooth objectives and finite-variance gradient noise, both frequently violated in practice. In contrast, we study nonconvex stochastic optimization under the joint relaxation of these assumptions: objectives with (L,s) -Hölder continuous gradients, s\in(0,1] , and gradient noise satisfying only a bounded \alpha -th moment condition for \alpha\in(1,2] . We establish three convergence results. Firstly, that standard SGD converges at rate O(T^-s/(1+s)) whenever \alpha\ge1+s , extending the classical nonconvex SGD rate to heavy-tailed noise and Hölder smoothness simultaneously. Secondly, we analyze \delta -regularized gradient clipping ( \delta -GClip), a provable trainer of wide and deep nets, and establish a stationarity rate of O(T^-2s(\alpha-1)/[(1+s)(2\alpha-1)]) under the same condition. Thirdly, we analyze standard gradient clipping (G-Clip) and show that it recovers the above rate for \alpha\ge1+s while in the very heavy-tailed regime \alpha1+s , it has a convergence rate O(T^-2s(\alpha-1)/[(\alpha-1)+s(2\alpha-1)]) — the first convergence guarantee in this regime for any stochastic gradient based method.
[LG-23] Curriculum-Based Adversarial Heterogeneous Agent Reinforcement Learning for Autonomous Quad-Copter Landing in Maritime Settings
链接: https://arxiv.org/abs/2609.12758
作者: Allan Minh-Tam Nguyen,Sree Showrya Kotala,Stefan Banioi-Crijman,Kurt Driessens,Rico Möckel
类目: Machine Learning (cs.LG)
*备注: 20 pages, 5 figures, submitted to BNAIC 2026
Abstract:Recovering unmanned aerial vehicles (UAVs) in maritime environments is challenging due to wind turbulence and ship-deck motion, making it a valuable test case for alternative control and learning approaches as conventional landing approaches often become unreliable. We study simulated mid-air capture of quadrotor UAVs by a ship-mounted robotic arm, learning robust cooperative control policies with Heterogeneous-Agent Proximal Policy Optimization (HAPPO) Reinforcement Learning. We train with HAPPO using a curriculum and an adversarial wind agent (HARL-AC) in NVIDIA Isaac Lab, and compare the obtained control policies against those generated through curriculum-based domain randomization and a benchmark trained on a single sea state. In-distribution evaluation on sea states 0/4/5 shows comparable success for HARL-AC and domain randomization of up to 97.5% . On out-of-distribution sea states 7/8/10 , HARL-AC generalizes better, achieving up to 16% higher median success rate at sea state 10, and substantially lower crash rates of up to 14% compared to the domain randomization policy. Furthermore, we show that the adversarially trained policy shows more cautious behavior, slightly increasing timeouts by 3% , but yields safer recovery behavior in severe, unseen conditions.
[LG-24] Optimizing for the decision not the prediction: an exploration of Smooth Net Benefit as a training objective
链接: https://arxiv.org/abs/2609.12752
作者: Koen M.F. Gorgels,Lasai Barreñada,Maarten van Smeden,Ben Van Calster,Ewout W. Steyerberg,Wouter A.C. van Amsterdam
类目: Machine Learning (cs.LG)
*备注: 22 pages, 5 figures, 2 tables. Code and supplementary data available
Abstract:Objective Prediction models are commonly trained using objectives such as Bernoulli negative log-likelihood (NLL), although downstream clinical decisions may depend on specific risk thresholds. We introduce Smooth Net Benefit ( \sigma NB), a differentiable approximation of Net Benefit designed to align model training with threshold-specific clinical utility. Materials and Methods We evaluated \sigma NB as a training objective for logistic regression, generalized additive models (GAMs), and XGBoost with three Hessian implementations. Experiments used the Framingham cardiovascular risk dataset and 44 TabZilla datasets comprising 72 dataset-threshold combinations. Results \sigma NB training did not consistently improve Net Benefit in Framingham. Across the TabZilla benchmark, mean standardized Net Benefit for logistic regression increased from 0.5669 with NLL to 0.5765 with \sigma NB (mean difference 0.0096, 95% CI -0.0001 to 0.0193). For GAMs, mean standardized Net Benefit decreased from 0.5921 to 0.5625 (mean difference -0.0296, 95% CI -0.0721 to 0.0129). For XGBoost, NLL achieved 0.6745 compared with 0.6723–0.6735 across \sigma NB implementations. In logistic regression, \sigma NB gains were positively associated with the performance advantage of XGBoost over NLL-trained logistic regression. Discussion The effect of \sigma NB was context dependent, with modest gains concentrated in logistic regression and little benefit for more flexible model classes. This suggests that decision-focused optimization may be most useful when limited model flexibility leaves greater scope for improvement. Conclusion Our results do not support \sigma NB as a general replacement for NLL training, but support further investigation of decision-focused objectives in settings where conventional likelihood-based training may not adequately capture decision-relevant structure. Comments: 22 pages, 5 figures, 2 tables. Code and supplementary data available Subjects: Machine Learning (cs.LG) Cite as: arXiv:2609.12752 [cs.LG] (or arXiv:2609.12752v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.12752 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Koen Gorgels [view email] [v1] Fri, 11 Sep 2026 12:01:05 UTC (967 KB) Full-text links: Access Paper: View a PDF of the paper titled Optimizing for the decision not the prediction: an exploration of Smooth Net Benefit as a training objective, by Koen M.F. Gorgels and 5 other authorsView PDF view license Current browse context: cs.LG prev | next new | recent | 2026-09 Change to browse by: cs References Citations NASA ADSGoogle Scholar Semantic Scholar export BibTeX citation Loading… BibTeX formatted citation loading… Data provided by: Bookmark checked="checked"class=“labs-tab-input”> Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) IArxiv recommender toggle IArxiv Recommender (What is IArxiv?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv’s community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?) mathjaxToggle(); We gratefully acknowledge support from our major funders, member institutions, , and all contributors. About Help Contact Subscribe Copyright Privacy Accessibility Operational Status (opens in new tab) Major funding support from
[LG-25] Write on Paper and Get the Online Digital Trace:newline A New Era for Handwriting
链接: https://arxiv.org/abs/2609.12702
作者: Florent Imbert,Yann Soullard,Eric Anquetil,Tanja Harbaum,Alexey Serdyuk,Fabian Kress,Tim Hamann,Peter Kampf
类目: Machine Learning (cs.LG)
*备注:
Abstract:Capturing the digital trace of handwriting usually requires a specific stylus and a compatible substrate, be it a capacitive touchscreen, an ElectroMagnetic Resonance (EMR) tablet as used in Wacom systems or special paper. While writing on regular paper offers rich haptics, no latency and is well known for improving information retention, no low-cost and widely accepted, effective solution exists to digitize such a pen trace. The challenge is to accurately track the pen’s trajectory without an external reference system while allowing unrestricted freedom of pen movement across a surface. We propose an innovative solution that combines a digital pen, advanced artificial intelligence algorithms, and adaptive AI techniques to reconstruct the digital trace of handwriting. Our approach integrates hardware development, focusing on a sensor-equipped pen, with software innovations to optimize trajectory reconstruction and processing in real time using an embedded AI. This work aims to advance the state-of-the-art in automated trace reconstruction of handwriting, enabling a seamless connection between traditional handwriting on paper and capturing the trace digitally.
[LG-26] SIFPBPNet: A Dual-Path Network for Wearable and Cuffless Blood Pressure Estimation via Individualized Steady-state Representation
链接: https://arxiv.org/abs/2609.12690
作者: Shuailong Tang,Xiaoyu Li,Donglin Xie,Wei Chen,Guangpu Zhu,Yelei Li,Yali Zheng
类目: Machine Learning (cs.LG)
*备注: Accepted for publication at the 48th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC 2026), Toronto, Canada
Abstract:Continuous and cuffless blood pressure (BP) monitoring using photoplethysmography (PPG) is of great interest for low-cost and personalized cardiovascular health management. However, significant population heterogeneity and the “one-to-many mapping” problem, where similar waveforms across individuals correspond to different BP levels, limit the accuracy of conventional population-based models. To address this challenge, we propose a dual-path architecture termed SIFPBPNet, which separately represents steady-state and instantaneous features, through a Steady-state Feature Path (SFP) and an Instantaneous Feature Path (IFP). The SFP employs a Graph Attention Network (GAT) to extract individual-specific and long-term characteristics from multi-day historical PPG trajectories. In parallel, the IFP captures short-term dynamics from current PPG segments and incorporates the steady-state prior via a cross-attention mechanism. Experiments on a large-scale wearable dataset demonstrate that SIFPBPNet achieves a Mean Absolute Error (MAE) of 8.57 and 5.97 mmHg for systolic and diastolic BP, respectively, outperforming state-of-the-art models. Furthermore, the SFP module consistently improves performance when integrated into various backbone architectures, yielding 2.8-13.1% relative MAE reductions for systolic BP. These results highlight the strong generalizability and plug-and-play transferability of the SFP module, underscoring its great potential for accurate cuffless BP monitoring.
[LG-27] ProactiveBench: Can Streaming Video Models Really Interact Like Humans?
链接: https://arxiv.org/abs/2609.12658
作者: Kaixuan Du,Xin Wan,YuKun Wang,Hang Zhang,Meng Cao,Dai Guan,Ming Chen,Ni Li
类目: Machine Learning (cs.LG)
*备注: Code and data is available at this https URL
Abstract:Streaming video understanding requires models to process continuous multimodal input while maintaining temporal context. Existing evaluations are predominantly reactive: they query a model at a selected timestamp and therefore do not assess when it should respond. Proactive interaction instead requires monitoring a standing request, responding within an appropriate interval after the target event, and otherwise remaining silent. We introduce ProactiveBench, which evaluates models at one-second stream intervals without an explicit response cue. Its six subtasks vary trigger ambiguity and timing tolerance. Event Sensitivity geometrically combines response and silence rates on the same recording; four window-based subtasks distinguish early, in-window, and missed responses; and Duplicate Counting penalizes omissions and repetitions. Premature responses outnumber missed responses for four of the six evaluated systems, revealing a substantial gap in the temporal decision-making required for human-like interaction.
[LG-28] Distortion of AI Alignment Revisited: RLHF is a Decent Utilitarian Aligner
链接: https://arxiv.org/abs/2609.12651
作者: Kazusato Oko,Annie Ulichney,Nika Haghtalab,Han Bao
类目: Machine Learning (cs.LG); Computer Science and Game Theory (cs.GT)
*备注:
Abstract:While Reinforcement Learning from Human Feedback (RLHF) is the standard paradigm for aligning large language models with human preferences, its effectiveness in pluralistic settings has been called into question. Notably, recent work by Gölz et al. (2025) demonstrated that the \textitdistortion – defined as the multiplicative gap between the average user utility of the RLHF policy and the optimal average utility – can scale exponentially with the Bradley-Terry temperature parameter \beta when users have heterogeneous preferences. In this work, we present a fine-grained analysis of the distortion of RLHF with reward clipping and demonstrate that such exponential degradation is not a fundamental property of the algorithm but rather a consequence of distribution mismatch between the distribution generating preference data ( \mu ) and the KL reference policy ( \pi_\mathrmref ). To this end, we establish tight upper and lower bounds on the distortion of RLHF across multiple regimes of the KL regularization strength. We show that in a representative regime, under the Bradley-Terry model, the distortion is \tilde\Theta(\beta B + \beta) , where B is an upper bound on the log density ratio between \mu and \pi_\mathrmref . In particular, when there is no distribution mismatch (i.e., \mu = \pi_\mathrmref ), RLHF achieves the optimal distortion of O(\beta) up to a constant. Our results suggest that, to reasonably maximize average utility with RLHF, it is preferable to use on-policy sampled preference data or to fine-tune before RLHF on data from a source close to \mu .
[LG-29] Poisson-Corrector Complexity Bounds for Moreau–Yosida Unadjusted Langevin Sampling
链接: https://arxiv.org/abs/2609.12594
作者: Yuchen Xin,Zhihua Zhang
类目: Machine Learning (cs.LG)
*备注:
Abstract:We study the classical Moreau–Yosida unadjusted Langevin algorithm (MYULA) for \pi(,\mathrmd x)\propto e^-f(x)-g(x),\mathrmd x , where f\in C^2(\mathbbR^d) is m -strongly convex with L_f -Lipschitz gradient and g:\mathbbR^d\to\mathbbR is convex and globally G -Lipschitz. For the Moreau-smoothed target \pi_\lambda and the MYULA invariant law \widehat\pi_\lambda,h , we prove [ \sqrt m,W_2(\pi_\lambda,\widehat\pi_\lambda,h) =O(h)+\widetilde O(h^3/4) ] under 0h(L_f+\lambda^-1)\le c , with only logarithmic dependence on \lambda^-1 in the error coefficients. Combining this estimate with the Moreau approximation bias yields \widetilde O(\varepsilon^-4/3) iterations to achieve \sqrt m,W_2(\mu_N,\pi)\le\varepsilon , for fixed model parameters and initialization. The proof combines a discrete Poisson corrector with active-trace estimates and a shared-noise bound for the exact–Euler two-point curvature. Subjects: Machine Learning (cs.LG) Cite as: arXiv:2609.12594 [cs.LG] (or arXiv:2609.12594v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.12594 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-30] Where Decoder Cosine Similarity Fails for SAE Feature Flow Discovery
链接: https://arxiv.org/abs/2609.12591
作者: Hendrik Droste,Christian Medeiros Adriano,Kathrin Korte,Holger Giese
类目: Machine Learning (cs.LG)
*备注: 4 pages, 2 figures
Abstract:Foundation models are increasingly adapted through fine-tuning, model editing, and alignment procedures while retaining previously acquired capabilities. Understanding the internal computations that support these adaptations is therefore becoming increasingly important for continual model evolution. Sparse autoencoders (SAEs) provide interpretable feature dictionaries for residual-stream activations and sublayer outputs, but it remains unclear how state features and update features interact to produce downstream residual features. In this work, we focus on MLP updates as a first test case. We construct a transition atlas of triples s_k + u_j \rightarrow t_\ell , where a residual-state feature and an MLP-update feature jointly predict a target residual feature, and validate candidate triples by ablating the decoded update feature. In a 20M-token Pythia-160M L_7 \rightarrow L_8 run, we find 38,125 strong ablation-effect transitions, but 88.0% have both state-target and update-target decoder cosine similarity below 0.7. As a preliminary cross-model check, a run of 20M-token Gemma-3-4B L_21 \rightarrow L_22 causally validates only the top 30,000 ranked candidate triples by ablating the decoded update feature, and 53.6% of strong-effect triples have both state-target and update-target decoder cosine similarity below 0.7. The Gemma result is directionally consistent with Pythia, but weaker, since update-target cosine recovers many of the strongest Gemma effects and the run is not a full-atlas causal validation. Ultimately, our results suggest that feature flow atlases can serve as diagnostics of representation-update mechanisms and thereby inform tools for steering model updates. Future work will validate more complex patterns across layers, models, and SAE families.
[LG-31] Quality-Constrained Routing over a Fixed Pool of Quantized Mixture-of-Experts Instances
链接: https://arxiv.org/abs/2609.12550
作者: Zhenghong Huang,Hongfan Wu,Jiheng Zhang
类目: Machine Learning (cs.LG)
*备注: 25 pages, 5 tables, and 1 figure
Abstract:Quantized Mixture-of-Experts (MoE) services can hold several pre-materialized instances of one base model, but quantization damage varies sharply across requests and bitwidths. Because instance materialization and replica counts consume memory and require slow reconfiguration, we treat them as upstream provisioning decisions and study routing within a fixed resident pool. Within this fixed-pool boundary, we route each request to maximize modeled throughput under a class-level expected quality-degradation budget and measured instance capacities. To predict this request-specific risk, we introduce FWP (Fragility-Weighted Perplexity), computed from prompt tokens on a reference-instance prefill and calibrated to candidate-instance degradation. Underlying FWP is an exact two-expert affinity–fragility decomposition and a conditional multi-layer top- k expansion whose bias, interaction, route-change, separability, and higher-order terms remain explicit. Using these calibrated risks, a window-level linear program yields a signed reduced-reward score that is KKT-consistent with the LP optimum under optimal prices and primal-feasible tie allocation. On 88 extended Qwen prompts, complete W2, W3, and W4 instances quantizing all 6,144 expert blocks incur mean \Delta NLL of 0.9437 , 0.1832 , and 0.0513 . Under the same population and \tau=0.1513 , FWP allocation reaches a 1.284\times offline model-based multiplier versus 1.253\times for request-agnostic mixing and 1.000\times for static W4, an incremental 2.5% relative FWP gain.
[LG-32] textGSF-χ: Global Stereochemical Fields for Chiral Graph Transformers
链接: https://arxiv.org/abs/2609.12532
作者: Jiaqing Xie,Yuxin Wang,Xipeng Qiu
类目: Machine Learning (cs.LG)
*备注:
Abstract:Enantiomers share atoms, bonds, and pairwise distances yet can behave differently in chiral environments, so molecular encoders must respect atom relabelings and proper rotations without becoming blind to reflection. We introduce GSF- \chi , a graph transformer in which stereogenic units modulate all pairwise interactions rather than single out one atom as special. Each central or axial stereogenic unit creates a reflection-even phase field over all atoms, a handedness pseudoscalar \chi sets the direction of a relative rotation on latent query–key blocks, giving a \textbfChiral-RoPE that reflection inverts rather than leaves fixed. A C_2 projection separates mirror-even ECD peak counts and positions from mirror-odd peak signs. We prove the operator’s even–odd decomposition and its annotation-inversion, permutation, and unit-order identities under explicit canonical-role conditions; property tests and a coordinate-reflection audit verify the laws end to end. GSF- \chi leads every central-ECD output and improves axial Rotation and Symbol by 12.6% and 7.9% over the strongest baseline. Equal-budget controls attribute the Rotation advantage to global signed support rather than parameter or edge count; the C_2 projection yields exact enantiomer-pair consistency at a small raw-accuracy cost under complete supervision and becomes predictive when mirror supervision is scarce.
[LG-33] mporal Recurrence Favors Fewer Layers
链接: https://arxiv.org/abs/2609.12531
作者: Ivan Anokhin,Johan Obando-Ceron,Irina Rish,Sebastian Risi
类目: Machine Learning (cs.LG)
*备注:
Abstract:In streaming tasks, recurrent models can carry latent computation across time, allowing each update to build on representations produced earlier. This raises a basic question: once temporal recurrence provides sequential computation across steps, how much depth is still needed within each step? Prior work has shown that recurrence can make shallow models competitive. We instead study this question as a compute-allocation problem, varying within-step depth, expert width, and the number of parallel experts per layer across several compute budgets. For each budget, we compare the best observed recurrent and non-recurrent allocations and the performance they achieve under approximately matched per-step computation. Across Sokoban and autoregressive FineWeb language modeling, we find that temporal recurrence shifts the best observed compute allocation toward substantially fewer layers, with comparable or better performance.
[LG-34] A Differentially Private Federated Proximal Optimization Framework for Customer Churn Prediction in Heterogeneous Federated Telecom Networks
链接: https://arxiv.org/abs/2609.12470
作者: Joydeb Kumar Sana,Subrata Chakraborty,M M Manjurul Islam
类目: Machine Learning (cs.LG)
*备注:
Abstract:Customer churn is one of the major issues in the telecommunication industry. To predict customer churn, conventional centralized machine learning approaches have been widely used. This centralized approach requires customer data to be stored in a central repository, which raises privacy concerns and may violate data protection regulations. Federated learning addresses this problem by allowing multiple telecom operators to collaboratively train a global model without transferring their raw customer data. However, real-world customer data are often heterogeneous (non-IID), which may negatively affect the performance of standard federated learning. Trained models can also suffer from privacy attacks. To address those issues, we propose a Differentially Private (DP) based Federated Proximal optimization (FedProx) framework. All experiments were performed on two publicly available telecom churn datasets. We trained Federated Averaging (FedAvg), DP-FedAvg, FedProx, and the proposed DP-FedProx framework. For baseline comparison, we also used several centralized and local models. To evaluate the models, we employed seven widely used evaluation metrics. The experimental results show that the FedProx based models consistently outperform the FedAvg based models. Compared with the best centralized model, the proposed DP-FedProx framework achieves competitive prediction performance with only a small reduction in accuracy while providing privacy guarantees. To explain our model, we conducted SHAP analysis which shows that DP-FedProx method priorities revenue group features. These results indicate that the proposed DP-FedProx framework provides a practical balance between prediction performance and data privacy protection.
[LG-35] SAGE-Loop: Reliable Closed-Loop LLM -Driven AutoML with Trial-and-Correction and Adaptive Ensembling
链接: https://arxiv.org/abs/2609.12455
作者: Junquan Gu,Shibo Cui,Xiangfeng Luo,Hang Yu
类目: Machine Learning (cs.LG)
*备注: 13 pages, 7 figures, 4 tables
Abstract:Automated machine learning (AutoML) is reshaping data-driven science and industrial practice, and as large language models are introduced into AutoML, pipeline reliability becomes as important as automation efficiency. However, existing AutoML still struggles to realize instant feedback and adaptive optimization during execution, so once a run drifts into a suboptimal or failed state, it lacks a process-level correction mechanism. The fundamental pathology lies in its one-way pipeline: intermediate failures are typically terminated or bypassed, while fixed paradigms often strengthen model generation but leave ensemble decisions static, weakening both execution reliability and the controlled use of structural diversity. This indicates that LLM-driven AutoML needs a closed-loop ability for trial-correction-improvement together with evidence-based use of model diversity. To this end, we propose SAGE-Loop, a reliable closed-loop, self-adaptive, LLM-driven AutoML framework that performs multi-round generation and validation for trial-and-repair, and adaptively selects ensemble strategies in both supervised and unsupervised tasks, thereby unifying how to generate with how to use models. Across 20 public datasets, SAGE-Loop consistently improves performance and stability on classification, regression, and clustering tasks. Additional results further show its ability to recover from execution failures and maintain robust pipeline behavior.
[LG-36] Granularity-Adaptive Credit Assignment for Long-Horizon LLM Agent Reinforcement Learning
链接: https://arxiv.org/abs/2609.12424
作者: Taoran Liang,Yang Liu,Shang Luo,Yingguang Yang,Rongrong Zhang,Yingzong Min,Yulin Huang,Jianshen Zhang,Yongzhi Qi,Kefu Xu,Congjing Ran,Bin Chong
类目: Machine Learning (cs.LG)
*备注: Preprint
Abstract:Reinforcement learning is now the standard way to train large language model agents on long-horizon tasks, where dozens of interdependent actions precede a single sparse reward. Critic-free, group-relative methods such as GRPO suit this regime, but they broadcast one trajectory-level scalar to every step and cannot say which decision drove the outcome. GiGPO recovers a step-level signal by grouping time steps that share an anchor state, yet it merges the step- and episode-level estimates under one fixed weight, spending the same resolution on a pivotal branching decision as on a routine, near-deterministic transition. We argue that the right resolution is state-dependent, and propose GACA, a critic-free estimator whose granularity follows an uncertainty-based criticality proxy. GACA scores every step by the negative log-likelihood its own rollout already records, then blends the two advantages with a per-step weight that grows with that score, so the gradient places more weight on the fine-grained signal at above-average NLL and on the episode-level signal below it. We derive an exact risk decomposition for the implemented mixture and show that sufficiently small modulation improves on fixed mixing under positive directional alignment. A separate conditional result bounds local action-value variation using expected NLL, while an error-projection analysis characterizes when mixing adds value beyond scalar uncertainty reweighting. On ALFWorld and WebShop, GACA improves task success over GRPO and GiGPO at both 1.5B and 7B scales.
[LG-37] HoliBench: A Cross-Platform Benchmarking and Deployment Toolkit for Foundation Models in CPS-IoT Applications
链接: https://arxiv.org/abs/2609.12412
作者: Inesh Chakrabarti,Zejun Xiong,Pragya Sharma,Mani Srivastava
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG); Performance (cs.PF)
*备注: 16 pages, 9 figures, 2 tables. Code: this https URL
Abstract:Foundation models, including large language models, vision-language models, and time-series foundation models, are increasingly deployed on embedded and edge platforms for CPS and IoT applications, where energy, latency, and memory are as critical as task accuracy. Existing benchmarking tools evaluate model capability in isolation, reporting accuracy assuming sufficient compute, while hardware profiling tools remain platform-specific and mutually incompatible. As a result, users lack a unified workflow for making deployment decisions across heterogeneous devices. We present HoliBench, a modular benchmarking and deployment toolkit that jointly characterizes accuracy, latency, and energy across platforms from single-board computers to GPU servers. Its platform abstraction layer calibrates cross-device measurement, and the toolkit supports multiple model modalities, inference engines, concurrencies, and existing evaluation harnesses. An interactive interface exposes constraint-aware configuration selection over a design space that is profiled once and reused across studies. Using HoliBench, we characterize 20 models across 7 device types, 3 quantization levels, 8 inference backends, and over 30 tasks, surfacing tradeoffs that existing tools miss: quantization reduces latency only on hardware with low-precision support, accuracy gains show diminishing returns relative to energy, and for autoregressive workloads, average inference power is approximately constant across output lengths. We further find that single-model profiles compose under sequential co-resident execution. In a multi-model CPS deployment, standalone profiles predict combined-pipeline latency and power within 1.2% and 2.5%, enabling deployment exploration without exhaustively profiling every pipeline configuration. We release HoliBench as open-source infrastructure for deployment-aware evaluation of foundation models.
[LG-38] Split Conformal Prediction with Label-Shift-Adjusted Bayesian Scores ICML2026
链接: https://arxiv.org/abs/2609.12386
作者: Hyeonsu Lee,Juyeon Kim,Erkhembayar Jadamba,Seungjin Choi,Hyunjin Shin
类目: Machine Learning (cs.LG)
*备注: 2nd Workshop on Epistemic Intelligence in Machine Learning (EIML@ICML 2026),
Abstract:Conformal prediction provides distribution-free uncertainty quantification under exchangeability. However, this assumption is violated by label shift, where the marginal distribution of labels changes while the conditional distribution of inputs given labels remains stable. Under such shifts, standard conformal procedures no longer maintain their intended coverage behavior. Existing approaches address this via importance weighting. They pair the reweighting with residual-based nonconformity scores that ignore predictive uncertainty. The resulting intervals have uniform width. Bayesian conformal methods produce adaptive intervals by leveraging predictive distributions. They evaluate conformity under the source predictive, which is misaligned with the target domain under label shift. We propose the \emphLabel-Shift-Adjusted Bayesian Score (LSA score), a nonconformity score derived from a posterior predictive tilting identity. This identity shows that the target predictive is an importance-weighted transformation of the source predictive. We use it to derive a direct correction to the Bayesian score. We evaluate the method on molecular property prediction under controlled label shift. The LSA score consistently yields shorter intervals than residual-based and source-based Bayesian scores. Coverage in the target domain remains comparable. Under stronger shift, all methods incur some coverage loss due to pseudo-label-based density-ratio estimation. The LSA score is defined for any source predictive with a tractable log-density. We instantiate it with Bayesian Ridge Regression, where the correction admits a closed form.
[LG-39] Certified AI Triage of ICU Alarms
链接: https://arxiv.org/abs/2609.12365
作者: Mohammed Sameer Syed,Rozhin Yasaei
类目: Machine Learning (cs.LG)
*备注:
Abstract:In the VTaC benchmark 71% of ventricular-tachycardia alarms are false, but silencing a real one can delay recognition of a dangerous arrhythmia. We reframe alarm reduction as three-way triage (retain, suppress, or defer) and bound the decision this analysis treats as harmful: among suppressed alarms, the fraction that were genuine stays below a user-set budget with 95% confidence, under i.i.d. event sampling. Alarms sharing a waveform record are dependent, so the clustered analysis is a sensitivity check. On the official split a 5% budget certifies in all three seeds, suppressing 74.8% of false alarms while silencing 1.5% of genuine ones, at AUROC 0.953 and Challenge Score 83.33, numerically comparable to the strongest of the eleven published systems. Our central finding measures what multiplicity costs: the correction charges for every candidate, so a finer grid can certify strictly less. Under held-out calibration the 885-cell grid we declared certifies 1 of 15 fold-runs, while choosing the grid on a separate selection partition certifies 8. We project the calibration volume each budget needs, making an uncertifiable budget a design parameter. Finally, adding a learned reliability dimension to the policy grid did not sharpen the certified frontier.
[LG-40] When Connected Does Not Mean Similar: Charting the Homophily Boundary of SNAP-KG for Streaming Entity Integration ISWC2026
链接: https://arxiv.org/abs/2609.12356
作者: Jui-Chien Lin,Oshani Seneviratne
类目: Machine Learning (cs.LG)
*备注: Accepted to the ISWC 2026 Companion Volume
Abstract:SNAP-KG is a framework for assigning newly arriving entities to semantic communities in a growing knowledge graph (KG) using only their raw features, with no graph access and no retraining at inference time. It was evaluated on five multi-view benchmarks and a 2.4M-node OGB-WikiKG2 KG. In each of these datasets, at least one graph view is homophilous, meaning that connected nodes usually belong to the same class, and SNAP-KG performs well on all of them. This paper asks what happens outside that setting. We extend the evaluation to three heterophilous graphs (Texas, Wisconsin, Chameleon) and measure the edge homophily of every view. When no homophilous view is available, clustering quality drops sharply for both SNAP-KG and the transductive baselines used in its original evaluation. What decides this is the homophily of the relation, not the number of relations. Multi-view fusion still helps, but only when at least one homophilous relation provides a reliable foundation. The homophily assumption is therefore shared by the whole method family, not specific to SNAP-KG. We argue that heterophilous multi-view clustering is a separate research problem, outside the scope of this work. As future work, we outline how a heterophily-aware teacher could be distilled into SNAP-KG’s projector to serve both homophilous and heterophilous KGs.
[LG-41] ParaRecover: A Process-Level Benchmark for Error Localization and Recovery in Parallel Tool-Use Agents EMNLP2026
链接: https://arxiv.org/abs/2609.12345
作者: Bowen Guan,Zhentao Yin,Yanming Shen
类目: Machine Learning (cs.LG); Software Engineering (cs.SE)
*备注: Accepted to EMNLP 2026 Main Conference
Abstract:Existing agent benchmarks mainly evaluate final task success or tool-call correctness, providing limited insight into whether agents can reliably diagnose and recover from intermediate execution failures. This limitation becomes particularly critical in multi-turn parallel tool-use scenarios, where errors may propagate across dependent branches and trigger cascading failures. We introduce ParaRecover, a process-level benchmark for evaluating error localization and recovery in multi-turn parallel tool-use agents. Built upon a fine-grained taxonomy of 14 error types covering planning dependencies, tool selection, and argument matching, the benchmark comprises 10,626 instances spanning two difficulty levels. To enable finegrained, process-oriented evaluation, we further propose the SDE rubric, which measures structural integrity, diagnostic reasoning, and evolutionary strategy during agent this http URL across more than ten mainstream LLMs reveal that even state-of-the-art models still struggle with multi-turn error propagation,implicit tool-use failures, and precise replanning. Moreover, we demonstrate that the SDE rubric provides effective supervision signals for improving agents’ reflective recovery capabilities. Our data and code are available at this https URL.
[LG-42] heoretical Guarantees for One-Shot Magnitude Pruning and Compute-Adaptive Early Exit ICML2026
链接: https://arxiv.org/abs/2609.12337
作者: Erdem Koyuncu
类目: Machine Learning (cs.LG)
*备注: Published in ICML 2026
Abstract:We study compute reduction in neural networks through a unified partial versus full computation view, captured by one-shot magnitude pruning in the static regime and early exit in the adaptive regime. In an asymptotic single-neuron model, we prove a concentration theorem for one-shot magnitude pruning with explicit rates. We also introduce the conditional perceptron for early exit and show that its excess generalization error decays as a power of the compute gap, with an exponent that grows to infinity as the alignment between partial and full computations tends to one. We then extend the analysis to deep networks, characterizing how pruning-induced distortions accumulate with depth and deriving a corresponding compute-accuracy tradeoff for frozen-backbone early exit under a neural network Gaussian process model. Numerical simulations corroborate the predicted scaling laws.
[LG-43] Simulating Disengaged Students to Evaluate LLM -based Tutors
链接: https://arxiv.org/abs/2609.12331
作者: Xianghui Meng,Jionghao Lin
类目: Machine Learning (cs.LG)
*备注:
Abstract:Simulated students generated by computational models provide a practical way to evaluate tutoring strategies and pedagogical approaches used by human and AI tutors. However, such simulations should account for disengaged behaviors, including gaming the system, wheel-spinning, and off-task behavior, because tutors may need different responses for different learner states. We present Disengagement-Aware Student Simulators (DAS2), a reproducible pre-deployment protocol that models five learner-engagement states: engaged, gaming, wheel-spinning, off-task, and mixed, and evaluates AI tutor performance across these states. Using ASSISTments09, two coders independently labeled 100 sampled tutoring sessions based on anonymized interaction-log summaries. They achieved 84% agreement (Cohen’s kappa = 0.78), and among agreed cases, human consensus labels matched DAS2 rule-based labels in 81% of cases (kappa = 0.75). Conditioning simulations on intended learner states reduced the correctness-rate gap between simulated and authentic sessions from 0.54 to 0.20 for gaming and from 0.51 to 0.18 for wheel-spinning. Fine-tuned Qwen2.5-7B better matched authentic response-time distributions, while prompt-only GPT-4o generated more distinguishable learner states. Evaluation of five AI tutors from the Claude, Llama, Gemini, Qwen, and GPT families showed that relative rankings remained stable across learner states and interaction lengths, while absolute performance varied, revealing state-specific differences in tutor support. Human validation further showed that automated tutor evaluation does not fully align with human judgment. DAS2 provides a pre-deployment framework for evaluating how AI tutors respond to diverse learner-engagement states before deployment.
[LG-44] Sampling via Decision-Flow: Training-Free Extraction of Improved Latent Reasoning Paths in Large Language Models
链接: https://arxiv.org/abs/2609.12317
作者: Zhendong Mi,Shaoyi Huang
类目: Machine Learning (cs.LG)
*备注:
Abstract:A central question in LLM reasoning is whether reinforcement learning (RL) instills genuinely new capabilities or merely reshapes how existing knowledge is expressed during inference. Building on the distribution-sharpening hypothesis, which holds that RL reallocates probability mass toward high-reward trajectories already latent in base models, we ask: can we unlock those latent paths without costly RL fine-tuning? We present Decision-Flow Sampling (DF-Sample), a training-free, data-free inference-time framework that constructs a hierarchical reasoning tree, scores terminal nodes for quality, and back-propagates utilities to inform each intermediate branching decision. Unlike conventional sampling strategies that make purely local step-wise choices, DF-Sample performs explicit global trajectory evaluation before committing to a path, recovering high-quality but low-probability reasoning chains that standard decoding overlooks. On GPQA, DF-Sample achieves 45.6% accuracy, surpassing power sampling (38.9%) and GRPO (39.9%), showing that a training-free method can outperform a trained one. Across three models and four benchmarks, DF-Sample consistently outperforms baselines, indicating substantial latent reasoning potential in pretrained base models.
[LG-45] FRIST: FMRI Representation Informed Shared-space Training Improves EEG-only Individual-Finger BCI Decoding
链接: https://arxiv.org/abs/2609.12298
作者: Jintao Zhang,Yidan Ding,Joshua Kosnoff,Maxim Karrenbach,Hanwen Wang,Bin He
类目: Machine Learning (cs.LG); Signal Processing (eess.SP)
*备注:
Abstract:Finger-level motor decoding is important for naturalistic brain-computer interface (BCI) control, yet individual-finger decoding from scalp electroencephalography (EEG) remains challenging because finger representations are spatially close in the sensorimotor cortex and blurred by volume conduction. Leveraging the high spatial resolution of functional MRI (fMRI), we introduce fMRI Representation-Informed Shared-Space Training (FRIST), a two-stage EEG decoding framework that first learns fMRI-informed spectral projections from simultaneous EEG-fMRI recordings and then uses fMRI-derived class geometry to guide residual refinement of EEG predictions. FRIST transfers information across recordings through shared finger labels without requiring paired trials and uses only EEG at inference. We evaluated 12 able-bodied participants during movement execution (ME) and motor imagery (MI) under two-class and three-class chronological session-held-out decoding simulating the online scenario. Using EEGNet as the EEG feature extractor, FRIST increased group average accuracy from 66.93% to 74.53% for two-class ME, from 44.83% to 56.58% for three-class ME, from 80.78% to 85.63% for two-class MI, and from 60.93% to 69.90% for three-class MI compared with the EEG-only EEGNet baseline. FRIST is also shown to improve EEG-only decoding when the target participant’s own fMRI data were unavailable. FRIST also generalized across multiple EEG decoding backbones, reaching 87.40% in two-class MI and 72.54% in three-class MI with EEG Conformer as the EEG feature extractor. These findings indicate that fMRI provide useful spatial constraints for EEG representation learning. FRIST improves noninvasive EEG-based finger-level BCI decoding, offering a multimodal strategy for integrating the spatial specificity of fMRI with real-time applicability of EEG.
[LG-46] Adaptive Chemotherapy Control under Tumor Heterogeneity via Reinforcement Learning
链接: https://arxiv.org/abs/2609.12264
作者: Bereket Sitotaw Kidane,Md Samiul Haque Motayed,Shuo Wang
类目: Machine Learning (cs.LG); Systems and Control (eess.SY)
*备注: 6 pages. Published in the 2026 American Control Conference (ACC)
Abstract:Designing effective chemotherapy regimens is hindered by tumor heterogeneity and drug resistance, which complicate the deployment of patient-specific model-based optimal control across diverse populations. We develop and compare closed-loop deep reinforcement learning (DRL) dosing policies with continuous (TD3) and discrete (DQN) action spaces trained on a high-dimensional heterogeneous tumor model. The DRL policies are benchmarked against a Pontryagin’s Maximum Principle (PMP)-derived open-loop benchmark. We assess generalization under parametric heterogeneity using a 100-patient virtual cohort with plus or minus 10 percent uniform perturbations in growth and drug-sensitivity parameters. Across this cohort, TD3 achieves higher average tumor reduction, while DQN yields tighter inter-patient dosing consistency, revealing a clear efficacy-consistency trade-off in this study. Our simulations assume full observation of all tumor subpopulations; translation to sparse and noisy clinical measurements will require partial-observability formulations and/or state estimation. Overall, the results show that simulation-trained DRL can learn state-dependent feedback dosing policies that complement open-loop optimal control benchmarks.
[LG-47] he Rank the Task Demands: A Causal Rank Law for Matrix Memories Trained on Group Composition
链接: https://arxiv.org/abs/2609.12259
作者: Samuel Larson
类目: Machine Learning (cs.LG)
*备注: 11 pages, 2 figures
Abstract:Matrix-valued memories make rank the natural budget of a learned representation: the number of independent directions a state spans bounds what it can bind, compose, and track. We report causal evidence, on a group-composition testbed trained under a hard single-state bottleneck with a fixed decoder that cannot launder rank, that gradient descent recruits precisely the rank the task’s algebra demands. A companion paper [Larson, 2026a] establishes the analogous recruitment and causal necessity pattern on a K -pair associative-binding testbed, where exact recovery provably requires state rank at least K ; this paper inherits that instrument and extends the rank law from a scalar capacity bound to a representation-theoretic one. We train toward chosen minimal faithful reference representations embedded in larger matrices. On group-composition state tracking over five finite groups spanning the solvable/non-solvable divide, the recruited rank equals the group’s minimal faithful real representation dimension d_\min (Spearman \rho = 0.9747 , the design’s tie-capped maximum), the dimension-matched solvable/non-solvable pair S_4 / A_5 is statistically equivalent under a pre-registered test, and a pre-registered force-rank test separates a guaranteed similarity ceiling from empirical recovery at the target dimension: one rank below d_\min , cosine similarity is capped by the target’s tied unit spectrum at \sqrt(d_\min-1)/d_\min \le 0.894 , below the 0.9 threshold in every group by construction, with observed cells at 86-95% (mean 91%) of that ceiling; at d_\min , not guaranteed a priori, recovery clears the pre-registered anchor-relative bar at four seeds per group in all five groups. Within this testbed, measured effective rank tracks representation dimension; the matched-dimension S_4 / A_5 comparison establishes equivalence within the pre-registered tolerance.
[LG-48] CRFCAN: A Complex-Valued Cross-Domain Residual Network for Joint Channel and Phase Noise Estimation in Sub-THz OFDM Systems
链接: https://arxiv.org/abs/2609.12244
作者: Ruilin Wang,Xiaodai Dong
类目: Machine Learning (cs.LG); Signal Processing (eess.SP)
*备注:
Abstract:In sub-terahertz (sub-THz) communications, the coupling of ultra-wide bandwidth and severe phase noise (PN) impairments renders conventional joint channel and PN estimation highly complex and computationally prohibitive. To address this, we propose CRFCAN, a complex-valued residual FFT convolutional attention network designed for joint channel and PN estimation. Unlike existing deep learning schemes that rely on cascaded networks or hybrid frameworks combining neural networks with conventional iterative estimators, CRFCAN performs joint recovery in a truly end-to-end fashion through a physics-inspired cross-domain structure. Specifically, Fast Fourier Transform (FFT) and inverse FFT modules are embedded within residual groups to enable iterative feature interaction across the time and frequency domains, thereby capturing both frequency-selective fading and time-varying phase distortions. In addition, two dedicated residual blocks are introduced for complex feature extraction and multiplicative phase-distortion modeling, respectively. A physics-aware PN output tail with soft normalization is further employed to improve estimation stability while preserving the physical characteristics of the effective PN process. Simulation results demonstrate that CRFCAN significantly outperforms conventional algorithms and state-of-the-art deep learning models in terms of normalized mean square error (NMSE) and bit error rate (BER). Notably, CRFCAN achieves superior performance with single-shot, fixed-complexity inference and generalizes well to unseen PN models without fine-tuning, highlighting its robustness and practicality for sub-THz receivers.
[LG-49] Patient-Reported Survey Data Improve Prediction of Opioid Use Disorder
链接: https://arxiv.org/abs/2609.12224
作者: Xiyue Jiang,Zihan Ding,Grace Han,Yinan Liu,Richard N. Rosenthal,Fusheng Wang
类目: Machine Learning (cs.LG)
*备注: 10 pages; submitted to the AMIA 2027 Amplify Informatics Summit
Abstract:Electronic health records (EHRs) may incompletely capture patient-reported factors associated with opioid use disorder (OUD). We evaluated whether survey data improve prediction of a first recorded OUD diagnosis among 267,747 All of Us participants with documented opioid exposure, including 15,287 OUD cases. We compared EHR-only and EHR+survey models across 6-, 12-, and 24-month look-back windows using logistic regression, random forest, XGBoost, LightGBM, multilayer perceptron, LSTM, GRU, and Transformer. Survey augmentation improved PR-AUC across all 24 model-window combinations by 0.0087-0.0505; the best 24-month LightGBM model improved from 0.6219 to 0.6603. Survey coverage increased with longer windows and differed by OUD status (24 months: 21.7% OUD-positive vs. 60.7% OUD-negative). Permutation analysis ranked survey features as the second most important information domain at 24 months in both evaluated models. Patient-reported data provide complementary predictive signals beyond structured EHRs while highlighting the importance of survey availability.
[LG-50] Estimating Pedestrian Volumes from GIS-Derived Built-Environment Features: A Machine Learning Framework
链接: https://arxiv.org/abs/2609.12173
作者: Bahareh Golchin,Banafsheh Rekabdar,Sirisha Kothuri,Joseph Broach
类目: Machine Learning (cs.LG)
*备注:
Abstract:Transportation agencies need pedestrian volume estimates across entire road networks to prioritize safety investments, yet manual counts are expensive and cover only a small share of intersections. We present a machine learning pipeline that predicts 2-hour PM peak pedestrian volume at 101 urban intersections in Portland, Oregon, from built-environment, land-use, and street-network features drawn from open GIS data. Starting from the Negative Binomial GLM used in practice, we add feature selection, count-aware gradient boosting, and repeated cross-validation, selecting one configuration by a combined rank over RMSE, MAPE, and SMAPE across four cross-validation strategies. The winner, a histogram-based gradient boosting model with Poisson loss and L1 Lasso feature selection, reduces cross-validated RMSE by 12% over the GLM baseline (89.8 to 78.7) and holdout RMSE by 19% (108.0 to 87.9). Code is released on GitHub.
[LG-51] Certifying Concept Unlearning in Text-to-Image Diffusion Models
链接: https://arxiv.org/abs/2609.12163
作者: Mansi,Luca Marzari,Francesco Leofante
类目: Machine Learning (cs.LG)
*备注:
Abstract:Existing evaluations of concept unlearning in text-to-image (T2I) diffusion models primarily rely on attack success rates obtained through automated adversarial prompt search. However, these metrics provide only empirical evidence over a finite set of queries and leave residual leakage over the broader prompt space largely unquantified. This limitation can lead to overestimating unlearning effectiveness and underestimating safety risks. To address this gap, we introduce a novel certification framework for T2I concept unlearning that provides high-confidence guarantees with bounded error on residual concept leakage. Our approach combines statistical certification with worst-case analysis along concept-relevant embedding directions to derive explicit upper bounds on leakage probability under user-specified confidence levels. We evaluate our framework across three major concept categories namely NSFW content, artistic styles, and celebrity identities, and six state-of-the-art unlearning methods. Certified leakage bounds consistently exceed standard attack success rates by 16.2%, uncovering substantial residual risks missed by existing evaluation protocols. Crucially, our results demonstrate that empirical attack-based evaluations can significantly underestimate residual leakage and establish certification as a necessary complement for reliable auditing of concept unlearning in T2I diffusion models.
[LG-52] Direct Topology Tracking in Continuous Implicit Models
链接: https://arxiv.org/abs/2609.12157
作者: Guanqun Ma,David Lenz,Kaiyuan Tang,Hanqi Guo,Chaoli Wang,Tom Peterka,Bei Wang
类目: Computational Geometry (cs.CG); Machine Learning (cs.LG)
*备注:
Abstract:We present a framework for tracking topological features directly within continuous implicit models. Such models, including implicit neural representations (INRs) and multivariate functional approximations (MFAs), are increasingly adopted to represent scientific data without the resolution constraints of discrete grids. They offer compact, smooth, and differentiable representations of complex fields, enabling new opportunities for high-performance data storage, reconstruction, and analysis. Given a continuous implicit model, our method tracks the evolution of critical points by querying the model and its derivatives, thereby eliminating the need to resample onto a grid. This approach enables faithful feature tracking while avoiding discretization-induced artifacts such as aliasing. We demonstrate the generality of our framework across a range of implicit representations, including analytic functions, MFAs, and INRs, and show that it produces smooth, coherent critical point trajectories. By enabling feature tracking directly on continuous representations, our method supports a new class of feature-driven visualization workflows centered on implicit models.
[LG-53] On Identifying Adversarial Intent Injection in AI-Native 6G Networks
链接: https://arxiv.org/abs/2609.12144
作者: Nilesh Chakraborty,Petar Djukic,Burak Kantarci
类目: Networking and Internet Architecture (cs.NI); Cryptography and Security (cs.CR); Machine Learning (cs.LG)
*备注: 6 pages, 3 figures, Accepted to IEEE Global Communications Conference (Globecom) 2026
Abstract:AI-native 6G networks have brought Intent-Based Networking (IBN) to the forefront, enabling high-level goals to be translated into network configurations. However, this abstraction opens new attack surfaces, primarily adversarial intent injection, where malicious policies are disguised within benign intent flows. The detection of attack instances might become significantly more difficult if the adversaries adopt a stealthy mode of malicious intent injection. With all these in mind, we first define a fine-grained threat model that facilitates the threat of malicious intent injection in an AI-native network. Alongside, we investigate four malicious intent injection strategies - stealth-mode, random distribution, increasing frequency, and decreasing frequency- and propose a dual-path detection framework: (i) a CNN using TF-IDF features for supervised malicious intent detection, and (ii) an AutoEncoder trained exclusively on benign data for one-class malicious intent detection. Our evaluation demonstrates strong detection performance, with accuracy improving to 0.97 (~9% gain) and F1-score to 0.98 (~36% gain) over the state-of-the-art baseline.
[LG-54] GUIDE: Generative Utility Inference and Decision Engine
链接: https://arxiv.org/abs/2609.12137
作者: Anagha Tiwari,Alexander G. Gray,Nick Feamster,Brian Jabarian,Alex Imas,Alex Kale
类目: Machine Learning (cs.LG)
*备注:
Abstract:Measuring the preferences of human users remains a fundamental challenge of AI alignment. Existing elicitation approaches struggle to efficiently discover multidimensional preferences or accurately ground these inferences in domain knowledge. To address this, we introduce GUIDE, an LLM-driven elicitation architecture that infers user preferences through conversations by combining Bayesian adaptive sampling for question selection and symbolic representation learning to initialize domain-specific preference models. GUIDE generalizes adaptive sampling to diverse elicitation questions through an extensible type system of transforms on a parameterized preference state. GUIDE produces domain-specific preference representations through an initialization process using symbolic rule-based learning to capture world knowledge and set priors over preference dimensions grounded in data about decision alternatives. The architecture provides observability and steerability to facilitate deployment and analyze elicitation processes. In silico experiments on investment portfolio optimization demonstrate that GUIDE improves cold-start and minimizes recommendation regret consistently within early elicitation interactions across user personas compared to prior work, LLM-only baselines, and ablated GUIDE versions.
[LG-55] Rank-Efficient LoRA via Joint Tangent-Space Optimization under Isotropic Curvature
链接: https://arxiv.org/abs/2609.12123
作者: Zihan Zhu,Zhehang Du,Xuyang Chen,Tim Tsz-Kit Lau,Jiayuan Wu,X. Y. Han,Qi Long,Weijie Su
类目: Machine Learning (cs.LG); Optimization and Control (math.OC); Machine Learning (stat.ML)
*备注:
Abstract:Low-Rank Adaptation (LoRA) is an effective approach for adapting large pretrained models by learning low-rank weight updates. In practice, the LoRA rank is used to control an adapter’s parameter budget and representational capacity. We show that this view is incomplete: while the nominal rank determines the representational capacity, the optimizer shapes how much of that capacity is used in the induced weight-space updates. In a case study of GPT-2 adaptation with LoRA, we observe a strong rank-dependent optimizer effect. Despite using the same nominal rank, AdamW often produces per-step updates with concentrated singular spectra and low effective rank, whereas Muon uses a richer set of directions and benefits more consistently from increasing LoRA rank. These observations motivate ISO-LoRA, an optimizer that couples the LoRA factor updates through spectral descent on the induced tangent perturbation in weight space. ISO-LoRA promotes updates that distribute energy more evenly across singular directions, improving rank utilization while preserving compatibility with the LoRA parameterization. We complement this design with theoretical guarantees showing that ISO-LoRA can achieve higher effective rank than standard factor-wise optimizers through a one-step analysis under a stylized spiked-gradient model. We validate this design on language-model adaptation across 0.1B-7B-parameter models, where ISO-LoRA improves effective rank and downstream performance, with the strongest gains at moderate-to-large LoRA ranks. Our results highlight rank utilization as a key factor in LoRA optimization and suggest that optimizer design offers an important path toward stronger parameter-efficient adaptation.
[LG-56] Almost Sure Convergence Analysis of Stochastic Gradient Methods with Clipping and Additive Noise
链接: https://arxiv.org/abs/2609.12119
作者: Amartya Mukherjee,Jun Liu
类目: Machine Learning (cs.LG); Optimization and Control (math.OC); Probability (math.PR)
*备注:
Abstract:Stochastic gradient descent (SGD) with gradient clipping and additive noise has become a standard technique for training machine learning models, particularly in applications requiring robustness or privacy guarantees. However, clipping introduces a bias in stochastic gradients, while additive noise introduces additional variance, making the long-run behaviour of individual optimization trajectories difficult to characterize. In this work, we prove that SGD with clipping and additive Gaussian noise (SGD-CN) converges almost surely (a.s.) under smoothness and uniformly bounded stochastic-gradient noise assumptions, provided the step sizes satisfy some standard decaying conditions. Our analysis extends to momentum variants such as the stochastic heavy ball and Nesterov’s accelerated gradient, where we show that careful energy constructions yield similar guarantees. These results provide stronger theoretical foundations for understanding the pathwise behaviour of clipped stochastic gradient methods and suggest that, despite the bias and noise introduced by clipping and perturbation, the algorithm remains stable in both convex and nonconvex regimes.
[LG-57] Score-based Outlier Generation via Controlling the Radon-Nikodym Derivative
链接: https://arxiv.org/abs/2609.12113
作者: Amartya Mukherjee,Tristan Milne,Kry Yik-Chau Lui,Stephanie Hazlewood,Jun Liu
类目: Machine Learning (cs.LG); Analysis of PDEs (math.AP); Optimization and Control (math.OC); Probability (math.PR); Machine Learning (stat.ML)
*备注:
Abstract:Outliers are important for stress-testing algorithms and understanding system behaviour under rare conditions. Despite being commonly described as low-likelihood events, existing generative approaches rarely control likelihood explicitly. In this work, we introduce a measure-theoretic notion of outliers based on the distribution of log-likelihood values, which is guaranteed to assign higher probability mass to low-likelihood events with a specifiable magnitude. Building on this formulation, we derive how likelihood reweighting modifies the diffusion score and use this relation to motivate a controlled modification of the reverse-time dynamics. In particular, likelihood reweighting implies a scaling of the score function with a control term derived from the Radon-Nikodym derivative of the likelihood distributions. Correspondingly, the updated score function can be obtained with no retraining of the diffusion model. We exploit the Ornstein-Uhlenbeck semigroup underlying diffusion models to motivate an exponentially interpolated controller which approximates the true control. Experiments demonstrate controlled generation of low-likelihood samples while remaining consistent with the data geometry.
[LG-58] Receiver-Surface Hit Patterns via Legendre Approximation for Molecular Signal Detection
链接: https://arxiv.org/abs/2609.12089
作者: Yasin Bastug,Erencem Ozbey,H. Birkan Yilmaz
类目: Information Theory (cs.IT); Machine Learning (cs.LG)
*备注: 5 pages, 4 figures
Abstract:Detecting whether a transmitter is actively communicating with a receiver is a fundamental problem in molecular communications. A spherical receiver may measure not only the number and arrival times of absorbed molecules, but also their absorption locations on the receiver surface. These locations contain a directional signature that is lost in count-only detection. In this letter, we develop a molecular signal detector based on a Legendre polynomial expansion of the receiver-surface hit density and extend it to a Legendre-based Viterbi sequence detector. By exploiting the surface-level directional signature of molecular arrivals, the proposed detectors provide geometry-aware, efficient, and better-performing alternatives to count-only detection.
[LG-59] Efficient Vision-Language-Action Management and Serving for Robot Factories
链接: https://arxiv.org/abs/2609.12075
作者: Dionysios Adamopoulos,Nattapol Chanpaisit,Basel Fakhri,Christina Giannoula
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Hardware Architecture (cs.AR); Machine Learning (cs.LG); Performance (cs.PF); Robotics (cs.RO)
*备注:
Abstract:Vision-Language-Action (VLA) models show high robotic manipulation capabilities via a two-stage design: a Vision-Language Model (VLM) stage followed by an Action Diffusion Transformer (ADiT) stage. Since robots must meet strict Service-Level Objectives (SLOs) for safety, VLA inference is inherently latency-critical. Meeting these SLOs requires high-end GPUs, yet weight, cost, and power constraints preclude integrating such GPUs on-robot. Prior works offload VLA inference to edge servers that serve many robots on VLA models. However, current VLA systems lack support for multi-request, multi-model execution on a multi-GPU server under SLOs, while existing serving systems for multi-stage models are optimized for throughput and stage disaggregation across separate GPUs, which are ill-suited for the millisecond-scale stages of VLA models. We design Robion, the first VLA serving and management system for multi-robot, multi-model requests on multi-GPU edge servers that meets SLOs. Our serving engine disaggregates the VLM and ADiT stages within a GPU via two streams, dynamically restricting the SMs on VLM stream so ADiT always finds SMs to run alongside it, and co-locates multiple models by sharing these streams across them, prioritizing requests by least remaining SLO time. Our management engine enables flexible model placements on multi-GPU servers, and integrates an intelligent traffic controller that maximizes per-model batching under the chosen placement while bounding each GPU’s load to meet SLOs. For individual models, Robion serves on average 6.7 \times and 1.5 \times higher robot load within 98% SLO attainment over vLLM-Omni, the most widely used multi-stage serving system, and Monolithic, which runs VLM and ADiT as a single pipeline, respectively. In a large-scale experiment of serving 8 different models on a 4-GPU server, Robion can serve up to 64 robots within 98% SLO attainment.
[LG-60] Scalable Discrete-to-Continuous Channel Simulation for Compression and Privacy
链接: https://arxiv.org/abs/2609.12067
作者: Joseph Rowan,Buu Phan,Ashish J. Khisti
类目: Machine Learning (cs.LG); Information Theory (cs.IT)
*备注: 39 pages, 9 figures
Abstract:Channel simulation has recently emerged as a useful component in machine learning systems where samples from a prescribed probability distribution are to be compressed. Yet, general channel simulation algorithms often suffer from high computational costs, random stopping times or, in the worst case, can require generating an infinite number of shared random samples. We introduce a scheme for both exact and approximate simulation of discrete-to-continuous channels which conversely uses a fixed number of random samples, and therefore has a runtime independent of the channel and the input. Unlike existing channel simulation schemes which generate a sequence of independent samples from a proposal distribution, our approach generates one sample, or alternatively a fixed number of samples, from each potential target distribution. We then apply a latent permutation to the samples before performing sample selection using an exponential race. Our scheme provides a flexible tradeoff between the number of generated samples and the compression rate. Using polar and multilevel coding, we scale our approach to handle long blocklengths in O(n \log n) time in order to benefit from reduced per-symbol overhead. We conclude by demonstrating applications to variable-rate compression with stochastic VQ-VAEs and communication-efficient differentially private distributed mean estimation via exact simulation of the Gaussian mechanism.
[LG-61] Reinforcement Learning for Syndrome Extraction
链接: https://arxiv.org/abs/2609.12020
作者: John Zhuoyang Ye,Aarav Pabla,Jens Palsberg
类目: Machine Learning (cs.LG); Quantum Physics (quant-ph)
*备注:
Abstract:A key subtask of quantum error correction is to extract a syndrome that, if nontrivial, signals an error. The number of possible ways to extract a syndrome grows exponentially with the syndrome size, and these implementations vary greatly in fault tolerance, as measured by their logical error rates. This creates a natural search problem: find an implementation with a low logical error rate. Previous work solves this problem but sacrifices either solution quality or scalability. In this paper, we use reinforcement learning and importance sampling to outperform previous work at all scales. Compared with the state of the art automatic scheduling tools AlphaSyndrome and PropHunt, our tool reduces the logical error rate by 25.9% and 71.7% on average, respectively, culminating with a reduction of 97.8% for a surface code with distance 15.
[LG-62] oward Reliable Railway-Bogie Response Prediction Using Multifidelity TDNN and Physics-Informed Residual Learning
链接: https://arxiv.org/abs/2609.12018
作者: Gyeolhee Lee,Moosun Kim,Taewook Kwon,Jaehun Kim,Dongjin Lee
类目: Machine Learning (cs.LG); Applied Physics (physics.app-ph)
*备注: 30 pages, 15 figures
Abstract:Railway engineers need simulation models that predict vehicle responses across operating scenarios that cannot be tested exhaustively. Agreement with representative measurements provides essential evidence, but calibration at a limited set of conditions does not guarantee accuracy elsewhere. We present a multifidelity railway-bogie response-correction method that treats multibody simulation histories as low-fidelity information and roller-rig measurements as high-fidelity evidence. This method combines an experiment-anchored fidelity assignment with physics-informed discrepancy learning for multichannel bogie-response histories. A time-delay neural network (TDNN) represents the condition-dependent simulation trend, and development-fitted amplitude alignment defines the low-fidelity baseline. A residual-correction network then models the reproducible response component not explained by this baseline and adds it to the baseline. An effective dynamic-balance equation constrains the learned discrepancy by representing differences in inertia, damping, stiffness, and external forcing between the simulated and physical systems. The training objective combines this constraint with residual matching, temporal smoothness, and selectively applied displacement-acceleration consistency terms. For the evaluated reconstruction case, the corrected response gives a mean coefficient of determination of 0.8197, a mean normalized root-mean-square error (NRMSE) of 4.6055 %, and a mean normalized mean absolute error (NMAE) of 1.9297 %. These results provide initial evidence of accurate response prediction at the held-out 385 km/h condition.
[LG-63] Inverting Self-Triggered Control: Adversarial Reinforcement Learning for Sparse Denial-of-Service Attacks
链接: https://arxiv.org/abs/2609.12016
作者: Adam Haroon,Erick J. Rodríguez-Seda,Tristan Schuler,Cody Fleming
类目: Machine Learning (cs.LG); Systems and Control (eess.SY)
*备注: 8 pages, 3 figures
Abstract:Self-triggered reinforcement learning control (RL-STC) learns the sparsest control schedule that preserves Lyapunov-decreasing stability under a Run-Time Assurance (RTA) override. We invert this: an adversarial RL agent learns the sparsest jamming or Denial-of-Service (DoS) schedule that destabilizes the closed loop, with a Lyapunov-increase admissibility predicate mirroring the defender’s safety certificate. We prove a plant-property lower bound on the minimum jam count required for an immediate hold-last medium-access-control adversary to force a crash against a self-triggered controller (STC) satisfying a Lyapunov contract, and recover a certificate-level analog of the consecutive-grouping optimality of prior count-budget DoS scheduling as a corollary. This extends the DoS-scheduling count-budget analysis from periodic and linear-time-invariant to STC controllers. Empirically, we train against four fixed defenders per plant (one Linear Quadratic Regulator (LQR) and three RL-STC) on Pendulum, CartPole, and Quadrotor2D. The learned adversary is the only adversary that crashes every defender on every plant at 100% : greedy misses Quadrotor2D LQR on 42% of episodes and periodic misses Pendulum LQR on 97% . On jam-time-per-failure it beats baselines by up to 2.8\times , and shows its widest absolute margin on Quadrotor2D LQR. Robustness ablations show that Gaussian observation noise exceeding the initial-state magnitude and position-only observation both preserve 100% failure rate and keep the learned adversary strictly ahead of both baselines on jam-time-per-failure.
[LG-64] Certified Safety Curation: Distribution-Free Guarantees for Safe Offline Reinforcement Learning
链接: https://arxiv.org/abs/2609.12014
作者: Adam Haroon,Cody Fleming
类目: Machine Learning (cs.LG)
*备注: 37 pages, 10 figures
Abstract:Safe offline reinforcement learning assumes a cost function on every transition. We ask what remains possible when safety can be judged only by comparing short clips and occasionally asking whether an episode exceeded its budget. Certified safety curation answers with a filter-then-clone pipeline: a state-only value trained from segment comparisons scores whole trajectories, Learn-then-Test calibration certifies a selection threshold under a distribution-free (\alpha, \delta) bound on the unsafe fraction of the selection, and behavior cloning follows. We are not aware of prior work certifying the composition of a training set for offline RL or imitation. Oracle controls justify the design: reweighting individual transitions fails even with an exact value, so the value selects whole trajectories. The policies satisfy the cost budget on eleven of fifteen DSRL tasks, one short of cloning the ground-truth safe subset, which needs a label on every trajectory; the uncertified variant reaches twelve. Retrained on the certified selection, the strongest full-label method becomes safe where no setting of its own cost target rescues it. Refusal is predictable: the certificate’s probability has a closed form in the purity the pool attains, which the calibration sample estimates and the scorer enters only through.
[LG-65] QTrans: A Quantum Transformer for Sentiment Classification
链接: https://arxiv.org/abs/2609.12011
作者: Ren-Xin Zhao,Xinjie Huang,Yahong Liu,Maoyu Ye,Jinjing Shi,Shi Wang,Yaonan Wang
类目: Machine Learning (cs.LG)
*备注:
Abstract:In small-scale binary sentiment classification scenarios, factors such as negation, contrastive shifts, and cross-word dependencies lead to the non-linear coupling of sentiment cues, making it difficult for conventional lightweight models to fully capture the contextual relationships between tokens. To address this issue, we propose a model named QTrans, which uses parameterized quantum circuits to construct query, key, and value features and derives attention coefficients from Gaussian distances between quantum measurements. By further integrating a quantum feed-forward neural network, residual connections, and layer normalization, the model establishes an end-to-end trainable quantum-classical hybrid framework for sentiment classification. Experimental results on the MR, CR, and MPQA datasets show that QTrans achieves test accuracies of 72.13%, 69.51%, and 63.45%, respectively, representing improvements of 2.88, 3.17, and 3.79 percentage points over the best-performing classical baselines for each dataset. Overall, QTrans expands the application of parameterized quantum circuits in lightweight sentiment analysis and lays an experimental foundation for further research into quantum multi-head self-attention for modeling textual relationships.
[LG-66] Fixed State Long Reach: What a Constant-Size Cache Buys Block Diffusion at Scale
链接: https://arxiv.org/abs/2609.11998
作者: Vaibhav Singh,Pierre-André Noël,Torsten Scholak,Eugene Belilovsky,Oleksiy Ostapenko
类目: Machine Learning (cs.LG)
*备注: 9 pages, 3 figures
Abstract:Diffusion language models decode tokens in parallel, but their bidirectional denoiser rules out the naive key–value (KV) cache behind fast autoregressive inference. Block diffusion restores caching by decoding block-by-block, and the block caches deployed on it so far are tied to attention: O(L)in memory and, if used as training-free retrofits, only an approximation of the model’s computation. Both constraints can be overcome: sequence mixers that summarize finalized blocks into a reusable state support block caching, and the corresponding block-causal training objective makes the cache exact. We study this recipe at scale, pretraining three 3B block-diffusion denoisers (attention, mamba, and hybrid) on 300B tokens under one single-frontier objective and decoding all three through a single cached interface. Only the state-space cache is O(1) in sequence length: its memory and per-step latency stay constant at any context length, while an attention cache remains O(L). At 256k tokens (where attention has grown to 82GB and 29 ms/step), the Mamba cache delivers 4.3x lower latency, 11x less memory, and 2.6x higher single-stream throughput; and because that footprint is constant it scales with batch as well, reaching 14x the aggregate throughput, where attention cannot run beyond a single stream. The same linear-state bias lets the Mamba and hybrid backbones keep retrieving out to 8-16x their training length, whereas attention’s retrieval collapses at 2x, at no measured quality cost.
[LG-67] DCRA: Diffusion-Conditioned Representation Alignment for Robust Time-Series Learning
链接: https://arxiv.org/abs/2609.11997
作者: Wenrui Xu,Anas Enanaa,Keshab K. Parhi
类目: Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:Learning robust representations for time-series signals under noise and distribution shifts remains challenging, especially in clinical applications such as electroencephalogram (EEG) and electrocardiogram (ECG) analysis. We propose Diffusion-Conditioned Representation Alignment (DCRA), a training framework that repurposes the forward diffusion process as a structured corruption scheduler for representation learning. Different from conventional augmentation and consistency-based methods that rely on independently sampled perturbations, DCRA introduces a structured corruption trajectory via the diffusion forward process, which enables continuous and controlled representation evolution across noise levels. We introduce a feature-level consistency objective that aligns representations across noise levels while preserving class-discriminative structure. This mechanism promotes structure-preserving consistency, which enables smooth and semantically coherent feature trajectories in latent space. The proposed framework is encoder-agnostic and can be integrated with state space models and Transformer architectures. The seizure detection experiments on the CHB-MIT EEG dataset show that DCRA consistently improves performance under multiple noise conditions and achieves higher sensitivity at low false-positive rates. Analysis reveals that DCRA produces more balanced and structured representations compared to baseline and diffusion-only models. These findings highlight the benefit of combining structured corruption with representation alignment for robust time-series learning.
[LG-68] Explainable Prediction from Mobile Sensing Data through LLM -guided Concept Integration
链接: https://arxiv.org/abs/2609.11995
作者: Yuning Wang,Iman Azimi,Amir M. Rahmani,Pasi Liljeberg
类目: Machine Learning (cs.LG)
*备注: 15 pages, 5 figures; 10-page supplementary information
Abstract:Mobile sensing enables longitudinal monitoring of behavioral and physiological patterns in everyday settings. However, accurate prediction remains challenging in small-cohort health-sensing studies, where task-specific outcome supervision is limited relative to heterogeneous sensing data. Interpretability is also important, as model outputs should reflect meaningful behavioral and physiological patterns rather than predictive scores alone. We develop a Concept-Integrated Transformer (CIT) with LLM-guided concept supervision for explainable prediction from mobile sensing data. CIT uses a pretrained large language model to generate baseline-aware concept abnormality targets with confidence weights without manual concept annotation. Across two longitudinal datasets, CIT achieves the highest F1 score on AFFECT (0.756) and ties for the highest on a PHQ-9 dataset (0.765). The learned concept scores also reveal interpretable behavioral and physiological patterns; in AFFECT, sleep quantity and quality show the clearest difference between high and low negative affect groups. These findings support LLM-guided concept integration for accurate and interpretable prediction in small-cohort mobile sensing studies.
[LG-69] FINESSE: An Agent -Based Simulator and Benchmark Dataset for Multimodal Financial Event Sequences
链接: https://arxiv.org/abs/2609.11993
作者: Tyler Farnan,Benjamin Eng,Adam Abate,Xirui Hou,Rizal Fathony,Nam H. Nguyen,Senthil Kumar
类目: Machine Learning (cs.LG)
*备注:
Abstract:Machine learning research in financial services is limited by the scarcity of representative open-source datasets. Existing resources are often narrowly focused on a single modality or task and fail to reflect the structured, multimodal, and dynamic nature inherent to many problems in financial services. In this paper, we introduce FINESSE, a Financial Event Sequence Simulation Environment, an agent-based simulation framework for generating synthetic, structured datasets composed of multiple interdependent event streams. Each stream corresponds to a distinct financial behavior such as transactions, payments, account status changes, and policy interventions, each with unique action spaces, schemas and variable types. These streams are coupled through agents’ latent evolving states, enabling the simulation of temporally rich interactions. We also introduce FINESSE-Bench, a benchmark dataset generated by the simulator, supporting four representative tasks: balance forecasting, transaction fraud detection, missed payment prediction, and next event prediction. We report baseline results using methods from time series forecasting, event sequence modeling, temporal graphs, and temporal point processes. We release the FINESSE framework, including the simulator and dataset to accelerate research on structured, multimodal event sequence modeling challenges in financial services. Subjects: Machine Learning (cs.LG) Cite as: arXiv:2609.11993 [cs.LG] (or arXiv:2609.11993v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.11993 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-70] Decoding Mixture Perception through Computational Modeling of Component Interactions
链接: https://arxiv.org/abs/2609.11958
作者: Fei Wang,Xiaoya Xie,Junfei Liu,Huihao Wang,Yixiao Wang,Yintao Wang,Yi Li,Hao Dong,Xing Chen
类目: Machine Learning (cs.LG)
*备注:
Abstract:Olfaction played an indispensable role throughout human evolution and civilization. Even in the contemporary era of advanced technology, olfaction remains a critical channel for person to conduct danger discrimination, emotional experience, and memory formation. However, most substances in nature exist as multi-molecule mixtures. The complexity of mixture compositions, as well as concentration dependent saturation effects and receptor specific activation thresholds, pose substantial challenges in identifying olfactory characteristics. In this study, we proposed a novel bio inspired deep learning framework for accurate odor perception recognition of mixtures. We robustly constructed neural response curves for molecule-receptor interactions, and developed a fusion strategy that integrates attention-weighted multi-receptor curves with concentration-dependent multi-molecule curves, replicating the competitive activation and synergistic integration of mixture components. Furthermore, by comparing the consistency of response curve patterns, the model can transfer knowledge from the semantically rich space of molecular associations to guide recognition of mixture perception characteristics. Therefore, we established a complete computational pathway from chemical blending, neural encoding, to perceptual formation. Finally, we conducted comprehensive evaluation, and results demonstrated exceptional superiority, achieving an accuracy of 92.2%. Consequently, our work provides a generalizable solution to the long standing mixture perception challenge. More importantly, it can be integrated into embodied cognitive systems to enhance the agents perceptual and interactive capabilities in complex scenarios.
[LG-71] Performance Efficiency and Collapse – Advantages and Challenges in Offline Post-training of Code LLM s
链接: https://arxiv.org/abs/2609.11956
作者: Abhinav Anand,Sanjana Reddy Pachika,Shweta Verma,Mira Mezini
类目: Machine Learning (cs.LG)
*备注:
Abstract:Post-training with reinforcement learning (RL) is a critical phase in the development of code-generating large language models (LLMs), as it ensures adherence to instructions and the production of functionally correct code. This process typically requires computationally intensive code sample generation from Transformer-based LLMs and substantial GPU-CPU communication for sequence verification. To address these computational challenges, this work examines whether RL-based post-training can be performed entirely offline by leveraging existing datasets rather than generating new samples. The findings indicate that, with only a few hours of training, zero-shot code generation performance of LLMs can be substantially improved without online sampling. Additionally, offline RL produces performance gains across models ranging from 0.5B to 7B parameters, although the extent of improvement varies among model families.
[LG-72] Efficient AI Model Deployment Using Quantization Analysis Tool
链接: https://arxiv.org/abs/2609.11954
作者: Dwith Chenna,Kanishka Macherla
类目: Machine Learning (cs.LG)
*备注:
Abstract:As deep learning models are increasingly deployed on resource constrained devices, the demand for efficient model optimization techniques continues to grow. Effective deployment of AI models on edge and low power platforms requires optimization methods that reduce model size and computational cost while maintaining high accuracy. This paper presents Quantization Analysis Tool, a practical system designed to streamline quantization workflows and support performance efficient model deployment. Built on the ONNX framework for broad interoperability, the tool provides detailed layer-wise sensitivity analysis, visualization of weight and activation distributions, and insights to guide precision selection. By identifying layers that are resilient or sensitive to reduced precision, the tool enables developers to make informed trade-offs between model size, latency, and accuracy. Experimental evaluations across multiple neural network architectures demonstrate that the tool effectively improves the quantized accuracy, leading to improved efficiency in real-world deployment scenarios. The tool also provides developers valuable insights into the effects on quantization on the model and its accuracy. This work highlights the tools capabilities, practical applications, and its role in enabling efficient AI model deployment through robust quantization analysis
[LG-73] Hyperion: An AI-powered HPC cluster for sciences and humanities research that utilizes ML for predicting job turnaround time
链接: https://arxiv.org/abs/2609.11946
作者: Jun Zhou,Nathan Elgar,Tawnee Benedetto,John Richards,Ming Hu,Greg Wilsbacher,Lawrence Miao,Paul Sagona
类目: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG)
*备注:
Abstract:Hyperion is an innovative high-performance computing (HPC) cluster developed for researchers in both science and humanities disciplines at the University of South Carolina (USC). Our approach involved constructing a HPC cluster designed to meet the current research needs while accommodating future expansion. Additionally, we developed and trained two machine learning (ML) models to predict turnaround time, including wait time and wall time, and seamlessly integrated them into the Slurm job submission. Finally, we showcase a variety of sample applications hosted on the Hyperion platform.
[LG-74] he Battery Price of edge AI: A study of the Environmental Impact of LLM Inference on Mobile Devices
链接: https://arxiv.org/abs/2609.11940
作者: Édouard Guégain,Tristan Coignion
类目: Performance (cs.PF); Machine Learning (cs.LG)
*备注: 12 pages. Currently under submission at a conference
Abstract:The rapid diffusion of generative artificial intelligence raises privacy, latency, and performance concerns that motivate a shift toward “local-first” AI, where inferences are performed on the user’s device instead of on remote cloud servers. This paradigm also places a significant computational load on battery-powered smartphones, potentially shortening battery life and increasing the overall replacement rate of mobile devices. This paper presents a systematic study of the energy consumption, performance, and accuracy of on-device large language model (LLM) inference. We evaluate 18 models from different model families, sizes, and quantization levels, on two modern smartphones and on a server, using the respective state-of-the-art for such deployments. We measure the energy per generated token, inter-token latency, model accuracy, and battery-cycle consumption. Our results show that (i) on-device inference is on average 3 times less energy-efficient than batched server inference; (ii) the relationship between quantization bit-width and energy per token is non-monotonic, with energy sweet spots on both tested smartphones; (iii) eight out of 18 model configurations lie on the Pareto front of accuracy and energy-efficiency, allowing practitioners to build battery-aware model routers; and (iv) realistic modeling assumptions do not allow local inference to be less environmentally impacting per token than batched server inference, with 88–90% of that impact attributable to device embodied carbon rather than electricity consumption. These findings challenge the premise that local AI is more sustainable than cloud inference, and motivate the need for context-aware and life-cycle-aware model selection when deploying edge AI on battery-powered mobile platforms.
[LG-75] Fed-Equilibrium Framework for Topological Pareto Control in Robust and Fair Clinical Federated Learning CEC
链接: https://arxiv.org/abs/2609.11937
作者: Ting Xu,Henry Leung
类目: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC)
*备注: Accepted to the IEEE Canadian Conference on Electrical and Computer Engineering (CCECE 2026)
Abstract:The deployment of Federated Learning (FL) in multi-center clinical networks faces the challenge of “knowledge dominance,” where high-volume hubs naturally overwhelm minority community nodes, implicitly treating the distinct clinical patterns of smaller cohorts as outliers. Existing geometric defenses provide a security baseline but leave this efficiency-fairness dilemma unresolved. To bridge this gap, we propose Fed-Equilibrium, a framework that advances the paradigm from simple defense to topological equilibrium. Unlike traditional aggregators, Fed-Equilibrium implements a sequential architectural synergy. It utilizes a two-stage gradient control cascade: Stage I (geometric quality assurance) enforces directional consistency via a cosine similarity funnel to filter malicious noise, creating a stabilized manifold; Stage II (topological Pareto control) then actively modulates verified contributions by identifying the optimal Pareto knee point. We validated this framework on a bi-national simulation integrating Canadian (CNODES) and U.S. (SyntheticMass) registries. Experimental results demonstrate that the system simultaneously secures the network against adversarial divergence while accommodating underrepresented signals. Notably, the minority U.S. spoke (representing less than 3% of data volume) achieved deep convergence comparable to the data-rich Canadian hub. This confirms that Fed-Equilibrium effectively counters “knowledge dominance,” establishing a true “knowledge commons” where global generalizability does not come at the cost of local clinical representation.
[LG-76] One Simple Trick for Improving the Performance of Energy-Limited Local Inference and Training ICML2026
链接: https://arxiv.org/abs/2609.11936
作者: Erik Schultheis,Maximilian Kleinegger,Dan Alistarh
类目: Performance (cs.PF); Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG)
*备注: AdaptFM Workshop Paper ICML 2026
Abstract:Energy supply and heat dissipation are two of the main challenges with modern GPU deployments. While typically discussed in the context of new datacenter constructions, the same constraints also apply to small form-factor consumer devices, such as the DGX spark. In workloads characterized by alternating compute-intensive tasks such as matmuls with memory-bound operations such as norms or cross-entropy, the compute-intensive parts might hit power and/or thermal limits and start throttling. In this short paper, we show that chunking the workload into smaller parts that alternate compute and memory in higher frequencies, these power and temperature spikes can be smoothed out, preventing throttling and resulting in considerably faster wall-clock time and reduced total energy consumption. We present several scenarios in which this effect can be exploited on a DGX Spark with up to 2% performance and energy improvements, and demonstrate that the same phenomenon also happens on less constrained systems, such as a multi-GPU server, albeit at significantly reduced effect size of 1-2%.
[LG-77] Physics-Informed Conformal Prediction: Embedding PDE Consistency into Distribution-Free Uncertainty Quantification for Neural Operators
链接: https://arxiv.org/abs/2609.11935
作者: Michael Chin
类目: Machine Learning (cs.LG)
*备注: 14 pages, 11 tables, 5 figures
Abstract:Neural operators such as the Fourier Neural Operator (FNO) achieve remarkable accuracy in approximating solutions to partial differential equations (PDEs). However, providing rigorous uncertainty estimates remains an open challenge. We propose Physics-Informed Conformal Prediction (PI-CP), a framework that embeds PDE residuals into the nonconformity score of split conformal prediction, producing prediction intervals that are (i) distribution-free with provable coverage guarantees, and (ii) spatially adaptive when the PDE residual correlates with prediction error – tighter where physics is well-satisfied, wider where it is violated. Additionally, we prove that FNO’s translation equivariance creates a fundamental approximation barrier for PDEs with Dirichlet boundary conditions, and show that coordinate channels resolve this with up to 63x error reduction. We validate PI-CP across six physics scenarios – heat conduction (2D/3D), structural mechanics (2D/3D), Darcy flow, and Navier-Stokes – demonstrating consistent 89-91% coverage for all four Conformal methods, while MC Dropout and Deep Ensembles are unstable (82-100%). FNO outperforms CNN and DeepONet by 10-12x.
[LG-78] Fundamental Dynamical Units for Physics-Informed Structural Inference from Perturbation Time-Series in Networked Systems
链接: https://arxiv.org/abs/2609.11934
作者: Nima Nouri
类目: Machine Learning (cs.LG); Computational Physics (physics.comp-ph)
*备注:
Abstract:In networked dynamical systems, the parameter of primary mechanistic interest is signed interaction structure. Recovering this structure from perturbation time-series data is a fundamental identification problem, compounded by three coupled obstacles: the combinatorial complexity of interaction architectures, ambiguity of causal attribution under limited interventions, and state-dependent dynamics that confound structural inference. Each obstacle is structural in origin and calls for a structural solution. We address these challenges by adopting a reductionist approach, introducing Fundamental Dynamical Units (FDUs): signed three-node interaction patterns as composable primitives that convert the interaction hypothesis space into a finite, constructive, and tractable representation. We show that local interaction structure determines the perturbation conditions required to disentangle direct from relayed influence, making intervention design a structural consequence of the FDU representation. We embed FDU-regularized structural inference within a physics-informed neural ordinary differential equation (ODE) whose governing-equation constraint transforms structural hypotheses into verifiable dynamical predictions, enabling joint recovery of interaction structure and perturbation-resolved trajectories. Validated on synthetic benchmarks with known ground truth, the framework supports structural commitment, expressed through FDU primitives, motif-prescribed intervention design, and physics-informed learning, as a principled basis for mechanistically interpretable inference in networked dynamical systems.
[LG-79] A Ranking Approach for Measuring Calibration
链接: https://arxiv.org/abs/2609.13100
作者: Anirban Chatterjee,Rina Foygel Barber
类目: Methodology (stat.ME); Machine Learning (cs.LG); Machine Learning (stat.ML)
*备注:
Abstract:When providing forecasted probabilities with a predictive model, the ideal model offers perfect calibration: the true probability of the outcome (i.e., the probability that Y=1 ) exactly matches the forecasted probability f(X) . In practice, models inevitably exhibit calibration error, and it is therefore important to be able to measure this miscalibration to assess a model’s reliability. The Expected Calibration Error (ECE) is the most widely used measure of miscalibration, but is known to be impossible to estimate the ECE with guaranteed accuracy in an assumption-free setting. In this work, we propose an alternative measure, the rankECE, that is based on comparing points with neighboring values of the predicted probability f(X) . Our theoretical guarantees and empirical results establish that rankECE provides a better proxy for ECE as compared to binned approximations to ECE, which are the most commonly-used approximations in practice.
[LG-80] Physics-enriched neural solvers for transient ice-flow simulation
链接: https://arxiv.org/abs/2609.12900
作者: Thomas Gregov,Sebastian Rosier,Brandon Finley,Andreas Vieli,Guillaume Jouvet
类目: Computational Physics (physics.comp-ph); Machine Learning (cs.LG); Geophysics (physics.geo-ph)
*备注:
Abstract:Transient glacier simulations with higher-order ice flow require the repeated solution of a nonlinear problem as the geometry evolves. In the online mode of the Instructed Glacier Model, the velocity field is represented by a neural network whose weights are warm-started from the previous time step and updated with a few optimizer iterations. We show that supplying the network with inexpensive input fields derived from low-order ice-flow balances improves this online solver. Unlike residual-based physics-informed neural networks, which incorporate physics through governing-equation penalties in the loss, our approach leaves the governing energy objective unchanged, adding physical structure through the network inputs. Across three real-world glacier configurations, the enriched solver is markedly more robust to solver settings. On the two alpine cases, it also improves the tuned accuracy–runtime trade-off, reducing surface-velocity errors by factors of two to four at fixed runtime and reaching few-percent relative errors with only 10^4 – 10^5 trainable parameters, far fewer than comparable raw-input baselines. A 300-year Aletsch simulation then completes in under one minute, and the larger Valais domain in about two minutes, on a single GPU—a budget once reserved for much simpler shallow-ice models. Gains are smaller for the fast marine-terminating glacier, where nonlocal stress coupling favors larger or spectral networks. More broadly, the results suggest that enriching a neural solver’s inputs with reduced-order physics can make repeated higher-order solves much cheaper, with no training data and no offline training.
[LG-81] High-Probability Convergence of SGD via Batched Updates
链接: https://arxiv.org/abs/2609.12765
作者: Feng Zhu,Robert W. Heath Jr.,Aritra Mitra
类目: Optimization and Control (math.OC); Machine Learning (cs.LG); Systems and Control (eess.SY)
*备注: To appear at the 65th IEEE Conference on Decision and Control
Abstract:Stochastic gradient descent (SGD) is the primary workhorse for large-scale optimization. While the average behavior of its iterates, typically characterized by mean-squared error bounds, is well-understood, obtaining high-probability guarantees for the last iterate remains challenging. Prior approaches to this problem have either imposed restrictive assumptions (such as bounded domains or gradients) or relied on complex proofs involving auxiliary sequences. In this work, we propose Batched SGD, a simple variant that partitions online samples into epochs and performs a single update per epoch using a refined, low-variance gradient estimate. Our main contribution demonstrates that this batching mechanism enables a surprisingly simple high-probability analysis that avoids both restrictive assumptions and auxiliary sequences. Under standard smoothness and norm-sub-Gaussian noise assumptions, we establish near-optimal rates for both strongly convex and non-convex objectives. Furthermore, we show that our batching idea extends naturally to federated learning (FL). We provide the first high-probability guarantees for FL, achieving logarithmic communication complexity, linear speedup in the number of agents, and resilience to data heterogeneity.
[LG-82] Prism-SQA: An Interpretable and Adaptable Neural Framework for Surface Electromyography Quality Assessment ALT
链接: https://arxiv.org/abs/2609.12724
作者: Kuan-Chen Wang,Kai-Chun Liu,Ping-Cheng Yeh,Sheng-Yu Peng,Yu Tsao
类目: ignal Processing (eess.SP); Machine Learning (cs.LG)
*备注: 14 pages, 9 figures, 7 tables, accepted by IEEE Journal of Biomedical and Health Informatics (JBHI)
Abstract:sEMG is vulnerable to various contaminants that distort signal morphology and spectral content. Accurate signal quality assessment (SQA) is essential for identifying such degradation and ensuring reliable clinical analyses and decisions. Recent neural network-based SQA methods achieve accurate quality estimation by learning complex contamination patterns, yet their black-box nature prevents clinicians from understanding or validating the reported scores and limits adaptability to application-specific quality definitions without retraining. To address these limitations, we propose Prism-SQA, an interpretable and adaptable neural framework that reformulates SQA as a physiology-aware source-separation and verification process. Prism-SQA decomposes each input signal into a clean sEMG component and five contaminant-specific components using a U-Net with bidirectional long short-term memory. Each separated contaminant component is examined by a Contaminant Fingerprint Verifier, which enforces physiological plausibility by comparing its temporal and spectral structure with canonical contaminant signatures. This design allows clinicians to inspect how each contaminant affects signal quality, grounding the assessment in transparent, signal-level evidence rather than opaque latent representations. Quality indices computed from the verified components further enable customization of quality criteria across clinical contexts without retraining. We evaluate Prism-SQA on continuous quality-score estimation using synthesized noisy sEMG from public Ninapro datasets and on binary quality classification using a clinical dysphagia dataset. Results show that Prism-SQA achieves competitive or better performance than contemporary black-box neural methods while providing explicit interpretability and adaptability, advancing toward practical and clinically aligned sEMG SQA.
[LG-83] Inferring Dislocation Microstructures from X-ray Diffraction via Cross-Modal Contrastive Learning
链接: https://arxiv.org/abs/2609.12713
作者: Benjamin Udofia,Nicolas Bertin,Markus Stricker
类目: Materials Science (cond-mat.mtrl-sci); Machine Learning (cs.LG)
*备注: 15 pages, 9 figures, preprint
Abstract:Understanding and inferring dislocation microstructures from diffraction patterns remains an open challenge in materials characterization, as diffraction measurements provide only indirect information about the underlying dislocation structure. In this work, a cross-modal learning framework is developed to enable the prediction of 3D dislocation structures directly from diffraction data. Dislocation density fields generated from discrete dislocation dynamics simulations are paired with corresponding virtual X-ray diffraction patterns and embedded into a shared 2D latent space using contrastive learning. The alignment between structural and diffraction representations of dislocation structures is evaluated directly in the learned latent space using correlations between corresponding latent features. To estimate the role of dataset size for this approach, farthest point sampling is employed to construct representative and diverse training subsets of varying sizes. The results show strong cross-modal alignment and that model performance improves rapidly with increasing dataset size. Near-saturation is achieved with approximately 500 representative observations from a dataset of 10,000 observations, enabling accurate prediction of dislocation density fields from previously unseen diffraction data of the same distribution. Qualitative comparisons confirm that the predicted structures capture the dominant spatial features of the underlying dislocation microstructures. These findings demonstrate an efficient approach for learning structure-diffraction relationships and highlight the potential for inferring structural characteristics of dislocation networks directly from diffraction patterns, providing a pathway toward diffraction-based structural analysis and future extension to experimental data.
[LG-84] ght Sampling Complexity with stochastic gradient oracles in Fixed Dimensions
链接: https://arxiv.org/abs/2609.12590
作者: Weiming Ou,Xiao Wang
类目: atistics Theory (math.ST); Machine Learning (cs.LG); Probability (math.PR)
*备注: 55 pages
Abstract:We investigate the stochastic-gradient query complexity of sampling smooth strongly log-concave distributions in any fixed Euclidean dimension. The potential is \mu -strongly convex and L -smooth, with an unknown mode in the ball of radius \mu^-1/2 about the origin. We have access to unbiased stochastic oracles with the variance at most \sigma^2 . For every \sigma^2\ge0 and total variation (TV) accuracy 0\varepsilon\le1/10 , we prove that the tight complexity of sampling a distribution within \epsilon -TV distance from the target distribution is [ N^\star_\textTV=\Theta!\left(\log(1+\kappa)+ \frac\sigma^2\mu\epsilon\right), ] where \kappa:=\frac L\mu is the condition number. Note that this complexity bound is simultaneously tight for the condition number \kappa and accuracy \epsilon . Besides, our tight complexity bound is adaptive to noiseless setting \sigma=0 , which is N^\star_\textTV=\Theta!\left(\log(1+\kappa)\right) . Comments: 55 pages Subjects: Statistics Theory (math.ST); Machine Learning (cs.LG); Probability (math.PR) Cite as: arXiv:2609.12590 [math.ST] (or arXiv:2609.12590v1 [math.ST] for this version) https://doi.org/10.48550/arXiv.2609.12590 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Weiming Ou [view email] [v1] Fri, 11 Sep 2026 08:44:06 UTC (57 KB)
[LG-85] Learning-Augmented Optimization for Strategic Two-Echelon Spare Parts Network Design
链接: https://arxiv.org/abs/2609.12524
作者: Donato Maragno,Marco Caserta,Alberto Sinigaglia,Komlanvi Ametana,David Corredor Montenegro,Luca D’Angelo
类目: Optimization and Control (math.OC); Machine Learning (cs.LG)
*备注:
Abstract:We study the strategic design of a two-echelon spare-parts inventory network where evaluating each candidate topology requires an expensive inventory optimization model. The design partitions hundreds of sites into feasible clusters and selects a central replenishment site for each cluster to reduce costs while maintaining service levels. Because the optimizer favors candidates with high predicted savings, it can exploit optimistic surrogate errors. We develop a conservative framework combining a graph neural network ensemble, variable neighborhood search, and set-partitioning recombination. The surrogate is trained on exact cluster evaluations, while a lower quantile of ensemble-predicted savings guides the search to limit optimism. Clusters found during the search are recombined through set partitioning using surrogate-based objective coefficients. The resulting network is evaluated with the exact inventory model, and only this evaluation is used to report performance. In a case study of 246 fulfillment centers in Amazon’s North American network, the framework improves combined savings by 30.5% over an optimization baseline based entirely on exact cluster evaluations, while maintaining approximately 99.8% service across six independent replications. Under equal computational budgets, graph-surrogate-guided search achieves higher mean exact savings than a tabular alternative under both scoring schemes. Conservative scoring improves mean savings for both surrogate classes and reduces the share of final-network clusters overestimated by the graph surrogate from 68% to 28%. Predictive and ranking accuracy deteriorate among search-generated candidates with high surrogate scores, indicating that random holdout performance can incompletely characterize surrogate quality during optimization. Subjects: Optimization and Control (math.OC); Machine Learning (cs.LG) Cite as: arXiv:2609.12524 [math.OC] (or arXiv:2609.12524v1 [math.OC] for this version) https://doi.org/10.48550/arXiv.2609.12524 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-86] A Splitting Method for SDE Terminal-Law Estimation
链接: https://arxiv.org/abs/2609.12513
作者: Rushil Gupta,Sandeep Juneja
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注:
Abstract:In many settings involving stochastic differential equations, including in diffusion based generative AI, our aim is to accurately generate samples from a terminal distribution. Typically, this is done by generating i.i.d. samples of diffusion paths. Given a fixed simulation budget, a reasonable way to gain efficiency may be to instead generate a tree of paths through appropriately split partial paths. This suggests improved performance, but one worries about the injected dependence. In this paper, we study this issue comprehensively. With Kolmogorov-Smirnov distance as a measure of accuracy, we identify the limiting errors of the associated empirical distributions as the simulation budget increases to infinity. We characterize a splitting strategy motivated by a corresponding asymptotic optimization problem. The theoretical results bring out the elegant underlying structure in the problem. Practical implementation involves two phases, an initial estimation phase and a final inference phase. Overall, we observe a 10-25% improvement in mean error over i.i.d. samples in many settings. In an exploratory CIFAR-10 study, our method reduces the maximum mean discrepancy by 8-13%.
[LG-87] Inference for Newton Methods with Accelerated Sketch-and-Project via Random Scaling
链接: https://arxiv.org/abs/2609.12421
作者: Xinchen Du,Elizaveta Rebrova,Michał Dereziński,Sen Na
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Numerical Analysis (math.NA); Optimization and Control (math.OC); Statistics Theory (math.ST)
*备注: 52 pages, 3 figures, 4 tables
Abstract:We study an online sketched Newton method that approximates the Newton direction at each step via a state-of-the-art sketching solver, called the generalized accelerated sketch-and-project solver (GAS), thereby mitigating the computational bottleneck of classical second-order methods. The GAS solver improves upon vanilla, unaccelerated sketch-and-project solvers by achieving accelerated convergence through Nesterov momentum updates, and accommodates a flexible projection metric whose proper choice further reduces computational cost. Building on this design, we establish asymptotic normality of the averaged sketched Newton iterates and characterize their limiting covariance matrix. The resulting covariance recovers that of the unaccelerated sketched Newton method under a specific choice of acceleration parameters, converges more rapidly (in the number of sketching steps) to the minimax-optimal covariance in general, and is smaller than that of the last iterate produced by the accelerated method. Finally, we strengthen these results by establishing a functional central limit theorem for the Newton iterates, which allows us to bypass explicit covariance estimation and develop an online inference procedure based on random scaling. Specifically, we construct a pivotal test statistic by appropriately rescaling the averaged iterates, so that its limiting distribution is free of any unknown parameters, enabling asymptotically valid online inference. Numerical experiments demonstrate superior performance of the proposed inference procedure.
[LG-88] Membership Inference via Pairwise Likelihood Ratios
链接: https://arxiv.org/abs/2609.12367
作者: Shengjie Niu,Zebin Yun,Yeheng Ge,Jian Huang
类目: Machine Learning (stat.ML); Machine Learning (cs.LG)
*备注: 33 pages, 7 figures and 11 tables
Abstract:Membership inference attacks (MIAs) are the standard tool for auditing the privacy risks of machine learning models. Given a query point, an MIA aims to determine whether that point was used to train the target model. In practice, such inference must rely on the statistical signals exposed by the model’s outputs, such as confidence scores, logits, and intermediate feature representations. However, existing methods often fail to efficiently summarize and combine these statistical signals. To address this limitation, we propose Pairwise Likelihood MIA (PL-MIA), a unified method that combines a Gaussian likelihood-ratio (GLR) statistic with population calibration and the Cauchy combination test. We characterize theoretically how the GLR retains variance-contraction signals and establish conditions under which population calibration and Cauchy combination improve attack power. We obtain p -values from pairwise comparisons between the query point and reference points not used for training, and aggregate these continuous signals using the Cauchy combination test. This preserves the evidence strength that is discarded when each pairwise comparison is reduced to a binary vote. Extensive experiments demonstrate that PL-MIA outperforms strong baselines, improving the true positive rate (TPR) by over 25% in the critical low-false-positive regime, corroborating our theoretical findings. These results demonstrate how statistical principles can turn noisy model outputs into more powerful, calibrated, and reproducible evidence for membership privacy auditing.
[LG-89] Fast BIB simulation at a future Muon Collider with generative machine learning
链接: https://arxiv.org/abs/2609.12054
作者: Radha Mastandrea,Shiyu Peng,Benjamin Rosser,Matt LeBlanc
类目: Instrumentation and Detectors (physics.ins-det); Machine Learning (cs.LG); High Energy Physics - Experiment (hep-ex)
*备注: 34 pages, 18 figures
Abstract:Beam-induced background (BIB) from muon decay products will be an overwhelming and unavoidable background at a future Muon Collider. In order to develop robust event reconstruction algorithms, we need large amounts of accurate BIB simulation to test on. BIB simulation is currently compute-limited: the simulated sample presently used for BIB overlay, which statistically represents approximately 10% of a single unique event’s worth of simulated BIB, requires on the order of 10^6 HS23 \cdot hours to generate and occupies approximately 100 GB on disk. In this work, we develop the first machine learning models for fast BIB generation in tracking detectors. We consider two classes of architectures: a slower but higher-fidelity tabular diffusion model, and a faster but lower fidelity circular spline flow model. We find that both classes of architectures produce BIB hits and tracks that closely resemble those of available full simulation hits and tracks, and that the machine learning models can produce BIB in over an order of magnitude less time than what is needed to produce full simulation BIB. We release the model weights with the paper so that the Muon Collider community can use these fast BIB hits for future RD.
[LG-90] Learning the Geometry of Collider Events with Metric-Aware Deep Sets
链接: https://arxiv.org/abs/2609.12024
作者: Lauren Hay,Rishabh Jain,Matt LeBlanc,Jennifer Roloff
类目: High Energy Physics - Phenomenology (hep-ph); Machine Learning (cs.LG)
*备注: 14 pages including appendices, 5 figures. Preliminary
Abstract:Optimal transport gives structured data a geometry, but exact evaluation is costly in large pairwise analyses that exploit relationships among distances. Learned surrogates are faster, but need not preserve this metric structure. We develop a Deep Sets surrogate for OT between variable-size weighted point clouds that enforces non-negativity, exchange symmetry, and zero self-distance, leaving the triangle inequality unconstrained. Applied to the Energy Mover’s Distance between collider events in a particle physics application, the Metric-Aware Particle Flow Network achieves percent-level mean absolute percentage error while significantly improving inference throughput over other exact and approximate methods surveyed. The architectural constraints are found to improve properties that are not explicitly enforced: across 10^6 held-out event triplets, triangle-inequality violations fall from 199 for a matched unconstrained network to 2, and the maximum from 149.5 to 5.8 GeV. These results demonstrate that targeted inductive biases can yield fast neural surrogates with substantially improved geometric fidelity.
[LG-91] Learning Interaction Kernels from Collective Steady States
链接: https://arxiv.org/abs/2609.12004
作者: Baoli Hao,Mauro Maggioni,Ming Zhong
类目: Machine Learning (stat.ML); Machine Learning (cs.LG); Dynamical Systems (math.DS); Numerical Analysis (math.NA); Adaptation and Self-Organizing Systems (nlin.AO)
*备注:
Abstract:We propose a learning procedure for system identification in interacting particle systems from single-snapshot observations of collective behaviors, unlike existing approaches that rely on observations of trajectories. This setting leads to a fundamentally ill-posed inverse problem, which we solve by using a regularization strategy based on the empirical distribution of observed configurations, drawn from different, unobserved initial conditions. We test our learning procedure on a variety of representative models with steady-state and quasi-stationary patterns, where collective behaviors encode implicit information about the interaction mechanisms, demonstrating that our approach enables stable and accurate recovery of the underlying interaction laws, leading to faithful reproduction of the collective behavior, and in many cases even of the dynamics leading up to it.
[LG-92] Impact of Multiple Non-Invasive Biosignals on Cardiovascular Biomarker Estimation via Simulation-Based Inference
链接: https://arxiv.org/abs/2609.11969
作者: Shusaku Maeda,Masahiro Nakano,Tomoharu Iwata,Kenji Komiya,Ryo Nishikimi,Kunio Kashino
类目: Medical Physics (physics.med-ph); Machine Learning (cs.LG); Applications (stat.AP)
*备注: 8 pages, 5 figures, 2 tables
Abstract:As the population ages, the number of patients with cardiovascular diseases continues to increase, highlighting the need for early detection before progression to severe and irreversible functional decline. Consequently, estimating cardiovascular biomarkers from non-invasive biosignals, such as photoplethysmography (PPG) and arterial pressure wave (APW) signals, has attracted increasing attention. These signals can be measured using wearable and cuff-type devices. Previous studies have used PPG and APW signals to estimate cardiovascular biomarkers. However, these signals exhibit strong similarities in both the temporal and frequency domains and primarily reflect peripheral and arterial pulse waveforms. Therefore, they may provide limited information about cardiac mechanical function. In contrast, the quantitative impact of additional biosignals, such as ballistocardiography (BCG), which reflect the body’s minute mechanical responses to cardiac ejection, remains unclear. In this study, we generated synthetic PPG, APW, and BCG signals from a unified whole-body cardiovascular circulation model and evaluated the complementary contribution of BCG to probabilistic cardiovascular biomarker estimation. We estimated posterior distributions of cardiovascular biomarkers using neural posterior estimation and simulation-based inference. The results showed that adding BCG signals significantly improved estimation performance. Furthermore, even in ill-posed cases where PPG and APW alone produced multimodal posterior distributions, adding BCG yielded unimodal posterior distributions. These findings provide fundamental insights into signal selection for estimating cardiovascular dynamics.
[LG-93] Exact ReLU realization of binary affine refinement iterates via reflection folding and cone switching
链接: https://arxiv.org/abs/2609.11962
作者: Boldsaikhan Bolorkhuu,Tsogtgerel Gantumur
类目: Optimization and Control (math.OC); Machine Learning (cs.LG)
*备注: 25 pages, no figures
Abstract:We study vector-valued binary affine refinement operators with finitely supported matrix masks and compactly supported continuous piecewise linear input and forcing data. We prove that every finite refinement iterate admits an exact ReLU realization of fixed width and depth linear in the number of iterations. No separation of the forcing profile from the binary cell seams is required. The main mechanism is universal reflection doubling. Pairing each residual profile with its reflection replaces the two binary transition matrices by one fixed block matrix together with a fixed swap involution. The cell-seam identity makes the two branch candidates agree at the tent fold, while their swap-odd component is bounded linearly by the distance to the fold. This permits exact branch selection by a fixed continuous piecewise linear cone switch, without multiplication by a variable selector. The resulting primal recursion requires the residual orbit in reverse order. We obtain exact backward replay from the residual memory controller developed previously for affine refinement, interpreted here through the reflection quotient of circle doubling. The construction propagates the full vectorized profiles rather than decomposing the input and forcing into reference atoms. We also treat stage-dependent forcing from a fixed finite-dimensional family and show that genuine reflection equivariance reduces the doubled cascade to a single parity sector. Comments: 25 pages, no figures Subjects: Optimization and Control (math.OC); Machine Learning (cs.LG) MSC classes: Primary 41A46, Secondary 68T07, 42C40 Cite as: arXiv:2609.11962 [math.OC] (or arXiv:2609.11962v1 [math.OC] for this version) https://doi.org/10.48550/arXiv.2609.11962 Focus to learn more arXiv-issued DOI via DataCite
[LG-94] owards Sustainable Hydrogen Systems: Supply Chain Optimization with Model Predictive Control and Reinforcement Learning
链接: https://arxiv.org/abs/2609.11933
作者: Mahammad Valiyev
类目: Optimization and Control (math.OC); Machine Learning (cs.LG)
*备注:
Abstract:Hydrogen supply chains are expected to play a central role in future low-carbon energy systems by enabling renewable energy integration, long-duration storage, and decarbonization of industrial and transportation sectors. However, their operation is challenged by renewable generation variability, electricity price fluctuations, uncertain hydrogen demand, and engineering constraints associated with electrolyzers, energy storage, and grid interaction. As hydrogen infrastructure expands toward commercial deployment, operational strategies must balance economic performance, reliability, and sustainability under dynamic and uncertain conditions. This paper investigates and compares four control approaches for a renewable-powered hydrogen supply chain: a rule-based controller (RBC), model predictive control (MPC), reinforcement learning without forecasts (RL-NF), and reinforcement learning with forecast-augmented observations (RL-F). All methods are evaluated within a unified, physically realistic framework incorporating electrolyzer minimum-load and ramp-rate constraints, battery and hydrogen storage dynamics, grid import limits, and consistent economic assumptions, enabling a fair comparison under identical operating conditions. Simulation results show that MPC achieves the highest economic performance by exploiting short-term forecasts to coordinate storage, reduce grid dependence, and improve efficiency. RL-NF demonstrates robust and competitive performance without future information, highlighting the capability of learning-based methods to discover effective policies from experience. RL-F does not consistently outperform its no-forecast counterpart, suggesting that forecast uncertainty and increased state complexity can limit forecast-augmented learning. The results provide guidance for selecting operational control strategies in future hydrogen energy systems. Subjects: Optimization and Control (math.OC); Machine Learning (cs.LG) Cite as: arXiv:2609.11933 [math.OC] (or arXiv:2609.11933v1 [math.OC] for this version) https://doi.org/10.48550/arXiv.2609.11933 Focus to learn more arXiv-issued DOI via DataCite
附件下载
